From ksrh at skicentral.tv Tue May 1 18:36:23 2007 From: ksrh at skicentral.tv (Darryl Wisneski) Date: Tue, 1 May 2007 14:36:23 -0400 Subject: drac ilo install via virtual cdrom In-Reply-To: <2992.165.236.183.58.1177942443.squirrel@raven.ontheside.net> References: <20070426155117.GA44382@commwebworks.com> <1c59d60b0704300618v32889d40y3015991c60393c38@mail.gmail.com> <2992.165.236.183.58.1177942443.squirrel@raven.ontheside.net> Message-ID: <20070501183623.GA60243@commwebworks.com> On Mon, Apr 30, 2007 at 08:14:03AM -0600, Matt Sturtz wrote: > Hi Brian-- > > >> I am trying to perform a kickstart install of RHEL4U4 exclusively > >> via the drac (Dell Remote Access Card) interface onto a PE 2950 > >> without utilizing the systems other embedded nics. > > Although I haven't tried it, you should be able to do this, as long as > it's a media-based install (meaning you're using CD's, even though they're > virtual -- the virtual drive shows up as a USB device, so Linux should be > happy to deal with it). > > Note that Linux can NOT see the NIC in the DRAC -- if you're trying to do > a PXE or other Net-based install, you will need to have one of the main > system NIC's connected. Hi Matt: No booting from pxe. Only virtual cdrom. > > Have you actually tried it? If so, what problem(s) did you have (sorry, I > missed the original post -- just got a reply)... http://www.redhat.com/archives/kickstart-list/2007-April/date.html Attached is my ks.cfg. I configured the bios to boot from "virtual cdrom." The kickstart proceeeds to load vmlinuz and initrd.img where /sbin/loader gets ks.cfg from. Embedding the ks.cfg in the initrd.img is the only way I can get the first step loader to find the ks.cfg. Emedding the ks.cfg is known to /sbin/loader via this statement in isolinux.cfg: isolinux/isolinux.cfg append initrd=initrd.img ramdisk_size=8192 ks=file:/ks.cfg Anaconda loads the megaraid and usb-storage driver, then asks for a RHEL Linux CD. tty0 shows "CD Not Found..." tty2 shows: getting kickstart file setting up kickstrt KickstartFromCD going to set lang to en_US.UTF-8 settting lang.. starting to STEP_URL trying to mount CD device scd0 tty3 shows: Initialized USB Mass Storage driver... sci4 : SCSI emulation for USB Mass Store decives Vendor: Dell Model: Viurtal CDROM .. Typel CD-ROM sr0: scsi4-mmc drive: 24x/24x/cd/rw/xa/form2 caddy Attached scsi CD-ROM sr0 at scsi3, channel 0, id 0, lun 0 USB Mass Storeage support registered. ISO 9660 Extensions: Microsfot Jolet Level 3 Unable to load NLS charset utf8 Unable to load NLS charset utf8 ISO 9660 Extensions: RRIP_1991A I am giving the command cdrom in ks.cfg I tried messing around with the bootloader parameters to change how anaconda looks for the cdrom. Maybe I am on the right track here. Anaconda needs to look at sr0 instead of scd0 for the cdrom. #bootloader --location=mbr --append="hdd=ide-scsi" bootloader --location=mbr --append="sdd=ide-scsi" #bootloader --location=mbr --append="sr0=ide-scsi" #bootloader --location=mbr --append="sr0=ide-scsi" --driveorder=sr0 #bootloader --location=mbr --append="scd0=ide-scsi" --driveorder=scd0,sr0 #bootloader --location=mbr --append="scd0=ide-scsi" Thanks for your help. Darryl Wisneski -dkw -------------- next part -------------- #System language lang en_US #Language modules to install langsupport --default=en_US #System keyboard keyboard us #System mouse mouse none #Sytem timezone timezone America/New_York #Root password rootpw --iscrypted ... #Reboot after installation reboot #Use text mode install text #Install Red Hat Linux instead of upgrade install #System bootloader configuration #url --url http://10.21.11.92/kickstart cdrom ################################################################################ # Setup the disk drives. Two SAS drives sda and sdb. # Partition two drives the same way. # Root partition can't do LVM so leave it alone # # SDA # sda1 - 1G - half of md0 device for /boot # sda2 - 4G - swap - not much point to RAID swap # sda3 - Remainder - half of md1 device for LVM # SDB # sdb1 - 1G - half of md0 device for /boot # sdb2 - 4G - swap some more # sdb3 - Remainder - half of md1 device for LVM # # It seems you'll have to setup grub on both drives. I don't think Anaconda # will do it for some reason. After you boot and get a blank screen just # boot from the rescue CD and do this # # chroot /mnt/sysimage # # grub # grub>root (hd0,0) # grub>setup (hd0) # grub>root (hd1,0) # grub>setup (hd1) # # Clear out the Master Boot Record zerombr yes # where will be put the boot loader? - test me #bootloader --location=partition #bootloader --location=mbr --append="hdd=ide-scsi" bootloader --location=mbr --append="sdd=ide-scsi" #bootloader --location=mbr --append="sr0=ide-scsi" #bootloader --location=mbr --append="sr0=ide-scsi" --driveorder=sr0 #bootloader --location=mbr --append="scd0=ide-scsi" --driveorder=scd0,sr0 #bootloader --location=mbr --append="scd0=ide-scsi" # Setup # hda part raid.a1 --noformat --size=1024 --ondisk=sda --asprimary part swap --size=4096 --ondisk=sda --asprimary part raid.a3 --noformat --size=1 --grow --ondisk=sda # hdb part raid.b1 --noformat --size=1024 --ondisk=sdb --asprimary part swap --size=4096 --ondisk=sdb --asprimary part raid.b3 --noformat --size=1 --grow --ondisk=sdb # # /boot on 1G RAID1 mirror raid /boot --level=1 --fstype=ext3 --device=md0 raid.a1 raid.b1 # LVM RAID1 array on 3rd partition of drives A,B raid pv.md1 --level=1 --fstype="physical volume (LVM)" --device=md1 raid.a3 raid.b3 # setup LVM for OS partitions # create volume group from md1 volgroup VGRAID pv.md1 # chop up volume group VGRAID into logical volumes logvol / --fstype=ext3 --name=LVRoot --vgname=VGRAID --size=1024 logvol /usr --fstype=ext3 --name=LVUsr --vgname=VGRAID --size=8192 logvol /usr/local --fstype=ext3 --name=LVUsrLocal --vgname=VGRAID --size=8192 logvol /home --fstype=ext3 --name=LVHome --vgname=VGRAID --size=10240 logvol /tmp --fstype=ext3 --name=LVTmp --vgname=VGRAID --size=4096 logvol /var --fstype=ext3 --name=LVVar --vgname=VGRAID --size=512 --grow #bootloader --location=mbr #Clear the Master Boot Record #zerombr yes #Partition clearing information #clearpart --all --initlabel #Disk partitioning information #part /boot --fstype "ext3" --size=256 --asprimary --ondisk=sda #part / --fstype "ext3" --size=1024 --ondisk=sda # at least 1x physical memory #part swap --size=8192 --ondisk=sda #part /usr --fstype "ext3" --size=8192 --ondisk=sda #part /usr/local --fstype "ext3" --size=8192 --ondisk=sda #part /home --fstype "ext3" --size=10240 --ondisk=sda #part /tmp --fstype "ext3" --size=2048 --ondisk=sda ##part /opt --fstype "ext3" --size=15360 --ondisk=sda #part /var --fstype "ext3" --grow --size=512 --ondisk=sda auth --useshadow --enablemd5 #network --bootproto=dhcp --device=eth0 firewall --disabled #Do not configure XWindows skipx ################################################################################ # pick software to install # This is a minimal software set for a webserver. It has no GUI nor other cruft %packages --resolvedeps @ Development Tools #@ Kernel Development kernel kernel-smp grub ntp zsh lynx curl lvm2 mdadm ################################################################################ # post install commands %post /usr/bin/eject /dev/hda /usr/bin/eject /dev/hdc # add users /usr/sbin/useradd -c "my user" -g 100 -d /home/myuser -s /bin/bash -m myuser # nfs mount on the new minted os a /mnt/temp, so we can put stuff on the machine /bin/mkdir /mnt/temp #/bin/mount -o nolock 10.21.11.92:/var/www/html/kickstart /mnt/temp # for update6 -dkw #/bin/rpm -i /mnt/temp/httpd-2.0.46-54.ent.i386.rpm /bin/sh /mnt/temp/sshk # josh added this for a bios update... not sure how it applies # /bin/cp /mnt/temp/4.5_OM_SUU_A00.iso /home/josh/4.5_OM_SUU_A00.iso /bin/tar xvpf /mnt/temp/fd.tar /bin/umount /mnt/temp /bin/rm -rf /mnt/temp # turn off chkconfig stuff /sbin/chkconfig --level 2345 xinetd off /sbin/chkconfig --level 2345 rhnsd off /sbin/chkconfig --level 2345 gpm off /sbin/chkconfig --level 2345 kudzu off /sbin/chkconfig --level 2345 apmd off /sbin/chkconfig --level 2345 acpid off /sbin/chkconfig --level 2345 iptables off /sbin/chkconfig --level 2345 ip6tables off /sbin/chkconfig --level 2345 pcmcia off /sbin/chkconfig --level 2345 mdmonitor off /sbin/chkconfig --level 2345 smartd off From forltran at yahoo.com Thu May 3 19:22:14 2007 From: forltran at yahoo.com (Lambert Tran) Date: Thu, 3 May 2007 12:22:14 -0700 (PDT) Subject: Kickstart from a thumb drive? In-Reply-To: <2992.165.236.183.58.1177942443.squirrel@raven.ontheside.net> Message-ID: <169033.76393.qm@web83209.mail.mud.yahoo.com> Greetings, Has anyone successfully done a kickstart installation from a USB thumb drive? I'd appreciate any info on this subject. -thanks -Lambert Tran. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From CallahanT at tessco.com Thu May 3 19:24:50 2007 From: CallahanT at tessco.com (Callahan, Tom) Date: Thu, 03 May 2007 15:24:50 -0400 Subject: Kickstart from a thumb drive? In-Reply-To: <169033.76393.qm@web83209.mail.mud.yahoo.com> Message-ID: You just define ks=sda or ks=xxx depending on how Linux will see the device On 5/3/07 3:22 PM, "Lambert Tran" wrote: > Greetings, > > Has anyone successfully done a kickstart installation from a USB thumb drive? > I'd appreciate any info on this subject. > -thanks > -Lambert Tran. > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > ------------------------------- Tom Callahan TDS Engineering 410-229-1361 Tel 410-588-7605 Mobile 410-229-1512 Fax callahant at TESSCO.com TESSCO Your Total Source for Wireless http://www.tessco.com Network Infrastructure Equipment | Mobile Devices & Accessories | Installation, Test, Equipment & Supplies From Klaus.Steden at thomson.net Thu May 3 19:30:59 2007 From: Klaus.Steden at thomson.net (Steden Klaus) Date: Thu, 3 May 2007 15:30:59 -0400 Subject: Kickstart from a thumb drive? Message-ID: I built a network-based installer that booted from USB. You can use syslinux to make the device bootable at which point it would present itself as a regular USB disk. hth, Klaus ----- Original Message ----- From: kickstart-list-bounces at redhat.com To: Discussion list about Kickstart Sent: Thu May 03 12:22:14 2007 Subject: Kickstart from a thumb drive? Greetings, Has anyone successfully done a kickstart installation from a USB thumb drive? I'd appreciate any info on this subject. -thanks -Lambert Tran. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From rspaulding at arc.nasa.gov Fri May 4 14:24:36 2007 From: rspaulding at arc.nasa.gov (Ryan C. Spaulding) Date: Fri, 4 May 2007 07:24:36 -0700 Subject: Kickstart from a thumb drive? In-Reply-To: References: Message-ID: Just to state the obvious but make sure your hardware/bios can do it. On May 3, 2007, at May 3, 2007 [12:30 PM], Steden Klaus wrote: > > I built a network-based installer that booted from USB. You can use > syslinux to make the device bootable at which point it would > present itself as a regular USB disk. > > hth, > Klaus > > > ----- Original Message ----- > From: kickstart-list-bounces at redhat.com bounces at redhat.com> > To: Discussion list about Kickstart > Sent: Thu May 03 12:22:14 2007 > Subject: Kickstart from a thumb drive? > > Greetings, > > Has anyone successfully done a kickstart installation from a USB > thumb drive? > I'd appreciate any info on this subject. > -thanks > -Lambert Tran. > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From Cameron.Kennedy at bovislendlease.com Sat May 5 01:17:37 2007 From: Cameron.Kennedy at bovislendlease.com (Kennedy, Cameron) Date: Fri, 4 May 2007 21:17:37 -0400 Subject: Kickstart from a thumb drive? In-Reply-To: <169033.76393.qm@web83209.mail.mud.yahoo.com> References: <2992.165.236.183.58.1177942443.squirrel@raven.ontheside.net> <169033.76393.qm@web83209.mail.mud.yahoo.com> Message-ID: <544663BCC5B1E34698C8150370A668F303B46093@USATL01ME253.amer.lendlease.com> The only issue I had was making the USB drive bootable. But I used the instructions found on the Redhat site which basically goes: Delete everything from the usb drive. Assuming the USB drive is registered as /dev/sda, you can do the following: dd if=diskboot.img of=/dev/sda Diskboot.img is from the install CD/DVD under images. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Lambert Tran Sent: Thursday, May 03, 2007 3:22 PM To: Discussion list about Kickstart Subject: Kickstart from a thumb drive? Greetings, Has anyone successfully done a kickstart installation from a USB thumb drive? I'd appreciate any info on this subject. -thanks -Lambert Tran. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list ----------------------------------------- This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country. From Bob.Huisman at filterworks.com Mon May 7 14:46:53 2007 From: Bob.Huisman at filterworks.com (Bob Huisman) Date: Mon, 7 May 2007 16:46:53 +0200 Subject: Kickstart from a thumb drive? In-Reply-To: <169033.76393.qm@web83209.mail.mud.yahoo.com> Message-ID: <9313697AF97C424595D8040FE226731E1D6F47@TINY.socs.lan> Yes. You need a diskboot.img corresponding to your to-be-installed edition (ie. You cannot install RHEL 5 Server with a RHEL 5 Client image). If you modify the file syslinux.cfg you can add the path to the ks.cfg so you will not have to type it each time. I believe there is some info in the RHEL Administration Guide, also found on redhat.com Bob Huisman -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Lambert Tran Sent: donderdag 3 mei 2007 21:22 To: Discussion list about Kickstart Subject: Kickstart from a thumb drive? Greetings, Has anyone successfully done a kickstart installation from a USB thumb drive? I'd appreciate any info on this subject. -thanks -Lambert Tran. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From msteinmann at vesbridge.com Mon May 7 17:10:34 2007 From: msteinmann at vesbridge.com (Martin Steinmann) Date: Mon, 7 May 2007 13:10:34 -0400 Subject: RHEL5 kickstart needs to mount cdrom in postinst Message-ID: On RHEL4 and FC5 we have the following code in the post install section of the ks.cfg file. It mounts the CD the system just installed from and copies the CD's content to a recovery partition on HD. In order to be able to mount the cdrom, we had to run start_udev. On RHEL5 this no longer seems to work. The ks.cfg post install section executes, but the /recovery/linux directory remains empty. It looks like as if it was not able to mount the cdrom. Does anyone have an idea what could cause this different behavior? #...Make copy of installation CD to the "/recovery" partition /sbin/start_udev mkdir /recovery/linux mkdir /tempdir mount /dev/cdrom /tempdir cp -a /tempdir/* /recovery/linux cp -a /tempdir/.discinfo /recovery/linux umount /tempdir rm -rf /tempdir --martin -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Mon May 7 17:35:48 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Mon, 07 May 2007 10:35:48 -0700 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: Message-ID: put: sleep 99999999 in your %post before you get to that part of your kickstart. Kickstart the box, open a terminal (Alt-F2), chroot to the build environment, and try to manually run your commands. Once you get it figured out, simply put those commands back into the kickstart. Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Martin Steinmann Sent: Monday, May 07, 2007 10:11 AM To: kickstart-list at redhat.com Subject: RHEL5 kickstart needs to mount cdrom in postinst On RHEL4 and FC5 we have the following code in the post install section of the ks.cfg file. It mounts the CD the system just installed from and copies the CD's content to a recovery partition on HD. In order to be able to mount the cdrom, we had to run start_udev. On RHEL5 this no longer seems to work. The ks.cfg post install section executes, but the /recovery/linux directory remains empty. It looks like as if it was not able to mount the cdrom. Does anyone have an idea what could cause this different behavior? #...Make copy of installation CD to the "/recovery" partition /sbin/start_udev mkdir /recovery/linux mkdir /tempdir mount /dev/cdrom /tempdir cp -a /tempdir/* /recovery/linux cp -a /tempdir/.discinfo /recovery/linux umount /tempdir rm -rf /tempdir --martin -------------- next part -------------- An HTML attachment was scrubbed... URL: From debian at herakles.homelinux.org Mon May 7 21:06:33 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Tue, 08 May 2007 05:06:33 +0800 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: References: Message-ID: <463F94D9.7050105@herakles.homelinux.org> Martin Steinmann wrote: > On RHEL4 and FC5 we have the following code in the post install section of the ks.cfg file. It mounts the CD the system just installed from and copies the CD's content to a recovery partition on HD. In order to be able to mount the cdrom, we had to run start_udev. > > On RHEL5 this no longer seems to work. The ks.cfg post install section executes, but the /recovery/linux directory remains empty. It looks like as if it was not able to mount the cdrom. > > Does anyone have an idea what could cause this different behavior? > > #...Make copy of installation CD to the "/recovery" partition > /sbin/start_udev > mkdir /recovery/linux > mkdir /tempdir > mount /dev/cdrom /tempdir > cp -a /tempdir/* /recovery/linux > cp -a /tempdir/.discinfo /recovery/linux > umount /tempdir > rm -rf /tempdir Seems to me there's enough people around who want the CD to be available to %post scripts that Anaconda should be keeping the CD mounted a little longer. I have several systems here, laptops and compact desktops, that cannot pull a CD in; once it's ejected I need to push the tray or CD back in. I think it's worth a bug report; perhaps you can allege a design fault. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From clumens at redhat.com Tue May 8 14:06:39 2007 From: clumens at redhat.com (Chris Lumens) Date: Tue, 8 May 2007 10:06:39 -0400 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: <463F94D9.7050105@herakles.homelinux.org> References: <463F94D9.7050105@herakles.homelinux.org> Message-ID: <20070508140638.GS17381@exeter.boston.redhat.com> > I have several systems here, laptops and compact desktops, that cannot > pull a CD in; once it's ejected I need to push the tray or CD back in. > > I think it's worth a bug report; perhaps you can allege a design fault. Yep, this was because some steps got moved around. It's tracked by bugs 238711 and 239002, and has been fixed in rawhide besides. - Chris From msteinmann at vesbridge.com Tue May 8 15:05:48 2007 From: msteinmann at vesbridge.com (Martin Steinmann) Date: Tue, 8 May 2007 11:05:48 -0400 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: <463F94D9.7050105@herakles.homelinux.org> Message-ID: > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of John Summerfield > Sent: Monday, May 07, 2007 5:07 PM > To: Discussion list about Kickstart > Subject: Re: RHEL5 kickstart needs to mount cdrom in postinst > > Martin Steinmann wrote: > > On RHEL4 and FC5 we have the following code in the post install section > of the ks.cfg file. It mounts the CD the system just installed from and > copies the CD's content to a recovery partition on HD. In order to be able > to mount the cdrom, we had to run start_udev. > > > > On RHEL5 this no longer seems to work. The ks.cfg post install section > executes, but the /recovery/linux directory remains empty. It looks like > as if it was not able to mount the cdrom. > > > > Does anyone have an idea what could cause this different behavior? > > > > #...Make copy of installation CD to the "/recovery" partition > > /sbin/start_udev > > mkdir /recovery/linux > > mkdir /tempdir > > mount /dev/cdrom /tempdir > > cp -a /tempdir/* /recovery/linux > > cp -a /tempdir/.discinfo /recovery/linux > > umount /tempdir > > rm -rf /tempdir > > > Seems to me there's enough people around who want the CD to be available > to %post scripts that Anaconda should be keeping the CD mounted a little > longer. > > I have several systems here, laptops and compact desktops, that cannot > pull a CD in; once it's ejected I need to push the tray or CD back in. > > I think it's worth a bug report; perhaps you can allege a design fault. You are correct; the problem is that the CD gets ejected before the post install script executes. This is different as compared to the behavior of RHEL4. Not knowing what the rational was to change this, I can only conclude that this is a major problem for us. We use Kickstart to custom install an application on an appliance style product using a slightly modified RHEL5 install CD and we need to be able to add a couple files during post install from the CD. Is there a way to patch Anaconda to get it to not eject the CD? Where should I start looking? --martin > > -- > > Cheers > John > > -- spambait > 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu > > Please do not reply off-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From msteinmann at vesbridge.com Tue May 8 15:26:26 2007 From: msteinmann at vesbridge.com (Martin Steinmann) Date: Tue, 8 May 2007 11:26:26 -0400 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: <20070508140638.GS17381@exeter.boston.redhat.com> Message-ID: > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Chris Lumens > Sent: Tuesday, May 08, 2007 10:07 AM > To: Discussion list about Kickstart > Subject: Re: RHEL5 kickstart needs to mount cdrom in postinst > > > I have several systems here, laptops and compact desktops, that cannot > > pull a CD in; once it's ejected I need to push the tray or CD back in. > > > > I think it's worth a bug report; perhaps you can allege a design fault. > > Yep, this was because some steps got moved around. It's tracked by bugs > 238711 and 239002, and has been fixed in rawhide besides. Could you be more specific? Would I need a new stage2.img file or what part do I need to update? Also, it says the rawhide repository is now part of Fedora. Can you give me a link to the right place? I assume that I should not mix Fedora packages with RHEL5. Thanks --martin > > - Chris > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From clumens at redhat.com Tue May 8 16:55:39 2007 From: clumens at redhat.com (Chris Lumens) Date: Tue, 8 May 2007 12:55:39 -0400 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: References: <20070508140638.GS17381@exeter.boston.redhat.com> Message-ID: <20070508165539.GU17381@exeter.boston.redhat.com> > > Yep, this was because some steps got moved around. It's tracked by > bugs > > 238711 and 239002, and has been fixed in rawhide besides. > > Could you be more specific? Would I need a new stage2.img file or what > part do I need to update? If you add updates=http://people.redhat.com/clumens/238711.img to your boot flags, that will get you the fix for this. You'll still need to remount your CD device in the %post script. And if you want the CD to be ejected at the end, you'll need to add --eject to your shutdown/reboot/halt/whatever command. > Also, it says the rawhide repository is now > part of Fedora. Can you give me a link to the right place? I assume > that I should not mix Fedora packages with RHEL5. No, you don't want to mix RHEL and Fedora packages especially with how divergent Fedora is getting from RHEL5 as time goes on. I don't know where to point you for the latest rawhide repo as a new one has not gone out for a while now. - Chris From msteinmann at vesbridge.com Tue May 8 18:48:28 2007 From: msteinmann at vesbridge.com (Martin Steinmann) Date: Tue, 8 May 2007 14:48:28 -0400 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: <20070508165539.GU17381@exeter.boston.redhat.com> Message-ID: > > If you add updates=http://people.redhat.com/clumens/238711.img to your > boot flags, that will get you the fix for this. Chris I am sorry but I get a 404 error going to the URL pointed out above. Is this an updated stage2.img? Adding it to the boot flags means adding it to isolinux.cfg? Thanks --martin > > - Chris > From clumens at redhat.com Tue May 8 19:22:36 2007 From: clumens at redhat.com (Chris Lumens) Date: Tue, 8 May 2007 15:22:36 -0400 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: References: <20070508165539.GU17381@exeter.boston.redhat.com> Message-ID: <20070508192235.GV17381@exeter.boston.redhat.com> > I am sorry but I get a 404 error going to the URL pointed out above. Whoops, copied it to the wrong place. Try it now. > Is > this an updated stage2.img? No, it's just an updated installmethod.py that will be used in place of the one that comes with anaconda. > Adding it to the boot flags means adding it > to isolinux.cfg? When you boot up and are staring at the boot prompt on the CD, you just add updates=... after the target you want to boot. You can do this with network booting as well. The updates procedure is documented in the docs/install-methods.txt file that comes with anaconda and at http://fedoraproject.org/wiki/Anaconda/Updates. - Chris From Chip.Shabazian at bankofamerica.com Wed May 9 07:14:47 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Wed, 09 May 2007 00:14:47 -0700 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: Message-ID: I'm not sure how much you need available from the cdrom in the %post, but I used a workaround in the past when I wanted to get information from %pre to %post which might help. What I did was create a ramdrive in %pre and copy what I needed to it. As long as you umount the ram drive, you can then mount it in %post and access everything. As long as what you need in your %post is small enough to fit on a ram drive, you can use this as a workaround. Chip -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Martin Steinmann Sent: Tuesday, May 08, 2007 8:06 AM To: Discussion list about Kickstart Subject: RE: RHEL5 kickstart needs to mount cdrom in postinst > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of John Summerfield > Sent: Monday, May 07, 2007 5:07 PM > To: Discussion list about Kickstart > Subject: Re: RHEL5 kickstart needs to mount cdrom in postinst > > Martin Steinmann wrote: > > On RHEL4 and FC5 we have the following code in the post install section > of the ks.cfg file. It mounts the CD the system just installed from and > copies the CD's content to a recovery partition on HD. In order to be able > to mount the cdrom, we had to run start_udev. > > > > On RHEL5 this no longer seems to work. The ks.cfg post install section > executes, but the /recovery/linux directory remains empty. It looks like > as if it was not able to mount the cdrom. > > > > Does anyone have an idea what could cause this different behavior? > > > > #...Make copy of installation CD to the "/recovery" partition > > /sbin/start_udev mkdir /recovery/linux mkdir /tempdir mount > > /dev/cdrom /tempdir cp -a /tempdir/* /recovery/linux cp -a > > /tempdir/.discinfo /recovery/linux umount /tempdir rm -rf /tempdir > > > Seems to me there's enough people around who want the CD to be available > to %post scripts that Anaconda should be keeping the CD mounted a little > longer. > > I have several systems here, laptops and compact desktops, that cannot > pull a CD in; once it's ejected I need to push the tray or CD back in. > > I think it's worth a bug report; perhaps you can allege a design fault. You are correct; the problem is that the CD gets ejected before the post install script executes. This is different as compared to the behavior of RHEL4. Not knowing what the rational was to change this, I can only conclude that this is a major problem for us. We use Kickstart to custom install an application on an appliance style product using a slightly modified RHEL5 install CD and we need to be able to add a couple files during post install from the CD. Is there a way to patch Anaconda to get it to not eject the CD? Where should I start looking? --martin > > -- > > Cheers > John > > -- spambait > 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu > > Please do not reply off-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From msteinmann at vesbridge.com Wed May 9 13:35:02 2007 From: msteinmann at vesbridge.com (Martin Steinmann) Date: Wed, 9 May 2007 09:35:02 -0400 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: <20070508192235.GV17381@exeter.boston.redhat.com> Message-ID: > > > Is > > this an updated stage2.img? > > No, it's just an updated installmethod.py that will be used in place of > the one that comes with anaconda. > > > Adding it to the boot flags means adding it > > to isolinux.cfg? > > When you boot up and are staring at the boot prompt on the CD, you just > add updates=... after the target you want to boot. You can do this with > network booting as well. > > The updates procedure is documented in the docs/install-methods.txt file > that comes with anaconda and at > http://fedoraproject.org/wiki/Anaconda/Updates. > Wow! It worked. Many thanks for all the help. I added the updates.img file into the tree on the CD and created a new CD. It now runs postinstall without ejecting the CD. Is there a similarly elegant way to add a driver disk image? I am still struggling embedding an additional RAID driver into the boot CD. I suppose I need to update the initial initrd file (using mkinitrd) and then tell Anaconda to load the driver image during stage2. Can the driver disk image be placed somewhere into the tree similar to the updates.img? Thanks --martin > - Chris > From jason at rampaginggeek.com Wed May 9 13:54:51 2007 From: jason at rampaginggeek.com (Jason Edgecombe) Date: Wed, 09 May 2007 09:54:51 -0400 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: References: Message-ID: <4641D2AB.9040300@rampaginggeek.com> Martin Steinmann wrote: >>> Is >>> this an updated stage2.img? >>> >> No, it's just an updated installmethod.py that will be used in place >> > of > >> the one that comes with anaconda. >> >> >>> Adding it to the boot flags means adding it >>> to isolinux.cfg? >>> >> When you boot up and are staring at the boot prompt on the CD, you >> > just > >> add updates=... after the target you want to boot. You can do this >> > with > >> network booting as well. >> >> The updates procedure is documented in the docs/install-methods.txt >> > file > >> that comes with anaconda and at >> http://fedoraproject.org/wiki/Anaconda/Updates. >> >> > > Wow! It worked. Many thanks for all the help. I added the updates.img > file into the tree on the CD and created a new CD. It now runs > postinstall without ejecting the CD. > > Is there a similarly elegant way to add a driver disk image? I am still > struggling embedding an additional RAID driver into the boot CD. I > suppose I need to update the initial initrd file (using mkinitrd) and > then tell Anaconda to load the driver image during stage2. Can the > driver disk image be placed somewhere into the tree similar to the > updates.img? > > Thanks > --martin > > > Yes. Look for the driverdisk option at https://www.redhat.com/docs/manuals/enterprise/RHEL-5-manual/Installation_Guide-en-US/s1-kickstart2-options.html That option is in RHEL4 & 5 that I know of. Sincerely, Jason From msteinmann at vesbridge.com Wed May 9 14:31:07 2007 From: msteinmann at vesbridge.com (Martin Steinmann) Date: Wed, 9 May 2007 10:31:07 -0400 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: <4641D2AB.9040300@rampaginggeek.com> Message-ID: > > > > Is there a similarly elegant way to add a driver disk image? I am still > > struggling embedding an additional RAID driver into the boot CD. I > > suppose I need to update the initial initrd file (using mkinitrd) and > > then tell Anaconda to load the driver image during stage2. Can the > > driver disk image be placed somewhere into the tree similar to the > > updates.img? > > > > Thanks > > --martin > > > > > > > Yes. > > Look for the driverdisk option at > https://www.redhat.com/docs/manuals/enterprise/RHEL-5- > manual/Installation_Guide-en-US/s1-kickstart2-options.html > > That option is in RHEL4 & 5 that I know of. Yes, I looked at this and tried several options. The problem is that I do not have a network during installation (no ftp, http or nfs available) and therefore the driver disk image needs to be in the tree on the CD. Also, the HD is empty when I start. --martin > > Sincerely, > Jason > From clumens at redhat.com Wed May 9 14:31:32 2007 From: clumens at redhat.com (Chris Lumens) Date: Wed, 9 May 2007 10:31:32 -0400 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: References: <20070508192235.GV17381@exeter.boston.redhat.com> Message-ID: <20070509143132.GW17381@exeter.boston.redhat.com> > Wow! It worked. Many thanks for all the help. I added the updates.img > file into the tree on the CD and created a new CD. It now runs > postinstall without ejecting the CD. Good, thanks for confirming my fix works. I'll try to get this into U2. I find it amazing that no one ran into this during testing or the beta cycles, but sometimes things get overlooked. - Chris From andymeany at yahoo.com Thu May 10 15:08:31 2007 From: andymeany at yahoo.com (Andy Meany) Date: Thu, 10 May 2007 08:08:31 -0700 (PDT) Subject: FC6 - How to update packages on kickstart server Message-ID: <582935.2565.qm@web63305.mail.re1.yahoo.com> Hello - I am new to kickstart. I am setting up a kickstart server for Fedora Core 6. I have all the RPMs from the original ISO images on my server. I'd like to update the packages on the server so users don't always need to update packages immediately after install. What is the best way to update the RPMs on my server? Regards, Andy --------------------------------- Now that's room service! Choose from over 150,000 hotels in 45,000 destinations on Yahoo! Travel to find your fit. -------------- next part -------------- An HTML attachment was scrubbed... URL: From kickstart at hastek.net Thu May 10 15:21:59 2007 From: kickstart at hastek.net (kickstart) Date: Thu, 10 May 2007 11:21:59 -0400 Subject: FC6 - How to update packages on kickstart server In-Reply-To: <582935.2565.qm@web63305.mail.re1.yahoo.com> References: <582935.2565.qm@web63305.mail.re1.yahoo.com> Message-ID: <20070510152158.GP45517@florian> Andy Meany: > server. I'd like to update the packages on the server so users don't > always need to update packages immediately after install. While there are several tools you can use for this - this OReilly book ($10.) covers the topic in very good detail - using Novi to maintain a "pre-patched" tree of your own. http://www.oreilly.com/catalog/9780596513825/index.html From Chip.Shabazian at bankofamerica.com Thu May 10 15:29:08 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Thu, 10 May 2007 08:29:08 -0700 Subject: FC6 - How to update packages on kickstart server In-Reply-To: <582935.2565.qm@web63305.mail.re1.yahoo.com> Message-ID: Look at FedoraUnity.Org Chip Shabazian VP ; Sr. Consultant - System Engineering Bank of America E&O OS Services -- Linux e: chip.shabazian at bankofamerica.com p: 925 692-7000 ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Andy Meany Sent: Thursday, May 10, 2007 8:09 AM To: kickstart-list at redhat.com Subject: FC6 - How to update packages on kickstart server Hello - I am new to kickstart. I am setting up a kickstart server for Fedora Core 6. I have all the RPMs from the original ISO images on my server. I'd like to update the packages on the server so users don't always need to update packages immediately after install. What is the best way to update the RPMs on my server? Regards, Andy ________________________________ Now that's room service! Choose from over 150,000 hotels in 45,000 destinations on Yahoo! Travel to find your fit. -------------- next part -------------- An HTML attachment was scrubbed... URL: From andymeany at yahoo.com Thu May 10 15:47:47 2007 From: andymeany at yahoo.com (Andy Meany) Date: Thu, 10 May 2007 08:47:47 -0700 (PDT) Subject: FC6 - How to update packages on kickstart server In-Reply-To: Message-ID: <787731.9279.qm@web63307.mail.re1.yahoo.com> Sorry, but I don't find anything dealing with my problem on FedoraUnity.org. Can you be more specific? "Shabazian, Chip" wrote: Look at FedoraUnity.Org Chip Shabazian VP ; Sr. Consultant - System Engineering Bank of America E&O OS Services -- Linux e: chip.shabazian at bankofamerica.com p: 925 692-7000 --------------------------------- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Andy Meany Sent: Thursday, May 10, 2007 8:09 AM To: kickstart-list at redhat.com Subject: FC6 - How to update packages on kickstart server Hello - I am new to kickstart. I am setting up a kickstart server for Fedora Core 6. I have all the RPMs from the original ISO images on my server. I'd like to update the packages on the server so users don't always need to update packages immediately after install. What is the best way to update the RPMs on my server? Regards, Andy --------------------------------- Now that's room service! Choose from over 150,000 hotels in 45,000 destinations on Yahoo! Travel to find your fit. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list --------------------------------- Ahhh...imagining that irresistible "new car" smell? Check outnew cars at Yahoo! Autos. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Thu May 10 15:56:13 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Thu, 10 May 2007 08:56:13 -0700 Subject: FC6 - How to update packages on kickstart server In-Reply-To: <787731.9279.qm@web63307.mail.re1.yahoo.com> Message-ID: http://fedoraunity.org/re-spins Chip Shabazian VP ; Sr. Consultant - System Engineering Bank of America E&O OS Services -- Linux e: chip.shabazian at bankofamerica.com p: 925 692-7000 ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Andy Meany Sent: Thursday, May 10, 2007 8:48 AM To: Discussion list about Kickstart Subject: RE: FC6 - How to update packages on kickstart server Sorry, but I don't find anything dealing with my problem on FedoraUnity.org. Can you be more specific? "Shabazian, Chip" wrote: Look at FedoraUnity.Org Chip Shabazian VP ; Sr. Consultant - System Engineering Bank of America E&O OS Services -- Linux e: chip.shabazian at bankofamerica.com p: 925 692-7000 ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Andy Meany Sent: Thursday, May 10, 2007 8:09 AM To: kickstart-list at redhat.com Subject: FC6 - How to update packages on kickstart server Hello - I am new to kickstart. I am setting up a kickstart server for Fedora Core 6. I have all the RPMs from the original ISO images on my server. I'd like to update the packages on the server so users don't always need to update packages immediately after install. What is the best way to update the RPMs on my server? Regards, Andy ________________________________ Now that's room service! Choose from over 150,000 hotels in 45,000 destinations on Yahoo! Travel to find your fit. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list ________________________________ Ahhh...imagining that irresistible "new car" smell? Check out new cars at Yahoo! Autos. -------------- next part -------------- An HTML attachment was scrubbed... URL: From andymeany at yahoo.com Thu May 10 17:41:45 2007 From: andymeany at yahoo.com (Andy Meany) Date: Thu, 10 May 2007 10:41:45 -0700 (PDT) Subject: FC6 - How to update packages on kickstart server In-Reply-To: Message-ID: <299736.34449.qm@web63314.mail.re1.yahoo.com> >From what I read, it seems like re-spins only work for ISO images. My server actually has all the RPMs rather than the ISO images. Can re-spins work for the RPMs also? I have installed jigdo and downloaded the latest re-spin but am unsure of how to proceed. "Shabazian, Chip" wrote: http://fedoraunity.org/re-spins Chip Shabazian VP ; Sr. Consultant - System Engineering Bank of America E&O OS Services -- Linux e: chip.shabazian at bankofamerica.com p: 925 692-7000 --------------------------------- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Andy Meany Sent: Thursday, May 10, 2007 8:48 AM To: Discussion list about Kickstart Subject: RE: FC6 - How to update packages on kickstart server Sorry, but I don't find anything dealing with my problem on FedoraUnity.org. Can you be more specific? "Shabazian, Chip" wrote: Look at FedoraUnity.Org Chip Shabazian VP ; Sr. Consultant - System Engineering Bank of America E&O OS Services -- Linux e: chip.shabazian at bankofamerica.com p: 925 692-7000 --------------------------------- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Andy Meany Sent: Thursday, May 10, 2007 8:09 AM To: kickstart-list at redhat.com Subject: FC6 - How to update packages on kickstart server Hello - I am new to kickstart. I am setting up a kickstart server for Fedora Core 6. I have all the RPMs from the original ISO images on my server. I'd like to update the packages on the server so users don't always need to update packages immediately after install. What is the best way to update the RPMs on my server? Regards, Andy --------------------------------- Now that's room service! Choose from over 150,000 hotels in 45,000 destinations on Yahoo! Travel to find your fit. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list --------------------------------- Ahhh...imagining that irresistible "new car" smell? Check out new cars at Yahoo! Autos. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list --------------------------------- Ahhh...imagining that irresistible "new car" smell? Check outnew cars at Yahoo! Autos. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Thu May 10 17:57:01 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Thu, 10 May 2007 10:57:01 -0700 Subject: FC6 - How to update packages on kickstart server In-Reply-To: <299736.34449.qm@web63314.mail.re1.yahoo.com> Message-ID: I haven't played with the respins (just saw them yesterday here at the Red Hat summit), but they should work just like the primary distribution where you simply replace the install tree with the one on the respin. BTW, there was a very interesting new product presented in the same session called revisor that will help you spin your own. You can search for it on the unity site, and it will soon work with RHEL. Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Andy Meany Sent: Thursday, May 10, 2007 10:42 AM To: Discussion list about Kickstart Subject: RE: FC6 - How to update packages on kickstart server >From what I read, it seems like re-spins only work for ISO images. My server actually has all the RPMs rather than the ISO images. Can re-spins work for the RPMs also? I have installed jigdo and downloaded the latest re-spin but am unsure of how to proceed. "Shabazian, Chip" wrote: http://fedoraunity.org/re-spins Chip Shabazian VP ; Sr. Consultant - System Engineering Bank of America E&O OS Services -- Linux e: chip.shabazian at bankofamerica.com p: 925 692-7000 ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Andy Meany Sent: Thursday, May 10, 2007 8:48 AM To: Discussion list about Kickstart Subject: RE: FC6 - How to update packages on kickstart server Sorry, but I don't find anything dealing with my problem on FedoraUnity.org. Can you be more specific? "Shabazian, Chip" wrote: Look at FedoraUnity.Org Chip Shabazian VP ; Sr. Consultant - System Engineering Bank of America E&O OS Services -- Linux e: chip.shabazian at bankofamerica.com p: 925 692-7000 ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Andy Meany Sent: Thursday, May 10, 2007 8:09 AM To: kickstart-list at redhat.com Subject: FC6 - How to update packages on kickstart server Hello - I am new to kickstart. I am setting up a kickstart server for Fedora Core 6. I have all the RPMs from the original ISO images on my server. I'd like to update the packages on the server so users don't always need to update packages immediately after install. What is the best way to update the RPMs on my server? Regards, Andy ________________________________ Now that's room service! Choose from over 150,000 hotels in 45,000 destinations on Yahoo! Travel to find your fit. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list ________________________________ Ahhh...imagining that irresistible "new car" smell? Check out new cars at Yahoo! Autos. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list ________________________________ Ahhh...imagining that irresistible "new car" smell? Check out new cars at Yahoo! Autos. -------------- next part -------------- An HTML attachment was scrubbed... URL: From mjenning at rackspace.com Thu May 10 18:38:20 2007 From: mjenning at rackspace.com (Mike Jennings) Date: Thu, 10 May 2007 13:38:20 -0500 Subject: FC6 - How to update packages on kickstart server In-Reply-To: <582935.2565.qm@web63305.mail.re1.yahoo.com> References: <582935.2565.qm@web63305.mail.re1.yahoo.com> Message-ID: <672D9620-7214-4B20-91BA-BB13DF832ADE@rackspace.com> On May 10, 2007, at 10:08 AM, Andy Meany wrote: > Hello - > > I am new to kickstart. I am setting up a kickstart server for > Fedora Core 6. I have all the RPMs from the original ISO images on > my server. I'd like to update the packages on the server so users > don't always need to update packages immediately after install. > What is the best way to update the RPMs on my server? You want mrepo, http://dag.wieers.com/home-made/mrepo/ It'll grab the latest updates for you. Mike > > Regards, > Andy > > Now that's room service! Choose from over 150,000 hotels > in 45,000 destinations on Yahoo! Travel to find your fit. > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list Confidentiality Notice: This e-mail message (including any attached or embedded documents) is intended for the exclusive and confidential use of the individual or entity to which this message is addressed, and unless otherwise expressly indicated, is confidential and privileged information of Rackspace Managed Hosting. Any dissemination, distribution or copying of the enclosed material is prohibited. If you receive this transmission in error, please notify us immediately by e-mail at abuse at rackspace.com, and delete the original message. Your cooperation is appreciated. -------------- next part -------------- An HTML attachment was scrubbed... URL: From andymeany at yahoo.com Thu May 10 18:43:42 2007 From: andymeany at yahoo.com (Andy Meany) Date: Thu, 10 May 2007 11:43:42 -0700 (PDT) Subject: FC6 - How to update packages on kickstart server In-Reply-To: Message-ID: <707081.72922.qm@web63311.mail.re1.yahoo.com> Shouldn't it also be possible to do this using a repo? I have all my original RPMs in a directory: fc6/Fedora/RPMS I downloaded the updated RPMs and put them in fc6/updates/Fedora/RPMS. I then ran the createrepo command like this: createrepo /fc6/updates Can I then use the --repo option in my ks.cfg? If so, what would it look like? "Shabazian, Chip" wrote: I haven't played with the respins (just saw them yesterday here at the Red Hat summit), but they should work just like the primary distribution where you simply replace the install tree with the one on the respin. BTW, there was a very interesting new product presented in the same session called revisor that will help you spin your own. You can search for it on the unity site, and it will soon work with RHEL. Chip --------------------------------- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Andy Meany Sent: Thursday, May 10, 2007 10:42 AM To: Discussion list about Kickstart Subject: RE: FC6 - How to update packages on kickstart server >From what I read, it seems like re-spins only work for ISO images. My server actually has all the RPMs rather than the ISO images. Can re-spins work for the RPMs also? I have installed jigdo and downloaded the latest re-spin but am unsure of how to proceed. "Shabazian, Chip" wrote: http://fedoraunity.org/re-spins Chip Shabazian VP ; Sr. Consultant - System Engineering Bank of America E&O OS Services -- Linux e: chip.shabazian at bankofamerica.com p: 925 692-7000 --------------------------------- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Andy Meany Sent: Thursday, May 10, 2007 8:48 AM To: Discussion list about Kickstart Subject: RE: FC6 - How to update packages on kickstart server Sorry, but I don't find anything dealing with my problem on FedoraUnity.org. Can you be more specific? "Shabazian, Chip" wrote: Look at FedoraUnity.Org Chip Shabazian VP ; Sr. Consultant - System Engineering Bank of America E&O OS Services -- Linux e: chip.shabazian at bankofamerica.com p: 925 692-7000 --------------------------------- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Andy Meany Sent: Thursday, May 10, 2007 8:09 AM To: kickstart-list at redhat.com Subject: FC6 - How to update packages on kickstart server Hello - I am new to kickstart. I am setting up a kickstart server for Fedora Core 6. I have all the RPMs from the original ISO images on my server. I'd like to update the packages on the server so users don't always need to update packages immediately after install. What is the best way to update the RPMs on my server? Regards, Andy --------------------------------- Now that's room service! Choose from over 150,000 hotels in 45,000 destinations on Yahoo! Travel to find your fit. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list --------------------------------- Ahhh...imagining that irresistible "new car" smell? Check out new cars at Yahoo! Autos. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list --------------------------------- Ahhh...imagining that irresistible "new car" smell? Check out new cars at Yahoo! Autos. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list --------------------------------- Food fight? Enjoy some healthy debate in the Yahoo! Answers Food & Drink Q&A. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ahpook at gmail.com Thu May 10 19:02:54 2007 From: ahpook at gmail.com (Eric Sorenson) Date: Thu, 10 May 2007 12:02:54 -0700 Subject: FC6 - How to update packages on kickstart server In-Reply-To: <672D9620-7214-4B20-91BA-BB13DF832ADE@rackspace.com> References: <582935.2565.qm@web63305.mail.re1.yahoo.com> <672D9620-7214-4B20-91BA-BB13DF832ADE@rackspace.com> Message-ID: <80145c330705101202p2899187ep3495a8b974dc5f81@mail.gmail.com> Actually mrepo (wonderful though it is) does not actually do what the original poster wants. He wants to take the RPMS from the cds and overlay downloaded updates to build a new install tree. Ethan McCallam's 'novi' does this: http://www.exmachinatech.net/01/novi/ On 5/10/07, Mike Jennings wrote: > > > On May 10, 2007, at 10:08 AM, Andy Meany wrote: > Hello - > > I am new to kickstart. I am setting up a kickstart server for Fedora Core > 6. I have all the RPMs from the original ISO images on my server. I'd like > to update the packages on the server so users don't always need to update > packages immediately after install. What is the best way to update the RPMs > on my server? > > > You want mrepo, http://dag.wieers.com/home-made/mrepo/ > > It'll grab the latest updates for you. > > Mike From kickstart at hastek.net Thu May 10 19:17:30 2007 From: kickstart at hastek.net (kickstart) Date: Thu, 10 May 2007 15:17:30 -0400 Subject: FC6 - How to update packages on kickstart server In-Reply-To: <80145c330705101202p2899187ep3495a8b974dc5f81@mail.gmail.com> References: <582935.2565.qm@web63305.mail.re1.yahoo.com> <672D9620-7214-4B20-91BA-BB13DF832ADE@rackspace.com> <80145c330705101202p2899187ep3495a8b974dc5f81@mail.gmail.com> Message-ID: <20070510191730.GS45517@florian> Eric Sorenson: > Actually mrepo (wonderful though it is) does not actually do what the > original poster wants. > He wants to take the RPMS from the cds and overlay downloaded updates > to build a new install tree. Ethan McCallam's 'novi' does this: > http://www.exmachinatech.net/01/novi/ Precisely. Which is why I mentioned novi earlier today:) What I am doing (an extension of the suggestion in the OReilly book by Ethan McCallum which I also mentioned) is to 1) fetch updates from multiple external repos into subdirectories of a new ("pre-patch") tree of your base OS, the then 2) run novi on these leaving the new prepatched tree with the latest RPMs. You'd then point new kickstarts at this prepatched tree. Of course, some testing would be in order every time you do a major upgrade to make sure nothing's suddenly broken. In my case I'm using CentOS44 so I also have genhdlist running in my automation scripts.. It has worked quite well so far.. -Harold From kanarip at kanarip.com Thu May 10 19:21:50 2007 From: kanarip at kanarip.com (Jeroen van Meeuwen) Date: Thu, 10 May 2007 21:21:50 +0200 Subject: FC6 - How to update packages on kickstart server In-Reply-To: <299736.34449.qm@web63314.mail.re1.yahoo.com> References: <299736.34449.qm@web63314.mail.re1.yahoo.com> Message-ID: <464370CE.90108@kanarip.com> You would share the mounted ISO over any protocol (HTTP/FTP/...) However this would only give you the core repository. The tool Shabazian is talking about (yesterday at the RH Summit), is Revisor. This tool will allow you to specify your own set or repositories, create custom package sets (including specific versions of packages), and create CD, DVD and Live Media. As a side effect, it'll also have a build tree or unified ISO which you can then use. The Revisor developers (Jonathan Steffan and myself) have submitted Revisor for review in the Fedora Repository, and it'll only work for F7+, but nonetheless should be able to create valid buildtrees for FC6+. A kickstart file (which it appears you already have), can be used to specify the package manifest, so you don't need to resolve dependencies over all 7000+ packages ;-) Kind regards, Jeroen van Meeuwen Andy Meany wrote: > >From what I read, it seems like re-spins only work for ISO images. My > server actually has all the RPMs rather than the ISO images. Can > re-spins work for the RPMs also? I have installed jigdo and downloaded > the latest re-spin but am unsure of how to proceed. > > */"Shabazian, Chip" /* wrote: > > http://fedoraunity.org/re-spins > > Chip Shabazian > VP ; Sr. Consultant - System Engineering > Bank of America > E&O OS Services -- Linux > e: chip.shabazian at bankofamerica.com > > p: 925 692-7000 > > > ------------------------------------------------------------------------ > *From:* kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] *On Behalf Of *Andy Meany > *Sent:* Thursday, May 10, 2007 8:48 AM > *To:* Discussion list about Kickstart > *Subject:* RE: FC6 - How to update packages on kickstart server > > Sorry, but I don't find anything dealing with my problem on > FedoraUnity.org. Can you be more specific? > > */"Shabazian, Chip" /* wrote: > > Look at FedoraUnity.Org > > Chip Shabazian > VP ; Sr. Consultant - System Engineering > Bank of America > E&O OS Services -- Linux > e: chip.shabazian at bankofamerica.com > > p: 925 692-7000 > > > ------------------------------------------------------------------------ > *From:* kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] *On Behalf Of *Andy Meany > *Sent:* Thursday, May 10, 2007 8:09 AM > *To:* kickstart-list at redhat.com > *Subject:* FC6 - How to update packages on kickstart server > > Hello - > > I am new to kickstart. I am setting up a kickstart server for > Fedora Core 6. I have all the RPMs from the original ISO images > on my server. I'd like to update the packages on the server so > users don't always need to update packages immediately after > install. What is the best way to update the RPMs on my server? > > Regards, > Andy > ------------------------------------------------------------------------ > Now that's room service! Choose from over 150,000 hotels > in 45,000 destinations on Yahoo! Travel > > to find your fit. _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > > ------------------------------------------------------------------------ > Ahhh...imagining that irresistible "new car" smell? > Check out new cars at Yahoo! Autos. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > > ------------------------------------------------------------------------ > Ahhh...imagining that irresistible "new car" smell? > Check out new cars at Yahoo! Autos. > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From kanarip at kanarip.com Thu May 10 19:26:32 2007 From: kanarip at kanarip.com (Jeroen van Meeuwen) Date: Thu, 10 May 2007 21:26:32 +0200 Subject: FC6 - How to update packages on kickstart server In-Reply-To: <80145c330705101202p2899187ep3495a8b974dc5f81@mail.gmail.com> References: <582935.2565.qm@web63305.mail.re1.yahoo.com> <672D9620-7214-4B20-91BA-BB13DF832ADE@rackspace.com> <80145c330705101202p2899187ep3495a8b974dc5f81@mail.gmail.com> Message-ID: <464371E8.1070406@kanarip.com> Eric Sorenson wrote: > Actually mrepo (wonderful though it is) does not actually do what the > original poster wants. > He wants to take the RPMS from the cds and overlay downloaded updates > to build a new install tree. Ethan McCallam's 'novi' does this: > http://www.exmachinatech.net/01/novi/ > However the kickstart code: repo --name=updates --url=http://localrepo/updates/ %packages @whatever-group whatever-package would just do the same assuming you sync you're core and updates repositories. Or am I missing the point here? Kind regards, Jeroen van Meeuwen From ahpook at gmail.com Thu May 10 19:31:13 2007 From: ahpook at gmail.com (Eric Sorenson) Date: Thu, 10 May 2007 12:31:13 -0700 Subject: FC6 - How to update packages on kickstart server In-Reply-To: <20070510191730.GS45517@florian> References: <582935.2565.qm@web63305.mail.re1.yahoo.com> <672D9620-7214-4B20-91BA-BB13DF832ADE@rackspace.com> <80145c330705101202p2899187ep3495a8b974dc5f81@mail.gmail.com> <20070510191730.GS45517@florian> Message-ID: <80145c330705101231m28090c59t353b88d8dc8ae6fa@mail.gmail.com> On 5/10/07, kickstart wrote: > Eric Sorenson: > > Actually mrepo (wonderful though it is) does not actually do what the > > original poster wants. > > He wants to take the RPMS from the cds and overlay downloaded updates > > to build a new install tree. Ethan McCallam's 'novi' does this: > > http://www.exmachinatech.net/01/novi/ > > Precisely. Which is why I mentioned novi earlier today:) Yeah, whups, that's what I get for jumping to the end of the thread. Consider your recommendation seconded ;) > > What I am doing (an extension of the suggestion in the OReilly book > by Ethan McCallum which I also mentioned) is to 1) fetch updates > from multiple external repos into subdirectories of a new > ("pre-patch") tree of your base OS, the then 2) run novi on these > leaving the new prepatched tree with the latest RPMs. You'd then > point new kickstarts at this prepatched tree. > > Of course, some testing would be in order every time you do a major > upgrade to make sure nothing's suddenly broken. > > In my case I'm using CentOS44 so I also have genhdlist running in my > automation scripts.. Have you had to modify comps.xml as well? > > It has worked quite well so far.. > > -Harold > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From ebrown at lanl.gov Thu May 10 22:54:07 2007 From: ebrown at lanl.gov (Ed Brown) Date: Thu, 10 May 2007 16:54:07 -0600 Subject: FC6 - How to update packages on kickstart server In-Reply-To: <464371E8.1070406@kanarip.com> References: <582935.2565.qm@web63305.mail.re1.yahoo.com> <672D9620-7214-4B20-91BA-BB13DF832ADE@rackspace.com> <80145c330705101202p2899187ep3495a8b974dc5f81@mail.gmail.com> <464371E8.1070406@kanarip.com> Message-ID: <4643A28F.10305@lanl.gov> Jeroen van Meeuwen wrote: > However the kickstart code: > > repo --name=updates --url=http://localrepo/updates/ > %packages > @whatever-group > whatever-package > > would just do the same assuming you sync you're core and updates > repositories. > > Or am I missing the point here? No, I think this WAS the point of Andy's original, and most recent, posts. He doesn't necessarily want to rebuild the tree, he just wants to avoid having to install, then update. This is by far the easiest solution. -Ed From yhhoo at tpl.com.sg Fri May 11 05:57:43 2007 From: yhhoo at tpl.com.sg (yhhoo at tpl.com.sg) Date: Fri, 11 May 2007 13:57:43 +0800 Subject: FC6 - How to update packages on kickstart server Message-ID: ----- Original Message ----- From: "Eric Sorenson" [ahpook at gmail.com] Sent: 05/11/2007 03:31 AM To: "Discussion list about Kickstart" Subject: Re: FC6 - How to update packages on kickstart server On 5/10/07, kickstart wrote: > Eric Sorenson: > > Actually mrepo (wonderful though it is) does not actually do what the > > original poster wants. > > He wants to take the RPMS from the cds and overlay downloaded updates > > to build a new install tree. Ethan McCallam's 'novi' does this: > > http://www.exmachinatech.net/01/novi/ > > Precisely. Which is why I mentioned novi earlier today:) Yeah, whups, that's what I get for jumping to the end of the thread. Consider your recommendation seconded ;) > > What I am doing (an extension of the suggestion in the OReilly book > by Ethan McCallum which I also mentioned) is to 1) fetch updates > from multiple external repos into subdirectories of a new > ("pre-patch") tree of your base OS, the then 2) run novi on these > leaving the new prepatched tree with the latest RPMs. You'd then > point new kickstarts at this prepatched tree. > > Of course, some testing would be in order every time you do a major > upgrade to make sure nothing's suddenly broken. > > In my case I'm using CentOS44 so I also have genhdlist running in my > automation scripts.. Have you had to modify comps.xml as well? > > It has worked quite well so far.. > > -Harold > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From yhhoo at tpl.com.sg Fri May 11 06:59:46 2007 From: yhhoo at tpl.com.sg (yhhoo at tpl.com.sg) Date: Fri, 11 May 2007 14:59:46 +0800 Subject: FC6 - How to update packages on kickstart server Message-ID: ----- Original Message ----- From: "Eric Sorenson" [ahpook at gmail.com] Sent: 05/11/2007 03:31 AM To: "Discussion list about Kickstart" Subject: Re: FC6 - How to update packages on kickstart server On 5/10/07, kickstart wrote: > Eric Sorenson: > > Actually mrepo (wonderful though it is) does not actually do what the > > original poster wants. > > He wants to take the RPMS from the cds and overlay downloaded updates > > to build a new install tree. Ethan McCallam's 'novi' does this: > > http://www.exmachinatech.net/01/novi/ > > Precisely. Which is why I mentioned novi earlier today:) Yeah, whups, that's what I get for jumping to the end of the thread. Consider your recommendation seconded ;) > > What I am doing (an extension of the suggestion in the OReilly book > by Ethan McCallum which I also mentioned) is to 1) fetch updates > from multiple external repos into subdirectories of a new > ("pre-patch") tree of your base OS, the then 2) run novi on these > leaving the new prepatched tree with the latest RPMs. You'd then > point new kickstarts at this prepatched tree. > > Of course, some testing would be in order every time you do a major > upgrade to make sure nothing's suddenly broken. > > In my case I'm using CentOS44 so I also have genhdlist running in my > automation scripts.. Have you had to modify comps.xml as well? > > It has worked quite well so far.. > > -Harold > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From SSTinsley at upsfreight.com Mon May 14 13:19:55 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Mon, 14 May 2007 09:19:55 -0400 Subject: Kickstart via PXE/NFS Message-ID: I am attempting to install RHEL ES V4 U5 on a Dell 2950 server. I am using a Sun server as the DHCP/TFTP/NFS server supplying all data to complete the install. I am able to start the PXE boot. The Dell server is able to download the pxelinux image and also read the pxelinux.cfg file. Here is the cfg file I am using. # Perform a local boot by default default installauto # Always prompt prompt 1 # Display the bootup message display pxeboot.msg # Boot automatically after 30 seconds in tenths of a second timeout 300 label local localboot 0 label rescue ipappend 2 kernel rhel_v4/vmlinuz append initrd=rhel_v4/initrd.img rescue label installauto ipappend 2 kernel rhel_v4/vmlinuz append initrd=rhel_v4/initrd.img ks=nfs:10.1.1.253:/export/redhat/ks_rhel_v4.cfg ksdevice=bootif The kernel image loads from the server via tftp. The initial blue startup screen appears at the start of installation. That is where things start to go wrong. First, even though I have designated KS to use the boot interfaces in the cfg file, I still get an interface selection screen at the start of the install. Why is the selection necessary? Why is the iterface not being set from the hsdevice option in the CFG file? Second, after selecting the interface, I switch to virt console 2 and see the message that the install is attemting to get the ks_rhel_v4.cfg file and the nfs URL is displayed. At the time this is happening I have a network snoop at the Sun server. There is no NFS traffic showing up from the Dell. I am snooping based on the Dell's MAC address. The output on the virt. console indicates that the cfg file cannot be loaded. The network interface is operable on the Dell. I can ping the configured address from the Sun DHCP server. I can also boot the Dell in rescue more and NFS mount the directory that contains the cfg file and also the directory that contains the operating system image. Any ideas why this installation will not complete? NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From CallahanT at tessco.com Mon May 14 13:53:40 2007 From: CallahanT at tessco.com (Callahan, Tom) Date: Mon, 14 May 2007 09:53:40 -0400 Subject: Kickstart via PXE/NFS Message-ID: <6982E319B698F143AEDB8F3D8398045702887FAC@EXCHANGE-1.tessco.com> Adding ksdevice=xxx to the append line for the boot should get rid of the interface options. Kickstart doesn't know the config until after it brings the interface up. As far as nfs not working, check to make sure the host is in dns, as we all know the rules of nfs and hostnames. You should still see traffic but its worth a try. Also, add ip=xxx.xxx.xxx.xxx , dns=xxx.xxx.xxx.xxx and gateway=xxx.xxx.xxx.xxx to the kernel append line to force dns,gateway,etc. I find that kickstart works much better that way. ================================ Sent from my wireless handheld device. Tom Callahan -----Original Message----- From: kickstart-list-bounces at redhat.com To: kickstart-list at redhat.com Sent: Mon May 14 09:19:55 2007 Subject: Kickstart via PXE/NFS I am attempting to install RHEL ES V4 U5 on a Dell 2950 server. I am using a Sun server as the DHCP/TFTP/NFS server supplying all data to complete the install. I am able to start the PXE boot. The Dell server is able to download the pxelinux image and also read the pxelinux.cfg file. Here is the cfg file I am using. # Perform a local boot by default default installauto # Always prompt prompt 1 # Display the bootup message display pxeboot.msg # Boot automatically after 30 seconds in tenths of a second timeout 300 label local localboot 0 label rescue ipappend 2 kernel rhel_v4/vmlinuz append initrd=rhel_v4/initrd.img rescue label installauto ipappend 2 kernel rhel_v4/vmlinuz append initrd=rhel_v4/initrd.img ks=nfs:10.1.1.253:/export/redhat/ks_rhel_v4.cfg ksdevice=bootif The kernel image loads from the server via tftp. The initial blue startup screen appears at the start of installation. That is where things start to go wrong. First, even though I have designated KS to use the boot interfaces in the cfg file, I still get an interface selection screen at the start of the install. Why is the selection necessary? Why is the iterface not being set from the hsdevice option in the CFG file? Second, after selecting the interface, I switch to virt console 2 and see the message that the install is attemting to get the ks_rhel_v4.cfg file and the nfs URL is displayed. At the time this is happening I have a network snoop at the Sun server. There is no NFS traffic showing up from the Dell. I am snooping based on the Dell's MAC address. The output on the virt. console indicates that the cfg file cannot be loaded. The network interface is operable on the Dell. I can ping the configured address from the Sun DHCP server. I can also boot the Dell in rescue more and NFS mount the directory that contains the cfg file and also the directory that contains the operating system image. Any ideas why this installation will not complete? NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From SSTinsley at upsfreight.com Mon May 14 14:29:46 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Mon, 14 May 2007 10:29:46 -0400 Subject: Kickstart via PXE/NFS In-Reply-To: <6982E319B698F143AEDB8F3D8398045702887FAC@EXCHANGE-1.tessco.com> Message-ID: kickstart-list-bounces at redhat.com wrote on 05/14/2007 09:53:40 AM: > Adding ksdevice=xxx to the append line for the boot should get rid > of the interface options. Kickstart doesn't know the config until > after it brings the interface up. > I am specifying the ksdevice in the pxelinux.cfg file. I also set the IPAPPEND 2 option which I understood would pass in the IP address of the boot server for use as the IP during the install. > As far as nfs not working, check to make sure the host is in dns, as > we all know the rules of nfs and hostnames. You should still see > traffic but its worth a try. For the NFS, I was giving the server's IP address so name resolution would not be required. Does DNS have to be configured? > > Also, add ip=xxx.xxx.xxx.xxx , dns=xxx.xxx.xxx.xxx and gateway=xxx. > xxx.xxx.xxx to the kernel append line to force dns,gateway,etc. I > find that kickstart works much better that way. > > > > ================================ > Sent from my wireless handheld device. > > Tom Callahan > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > To: kickstart-list at redhat.com > Sent: Mon May 14 09:19:55 2007 > Subject: Kickstart via PXE/NFS > > > I am attempting to install RHEL ES V4 U5 on a Dell 2950 server. I am > using a Sun server as the DHCP/TFTP/NFS server supplying all data to > complete the install. I am able to start the PXE boot. The Dell > server is able to download the pxelinux image and also read the > pxelinux.cfg file. Here is the cfg file I am using. > > # Perform a local boot by default > default installauto > # Always prompt > prompt 1 > # Display the bootup message > display pxeboot.msg > # Boot automatically after 30 seconds in tenths of a second > timeout 300 > label local > localboot 0 > label rescue > ipappend 2 > kernel rhel_v4/vmlinuz > append initrd=rhel_v4/initrd.img rescue > label installauto > ipappend 2 > kernel rhel_v4/vmlinuz > append initrd=rhel_v4/initrd.img ks=nfs:10.1.1.253: > /export/redhat/ks_rhel_v4.cfg ksdevice=bootif > > The kernel image loads from the server via tftp. The initial blue > startup screen appears at the start of installation. That is where > things start to go wrong. > > First, even though I have designated KS to use the boot interfaces > in the cfg file, I still get an interface selection screen at the > start of the install. Why is the selection necessary? Why is the > iterface not being set from the hsdevice option in the CFG file? > > Second, after selecting the interface, I switch to virt console 2 > and see the message that the install is attemting to get the > ks_rhel_v4.cfg file and the nfs URL is displayed. At the time this > is happening I have a network snoop at the Sun server. There is no > NFS traffic showing up from the Dell. I am snooping based on the > Dell's MAC address. The output on the virt. console indicates that > the cfg file cannot be loaded. The network interface is operable on > the Dell. I can ping the configured address from the Sun DHCP > server. I can also boot the Dell in rescue more and NFS mount the > directory that contains the cfg file and also the directory that > contains the operating system image. Any ideas why this installation > will not complete? > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC > INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR > ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO > RECEIVE IT. If you are not the intended recipient of this document, > you are notified that any review, dissemination, distribution or > copying of this communication is prohibited. If you have received > this communication in error, please notify me immediately by return > email, delete the electronic message and destroy any printed copies. > Thank you for your cooperation. > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From CallahanT at tessco.com Mon May 14 14:32:26 2007 From: CallahanT at tessco.com (Callahan, Tom) Date: Mon, 14 May 2007 10:32:26 -0400 Subject: Kickstart via PXE/NFS Message-ID: <6982E319B698F143AEDB8F3D8398045702887FB1@EXCHANGE-1.tessco.com> I think, don't quote me, that nfs is still going to try to reverse resolve the ip to dns. ================================ Sent from my wireless handheld device. Tom Callahan -----Original Message----- From: kickstart-list-bounces at redhat.com To: Discussion list about Kickstart Sent: Mon May 14 10:29:46 2007 Subject: Re: Kickstart via PXE/NFS kickstart-list-bounces at redhat.com wrote on 05/14/2007 09:53:40 AM: > Adding ksdevice=xxx to the append line for the boot should get rid > of the interface options. Kickstart doesn't know the config until > after it brings the interface up. > I am specifying the ksdevice in the pxelinux.cfg file. I also set the IPAPPEND 2 option which I understood would pass in the IP address of the boot server for use as the IP during the install. > As far as nfs not working, check to make sure the host is in dns, as > we all know the rules of nfs and hostnames. You should still see > traffic but its worth a try. For the NFS, I was giving the server's IP address so name resolution would not be required. Does DNS have to be configured? > > Also, add ip=xxx.xxx.xxx.xxx , dns=xxx.xxx.xxx.xxx and gateway=xxx. > xxx.xxx.xxx to the kernel append line to force dns,gateway,etc. I > find that kickstart works much better that way. > > > > ================================ > Sent from my wireless handheld device. > > Tom Callahan > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > To: kickstart-list at redhat.com > Sent: Mon May 14 09:19:55 2007 > Subject: Kickstart via PXE/NFS > > > I am attempting to install RHEL ES V4 U5 on a Dell 2950 server. I am > using a Sun server as the DHCP/TFTP/NFS server supplying all data to > complete the install. I am able to start the PXE boot. The Dell > server is able to download the pxelinux image and also read the > pxelinux.cfg file. Here is the cfg file I am using. > > # Perform a local boot by default > default installauto > # Always prompt > prompt 1 > # Display the bootup message > display pxeboot.msg > # Boot automatically after 30 seconds in tenths of a second > timeout 300 > label local > localboot 0 > label rescue > ipappend 2 > kernel rhel_v4/vmlinuz > append initrd=rhel_v4/initrd.img rescue > label installauto > ipappend 2 > kernel rhel_v4/vmlinuz > append initrd=rhel_v4/initrd.img ks=nfs:10.1.1.253: > /export/redhat/ks_rhel_v4.cfg ksdevice=bootif > > The kernel image loads from the server via tftp. The initial blue > startup screen appears at the start of installation. That is where > things start to go wrong. > > First, even though I have designated KS to use the boot interfaces > in the cfg file, I still get an interface selection screen at the > start of the install. Why is the selection necessary? Why is the > iterface not being set from the hsdevice option in the CFG file? > > Second, after selecting the interface, I switch to virt console 2 > and see the message that the install is attemting to get the > ks_rhel_v4.cfg file and the nfs URL is displayed. At the time this > is happening I have a network snoop at the Sun server. There is no > NFS traffic showing up from the Dell. I am snooping based on the > Dell's MAC address. The output on the virt. console indicates that > the cfg file cannot be loaded. The network interface is operable on > the Dell. I can ping the configured address from the Sun DHCP > server. I can also boot the Dell in rescue more and NFS mount the > directory that contains the cfg file and also the directory that > contains the operating system image. Any ideas why this installation > will not complete? > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC > INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR > ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO > RECEIVE IT. If you are not the intended recipient of this document, > you are notified that any review, dissemination, distribution or > copying of this communication is prohibited. If you have received > this communication in error, please notify me immediately by return > email, delete the electronic message and destroy any printed copies. > Thank you for your cooperation. > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From law at redhat.com Mon May 14 15:23:36 2007 From: law at redhat.com (Jeffrey Law) Date: Mon, 14 May 2007 09:23:36 -0600 Subject: Kickstart via PXE/NFS In-Reply-To: References: Message-ID: <1179156216.4904.42.camel@localhost.localdomain> On Mon, 2007-05-14 at 10:29 -0400, SSTinsley at upsfreight.com wrote: > > > kickstart-list-bounces at redhat.com wrote on 05/14/2007 09:53:40 AM: > > > Adding ksdevice=xxx to the append line for the boot should get rid > > of the interface options. Kickstart doesn't know the config until > > after it brings the interface up. > > > > I am specifying the ksdevice in the pxelinux.cfg file. I also set the > IPAPPEND 2 option which I understood would pass in the IP address of > the boot server for use as the IP during the > install. How long is the line specifying all your options (ks device, ks location, initrd, etc etc? There's a hard limit of 255 characters and if you exceed it the options are silently truncated which obviously leads to unexpected results. Jeff From SSTinsley at upsfreight.com Mon May 14 15:31:25 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Mon, 14 May 2007 11:31:25 -0400 Subject: Kickstart via PXE/NFS In-Reply-To: <1179156216.4904.42.camel@localhost.localdomain> Message-ID: kickstart-list-bounces at redhat.com wrote on 05/14/2007 11:23:36 AM: > On Mon, 2007-05-14 at 10:29 -0400, SSTinsley at upsfreight.com wrote: > > > > > > kickstart-list-bounces at redhat.com wrote on 05/14/2007 09:53:40 AM: > > > > > Adding ksdevice=xxx to the append line for the boot should get rid > > > of the interface options. Kickstart doesn't know the config until > > > after it brings the interface up. > > > > > > > I am specifying the ksdevice in the pxelinux.cfg file. I also set the > > IPAPPEND 2 option which I understood would pass in the IP address of > > the boot server for use as the IP during the > > install. > How long is the line specifying all your options (ks device, ks > location, initrd, etc etc? There's a hard limit of 255 characters > and if you exceed it the options are silently truncated which > obviously leads to unexpected results. It is about 120 characters, give or take, but nowhere near 255. > > Jeff > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From secroft at micron.com Mon May 14 17:05:14 2007 From: secroft at micron.com (secroft at micron.com) Date: Mon, 14 May 2007 11:05:14 -0600 Subject: Kickstart via PXE/NFS In-Reply-To: Message-ID: <3BA8753665DB594C8C35CFA8801AD522BEFB21@ntxboimbx32.micron.com> Are both NIC's turned on? You May want to turn one off in the bios to bypass the enumeration issue. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com Sent: Monday, May 14, 2007 9:31 AM To: law at redhat.com; Discussion list about Kickstart Subject: Re: Kickstart via PXE/NFS kickstart-list-bounces at redhat.com wrote on 05/14/2007 11:23:36 AM: > On Mon, 2007-05-14 at 10:29 -0400, SSTinsley at upsfreight.com wrote: > > > > > > kickstart-list-bounces at redhat.com wrote on 05/14/2007 09:53:40 AM: > > > > > Adding ksdevice=xxx to the append line for the boot should get rid > > > of the interface options. Kickstart doesn't know the config until > > > after it brings the interface up. > > > > > > > I am specifying the ksdevice in the pxelinux.cfg file. I also set the > > IPAPPEND 2 option which I understood would pass in the IP address of > > the boot server for use as the IP during the > > install. > How long is the line specifying all your options (ks device, ks > location, initrd, etc etc? There's a hard limit of 255 characters > and if you exceed it the options are silently truncated which > obviously leads to unexpected results. It is about 120 characters, give or take, but nowhere near 255. > > Jeff > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From SSTinsley at upsfreight.com Mon May 14 17:08:47 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Mon, 14 May 2007 13:08:47 -0400 Subject: Kickstart via PXE/NFS In-Reply-To: <3BA8753665DB594C8C35CFA8801AD522BEFB21@ntxboimbx32.micron.com> Message-ID: kickstart-list-bounces at redhat.com wrote on 05/14/2007 01:05:14 PM: > Are both NIC's turned on? You May want to turn one off in the bios > to bypass the enumeration issue. > Second interface is not active, no link > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com > Sent: Monday, May 14, 2007 9:31 AM > To: law at redhat.com; Discussion list about Kickstart > Subject: Re: Kickstart via PXE/NFS > > > kickstart-list-bounces at redhat.com wrote on 05/14/2007 11:23:36 AM: > > > On Mon, 2007-05-14 at 10:29 -0400, SSTinsley at upsfreight.com wrote: > > > > > > > > > kickstart-list-bounces at redhat.com wrote on 05/14/2007 09:53:40 AM: > > > > > > > Adding ksdevice=xxx to the append line for the boot should get rid > > > > of the interface options. Kickstart doesn't know the config until > > > > after it brings the interface up. > > > > > > > > > > I am specifying the ksdevice in the pxelinux.cfg file. I also set the > > > IPAPPEND 2 option which I understood would pass in the IP address of > > > the boot server for use as the IP during the > > > install. > > How long is the line specifying all your options (ks device, ks > > location, initrd, etc etc? There's a hard limit of 255 characters > > and if you exceed it the options are silently truncated which > > obviously leads to unexpected results. > > It is about 120 characters, give or take, but nowhere near 255. > > > > > > Jeff > > > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC > INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR > ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO > RECEIVE IT. If you are not the intended recipient of this document, > you are notified that any review, dissemination, distribution or > copying of this communication is prohibited. If you have received > this communication in error, please notify me immediately by return > email, delete the electronic message and destroy any printed copies. > Thank you for your cooperation._______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ahpook at gmail.com Mon May 14 17:29:54 2007 From: ahpook at gmail.com (Eric Sorenson) Date: Mon, 14 May 2007 10:29:54 -0700 Subject: Kickstart via PXE/NFS In-Reply-To: References: <3BA8753665DB594C8C35CFA8801AD522BEFB21@ntxboimbx32.micron.com> Message-ID: <80145c330705141029k59c1b4fh6f97d0dda75b3afe@mail.gmail.com> if you page-up in vty3, can you see what your kernel command line is? alternately if it makes it to shell in vty2, 'cat /proc/cmdline' - i wonder if it's actually doing the ipappend stuff. you might need a newer pxelinux.0 if not. also, instead of 'ksdevice=bootif' you can do 'ksdevice=link ip=dhcp' ; the bootif argument is more useful if you have 2 nics enabled/link-up and want to make sure it doesnt pxeboot out one interface and then ks out the other. -=Eric From SSTinsley at upsfreight.com Mon May 14 18:35:36 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Mon, 14 May 2007 14:35:36 -0400 Subject: Kickstart via PXE/NFS In-Reply-To: <80145c330705141029k59c1b4fh6f97d0dda75b3afe@mail.gmail.com> Message-ID: I made a change to pxe cfg file to have ks.cfg file read via http. It fails too. However, if I look in the anaconda log I see the message getting ks.cfg but then failed to transfer ks.cfg to fd. The snoop from the KS server side shows the HTTP retrieval of the KS config file. Is it really trying to use an fd device, a floppy? These Dell servers do not include floppys. Redhat sure has taken something that should be relatively easy and made it a pain in the butt. kickstart-list-bounces at redhat.com wrote on 05/14/2007 01:29:54 PM: > if you page-up in vty3, can you see what your kernel command line is? > alternately if it makes it to shell in vty2, 'cat /proc/cmdline' - i > wonder if it's actually doing the ipappend stuff. you might need a > newer pxelinux.0 if not. also, instead of 'ksdevice=bootif' you can do > 'ksdevice=link ip=dhcp' ; the bootif argument is more useful if you > have 2 nics enabled/link-up and want to make sure it doesnt pxeboot > out one interface and then ks out the other. > > -=Eric > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Mon May 14 18:48:04 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Mon, 14 May 2007 11:48:04 -0700 Subject: Kickstart via PXE/NFS In-Reply-To: Message-ID: How many NIC's do you have in the box? Are they all embedded, or are there some on cards? Kickstart is a very powerful and useful tool that we use globally to build over 1000 distributed systems each year from a centralized and managed location, and although it has bugs like any product, most issues like you are having tend to be infrastructure or manufacturer related. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com Sent: Monday, May 14, 2007 11:36 AM To: Discussion list about Kickstart Cc: Discussion list about Kickstart; kickstart-list-bounces at redhat.com Subject: Re: Kickstart via PXE/NFS I made a change to pxe cfg file to have ks.cfg file read via http. It fails too. However, if I look in the anaconda log I see the message getting ks.cfg but then failed to transfer ks.cfg to fd. The snoop from the KS server side shows the HTTP retrieval of the KS config file. Is it really trying to use an fd device, a floppy? These Dell servers do not include floppys. Redhat sure has taken something that should be relatively easy and made it a pain in the butt. kickstart-list-bounces at redhat.com wrote on 05/14/2007 01:29:54 PM: > if you page-up in vty3, can you see what your kernel command line is? > alternately if it makes it to shell in vty2, 'cat /proc/cmdline' - i > wonder if it's actually doing the ipappend stuff. you might need a > newer pxelinux.0 if not. also, instead of 'ksdevice=bootif' you can do > 'ksdevice=link ip=dhcp' ; the bootif argument is more useful if you > have 2 nics enabled/link-up and want to make sure it doesnt pxeboot > out one interface and then ks out the other. > > -=Eric > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From SSTinsley at upsfreight.com Mon May 14 19:38:45 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Mon, 14 May 2007 15:38:45 -0400 Subject: Kickstart via PXE/NFS In-Reply-To: Message-ID: kickstart-list-bounces at redhat.com wrote on 05/14/2007 02:48:04 PM: > How many NIC's do you have in the box? Are they all embedded, or > are there some on cards? 2 NICs. Only one is active an plugged in to a switch. While it may be powerful, it seems to be way more complicated than should be required. If the server can find its initial IP via DHCP to load pxelinux and to load the kernel, why is it unable to complete the last step and locate the kickstart file? I have specified the ksdevice several different ways and none have been successful (ksdevice=eth0, =link, =bootip with ipappend, =eth1) and in all cases anaconda ends up unable to get the IP and prompts me for static network configuration. I enter the needed network information and provide the NFS information to reach the RH install image and the box is able to read the o/s image and start an interactive install. I have also dropped to command line (ctl-alt-f2) and verified that the nfs directory holding the KS file can be mounted. And I have verified it can be retrieved with http using wget. And still anaconda is unable to retrieve the file with either nfs or http. > > Kickstart is a very powerful and useful tool that we use globally to > build over 1000 distributed systems each year from a centralized and > managed location, and although it has bugs like any product, most > issues like you are having tend to be infrastructure or manufacturer related. > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com > Sent: Monday, May 14, 2007 11:36 AM > To: Discussion list about Kickstart > Cc: Discussion list about Kickstart; kickstart-list-bounces at redhat.com > Subject: Re: Kickstart via PXE/NFS > > I made a change to pxe cfg file to have ks.cfg file read via http. > > It fails too. However, if I look in the anaconda log I see the > message getting ks.cfg > but then failed to transfer ks.cfg to fd. The snoop from the KS > server side shows the HTTP > retrieval of the KS config file. Is it really trying to use an fd > device, a floppy? These > Dell servers do not include floppys. > > Redhat sure has taken something that should be relatively easy and > made it a pain in the butt. > > > kickstart-list-bounces at redhat.com wrote on 05/14/2007 01:29:54 PM: > > > if you page-up in vty3, can you see what your kernel command line is? > > alternately if it makes it to shell in vty2, 'cat /proc/cmdline' - i > > wonder if it's actually doing the ipappend stuff. you might need a > > newer pxelinux.0 if not. also, instead of 'ksdevice=bootif' you can do > > 'ksdevice=link ip=dhcp' ; the bootif argument is more useful if you > > have 2 nics enabled/link-up and want to make sure it doesnt pxeboot > > out one interface and then ks out the other. > > > > -=Eric > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC > INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR > ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO > RECEIVE IT. If you are not the intended recipient of this document, > you are notified that any review, dissemination, distribution or > copying of this communication is prohibited. If you have received > this communication in error, please notify me immediately by return > email, delete the electronic message and destroy any printed copies. > Thank you for your cooperation._______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From secroft at micron.com Mon May 14 20:01:42 2007 From: secroft at micron.com (secroft at micron.com) Date: Mon, 14 May 2007 14:01:42 -0600 Subject: Kickstart via PXE/NFS In-Reply-To: Message-ID: <3BA8753665DB594C8C35CFA8801AD522BEFB27@ntxboimbx32.micron.com> For the second interface, try turning it off in the bios. I believe you're running into an enumeration problem with the onboard NIC's. Whether you have it plugged in or not, it is probably where the issue is. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com Sent: Monday, May 14, 2007 1:39 PM To: Discussion list about Kickstart Cc: Discussion list about Kickstart; kickstart-list-bounces at redhat.com Subject: RE: Kickstart via PXE/NFS kickstart-list-bounces at redhat.com wrote on 05/14/2007 02:48:04 PM: > How many NIC's do you have in the box? Are they all embedded, or > are there some on cards? 2 NICs. Only one is active an plugged in to a switch. While it may be powerful, it seems to be way more complicated than should be required. If the server can find its initial IP via DHCP to load pxelinux and to load the kernel, why is it unable to complete the last step and locate the kickstart file? I have specified the ksdevice several different ways and none have been successful (ksdevice=eth0, =link, =bootip with ipappend, =eth1) and in all cases anaconda ends up unable to get the IP and prompts me for static network configuration. I enter the needed network information and provide the NFS information to reach the RH install image and the box is able to read the o/s image and start an interactive install. I have also dropped to command line (ctl-alt-f2) and verified that the nfs directory holding the KS file can be mounted. And I have verified it can be retrieved with http using wget. And still anaconda is unable to retrieve the file with either nfs or http. > > Kickstart is a very powerful and useful tool that we use globally to > build over 1000 distributed systems each year from a centralized and > managed location, and although it has bugs like any product, most > issues like you are having tend to be infrastructure or manufacturer related. > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com > Sent: Monday, May 14, 2007 11:36 AM > To: Discussion list about Kickstart > Cc: Discussion list about Kickstart; kickstart-list-bounces at redhat.com > Subject: Re: Kickstart via PXE/NFS > > I made a change to pxe cfg file to have ks.cfg file read via http. > > It fails too. However, if I look in the anaconda log I see the > message getting ks.cfg > but then failed to transfer ks.cfg to fd. The snoop from the KS > server side shows the HTTP > retrieval of the KS config file. Is it really trying to use an fd > device, a floppy? These > Dell servers do not include floppys. > > Redhat sure has taken something that should be relatively easy and > made it a pain in the butt. > > > kickstart-list-bounces at redhat.com wrote on 05/14/2007 01:29:54 PM: > > > if you page-up in vty3, can you see what your kernel command line is? > > alternately if it makes it to shell in vty2, 'cat /proc/cmdline' - i > > wonder if it's actually doing the ipappend stuff. you might need a > > newer pxelinux.0 if not. also, instead of 'ksdevice=bootif' you can do > > 'ksdevice=link ip=dhcp' ; the bootif argument is more useful if you > > have 2 nics enabled/link-up and want to make sure it doesnt pxeboot > > out one interface and then ks out the other. > > > > -=Eric > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC > INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR > ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO > RECEIVE IT. If you are not the intended recipient of this document, > you are notified that any review, dissemination, distribution or > copying of this communication is prohibited. If you have received > this communication in error, please notify me immediately by return > email, delete the electronic message and destroy any printed copies. > Thank you for your cooperation._______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Elizabeth.Brosch at thomson.com Mon May 14 20:03:21 2007 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Mon, 14 May 2007 16:03:21 -0400 Subject: Kickstart via PXE/NFS In-Reply-To: <3BA8753665DB594C8C35CFA8801AD522BEFB27@ntxboimbx32.micron.com> Message-ID: <2213CAA79DA9094AB52517FDADEF07FA057A942E@tshuspaphimbx01.ERF.THOMSON.COM> I had a similar issue. I had only one NIC cabled to the switch. The initial DHCP request was successful but it would fail on finding the ks.cfg file. I ended up turning off all the NICs in the bios and this helped. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of secroft at micron.com Sent: Monday, May 14, 2007 4:02 PM To: kickstart-list at redhat.com Cc: kickstart-list-bounces at redhat.com Subject: RE: Kickstart via PXE/NFS For the second interface, try turning it off in the bios. I believe you're running into an enumeration problem with the onboard NIC's. Whether you have it plugged in or not, it is probably where the issue is. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com Sent: Monday, May 14, 2007 1:39 PM To: Discussion list about Kickstart Cc: Discussion list about Kickstart; kickstart-list-bounces at redhat.com Subject: RE: Kickstart via PXE/NFS kickstart-list-bounces at redhat.com wrote on 05/14/2007 02:48:04 PM: > How many NIC's do you have in the box? Are they all embedded, or > are there some on cards? 2 NICs. Only one is active an plugged in to a switch. While it may be powerful, it seems to be way more complicated than should be required. If the server can find its initial IP via DHCP to load pxelinux and to load the kernel, why is it unable to complete the last step and locate the kickstart file? I have specified the ksdevice several different ways and none have been successful (ksdevice=eth0, =link, =bootip with ipappend, =eth1) and in all cases anaconda ends up unable to get the IP and prompts me for static network configuration. I enter the needed network information and provide the NFS information to reach the RH install image and the box is able to read the o/s image and start an interactive install. I have also dropped to command line (ctl-alt-f2) and verified that the nfs directory holding the KS file can be mounted. And I have verified it can be retrieved with http using wget. And still anaconda is unable to retrieve the file with either nfs or http. > > Kickstart is a very powerful and useful tool that we use globally to > build over 1000 distributed systems each year from a centralized and > managed location, and although it has bugs like any product, most > issues like you are having tend to be infrastructure or manufacturer related. > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com > Sent: Monday, May 14, 2007 11:36 AM > To: Discussion list about Kickstart > Cc: Discussion list about Kickstart; kickstart-list-bounces at redhat.com > Subject: Re: Kickstart via PXE/NFS > > I made a change to pxe cfg file to have ks.cfg file read via http. > > It fails too. However, if I look in the anaconda log I see the > message getting ks.cfg > but then failed to transfer ks.cfg to fd. The snoop from the KS > server side shows the HTTP > retrieval of the KS config file. Is it really trying to use an fd > device, a floppy? These > Dell servers do not include floppys. > > Redhat sure has taken something that should be relatively easy and > made it a pain in the butt. > > > kickstart-list-bounces at redhat.com wrote on 05/14/2007 01:29:54 PM: > > > if you page-up in vty3, can you see what your kernel command line is? > > alternately if it makes it to shell in vty2, 'cat /proc/cmdline' - i > > wonder if it's actually doing the ipappend stuff. you might need a > > newer pxelinux.0 if not. also, instead of 'ksdevice=bootif' you can do > > 'ksdevice=link ip=dhcp' ; the bootif argument is more useful if you > > have 2 nics enabled/link-up and want to make sure it doesnt pxeboot > > out one interface and then ks out the other. > > > > -=Eric > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC > INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR > ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO > RECEIVE IT. If you are not the intended recipient of this document, > you are notified that any review, dissemination, distribution or > copying of this communication is prohibited. If you have received > this communication in error, please notify me immediately by return > email, delete the electronic message and destroy any printed copies. > Thank you for your cooperation._______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Mon May 14 20:14:11 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Mon, 14 May 2007 13:14:11 -0700 Subject: Kickstart via PXE/NFS In-Reply-To: Message-ID: You still have never answered the question. We are trying to help you here. It doesn't matter whether or not your second NIC is plugged in, it does however matter whether or not it is on the same bus as the NIC you are pxebooting from. As for why it works in pxe and doesn't work in kickstart, they are two entirely different systems. pxe comes from the system BIOS, and the kickstart installer comes from RHEL. The reason this is important is because RHEL 4 + enumerates the bus differently (and according to Red Hat properly) than either pxe or previous RHEL versions. As for "complicated", my time goes back to building "unattend.txt" files for Windows 95 (and every version since), and kickstart is FAR more powerful and easier to use than any of the windows technologies that are included with the OS. And if you don't like kickstart, try using YAST sometime, that will make you old before your time. This list is very helpful and has gotten many people to quality build setups, but disparaging the tool that we are trying to help you deploy isn't going to win friends or help. At the end of the day, if you don't like anaconda, do something about it. It's open source and you can dive in and fix the things you don't like, and submit the patches upstream. There are many things *I* would like to see changed, but I don't have the time to write a patch, so I have to live with it or work around it. At least now I can make the changes if I wanted to... Sure wish I could have done that in Windows 10 years ago when I was trying to get an unattend.txt file to bring up a backup domain controller that wasn't online with the primary since it hadn't been built yet and was located in a different state. Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com Sent: Monday, May 14, 2007 12:39 PM To: Discussion list about Kickstart Cc: Discussion list about Kickstart; kickstart-list-bounces at redhat.com Subject: RE: Kickstart via PXE/NFS kickstart-list-bounces at redhat.com wrote on 05/14/2007 02:48:04 PM: > How many NIC's do you have in the box? Are they all embedded, or > are there some on cards? 2 NICs. Only one is active an plugged in to a switch. While it may be powerful, it seems to be way more complicated than should be required. If the server can find its initial IP via DHCP to load pxelinux and to load the kernel, why is it unable to complete the last step and locate the kickstart file? I have specified the ksdevice several different ways and none have been successful (ksdevice=eth0, =link, =bootip with ipappend, =eth1) and in all cases anaconda ends up unable to get the IP and prompts me for static network configuration. I enter the needed network information and provide the NFS information to reach the RH install image and the box is able to read the o/s image and start an interactive install. I have also dropped to command line (ctl-alt-f2) and verified that the nfs directory holding the KS file can be mounted. And I have verified it can be retrieved with http using wget. And still anaconda is unable to retrieve the file with either nfs or http. > > Kickstart is a very powerful and useful tool that we use globally to > build over 1000 distributed systems each year from a centralized and > managed location, and although it has bugs like any product, most > issues like you are having tend to be infrastructure or manufacturer related. > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com > Sent: Monday, May 14, 2007 11:36 AM > To: Discussion list about Kickstart > Cc: Discussion list about Kickstart; kickstart-list-bounces at redhat.com > Subject: Re: Kickstart via PXE/NFS > > I made a change to pxe cfg file to have ks.cfg file read via http. > > It fails too. However, if I look in the anaconda log I see the > message getting ks.cfg > but then failed to transfer ks.cfg to fd. The snoop from the KS > server side shows the HTTP > retrieval of the KS config file. Is it really trying to use an fd > device, a floppy? These > Dell servers do not include floppys. > > Redhat sure has taken something that should be relatively easy and > made it a pain in the butt. > > > kickstart-list-bounces at redhat.com wrote on 05/14/2007 01:29:54 PM: > > > if you page-up in vty3, can you see what your kernel command line is? > > alternately if it makes it to shell in vty2, 'cat /proc/cmdline' - i > > wonder if it's actually doing the ipappend stuff. you might need a > > newer pxelinux.0 if not. also, instead of 'ksdevice=bootif' you can do > > 'ksdevice=link ip=dhcp' ; the bootif argument is more useful if you > > have 2 nics enabled/link-up and want to make sure it doesnt pxeboot > > out one interface and then ks out the other. > > > > -=Eric > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC > INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR > ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO > RECEIVE IT. If you are not the intended recipient of this document, > you are notified that any review, dissemination, distribution or > copying of this communication is prohibited. If you have received > this communication in error, please notify me immediately by return > email, delete the electronic message and destroy any printed copies. > Thank you for your cooperation._______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From debian at herakles.homelinux.org Mon May 14 23:29:32 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Tue, 15 May 2007 07:29:32 +0800 Subject: Kickstart via PXE/NFS In-Reply-To: <3BA8753665DB594C8C35CFA8801AD522BEFB27@ntxboimbx32.micron.com> References: <3BA8753665DB594C8C35CFA8801AD522BEFB27@ntxboimbx32.micron.com> Message-ID: <4648F0DC.4030502@herakles.homelinux.org> secroft at micron.com wrote: > For the second interface, try turning it off in the bios. I believe > you're running into an enumeration problem with the onboard NIC's. > Whether you have it plugged in or not, it is probably where the issue > is. Why not plug the other one in, or both? -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From SSTinsley at upsfreight.com Tue May 15 10:43:21 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Tue, 15 May 2007 06:43:21 -0400 Subject: Kickstart via PXE/NFS In-Reply-To: <3BA8753665DB594C8C35CFA8801AD522BEFB27@ntxboimbx32.micron.com> Message-ID: kickstart-list-bounces at redhat.com wrote on 05/14/2007 04:14:11 PM: > You still have never answered the question. We are trying to help > you here. It doesn't matter whether or not your second NIC is > plugged in, it does however matter whether or not it is on the same > bus as the NIC you are pxebooting from. As for why it works in pxe > and doesn't work in kickstart, they are two entirely different > systems. pxe comes from the system BIOS, and the kickstart > installer comes from RHEL. The reason this is important is because > RHEL 4 + enumerates the bus differently (and according to Red Hat > properly) than either pxe or previous RHEL versions. > > As for "complicated", my time goes back to building "unattend.txt" > files for Windows 95 (and every version since), and kickstart is FAR > more powerful and easier to use than any of the windows technologies > that are included with the OS. And if you don't like kickstart, try > using YAST sometime, that will make you old before your time. > > This list is very helpful and has gotten many people to quality > build setups, but disparaging the tool that we are trying to help > you deploy isn't going to win friends or help. > > At the end of the day, if you don't like anaconda, do something > about it. It's open source and you can dive in and fix the things > you don't like, and submit the patches upstream. There are many > things *I* would like to see changed, but I don't have the time to > write a patch, so I have to live with it or work around it. At > least now I can make the changes if I wanted to... Sure wish I could > have done that in Windows 10 years ago when I was trying to get an > unattend.txt file to bring up a backup domain controller that wasn't > online with the primary since it hadn't been built yet and was > located in a different state. > > Chip I appreciate your time and help. Just a bit frustrated after working on this for 2 solid days and watching the PC boot about 50 times. If you want free software to be successful, the standard answer can't be "fix it yourself if you don't like it". Many people, like me are not programmers. You don't want me in messing with code. People will eventually go where they can find someone to supply the fix. In my case, that is RedHat. And I have a case open with them. I was hoping I might get the problem solved faster going this route. kickstart-list-bounces at redhat.com wrote on 05/14/2007 04:01:42 PM: > For the second interface, try turning it off in the bios. I believe > you're running into an enumeration problem with the onboard NIC's. > Whether you have it plugged in or not, it is probably where the issue is. > kickstart-list-bounces at redhat.com wrote on 05/14/2007 04:03:21 PM: > I had a similar issue. I had only one NIC cabled to the switch. > The initial DHCP request was successful but it would fail on finding > the ks.cfg file. > > I ended up turning off all the NICs in the bios and this helped. > > I disabled the second interface in the bios as suggested. No difference. Let me describe what happened and show you the anacaonda log and you may see why this is so puzzling. The process up through the kernel load works fine. The initial DHCP request shows up on the network snoop, the pxelinux downloads. I get the PXE menu and kick off the auto install. The kernel and initrd download successfully. Here is the cmdline passed to the kernel based on the contents of /proc/cmdline. initrd=rhel_v4/initrd.img ks=http://10.1.1.253/redhat/ks_rhel_v4.cfg BOOT_IMAGE=rhel_v4/vmlinuz On the screen, I see the initial Anaconda screen with various messages on loading drivers. Then a message is displayed that DHCP is attempting to start eth0. It fails and the net config screen appears. I set it for DHCP again. No luck. Here is what shows up in the anaconda log after several DHCP attempts from the network config screen. only have one network device: eth0 * sending dhcp request through device eth0 * waiting for link... * 0 seconds. * running dhcp for eth0 * pump told us: No DHCP reply received * eth0 isn't a wireless adaptor * waiting for link... * 5 seconds. * running dhcp for eth0 * pump told us: No DHCP reply received * waiting for link... * 5 seconds. * running dhcp for eth0 * pump told us: No DHCP reply received * waiting for link... * 5 seconds. * running dhcp for eth0 * pump told us: No DHCP reply received * waiting for link... * 2 seconds. * no DNS servers, can't look up hostname * ks location: http://10.1.1.253/redhat/ks_rhel_v4.cfg * transferring http://10.1.1.253//./redhat/ks_rhel_v4.cfg to a fd * failed to retrieve http://10.1.1.253///redhat/ks_rhel_v4.cfg * trying to mount CD device hda * trying to mount CD device scd0 * going to set language to en_US.UTF-8 I finally assign the IP address statically on the network config screen. Anaconda moves on to ask for the location of the RedHat image. I supply the NFS information to get the release. Here is the odd part. Anaconda again shows that it is running DHCP to start interface eth0 even though I just supplied the static information. I am still snooping the network and I see this DHCP request and the same IP gets assigned that I had statically entered earlier. Anaconda then starts an interactive install. * trying to mount CD device hda * trying to mount CD device scd0 * going to set language to en_US.UTF-8 * setting language to en_US.UTF-8 * 164 keymaps are available * need to set up networking * going to pick interface * going to do getNetConfig * sending dhcp request through device eth0 * waiting for link... * 0 seconds. * running dhcp for eth0 * doing kickstart... setting it up * waiting for link... * 0 seconds. * starting to STEP_URL * going to do nfsGetSetup * mounting nfs path 10.1.1.253:/export/redhat/rhel4.0_ES_U5 * mounting nfs path 10.1.1.253:/export/redhat/rhel4.0_ES_U5 * mounted 10.1.1.253:/export/redhat/rhel4.0_ES_U5 on /mnt/source * can access /mnt/source/RedHat/base/stage2.img So, with only one interface active in the BIOS, why do the first several DHCP attempts fail, showing no network traffic at all? Then this last DHCP request shows up on the network and succeeds. However, at this point it is too late as the process is past the point where the kickstart file needs to be retrieved. Crazy!! NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From secroft at micron.com Tue May 15 13:42:03 2007 From: secroft at micron.com (secroft at micron.com) Date: Tue, 15 May 2007 07:42:03 -0600 Subject: Kickstart via PXE/NFS In-Reply-To: Message-ID: <3BA8753665DB594C8C35CFA8801AD522BEFB2B@ntxboimbx32.micron.com> Could you post your ks.cfg file, specifically the networking section. I have seen this, even when the kernel loads correctly, because the networking information is incorrect. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com Sent: Tuesday, May 15, 2007 4:43 AM To: Discussion list about Kickstart Cc: kickstart-list at redhat.com; kickstart-list-bounces at redhat.com Subject: RE: Kickstart via PXE/NFS kickstart-list-bounces at redhat.com wrote on 05/14/2007 04:14:11 PM: > You still have never answered the question. We are trying to help > you here. It doesn't matter whether or not your second NIC is > plugged in, it does however matter whether or not it is on the same > bus as the NIC you are pxebooting from. As for why it works in pxe > and doesn't work in kickstart, they are two entirely different > systems. pxe comes from the system BIOS, and the kickstart > installer comes from RHEL. The reason this is important is because > RHEL 4 + enumerates the bus differently (and according to Red Hat > properly) than either pxe or previous RHEL versions. > > As for "complicated", my time goes back to building "unattend.txt" > files for Windows 95 (and every version since), and kickstart is FAR > more powerful and easier to use than any of the windows technologies > that are included with the OS. And if you don't like kickstart, try > using YAST sometime, that will make you old before your time. > > This list is very helpful and has gotten many people to quality > build setups, but disparaging the tool that we are trying to help > you deploy isn't going to win friends or help. > > At the end of the day, if you don't like anaconda, do something > about it. It's open source and you can dive in and fix the things > you don't like, and submit the patches upstream. There are many > things *I* would like to see changed, but I don't have the time to > write a patch, so I have to live with it or work around it. At > least now I can make the changes if I wanted to... Sure wish I could > have done that in Windows 10 years ago when I was trying to get an > unattend.txt file to bring up a backup domain controller that wasn't > online with the primary since it hadn't been built yet and was > located in a different state. > > Chip I appreciate your time and help. Just a bit frustrated after working on this for 2 solid days and watching the PC boot about 50 times. If you want free software to be successful, the standard answer can't be "fix it yourself if you don't like it". Many people, like me are not programmers. You don't want me in messing with code. People will eventually go where they can find someone to supply the fix. In my case, that is RedHat. And I have a case open with them. I was hoping I might get the problem solved faster going this route. kickstart-list-bounces at redhat.com wrote on 05/14/2007 04:01:42 PM: > For the second interface, try turning it off in the bios. I believe > you're running into an enumeration problem with the onboard NIC's. > Whether you have it plugged in or not, it is probably where the issue is. > kickstart-list-bounces at redhat.com wrote on 05/14/2007 04:03:21 PM: > I had a similar issue. I had only one NIC cabled to the switch. > The initial DHCP request was successful but it would fail on finding > the ks.cfg file. > > I ended up turning off all the NICs in the bios and this helped. > > I disabled the second interface in the bios as suggested. No difference. Let me describe what happened and show you the anacaonda log and you may see why this is so puzzling. The process up through the kernel load works fine. The initial DHCP request shows up on the network snoop, the pxelinux downloads. I get the PXE menu and kick off the auto install. The kernel and initrd download successfully. Here is the cmdline passed to the kernel based on the contents of /proc/cmdline. initrd=rhel_v4/initrd.img ks=http://10.1.1.253/redhat/ks_rhel_v4.cfg BOOT_IMAGE=rhel_v4/vmlinuz On the screen, I see the initial Anaconda screen with various messages on loading drivers. Then a message is displayed that DHCP is attempting to start eth0. It fails and the net config screen appears. I set it for DHCP again. No luck. Here is what shows up in the anaconda log after several DHCP attempts from the network config screen. only have one network device: eth0 * sending dhcp request through device eth0 * waiting for link... * 0 seconds. * running dhcp for eth0 * pump told us: No DHCP reply received * eth0 isn't a wireless adaptor * waiting for link... * 5 seconds. * running dhcp for eth0 * pump told us: No DHCP reply received * waiting for link... * 5 seconds. * running dhcp for eth0 * pump told us: No DHCP reply received * waiting for link... * 5 seconds. * running dhcp for eth0 * pump told us: No DHCP reply received * waiting for link... * 2 seconds. * no DNS servers, can't look up hostname * ks location: http://10.1.1.253/redhat/ks_rhel_v4.cfg * transferring http://10.1.1.253//./redhat/ks_rhel_v4.cfg to a fd * failed to retrieve http://10.1.1.253///redhat/ks_rhel_v4.cfg * trying to mount CD device hda * trying to mount CD device scd0 * going to set language to en_US.UTF-8 I finally assign the IP address statically on the network config screen. Anaconda moves on to ask for the location of the RedHat image. I supply the NFS information to get the release. Here is the odd part. Anaconda again shows that it is running DHCP to start interface eth0 even though I just supplied the static information. I am still snooping the network and I see this DHCP request and the same IP gets assigned that I had statically entered earlier. Anaconda then starts an interactive install. * trying to mount CD device hda * trying to mount CD device scd0 * going to set language to en_US.UTF-8 * setting language to en_US.UTF-8 * 164 keymaps are available * need to set up networking * going to pick interface * going to do getNetConfig * sending dhcp request through device eth0 * waiting for link... * 0 seconds. * running dhcp for eth0 * doing kickstart... setting it up * waiting for link... * 0 seconds. * starting to STEP_URL * going to do nfsGetSetup * mounting nfs path 10.1.1.253:/export/redhat/rhel4.0_ES_U5 * mounting nfs path 10.1.1.253:/export/redhat/rhel4.0_ES_U5 * mounted 10.1.1.253:/export/redhat/rhel4.0_ES_U5 on /mnt/source * can access /mnt/source/RedHat/base/stage2.img So, with only one interface active in the BIOS, why do the first several DHCP attempts fail, showing no network traffic at all? Then this last DHCP request shows up on the network and succeeds. However, at this point it is too late as the process is past the point where the kickstart file needs to be retrieved. Crazy!! NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From secroft at micron.com Tue May 15 13:42:50 2007 From: secroft at micron.com (secroft at micron.com) Date: Tue, 15 May 2007 07:42:50 -0600 Subject: Kickstart via PXE/NFS In-Reply-To: <4648F0DC.4030502@herakles.homelinux.org> Message-ID: <3BA8753665DB594C8C35CFA8801AD522BEFB2C@ntxboimbx32.micron.com> No reason too unless you need it, and most times, people are short on network lines. If you have them, no reason not too..... -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John Summerfield Sent: Monday, May 14, 2007 5:30 PM To: Discussion list about Kickstart Subject: Re: Kickstart via PXE/NFS secroft at micron.com wrote: > For the second interface, try turning it off in the bios. I believe > you're running into an enumeration problem with the onboard NIC's. > Whether you have it plugged in or not, it is probably where the issue > is. Why not plug the other one in, or both? -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From pfhonore at gmail.com Tue May 15 14:49:43 2007 From: pfhonore at gmail.com (=?ISO-8859-1?Q?Pierre-Fran=E7ois_Honor=E9?=) Date: Tue, 15 May 2007 16:49:43 +0200 Subject: Kickstart via PXE/NFS In-Reply-To: References: Message-ID: <198fe3620705150749u280f235bjad06449dde2117fe@mail.gmail.com> Are you also using Dell hardware for network ? If so look at this in Rocks cluster: https://lists.sdsc.edu/pipermail/npaci-rocks-discussion/2007-March/024280.html Dell switches need to be configured with 'fast-port enable' for all ports to handle correctly the DCHP traffic. Or the first dhcp request by PXE is ok but the second one issued by Anaconda does not go through. 2007/5/14, SSTinsley at upsfreight.com : > > I am attempting to install RHEL ES V4 U5 on a Dell 2950 server. I am using a > Sun server as the DHCP/TFTP/NFS server supplying all data to complete the > install. I am able to start the PXE boot. The Dell server is able to > download the pxelinux image and also read the pxelinux.cfg file. Here is the > cfg file I am using. > > # Perform a local boot by default > default installauto > # Always prompt > prompt 1 > # Display the bootup message > display pxeboot.msg > # Boot automatically after 30 seconds in tenths of a second > timeout 300 > label local > localboot 0 > label rescue > ipappend 2 > kernel rhel_v4/vmlinuz > append initrd=rhel_v4/initrd.img rescue > label installauto > ipappend 2 > kernel rhel_v4/vmlinuz > append initrd=rhel_v4/initrd.img > ks=nfs:10.1.1.253:/export/redhat/ks_rhel_v4.cfg > ksdevice=bootif > > The kernel image loads from the server via tftp. The initial blue startup > screen appears at the start of installation. That is where things start to > go wrong. > > First, even though I have designated KS to use the boot interfaces in the > cfg file, I still get an interface selection screen at the start of the > install. Why is the selection necessary? Why is the iterface not being set > from the hsdevice option in the CFG file? > > Second, after selecting the interface, I switch to virt console 2 and see > the message that the install is attemting to get the ks_rhel_v4.cfg file and > the nfs URL is displayed. At the time this is happening I have a network > snoop at the Sun server. There is no NFS traffic showing up from the Dell. I > am snooping based on the Dell's MAC address. The output on the virt. console > indicates that the cfg file cannot be loaded. The network interface is > operable on the Dell. I can ping the configured address from the Sun DHCP > server. I can also boot the Dell in rescue more and NFS mount the directory > that contains the cfg file and also the directory that contains the > operating system image. Any ideas why this installation will not complete? > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT > IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED > ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the > intended recipient of this document, you are notified that any review, > dissemination, distribution or copying of this communication is prohibited. > If you have received this communication in error, please notify me > immediately by return email, delete the electronic message and destroy any > printed copies. Thank you for your cooperation. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -- Pierre-Fran?ois.Honor? PFHonore at gmail.com From SSTinsley at upsfreight.com Tue May 15 15:47:30 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Tue, 15 May 2007 11:47:30 -0400 Subject: Kickstart via PXE/NFS In-Reply-To: <3BA8753665DB594C8C35CFA8801AD522BEFB2B@ntxboimbx32.micron.com> Message-ID: kickstart-list-bounces at redhat.com wrote on 05/15/2007 09:42:03 AM: > Could you post your ks.cfg file, specifically the networking > section. I have seen this, even when the kernel loads correctly, > because the networking information is incorrect. > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com > Sent: Tuesday, May 15, 2007 4:43 AM > To: Discussion list about Kickstart > Cc: kickstart-list at redhat.com; kickstart-list-bounces at redhat.com > Subject: RE: Kickstart via PXE/NFS Curious why you would want to see the kickstart cfg since the log indicates it never loads. Here you go. #System language lang en_US #Language modules to install langsupport --default=en_US #System keyboard keyboard us #System mouse mouse none #Sytem timezone timezone America/New_York #Root password rootpw --iscrypted $1$DiNcd4k4$cm9CuU4ZpdCLI92saMZ4K/ #Reboot after installation reboot #Use text mode install text #Use interactive kickstart installation method ##interactive #Install Red Hat Linux instead of upgrade install #Use Web installation #url --url http://172.16.9.147/redhat/rhel3.0_ES_U8 nfs --server=10.1.1.253 --dir=/export/redhat/rhel4.0_ES_U5 #System bootloader configuration bootloader --location=mbr #Partition clearing information clearpart --all #Disk partitioning information part /boot --fstype=ext3 --ondisk=sda --size=256 part pv.11 --noformat --grow --ondisk=sda --size=0 volgroup rootvg pv.11 logvol / --fstype ext3 --vgname=rootvg --size=5192 --name=root_vol logvol swap --vgname=rootvg --size=2096 --name=swap_vol logvol /var --fstype ext3 --vgname=rootvg --size=2096 --name=var_vol #logvol /export --fstype ext3 --vgname=rootvg --size=4096 --name=export_vol #logvol /users --fstype ext3 --vgname=rootvg --size=4096 --name=users_vol #part /extra --fstype=ext3 --grow --ondisk=sda --size=0 #part pv.21 --noformat --grow --ondisk=sdb --size=10 #part pv.22 --noformat --grow --ondisk=sdb --size=10 ##volgroup uservg pv.21 pv.22 ##logvol /u002 --fstype=ext3 --vgname=uservg --size=4096 --name=u001_vol #System authorization infomation auth --useshadow --enablemd5 #Network information network --bootproto=dhcp --device=eth0 --ethtool="autoneg off duplex full" #network --bootproto=dhcp --device=eth1 --ethtool="autoneg off duplex full speed 100" #Firewall configuration firewall --enabled #Do not configure XWindows skipx #Package install information %packages --resolvedeps kernel NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From srobson at cadence.com Tue May 15 16:06:30 2007 From: srobson at cadence.com (Steve Robson) Date: Tue, 15 May 2007 17:06:30 +0100 Subject: Kickstart via PXE/NFS In-Reply-To: <20070515160013.4EDE0737D2@hormel.redhat.com> References: <20070515160013.4EDE0737D2@hormel.redhat.com> Message-ID: <4649DA86.6010708@cadence.com> kickstart-list-request at redhat.com wrote: > #Network information > network --bootproto=dhcp --device=eth0 --ethtool="autoneg off duplex full" Greetings, Try leaving the port to autonegotiate. You can always lock it once the system is built. -- Regards, Steve IT Support - UNIX/Linux Cadence Design Systems Tel: +44 (0)1344 866523 Bagshot Road Mobile: +44 (0)7901 855046 Bracknell BERKSHIRE Fax: +44 (0)1344 869647 RG12 0PH UK From secroft at micron.com Tue May 15 16:22:02 2007 From: secroft at micron.com (secroft at micron.com) Date: Tue, 15 May 2007 10:22:02 -0600 Subject: Kickstart via PXE/NFS In-Reply-To: Message-ID: <3BA8753665DB594C8C35CFA8801AD522BEFB31@ntxboimbx32.micron.com> You are correct that it is not necessarily relevant, was just curious. The previous post about the switch is something I had forgotten about and is definitely a problem that we corrected and all worked well. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com Sent: Tuesday, May 15, 2007 9:48 AM To: Discussion list about Kickstart Subject: RE: Kickstart via PXE/NFS kickstart-list-bounces at redhat.com wrote on 05/15/2007 09:42:03 AM: > Could you post your ks.cfg file, specifically the networking > section. I have seen this, even when the kernel loads correctly, > because the networking information is incorrect. > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com > Sent: Tuesday, May 15, 2007 4:43 AM > To: Discussion list about Kickstart > Cc: kickstart-list at redhat.com; kickstart-list-bounces at redhat.com > Subject: RE: Kickstart via PXE/NFS Curious why you would want to see the kickstart cfg since the log indicates it never loads. Here you go. #System language lang en_US #Language modules to install langsupport --default=en_US #System keyboard keyboard us #System mouse mouse none #Sytem timezone timezone America/New_York #Root password rootpw --iscrypted $1$DiNcd4k4$cm9CuU4ZpdCLI92saMZ4K/ #Reboot after installation reboot #Use text mode install text #Use interactive kickstart installation method ##interactive #Install Red Hat Linux instead of upgrade install #Use Web installation #url --url http://172.16.9.147/redhat/rhel3.0_ES_U8 nfs --server=10.1.1.253 --dir=/export/redhat/rhel4.0_ES_U5 #System bootloader configuration bootloader --location=mbr #Partition clearing information clearpart --all #Disk partitioning information part /boot --fstype=ext3 --ondisk=sda --size=256 part pv.11 --noformat --grow --ondisk=sda --size=0 volgroup rootvg pv.11 logvol / --fstype ext3 --vgname=rootvg --size=5192 --name=root_vol logvol swap --vgname=rootvg --size=2096 --name=swap_vol logvol /var --fstype ext3 --vgname=rootvg --size=2096 --name=var_vol #logvol /export --fstype ext3 --vgname=rootvg --size=4096 --name=export_vol #logvol /users --fstype ext3 --vgname=rootvg --size=4096 --name=users_vol #part /extra --fstype=ext3 --grow --ondisk=sda --size=0 #part pv.21 --noformat --grow --ondisk=sdb --size=10 #part pv.22 --noformat --grow --ondisk=sdb --size=10 ##volgroup uservg pv.21 pv.22 ##logvol /u002 --fstype=ext3 --vgname=uservg --size=4096 --name=u001_vol #System authorization infomation auth --useshadow --enablemd5 #Network information network --bootproto=dhcp --device=eth0 --ethtool="autoneg off duplex full" #network --bootproto=dhcp --device=eth1 --ethtool="autoneg off duplex full speed 100" #Firewall configuration firewall --enabled #Do not configure XWindows skipx #Package install information %packages --resolvedeps kernel NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From SSTinsley at upsfreight.com Tue May 15 16:48:32 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Tue, 15 May 2007 12:48:32 -0400 Subject: Kickstart via PXE/NFS In-Reply-To: <3BA8753665DB594C8C35CFA8801AD522BEFB31@ntxboimbx32.micron.com> Message-ID: > Are you also using Dell hardware for network ? > > If so look at this in Rocks cluster: > https://lists.sdsc.edu/pipermail/npaci-rocks-discussion/2007-March/024280.html > > Dell switches need to be configured with 'fast-port enable' for all > ports to handle correctly the DCHP traffic. Or the first dhcp request > by PXE is ok but the second one issued by Anaconda does not go > through. kickstart-list-bounces at redhat.com wrote on 05/15/2007 12:22:02 PM: > You are correct that it is not necessarily relevant, was just > curious. The previous post about the switch is something I had > forgotten about and is definitely a problem that we corrected and > all worked well. > I am plugged in to a CISCO switch, but I will check with the network guys to find out if the setting may still be applicable. Nothing back from Redhat yet at this point. NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From SSTinsley at upsfreight.com Tue May 15 17:19:13 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Tue, 15 May 2007 13:19:13 -0400 Subject: Kickstart via PXE/NFS In-Reply-To: Message-ID: I have now tried the install by using DHCP and fixed IP address. I would hope that assigning the IP with kernel arguments through the pxelinux cfg would work. Assuming it does, the only other common error is related to DNS. I do not have DNS available on this private install net. Does Anaconda require DNS for a completely hands-off install? kickstart-list-bounces at redhat.com wrote on 05/15/2007 12:48:32 PM: > > > Are you also using Dell hardware for network ? > > > > If so look at this in Rocks cluster: > > https://lists.sdsc.edu/pipermail/npaci-rocks-discussion/2007- > March/024280.html > > > > Dell switches need to be configured with 'fast-port enable' for all > > ports to handle correctly the DCHP traffic. Or the first dhcp request > > by PXE is ok but the second one issued by Anaconda does not go > > through. > > > kickstart-list-bounces at redhat.com wrote on 05/15/2007 12:22:02 PM: > > > You are correct that it is not necessarily relevant, was just > > curious. The previous post about the switch is something I had > > forgotten about and is definitely a problem that we corrected and > > all worked well. > > > > > I am plugged in to a CISCO switch, but I will check with the network > guys to find out if the setting may still be applicable. > > Nothing back from Redhat yet at this point. > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC > INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR > ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO > RECEIVE IT. If you are not the intended recipient of this document, > you are notified that any review, dissemination, distribution or > copying of this communication is prohibited. If you have received > this communication in error, please notify me immediately by return > email, delete the electronic message and destroy any printed copies. > Thank you for your cooperation._______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From patchu1 at yahoo.com Tue May 15 17:57:41 2007 From: patchu1 at yahoo.com (Pat) Date: Tue, 15 May 2007 10:57:41 -0700 (PDT) Subject: Location of installation media no longer /tmp/cdrom? Message-ID: <963676.55991.qm@web32504.mail.mud.yahoo.com> I copy over a compressed file (contains custom files) from a custom installation dvd to /root of the installed system in the %post section of my kickstart file. It seems to no longer work with CentOS 5(x86_64). Here's what the relevant section of my kickstart file looks like: %post --nochroot mkdir -p /mnt/source mount /tmp/cdrom /mnt/source cp /mnt/source/mystuff.tar.gz /mnt/sysimage/root/ umount /mnt/source However I get the following error during the install: mount: Mounting /tmp/cdrom on /mnt/source failed: No such file or directory So it looks like the location of the install media is no longer /tmp/cdrom? I poked around but could not find where it is now located. Can someone clue me in? Thanks! ____________________________________________________________________________________Sick sense of humor? Visit Yahoo! TV's Comedy with an Edge to see what's on, when. http://tv.yahoo.com/collections/222 From andymeany at yahoo.com Tue May 15 18:32:43 2007 From: andymeany at yahoo.com (Andy Meany) Date: Tue, 15 May 2007 11:32:43 -0700 (PDT) Subject: FC6 - How to update packages on kickstart server In-Reply-To: <80145c330705101231m28090c59t353b88d8dc8ae6fa@mail.gmail.com> Message-ID: <993340.2124.qm@web63304.mail.re1.yahoo.com> Thank you all for your input. I used novi to create a patched installation tree. It was actually quite simple to do and it worked great! Eric Sorenson wrote: On 5/10/07, kickstart wrote: > Eric Sorenson: > > Actually mrepo (wonderful though it is) does not actually do what the > > original poster wants. > > He wants to take the RPMS from the cds and overlay downloaded updates > > to build a new install tree. Ethan McCallam's 'novi' does this: > > http://www.exmachinatech.net/01/novi/ > > Precisely. Which is why I mentioned novi earlier today:) Yeah, whups, that's what I get for jumping to the end of the thread. Consider your recommendation seconded ;) > > What I am doing (an extension of the suggestion in the OReilly book > by Ethan McCallum which I also mentioned) is to 1) fetch updates > from multiple external repos into subdirectories of a new > ("pre-patch") tree of your base OS, the then 2) run novi on these > leaving the new prepatched tree with the latest RPMs. You'd then > point new kickstarts at this prepatched tree. > > Of course, some testing would be in order every time you do a major > upgrade to make sure nothing's suddenly broken. > > In my case I'm using CentOS44 so I also have genhdlist running in my > automation scripts.. Have you had to modify comps.xml as well? > > It has worked quite well so far.. > > -Harold > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list --------------------------------- Sucker-punch spam with award-winning protection. Try the free Yahoo! Mail Beta. -------------- next part -------------- An HTML attachment was scrubbed... URL: From kickstart at hastek.net Tue May 15 19:03:19 2007 From: kickstart at hastek.net (kickstart) Date: Tue, 15 May 2007 15:03:19 -0400 Subject: FC6 - How to update packages on kickstart server In-Reply-To: <993340.2124.qm@web63304.mail.re1.yahoo.com> References: <80145c330705101231m28090c59t353b88d8dc8ae6fa@mail.gmail.com> <993340.2124.qm@web63304.mail.re1.yahoo.com> Message-ID: <20070515190318.GI80015@florian> Andy Meany: > Thank you all for your input. I used novi to create a patched > installation tree. It was actually quite simple to do and it worked Novi does work well for this! I don't know if this is helpful for coding ideas or shortcuts, but here is my script for fetching from the CentOS repo using rsync. Once you do the initial pull it becomes very lightweight what with rsync. http://hastek.net/novi/centos-x86_64-updater.html The branch /data1/kickstart/CentOS44-prepatched/ was first populated by copying a centos44 DVD install image to it. -Harold From debian at herakles.homelinux.org Tue May 15 15:34:51 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Tue, 15 May 2007 23:34:51 +0800 Subject: Kickstart via PXE/NFS In-Reply-To: <3BA8753665DB594C8C35CFA8801AD522BEFB2C@ntxboimbx32.micron.com> References: <3BA8753665DB594C8C35CFA8801AD522BEFB2C@ntxboimbx32.micron.com> Message-ID: <4649D31B.3090201@herakles.homelinux.org> secroft at micron.com wrote: > No reason too unless you need it, and most times, people are short on > network lines. If you have them, no reason not too..... > Seems to me in these particular circumstances it might be a simple way around a problem that's proven hard to fix properly. > secroft at micron.com wrote: >> For the second interface, try turning it off in the bios. I believe >> you're running into an enumeration problem with the onboard NIC's. >> Whether you have it plugged in or not, it is probably where the issue >> is. > > Why not plug the other one in, or both? > > -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From muksyed at stanford.edu Tue May 15 23:49:51 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Tue, 15 May 2007 16:49:51 -0700 Subject: Unable to get any http requests: Unable to retreive netstg2.img file Message-ID: <001601c7974b$bb1b0500$6d1b42ab@stanford.edu> Hi I am sure there is some information regarding my request, but I am unable to find it, hence the post. I configured my kickstart server using http. I am able to pxeboot the client, but my client is not able to get the netstg2.img file from the server. I am able to http into my netstg2.img directory from a web browser from another server in the same subnet successfully. Also I tried to put in a boot cdrom and run the following: Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg ip= netmask=255.255.255.224 gateway= Still no luck. It goes straight to the cd installer. I tried to tcpdump the traffic on the MAC address of the client, it outputs the tftp information and after that it stops. I don't see any http traffic coming from the client. I'd appreciate some help in this regard. I've spent all day trying to troubleshoot this but no luck. Thanks Much Mukarram Syed Unix Systems Administrator, Stanford University. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Colin.Coe at woodside.com.au Tue May 15 23:51:32 2007 From: Colin.Coe at woodside.com.au (Coe, Colin C. (Unix Engineer)) Date: Wed, 16 May 2007 07:51:32 +0800 Subject: Unable to get any http requests: Unable to retreive netstg2.img file In-Reply-To: <001601c7974b$bb1b0500$6d1b42ab@stanford.edu> Message-ID: Does the machine you're on have multiple Ethernet interfaces? If so, what you may find is adding 'ksdevice=link' will assist. CC ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed Sent: Wednesday, 16 May 2007 7:50 AM To: kickstart-list at redhat.com Subject: Unable to get any http requests: Unable to retreive netstg2.img file Hi I am sure there is some information regarding my request, but I am unable to find it, hence the post. I configured my kickstart server using http. I am able to pxeboot the client, but my client is not able to get the netstg2.img file from the server. I am able to http into my netstg2.img directory from a web browser from another server in the same subnet successfully. Also I tried to put in a boot cdrom and run the following: Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg ip= netmask=255.255.255.224 gateway= Still no luck. It goes straight to the cd installer. I tried to tcpdump the traffic on the MAC address of the client, it outputs the tftp information and after that it stops. I don't see any http traffic coming from the client. I'd appreciate some help in this regard. I've spent all day trying to troubleshoot this but no luck. Thanks Much Mukarram Syed Unix Systems Administrator, Stanford University. NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- An HTML attachment was scrubbed... URL: From muksyed at stanford.edu Wed May 16 00:04:47 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Tue, 15 May 2007 17:04:47 -0700 Subject: Unable to get any http requests: Unable to retreive netstg2.imgfile In-Reply-To: Message-ID: <002701c7974d$d1bb7db0$6d1b42ab@stanford.edu> Thanks Colin for the quick response Yes. It does have. It's a Dell PE2950. I have already tried that option as well. I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. -Mukarram. _____ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: Tuesday, May 15, 2007 4:52 PM To: Discussion list about Kickstart Subject: RE: Unable to get any http requests: Unable to retreive netstg2.imgfile Does the machine you're on have multiple Ethernet interfaces? If so, what you may find is adding 'ksdevice=link' will assist. CC _____ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed Sent: Wednesday, 16 May 2007 7:50 AM To: kickstart-list at redhat.com Subject: Unable to get any http requests: Unable to retreive netstg2.img file Hi I am sure there is some information regarding my request, but I am unable to find it, hence the post. I configured my kickstart server using http. I am able to pxeboot the client, but my client is not able to get the netstg2.img file from the server. I am able to http into my netstg2.img directory from a web browser from another server in the same subnet successfully. Also I tried to put in a boot cdrom and run the following: Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg ip= netmask=255.255.255.224 gateway= Still no luck. It goes straight to the cd installer. I tried to tcpdump the traffic on the MAC address of the client, it outputs the tftp information and after that it stops. I don't see any http traffic coming from the client. I'd appreciate some help in this regard. I've spent all day trying to troubleshoot this but no luck. Thanks Much Mukarram Syed Unix Systems Administrator, Stanford University. NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Colin.Coe at woodside.com.au Wed May 16 00:15:48 2007 From: Colin.Coe at woodside.com.au (Coe, Colin C. (Unix Engineer)) Date: Wed, 16 May 2007 08:15:48 +0800 Subject: Unable to get any http requests: Unable to retreivenetstg2.imgfile In-Reply-To: <002701c7974d$d1bb7db0$6d1b42ab@stanford.edu> Message-ID: What distro are you trying to install (RHEL3,RHEL4,FC6, etc)? ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed Sent: Wednesday, 16 May 2007 8:05 AM To: 'Discussion list about Kickstart' Subject: RE: Unable to get any http requests: Unable to retreivenetstg2.imgfile Thanks Colin for the quick response Yes. It does have. It's a Dell PE2950. I have already tried that option as well. I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. -Mukarram. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: Tuesday, May 15, 2007 4:52 PM To: Discussion list about Kickstart Subject: RE: Unable to get any http requests: Unable to retreive netstg2.imgfile Does the machine you're on have multiple Ethernet interfaces? If so, what you may find is adding 'ksdevice=link' will assist. CC ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed Sent: Wednesday, 16 May 2007 7:50 AM To: kickstart-list at redhat.com Subject: Unable to get any http requests: Unable to retreive netstg2.img file Hi I am sure there is some information regarding my request, but I am unable to find it, hence the post. I configured my kickstart server using http. I am able to pxeboot the client, but my client is not able to get the netstg2.img file from the server. I am able to http into my netstg2.img directory from a web browser from another server in the same subnet successfully. Also I tried to put in a boot cdrom and run the following: Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg ip= netmask=255.255.255.224 gateway= Still no luck. It goes straight to the cd installer. I tried to tcpdump the traffic on the MAC address of the client, it outputs the tftp information and after that it stops. I don't see any http traffic coming from the client. I'd appreciate some help in this regard. I've spent all day trying to troubleshoot this but no luck. Thanks Much Mukarram Syed Unix Systems Administrator, Stanford University. NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- An HTML attachment was scrubbed... URL: From CallahanT at tessco.com Wed May 16 00:13:01 2007 From: CallahanT at tessco.com (Callahan, Tom) Date: Tue, 15 May 2007 20:13:01 -0400 Subject: Unable to get any http requests: Unable to retreivenetstg2.imgfile References: <002701c7974d$d1bb7db0$6d1b42ab@stanford.edu> Message-ID: <6982E319B698F143AEDB8F3D83980457524192@EXCHANGE-1.tessco.com> The path and network settings from the kickstart "can" apply once it pulls the ks file. Make sure the network settings are correct in the kickstart file. You could create a %pre section in your installer, and attempt to pull the file using something like wget. Write a small script to do it, and on failure, tell it to just sit there. Then you can use CTRL-F1 thru CTRL-F4 to see what the errors were. Botched kickstarts can get tricky, especially when it seems like a network issue. Thanks, Tom Callahan -----Original Message----- From: kickstart-list-bounces at redhat.com on behalf of Mukarram Syed Sent: Tue 5/15/2007 8:04 PM To: 'Discussion list about Kickstart' Subject: RE: Unable to get any http requests: Unable to retreive netstg2.imgfile Thanks Colin for the quick response Yes. It does have. It's a Dell PE2950. I have already tried that option as well. I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. -Mukarram. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: Tuesday, May 15, 2007 4:52 PM To: Discussion list about Kickstart Subject: RE: Unable to get any http requests: Unable to retreive netstg2.imgfile Does the machine you're on have multiple Ethernet interfaces? If so, what you may find is adding 'ksdevice=link' will assist. CC ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed Sent: Wednesday, 16 May 2007 7:50 AM To: kickstart-list at redhat.com Subject: Unable to get any http requests: Unable to retreive netstg2.img file Hi I am sure there is some information regarding my request, but I am unable to find it, hence the post. I configured my kickstart server using http. I am able to pxeboot the client, but my client is not able to get the netstg2.img file from the server. I am able to http into my netstg2.img directory from a web browser from another server in the same subnet successfully. Also I tried to put in a boot cdrom and run the following: Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg ip= netmask=255.255.255.224 gateway= Still no luck. It goes straight to the cd installer. I tried to tcpdump the traffic on the MAC address of the client, it outputs the tftp information and after that it stops. I don't see any http traffic coming from the client. I'd appreciate some help in this regard. I've spent all day trying to troubleshoot this but no luck. Thanks Much Mukarram Syed Unix Systems Administrator, Stanford University. NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- An HTML attachment was scrubbed... URL: From muksyed at stanford.edu Wed May 16 00:25:55 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Tue, 15 May 2007 17:25:55 -0700 Subject: Unable to get any http requests: Unable toretreivenetstg2.imgfile In-Reply-To: Message-ID: <003e01c79750$c56ab0f0$6d1b42ab@stanford.edu> Excellent question. Oracle Enterprise Linux. But it's pretty much RedHat4. Just that it has Enterprise written all over it and the kernel has a bunch of 0.0.1 prefixed to it. -Mukarram. _____ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: Tuesday, May 15, 2007 5:16 PM To: Discussion list about Kickstart Subject: RE: Unable to get any http requests: Unable toretreivenetstg2.imgfile What distro are you trying to install (RHEL3,RHEL4,FC6, etc)? _____ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed Sent: Wednesday, 16 May 2007 8:05 AM To: 'Discussion list about Kickstart' Subject: RE: Unable to get any http requests: Unable to retreivenetstg2.imgfile Thanks Colin for the quick response Yes. It does have. It's a Dell PE2950. I have already tried that option as well. I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. -Mukarram. _____ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: Tuesday, May 15, 2007 4:52 PM To: Discussion list about Kickstart Subject: RE: Unable to get any http requests: Unable to retreive netstg2.imgfile Does the machine you're on have multiple Ethernet interfaces? If so, what you may find is adding 'ksdevice=link' will assist. CC _____ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed Sent: Wednesday, 16 May 2007 7:50 AM To: kickstart-list at redhat.com Subject: Unable to get any http requests: Unable to retreive netstg2.img file Hi I am sure there is some information regarding my request, but I am unable to find it, hence the post. I configured my kickstart server using http. I am able to pxeboot the client, but my client is not able to get the netstg2.img file from the server. I am able to http into my netstg2.img directory from a web browser from another server in the same subnet successfully. Also I tried to put in a boot cdrom and run the following: Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg ip= netmask=255.255.255.224 gateway= Still no luck. It goes straight to the cd installer. I tried to tcpdump the traffic on the MAC address of the client, it outputs the tftp information and after that it stops. I don't see any http traffic coming from the client. I'd appreciate some help in this regard. I've spent all day trying to troubleshoot this but no luck. Thanks Much Mukarram Syed Unix Systems Administrator, Stanford University. NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- An HTML attachment was scrubbed... URL: From muksyed at stanford.edu Wed May 16 00:34:05 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Tue, 15 May 2007 17:34:05 -0700 Subject: Unable to get any http requests: Unable toretreivenetstg2.imgfile In-Reply-To: <6982E319B698F143AEDB8F3D83980457524192@EXCHANGE-1.tessco.com> Message-ID: <004901c79751$e8e6e660$6d1b42ab@stanford.edu> Thanks Tom. It looks like it's not reading the ks.cfg file from the http server or maybe can't find it. The installer goes to Language selection after booting off pxe/tftp. I created my ks.cfg file from the system-config-kickstart gui after using my other ks.cfg file. Just to make sure that the software creates it. But same issue. Here is my default file: label 1 kernel ENT_LINUX_64/vmlinuz append nofb ks=http://171.67.39.130/ENT_LINUX_64/Enterprise/ks.cfg initrd=ENT_LINUX_64/initrd.img ramdisk_size=10000 lang= devfs=nomount ramdisk_size=9216 ksdevice=eth0 ip=171.67.39.139 netmask=255.255.255.224 gateway=171.67.39.129 console=ttyS0,9600 and my ks.cfg file: #platform=x86, AMD64, or Intel EM64T #System language lang en_US #Language modules to install langsupport en_US #System keyboard keyboard us #System mouse mouse #Sytem timezone timezone America/Los_Angeles #Root password rootpw welcome #Reboot after installation reboot #Install OS instead of upgrade install #Use Web installation url --url http://171.67.39.130/ENT_LINUX_64/Enterprise #System bootloader configuration bootloader --location=mbr #Clear the Master Boot Record zerombr yes #Partition clearing information clearpart --all --initlabel #Disk partitioning information part /boot --fstype ext3 --size 100 --asprimary part / --fstype ext3 --size 1 --grow --asprimary part swap --recommended --asprimary #System authorization infomation auth --useshadow --enablemd5 #Network information network --bootproto=static --ip=171.67.39.139 --netmask=255.255.255.224 --gateway=171.67.39.129 --nameserver=000.000.000.000 --device=eth0 #Firewall configuration firewall --disabled #Do not configure XWindows skipx #Package install information %packages --resolvedeps @ admin-tools @ system-tools @ printing @ compat-arch-support _____ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Callahan, Tom Sent: Tuesday, May 15, 2007 5:13 PM To: Discussion list about Kickstart; Discussion list about Kickstart Subject: RE: Unable to get any http requests: Unable toretreivenetstg2.imgfile The path and network settings from the kickstart "can" apply once it pulls the ks file. Make sure the network settings are correct in the kickstart file. You could create a %pre section in your installer, and attempt to pull the file using something like wget. Write a small script to do it, and on failure, tell it to just sit there. Then you can use CTRL-F1 thru CTRL-F4 to see what the errors were. Botched kickstarts can get tricky, especially when it seems like a network issue. Thanks, Tom Callahan -----Original Message----- From: kickstart-list-bounces at redhat.com on behalf of Mukarram Syed Sent: Tue 5/15/2007 8:04 PM To: 'Discussion list about Kickstart' Subject: RE: Unable to get any http requests: Unable to retreive netstg2.imgfile Thanks Colin for the quick response Yes. It does have. It's a Dell PE2950. I have already tried that option as well. I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. -Mukarram. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: Tuesday, May 15, 2007 4:52 PM To: Discussion list about Kickstart Subject: RE: Unable to get any http requests: Unable to retreive netstg2.imgfile Does the machine you're on have multiple Ethernet interfaces? If so, what you may find is adding 'ksdevice=link' will assist. CC ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed Sent: Wednesday, 16 May 2007 7:50 AM To: kickstart-list at redhat.com Subject: Unable to get any http requests: Unable to retreive netstg2.img file Hi I am sure there is some information regarding my request, but I am unable to find it, hence the post. I configured my kickstart server using http. I am able to pxeboot the client, but my client is not able to get the netstg2.img file from the server. I am able to http into my netstg2.img directory from a web browser from another server in the same subnet successfully. Also I tried to put in a boot cdrom and run the following: Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg ip= netmask=255.255.255.224 gateway= Still no luck. It goes straight to the cd installer. I tried to tcpdump the traffic on the MAC address of the client, it outputs the tftp information and after that it stops. I don't see any http traffic coming from the client. I'd appreciate some help in this regard. I've spent all day trying to troubleshoot this but no luck. Thanks Much Mukarram Syed Unix Systems Administrator, Stanford University. NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- An HTML attachment was scrubbed... URL: From debian at herakles.homelinux.org Wed May 16 00:32:16 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Wed, 16 May 2007 08:32:16 +0800 Subject: Unable to get any http requests: Unable to retreive netstg2.imgfile In-Reply-To: <002701c7974d$d1bb7db0$6d1b42ab@stanford.edu> References: <002701c7974d$d1bb7db0$6d1b42ab@stanford.edu> Message-ID: <464A5110.9040202@herakles.homelinux.org> Mukarram Syed wrote: > Thanks Colin for the quick response > > > > Yes. It does have. It's a Dell PE2950. > > > > I have already tried that option as well. > > I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. > If you can connect all network interfaces, that might help, because it then doesn't matter whether either one or both is used. If you run arpwatch that will report on which interfaces are used (by mac address and IP). More below... > > > -Mukarram. > > > > > > _____ > > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix > Engineer) > Sent: Tuesday, May 15, 2007 4:52 PM > To: Discussion list about Kickstart > Subject: RE: Unable to get any http requests: Unable to retreive > netstg2.imgfile > > > > Does the machine you're on have multiple Ethernet interfaces? If so, what > you may find is adding 'ksdevice=link' will assist. > > > > CC > > > > > _____ > > > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed > Sent: Wednesday, 16 May 2007 7:50 AM > To: kickstart-list at redhat.com > Subject: Unable to get any http requests: Unable to retreive netstg2.img > file > > Hi > > I am sure there is some information regarding my request, but I am unable to > find it, hence the post. > > > > I configured my kickstart server using http. > > I am able to pxeboot the client, but my client is not able to get the > netstg2.img file from the server. > > I am able to http into my netstg2.img directory from a web browser from > another server in the same subnet successfully. > > > > Also I tried to put in a boot cdrom and run the following: > > Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg ip= > netmask=255.255.255.224 gateway= I have never used static IP for ks, I alway configure everything I can with dhcp. However, I do know that that's not the format the kernel expects for the ip parameter. Check your documentation, if I were writing a script tp parse ip= then I'd conform to the existing standards (that is, the way the kernel expects it), and that's what works with Ubuntu. > > > > Still no luck. It goes straight to the cd installer. > > > > I tried to tcpdump the traffic on the MAC address of the client, it outputs > the tftp information and after that it stops. > > I don't see any http traffic coming from the client. That's fairly consistent with you specifying ip= incorrectly. btw, if you're using PXE, the why are you trying to spectify static IP info? btw, as I recall, pxelinux has an option to pass on the IP address info it gets in the proper format for the kernel to parse. > > > > I'd appreciate some help in this regard. > > I've spent all day trying to troubleshoot this but no luck. > > > > Thanks Much > > > > > > Mukarram Syed > > Unix Systems Administrator, > > Stanford University. > > > > > NOTICE: This email and any attachments are confidential. > They may contain legally privileged information or > copyright material. You must not read, copy, use or > disclose them without authorisation. If you are not an > intended recipient, please contact us at once by return > email and then delete both messages and all attachments. > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From muksyed at stanford.edu Wed May 16 00:42:11 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Tue, 15 May 2007 17:42:11 -0700 Subject: Unable to get any http requests: Unable toretreive netstg2.imgfile In-Reply-To: <464A5110.9040202@herakles.homelinux.org> Message-ID: <005401c79753$0fcbc240$6d1b42ab@stanford.edu> John, Sorry...just to make things clear. eth0 works to get the dhcp/tftp request and eth1 does not. However, I tried using both is what I meant. We have a problem with dhcp, the networking folks here did not enable "spanning tree" on the switch ports, until that happens I am using static ip. I double checked the syntax etc. Thanks -Mukarram. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John Summerfield Sent: Tuesday, May 15, 2007 5:32 PM To: Discussion list about Kickstart Subject: Re: Unable to get any http requests: Unable toretreive netstg2.imgfile Mukarram Syed wrote: > Thanks Colin for the quick response > > > > Yes. It does have. It's a Dell PE2950. > > > > I have already tried that option as well. > > I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. > If you can connect all network interfaces, that might help, because it then doesn't matter whether either one or both is used. If you run arpwatch that will report on which interfaces are used (by mac address and IP). More below... > > > -Mukarram. > > > > > > _____ > > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix > Engineer) > Sent: Tuesday, May 15, 2007 4:52 PM > To: Discussion list about Kickstart > Subject: RE: Unable to get any http requests: Unable to retreive > netstg2.imgfile > > > > Does the machine you're on have multiple Ethernet interfaces? If so, what > you may find is adding 'ksdevice=link' will assist. > > > > CC > > > > > _____ > > > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed > Sent: Wednesday, 16 May 2007 7:50 AM > To: kickstart-list at redhat.com > Subject: Unable to get any http requests: Unable to retreive netstg2.img > file > > Hi > > I am sure there is some information regarding my request, but I am unable to > find it, hence the post. > > > > I configured my kickstart server using http. > > I am able to pxeboot the client, but my client is not able to get the > netstg2.img file from the server. > > I am able to http into my netstg2.img directory from a web browser from > another server in the same subnet successfully. > > > > Also I tried to put in a boot cdrom and run the following: > > Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg ip= > netmask=255.255.255.224 gateway= I have never used static IP for ks, I alway configure everything I can with dhcp. However, I do know that that's not the format the kernel expects for the ip parameter. Check your documentation, if I were writing a script tp parse ip= then I'd conform to the existing standards (that is, the way the kernel expects it), and that's what works with Ubuntu. > > > > Still no luck. It goes straight to the cd installer. > > > > I tried to tcpdump the traffic on the MAC address of the client, it outputs > the tftp information and after that it stops. > > I don't see any http traffic coming from the client. That's fairly consistent with you specifying ip= incorrectly. btw, if you're using PXE, the why are you trying to spectify static IP info? btw, as I recall, pxelinux has an option to pass on the IP address info it gets in the proper format for the kernel to parse. > > > > I'd appreciate some help in this regard. > > I've spent all day trying to troubleshoot this but no luck. > > > > Thanks Much > > > > > > Mukarram Syed > > Unix Systems Administrator, > > Stanford University. > > > > > NOTICE: This email and any attachments are confidential. > They may contain legally privileged information or > copyright material. You must not read, copy, use or > disclose them without authorisation. If you are not an > intended recipient, please contact us at once by return > email and then delete both messages and all attachments. > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Colin.Coe at woodside.com.au Wed May 16 00:56:08 2007 From: Colin.Coe at woodside.com.au (Coe, Colin C. (Unix Engineer)) Date: Wed, 16 May 2007 08:56:08 +0800 Subject: Unable to get any http requests: Unabletoretreive netstg2.imgfile In-Reply-To: <005401c79753$0fcbc240$6d1b42ab@stanford.edu> Message-ID: Try changing 'ksdevice=eth0' to 'ksdevice=link'. With 2.6 kernels, the kernel's idea of what eth0 and eth1 are often differs from the BIOS/PXE stack. CC > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed > Sent: Wednesday, 16 May 2007 8:42 AM > To: 'Discussion list about Kickstart' > Subject: RE: Unable to get any http requests: > Unabletoretreive netstg2.imgfile > > John, > Sorry...just to make things clear. > eth0 works to get the dhcp/tftp request and eth1 does not. > However, I tried using both is what I meant. > > We have a problem with dhcp, the networking folks here did > not enable "spanning tree" on the switch ports, until that > happens I am using static > ip. I double checked the syntax etc. > > Thanks > -Mukarram. > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John > Summerfield > Sent: Tuesday, May 15, 2007 5:32 PM > To: Discussion list about Kickstart > Subject: Re: Unable to get any http requests: Unable > toretreive netstg2.imgfile > > Mukarram Syed wrote: > > Thanks Colin for the quick response > > > > > > > > Yes. It does have. It's a Dell PE2950. > > > > > > > > I have already tried that option as well. > > > > I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. > > > > If you can connect all network interfaces, that might help, > because it then doesn't matter whether either one or both is used. > > If you run arpwatch that will report on which interfaces are > used (by mac address and IP). > > More below... > > > > > > > -Mukarram. > > > > > > > > > > > > _____ > > > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > Coe, Colin C. > (Unix > > Engineer) > > Sent: Tuesday, May 15, 2007 4:52 PM > > To: Discussion list about Kickstart > > Subject: RE: Unable to get any http requests: Unable to retreive > > netstg2.imgfile > > > > > > > > Does the machine you're on have multiple Ethernet > interfaces? If so, > > what you may find is adding 'ksdevice=link' will assist. > > > > > > > > CC > > > > > > > > > > _____ > > > > > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > Mukarram Syed > > Sent: Wednesday, 16 May 2007 7:50 AM > > To: kickstart-list at redhat.com > > Subject: Unable to get any http requests: Unable to retreive > > netstg2.img file > > > > Hi > > > > I am sure there is some information regarding my request, but I am > > unable > to > > find it, hence the post. > > > > > > > > I configured my kickstart server using http. > > > > I am able to pxeboot the client, but my client is not able > to get the > > netstg2.img file from the server. > > > > I am able to http into my netstg2.img directory from a web browser > > from another server in the same subnet successfully. > > > > > > > > Also I tried to put in a boot cdrom and run the following: > > > > Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg > > ip= > > netmask=255.255.255.224 gateway= > > > I have never used static IP for ks, I alway configure > everything I can with dhcp. However, I do know that that's > not the format the kernel expects for the ip parameter. Check > your documentation, if I were writing a script tp parse ip= > then I'd conform to the existing standards (that is, the way > the kernel expects it), and that's what works with Ubuntu. > > > > > > > > > > > Still no luck. It goes straight to the cd installer. > > > > > > > > I tried to tcpdump the traffic on the MAC address of the client, it > outputs > > the tftp information and after that it stops. > > > > I don't see any http traffic coming from the client. > > That's fairly consistent with you specifying ip= incorrectly. > > btw, if you're using PXE, the why are you trying to spectify > static IP info? > > btw, as I recall, pxelinux has an option to pass on the IP > address info it gets in the proper format for the kernel to parse. > > > > > > > > > > I'd appreciate some help in this regard. > > > > I've spent all day trying to troubleshoot this but no luck. > > > > > > > > Thanks Much > > > > > > > > > > > > Mukarram Syed > > > > Unix Systems Administrator, > > > > Stanford University. > > > > > > > > > > NOTICE: This email and any attachments are confidential. > > They may contain legally privileged information or > > copyright material. You must not read, copy, use or > > disclose them without authorisation. If you are not an > > intended recipient, please contact us at once by return > > email and then delete both messages and all attachments. > > > > > > > > > -------------------------------------------------------------- > ---------- > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > -- > > Cheers > John > > -- spambait > 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu > > Please do not reply off-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. From debian at herakles.homelinux.org Wed May 16 01:05:10 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Wed, 16 May 2007 09:05:10 +0800 Subject: Unable to get any http requests: Unable toretreive netstg2.imgfile In-Reply-To: <005401c79753$0fcbc240$6d1b42ab@stanford.edu> References: <005401c79753$0fcbc240$6d1b42ab@stanford.edu> Message-ID: <464A58C6.7030801@herakles.homelinux.org> Mukarram Syed wrote: > John, > Sorry...just to make things clear. > eth0 works to get the dhcp/tftp request and eth1 does not. However, I tried > using both is what I meant. > > We have a problem with dhcp, the networking folks here did not enable > "spanning tree" on the switch ports, until that happens I am using static > ip. I double checked the syntax etc. I don't understand... pxelinux works and can load the kernel/initrd, and the kernel/initrd doesn't work. Is that what you said? How does the BIOS know to load pxelinux, and where to get it? Doesn't that require dhcp (or at least bootp)? How does pxelinux get its menus and then the kernel? Doesn't that require dhcp (or at least bootp) too? And then the kernel/initrd can't get the IP address the same way. Have I got the facts straight? > > Thanks > -Mukarram. > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John Summerfield > Sent: Tuesday, May 15, 2007 5:32 PM > To: Discussion list about Kickstart > Subject: Re: Unable to get any http requests: Unable toretreive > netstg2.imgfile > > Mukarram Syed wrote: >> Thanks Colin for the quick response >> >> >> >> Yes. It does have. It's a Dell PE2950. >> >> >> >> I have already tried that option as well. >> >> I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. >> > > If you can connect all network interfaces, that might help, because it > then doesn't matter whether either one or both is used. > > If you run arpwatch that will report on which interfaces are used (by > mac address and IP). > > More below... > >> >> >> -Mukarram. >> >> >> >> >> >> _____ >> >> From: kickstart-list-bounces at redhat.com >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. > (Unix >> Engineer) >> Sent: Tuesday, May 15, 2007 4:52 PM >> To: Discussion list about Kickstart >> Subject: RE: Unable to get any http requests: Unable to retreive >> netstg2.imgfile >> >> >> >> Does the machine you're on have multiple Ethernet interfaces? If so, what >> you may find is adding 'ksdevice=link' will assist. >> >> >> >> CC >> >> >> >> >> _____ >> >> >> From: kickstart-list-bounces at redhat.com >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed >> Sent: Wednesday, 16 May 2007 7:50 AM >> To: kickstart-list at redhat.com >> Subject: Unable to get any http requests: Unable to retreive netstg2.img >> file >> >> Hi >> >> I am sure there is some information regarding my request, but I am unable > to >> find it, hence the post. >> >> >> >> I configured my kickstart server using http. >> >> I am able to pxeboot the client, but my client is not able to get the >> netstg2.img file from the server. >> >> I am able to http into my netstg2.img directory from a web browser from >> another server in the same subnet successfully. >> >> >> >> Also I tried to put in a boot cdrom and run the following: >> >> Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg ip= >> netmask=255.255.255.224 gateway= > > > I have never used static IP for ks, I alway configure everything I can > with dhcp. However, I do know that that's not the format the kernel > expects for the ip parameter. Check your documentation, if I were > writing a script tp parse ip= then I'd conform to the existing standards > (that is, the way the kernel expects it), and that's what works with Ubuntu. > > > >> >> >> Still no luck. It goes straight to the cd installer. >> >> >> >> I tried to tcpdump the traffic on the MAC address of the client, it > outputs >> the tftp information and after that it stops. >> >> I don't see any http traffic coming from the client. > > That's fairly consistent with you specifying ip= incorrectly. > > btw, if you're using PXE, the why are you trying to spectify static IP info? > > btw, as I recall, pxelinux has an option to pass on the IP address info > it gets in the proper format for the kernel to parse. > > >> >> >> I'd appreciate some help in this regard. >> >> I've spent all day trying to troubleshoot this but no luck. >> >> >> >> Thanks Much >> >> >> >> >> >> Mukarram Syed >> >> Unix Systems Administrator, >> >> Stanford University. >> >> >> >> >> NOTICE: This email and any attachments are confidential. >> They may contain legally privileged information or >> copyright material. You must not read, copy, use or >> disclose them without authorisation. If you are not an >> intended recipient, please contact us at once by return >> email and then delete both messages and all attachments. >> >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list > > -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From muksyed at stanford.edu Wed May 16 02:24:52 2007 From: muksyed at stanford.edu (muksyed at stanford.edu) Date: Tue, 15 May 2007 19:24:52 -0700 Subject: Unable to get any http requests: Unable toretreive netstg2.imgfile In-Reply-To: <464A58C6.7030801@herakles.homelinux.org> References: <005401c79753$0fcbc240$6d1b42ab@stanford.edu> <464A58C6.7030801@herakles.homelinux.org> Message-ID: <20070515192452.1e1ofj8em7b400gk@webmail.stanford.edu> Quoting John Summerfield : > Mukarram Syed wrote: >> John, >> Sorry...just to make things clear. >> eth0 works to get the dhcp/tftp request and eth1 does not. However, I tried >> using both is what I meant. >> >> We have a problem with dhcp, the networking folks here did not enable >> "spanning tree" on the switch ports, until that happens I am using static >> ip. I double checked the syntax etc. > > I don't understand... pxelinux works and can load the kernel/initrd, > and the kernel/initrd doesn't work. Is that what you said? Sorry for the confusion John. pxelinux works and it loads the vmlinuz and initrd. > > How does the BIOS know to load pxelinux, and where to get it? Doesn't > that require dhcp (or at least bootp)? > > How does pxelinux get its menus and then the kernel? Doesn't that > require dhcp (or at least bootp) too? > > And then the kernel/initrd can't get the IP address the same way. Have > I got the facts straight? > Actually I have dhcpd configured for static IP address. Not for a range of IP addresses to pick from. dhcpd is running. After PXE loads up the kernel, I think the installer anaconda uses dhcp to again look for the ks.cfg file. This is after the menu. That's the part which does not work. That's why I put in the static IP's in my default file under label 1. Putting the static IP's don't work as well. Thanks -Mukarram. > > >> >> Thanks >> -Mukarram. >> >> -----Original Message----- >> From: kickstart-list-bounces at redhat.com >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John Summerfield >> Sent: Tuesday, May 15, 2007 5:32 PM >> To: Discussion list about Kickstart >> Subject: Re: Unable to get any http requests: Unable toretreive >> netstg2.imgfile >> >> Mukarram Syed wrote: >>> Thanks Colin for the quick response Yes. It does have. It's a >>> Dell PE2950. >>> >>> I have already tried that option as well. I tried both >>> ksdevice=eth0 and ksdevice=eth1 and they don't work. >>> >> >> If you can connect all network interfaces, that might help, because >> it then doesn't matter whether either one or both is used. >> >> If you run arpwatch that will report on which interfaces are used >> (by mac address and IP). >> >> More below... >> >>> -Mukarram. >>> >>> _____ From: kickstart-list-bounces at redhat.com >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. >> (Unix >>> Engineer) >>> Sent: Tuesday, May 15, 2007 4:52 PM >>> To: Discussion list about Kickstart >>> Subject: RE: Unable to get any http requests: Unable to retreive >>> netstg2.imgfile >>> >>> Does the machine you're on have multiple Ethernet interfaces? If so, what >>> you may find is adding 'ksdevice=link' will assist. >>> >>> CC >>> >>> _____ From: kickstart-list-bounces at redhat.com >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed >>> Sent: Wednesday, 16 May 2007 7:50 AM >>> To: kickstart-list at redhat.com >>> Subject: Unable to get any http requests: Unable to retreive netstg2.img >>> file >>> >>> Hi I am sure there is some information regarding my request, but I >>> am unable >> to >>> find it, hence the post. >>> >>> I configured my kickstart server using http. >>> >>> I am able to pxeboot the client, but my client is not able to get the >>> netstg2.img file from the server. >>> >>> I am able to http into my netstg2.img directory from a web browser from >>> another server in the same subnet successfully. >>> >>> Also I tried to put in a boot cdrom and run the following: >>> >>> Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg ip= >>> netmask=255.255.255.224 gateway= >> >> >> I have never used static IP for ks, I alway configure everything I >> can with dhcp. However, I do know that that's not the format the >> kernel expects for the ip parameter. Check your documentation, if I >> were writing a script tp parse ip= then I'd conform to the >> existing standards (that is, the way the kernel expects it), and >> that's what works with Ubuntu. >> >> >> >>> Still no luck. It goes straight to the cd installer. >>> >>> I tried to tcpdump the traffic on the MAC address of the client, it >> outputs >>> the tftp information and after that it stops. >>> >>> I don't see any http traffic coming from the client. >> >> That's fairly consistent with you specifying ip= incorrectly. >> >> btw, if you're using PXE, the why are you trying to spectify static IP info? >> >> btw, as I recall, pxelinux has an option to pass on the IP address >> info it gets in the proper format for the kernel to parse. >> >> >>> I'd appreciate some help in this regard. I've spent all day >>> trying to troubleshoot this but no luck. >>> >>> Thanks Much >>> >>> Mukarram Syed >>> >>> Unix Systems Administrator, Stanford University. >>> >>> NOTICE: This email and any attachments are confidential. They may >>> contain legally privileged information or copyright material. You >>> must not read, copy, use or disclose them without authorisation. >>> If you are not an intended recipient, please contact us at once by >>> return email and then delete both messages and all attachments. >>> >>> >>> >>> ------------------------------------------------------------------------ >>> >>> _______________________________________________ >>> Kickstart-list mailing list >>> Kickstart-list at redhat.com >>> https://www.redhat.com/mailman/listinfo/kickstart-list >> >> > > > -- > > Cheers > John > > -- spambait > 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu > > Please do not reply off-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From muksyed at stanford.edu Wed May 16 03:39:38 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Tue, 15 May 2007 20:39:38 -0700 Subject: Unable to get any http requests:Unable to retreive netstg2.img file In-Reply-To: Message-ID: <005501c7976b$d4ed6110$6d1b42ab@stanford.edu> Colin, I changed it from ksdevice=eth0 to ksdevice=link Sorry, that didn't help. Wonder what else could it be, why is it not connecting to the http server? I don't see any traffic in tcpdump after pxeboot. -Mukarram. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: Tuesday, May 15, 2007 5:56 PM To: Discussion list about Kickstart Subject: RE: Unable to get any http requests:Unabletoretreive netstg2.imgfile Try changing 'ksdevice=eth0' to 'ksdevice=link'. With 2.6 kernels, the kernel's idea of what eth0 and eth1 are often differs from the BIOS/PXE stack. CC > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed > Sent: Wednesday, 16 May 2007 8:42 AM > To: 'Discussion list about Kickstart' > Subject: RE: Unable to get any http requests: > Unabletoretreive netstg2.imgfile > > John, > Sorry...just to make things clear. > eth0 works to get the dhcp/tftp request and eth1 does not. > However, I tried using both is what I meant. > > We have a problem with dhcp, the networking folks here did > not enable "spanning tree" on the switch ports, until that > happens I am using static > ip. I double checked the syntax etc. > > Thanks > -Mukarram. > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John > Summerfield > Sent: Tuesday, May 15, 2007 5:32 PM > To: Discussion list about Kickstart > Subject: Re: Unable to get any http requests: Unable > toretreive netstg2.imgfile > > Mukarram Syed wrote: > > Thanks Colin for the quick response > > > > > > > > Yes. It does have. It's a Dell PE2950. > > > > > > > > I have already tried that option as well. > > > > I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. > > > > If you can connect all network interfaces, that might help, > because it then doesn't matter whether either one or both is used. > > If you run arpwatch that will report on which interfaces are > used (by mac address and IP). > > More below... > > > > > > > -Mukarram. > > > > > > > > > > > > _____ > > > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > Coe, Colin C. > (Unix > > Engineer) > > Sent: Tuesday, May 15, 2007 4:52 PM > > To: Discussion list about Kickstart > > Subject: RE: Unable to get any http requests: Unable to retreive > > netstg2.imgfile > > > > > > > > Does the machine you're on have multiple Ethernet > interfaces? If so, > > what you may find is adding 'ksdevice=link' will assist. > > > > > > > > CC > > > > > > > > > > _____ > > > > > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > Mukarram Syed > > Sent: Wednesday, 16 May 2007 7:50 AM > > To: kickstart-list at redhat.com > > Subject: Unable to get any http requests: Unable to retreive > > netstg2.img file > > > > Hi > > > > I am sure there is some information regarding my request, but I am > > unable > to > > find it, hence the post. > > > > > > > > I configured my kickstart server using http. > > > > I am able to pxeboot the client, but my client is not able > to get the > > netstg2.img file from the server. > > > > I am able to http into my netstg2.img directory from a web browser > > from another server in the same subnet successfully. > > > > > > > > Also I tried to put in a boot cdrom and run the following: > > > > Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg > > ip= > > netmask=255.255.255.224 gateway= > > > I have never used static IP for ks, I alway configure > everything I can with dhcp. However, I do know that that's > not the format the kernel expects for the ip parameter. Check > your documentation, if I were writing a script tp parse ip= > then I'd conform to the existing standards (that is, the way > the kernel expects it), and that's what works with Ubuntu. > > > > > > > > > > > Still no luck. It goes straight to the cd installer. > > > > > > > > I tried to tcpdump the traffic on the MAC address of the client, it > outputs > > the tftp information and after that it stops. > > > > I don't see any http traffic coming from the client. > > That's fairly consistent with you specifying ip= incorrectly. > > btw, if you're using PXE, the why are you trying to spectify > static IP info? > > btw, as I recall, pxelinux has an option to pass on the IP > address info it gets in the proper format for the kernel to parse. > > > > > > > > > > I'd appreciate some help in this regard. > > > > I've spent all day trying to troubleshoot this but no luck. > > > > > > > > Thanks Much > > > > > > > > > > > > Mukarram Syed > > > > Unix Systems Administrator, > > > > Stanford University. > > > > > > > > > > NOTICE: This email and any attachments are confidential. > > They may contain legally privileged information or > > copyright material. You must not read, copy, use or > > disclose them without authorisation. If you are not an > > intended recipient, please contact us at once by return > > email and then delete both messages and all attachments. > > > > > > > > > -------------------------------------------------------------- > ---------- > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > -- > > Cheers > John > > -- spambait > 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu > > Please do not reply off-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From mjenning at rackspace.com Wed May 16 03:47:11 2007 From: mjenning at rackspace.com (Mike Jennings) Date: Tue, 15 May 2007 22:47:11 -0500 Subject: Unable to get any http requests:Unable to retreive netstg2.img file In-Reply-To: <005501c7976b$d4ed6110$6d1b42ab@stanford.edu> References: <005501c7976b$d4ed6110$6d1b42ab@stanford.edu> Message-ID: <63087C0F-8A83-4B6F-BC15-7068100C7AFC@rackspace.com> Is the apache log showing any 404 errors? My stupid screwups are always shown in the error log when I typo something. Mike On May 15, 2007, at 10:39 PM, Mukarram Syed wrote: > > Colin, > I changed it from ksdevice=eth0 to ksdevice=link > > Sorry, that didn't help. > > Wonder what else could it be, why is it not connecting to the http > server? > I don't see any traffic in tcpdump after pxeboot. > > -Mukarram. > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin > C. (Unix > Engineer) > Sent: Tuesday, May 15, 2007 5:56 PM > To: Discussion list about Kickstart > Subject: RE: Unable to get any http requests:Unabletoretreive > netstg2.imgfile > > > Try changing 'ksdevice=eth0' to 'ksdevice=link'. With 2.6 kernels, > the > kernel's idea of what eth0 and eth1 are often differs from the BIOS/ > PXE > stack. > > CC > >> -----Original Message----- >> From: kickstart-list-bounces at redhat.com >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed >> Sent: Wednesday, 16 May 2007 8:42 AM >> To: 'Discussion list about Kickstart' >> Subject: RE: Unable to get any http requests: >> Unabletoretreive netstg2.imgfile >> >> John, >> Sorry...just to make things clear. >> eth0 works to get the dhcp/tftp request and eth1 does not. >> However, I tried using both is what I meant. >> >> We have a problem with dhcp, the networking folks here did >> not enable "spanning tree" on the switch ports, until that >> happens I am using static >> ip. I double checked the syntax etc. >> >> Thanks >> -Mukarram. >> >> -----Original Message----- >> From: kickstart-list-bounces at redhat.com >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John >> Summerfield >> Sent: Tuesday, May 15, 2007 5:32 PM >> To: Discussion list about Kickstart >> Subject: Re: Unable to get any http requests: Unable >> toretreive netstg2.imgfile >> >> Mukarram Syed wrote: >>> Thanks Colin for the quick response >>> >>> >>> >>> Yes. It does have. It's a Dell PE2950. >>> >>> >>> >>> I have already tried that option as well. >>> >>> I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. >>> >> >> If you can connect all network interfaces, that might help, >> because it then doesn't matter whether either one or both is used. >> >> If you run arpwatch that will report on which interfaces are >> used (by mac address and IP). >> >> More below... >> >>> >>> >>> -Mukarram. >>> >>> >>> >>> >>> >>> _____ >>> >>> From: kickstart-list-bounces at redhat.com >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of >> Coe, Colin C. >> (Unix >>> Engineer) >>> Sent: Tuesday, May 15, 2007 4:52 PM >>> To: Discussion list about Kickstart >>> Subject: RE: Unable to get any http requests: Unable to retreive >>> netstg2.imgfile >>> >>> >>> >>> Does the machine you're on have multiple Ethernet >> interfaces? If so, >>> what you may find is adding 'ksdevice=link' will assist. >>> >>> >>> >>> CC >>> >>> >>> >>> >>> _____ >>> >>> >>> From: kickstart-list-bounces at redhat.com >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of >> Mukarram Syed >>> Sent: Wednesday, 16 May 2007 7:50 AM >>> To: kickstart-list at redhat.com >>> Subject: Unable to get any http requests: Unable to retreive >>> netstg2.img file >>> >>> Hi >>> >>> I am sure there is some information regarding my request, but I am >>> unable >> to >>> find it, hence the post. >>> >>> >>> >>> I configured my kickstart server using http. >>> >>> I am able to pxeboot the client, but my client is not able >> to get the >>> netstg2.img file from the server. >>> >>> I am able to http into my netstg2.img directory from a web browser >>> from another server in the same subnet successfully. >>> >>> >>> >>> Also I tried to put in a boot cdrom and run the following: >>> >>> Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg >>> ip= >>> netmask=255.255.255.224 gateway= >> >> >> I have never used static IP for ks, I alway configure >> everything I can with dhcp. However, I do know that that's >> not the format the kernel expects for the ip parameter. Check >> your documentation, if I were writing a script tp parse ip= >> then I'd conform to the existing standards (that is, the way >> the kernel expects it), and that's what works with Ubuntu. >> >> >> >>> >>> >>> >>> Still no luck. It goes straight to the cd installer. >>> >>> >>> >>> I tried to tcpdump the traffic on the MAC address of the client, it >> outputs >>> the tftp information and after that it stops. >>> >>> I don't see any http traffic coming from the client. >> >> That's fairly consistent with you specifying ip= incorrectly. >> >> btw, if you're using PXE, the why are you trying to spectify >> static IP info? >> >> btw, as I recall, pxelinux has an option to pass on the IP >> address info it gets in the proper format for the kernel to parse. >> >> >>> >>> >>> >>> I'd appreciate some help in this regard. >>> >>> I've spent all day trying to troubleshoot this but no luck. >>> >>> >>> >>> Thanks Much >>> >>> >>> >>> >>> >>> Mukarram Syed >>> >>> Unix Systems Administrator, >>> >>> Stanford University. >>> >>> >>> >>> >>> NOTICE: This email and any attachments are confidential. >>> They may contain legally privileged information or >>> copyright material. You must not read, copy, use or >>> disclose them without authorisation. If you are not an >>> intended recipient, please contact us at once by return >>> email and then delete both messages and all attachments. >>> >>> >>> >>> >> -------------------------------------------------------------- >> ---------- >>> >>> _______________________________________________ >>> Kickstart-list mailing list >>> Kickstart-list at redhat.com >>> https://www.redhat.com/mailman/listinfo/kickstart-list >> >> >> -- >> >> Cheers >> John >> >> -- spambait >> 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu >> >> Please do not reply off-list >> >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list >> >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list >> > > NOTICE: This email and any attachments are confidential. > They may contain legally privileged information or > copyright material. You must not read, copy, use or > disclose them without authorisation. If you are not an > intended recipient, please contact us at once by return > email and then delete both messages and all attachments. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From jce at zot.com Wed May 16 05:27:22 2007 From: jce at zot.com (Chris Edillon) Date: Wed, 16 May 2007 01:27:22 -0400 Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: References: Message-ID: <464A963A.5030702@zot.com> Martin Steinmann wrote: >>> Is there a similarly elegant way to add a driver disk image? I am > still >>> struggling embedding an additional RAID driver into the boot CD. I >>> suppose I need to update the initial initrd file (using mkinitrd) > and >>> then tell Anaconda to load the driver image during stage2. Can the >>> driver disk image be placed somewhere into the tree similar to the >>> updates.img? >> >> Yes. >> >> Look for the driverdisk option at >> https://www.redhat.com/docs/manuals/enterprise/RHEL-5- >> manual/Installation_Guide-en-US/s1-kickstart2-options.html >> >> That option is in RHEL4 & 5 that I know of. > > Yes, I looked at this and tried several options. The problem is that I > do not have a network during installation (no ftp, http or nfs > available) and therefore the driver disk image needs to be in the tree > on the CD. Also, the HD is empty when I start. > are you adding a new driver or replacing an existing one? you can place an updated driver in the initrd.img, although not with mkinitrd because the resultant ramdisk images are vastly different (the kickstart image is more complex, a standard initrd.img just has drivers and not much else). the steps are: 1) uncompress initial ramdisk image to a temporary file and do a loopback mount of the file: # gzip -dc initrd.img > /tmp/initrd.tmp # mount -o rw,loop /tmp/initrd.tmp /mnt 2) uncompress and unarchive the modules bundle in the initrd: # cd /tmp # gzip -dc /mnt/modules/modules.cgz | cpio -idv you'll end up with a directory in /tmp named after the boot kernel, e.g. /tmp/2.6.9-34.EL or similar. 3) drop your driver module in the module tree created above: # cp your_module.ko /tmp/2.6.9-34.EL/ where is probably i686 or x86_64 based on your OS architecture (32- or 64-bit). 4) overwrite the existing module archive with your new modules: # cd /tmp # find 2.6.9-34.EL | cpio -ov -H crc | gzip -c > /mnt/modules/modules.cgz (that should all be on one line, if my email client wrapped it across two) 5) unmount and compress the temp initrd: # umount /mnt # gzip -c /tmp/initrd.tmp > /tmp/initrd-new.img i haven't tried adding previously nonexisting modules to the initial ramdisk before, but it should be identical to the above except for an added step where you edit /mnt/modules/modules.pcimap and /mnt/modules/pcitable to put the correct entries in for the new module in question. if you have a driver disk hopefully you'll also have the pci information buried somewhere in it; otherwise, i have no idea how to generate that data. :) i've done this numerous times for RHEL3 and RHEL4, and i imagine it should be no different for RHEL5. if you update the initrd.img you shouldn't need to update the stage2.img image. hope this is useful, chris From jce at zot.com Wed May 16 06:02:14 2007 From: jce at zot.com (Chris Edillon) Date: Wed, 16 May 2007 02:02:14 -0400 Subject: Kickstart via PXE/NFS In-Reply-To: References: Message-ID: <464A9E66.9000603@zot.com> SSTinsley at upsfreight.com wrote: > I have now tried the install by using DHCP and fixed IP address. I would > hope that > assigning the IP with kernel arguments through the pxelinux cfg would > work. Assuming > it does, the only other common error is related to DNS. I do not have > DNS available > on this private install net. Does Anaconda require DNS for a completely > hands-off > install? > dns should only be required if you use hostnames instead of ip addresses when referencing your ks.cfg and media locations. that being said, i've seen problems where i used ip addreses, dns was available and configured for the kickstart environment by DHCP, and the reverse entry for the ip address did not match the forward address. in that case, the kickstart barfed, but i remember it clearly indicated that it had made a reverse lookup successfully but that there was a mismatch. as others have mentioned, i've seen the portfast problem come up before with cisco equipment as well, so definitely check with your network guys. HTH, chris From kanarip at kanarip.com Wed May 16 08:30:44 2007 From: kanarip at kanarip.com (Jeroen van Meeuwen) Date: Wed, 16 May 2007 10:30:44 +0200 Subject: Location of installation media no longer /tmp/cdrom? In-Reply-To: <963676.55991.qm@web32504.mail.mud.yahoo.com> References: <963676.55991.qm@web32504.mail.mud.yahoo.com> Message-ID: <464AC134.6070205@kanarip.com> Pat wrote: > However I get the following error during the install: > > mount: Mounting /tmp/cdrom on /mnt/source failed: No > such file or directory > > Are you sure the /mnt/source directory exists? -kanarip From muksyed at stanford.edu Wed May 16 08:38:30 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Wed, 16 May 2007 01:38:30 -0700 Subject: Unable to get any http requests:Unable to retreive netstg2.imgfile In-Reply-To: <63087C0F-8A83-4B6F-BC15-7068100C7AFC@rackspace.com> Message-ID: <005c01c79795$956fc580$6d1b42ab@stanford.edu> Nope. Nothing in the error logs. In my troubleshooting, this is what I've found... When I remove the ks=http:////ks.cfg from my "default" file Client PXE boots, shows menu and loads vmlinuz/initrd. Then it shows up the Language selection. No luck. However, when I change my default file to read: ks=//ks.cfg, Client PXE boots, shows menu and loads vmlinuz/initrd. Then it gives the error "Error opening kickstart file (null): Bad address" I do some more research (see link below) http://linux.derkeiler.com/Mailing-Lists/Fedora/2004-06/5939.html and modified the ks.cfg file...add more network parameters like so: network --device=eth0 --bootproto=static --ip=171.67.39.139 --netmask=255.255.255.224 --gateway=171.67.39.129 --nameserver=171.64.7.77 --hostname=as-test.stanford.edu Still no luck. I then changed my --device=eth0 to --device=link and then to --device=eth1...still no luck. Then put it back to --device=eth0. Then I change my PATH from: url --url http://171.67.39.130/ENT_LINUX_64/Enterprise to url --url http://171.67.39.130/ENT_LINUX_64 Still no luck. Again nothing in the http logs. It looks like the installer is reading the ks.cfg file silently since it complained that it couldn't find it from the default file when I didn't specify the http://. Can I figure out what is reading and stopping. Is there a way to look at the kickstart logs (If there is any...I haven't seen one). I tried to create logging as well in my ks.cfg file...no luck...no output: %pre --log=/tmp/my-pre-log echo 'Hello, World!' Here is my directory/file setup: TFTP: /tftpboot/linux-install/pxelinux.cfg [root at as-kickball pxelinux.cfg]# ls default pxeos.xml [root at as-kickball linux-install]# ls -lR .: total 32 drwxr-xr-x 2 root root 4096 May 14 00:31 ENT_LINUX_32 drwxr-xr-x 2 root root 4096 May 16 00:30 ENT_LINUX_64 drwxr-xr-x 2 root root 4096 May 14 21:09 msgs -rwxr-xr-x 1 root root 12952 Oct 24 2006 pxelinux.0 drwxr-xr-x 2 root root 4096 May 16 00:25 pxelinux.cfg ./ENT_LINUX_32: total 0 ./ENT_LINUX_64: total 5984 -rwxr-xr-x 1 root root 4310479 May 14 15:29 initrd.img -rwxr-xr-x 1 root root 85 May 16 00:30 ks.cfg -rwxr-xr-x 1 root root 1792835 May 14 15:29 vmlinuz ./msgs: total 24 -rwxr-xr-x 1 root root 534 May 14 21:00 boot.msg -rwxr-xr-x 1 root root 668 Oct 24 2006 expert.msg -rwxr-xr-x 1 root root 871 Oct 24 2006 general.msg -rwxr-xr-x 1 root root 860 Oct 24 2006 param.msg -rwxr-xr-x 1 root root 530 Oct 24 2006 rescue.msg -rwxr-xr-x 1 root root 545 Oct 24 2006 snake.msg ./pxelinux.cfg: total 20 -rwxr-xr-x 1 root root 511 May 16 00:25 default -rwxr-xr-x 1 root root 511 May 15 23:34 default.bak -rwxr-xr-x 1 root root 461 May 15 14:44 default.dhcp -rwxr-xr-x 1 root root 573 May 15 13:30 default.staticIP -rwxr-xr-x 1 root root 42 May 14 16:28 pxeos.xml My HTTP directory: [root at as-kickball /]# ls -lart /var/www/html total 24 drwxr-xr-x 8 root root 4096 Mar 21 19:17 .. drwxr-xr-x 4 root root 4096 May 14 00:35 . drwxr-xr-x 3 root root 4096 May 14 00:40 ENT_LINUX_32 drwxr-xr-x 3 root root 4096 May 15 15:49 ENT_LINUX_64 [root at as-kickball /]# ls -lart /var/www/html/ENT_LINUX_64/Enterprise/ total 136 drwxr-xr-x 2 root root 114688 Feb 24 13:15 RPMS drwxr-xr-x 2 root root 4096 May 14 15:53 base -rwxr-xr-x 1 root root 1386 May 15 15:05 ks_ent_linux_64.cfg drwxr-xr-x 3 root root 4096 May 15 15:49 .. -rw-r--r-- 1 root root 1242 May 16 01:08 ks.cfg drwxr-xr-x 4 root root 4096 May 16 01:08 . I am at a loss now. I am sure it's something simple as it mostly is. Help is appreciated. Tomorrow, I'll tell my networking guy to enable "spanning tree" so I could use dhcp and get the static crap out of my way. Will let you know more. -Mukarram. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mike Jennings Sent: Tuesday, May 15, 2007 8:47 PM To: Discussion list about Kickstart Subject: Re: Unable to get any http requests:Unable to retreive netstg2.imgfile Is the apache log showing any 404 errors? My stupid screwups are always shown in the error log when I typo something. Mike On May 15, 2007, at 10:39 PM, Mukarram Syed wrote: > > Colin, > I changed it from ksdevice=eth0 to ksdevice=link > > Sorry, that didn't help. > > Wonder what else could it be, why is it not connecting to the http > server? > I don't see any traffic in tcpdump after pxeboot. > > -Mukarram. > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin > C. (Unix > Engineer) > Sent: Tuesday, May 15, 2007 5:56 PM > To: Discussion list about Kickstart > Subject: RE: Unable to get any http requests:Unabletoretreive > netstg2.imgfile > > > Try changing 'ksdevice=eth0' to 'ksdevice=link'. With 2.6 kernels, > the > kernel's idea of what eth0 and eth1 are often differs from the BIOS/ > PXE > stack. > > CC > >> -----Original Message----- >> From: kickstart-list-bounces at redhat.com >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed >> Sent: Wednesday, 16 May 2007 8:42 AM >> To: 'Discussion list about Kickstart' >> Subject: RE: Unable to get any http requests: >> Unabletoretreive netstg2.imgfile >> >> John, >> Sorry...just to make things clear. >> eth0 works to get the dhcp/tftp request and eth1 does not. >> However, I tried using both is what I meant. >> >> We have a problem with dhcp, the networking folks here did >> not enable "spanning tree" on the switch ports, until that >> happens I am using static >> ip. I double checked the syntax etc. >> >> Thanks >> -Mukarram. >> >> -----Original Message----- >> From: kickstart-list-bounces at redhat.com >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John >> Summerfield >> Sent: Tuesday, May 15, 2007 5:32 PM >> To: Discussion list about Kickstart >> Subject: Re: Unable to get any http requests: Unable >> toretreive netstg2.imgfile >> >> Mukarram Syed wrote: >>> Thanks Colin for the quick response >>> >>> >>> >>> Yes. It does have. It's a Dell PE2950. >>> >>> >>> >>> I have already tried that option as well. >>> >>> I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. >>> >> >> If you can connect all network interfaces, that might help, >> because it then doesn't matter whether either one or both is used. >> >> If you run arpwatch that will report on which interfaces are >> used (by mac address and IP). >> >> More below... >> >>> >>> >>> -Mukarram. >>> >>> >>> >>> >>> >>> _____ >>> >>> From: kickstart-list-bounces at redhat.com >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of >> Coe, Colin C. >> (Unix >>> Engineer) >>> Sent: Tuesday, May 15, 2007 4:52 PM >>> To: Discussion list about Kickstart >>> Subject: RE: Unable to get any http requests: Unable to retreive >>> netstg2.imgfile >>> >>> >>> >>> Does the machine you're on have multiple Ethernet >> interfaces? If so, >>> what you may find is adding 'ksdevice=link' will assist. >>> >>> >>> >>> CC >>> >>> >>> >>> >>> _____ >>> >>> >>> From: kickstart-list-bounces at redhat.com >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of >> Mukarram Syed >>> Sent: Wednesday, 16 May 2007 7:50 AM >>> To: kickstart-list at redhat.com >>> Subject: Unable to get any http requests: Unable to retreive >>> netstg2.img file >>> >>> Hi >>> >>> I am sure there is some information regarding my request, but I am >>> unable >> to >>> find it, hence the post. >>> >>> >>> >>> I configured my kickstart server using http. >>> >>> I am able to pxeboot the client, but my client is not able >> to get the >>> netstg2.img file from the server. >>> >>> I am able to http into my netstg2.img directory from a web browser >>> from another server in the same subnet successfully. >>> >>> >>> >>> Also I tried to put in a boot cdrom and run the following: >>> >>> Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg >>> ip= >>> netmask=255.255.255.224 gateway= >> >> >> I have never used static IP for ks, I alway configure >> everything I can with dhcp. However, I do know that that's >> not the format the kernel expects for the ip parameter. Check >> your documentation, if I were writing a script tp parse ip= >> then I'd conform to the existing standards (that is, the way >> the kernel expects it), and that's what works with Ubuntu. >> >> >> >>> >>> >>> >>> Still no luck. It goes straight to the cd installer. >>> >>> >>> >>> I tried to tcpdump the traffic on the MAC address of the client, it >> outputs >>> the tftp information and after that it stops. >>> >>> I don't see any http traffic coming from the client. >> >> That's fairly consistent with you specifying ip= incorrectly. >> >> btw, if you're using PXE, the why are you trying to spectify >> static IP info? >> >> btw, as I recall, pxelinux has an option to pass on the IP >> address info it gets in the proper format for the kernel to parse. >> >> >>> >>> >>> >>> I'd appreciate some help in this regard. >>> >>> I've spent all day trying to troubleshoot this but no luck. >>> >>> >>> >>> Thanks Much >>> >>> >>> >>> >>> >>> Mukarram Syed >>> >>> Unix Systems Administrator, >>> >>> Stanford University. >>> >>> >>> >>> >>> NOTICE: This email and any attachments are confidential. >>> They may contain legally privileged information or >>> copyright material. You must not read, copy, use or >>> disclose them without authorisation. If you are not an >>> intended recipient, please contact us at once by return >>> email and then delete both messages and all attachments. >>> >>> >>> >>> >> -------------------------------------------------------------- >> ---------- >>> >>> _______________________________________________ >>> Kickstart-list mailing list >>> Kickstart-list at redhat.com >>> https://www.redhat.com/mailman/listinfo/kickstart-list >> >> >> -- >> >> Cheers >> John >> >> -- spambait >> 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu >> >> Please do not reply off-list >> >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list >> >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list >> > > NOTICE: This email and any attachments are confidential. > They may contain legally privileged information or > copyright material. You must not read, copy, use or > disclose them without authorisation. If you are not an > intended recipient, please contact us at once by return > email and then delete both messages and all attachments. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Eric.KENNEDYvanDAM at rvponp.fgov.be Wed May 16 08:46:19 2007 From: Eric.KENNEDYvanDAM at rvponp.fgov.be (KENNEDY VAN DAM Eric) Date: Wed, 16 May 2007 10:46:19 +0200 Subject: Unable to get any http requests:Unable to retreive netstg2.imgfile References: <005c01c79795$956fc580$6d1b42ab@stanford.edu> Message-ID: <2F4F96E90349F241809293630A893B3D02C50BC3@MAIL03.rvponp.fgov.be> > -----Message d'origine----- > De : kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] De la part de Mukarram Syed > Envoy? : mercredi 16 mai 2007 10:39 > ? : 'Discussion list about Kickstart' > Objet : RE: Unable to get any http requests:Unable to > retreive netstg2.imgfile > > Nope. Nothing in the error logs. > > In my troubleshooting, this is what I've found... > > When I remove the ks=http:////ks.cfg from my > "default" file > Client PXE boots, shows menu and loads vmlinuz/initrd. > Then it shows up the Language selection. No luck. > However, when I change my default file to read: ks=//ks.cfg, > Client PXE boots, shows menu and loads vmlinuz/initrd. > Then it gives the error "Error opening kickstart file (null): > Bad address" > I do some more research (see link below) > http://linux.derkeiler.com/Mailing-Lists/Fedora/2004-06/5939.html > and modified the ks.cfg file...add more network parameters like so: > network --device=eth0 --bootproto=static --ip=171.67.39.139 > --netmask=255.255.255.224 --gateway=171.67.39.129 > --nameserver=171.64.7.77 > --hostname=as-test.stanford.edu > Still no luck. > I then changed my --device=eth0 to --device=link and then to > --device=eth1...still no luck. Then put it back to --device=eth0. > Then I change my PATH from: > url --url http://171.67.39.130/ENT_LINUX_64/Enterprise > to > url --url http://171.67.39.130/ENT_LINUX_64 > Still no luck. > Again nothing in the http logs. > > It looks like the installer is reading the ks.cfg file > silently since it > complained that it couldn't find it from the default file > when I didn't > specify the http://. > Can I figure out what is reading and stopping. Is there a > way to look at > the kickstart logs (If there is any...I haven't seen one). > > I tried to create logging as well in my ks.cfg file...no > luck...no output: > %pre --log=/tmp/my-pre-log > echo 'Hello, World!' Sorry if the suggestion has already been made. Did you try to go to the other consoles ? There is a lot of information there which can help to debu your problem. From SSTinsley at upsfreight.com Wed May 16 13:00:13 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Wed, 16 May 2007 09:00:13 -0400 Subject: Unable to get any http requests:Unable to retrieve netstg2.imgfile In-Reply-To: <005c01c79795$956fc580$6d1b42ab@stanford.edu> Message-ID: I have been fighting almost the identical problem. See the thread entitled "Re: Kickstart via PXE/NFS" After the kernel is loaded via configuration set up in the pxelinux config file, the network will not come up and the ks.cfg file can then not be loaded. I get the network config screen asking to use DHCP or to enter static info for setup of eth0. It has really bee frustrating. I finally just placed my ks.cfg on CDROM and was able to get the rest of the PXE auto install to complete. If you get a resolution, I will be very interested. I have had a case open with RedHat for 2 days waiting for a response. kickstart-list-bounces at redhat.com wrote on 05/16/2007 04:38:30 AM: > Nope. Nothing in the error logs. > > In my troubleshooting, this is what I've found... > > When I remove the ks=http:////ks.cfg from my "default" file > Client PXE boots, shows menu and loads vmlinuz/initrd. > Then it shows up the Language selection. No luck. > However, when I change my default file to read: ks=//ks.cfg, > Client PXE boots, shows menu and loads vmlinuz/initrd. > Then it gives the error "Error opening kickstart file (null): Bad address" > I do some more research (see link below) > http://linux.derkeiler.com/Mailing-Lists/Fedora/2004-06/5939.html > and modified the ks.cfg file...add more network parameters like so: > network --device=eth0 --bootproto=static --ip=171.67.39.139 > --netmask=255.255.255.224 --gateway=171.67.39.129 --nameserver=171.64.7.77 > --hostname=as-test.stanford.edu > Still no luck. > I then changed my --device=eth0 to --device=link and then to > --device=eth1...still no luck. Then put it back to --device=eth0. > Then I change my PATH from: > url --url http://171.67.39.130/ENT_LINUX_64/Enterprise > to > url --url http://171.67.39.130/ENT_LINUX_64 > Still no luck. > Again nothing in the http logs. > > It looks like the installer is reading the ks.cfg file silently since it > complained that it couldn't find it from the default file when I didn't > specify the http://. > Can I figure out what is reading and stopping. Is there a way to look at > the kickstart logs (If there is any...I haven't seen one). > > I tried to create logging as well in my ks.cfg file...no luck...no output: > %pre --log=/tmp/my-pre-log > echo 'Hello, World!' > > Here is my directory/file setup: > > TFTP: > > /tftpboot/linux-install/pxelinux.cfg > [root at as-kickball pxelinux.cfg]# ls > default pxeos.xml > > [root at as-kickball linux-install]# ls -lR > .: > total 32 > drwxr-xr-x 2 root root 4096 May 14 00:31 ENT_LINUX_32 > drwxr-xr-x 2 root root 4096 May 16 00:30 ENT_LINUX_64 > drwxr-xr-x 2 root root 4096 May 14 21:09 msgs > -rwxr-xr-x 1 root root 12952 Oct 24 2006 pxelinux.0 > drwxr-xr-x 2 root root 4096 May 16 00:25 pxelinux.cfg > > ./ENT_LINUX_32: > total 0 > > ./ENT_LINUX_64: > total 5984 > -rwxr-xr-x 1 root root 4310479 May 14 15:29 initrd.img > -rwxr-xr-x 1 root root 85 May 16 00:30 ks.cfg > -rwxr-xr-x 1 root root 1792835 May 14 15:29 vmlinuz > > ./msgs: > total 24 > -rwxr-xr-x 1 root root 534 May 14 21:00 boot.msg > -rwxr-xr-x 1 root root 668 Oct 24 2006 expert.msg > -rwxr-xr-x 1 root root 871 Oct 24 2006 general.msg > -rwxr-xr-x 1 root root 860 Oct 24 2006 param.msg > -rwxr-xr-x 1 root root 530 Oct 24 2006 rescue.msg > -rwxr-xr-x 1 root root 545 Oct 24 2006 snake.msg > > ./pxelinux.cfg: > total 20 > -rwxr-xr-x 1 root root 511 May 16 00:25 default > -rwxr-xr-x 1 root root 511 May 15 23:34 default.bak > -rwxr-xr-x 1 root root 461 May 15 14:44 default.dhcp > -rwxr-xr-x 1 root root 573 May 15 13:30 default.staticIP > -rwxr-xr-x 1 root root 42 May 14 16:28 pxeos.xml > > My HTTP directory: > > [root at as-kickball /]# ls -lart /var/www/html > total 24 > drwxr-xr-x 8 root root 4096 Mar 21 19:17 .. > drwxr-xr-x 4 root root 4096 May 14 00:35 . > drwxr-xr-x 3 root root 4096 May 14 00:40 ENT_LINUX_32 > drwxr-xr-x 3 root root 4096 May 15 15:49 ENT_LINUX_64 > > [root at as-kickball /]# ls -lart /var/www/html/ENT_LINUX_64/Enterprise/ > total 136 > drwxr-xr-x 2 root root 114688 Feb 24 13:15 RPMS > drwxr-xr-x 2 root root 4096 May 14 15:53 base > -rwxr-xr-x 1 root root 1386 May 15 15:05 ks_ent_linux_64.cfg > drwxr-xr-x 3 root root 4096 May 15 15:49 .. > -rw-r--r-- 1 root root 1242 May 16 01:08 ks.cfg > drwxr-xr-x 4 root root 4096 May 16 01:08 . > > I am at a loss now. I am sure it's something simple as it mostly is. > Help is appreciated. > > Tomorrow, I'll tell my networking guy to enable "spanning tree" so I could > use dhcp and get the static crap out of my way. > Will let you know more. > > -Mukarram. > > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mike Jennings > Sent: Tuesday, May 15, 2007 8:47 PM > To: Discussion list about Kickstart > Subject: Re: Unable to get any http requests:Unable to retreive > netstg2.imgfile > > Is the apache log showing any 404 errors? My stupid screwups are > always shown in the error log when I typo something. > > Mike > > > > On May 15, 2007, at 10:39 PM, Mukarram Syed wrote: > > > > > Colin, > > I changed it from ksdevice=eth0 to ksdevice=link > > > > Sorry, that didn't help. > > > > Wonder what else could it be, why is it not connecting to the http > > server? > > I don't see any traffic in tcpdump after pxeboot. > > > > -Mukarram. > > > > -----Original Message----- > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin > > C. (Unix > > Engineer) > > Sent: Tuesday, May 15, 2007 5:56 PM > > To: Discussion list about Kickstart > > Subject: RE: Unable to get any http requests:Unabletoretreive > > netstg2.imgfile > > > > > > Try changing 'ksdevice=eth0' to 'ksdevice=link'. With 2.6 kernels, > > the > > kernel's idea of what eth0 and eth1 are often differs from the BIOS/ > > PXE > > stack. > > > > CC > > > >> -----Original Message----- > >> From: kickstart-list-bounces at redhat.com > >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed > >> Sent: Wednesday, 16 May 2007 8:42 AM > >> To: 'Discussion list about Kickstart' > >> Subject: RE: Unable to get any http requests: > >> Unabletoretreive netstg2.imgfile > >> > >> John, > >> Sorry...just to make things clear. > >> eth0 works to get the dhcp/tftp request and eth1 does not. > >> However, I tried using both is what I meant. > >> > >> We have a problem with dhcp, the networking folks here did > >> not enable "spanning tree" on the switch ports, until that > >> happens I am using static > >> ip. I double checked the syntax etc. > >> > >> Thanks > >> -Mukarram. > >> > >> -----Original Message----- > >> From: kickstart-list-bounces at redhat.com > >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John > >> Summerfield > >> Sent: Tuesday, May 15, 2007 5:32 PM > >> To: Discussion list about Kickstart > >> Subject: Re: Unable to get any http requests: Unable > >> toretreive netstg2.imgfile > >> > >> Mukarram Syed wrote: > >>> Thanks Colin for the quick response > >>> > >>> > >>> > >>> Yes. It does have. It's a Dell PE2950. > >>> > >>> > >>> > >>> I have already tried that option as well. > >>> > >>> I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. > >>> > >> > >> If you can connect all network interfaces, that might help, > >> because it then doesn't matter whether either one or both is used. > >> > >> If you run arpwatch that will report on which interfaces are > >> used (by mac address and IP). > >> > >> More below... > >> > >>> > >>> > >>> -Mukarram. > >>> > >>> > >>> > >>> > >>> > >>> _____ > >>> > >>> From: kickstart-list-bounces at redhat.com > >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > >> Coe, Colin C. > >> (Unix > >>> Engineer) > >>> Sent: Tuesday, May 15, 2007 4:52 PM > >>> To: Discussion list about Kickstart > >>> Subject: RE: Unable to get any http requests: Unable to retreive > >>> netstg2.imgfile > >>> > >>> > >>> > >>> Does the machine you're on have multiple Ethernet > >> interfaces? If so, > >>> what you may find is adding 'ksdevice=link' will assist. > >>> > >>> > >>> > >>> CC > >>> > >>> > >>> > >>> > >>> _____ > >>> > >>> > >>> From: kickstart-list-bounces at redhat.com > >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > >> Mukarram Syed > >>> Sent: Wednesday, 16 May 2007 7:50 AM > >>> To: kickstart-list at redhat.com > >>> Subject: Unable to get any http requests: Unable to retreive > >>> netstg2.img file > >>> > >>> Hi > >>> > >>> I am sure there is some information regarding my request, but I am > >>> unable > >> to > >>> find it, hence the post. > >>> > >>> > >>> > >>> I configured my kickstart server using http. > >>> > >>> I am able to pxeboot the client, but my client is not able > >> to get the > >>> netstg2.img file from the server. > >>> > >>> I am able to http into my netstg2.img directory from a web browser > >>> from another server in the same subnet successfully. > >>> > >>> > >>> > >>> Also I tried to put in a boot cdrom and run the following: > >>> > >>> Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg > >>> ip= > >>> netmask=255.255.255.224 gateway= > >> > >> > >> I have never used static IP for ks, I alway configure > >> everything I can with dhcp. However, I do know that that's > >> not the format the kernel expects for the ip parameter. Check > >> your documentation, if I were writing a script tp parse ip= > >> then I'd conform to the existing standards (that is, the way > >> the kernel expects it), and that's what works with Ubuntu. > >> > >> > >> > >>> > >>> > >>> > >>> Still no luck. It goes straight to the cd installer. > >>> > >>> > >>> > >>> I tried to tcpdump the traffic on the MAC address of the client, it > >> outputs > >>> the tftp information and after that it stops. > >>> > >>> I don't see any http traffic coming from the client. > >> > >> That's fairly consistent with you specifying ip= incorrectly. > >> > >> btw, if you're using PXE, the why are you trying to spectify > >> static IP info? > >> > >> btw, as I recall, pxelinux has an option to pass on the IP > >> address info it gets in the proper format for the kernel to parse. > >> > >> > >>> > >>> > >>> > >>> I'd appreciate some help in this regard. > >>> > >>> I've spent all day trying to troubleshoot this but no luck. > >>> > >>> > >>> > >>> Thanks Much > >>> > >>> > >>> > >>> > >>> > >>> Mukarram Syed > >>> > >>> Unix Systems Administrator, > >>> > >>> Stanford University. > >>> > >>> > >>> > >>> > >>> NOTICE: This email and any attachments are confidential. > >>> They may contain legally privileged information or > >>> copyright material. You must not read, copy, use or > >>> disclose them without authorisation. If you are not an > >>> intended recipient, please contact us at once by return > >>> email and then delete both messages and all attachments. > >>> > >>> > >>> > >>> > >> -------------------------------------------------------------- > >> ---------- > >>> > >>> _______________________________________________ > >>> Kickstart-list mailing list > >>> Kickstart-list at redhat.com > >>> https://www.redhat.com/mailman/listinfo/kickstart-list > >> > >> > >> -- > >> > >> Cheers > >> John > >> > >> -- spambait > >> 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu > >> > >> Please do not reply off-list > >> > >> _______________________________________________ > >> Kickstart-list mailing list > >> Kickstart-list at redhat.com > >> https://www.redhat.com/mailman/listinfo/kickstart-list > >> > >> _______________________________________________ > >> Kickstart-list mailing list > >> Kickstart-list at redhat.com > >> https://www.redhat.com/mailman/listinfo/kickstart-list > >> > > > > NOTICE: This email and any attachments are confidential. > > They may contain legally privileged information or > > copyright material. You must not read, copy, use or > > disclose them without authorisation. If you are not an > > intended recipient, please contact us at once by return > > email and then delete both messages and all attachments. > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From muksyed at stanford.edu Wed May 16 14:59:47 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Wed, 16 May 2007 07:59:47 -0700 Subject: Unable to get any http requests:Unable toretreive netstg2.imgfile In-Reply-To: <2F4F96E90349F241809293630A893B3D02C50BC3@MAIL03.rvponp.fgov.be> Message-ID: <007101c797ca$d908a980$6d1b42ab@stanford.edu> I tried Alt+f1, Alt+f2 etc. no effect. I am doing all this from the cyclades console. I might have to go in front of the box and try it out, which I should be doing in a few hours. Thanks for the suggestion. -Mukarram. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of KENNEDY VAN DAM Eric Sent: Wednesday, May 16, 2007 1:46 AM To: Discussion list about Kickstart Subject: RE: Unable to get any http requests:Unable toretreive netstg2.imgfile > -----Message d'origine----- > De : kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] De la part de Mukarram Syed > Envoy? : mercredi 16 mai 2007 10:39 > ? : 'Discussion list about Kickstart' > Objet : RE: Unable to get any http requests:Unable to > retreive netstg2.imgfile > > Nope. Nothing in the error logs. > > In my troubleshooting, this is what I've found... > > When I remove the ks=http:////ks.cfg from my > "default" file > Client PXE boots, shows menu and loads vmlinuz/initrd. > Then it shows up the Language selection. No luck. > However, when I change my default file to read: ks=//ks.cfg, > Client PXE boots, shows menu and loads vmlinuz/initrd. > Then it gives the error "Error opening kickstart file (null): > Bad address" > I do some more research (see link below) > http://linux.derkeiler.com/Mailing-Lists/Fedora/2004-06/5939.html > and modified the ks.cfg file...add more network parameters like so: > network --device=eth0 --bootproto=static --ip=171.67.39.139 > --netmask=255.255.255.224 --gateway=171.67.39.129 > --nameserver=171.64.7.77 > --hostname=as-test.stanford.edu > Still no luck. > I then changed my --device=eth0 to --device=link and then to > --device=eth1...still no luck. Then put it back to --device=eth0. > Then I change my PATH from: > url --url http://171.67.39.130/ENT_LINUX_64/Enterprise > to > url --url http://171.67.39.130/ENT_LINUX_64 > Still no luck. > Again nothing in the http logs. > > It looks like the installer is reading the ks.cfg file > silently since it > complained that it couldn't find it from the default file > when I didn't > specify the http://. > Can I figure out what is reading and stopping. Is there a > way to look at > the kickstart logs (If there is any...I haven't seen one). > > I tried to create logging as well in my ks.cfg file...no > luck...no output: > %pre --log=/tmp/my-pre-log > echo 'Hello, World!' Sorry if the suggestion has already been made. Did you try to go to the other consoles ? There is a lot of information there which can help to debu your problem. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From muksyed at stanford.edu Wed May 16 15:03:13 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Wed, 16 May 2007 08:03:13 -0700 Subject: Unable to get any http requests:Unable toretrieve netstg2.imgfile In-Reply-To: Message-ID: <007201c797cb$53d65860$6d1b42ab@stanford.edu> I feel your pain. I'll let you know if I find anything. Thanks -Mukarram. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com Sent: Wednesday, May 16, 2007 6:00 AM To: Discussion list about Kickstart Subject: RE: Unable to get any http requests:Unable toretrieve netstg2.imgfile I have been fighting almost the identical problem. See the thread entitled "Re: Kickstart via PXE/NFS" After the kernel is loaded via configuration set up in the pxelinux config file, the network will not come up and the ks.cfg file can then not be loaded. I get the network config screen asking to use DHCP or to enter static info for setup of eth0. It has really bee frustrating. I finally just placed my ks.cfg on CDROM and was able to get the rest of the PXE auto install to complete. If you get a resolution, I will be very interested. I have had a case open with RedHat for 2 days waiting for a response. kickstart-list-bounces at redhat.com wrote on 05/16/2007 04:38:30 AM: > Nope. Nothing in the error logs. > > In my troubleshooting, this is what I've found... > > When I remove the ks=http:////ks.cfg from my "default" file > Client PXE boots, shows menu and loads vmlinuz/initrd. > Then it shows up the Language selection. No luck. > However, when I change my default file to read: ks=//ks.cfg, > Client PXE boots, shows menu and loads vmlinuz/initrd. > Then it gives the error "Error opening kickstart file (null): Bad address" > I do some more research (see link below) > http://linux.derkeiler.com/Mailing-Lists/Fedora/2004-06/5939.html > and modified the ks.cfg file...add more network parameters like so: > network --device=eth0 --bootproto=static --ip=171.67.39.139 > --netmask=255.255.255.224 --gateway=171.67.39.129 --nameserver=171.64.7.77 > --hostname=as-test.stanford.edu > Still no luck. > I then changed my --device=eth0 to --device=link and then to > --device=eth1...still no luck. Then put it back to --device=eth0. > Then I change my PATH from: > url --url http://171.67.39.130/ENT_LINUX_64/Enterprise > to > url --url http://171.67.39.130/ENT_LINUX_64 > Still no luck. > Again nothing in the http logs. > > It looks like the installer is reading the ks.cfg file silently since it > complained that it couldn't find it from the default file when I didn't > specify the http://. > Can I figure out what is reading and stopping. Is there a way to look at > the kickstart logs (If there is any...I haven't seen one). > > I tried to create logging as well in my ks.cfg file...no luck...no output: > %pre --log=/tmp/my-pre-log > echo 'Hello, World!' > > Here is my directory/file setup: > > TFTP: > > /tftpboot/linux-install/pxelinux.cfg > [root at as-kickball pxelinux.cfg]# ls > default pxeos.xml > > [root at as-kickball linux-install]# ls -lR > .: > total 32 > drwxr-xr-x 2 root root 4096 May 14 00:31 ENT_LINUX_32 > drwxr-xr-x 2 root root 4096 May 16 00:30 ENT_LINUX_64 > drwxr-xr-x 2 root root 4096 May 14 21:09 msgs > -rwxr-xr-x 1 root root 12952 Oct 24 2006 pxelinux.0 > drwxr-xr-x 2 root root 4096 May 16 00:25 pxelinux.cfg > > ./ENT_LINUX_32: > total 0 > > ./ENT_LINUX_64: > total 5984 > -rwxr-xr-x 1 root root 4310479 May 14 15:29 initrd.img > -rwxr-xr-x 1 root root 85 May 16 00:30 ks.cfg > -rwxr-xr-x 1 root root 1792835 May 14 15:29 vmlinuz > > ./msgs: > total 24 > -rwxr-xr-x 1 root root 534 May 14 21:00 boot.msg > -rwxr-xr-x 1 root root 668 Oct 24 2006 expert.msg > -rwxr-xr-x 1 root root 871 Oct 24 2006 general.msg > -rwxr-xr-x 1 root root 860 Oct 24 2006 param.msg > -rwxr-xr-x 1 root root 530 Oct 24 2006 rescue.msg > -rwxr-xr-x 1 root root 545 Oct 24 2006 snake.msg > > ./pxelinux.cfg: > total 20 > -rwxr-xr-x 1 root root 511 May 16 00:25 default > -rwxr-xr-x 1 root root 511 May 15 23:34 default.bak > -rwxr-xr-x 1 root root 461 May 15 14:44 default.dhcp > -rwxr-xr-x 1 root root 573 May 15 13:30 default.staticIP > -rwxr-xr-x 1 root root 42 May 14 16:28 pxeos.xml > > My HTTP directory: > > [root at as-kickball /]# ls -lart /var/www/html > total 24 > drwxr-xr-x 8 root root 4096 Mar 21 19:17 .. > drwxr-xr-x 4 root root 4096 May 14 00:35 . > drwxr-xr-x 3 root root 4096 May 14 00:40 ENT_LINUX_32 > drwxr-xr-x 3 root root 4096 May 15 15:49 ENT_LINUX_64 > > [root at as-kickball /]# ls -lart /var/www/html/ENT_LINUX_64/Enterprise/ > total 136 > drwxr-xr-x 2 root root 114688 Feb 24 13:15 RPMS > drwxr-xr-x 2 root root 4096 May 14 15:53 base > -rwxr-xr-x 1 root root 1386 May 15 15:05 ks_ent_linux_64.cfg > drwxr-xr-x 3 root root 4096 May 15 15:49 .. > -rw-r--r-- 1 root root 1242 May 16 01:08 ks.cfg > drwxr-xr-x 4 root root 4096 May 16 01:08 . > > I am at a loss now. I am sure it's something simple as it mostly is. > Help is appreciated. > > Tomorrow, I'll tell my networking guy to enable "spanning tree" so I could > use dhcp and get the static crap out of my way. > Will let you know more. > > -Mukarram. > > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mike Jennings > Sent: Tuesday, May 15, 2007 8:47 PM > To: Discussion list about Kickstart > Subject: Re: Unable to get any http requests:Unable to retreive > netstg2.imgfile > > Is the apache log showing any 404 errors? My stupid screwups are > always shown in the error log when I typo something. > > Mike > > > > On May 15, 2007, at 10:39 PM, Mukarram Syed wrote: > > > > > Colin, > > I changed it from ksdevice=eth0 to ksdevice=link > > > > Sorry, that didn't help. > > > > Wonder what else could it be, why is it not connecting to the http > > server? > > I don't see any traffic in tcpdump after pxeboot. > > > > -Mukarram. > > > > -----Original Message----- > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin > > C. (Unix > > Engineer) > > Sent: Tuesday, May 15, 2007 5:56 PM > > To: Discussion list about Kickstart > > Subject: RE: Unable to get any http requests:Unabletoretreive > > netstg2.imgfile > > > > > > Try changing 'ksdevice=eth0' to 'ksdevice=link'. With 2.6 kernels, > > the > > kernel's idea of what eth0 and eth1 are often differs from the BIOS/ > > PXE > > stack. > > > > CC > > > >> -----Original Message----- > >> From: kickstart-list-bounces at redhat.com > >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed > >> Sent: Wednesday, 16 May 2007 8:42 AM > >> To: 'Discussion list about Kickstart' > >> Subject: RE: Unable to get any http requests: > >> Unabletoretreive netstg2.imgfile > >> > >> John, > >> Sorry...just to make things clear. > >> eth0 works to get the dhcp/tftp request and eth1 does not. > >> However, I tried using both is what I meant. > >> > >> We have a problem with dhcp, the networking folks here did > >> not enable "spanning tree" on the switch ports, until that > >> happens I am using static > >> ip. I double checked the syntax etc. > >> > >> Thanks > >> -Mukarram. > >> > >> -----Original Message----- > >> From: kickstart-list-bounces at redhat.com > >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John > >> Summerfield > >> Sent: Tuesday, May 15, 2007 5:32 PM > >> To: Discussion list about Kickstart > >> Subject: Re: Unable to get any http requests: Unable > >> toretreive netstg2.imgfile > >> > >> Mukarram Syed wrote: > >>> Thanks Colin for the quick response > >>> > >>> > >>> > >>> Yes. It does have. It's a Dell PE2950. > >>> > >>> > >>> > >>> I have already tried that option as well. > >>> > >>> I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. > >>> > >> > >> If you can connect all network interfaces, that might help, > >> because it then doesn't matter whether either one or both is used. > >> > >> If you run arpwatch that will report on which interfaces are > >> used (by mac address and IP). > >> > >> More below... > >> > >>> > >>> > >>> -Mukarram. > >>> > >>> > >>> > >>> > >>> > >>> _____ > >>> > >>> From: kickstart-list-bounces at redhat.com > >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > >> Coe, Colin C. > >> (Unix > >>> Engineer) > >>> Sent: Tuesday, May 15, 2007 4:52 PM > >>> To: Discussion list about Kickstart > >>> Subject: RE: Unable to get any http requests: Unable to retreive > >>> netstg2.imgfile > >>> > >>> > >>> > >>> Does the machine you're on have multiple Ethernet > >> interfaces? If so, > >>> what you may find is adding 'ksdevice=link' will assist. > >>> > >>> > >>> > >>> CC > >>> > >>> > >>> > >>> > >>> _____ > >>> > >>> > >>> From: kickstart-list-bounces at redhat.com > >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > >> Mukarram Syed > >>> Sent: Wednesday, 16 May 2007 7:50 AM > >>> To: kickstart-list at redhat.com > >>> Subject: Unable to get any http requests: Unable to retreive > >>> netstg2.img file > >>> > >>> Hi > >>> > >>> I am sure there is some information regarding my request, but I am > >>> unable > >> to > >>> find it, hence the post. > >>> > >>> > >>> > >>> I configured my kickstart server using http. > >>> > >>> I am able to pxeboot the client, but my client is not able > >> to get the > >>> netstg2.img file from the server. > >>> > >>> I am able to http into my netstg2.img directory from a web browser > >>> from another server in the same subnet successfully. > >>> > >>> > >>> > >>> Also I tried to put in a boot cdrom and run the following: > >>> > >>> Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg > >>> ip= > >>> netmask=255.255.255.224 gateway= > >> > >> > >> I have never used static IP for ks, I alway configure > >> everything I can with dhcp. However, I do know that that's > >> not the format the kernel expects for the ip parameter. Check > >> your documentation, if I were writing a script tp parse ip= > >> then I'd conform to the existing standards (that is, the way > >> the kernel expects it), and that's what works with Ubuntu. > >> > >> > >> > >>> > >>> > >>> > >>> Still no luck. It goes straight to the cd installer. > >>> > >>> > >>> > >>> I tried to tcpdump the traffic on the MAC address of the client, it > >> outputs > >>> the tftp information and after that it stops. > >>> > >>> I don't see any http traffic coming from the client. > >> > >> That's fairly consistent with you specifying ip= incorrectly. > >> > >> btw, if you're using PXE, the why are you trying to spectify > >> static IP info? > >> > >> btw, as I recall, pxelinux has an option to pass on the IP > >> address info it gets in the proper format for the kernel to parse. > >> > >> > >>> > >>> > >>> > >>> I'd appreciate some help in this regard. > >>> > >>> I've spent all day trying to troubleshoot this but no luck. > >>> > >>> > >>> > >>> Thanks Much > >>> > >>> > >>> > >>> > >>> > >>> Mukarram Syed > >>> > >>> Unix Systems Administrator, > >>> > >>> Stanford University. > >>> > >>> > >>> > >>> > >>> NOTICE: This email and any attachments are confidential. > >>> They may contain legally privileged information or > >>> copyright material. You must not read, copy, use or > >>> disclose them without authorisation. If you are not an > >>> intended recipient, please contact us at once by return > >>> email and then delete both messages and all attachments. > >>> > >>> > >>> > >>> > >> -------------------------------------------------------------- > >> ---------- > >>> > >>> _______________________________________________ > >>> Kickstart-list mailing list > >>> Kickstart-list at redhat.com > >>> https://www.redhat.com/mailman/listinfo/kickstart-list > >> > >> > >> -- > >> > >> Cheers > >> John > >> > >> -- spambait > >> 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu > >> > >> Please do not reply off-list > >> > >> _______________________________________________ > >> Kickstart-list mailing list > >> Kickstart-list at redhat.com > >> https://www.redhat.com/mailman/listinfo/kickstart-list > >> > >> _______________________________________________ > >> Kickstart-list mailing list > >> Kickstart-list at redhat.com > >> https://www.redhat.com/mailman/listinfo/kickstart-list > >> > > > > NOTICE: This email and any attachments are confidential. > > They may contain legally privileged information or > > copyright material. You must not read, copy, use or > > disclose them without authorisation. If you are not an > > intended recipient, please contact us at once by return > > email and then delete both messages and all attachments. > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. From patchu1 at yahoo.com Wed May 16 15:05:26 2007 From: patchu1 at yahoo.com (Pat) Date: Wed, 16 May 2007 08:05:26 -0700 (PDT) Subject: Location of installation media no longer /tmp/cdrom? In-Reply-To: <20070516083841.4A3F2730C1@hormel.redhat.com> Message-ID: <819591.59305.qm@web32503.mail.mud.yahoo.com> --- kickstart-list-request at redhat.com wrote: > > From: Jeroen van Meeuwen > To: Discussion list about Kickstart > > Date: Wed, 16 May 2007 10:30:44 +0200 > Subject: Re: Location of installation media no > longer /tmp/cdrom? > > Pat wrote: > > However I get the following error during the > install: > > > > mount: Mounting /tmp/cdrom on /mnt/source failed: > No > > such file or directory > > > > > > Are you sure the /mnt/source directory exists? > > -kanarip > Yes, /mnt/source was created. I did not see /tmp/cdrom though. The same kickstart file worked with Redhat/CentOS 4.x, it looks like something changed with 5? ____________________________________________________________________________________Building a website is a piece of cake. Yahoo! Small Business gives you all the tools to get online. http://smallbusiness.yahoo.com/webhosting From muksyed at stanford.edu Wed May 16 17:04:46 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Wed, 16 May 2007 10:04:46 -0700 Subject: Unable to get any http requests:Unable to retreive netstg2.img file Message-ID: <009001c797dc$4eb23690$6d1b42ab@stanford.edu> I did what Kennedy suggested. Thanks much. Connected the KVM and did the Alt-F commands. What I found is this: " ... ... getting kickstart file looking for first netDev with link eth0 has link, using it doing kickstart...setting it up waiting for link... 4 seconds no DNS servers, can't look up hostname ks location: http://171.67.39.130/ENT_LINUX_64/Enterprise/ks.cfg transferring http://171.67.39.130//./ENT_LINUX_64/Enterprise/ks.cfg to a fd failed to retrieve http://171.67.39.130///ENT_LINUX_64/Enterprise/ks.cfg trying to mount CD device hda " As I said before, I could retrieve all the above http links from another system on the same subnet as my kickstart install server and the client. After which I modified my default file to the following: label 1 kernel ENT_LINUX_64/vmlinuz append nofb initrd=ENT_LINUX_64/initrd.img ramdisk_size=10000 lang= devfs=nomount ramdisk_size=9216 ksdevice=link ip=171.67.39.139 netmask=255.255.255.224 gateway=171.67.39.129 dns=171.64.7.77 ks=http://171.67.39.130/ENT_LINUX_64/Enterprise/ks.cfg I put the dns= entry in there. The I get the following: " ... Adding 171.64.7.77 dnsserver is 171.64.7.77 doing kickstart...setting it up waiting for link 4 seconds reverse name lookup failed ks location: http://171.67.39.130/ENT_LINUX_64/Enterprise/ks.cfg transferring http://171.67.39.130//./ENT_LINUX_64/Enterprise/ks.cfg to a fd failed to retrieve http://171.67.39.130///ENT_LINUX_64/Enterprise/ks.cfg trying to mount CD device had " Here is the network information from my ks.cfg file: #Network information network --device=eth0 --bootproto=static --ip=171.67.39.130 --netmask=255.255.255.224 --gateway=171.67.39.129 --nameserver=171.64.7.77 --hostname=as-kickball.stanford.edu The hostname is invalid. It?s a bogus name. I?ll try to make it a valid name and check it out. Should there be a hostname in the default file as well since I am using static IP? I think I am getting close here. Something else is missing for the reverse lookup to work. Suggestions would be helpful at this point. Thanks -Mukarram. -----Original Message----- From: Mukarram Syed [mailto:muksyed at stanford.edu] Sent: Wednesday, May 16, 2007 8:00 AM To: 'Discussion list about Kickstart' Subject: RE: Unable to get any http requests:Unable toretreive netstg2.imgfile I tried Alt+f1, Alt+f2 etc. no effect. I am doing all this from the cyclades console. I might have to go in front of the box and try it out, which I should be doing in a few hours. Thanks for the suggestion. -Mukarram. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of KENNEDY VAN DAM Eric Sent: Wednesday, May 16, 2007 1:46 AM To: Discussion list about Kickstart Subject: RE: Unable to get any http requests:Unable toretreive netstg2.imgfile > -----Message d'origine----- > De : kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] De la part de Mukarram Syed > Envoy? : mercredi 16 mai 2007 10:39 > ? : 'Discussion list about Kickstart' > Objet : RE: Unable to get any http requests:Unable to > retreive netstg2.imgfile > > Nope. Nothing in the error logs. > > In my troubleshooting, this is what I've found... > > When I remove the ks=http:////ks.cfg from my > "default" file > Client PXE boots, shows menu and loads vmlinuz/initrd. > Then it shows up the Language selection. No luck. > However, when I change my default file to read: ks=//ks.cfg, > Client PXE boots, shows menu and loads vmlinuz/initrd. > Then it gives the error "Error opening kickstart file (null): > Bad address" > I do some more research (see link below) > http://linux.derkeiler.com/Mailing-Lists/Fedora/2004-06/5939.html > and modified the ks.cfg file...add more network parameters like so: > network --device=eth0 --bootproto=static --ip=171.67.39.139 > --netmask=255.255.255.224 --gateway=171.67.39.129 > --nameserver=171.64.7.77 > --hostname=as-test.stanford.edu > Still no luck. > I then changed my --device=eth0 to --device=link and then to > --device=eth1...still no luck. Then put it back to --device=eth0. > Then I change my PATH from: > url --url http://171.67.39.130/ENT_LINUX_64/Enterprise > to > url --url http://171.67.39.130/ENT_LINUX_64 > Still no luck. > Again nothing in the http logs. > > It looks like the installer is reading the ks.cfg file > silently since it > complained that it couldn't find it from the default file > when I didn't > specify the http://. > Can I figure out what is reading and stopping. Is there a > way to look at > the kickstart logs (If there is any...I haven't seen one). > > I tried to create logging as well in my ks.cfg file...no > luck...no output: > %pre --log=/tmp/my-pre-log > echo 'Hello, World!' Sorry if the suggestion has already been made. Did you try to go to the other consoles ? There is a lot of information there which can help to debu your problem. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From patchu1 at yahoo.com Wed May 16 20:45:02 2007 From: patchu1 at yahoo.com (Pat) Date: Wed, 16 May 2007 13:45:02 -0700 (PDT) Subject: RHEL5 kickstart needs to mount cdrom in postinst In-Reply-To: <20070516083841.4A3F2730C1@hormel.redhat.com> Message-ID: <124890.72270.qm@web32505.mail.mud.yahoo.com> Martin Steinmann wrote: >>> Is there a similarly elegant way to add a driver disk image? I am > still >>> struggling embedding an additional RAID driver into the boot CD. I >>> suppose I need to update the initial initrd file (using mkinitrd) > and >>> then tell Anaconda to load the driver image during stage2. Can the >>> driver disk image be placed somewhere into the tree similar to the >>> updates.img? >> >> Yes. >> >> Look for the driverdisk option at >> https://www.redhat.com/docs/manuals/enterprise/RHEL-5- >> manual/Installation_Guide-en-US/s1-kickstart2-options.html >> >> That option is in RHEL4 & 5 that I know of. > > Yes, I looked at this and tried several options. The problem is that I > do not have a network during installation (no ftp, http or nfs > available) and therefore the driver disk image needs to be in the tree > on the CD. Also, the HD is empty when I start. I had to same thing awhile ago. It was beyond my expertise to add a driver to initrd.img, although I may give it another shot after reading Chris's steps. What I ended up doing was putting the driver files (modinfo, modules.cgz, etc.) in the root of my custom install DVD, then I added the following line to my kickstart file: driverdisk hda --type=iso9660 The RAID drive should load automatically. You'd be limited to the first CD/DVD drive using the above call BTW. It's not the most elegant solution, but it got me up and running. ____________________________________________________________________________________ Now that's room service! Choose from over 150,000 hotels in 45,000 destinations on Yahoo! Travel to find your fit. http://farechase.yahoo.com/promo-generic-14795097 From SSTinsley at upsfreight.com Wed May 16 21:19:05 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Wed, 16 May 2007 17:19:05 -0400 Subject: Unable to get any http requests:Unable toretrieve netstg2.imgfile In-Reply-To: <007201c797cb$53d65860$6d1b42ab@stanford.edu> Message-ID: I finally gave up and copied my ks.cfg to a CDROM and changed my pxelinux.cfg to load the KS cfg file from CD the drive. Anaconda could read the ks.cfg. The ks.cfg file pointed to a RedHat release image on an NFS server. However, Anaconda said it could not read the NFS release and displayed a screen to input the corrected NFS image information. The defaults from the KS cfg file where displayed as defaults. I entered nothing different and hit OK. Anaconda was able to now mount the RHEL image using the same information as was specified in the KS cfg file. I think Anaconda has a few bugs in handling network interfaces and protocols. I would love to figure out a fix, but I got to get some servers installed. I'll keep and eye out for your posts to see if you have any better luck. I still have not heard back from Redhat. So much for paid support. Get me back on Solaris!! kickstart-list-bounces at redhat.com wrote on 05/16/2007 11:03:13 AM: > I feel your pain. > I'll let you know if I find anything. > Thanks > > -Mukarram. > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > SSTinsley at upsfreight.com > Sent: Wednesday, May 16, 2007 6:00 AM > To: Discussion list about Kickstart > Subject: RE: Unable to get any http requests:Unable toretrieve > netstg2.imgfile > > > I have been fighting almost the identical problem. See the thread entitled > > "Re: Kickstart via PXE/NFS" > > After the kernel is loaded via configuration set up in the pxelinux config > file, the > network will not come up and the ks.cfg file can then not be loaded. I get > the > network config screen asking to use DHCP or to enter static info for setup > of eth0. It > has really bee frustrating. I finally just placed my ks.cfg on CDROM and was > able > to get the rest of the PXE auto install to complete. > > If you get a resolution, I will be very interested. I have had a case open > with > RedHat for 2 days waiting for a response. > > kickstart-list-bounces at redhat.com wrote on 05/16/2007 04:38:30 AM: > > > Nope. Nothing in the error logs. > > > > In my troubleshooting, this is what I've found... > > > > When I remove the ks=http:////ks.cfg from my "default" > file > > Client PXE boots, shows menu and loads vmlinuz/initrd. > > Then it shows up the Language selection. No luck. > > However, when I change my default file to read: ks=//ks.cfg, > > Client PXE boots, shows menu and loads vmlinuz/initrd. > > Then it gives the error "Error opening kickstart file (null): Bad address" > > I do some more research (see link below) > > http://linux.derkeiler.com/Mailing-Lists/Fedora/2004-06/5939.html > > and modified the ks.cfg file...add more network parameters like so: > > network --device=eth0 --bootproto=static --ip=171.67.39.139 > > --netmask=255.255.255.224 --gateway=171.67.39.129 --nameserver=171.64.7.77 > > --hostname=as-test.stanford.edu > > Still no luck. > > I then changed my --device=eth0 to --device=link and then to > > --device=eth1...still no luck. Then put it back to --device=eth0. > > Then I change my PATH from: > > url --url http://171.67.39.130/ENT_LINUX_64/Enterprise > > to > > url --url http://171.67.39.130/ENT_LINUX_64 > > Still no luck. > > Again nothing in the http logs. > > > > It looks like the installer is reading the ks.cfg file silently since it > > complained that it couldn't find it from the default file when I didn't > > specify the http://. > > Can I figure out what is reading and stopping. Is there a way to look at > > the kickstart logs (If there is any...I haven't seen one). > > > > I tried to create logging as well in my ks.cfg file...no luck...no output: > > %pre --log=/tmp/my-pre-log > > echo 'Hello, World!' > > > > Here is my directory/file setup: > > > > TFTP: > > > > /tftpboot/linux-install/pxelinux.cfg > > [root at as-kickball pxelinux.cfg]# ls > > default pxeos.xml > > > > [root at as-kickball linux-install]# ls -lR > > .: > > total 32 > > drwxr-xr-x 2 root root 4096 May 14 00:31 ENT_LINUX_32 > > drwxr-xr-x 2 root root 4096 May 16 00:30 ENT_LINUX_64 > > drwxr-xr-x 2 root root 4096 May 14 21:09 msgs > > -rwxr-xr-x 1 root root 12952 Oct 24 2006 pxelinux.0 > > drwxr-xr-x 2 root root 4096 May 16 00:25 pxelinux.cfg > > > > ./ENT_LINUX_32: > > total 0 > > > > ./ENT_LINUX_64: > > total 5984 > > -rwxr-xr-x 1 root root 4310479 May 14 15:29 initrd.img > > -rwxr-xr-x 1 root root 85 May 16 00:30 ks.cfg > > -rwxr-xr-x 1 root root 1792835 May 14 15:29 vmlinuz > > > > ./msgs: > > total 24 > > -rwxr-xr-x 1 root root 534 May 14 21:00 boot.msg > > -rwxr-xr-x 1 root root 668 Oct 24 2006 expert.msg > > -rwxr-xr-x 1 root root 871 Oct 24 2006 general.msg > > -rwxr-xr-x 1 root root 860 Oct 24 2006 param.msg > > -rwxr-xr-x 1 root root 530 Oct 24 2006 rescue.msg > > -rwxr-xr-x 1 root root 545 Oct 24 2006 snake.msg > > > > ./pxelinux.cfg: > > total 20 > > -rwxr-xr-x 1 root root 511 May 16 00:25 default > > -rwxr-xr-x 1 root root 511 May 15 23:34 default.bak > > -rwxr-xr-x 1 root root 461 May 15 14:44 default.dhcp > > -rwxr-xr-x 1 root root 573 May 15 13:30 default.staticIP > > -rwxr-xr-x 1 root root 42 May 14 16:28 pxeos.xml > > > > My HTTP directory: > > > > [root at as-kickball /]# ls -lart /var/www/html > > total 24 > > drwxr-xr-x 8 root root 4096 Mar 21 19:17 .. > > drwxr-xr-x 4 root root 4096 May 14 00:35 . > > drwxr-xr-x 3 root root 4096 May 14 00:40 ENT_LINUX_32 > > drwxr-xr-x 3 root root 4096 May 15 15:49 ENT_LINUX_64 > > > > [root at as-kickball /]# ls -lart /var/www/html/ENT_LINUX_64/Enterprise/ > > total 136 > > drwxr-xr-x 2 root root 114688 Feb 24 13:15 RPMS > > drwxr-xr-x 2 root root 4096 May 14 15:53 base > > -rwxr-xr-x 1 root root 1386 May 15 15:05 ks_ent_linux_64.cfg > > drwxr-xr-x 3 root root 4096 May 15 15:49 .. > > -rw-r--r-- 1 root root 1242 May 16 01:08 ks.cfg > > drwxr-xr-x 4 root root 4096 May 16 01:08 . > > > > I am at a loss now. I am sure it's something simple as it mostly is. > > Help is appreciated. > > > > Tomorrow, I'll tell my networking guy to enable "spanning tree" so I could > > use dhcp and get the static crap out of my way. > > Will let you know more. > > > > -Mukarram. > > > > > > -----Original Message----- > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mike Jennings > > Sent: Tuesday, May 15, 2007 8:47 PM > > To: Discussion list about Kickstart > > Subject: Re: Unable to get any http requests:Unable to retreive > > netstg2.imgfile > > > > Is the apache log showing any 404 errors? My stupid screwups are > > always shown in the error log when I typo something. > > > > Mike > > > > > > > > On May 15, 2007, at 10:39 PM, Mukarram Syed wrote: > > > > > > > > Colin, > > > I changed it from ksdevice=eth0 to ksdevice=link > > > > > > Sorry, that didn't help. > > > > > > Wonder what else could it be, why is it not connecting to the http > > > server? > > > I don't see any traffic in tcpdump after pxeboot. > > > > > > -Mukarram. > > > > > > -----Original Message----- > > > From: kickstart-list-bounces at redhat.com > > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin > > > C. (Unix > > > Engineer) > > > Sent: Tuesday, May 15, 2007 5:56 PM > > > To: Discussion list about Kickstart > > > Subject: RE: Unable to get any http requests:Unabletoretreive > > > netstg2.imgfile > > > > > > > > > Try changing 'ksdevice=eth0' to 'ksdevice=link'. With 2.6 kernels, > > > the > > > kernel's idea of what eth0 and eth1 are often differs from the BIOS/ > > > PXE > > > stack. > > > > > > CC > > > > > >> -----Original Message----- > > >> From: kickstart-list-bounces at redhat.com > > >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed > > >> Sent: Wednesday, 16 May 2007 8:42 AM > > >> To: 'Discussion list about Kickstart' > > >> Subject: RE: Unable to get any http requests: > > >> Unabletoretreive netstg2.imgfile > > >> > > >> John, > > >> Sorry...just to make things clear. > > >> eth0 works to get the dhcp/tftp request and eth1 does not. > > >> However, I tried using both is what I meant. > > >> > > >> We have a problem with dhcp, the networking folks here did > > >> not enable "spanning tree" on the switch ports, until that > > >> happens I am using static > > >> ip. I double checked the syntax etc. > > >> > > >> Thanks > > >> -Mukarram. > > >> > > >> -----Original Message----- > > >> From: kickstart-list-bounces at redhat.com > > >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John > > >> Summerfield > > >> Sent: Tuesday, May 15, 2007 5:32 PM > > >> To: Discussion list about Kickstart > > >> Subject: Re: Unable to get any http requests: Unable > > >> toretreive netstg2.imgfile > > >> > > >> Mukarram Syed wrote: > > >>> Thanks Colin for the quick response > > >>> > > >>> > > >>> > > >>> Yes. It does have. It's a Dell PE2950. > > >>> > > >>> > > >>> > > >>> I have already tried that option as well. > > >>> > > >>> I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. > > >>> > > >> > > >> If you can connect all network interfaces, that might help, > > >> because it then doesn't matter whether either one or both is used. > > >> > > >> If you run arpwatch that will report on which interfaces are > > >> used (by mac address and IP). > > >> > > >> More below... > > >> > > >>> > > >>> > > >>> -Mukarram. > > >>> > > >>> > > >>> > > >>> > > >>> > > >>> _____ > > >>> > > >>> From: kickstart-list-bounces at redhat.com > > >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > > >> Coe, Colin C. > > >> (Unix > > >>> Engineer) > > >>> Sent: Tuesday, May 15, 2007 4:52 PM > > >>> To: Discussion list about Kickstart > > >>> Subject: RE: Unable to get any http requests: Unable to retreive > > >>> netstg2.imgfile > > >>> > > >>> > > >>> > > >>> Does the machine you're on have multiple Ethernet > > >> interfaces? If so, > > >>> what you may find is adding 'ksdevice=link' will assist. > > >>> > > >>> > > >>> > > >>> CC > > >>> > > >>> > > >>> > > >>> > > >>> _____ > > >>> > > >>> > > >>> From: kickstart-list-bounces at redhat.com > > >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > > >> Mukarram Syed > > >>> Sent: Wednesday, 16 May 2007 7:50 AM > > >>> To: kickstart-list at redhat.com > > >>> Subject: Unable to get any http requests: Unable to retreive > > >>> netstg2.img file > > >>> > > >>> Hi > > >>> > > >>> I am sure there is some information regarding my request, but I am > > >>> unable > > >> to > > >>> find it, hence the post. > > >>> > > >>> > > >>> > > >>> I configured my kickstart server using http. > > >>> > > >>> I am able to pxeboot the client, but my client is not able > > >> to get the > > >>> netstg2.img file from the server. > > >>> > > >>> I am able to http into my netstg2.img directory from a web browser > > >>> from another server in the same subnet successfully. > > >>> > > >>> > > >>> > > >>> Also I tried to put in a boot cdrom and run the following: > > >>> > > >>> Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg > > >>> ip= > > >>> netmask=255.255.255.224 gateway= > > >> > > >> > > >> I have never used static IP for ks, I alway configure > > >> everything I can with dhcp. However, I do know that that's > > >> not the format the kernel expects for the ip parameter. Check > > >> your documentation, if I were writing a script tp parse ip= > > >> then I'd conform to the existing standards (that is, the way > > >> the kernel expects it), and that's what works with Ubuntu. > > >> > > >> > > >> > > >>> > > >>> > > >>> > > >>> Still no luck. It goes straight to the cd installer. > > >>> > > >>> > > >>> > > >>> I tried to tcpdump the traffic on the MAC address of the client, it > > >> outputs > > >>> the tftp information and after that it stops. > > >>> > > >>> I don't see any http traffic coming from the client. > > >> > > >> That's fairly consistent with you specifying ip= incorrectly. > > >> > > >> btw, if you're using PXE, the why are you trying to spectify > > >> static IP info? > > >> > > >> btw, as I recall, pxelinux has an option to pass on the IP > > >> address info it gets in the proper format for the kernel to parse. > > >> > > >> > > >>> > > >>> > > >>> > > >>> I'd appreciate some help in this regard. > > >>> > > >>> I've spent all day trying to troubleshoot this but no luck. > > >>> > > >>> > > >>> > > >>> Thanks Much > > >>> > > >>> > > >>> > > >>> > > >>> > > >>> Mukarram Syed > > >>> > > >>> Unix Systems Administrator, > > >>> > > >>> Stanford University. > > >>> > > >>> > > >>> > > >>> > > >>> NOTICE: This email and any attachments are confidential. > > >>> They may contain legally privileged information or > > >>> copyright material. You must not read, copy, use or > > >>> disclose them without authorisation. If you are not an > > >>> intended recipient, please contact us at once by return > > >>> email and then delete both messages and all attachments. > > >>> > > >>> > > >>> > > >>> > > >> -------------------------------------------------------------- > > >> ---------- > > >>> > > >>> _______________________________________________ > > >>> Kickstart-list mailing list > > >>> Kickstart-list at redhat.com > > >>> https://www.redhat.com/mailman/listinfo/kickstart-list > > >> > > >> > > >> -- > > >> > > >> Cheers > > >> John > > >> > > >> -- spambait > > >> 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu > > >> > > >> Please do not reply off-list > > >> > > >> _______________________________________________ > > >> Kickstart-list mailing list > > >> Kickstart-list at redhat.com > > >> https://www.redhat.com/mailman/listinfo/kickstart-list > > >> > > >> _______________________________________________ > > >> Kickstart-list mailing list > > >> Kickstart-list at redhat.com > > >> https://www.redhat.com/mailman/listinfo/kickstart-list > > >> > > > > > > NOTICE: This email and any attachments are confidential. > > > They may contain legally privileged information or > > > copyright material. You must not read, copy, use or > > > disclose them without authorisation. If you are not an > > > intended recipient, please contact us at once by return > > > email and then delete both messages and all attachments. > > > > > > _______________________________________________ > > > Kickstart-list mailing list > > > Kickstart-list at redhat.com > > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > > > _______________________________________________ > > > Kickstart-list mailing list > > > Kickstart-list at redhat.com > > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT > IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED > ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the > intended recipient of this document, you are notified that any review, > dissemination, distribution or copying of this communication is prohibited. > If you have received this communication in error, please notify me > immediately by return email, delete the electronic message and destroy any > printed copies. Thank you for your cooperation. > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From debian at herakles.homelinux.org Wed May 16 23:00:29 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Thu, 17 May 2007 07:00:29 +0800 Subject: Unable to get any http requests:Unable to retreive netstg2.imgfile In-Reply-To: <005c01c79795$956fc580$6d1b42ab@stanford.edu> References: <005c01c79795$956fc580$6d1b42ab@stanford.edu> Message-ID: <464B8D0D.8040206@herakles.homelinux.org> Mukarram Syed wrote: > > I am at a loss now. I am sure it's something simple as it mostly is. > Help is appreciated. > > Tomorrow, I'll tell my networking guy to enable "spanning tree" so I could > use dhcp and get the static crap out of my way. > Will let you know more. > Please, do this: 1. Plug all NICs into the same network 2. run tcpdump to log traffic to a file 3. When it's failed, use tcpdump (or wireshark) to analyse the file. If you do a minimum of selection, if it's on the wire it's in your dump. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From debian at herakles.homelinux.org Wed May 16 23:06:06 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Thu, 17 May 2007 07:06:06 +0800 Subject: Unable to get any http requests: Unable toretreive netstg2.imgfile In-Reply-To: <20070515192452.1e1ofj8em7b400gk@webmail.stanford.edu> References: <005401c79753$0fcbc240$6d1b42ab@stanford.edu> <464A58C6.7030801@herakles.homelinux.org> <20070515192452.1e1ofj8em7b400gk@webmail.stanford.edu> Message-ID: <464B8E5E.10706@herakles.homelinux.org> muksyed at stanford.edu wrote: > Quoting John Summerfield : > >> Mukarram Syed wrote: >>> John, >>> Sorry...just to make things clear. >>> eth0 works to get the dhcp/tftp request and eth1 does not. However, >>> I tried >>> using both is what I meant. >>> >>> We have a problem with dhcp, the networking folks here did not enable >>> "spanning tree" on the switch ports, until that happens I am using >>> static >>> ip. I double checked the syntax etc. >> >> I don't understand... pxelinux works and can load the kernel/initrd, >> and the kernel/initrd doesn't work. Is that what you said? > > Sorry for the confusion John. > > pxelinux works and it loads the vmlinuz and initrd. > > >> >> How does the BIOS know to load pxelinux, and where to get it? Doesn't >> that require dhcp (or at least bootp)? >> >> How does pxelinux get its menus and then the kernel? Doesn't that >> require dhcp (or at least bootp) too? >> >> And then the kernel/initrd can't get the IP address the same way. Have >> I got the facts straight? >> > > Actually I have dhcpd configured for static IP address. Not for a range > of IP addresses to pick from. dhcpd is running. After PXE loads up the > kernel, I think the installer anaconda uses dhcp to again look for the > ks.cfg file. This is after the menu. That's the part which does not > work. That's why I put in the static IP's in my default file under > label 1. Putting the static IP's don't work as well. > What I don't understand is how the BIOS and pxelinux can both get an IP address, and Anaconda can't. Have you tried using IPAPPEND instead of ip= on the kernel parameters. Read syslinux.doc for clarification. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From muksyed at stanford.edu Wed May 16 23:15:34 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Wed, 16 May 2007 16:15:34 -0700 Subject: Unable to get any http requests: Unabletoretreive netstg2.imgfile In-Reply-To: <464B8E5E.10706@herakles.homelinux.org> Message-ID: <000c01c79810$1e40f260$2b1b42ab@stanford.edu> Hi John, Anaconda depends on DHCP even if you use static IPs. I have made some progress on this... I have posted this on the group but here it is for you anyways (it's rather long): I did what Kennedy suggested. Thanks much. Connected the KVM and did the Alt-F commands. What I found is this: " ... ... getting kickstart file looking for first netDev with link eth0 has link, using it doing kickstart...setting it up waiting for link... 4 seconds no DNS servers, can't look up hostname ks location: http://171.67.39.130/ENT_LINUX_64/Enterprise/ks.cfg transferring http://171.67.39.130//./ENT_LINUX_64/Enterprise/ks.cfg to a fd failed to retrieve http://171.67.39.130///ENT_LINUX_64/Enterprise/ks.cfg trying to mount CD device hda " As I said before, I could retrieve all the above http links from another system on the same subnet as my kickstart install server and the client. After which I modified my default file to the following: label 1 kernel ENT_LINUX_64/vmlinuz append nofb initrd=ENT_LINUX_64/initrd.img ramdisk_size=10000 lang= devfs=nomount ramdisk_size=9216 ksdevice=link ip=171.67.39.139 netmask=255.255.255.224 gateway=171.67.39.129 dns=171.64.7.77 ks=http://171.67.39.130/ENT_LINUX_64/Enterprise/ks.cfg I put the dns= entry in there. The I get the following: " ... Adding 171.64.7.77 dnsserver is 171.64.7.77 doing kickstart...setting it up waiting for link 4 seconds reverse name lookup failed ks location: http://171.67.39.130/ENT_LINUX_64/Enterprise/ks.cfg transferring http://171.67.39.130//./ENT_LINUX_64/Enterprise/ks.cfg to a fd failed to retrieve http://171.67.39.130///ENT_LINUX_64/Enterprise/ks.cfg trying to mount CD device had " Here is the network information from my ks.cfg file: #Network information network --device=eth0 --bootproto=static --ip=171.67.39.130 --netmask=255.255.255.224 --gateway=171.67.39.129 --nameserver=171.64.7.77 --hostname=as-kickball.stanford.edu The hostname is invalid. It's a bogus name. I'll try to make it a valid name and check it out. Should there be a hostname in the default file as well since I am using static IP? I think I am getting close here. Something else is missing for the reverse lookup to work. Suggestions would be helpful at this point. Thanks -Mukarram. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John Summerfield Sent: Wednesday, May 16, 2007 4:06 PM To: Discussion list about Kickstart Subject: Re: Unable to get any http requests: Unabletoretreive netstg2.imgfile muksyed at stanford.edu wrote: > Quoting John Summerfield : > >> Mukarram Syed wrote: >>> John, >>> Sorry...just to make things clear. >>> eth0 works to get the dhcp/tftp request and eth1 does not. However, >>> I tried >>> using both is what I meant. >>> >>> We have a problem with dhcp, the networking folks here did not enable >>> "spanning tree" on the switch ports, until that happens I am using >>> static >>> ip. I double checked the syntax etc. >> >> I don't understand... pxelinux works and can load the kernel/initrd, >> and the kernel/initrd doesn't work. Is that what you said? > > Sorry for the confusion John. > > pxelinux works and it loads the vmlinuz and initrd. > > >> >> How does the BIOS know to load pxelinux, and where to get it? Doesn't >> that require dhcp (or at least bootp)? >> >> How does pxelinux get its menus and then the kernel? Doesn't that >> require dhcp (or at least bootp) too? >> >> And then the kernel/initrd can't get the IP address the same way. Have >> I got the facts straight? >> > > Actually I have dhcpd configured for static IP address. Not for a range > of IP addresses to pick from. dhcpd is running. After PXE loads up the > kernel, I think the installer anaconda uses dhcp to again look for the > ks.cfg file. This is after the menu. That's the part which does not > work. That's why I put in the static IP's in my default file under > label 1. Putting the static IP's don't work as well. > What I don't understand is how the BIOS and pxelinux can both get an IP address, and Anaconda can't. Have you tried using IPAPPEND instead of ip= on the kernel parameters. Read syslinux.doc for clarification. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From Colin.Coe at woodside.com.au Wed May 16 23:15:43 2007 From: Colin.Coe at woodside.com.au (Coe, Colin C. (Unix Engineer)) Date: Thu, 17 May 2007 07:15:43 +0800 Subject: Unable to get any httprequests:Unable toretrieve netstg2.imgfile In-Reply-To: Message-ID: What do your HTTPD logs look like? Do you see a GET for your ks.cfg in (assuming Apache HTTPD) access_log? Do you see any errors in error_log? CC ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com Sent: Thursday, 17 May 2007 5:19 AM To: Discussion list about Kickstart Subject: RE: Unable to get any httprequests:Unable toretrieve netstg2.imgfile I finally gave up and copied my ks.cfg to a CDROM and changed my pxelinux.cfg to load the KS cfg file from CD the drive. Anaconda could read the ks.cfg. The ks.cfg file pointed to a RedHat release image on an NFS server. However, Anaconda said it could not read the NFS release and displayed a screen to input the corrected NFS image information. The defaults from the KS cfg file where displayed as defaults. I entered nothing different and hit OK. Anaconda was able to now mount the RHEL image using the same information as was specified in the KS cfg file. I think Anaconda has a few bugs in handling network interfaces and protocols. I would love to figure out a fix, but I got to get some servers installed. I'll keep and eye out for your posts to see if you have any better luck. I still have not heard back from Redhat. So much for paid support. Get me back on Solaris!! kickstart-list-bounces at redhat.com wrote on 05/16/2007 11:03:13 AM: > I feel your pain. > I'll let you know if I find anything. > Thanks > > -Mukarram. > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > SSTinsley at upsfreight.com > Sent: Wednesday, May 16, 2007 6:00 AM > To: Discussion list about Kickstart > Subject: RE: Unable to get any http requests:Unable toretrieve > netstg2.imgfile > > > I have been fighting almost the identical problem. See the thread entitled > > "Re: Kickstart via PXE/NFS" > > After the kernel is loaded via configuration set up in the pxelinux config > file, the > network will not come up and the ks.cfg file can then not be loaded. I get > the > network config screen asking to use DHCP or to enter static info for setup > of eth0. It > has really bee frustrating. I finally just placed my ks.cfg on CDROM and was > able > to get the rest of the PXE auto install to complete. > > If you get a resolution, I will be very interested. I have had a case open > with > RedHat for 2 days waiting for a response. > > kickstart-list-bounces at redhat.com wrote on 05/16/2007 04:38:30 AM: > > > Nope. Nothing in the error logs. > > > > In my troubleshooting, this is what I've found... > > > > When I remove the ks=http:////ks.cfg from my "default" > file > > Client PXE boots, shows menu and loads vmlinuz/initrd. > > Then it shows up the Language selection. No luck. > > However, when I change my default file to read: ks=//ks.cfg, > > Client PXE boots, shows menu and loads vmlinuz/initrd. > > Then it gives the error "Error opening kickstart file (null): Bad address" > > I do some more research (see link below) > > http://linux.derkeiler.com/Mailing-Lists/Fedora/2004-06/5939.html > > and modified the ks.cfg file...add more network parameters like so: > > network --device=eth0 --bootproto=static --ip=171.67.39.139 > > --netmask=255.255.255.224 --gateway=171.67.39.129 --nameserver=171.64.7.77 > > --hostname=as-test.stanford.edu > > Still no luck. > > I then changed my --device=eth0 to --device=link and then to > > --device=eth1...still no luck. Then put it back to --device=eth0. > > Then I change my PATH from: > > url --url http://171.67.39.130/ENT_LINUX_64/Enterprise > > to > > url --url http://171.67.39.130/ENT_LINUX_64 > > Still no luck. > > Again nothing in the http logs. > > > > It looks like the installer is reading the ks.cfg file silently since it > > complained that it couldn't find it from the default file when I didn't > > specify the http://. > > Can I figure out what is reading and stopping. Is there a way to look at > > the kickstart logs (If there is any...I haven't seen one). > > > > I tried to create logging as well in my ks.cfg file...no luck...no output: > > %pre --log=/tmp/my-pre-log > > echo 'Hello, World!' > > > > Here is my directory/file setup: > > > > TFTP: > > > > /tftpboot/linux-install/pxelinux.cfg > > [root at as-kickball pxelinux.cfg]# ls > > default pxeos.xml > > > > [root at as-kickball linux-install]# ls -lR > > .: > > total 32 > > drwxr-xr-x 2 root root 4096 May 14 00:31 ENT_LINUX_32 > > drwxr-xr-x 2 root root 4096 May 16 00:30 ENT_LINUX_64 > > drwxr-xr-x 2 root root 4096 May 14 21:09 msgs > > -rwxr-xr-x 1 root root 12952 Oct 24 2006 pxelinux.0 > > drwxr-xr-x 2 root root 4096 May 16 00:25 pxelinux.cfg > > > > ./ENT_LINUX_32: > > total 0 > > > > ./ENT_LINUX_64: > > total 5984 > > -rwxr-xr-x 1 root root 4310479 May 14 15:29 initrd.img > > -rwxr-xr-x 1 root root 85 May 16 00:30 ks.cfg > > -rwxr-xr-x 1 root root 1792835 May 14 15:29 vmlinuz > > > > ./msgs: > > total 24 > > -rwxr-xr-x 1 root root 534 May 14 21:00 boot.msg > > -rwxr-xr-x 1 root root 668 Oct 24 2006 expert.msg > > -rwxr-xr-x 1 root root 871 Oct 24 2006 general.msg > > -rwxr-xr-x 1 root root 860 Oct 24 2006 param.msg > > -rwxr-xr-x 1 root root 530 Oct 24 2006 rescue.msg > > -rwxr-xr-x 1 root root 545 Oct 24 2006 snake.msg > > > > ./pxelinux.cfg: > > total 20 > > -rwxr-xr-x 1 root root 511 May 16 00:25 default > > -rwxr-xr-x 1 root root 511 May 15 23:34 default.bak > > -rwxr-xr-x 1 root root 461 May 15 14:44 default.dhcp > > -rwxr-xr-x 1 root root 573 May 15 13:30 default.staticIP > > -rwxr-xr-x 1 root root 42 May 14 16:28 pxeos.xml > > > > My HTTP directory: > > > > [root at as-kickball /]# ls -lart /var/www/html > > total 24 > > drwxr-xr-x 8 root root 4096 Mar 21 19:17 .. > > drwxr-xr-x 4 root root 4096 May 14 00:35 . > > drwxr-xr-x 3 root root 4096 May 14 00:40 ENT_LINUX_32 > > drwxr-xr-x 3 root root 4096 May 15 15:49 ENT_LINUX_64 > > > > [root at as-kickball /]# ls -lart /var/www/html/ENT_LINUX_64/Enterprise/ > > total 136 > > drwxr-xr-x 2 root root 114688 Feb 24 13:15 RPMS > > drwxr-xr-x 2 root root 4096 May 14 15:53 base > > -rwxr-xr-x 1 root root 1386 May 15 15:05 ks_ent_linux_64.cfg > > drwxr-xr-x 3 root root 4096 May 15 15:49 .. > > -rw-r--r-- 1 root root 1242 May 16 01:08 ks.cfg > > drwxr-xr-x 4 root root 4096 May 16 01:08 . > > > > I am at a loss now. I am sure it's something simple as it mostly is. > > Help is appreciated. > > > > Tomorrow, I'll tell my networking guy to enable "spanning tree" so I could > > use dhcp and get the static crap out of my way. > > Will let you know more. > > > > -Mukarram. > > > > > > -----Original Message----- > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mike Jennings > > Sent: Tuesday, May 15, 2007 8:47 PM > > To: Discussion list about Kickstart > > Subject: Re: Unable to get any http requests:Unable to retreive > > netstg2.imgfile > > > > Is the apache log showing any 404 errors? My stupid screwups are > > always shown in the error log when I typo something. > > > > Mike > > > > > > > > On May 15, 2007, at 10:39 PM, Mukarram Syed wrote: > > > > > > > > Colin, > > > I changed it from ksdevice=eth0 to ksdevice=link > > > > > > Sorry, that didn't help. > > > > > > Wonder what else could it be, why is it not connecting to the http > > > server? > > > I don't see any traffic in tcpdump after pxeboot. > > > > > > -Mukarram. > > > > > > -----Original Message----- > > > From: kickstart-list-bounces at redhat.com > > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin > > > C. (Unix > > > Engineer) > > > Sent: Tuesday, May 15, 2007 5:56 PM > > > To: Discussion list about Kickstart > > > Subject: RE: Unable to get any http requests:Unabletoretreive > > > netstg2.imgfile > > > > > > > > > Try changing 'ksdevice=eth0' to 'ksdevice=link'. With 2.6 kernels, > > > the > > > kernel's idea of what eth0 and eth1 are often differs from the BIOS/ > > > PXE > > > stack. > > > > > > CC > > > > > >> -----Original Message----- > > >> From: kickstart-list-bounces at redhat.com > > >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed > > >> Sent: Wednesday, 16 May 2007 8:42 AM > > >> To: 'Discussion list about Kickstart' > > >> Subject: RE: Unable to get any http requests: > > >> Unabletoretreive netstg2.imgfile > > >> > > >> John, > > >> Sorry...just to make things clear. > > >> eth0 works to get the dhcp/tftp request and eth1 does not. > > >> However, I tried using both is what I meant. > > >> > > >> We have a problem with dhcp, the networking folks here did > > >> not enable "spanning tree" on the switch ports, until that > > >> happens I am using static > > >> ip. I double checked the syntax etc. > > >> > > >> Thanks > > >> -Mukarram. > > >> > > >> -----Original Message----- > > >> From: kickstart-list-bounces at redhat.com > > >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John > > >> Summerfield > > >> Sent: Tuesday, May 15, 2007 5:32 PM > > >> To: Discussion list about Kickstart > > >> Subject: Re: Unable to get any http requests: Unable > > >> toretreive netstg2.imgfile > > >> > > >> Mukarram Syed wrote: > > >>> Thanks Colin for the quick response > > >>> > > >>> > > >>> > > >>> Yes. It does have. It's a Dell PE2950. > > >>> > > >>> > > >>> > > >>> I have already tried that option as well. > > >>> > > >>> I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. > > >>> > > >> > > >> If you can connect all network interfaces, that might help, > > >> because it then doesn't matter whether either one or both is used. > > >> > > >> If you run arpwatch that will report on which interfaces are > > >> used (by mac address and IP). > > >> > > >> More below... > > >> > > >>> > > >>> > > >>> -Mukarram. > > >>> > > >>> > > >>> > > >>> > > >>> > > >>> _____ > > >>> > > >>> From: kickstart-list-bounces at redhat.com > > >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > > >> Coe, Colin C. > > >> (Unix > > >>> Engineer) > > >>> Sent: Tuesday, May 15, 2007 4:52 PM > > >>> To: Discussion list about Kickstart > > >>> Subject: RE: Unable to get any http requests: Unable to retreive > > >>> netstg2.imgfile > > >>> > > >>> > > >>> > > >>> Does the machine you're on have multiple Ethernet > > >> interfaces? If so, > > >>> what you may find is adding 'ksdevice=link' will assist. > > >>> > > >>> > > >>> > > >>> CC > > >>> > > >>> > > >>> > > >>> > > >>> _____ > > >>> > > >>> > > >>> From: kickstart-list-bounces at redhat.com > > >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > > >> Mukarram Syed > > >>> Sent: Wednesday, 16 May 2007 7:50 AM > > >>> To: kickstart-list at redhat.com > > >>> Subject: Unable to get any http requests: Unable to retreive > > >>> netstg2.img file > > >>> > > >>> Hi > > >>> > > >>> I am sure there is some information regarding my request, but I am > > >>> unable > > >> to > > >>> find it, hence the post. > > >>> > > >>> > > >>> > > >>> I configured my kickstart server using http. > > >>> > > >>> I am able to pxeboot the client, but my client is not able > > >> to get the > > >>> netstg2.img file from the server. > > >>> > > >>> I am able to http into my netstg2.img directory from a web browser > > >>> from another server in the same subnet successfully. > > >>> > > >>> > > >>> > > >>> Also I tried to put in a boot cdrom and run the following: > > >>> > > >>> Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg > > >>> ip= > > >>> netmask=255.255.255.224 gateway= > > >> > > >> > > >> I have never used static IP for ks, I alway configure > > >> everything I can with dhcp. However, I do know that that's > > >> not the format the kernel expects for the ip parameter. Check > > >> your documentation, if I were writing a script tp parse ip= > > >> then I'd conform to the existing standards (that is, the way > > >> the kernel expects it), and that's what works with Ubuntu. > > >> > > >> > > >> > > >>> > > >>> > > >>> > > >>> Still no luck. It goes straight to the cd installer. > > >>> > > >>> > > >>> > > >>> I tried to tcpdump the traffic on the MAC address of the client, it > > >> outputs > > >>> the tftp information and after that it stops. > > >>> > > >>> I don't see any http traffic coming from the client. > > >> > > >> That's fairly consistent with you specifying ip= incorrectly. > > >> > > >> btw, if you're using PXE, the why are you trying to spectify > > >> static IP info? > > >> > > >> btw, as I recall, pxelinux has an option to pass on the IP > > >> address info it gets in the proper format for the kernel to parse. > > >> > > >> > > >>> > > >>> > > >>> > > >>> I'd appreciate some help in this regard. > > >>> > > >>> I've spent all day trying to troubleshoot this but no luck. > > >>> > > >>> > > >>> > > >>> Thanks Much > > >>> > > >>> > > >>> > > >>> > > >>> > > >>> Mukarram Syed > > >>> > > >>> Unix Systems Administrator, > > >>> > > >>> Stanford University. > > >>> NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- An HTML attachment was scrubbed... URL: From debian at herakles.homelinux.org Wed May 16 23:29:55 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Thu, 17 May 2007 07:29:55 +0800 Subject: Unable to get any http requests: Unabletoretreive netstg2.imgfile In-Reply-To: <000c01c79810$1e40f260$2b1b42ab@stanford.edu> References: <000c01c79810$1e40f260$2b1b42ab@stanford.edu> Message-ID: <464B93F3.7060404@herakles.homelinux.org> Mukarram Syed wrote: > Hi John, > > Anaconda depends on DHCP even if you use static IPs. > > I have made some progress on this... > > I have posted this on the group but here it is for you anyways (it's rather > long): I don't read the whole thread before responding to parts of it, and email arrives out of order, but I have read it now. I can't really help if you don't try my suggestions. There are good reasons to snip, and to not top-post. email isn't like paper mail, you can and should reply to points in context. >> Actually I have dhcpd configured for static IP address. Not for a range > >> of IP addresses to pick from. dhcpd is running. After PXE loads up the > >> kernel, I think the installer anaconda uses dhcp to again look for the > >> ks.cfg file. This is after the menu. That's the part which does not > >> work. That's why I put in the static IP's in my default file under > >> label 1. Putting the static IP's don't work as well. > > > > > What I don't understand is how the BIOS and pxelinux can both get an IP > > address, and Anaconda can't. > > > > Have you tried using IPAPPEND instead of ip= on the kernel parameters. > > Read syslinux.doc for clarification. > > > > > > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From Chip.Shabazian at bankofamerica.com Thu May 17 00:21:42 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Wed, 16 May 2007 17:21:42 -0700 Subject: Unable to get any http requests:Unable toretrieve netstg2.imgfile In-Reply-To: Message-ID: This is exactly what happens when you have a portfast issue on your switch. It appears your problem was infrastructure after all. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com Sent: Wednesday, May 16, 2007 2:19 PM To: Discussion list about Kickstart Subject: RE: Unable to get any http requests:Unable toretrieve netstg2.imgfile I finally gave up and copied my ks.cfg to a CDROM and changed my pxelinux.cfg to load the KS cfg file from CD the drive. Anaconda could read the ks.cfg. The ks.cfg file pointed to a RedHat release image on an NFS server. However, Anaconda said it could not read the NFS release and displayed a screen to input the corrected NFS image information. The defaults from the KS cfg file where displayed as defaults. I entered nothing different and hit OK. Anaconda was able to now mount the RHEL image using the same information as was specified in the KS cfg file. I think Anaconda has a few bugs in handling network interfaces and protocols. I would love to figure out a fix, but I got to get some servers installed. I'll keep and eye out for your posts to see if you have any better luck. I still have not heard back from Redhat. So much for paid support. Get me back on Solaris!! kickstart-list-bounces at redhat.com wrote on 05/16/2007 11:03:13 AM: > I feel your pain. > I'll let you know if I find anything. > Thanks > > -Mukarram. > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > SSTinsley at upsfreight.com > Sent: Wednesday, May 16, 2007 6:00 AM > To: Discussion list about Kickstart > Subject: RE: Unable to get any http requests:Unable toretrieve > netstg2.imgfile > > > I have been fighting almost the identical problem. See the thread entitled > > "Re: Kickstart via PXE/NFS" > > After the kernel is loaded via configuration set up in the pxelinux config > file, the > network will not come up and the ks.cfg file can then not be loaded. I get > the > network config screen asking to use DHCP or to enter static info for setup > of eth0. It > has really bee frustrating. I finally just placed my ks.cfg on CDROM and was > able > to get the rest of the PXE auto install to complete. > > If you get a resolution, I will be very interested. I have had a case open > with > RedHat for 2 days waiting for a response. > > kickstart-list-bounces at redhat.com wrote on 05/16/2007 04:38:30 AM: > > > Nope. Nothing in the error logs. > > > > In my troubleshooting, this is what I've found... > > > > When I remove the ks=http:////ks.cfg from my "default" > file > > Client PXE boots, shows menu and loads vmlinuz/initrd. > > Then it shows up the Language selection. No luck. > > However, when I change my default file to read: ks=//ks.cfg, > > Client PXE boots, shows menu and loads vmlinuz/initrd. > > Then it gives the error "Error opening kickstart file (null): Bad address" > > I do some more research (see link below) > > http://linux.derkeiler.com/Mailing-Lists/Fedora/2004-06/5939.html > > and modified the ks.cfg file...add more network parameters like so: > > network --device=eth0 --bootproto=static --ip=171.67.39.139 > > --netmask=255.255.255.224 --gateway=171.67.39.129 --nameserver=171.64.7.77 > > --hostname=as-test.stanford.edu > > Still no luck. > > I then changed my --device=eth0 to --device=link and then to > > --device=eth1...still no luck. Then put it back to --device=eth0. > > Then I change my PATH from: > > url --url http://171.67.39.130/ENT_LINUX_64/Enterprise > > to > > url --url http://171.67.39.130/ENT_LINUX_64 > > Still no luck. > > Again nothing in the http logs. > > > > It looks like the installer is reading the ks.cfg file silently since it > > complained that it couldn't find it from the default file when I didn't > > specify the http://. > > Can I figure out what is reading and stopping. Is there a way to look at > > the kickstart logs (If there is any...I haven't seen one). > > > > I tried to create logging as well in my ks.cfg file...no luck...no output: > > %pre --log=/tmp/my-pre-log > > echo 'Hello, World!' > > > > Here is my directory/file setup: > > > > TFTP: > > > > /tftpboot/linux-install/pxelinux.cfg > > [root at as-kickball pxelinux.cfg]# ls > > default pxeos.xml > > > > [root at as-kickball linux-install]# ls -lR > > .: > > total 32 > > drwxr-xr-x 2 root root 4096 May 14 00:31 ENT_LINUX_32 > > drwxr-xr-x 2 root root 4096 May 16 00:30 ENT_LINUX_64 > > drwxr-xr-x 2 root root 4096 May 14 21:09 msgs > > -rwxr-xr-x 1 root root 12952 Oct 24 2006 pxelinux.0 > > drwxr-xr-x 2 root root 4096 May 16 00:25 pxelinux.cfg > > > > ./ENT_LINUX_32: > > total 0 > > > > ./ENT_LINUX_64: > > total 5984 > > -rwxr-xr-x 1 root root 4310479 May 14 15:29 initrd.img > > -rwxr-xr-x 1 root root 85 May 16 00:30 ks.cfg > > -rwxr-xr-x 1 root root 1792835 May 14 15:29 vmlinuz > > > > ./msgs: > > total 24 > > -rwxr-xr-x 1 root root 534 May 14 21:00 boot.msg > > -rwxr-xr-x 1 root root 668 Oct 24 2006 expert.msg > > -rwxr-xr-x 1 root root 871 Oct 24 2006 general.msg > > -rwxr-xr-x 1 root root 860 Oct 24 2006 param.msg > > -rwxr-xr-x 1 root root 530 Oct 24 2006 rescue.msg > > -rwxr-xr-x 1 root root 545 Oct 24 2006 snake.msg > > > > ./pxelinux.cfg: > > total 20 > > -rwxr-xr-x 1 root root 511 May 16 00:25 default > > -rwxr-xr-x 1 root root 511 May 15 23:34 default.bak > > -rwxr-xr-x 1 root root 461 May 15 14:44 default.dhcp > > -rwxr-xr-x 1 root root 573 May 15 13:30 default.staticIP > > -rwxr-xr-x 1 root root 42 May 14 16:28 pxeos.xml > > > > My HTTP directory: > > > > [root at as-kickball /]# ls -lart /var/www/html > > total 24 > > drwxr-xr-x 8 root root 4096 Mar 21 19:17 .. > > drwxr-xr-x 4 root root 4096 May 14 00:35 . > > drwxr-xr-x 3 root root 4096 May 14 00:40 ENT_LINUX_32 > > drwxr-xr-x 3 root root 4096 May 15 15:49 ENT_LINUX_64 > > > > [root at as-kickball /]# ls -lart /var/www/html/ENT_LINUX_64/Enterprise/ > > total 136 > > drwxr-xr-x 2 root root 114688 Feb 24 13:15 RPMS > > drwxr-xr-x 2 root root 4096 May 14 15:53 base > > -rwxr-xr-x 1 root root 1386 May 15 15:05 ks_ent_linux_64.cfg > > drwxr-xr-x 3 root root 4096 May 15 15:49 .. > > -rw-r--r-- 1 root root 1242 May 16 01:08 ks.cfg > > drwxr-xr-x 4 root root 4096 May 16 01:08 . > > > > I am at a loss now. I am sure it's something simple as it mostly is. > > Help is appreciated. > > > > Tomorrow, I'll tell my networking guy to enable "spanning tree" so I could > > use dhcp and get the static crap out of my way. > > Will let you know more. > > > > -Mukarram. > > > > > > -----Original Message----- > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mike Jennings > > Sent: Tuesday, May 15, 2007 8:47 PM > > To: Discussion list about Kickstart > > Subject: Re: Unable to get any http requests:Unable to retreive > > netstg2.imgfile > > > > Is the apache log showing any 404 errors? My stupid screwups are > > always shown in the error log when I typo something. > > > > Mike > > > > > > > > On May 15, 2007, at 10:39 PM, Mukarram Syed wrote: > > > > > > > > Colin, > > > I changed it from ksdevice=eth0 to ksdevice=link > > > > > > Sorry, that didn't help. > > > > > > Wonder what else could it be, why is it not connecting to the http > > > server? > > > I don't see any traffic in tcpdump after pxeboot. > > > > > > -Mukarram. > > > > > > -----Original Message----- > > > From: kickstart-list-bounces at redhat.com > > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin > > > C. (Unix > > > Engineer) > > > Sent: Tuesday, May 15, 2007 5:56 PM > > > To: Discussion list about Kickstart > > > Subject: RE: Unable to get any http requests:Unabletoretreive > > > netstg2.imgfile > > > > > > > > > Try changing 'ksdevice=eth0' to 'ksdevice=link'. With 2.6 kernels, > > > the > > > kernel's idea of what eth0 and eth1 are often differs from the BIOS/ > > > PXE > > > stack. > > > > > > CC > > > > > >> -----Original Message----- > > >> From: kickstart-list-bounces at redhat.com > > >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mukarram Syed > > >> Sent: Wednesday, 16 May 2007 8:42 AM > > >> To: 'Discussion list about Kickstart' > > >> Subject: RE: Unable to get any http requests: > > >> Unabletoretreive netstg2.imgfile > > >> > > >> John, > > >> Sorry...just to make things clear. > > >> eth0 works to get the dhcp/tftp request and eth1 does not. > > >> However, I tried using both is what I meant. > > >> > > >> We have a problem with dhcp, the networking folks here did > > >> not enable "spanning tree" on the switch ports, until that > > >> happens I am using static > > >> ip. I double checked the syntax etc. > > >> > > >> Thanks > > >> -Mukarram. > > >> > > >> -----Original Message----- > > >> From: kickstart-list-bounces at redhat.com > > >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John > > >> Summerfield > > >> Sent: Tuesday, May 15, 2007 5:32 PM > > >> To: Discussion list about Kickstart > > >> Subject: Re: Unable to get any http requests: Unable > > >> toretreive netstg2.imgfile > > >> > > >> Mukarram Syed wrote: > > >>> Thanks Colin for the quick response > > >>> > > >>> > > >>> > > >>> Yes. It does have. It's a Dell PE2950. > > >>> > > >>> > > >>> > > >>> I have already tried that option as well. > > >>> > > >>> I tried both ksdevice=eth0 and ksdevice=eth1 and they don't work. > > >>> > > >> > > >> If you can connect all network interfaces, that might help, > > >> because it then doesn't matter whether either one or both is used. > > >> > > >> If you run arpwatch that will report on which interfaces are > > >> used (by mac address and IP). > > >> > > >> More below... > > >> > > >>> > > >>> > > >>> -Mukarram. > > >>> > > >>> > > >>> > > >>> > > >>> > > >>> _____ > > >>> > > >>> From: kickstart-list-bounces at redhat.com > > >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > > >> Coe, Colin C. > > >> (Unix > > >>> Engineer) > > >>> Sent: Tuesday, May 15, 2007 4:52 PM > > >>> To: Discussion list about Kickstart > > >>> Subject: RE: Unable to get any http requests: Unable to retreive > > >>> netstg2.imgfile > > >>> > > >>> > > >>> > > >>> Does the machine you're on have multiple Ethernet > > >> interfaces? If so, > > >>> what you may find is adding 'ksdevice=link' will assist. > > >>> > > >>> > > >>> > > >>> CC > > >>> > > >>> > > >>> > > >>> > > >>> _____ > > >>> > > >>> > > >>> From: kickstart-list-bounces at redhat.com > > >>> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > > >> Mukarram Syed > > >>> Sent: Wednesday, 16 May 2007 7:50 AM > > >>> To: kickstart-list at redhat.com > > >>> Subject: Unable to get any http requests: Unable to retreive > > >>> netstg2.img file > > >>> > > >>> Hi > > >>> > > >>> I am sure there is some information regarding my request, but I am > > >>> unable > > >> to > > >>> find it, hence the post. > > >>> > > >>> > > >>> > > >>> I configured my kickstart server using http. > > >>> > > >>> I am able to pxeboot the client, but my client is not able > > >> to get the > > >>> netstg2.img file from the server. > > >>> > > >>> I am able to http into my netstg2.img directory from a web browser > > >>> from another server in the same subnet successfully. > > >>> > > >>> > > >>> > > >>> Also I tried to put in a boot cdrom and run the following: > > >>> > > >>> Linux ks=http:///ENT_LINUX_64/Enterprise/ks.cfg > > >>> ip= > > >>> netmask=255.255.255.224 gateway= > > >> > > >> > > >> I have never used static IP for ks, I alway configure > > >> everything I can with dhcp. However, I do know that that's > > >> not the format the kernel expects for the ip parameter. Check > > >> your documentation, if I were writing a script tp parse ip= > > >> then I'd conform to the existing standards (that is, the way > > >> the kernel expects it), and that's what works with Ubuntu. > > >> > > >> > > >> > > >>> > > >>> > > >>> > > >>> Still no luck. It goes straight to the cd installer. > > >>> > > >>> > > >>> > > >>> I tried to tcpdump the traffic on the MAC address of the client, it > > >> outputs > > >>> the tftp information and after that it stops. > > >>> > > >>> I don't see any http traffic coming from the client. > > >> > > >> That's fairly consistent with you specifying ip= incorrectly. > > >> > > >> btw, if you're using PXE, the why are you trying to spectify > > >> static IP info? > > >> > > >> btw, as I recall, pxelinux has an option to pass on the IP > > >> address info it gets in the proper format for the kernel to parse. > > >> > > >> > > >>> > > >>> > > >>> > > >>> I'd appreciate some help in this regard. > > >>> > > >>> I've spent all day trying to troubleshoot this but no luck. > > >>> > > >>> > > >>> > > >>> Thanks Much > > >>> > > >>> > > >>> > > >>> > > >>> > > >>> Mukarram Syed > > >>> > > >>> Unix Systems Administrator, > > >>> > > >>> Stanford University. > > >>> > > >>> > > >>> > > >>> > > >>> NOTICE: This email and any attachments are confidential. > > >>> They may contain legally privileged information or > > >>> copyright material. You must not read, copy, use or > > >>> disclose them without authorisation. If you are not an > > >>> intended recipient, please contact us at once by return > > >>> email and then delete both messages and all attachments. > > >>> > > >>> > > >>> > > >>> > > >> -------------------------------------------------------------- > > >> ---------- > > >>> > > >>> _______________________________________________ > > >>> Kickstart-list mailing list > > >>> Kickstart-list at redhat.com > > >>> https://www.redhat.com/mailman/listinfo/kickstart-list > > >> > > >> > > >> -- > > >> > > >> Cheers > > >> John > > >> > > >> -- spambait > > >> 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu > > >> > > >> Please do not reply off-list > > >> > > >> _______________________________________________ > > >> Kickstart-list mailing list > > >> Kickstart-list at redhat.com > > >> https://www.redhat.com/mailman/listinfo/kickstart-list > > >> > > >> _______________________________________________ > > >> Kickstart-list mailing list > > >> Kickstart-list at redhat.com > > >> https://www.redhat.com/mailman/listinfo/kickstart-list > > >> > > > > > > NOTICE: This email and any attachments are confidential. > > > They may contain legally privileged information or > > > copyright material. You must not read, copy, use or > > > disclose them without authorisation. If you are not an > > > intended recipient, please contact us at once by return > > > email and then delete both messages and all attachments. > > > > > > _______________________________________________ > > > Kickstart-list mailing list > > > Kickstart-list at redhat.com > > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > > > _______________________________________________ > > > Kickstart-list mailing list > > > Kickstart-list at redhat.com > > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT > IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED > ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the > intended recipient of this document, you are notified that any review, > dissemination, distribution or copying of this communication is prohibited. > If you have received this communication in error, please notify me > immediately by return email, delete the electronic message and destroy any > printed copies. Thank you for your cooperation. > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Thu May 17 00:28:48 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Wed, 16 May 2007 17:28:48 -0700 Subject: Unable to get any http requests: Unable toretreive netstg2.imgfile In-Reply-To: <464B8E5E.10706@herakles.homelinux.org> Message-ID: The reason you can get an IP and PXEboot is because the NIC is up for more than 30 seconds, which gives the switch time to negotiate with the NIC and turn up the port. When anaconda starts, it recycles the NIC (to load the driver). If you have portfast issues, anaconda will timeout before your switch makes the port live. Ananacoda will also recycle the NIC one more time after it gets the ks.cfg, but that one never causes portfast timeout issues. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John Summerfield Sent: Wednesday, May 16, 2007 4:06 PM To: Discussion list about Kickstart Subject: Re: Unable to get any http requests: Unable toretreive netstg2.imgfile muksyed at stanford.edu wrote: > Quoting John Summerfield : > >> Mukarram Syed wrote: >>> John, >>> Sorry...just to make things clear. >>> eth0 works to get the dhcp/tftp request and eth1 does not. However, >>> I tried using both is what I meant. >>> >>> We have a problem with dhcp, the networking folks here did not >>> enable "spanning tree" on the switch ports, until that happens I am >>> using static >>> ip. I double checked the syntax etc. >> >> I don't understand... pxelinux works and can load the kernel/initrd, >> and the kernel/initrd doesn't work. Is that what you said? > > Sorry for the confusion John. > > pxelinux works and it loads the vmlinuz and initrd. > > >> >> How does the BIOS know to load pxelinux, and where to get it? Doesn't >> that require dhcp (or at least bootp)? >> >> How does pxelinux get its menus and then the kernel? Doesn't that >> require dhcp (or at least bootp) too? >> >> And then the kernel/initrd can't get the IP address the same way. >> Have I got the facts straight? >> > > Actually I have dhcpd configured for static IP address. Not for a > range of IP addresses to pick from. dhcpd is running. After PXE > loads up the kernel, I think the installer anaconda uses dhcp to again > look for the ks.cfg file. This is after the menu. That's the part > which does not work. That's why I put in the static IP's in my > default file under label 1. Putting the static IP's don't work as well. > What I don't understand is how the BIOS and pxelinux can both get an IP address, and Anaconda can't. Have you tried using IPAPPEND instead of ip= on the kernel parameters. Read syslinux.doc for clarification. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From SSTinsley at upsfreight.com Thu May 17 11:20:11 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Thu, 17 May 2007 07:20:11 -0400 Subject: Unable to get any http requests: Unable toretreive netstg2.imgfile In-Reply-To: Message-ID: kickstart-list-bounces at redhat.com wrote on 05/16/2007 08:28:48 PM: > The reason you can get an IP and PXEboot is because the NIC is up for > more than 30 seconds, which gives the switch time to negotiate with the > NIC and turn up the port. When anaconda starts, it recycles the NIC (to > load the driver). If you have portfast issues, anaconda will timeout > before your switch makes the port live. > > Ananacoda will also recycle the NIC one more time after it gets the > ks.cfg, but that one never causes portfast timeout issues. > I here what your saying. I took your suggestion to the network guys and they confirmed the switch is configured correctly. So short of calling them liars, I am pretty much at a standstill and have to go with the ks.cfg on CDROM. Thanks for your time. NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From klandreth at theplanet.com Thu May 17 12:46:19 2007 From: klandreth at theplanet.com (Landreth, Kevin ) Date: Thu, 17 May 2007 07:46:19 -0500 Subject: Unable to get any httprequests: Unable toretreive netstg2.imgfile In-Reply-To: References: Message-ID: <1179405980.8184.3.camel@kevin-desktop> On Thu, 2007-05-17 at 07:20 -0400, SSTinsley at upsfreight.com wrote: > kickstart-list-bounces at redhat.com wrote on 05/16/2007 08:28:48 PM: > > The reason you can get an IP and PXEboot is because the NIC is up > for > > more than 30 seconds, which gives the switch time to negotiate with > the > > NIC and turn up the port. When anaconda starts, it recycles the NIC > (to > > load the driver). If you have portfast issues, anaconda will > timeout > > before your switch makes the port live. > > > > Ananacoda will also recycle the NIC one more time after it gets the > > ks.cfg, but that one never causes portfast timeout issues. > > > > I here what your saying. I took your suggestion to the network guys > and they confirmed the > switch is configured correctly. So short of calling them liars, I am > pretty much at a > standstill and have to go with the ks.cfg on CDROM. you may want to add "nicdelay=40 linksleep=30" to you kernel command line. I didn't see that in any of your posts so I thought I'd mention it. https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189795 Note: you can add these to your ifcfg-eth0 conf file as well # No portfast, wait 30s LINKDELAY=30 # DISABLE stupid ZeroConf address NOZEROCONF=1 I've been on this list for a long time and been quite mostly but this is a long thread and I thought I would hop in share my thoughts with the rest of the Internet as well :) I'm not calling your networking team a liar, but in the same effect, you appear to be experiencing an issue where linkdelay and nicdelay my come in handy for you (works for both static and dhcp) -- - Kevin Landreth - RHCE - Sr. Systems Architect - The Planet (https://www.theplanet.com) -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From Chip.Shabazian at bankofamerica.com Thu May 17 15:38:31 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Thu, 17 May 2007 08:38:31 -0700 Subject: Unable to get any httprequests: Unable toretreive netstg2.imgfile In-Reply-To: <1179405980.8184.3.camel@kevin-desktop> Message-ID: In addition to: nicdelay=40 linksleep=30 adding: eth0_ethtool="autoneg=on speed=1000 duplex=full" (for a GB NIC) can help when the issue is portfast, as it seems to be 90% of the time. For 100MB, just adjust the settings appropriately From SSTinsley at upsfreight.com Thu May 17 16:12:11 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Thu, 17 May 2007 12:12:11 -0400 Subject: Unable to get any httprequests: Unable toretreive netstg2.imgfile In-Reply-To: <1179405980.8184.3.camel@kevin-desktop> Message-ID: kickstart-list-bounces at redhat.com wrote on 05/17/2007 08:46:19 AM: > On Thu, 2007-05-17 at 07:20 -0400, SSTinsley at upsfreight.com wrote: > > kickstart-list-bounces at redhat.com wrote on 05/16/2007 08:28:48 PM: > > > The reason you can get an IP and PXEboot is because the NIC is up > > for > > > more than 30 seconds, which gives the switch time to negotiate with > > the > > > NIC and turn up the port. When anaconda starts, it recycles the NIC > > (to > > > load the driver). If you have portfast issues, anaconda will > > timeout > > > before your switch makes the port live. > > > > > > Ananacoda will also recycle the NIC one more time after it gets the > > > ks.cfg, but that one never causes portfast timeout issues. > > > > > > > I here what your saying. I took your suggestion to the network guys > > and they confirmed the > > switch is configured correctly. So short of calling them liars, I am > > pretty much at a > > standstill and have to go with the ks.cfg on CDROM. > > you may want to add "nicdelay=40 linksleep=30" to you kernel command > line. > > I didn't see that in any of your posts so I thought I'd mention it. > https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189795 > > Note: you can add these to your ifcfg-eth0 conf file as well > # No portfast, wait 30s > LINKDELAY=30 > # DISABLE stupid ZeroConf address > NOZEROCONF=1 > > > I've been on this list for a long time and been quite mostly but this is > a long thread and I thought I would hop in share my thoughts with the > rest of the Internet as well :) > > I'm not calling your networking team a liar, but in the same effect, you > appear to be experiencing an issue where linkdelay and nicdelay my come > in handy for you (works for both static and dhcp) > > -- > - Kevin Landreth - RHCE > - Sr. Systems Architect While this may solve the problem (and I will try it if I have time), I think the thread shows that there is an issue. The fact that DHCP works fine during the PXE boot and download of the kernel and fails in Anaconda should be reason enough to a fix to be made. Make the DHCP operate consistently during the whole process. Thanks for your help NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Thu May 17 17:11:27 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Thu, 17 May 2007 10:11:27 -0700 Subject: Unable to get any httprequests: Unable toretreive netstg2.imgfile In-Reply-To: Message-ID: While I completely agree that this is something that has been a hassle and has been the source of more problems than anything I've seen related to kickstarting systems, we need to keep in mind that these workarounds are being put into place to deal with a "feature" of the switch. Wouldn't it be better for Cisco to fix STP so that if it saw a DHCP packet, it would forward that packet while it determines if it's going to allow that port on the network? I can't imagine any loops being created by forwarding and responding to DHCP packets while ensuring there are no loops. We always see this problem on Cisco gear, and in the past 4+ years, I've only seen this issue as a possibility on one other type of switch (but the person never responded to the list on whether or not the known Cisco workarounds helped), yet people think this is an "anaconda" or "kickstart" problem, not a "Cisco" problem. Now, *I* sure would like to see the timeout for pump increased from 30 seconds to 60 seconds, but that would probably piss off even MORE people since I'm sure there are a lot more people who get frustrated waiting 30 seconds for pump to timeout during boot if their dhcp server is down, or if they have more than one interface and don't know how to turn off dhcp on the unused interfaces, etc. I guess the ideal solution would be an option for pump that does increase the timeout that could be passed from anaconda, or an option in anaconda such as dhcpretry=X where you could set the system to retry dhcp X number of times without recycling the NIC and starting the negotiation dance again. But at the end of the day, please keep in mind that these are all work arounds for the Cisco STP feature, that while valuable, is the actual root cause. Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com Sent: Thursday, May 17, 2007 9:12 AM To: Discussion list about Kickstart Subject: RE: Unable to get any httprequests: Unable toretreive netstg2.imgfile kickstart-list-bounces at redhat.com wrote on 05/17/2007 08:46:19 AM: > On Thu, 2007-05-17 at 07:20 -0400, SSTinsley at upsfreight.com wrote: > > kickstart-list-bounces at redhat.com wrote on 05/16/2007 08:28:48 PM: > > > The reason you can get an IP and PXEboot is because the NIC is up > > for > > > more than 30 seconds, which gives the switch time to negotiate with > > the > > > NIC and turn up the port. When anaconda starts, it recycles the NIC > > (to > > > load the driver). If you have portfast issues, anaconda will > > timeout > > > before your switch makes the port live. > > > > > > Ananacoda will also recycle the NIC one more time after it gets the > > > ks.cfg, but that one never causes portfast timeout issues. > > > > > > > I here what your saying. I took your suggestion to the network guys > > and they confirmed the > > switch is configured correctly. So short of calling them liars, I am > > pretty much at a > > standstill and have to go with the ks.cfg on CDROM. > > you may want to add "nicdelay=40 linksleep=30" to you kernel command > line. > > I didn't see that in any of your posts so I thought I'd mention it. > https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189795 > > Note: you can add these to your ifcfg-eth0 conf file as well > # No portfast, wait 30s > LINKDELAY=30 > # DISABLE stupid ZeroConf address > NOZEROCONF=1 > > > I've been on this list for a long time and been quite mostly but this is > a long thread and I thought I would hop in share my thoughts with the > rest of the Internet as well :) > > I'm not calling your networking team a liar, but in the same effect, you > appear to be experiencing an issue where linkdelay and nicdelay my come > in handy for you (works for both static and dhcp) > > -- > - Kevin Landreth - RHCE > - Sr. Systems Architect While this may solve the problem (and I will try it if I have time), I think the thread shows that there is an issue. The fact that DHCP works fine during the PXE boot and download of the kernel and fails in Anaconda should be reason enough to a fix to be made. Make the DHCP operate consistently during the whole process. Thanks for your help NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From amm at lanl.gov Thu May 17 17:22:21 2007 From: amm at lanl.gov (Aaron M Morrison) Date: Thu, 17 May 2007 11:22:21 -0600 Subject: Unable to get any httprequests: Unable toretreive netstg2.imgfile In-Reply-To: References: Message-ID: <1179422541.7224.22.camel@itchy> SSTinsley at upsfreight.com; Apologies if I've missed part of the thread, but does the hardware you are attempting to pxeboot/ks have more than one nic? Sometimes you need to put --dev=eth(N) as the install kernel might probe and set up the device in a different order than the bios... bios device 1 might show up as eth(N) to the installer... so you can get a ip via pxeboot, download installer, etc, but anaconda set's up eth0 as the other nic, so it times out... using --dev=eth(N) usually fixes this. On Thu, 2007-05-17 at 12:12 -0400, SSTinsley at upsfreight.com wrote: > > > kickstart-list-bounces at redhat.com wrote on 05/17/2007 08:46:19 AM: > > > On Thu, 2007-05-17 at 07:20 -0400, SSTinsley at upsfreight.com wrote: > > > kickstart-list-bounces at redhat.com wrote on 05/16/2007 08:28:48 PM: > > > > The reason you can get an IP and PXEboot is because the NIC is > up > > > for > > > > more than 30 seconds, which gives the switch time to negotiate > with > > > the > > > > NIC and turn up the port. When anaconda starts, it recycles the > NIC > > > (to > > > > load the driver). If you have portfast issues, anaconda will > > > timeout > > > > before your switch makes the port live. > > > > > > > > Ananacoda will also recycle the NIC one more time after it gets > the > > > > ks.cfg, but that one never causes portfast timeout issues. > > > > > > > > > > I here what your saying. I took your suggestion to the network > guys > > > and they confirmed the > > > switch is configured correctly. So short of calling them liars, I > am > > > pretty much at a > > > standstill and have to go with the ks.cfg on CDROM. > > > > you may want to add "nicdelay=40 linksleep=30" to you kernel command > > line. > > > > I didn't see that in any of your posts so I thought I'd mention it. > > https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189795 > > > > Note: you can add these to your ifcfg-eth0 conf file as well > > # No portfast, wait 30s > > LINKDELAY=30 > > # DISABLE stupid ZeroConf address > > NOZEROCONF=1 > > > > > > I've been on this list for a long time and been quite mostly but > this is > > a long thread and I thought I would hop in share my thoughts with > the > > rest of the Internet as well :) > > > > I'm not calling your networking team a liar, but in the same effect, > you > > appear to be experiencing an issue where linkdelay and nicdelay my > come > > in handy for you (works for both static and dhcp) > > > > -- > > - Kevin Landreth - RHCE > > - Sr. Systems Architect > > > While this may solve the problem (and I will try it if I have time), I > think > the thread shows that there is an issue. The fact that DHCP works fine > during > the PXE boot and download of the kernel and fails in Anaconda should > be > reason enough to a fix to be made. Make the DHCP operate consistently > during the > whole process. > > Thanks for your help > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC > INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR > ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE > IT. If you are not the intended recipient of this document, you are > notified that any review, dissemination, distribution or copying of > this communication is prohibited. If you have received this > communication in error, please notify me immediately by return email, > delete the electronic message and destroy any printed copies. Thank > you for your cooperation. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From SSTinsley at upsfreight.com Thu May 17 17:31:57 2007 From: SSTinsley at upsfreight.com (SSTinsley at upsfreight.com) Date: Thu, 17 May 2007 13:31:57 -0400 Subject: Unable to get any httprequests: Unable toretreive netstg2.imgfile In-Reply-To: <1179422541.7224.22.camel@itchy> Message-ID: kickstart-list-bounces at redhat.com wrote on 05/17/2007 01:22:21 PM: > SSTinsley at upsfreight.com; > Apologies if I've missed part of the thread, but > does the hardware you are attempting to pxeboot/ks have more than one > nic? > Sometimes you need to put --dev=eth(N) as the install kernel might probe > and set up the device in a different order than the bios... > > bios device 1 might show up as eth(N) to the installer... > > so you can get a ip via pxeboot, download installer, etc, but anaconda > set's up eth0 as the other nic, so it times out... > using --dev=eth(N) usually fixes this. > Yes these suggestions were made earlier and tried. I disabled the second NIC in the BIOS and still had the problems. NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From muksyed at stanford.edu Thu May 17 18:02:09 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Thu, 17 May 2007 11:02:09 -0700 Subject: Unable toget any httprequests: Unable toretreive netstg2.imgfile In-Reply-To: Message-ID: <00ab01c798ad$7d90e340$2b1b42ab@stanford.edu> Hi All, Since I started this thread, let me finish it. I got it to work. Thanks to everyone in this list. I couldn't have done it without your help. There were 3 problems: 1. Spanning Tree/portfast was not enabled on the switch ports. 2. I was working on Dell PE2950's. They have a Link Enumeration issue. NIC Port1 becomes eth1 and NIC port2 becomes eth0. Hence, what I did was this in my default file: label 1 kernel ENT_LINUX_64/vmlinuz append nofb ks=http://171.67.39.130/ENT_LINUX_64/Enterprise/ks.cfg initrd=ENT_LINUX_64/initrd.img ramdisk_size=10000 lang= devfs=nomount ramdisk_size=9216 ksdevice=eth1 console=ttyS0,9600 ipappend 1 I changed ksdevice=link to ksdevice=eth1 3. The third problem was I had to change the following link in my ks.cfg file to reflect eth1 instead of eth0 network --bootproto=dhcp --device=eth1 That was it Thanks again. -Mukarram. _____ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of SSTinsley at upsfreight.com Sent: Thursday, May 17, 2007 10:32 AM To: amm at lanl.gov; Discussion list about Kickstart Subject: RE: Unable toget any httprequests: Unable toretreive netstg2.imgfile kickstart-list-bounces at redhat.com wrote on 05/17/2007 01:22:21 PM: > SSTinsley at upsfreight.com; > Apologies if I've missed part of the thread, but > does the hardware you are attempting to pxeboot/ks have more than one > nic? > Sometimes you need to put --dev=eth(N) as the install kernel might probe > and set up the device in a different order than the bios... > > bios device 1 might show up as eth(N) to the installer... > > so you can get a ip via pxeboot, download installer, etc, but anaconda > set's up eth0 as the other nic, so it times out... > using --dev=eth(N) usually fixes this. > Yes these suggestions were made earlier and tried. I disabled the second NIC in the BIOS and still had the problems. NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ebrown at lanl.gov Thu May 17 18:39:13 2007 From: ebrown at lanl.gov (Ed Brown) Date: Thu, 17 May 2007 12:39:13 -0600 Subject: Unable to get any httprequests: Unable toretreive netstg2.imgfile In-Reply-To: References: Message-ID: <464CA151.1060207@lanl.gov> Thanks for bringing the historical perspective to this question, it has indeed been kicked around for years now. But I disagree with the assertion that it's only a Cisco problem. I remember at least two other switch manufacturers involved, including my direct experience with 3com gigabit switches. Also, it hasn't only caused problems for pxe/dhcp installs, but also static ip installs using http and ftp. The only common factors seem to have been gigabit switches/nics, STP, and anaconda. Wherever the root causes lay, it's a problem during redhat installations, and no other circumstances, that I know about anyway. If it CAN be fixed or worked around in anaconda, then it should be, if only for the sake of RedHat's reputation and their customers' experience. While I don't understand all the issues involved, I do appreciate the recent options intended to help in these situations, and hope your additional suggestions are acted on too. -Ed Shabazian, Chip wrote: > > While I completely agree that this is something that has been a hassle > and has been the source of more problems than anything I've seen related > to kickstarting systems, we need to keep in mind that these workarounds > are being put into place to deal with a "feature" of the switch. > Wouldn't it be better for Cisco to fix STP so that if it saw a DHCP > packet, it would forward that packet while it determines if it's going > to allow that port on the network? I can't imagine any loops being > created by forwarding and responding to DHCP packets while ensuring > there are no loops. > > We always see this problem on Cisco gear, and in the past 4+ years, I've > only seen this issue as a possibility on one other type of switch (but > the person never responded to the list on whether or not the known Cisco > workarounds helped), yet people think this is an "anaconda" or > "kickstart" problem, not a "Cisco" problem. > > Now, *I* sure would like to see the timeout for pump increased from 30 > seconds to 60 seconds, but that would probably piss off even MORE people > since I'm sure there are a lot more people who get frustrated waiting 30 > seconds for pump to timeout during boot if their dhcp server is down, or > if they have more than one interface and don't know how to turn off dhcp > on the unused interfaces, etc. > > I guess the ideal solution would be an option for pump that does > increase the timeout that could be passed from anaconda, or an option in > anaconda such as dhcpretry=X where you could set the system to retry > dhcp X number of times without recycling the NIC and starting the > negotiation dance again. > > But at the end of the day, please keep in mind that these are all work > arounds for the Cisco STP feature, that while valuable, is the actual > root cause. > > Chip From debian at herakles.homelinux.org Thu May 17 22:28:07 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Fri, 18 May 2007 06:28:07 +0800 Subject: Unable to get any http requests: Unable toretreive netstg2.imgfile In-Reply-To: References: Message-ID: <464CD6F7.8020309@herakles.homelinux.org> Shabazian, Chip wrote: > The reason you can get an IP and PXEboot is because the NIC is up for > more than 30 seconds, which gives the switch time to negotiate with the > NIC and turn up the port. When anaconda starts, it recycles the NIC (to > load the driver). If you have portfast issues, anaconda will timeout > before your switch makes the port live. So there's an identifiable bug in Anaconda. Or somewhere. > > Ananacoda will also recycle the NIC one more time after it gets the > ks.cfg, but that one never causes portfast timeout issues. Did you try IPAPPEND? > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John Summerfield > Sent: Wednesday, May 16, 2007 4:06 PM > To: Discussion list about Kickstart > Subject: Re: Unable to get any http requests: Unable toretreive > netstg2.imgfile > > muksyed at stanford.edu wrote: >> Quoting John Summerfield : >> >>> Mukarram Syed wrote: >>>> John, >>>> Sorry...just to make things clear. >>>> eth0 works to get the dhcp/tftp request and eth1 does not. However, > >>>> I tried using both is what I meant. >>>> >>>> We have a problem with dhcp, the networking folks here did not >>>> enable "spanning tree" on the switch ports, until that happens I am >>>> using static >>>> ip. I double checked the syntax etc. >>> I don't understand... pxelinux works and can load the kernel/initrd, >>> and the kernel/initrd doesn't work. Is that what you said? >> Sorry for the confusion John. >> >> pxelinux works and it loads the vmlinuz and initrd. >> >> >>> How does the BIOS know to load pxelinux, and where to get it? Doesn't > >>> that require dhcp (or at least bootp)? >>> >>> How does pxelinux get its menus and then the kernel? Doesn't that >>> require dhcp (or at least bootp) too? >>> >>> And then the kernel/initrd can't get the IP address the same way. >>> Have I got the facts straight? >>> >> Actually I have dhcpd configured for static IP address. Not for a >> range of IP addresses to pick from. dhcpd is running. After PXE >> loads up the kernel, I think the installer anaconda uses dhcp to again > >> look for the ks.cfg file. This is after the menu. That's the part >> which does not work. That's why I put in the static IP's in my >> default file under label 1. Putting the static IP's don't work as > well. > > What I don't understand is how the BIOS and pxelinux can both get an IP > address, and Anaconda can't. > > Have you tried using IPAPPEND instead of ip= on the kernel parameters. > Read syslinux.doc for clarification. > > > -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From debian at herakles.homelinux.org Thu May 17 22:35:53 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Fri, 18 May 2007 06:35:53 +0800 Subject: Unable to get any http requests: Unable toretreive netstg2.imgfile In-Reply-To: References: Message-ID: <464CD8C9.8050501@herakles.homelinux.org> SSTinsley at upsfreight.com wrote: > kickstart-list-bounces at redhat.com wrote on 05/16/2007 08:28:48 PM: > >> The reason you can get an IP and PXEboot is because the NIC is up for >> more than 30 seconds, which gives the switch time to negotiate with the >> NIC and turn up the port. When anaconda starts, it recycles the NIC (to >> load the driver). If you have portfast issues, anaconda will timeout >> before your switch makes the port live. >> >> Ananacoda will also recycle the NIC one more time after it gets the >> ks.cfg, but that one never causes portfast timeout issues. >> > > I here what your saying. I took your suggestion to the network guys and > they confirmed the > switch is configured correctly. So short of calling them liars, I am > pretty much at a > standstill and have to go with the ks.cfg on CDROM. You don't have to call them rude names, and if you asked, "Is your switch configured properly?" then of course it is, but if you can explain "we have a problem, is there something we can try to get around it?" the response might be more accommodating. "Can we change timeouts?" It might be increasing the time before something times out won't have any effect in the ordinary course of events, but help in your case. 'Can we ...?" There's never only one correct configuration for anything computer-related, and if there's something not working there may be improvements available. > > Thanks for your time. > > NOTE: THIS DOCUMENT MAY CONTAIN CONFIDENTIAL AND NONPUBLIC INFORMATION. IT IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL(S) OR ENTITY(IES) NAMED ABOVE, AND OTHERS SPECIFICALLY AUTHORIZED TO RECEIVE IT. If you are not the intended recipient of this document, you are notified that any review, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify me immediately by return email, delete the electronic message and destroy any printed copies. Thank you for your cooperation. > > > ------------------------------------------------------------------------ > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From debian at herakles.homelinux.org Thu May 17 22:44:37 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Fri, 18 May 2007 06:44:37 +0800 Subject: Unable to get any httprequests: Unable toretreive netstg2.imgfile In-Reply-To: References: Message-ID: <464CDAD5.8020504@herakles.homelinux.org> Shabazian, Chip wrote: > Now, *I* sure would like to see the timeout for pump increased from 30 > seconds to 60 seconds, but that would probably piss off even MORE people > since I'm sure there are a lot more people who get frustrated waiting 30 > seconds for pump to timeout during boot if their dhcp server is down, or > if they have more than one interface and don't know how to turn off dhcp > on the unused interfaces, etc. If my dhcp server's down, I got problems regardless of the timeout. > > I guess the ideal solution would be an option for pump that does > increase the timeout that could be passed from anaconda, or an option in > anaconda such as dhcpretry=X where you could set the system to retry > dhcp X number of times without recycling the NIC and starting the > negotiation dance again. > > But at the end of the day, please keep in mind that these are all work > arounds for the Cisco STP feature, that while valuable, is the actual > root cause. What _I_ would like to see is an attempt to use them all, configure all the network interfaces in parallel. Use the first one that comes up and allows access to the ks files. Then it wouldn't matter much what the timeout is, if eth1 comes up and we can get the ks file from it (Or later, can get the install source) and eth0 is still dithering around, well we're on the way with eth1 and who cares? -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From Chip.Shabazian at bankofamerica.com Thu May 17 23:09:38 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Thu, 17 May 2007 16:09:38 -0700 Subject: Unable to get any httprequests: Unable toretreive netstg2.imgfile In-Reply-To: <464CDAD5.8020504@herakles.homelinux.org> Message-ID: > Then it wouldn't matter much what the timeout is, if eth1 comes up and we can get the ks file from it >(Or later, can get the install source) and eth0 is still dithering around, well we're on the way with > eth1 and who cares? This doesn't work in environments such as a DMZ environment where you have to make sure that the one NIC (out of 3, 4, or more) that is on the only subnet that allows this traffic, is the one that is used. Also, this particular problem isn't whether or not you are accessing the right NIC, but the problem that pump times out before portfast turns the port up. It shouldn't matter if you have one or 20 NIC's, they should all have the same portfast problem. Dialog is good though, if we keep discussing the core issue and come up with a good resolution, we can get this incorporated into the process somewhere to make sure the problem is worked around. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John Summerfield Sent: Thursday, May 17, 2007 3:45 PM To: Discussion list about Kickstart Subject: Re: Unable to get any httprequests: Unable toretreive netstg2.imgfile Shabazian, Chip wrote: > Now, *I* sure would like to see the timeout for pump increased from 30 > seconds to 60 seconds, but that would probably piss off even MORE > people since I'm sure there are a lot more people who get frustrated > waiting 30 seconds for pump to timeout during boot if their dhcp > server is down, or if they have more than one interface and don't know > how to turn off dhcp on the unused interfaces, etc. If my dhcp server's down, I got problems regardless of the timeout. > > I guess the ideal solution would be an option for pump that does > increase the timeout that could be passed from anaconda, or an option > in anaconda such as dhcpretry=X where you could set the system to > retry dhcp X number of times without recycling the NIC and starting > the negotiation dance again. > > But at the end of the day, please keep in mind that these are all work > arounds for the Cisco STP feature, that while valuable, is the actual > root cause. What _I_ would like to see is an attempt to use them all, configure all the network interfaces in parallel. Use the first one that comes up and allows access to the ks files. Then it wouldn't matter much what the timeout is, if eth1 comes up and we can get the ks file from it (Or later, can get the install source) and eth0 is still dithering around, well we're on the way with eth1 and who cares? -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Klaus.Steden at thomson.net Fri May 18 02:03:52 2007 From: Klaus.Steden at thomson.net (Steden Klaus) Date: Thu, 17 May 2007 22:03:52 -0400 Subject: Unable to get any httprequests: Unable toretreive netstg2.imgfile References: Message-ID: Why not add command-line options to kickstart: dhcpretry=N (attempt to obtain a lease at least N times) dhcpwait=N (wait at least N seconds for a physical link to become active before transmitting a DHCP request) dhcptimeout=N (wait at least N seconds before retransmitting a DHCP request) I don't see Cisco changing the way it does things any time soon, so it's just something as a user community we will have to learn to live with -- but I think it can be mitigated until it's at least tolerable and usable. The critical thing is, I think, to allow application behaviours that relate to this problem to be modifiable easily, rather than forcing users to adopt "suggested" behaviour (I'm looking in your direction, Cisco!). Cheers, Klaus > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Shabazian, Chip > Sent: Thursday, May 17, 2007 4:10 PM > To: Discussion list about Kickstart > Subject: RE: Unable to get any httprequests: Unable toretreive > netstg2.imgfile > > > Then it wouldn't matter much what the timeout is, if eth1 comes up and > we can get the ks file from it > >(Or later, can get the install source) and eth0 is still dithering > around, well we're on the way with > > eth1 and who cares? > > This doesn't work in environments such as a DMZ environment where you > have to make sure that the one NIC (out of 3, 4, or more) that is on the > only subnet that allows this traffic, is the one that is used. > > Also, this particular problem isn't whether or not you are accessing the > right NIC, but the problem that pump times out before portfast turns the > port up. It shouldn't matter if you have one or 20 NIC's, they should > all have the same portfast problem. > > Dialog is good though, if we keep discussing the core issue and come up > with a good resolution, we can get this incorporated into the process > somewhere to make sure the problem is worked around. > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John Summerfield > Sent: Thursday, May 17, 2007 3:45 PM > To: Discussion list about Kickstart > Subject: Re: Unable to get any httprequests: Unable toretreive > netstg2.imgfile > > Shabazian, Chip wrote: > > > Now, *I* sure would like to see the timeout for pump increased from 30 > > > seconds to 60 seconds, but that would probably piss off even MORE > > people since I'm sure there are a lot more people who get frustrated > > waiting 30 seconds for pump to timeout during boot if their dhcp > > server is down, or if they have more than one interface and don't know > > > how to turn off dhcp on the unused interfaces, etc. > > > If my dhcp server's down, I got problems regardless of the timeout. > > > > > I guess the ideal solution would be an option for pump that does > > increase the timeout that could be passed from anaconda, or an option > > in anaconda such as dhcpretry=X where you could set the system to > > retry dhcp X number of times without recycling the NIC and starting > > the negotiation dance again. > > > > But at the end of the day, please keep in mind that these are all work > > > arounds for the Cisco STP feature, that while valuable, is the actual > > root cause. > > What _I_ would like to see is an attempt to use them all, configure all > the network interfaces in parallel. > > Use the first one that comes up and allows access to the ks files. > > Then it wouldn't matter much what the timeout is, if eth1 comes up and > we can get the ks file from it (Or later, can get the install source) > and eth0 is still dithering around, well we're on the way with eth1 and > who cares? > > > > > -- > > Cheers > John > > -- spambait > 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu > > Please do not reply off-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From jgranado at redhat.com Fri May 18 10:44:49 2007 From: jgranado at redhat.com (Joel Andres Granados) Date: Fri, 18 May 2007 12:44:49 +0200 Subject: Location of installation media no longer /tmp/cdrom? In-Reply-To: <819591.59305.qm@web32503.mail.mud.yahoo.com> References: <819591.59305.qm@web32503.mail.mud.yahoo.com> Message-ID: <464D83A1.6020204@redhat.com> Pat wrote: > --- kickstart-list-request at redhat.com wrote: > >>> From: Jeroen van Meeuwen >>> >> To: Discussion list about Kickstart >> >> Date: Wed, 16 May 2007 10:30:44 +0200 >> Subject: Re: Location of installation media no >> longer /tmp/cdrom? >> >> Pat wrote: >> >>> However I get the following error during the >>> >> install: >> >>> mount: Mounting /tmp/cdrom on /mnt/source failed: >>> >> No >> >>> such file or directory >>> >>> >>> >> Are you sure the /mnt/source directory exists? >> >> -kanarip >> >> > > > Yes, /mnt/source was created. I did not see /tmp/cdrom > though. The same kickstart file worked with > Redhat/CentOS 4.x, it looks like something changed > with 5? > > > > ____________________________________________________________________________________Building a website is a piece of cake. Yahoo! Small Business gives you all the tools to get online. > http://smallbusiness.yahoo.com/webhosting > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > In case you are still having some issues, I was able to mount it with "mount /dev/hdb ..". Hope it helps!!! Regards From nicholas.byrne at quadriga.com Fri May 18 12:04:07 2007 From: nicholas.byrne at quadriga.com (Nicholas Byrne) Date: Fri, 18 May 2007 13:04:07 +0100 Subject: Location of installation media no longer /tmp/cdrom? In-Reply-To: <464D83A1.6020204@redhat.com> References: <819591.59305.qm@web32503.mail.mud.yahoo.com> <464D83A1.6020204@redhat.com> Message-ID: <464D9637.9050908@quadriga.com> see this thread - https://www.redhat.com/archives/kickstart-list/2007-May/msg00007.html. It's a bug, the cdrom drive is being unmounted to early in the ks process, it manifests itself most commonly on "eject only" mechanism type drives because the mount request normally retracts the draw on capable devices which is also probably why it wasn't picked up during testing. To fix manually, open up stage2.img and change dispatch.py, swapping the lines: ("methodcomplete", doMethodComplete, ), ("dopostaction", doPostAction, ), for ("dopostaction", doPostAction, ), ("methodcomplete", doMethodComplete, ), Joel Andres Granados wrote: > Pat wrote: >> --- kickstart-list-request at redhat.com wrote: >> >>>> From: Jeroen van Meeuwen >>>> >>> To: Discussion list about Kickstart >>> >>> Date: Wed, 16 May 2007 10:30:44 +0200 >>> Subject: Re: Location of installation media no >>> longer /tmp/cdrom? >>> >>> Pat wrote: >>> >>>> However I get the following error during the >>>> >>> install: >>> >>>> mount: Mounting /tmp/cdrom on /mnt/source failed: >>>> >>> No >>> >>>> such file or directory >>>> >>>> >>> Are you sure the /mnt/source directory exists? >>> >>> -kanarip >>> >>> >> >> >> Yes, /mnt/source was created. I did not see /tmp/cdrom >> though. The same kickstart file worked with >> Redhat/CentOS 4.x, it looks like something changed >> with 5? >> >> >> >> ____________________________________________________________________________________Building >> a website is a piece of cake. Yahoo! Small Business gives you all the >> tools to get online. >> http://smallbusiness.yahoo.com/webhosting >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list >> > In case you are still having some issues, I was able to mount it with > "mount /dev/hdb ..". > Hope it helps!!! > Regards > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > This e-mail is the property of Quadriga Worldwide Ltd, intended for the addressee only and confidential. Any dissemination, copying or distribution of this message or any attachments is strictly prohibited. If you have received this message in error, please notify us immediately by replying to the message and deleting it from your computer. Messages sent to and from Quadriga may be monitored. Quadriga cannot guarantee any message delivery method is secure or error-free. Information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. We do not accept responsibility for any errors or omissions in this message and/or attachment that arise as a result of transmission. You should carry out your own virus checks before opening any attachment. Any views or opinions presented are solely those of the author and do not necessarily represent those of Quadriga. From debian at herakles.homelinux.org Sun May 20 00:39:33 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Sun, 20 May 2007 08:39:33 +0800 Subject: Unable to get any httprequests: Unable toretreive netstg2.imgfile In-Reply-To: References: Message-ID: <464F98C5.2060702@herakles.homelinux.org> Shabazian, Chip wrote: >> Then it wouldn't matter much what the timeout is, if eth1 comes up and > we can get the ks file from it >> (Or later, can get the install source) and eth0 is still dithering > around, well we're on the way with >> eth1 and who cares? Your pruning was a little severe:-( > > This doesn't work in environments such as a DMZ environment where you > have to make sure that the one NIC (out of 3, 4, or more) that is on the > only subnet that allows this traffic, is the one that is used. I think the time for attaching to networks required "in production" (whatever that means in your circumstances) is after the installation's done, the configuration is done sufficiently and the system's checked out. Then, you can worry about what goes where and plug all the right wires into all the right holes. > > Also, this particular problem isn't whether or not you are accessing the > right NIC, but the problem that pump times out before portfast turns the > port up. It shouldn't matter if you have one or 20 NIC's, they should > all have the same portfast problem. I was responding to to someone's point about the cumulative effect of several long timeouts. By trying all NICs in parallel, there's only one timeout that matters, and mostly it's fairly short. > > Dialog is good though, if we keep discussing the core issue and come up > with a good resolution, we can get this incorporated into the process > somewhere to make sure the problem is worked around. > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John Summerfield > Sent: Thursday, May 17, 2007 3:45 PM > To: Discussion list about Kickstart > Subject: Re: Unable to get any httprequests: Unable toretreive > netstg2.imgfile > > Shabazian, Chip wrote: > >> Now, *I* sure would like to see the timeout for pump increased from 30 > >> seconds to 60 seconds, but that would probably piss off even MORE >> people since I'm sure there are a lot more people who get frustrated >> waiting 30 seconds for pump to timeout during boot if their dhcp >> server is down, or if they have more than one interface and don't know > >> how to turn off dhcp on the unused interfaces, etc. > > > If my dhcp server's down, I got problems regardless of the timeout. > >> >> I guess the ideal solution would be an option for pump that does >> increase the timeout that could be passed from anaconda, or an option >> in anaconda such as dhcpretry=X where you could set the system to >> retry dhcp X number of times without recycling the NIC and starting >> the negotiation dance again. >> >> But at the end of the day, please keep in mind that these are all work > >> arounds for the Cisco STP feature, that while valuable, is the actual >> root cause. > > What _I_ would like to see is an attempt to use them all, configure all > the network interfaces in parallel. > > Use the first one that comes up and allows access to the ks files. > > Then it wouldn't matter much what the timeout is, if eth1 comes up and > we can get the ks file from it (Or later, can get the install source) > and eth0 is still dithering around, well we're on the way with eth1 and > who cares? > > > > -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From guolin at alexa.com Mon May 21 17:54:16 2007 From: guolin at alexa.com (Guolin Cheng) Date: Mon, 21 May 2007 10:54:16 -0700 Subject: trying to run sshd daemon in kickstart environment Message-ID: Hi, I tried to make kickstart environment a little bit more charming by firing up openssh daemon in the limited environment. I edited/checked a bunch of files and almost succeeded but still not. I can reach the machine to run a command like 'ssh -x ' and it does work for %pre and both chroot and non-chroot %post sections, in fact, you can run a command on kickstartClient through ssh connections without problems. But it still fails because I can not ssh into the kickstartClient to get a local command shell. In short, 'ssh -x ' fails, 'ssh -x ' works. Any one has succeeded in this topic? Life will be much happier if we can freely turn on/off sshd daemon for Kickstart environment. Thanks. --Guolin -------------- next part -------------- An HTML attachment was scrubbed... URL: From patchu1 at yahoo.com Tue May 22 01:02:23 2007 From: patchu1 at yahoo.com (Pat) Date: Mon, 21 May 2007 18:02:23 -0700 (PDT) Subject: Location of installation media no longer /tmp/cdrom? In-Reply-To: <20070518160011.E571573533@hormel.redhat.com> Message-ID: <500228.25162.qm@web32504.mail.mud.yahoo.com> --- kickstart-list-request at redhat.com wrote: > > From: Nicholas Byrne > To: Discussion list about Kickstart > > Date: Fri, 18 May 2007 13:04:07 +0100 > Subject: Re: Location of installation media no > longer /tmp/cdrom? > > see this thread - > https://www.redhat.com/archives/kickstart-list/2007-May/msg00007.html. > It's a bug, the cdrom drive is being unmounted to > early in the ks > process, it manifests itself most commonly on "eject > only" mechanism > type drives because the mount request normally > retracts the draw on > capable devices which is also probably why it wasn't > picked up during > testing. > To fix manually, open up stage2.img and change > dispatch.py, swapping the > lines: > > ("methodcomplete", doMethodComplete, ), > ("dopostaction", doPostAction, ), > > for > > ("dopostaction", doPostAction, ), > ("methodcomplete", doMethodComplete, ), > Thanks for the tip, I tried editing the stage2.img and creating a new install CD, however I received the following error during the install (from my notes) with that CD: No CentOS CD was found which matches your boot media. Please insert the CentOS CD and press OK to retry. If I press CONTROL-ALT-F3 I see the following messages: -mntloop loop0 on /mnt/runtime as /mnt/source/images/stage2.img fd is 10 -umounting loopback /mnt/runtime loop 0 -mounting stage2 failed -ejecting /tmp/cdrom CONTROL-ALT-F3 show the following messages: <7>ISO 9660 Extensions: Microsoft Juliet Level 3 <4> Unable to load NLF charset utf8 <4> Unable to load NLF charset utf8 <7>ISO 9660 Extensions: RRIP_1991A <4>Unable to identify CD-ROM format <4>VFS: Can't find an ext2 fileystem on dev loop0 Here's the steps I took to create the new install CD: # cp /mnt/cdrom/images/stage2.img . # mkdir anaconda # mount -rw -t squashfs -o loop stage2.img anaconda/ # cp -av anaconda/* new/ # cd new # mksquashfs . ../stage2.img.new I then copied that stage2.img.new to a copy of my current install DVD, overwriting the original stage2.img, and burned the new image. However, I got that error above. Did I need to change any other files since I altered stage2.img? BTW I noticed when I do a file command on the original stage2.img and the stage2.img.new I get the following info: # file stage2.img stage2.img: Squashfs filesystem, little endian, version 3.0, 0 bytes, 7905 inodes, blocksize: 65536 bytes, created: Tue Apr 10 12:50:46 2007 # file stage2.img.new stage2.img.new: Squashfs filesystem, little endian, version 3.0, 1268861952 bytes, 7904 inodes, blocksize: 65536 bytes, created: Fri May 18 16:24:11 2007 Notice how the size of the original stage2.img say 0 bytes versus 1268861952 bytes for the new one I created. In order to create stage2.img.new, I downloaded and compiled squashfs3.2-r2.tar.gz so I could get mksquashfs. If you can spot where I went wrong, please let me know. Thanks! ____________________________________________________________________________________Yahoo! oneSearch: Finally, mobile search that gives answers, not web links. http://mobile.yahoo.com/mobileweb/onesearch?refer=1ONXIC From debian at herakles.homelinux.org Tue May 22 01:10:31 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Tue, 22 May 2007 09:10:31 +0800 Subject: Location of installation media no longer /tmp/cdrom? In-Reply-To: <500228.25162.qm@web32504.mail.mud.yahoo.com> References: <500228.25162.qm@web32504.mail.mud.yahoo.com> Message-ID: <46524307.3090606@herakles.homelinux.org> Pat wrote: > --- kickstart-list-request at redhat.com wrote: > >>> From: Nicholas Byrne >> To: Discussion list about Kickstart >> >> Date: Fri, 18 May 2007 13:04:07 +0100 >> Subject: Re: Location of installation media no >> longer /tmp/cdrom? >> >> see this thread - >> > https://www.redhat.com/archives/kickstart-list/2007-May/msg00007.html. >> It's a bug, the cdrom drive is being unmounted to >> early in the ks >> process, it manifests itself most commonly on "eject >> only" mechanism >> type drives because the mount request normally >> retracts the draw on >> capable devices which is also probably why it wasn't >> picked up during >> testing. >> To fix manually, open up stage2.img and change >> dispatch.py, swapping the >> lines: >> >> ("methodcomplete", doMethodComplete, ), >> ("dopostaction", doPostAction, ), >> >> for >> >> ("dopostaction", doPostAction, ), >> ("methodcomplete", doMethodComplete, ), >> > > Thanks for the tip, I tried editing the stage2.img and > creating a new install CD, however I received the > following error during the install (from my notes) > with that CD: > > > No CentOS CD was found which matches your boot media. > Please insert the CentOS CD and press OK to retry. Perhaps you should grep the source tree to see where that message comes from, it seems well within your capabilities. Also, the commandline for mkisofs is important. I'm wondering whether there is some little hidden file or directory you missed. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From jgranado at redhat.com Tue May 22 09:05:01 2007 From: jgranado at redhat.com (Joel Andres Granados) Date: Tue, 22 May 2007 11:05:01 +0200 Subject: Location of installation media no longer /tmp/cdrom? In-Reply-To: <46524307.3090606@herakles.homelinux.org> References: <500228.25162.qm@web32504.mail.mud.yahoo.com> <46524307.3090606@herakles.homelinux.org> Message-ID: <4652B23D.9060908@redhat.com> John Summerfield wrote: > Pat wrote: >> --- kickstart-list-request at redhat.com wrote: >> >>>> From: Nicholas Byrne >>> To: Discussion list about Kickstart >>> >>> Date: Fri, 18 May 2007 13:04:07 +0100 >>> Subject: Re: Location of installation media no >>> longer /tmp/cdrom? >>> >>> see this thread - >>> >> https://www.redhat.com/archives/kickstart-list/2007-May/msg00007.html. >>> It's a bug, the cdrom drive is being unmounted to >>> early in the ks >>> process, it manifests itself most commonly on "eject >>> only" mechanism >>> type drives because the mount request normally >>> retracts the draw on >>> capable devices which is also probably why it wasn't >>> picked up during >>> testing. >>> To fix manually, open up stage2.img and change >>> dispatch.py, swapping the >>> lines: >>> >>> ("methodcomplete", doMethodComplete, ), >>> ("dopostaction", doPostAction, ), >>> >>> for >>> >>> ("dopostaction", doPostAction, ), >>> ("methodcomplete", doMethodComplete, ), >>> >> >> Thanks for the tip, I tried editing the stage2.img and >> creating a new install CD, however I received the >> following error during the install (from my notes) >> with that CD: >> >> >> No CentOS CD was found which matches your boot media. >> Please insert the CentOS CD and press OK to retry. > > Perhaps you should grep the source tree to see where that message > comes from, it seems well within your capabilities. > > Also, the commandline for mkisofs is important. I'm wondering whether > there is some little hidden file or directory you missed. Maybe your missing the .discinfo file Hope it helps. :) From jce at zot.com Tue May 22 16:41:32 2007 From: jce at zot.com (Chris Edillon) Date: Tue, 22 May 2007 12:41:32 -0400 Subject: trying to run sshd daemon in kickstart environment In-Reply-To: References: Message-ID: <1179852093.5178.7.camel@ruthless> On Mon, 2007-05-21 at 10:54 -0700, Guolin Cheng wrote: > I tried to make kickstart environment a little bit more charming by > firing up openssh daemon in the limited environment. I edited/checked > a bunch of files and almost succeeded but still not. > > I can reach the machine to run a command like ?ssh ?x > ? and it does work for %pre and both chroot > and non-chroot %post sections, in fact, you can run a command on > kickstartClient through ssh connections without problems. But it > still fails because I can not ssh into the kickstartClient to get a > local command shell. > > In short, ?ssh ?x ? fails, ?ssh ?x > ? works. > > Any one has succeeded in this topic? Life will be much happier if we > can freely turn on/off sshd daemon for Kickstart environment. Thanks. you may need to force ssh to allocate a pseudo-tty, try "ssh -t -x ". the -t option is described in the ssh manpage. chris From guolin at alexa.com Wed May 23 21:39:03 2007 From: guolin at alexa.com (Guolin Cheng) Date: Wed, 23 May 2007 14:39:03 -0700 Subject: It works now -- RE: trying to run sshd daemon in kickstart environment In-Reply-To: <1179852093.5178.7.camel@ruthless> Message-ID: Hi, Thanks a lot for so many helpful responses. The fix is pretty straight: just mount /dev/pts again at the chrooted %post environment where my original sshd fires up from. One command 'Mount -t devpts -o rw,gid=5,mode=620 devpts /dev/pts' or just a bare 'mount /dev/pts' command fix the problem since the devpts entry is in fstab already. I also hacked image stage2.img to include sshd daemon binary, libraries and configuration files, and fire up the ssh daemon at the first line of %pre script section. Then disable my original fired up sshd daemon at %post section. Now kickstart environment is not 'dump' at all -- I can kick into and monitor the kickstart clients any time and stop run-away kickstart processes with 'reboot' command immediately should my home directory server upgrade turn into a disaster. --Guolin -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Chris Edillon Sent: Tuesday, May 22, 2007 9:42 AM To: Discussion list about Kickstart Cc: centos at centos.org Subject: Re: trying to run sshd daemon in kickstart environment On Mon, 2007-05-21 at 10:54 -0700, Guolin Cheng wrote: > I tried to make kickstart environment a little bit more charming by > firing up openssh daemon in the limited environment. I edited/checked > a bunch of files and almost succeeded but still not. > > I can reach the machine to run a command like 'ssh -x > ' and it does work for %pre and both chroot > and non-chroot %post sections, in fact, you can run a command on > kickstartClient through ssh connections without problems. But it > still fails because I can not ssh into the kickstartClient to get a > local command shell. > > In short, 'ssh -x ' fails, 'ssh -x > ' works. > > Any one has succeeded in this topic? Life will be much happier if we > can freely turn on/off sshd daemon for Kickstart environment. Thanks. you may need to force ssh to allocate a pseudo-tty, try "ssh -t -x ". the -t option is described in the ssh manpage. chris _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Klaus.Steden at thomson.net Wed May 23 21:41:20 2007 From: Klaus.Steden at thomson.net (Steden Klaus) Date: Wed, 23 May 2007 17:41:20 -0400 Subject: It works now -- RE: trying to run sshd daemon in kickstartenvironment References: Message-ID: Hmmm, this sounds like a useful addition. Could you post your steps, etc. to the Fedora Wiki? Cheers, Klaus > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Guolin Cheng > Sent: Wednesday, May 23, 2007 2:39 PM > To: Discussion list about Kickstart > Cc: centos at centos.org; kickstart-list at redhat.com > Subject: It works now -- RE: trying to run sshd daemon in > kickstartenvironment > > Hi, > > Thanks a lot for so many helpful responses. > > The fix is pretty straight: just mount /dev/pts again at the chrooted > %post environment where my original sshd fires up from. > > One command 'Mount -t devpts -o rw,gid=5,mode=620 devpts /dev/pts' or > just a bare 'mount /dev/pts' command fix the problem since the devpts > entry is in fstab already. > > I also hacked image stage2.img to include sshd daemon binary, libraries > and configuration files, and fire up the ssh daemon at the first line of > %pre script section. Then disable my original fired up sshd daemon at > %post section. > > Now kickstart environment is not 'dump' at all -- I can kick into and > monitor the kickstart clients any time and stop run-away kickstart > processes with 'reboot' command immediately should my home directory > server upgrade turn into a disaster. > > --Guolin > > > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Chris Edillon > Sent: Tuesday, May 22, 2007 9:42 AM > To: Discussion list about Kickstart > Cc: centos at centos.org > Subject: Re: trying to run sshd daemon in kickstart environment > > On Mon, 2007-05-21 at 10:54 -0700, Guolin Cheng wrote: > > > I tried to make kickstart environment a little bit more charming by > > firing up openssh daemon in the limited environment. I edited/checked > > a bunch of files and almost succeeded but still not. > > > > I can reach the machine to run a command like 'ssh -x > > ' and it does work for %pre and both chroot > > and non-chroot %post sections, in fact, you can run a command on > > kickstartClient through ssh connections without problems. But it > > still fails because I can not ssh into the kickstartClient to get a > > local command shell. > > > > In short, 'ssh -x ' fails, 'ssh -x > > ' works. > > > > Any one has succeeded in this topic? Life will be much happier if we > > can freely turn on/off sshd daemon for Kickstart environment. Thanks. > > you may need to force ssh to allocate a pseudo-tty, > try "ssh -t -x ". the -t option is described > in the ssh manpage. > > chris > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From guolin at alexa.com Wed May 23 22:55:05 2007 From: guolin at alexa.com (Guolin Cheng) Date: Wed, 23 May 2007 15:55:05 -0700 Subject: It works now -- RE: trying to run sshd daemon inkickstartenvironment In-Reply-To: Message-ID: Steden and all, Sure, I'll find some time to document my steps in near future. Have fun. --Guolin -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Steden Klaus Sent: Wednesday, May 23, 2007 2:41 PM To: Discussion list about Kickstart Cc: centos at centos.org Subject: RE: It works now -- RE: trying to run sshd daemon inkickstartenvironment Hmmm, this sounds like a useful addition. Could you post your steps, etc. to the Fedora Wiki? Cheers, Klaus > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Guolin Cheng > Sent: Wednesday, May 23, 2007 2:39 PM > To: Discussion list about Kickstart > Cc: centos at centos.org; kickstart-list at redhat.com > Subject: It works now -- RE: trying to run sshd daemon in > kickstartenvironment > > Hi, > > Thanks a lot for so many helpful responses. > > The fix is pretty straight: just mount /dev/pts again at the chrooted > %post environment where my original sshd fires up from. > > One command 'Mount -t devpts -o rw,gid=5,mode=620 devpts /dev/pts' or > just a bare 'mount /dev/pts' command fix the problem since the devpts > entry is in fstab already. > > I also hacked image stage2.img to include sshd daemon binary, libraries > and configuration files, and fire up the ssh daemon at the first line of > %pre script section. Then disable my original fired up sshd daemon at > %post section. > > Now kickstart environment is not 'dump' at all -- I can kick into and > monitor the kickstart clients any time and stop run-away kickstart > processes with 'reboot' command immediately should my home directory > server upgrade turn into a disaster. > > --Guolin > > > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Chris Edillon > Sent: Tuesday, May 22, 2007 9:42 AM > To: Discussion list about Kickstart > Cc: centos at centos.org > Subject: Re: trying to run sshd daemon in kickstart environment > > On Mon, 2007-05-21 at 10:54 -0700, Guolin Cheng wrote: > > > I tried to make kickstart environment a little bit more charming by > > firing up openssh daemon in the limited environment. I edited/checked > > a bunch of files and almost succeeded but still not. > > > > I can reach the machine to run a command like 'ssh -x > > ' and it does work for %pre and both chroot > > and non-chroot %post sections, in fact, you can run a command on > > kickstartClient through ssh connections without problems. But it > > still fails because I can not ssh into the kickstartClient to get a > > local command shell. > > > > In short, 'ssh -x ' fails, 'ssh -x > > ' works. > > > > Any one has succeeded in this topic? Life will be much happier if we > > can freely turn on/off sshd daemon for Kickstart environment. Thanks. > > you may need to force ssh to allocate a pseudo-tty, > try "ssh -t -x ". the -t option is described > in the ssh manpage. > > chris > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From guolin at alexa.com Wed May 30 17:49:17 2007 From: guolin at alexa.com (Guolin Cheng) Date: Wed, 30 May 2007 10:49:17 -0700 Subject: custom kernel for Centos 5 kickstart? Message-ID: Hi, Has any one tried to merge a custom kernel into vmlinuz/initrd.img for Centos 5 kickstart? Or similar Linux distribution? I am rebuilding a custom kernel from stock kernel source to build some hardware modules into static form, so that the hardware detection sequence can be altered in a hard way. Quick examples are: tg3 drivers will be loaded before e1000, 3w-9xxx be loaded before 3w-9xxx, etc. Please advice. If the object could be achieved through adjustment to /etc/modules.conf or similar files, then it could be even more fantastic. -------------- next part -------------- An HTML attachment was scrubbed... URL: From CallahanT at tessco.com Wed May 30 17:58:05 2007 From: CallahanT at tessco.com (Callahan, Tom) Date: Wed, 30 May 2007 13:58:05 -0400 Subject: custom kernel for Centos 5 kickstart? In-Reply-To: Message-ID: I remember there being an option to define which MAC addr is defined as eth0, etc. This was something you could define on the kickstart command line.... Check out the documentation for anaconda. On 5/30/07 1:49 PM, "Guolin Cheng" wrote: > Hi, > > Has any one tried to merge a custom kernel into vmlinuz/initrd.img for Centos > 5 kickstart? Or similar Linux distribution? > > I am rebuilding a custom kernel from stock kernel source to build some > hardware modules into static form, so that the hardware detection sequence can > be altered in a hard way. Quick examples are: tg3 drivers will be loaded > before e1000, 3w-9xxx be loaded before 3w-9xxx, etc. > > Please advice. If the object could be achieved through adjustment to > /etc/modules.conf or similar files, then it could be even more fantastic. > > > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list ------------------------------- Tom Callahan TDS Engineering 410-229-1361 Tel 410-588-7605 Mobile 410-229-1512 Fax callahant at TESSCO.com TESSCO Your Total Source for Wireless http://www.tessco.com Network Infrastructure Equipment | Mobile Devices & Accessories | Installation, Test, Equipment & Supplies -------------- next part -------------- An HTML attachment was scrubbed... URL: From Nicholas.Byrne at quadriga.com Wed May 30 22:40:25 2007 From: Nicholas.Byrne at quadriga.com (Byrne, Nicholas) Date: Wed, 30 May 2007 23:40:25 +0100 Subject: custom kernel for Centos 5 kickstart? References: Message-ID: <9510285C413E404895C083084CD0D43D9257D7@VSGLMAIL01.quadriga.com> possibly "ksdevice=" which has a number of options that are useful but it depends on what exactly he's trying to achieve. The other method i thought of is to not setup network in main kickstart but instead do it with netconfig in %post where you can use the --hwaddr switch (and determine the target interface by it's mac address). Or perhaps, since anaconda does a good job of setting up interfaces for you, so just configure them up as you would like ignoring the fact that they may be pointing to the wrong NIC and then change /etc/modprobe.conf and /etc/sysconfig/network-interfaces/ifcfg-ethX files in %post to suite your needs (again determine the target via mac address) The first 3 bytes of the mac addresses are the manufacture code so you should be able to work our a list for your hardware using http://coffer.com/mac_find/ ________________________________ From: kickstart-list-bounces at redhat.com on behalf of Callahan, Tom Sent: Wed 30/05/2007 18:58 To: Discussion list about Kickstart; centos at centos.org Subject: Re: custom kernel for Centos 5 kickstart? I remember there being an option to define which MAC addr is defined as eth0, etc. This was something you could define on the kickstart command line.... Check out the documentation for anaconda. On 5/30/07 1:49 PM, "Guolin Cheng" wrote: Hi, Has any one tried to merge a custom kernel into vmlinuz/initrd.img for Centos 5 kickstart? Or similar Linux distribution? I am rebuilding a custom kernel from stock kernel source to build some hardware modules into static form, so that the hardware detection sequence can be altered in a hard way. Quick examples are: tg3 drivers will be loaded before e1000, 3w-9xxx be loaded before 3w-9xxx, etc. Please advice. If the object could be achieved through adjustment to /etc/modules.conf or similar files, then it could be even more fantastic. ________________________________ _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list ------------------------------- Tom Callahan TDS Engineering 410-229-1361 Tel 410-588-7605 Mobile 410-229-1512 Fax callahant at TESSCO.com TESSCO Your Total Source for Wireless http://www.tessco.com Network Infrastructure Equipment | Mobile Devices & Accessories | Installation, Test, Equipment & Supplies This e-mail is the property of Quadriga Worldwide Ltd, intended for the addressee only and confidential. Any dissemination, copying or distribution of this message or any attachments is strictly prohibited. If you have received this message in error, please notify us immediately by replying to the message and deleting it from your computer. Messages sent to and from Quadriga may be monitored. Quadriga cannot guarantee any message delivery method is secure or error-free. Information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. We do not accept responsibility for any errors or omissions in this message and/or attachment that arise as a result of transmission. You should carry out your own virus checks before opening any attachment. Any views or opinions presented are solely those of the author and do not necessarily represent those of Quadriga. From guolin at alexa.com Wed May 30 23:34:10 2007 From: guolin at alexa.com (Guolin Cheng) Date: Wed, 30 May 2007 16:34:10 -0700 Subject: custom kernel for Centos 5 kickstart? In-Reply-To: <9510285C413E404895C083084CD0D43D9257D7@VSGLMAIL01.quadriga.com> Message-ID: Hi, Thanks a lot for your great suggestions. I've tried ksdevice options as well. It does work for some cases. But the main problems. We have tens different network/harddrive combinations. And so it turns a simple issue into a unmanageable complex. Originally we use a custom booting kernel plus in-house sysimager solution to handle it, custom kernel fix the detection sequence problem and sysimager dumps a few tarball/images to hard drives. If we can find a way to use a custom kernel to replace default Kickstart kernel+initrd.img. Then we can fix the problem very elegantly. Any ideas on hacking Kickstart initrd.img? --Guolin -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Byrne, Nicholas Sent: Wednesday, May 30, 2007 3:40 PM To: Discussion list about Kickstart; centos at centos.org Subject: RE: custom kernel for Centos 5 kickstart? possibly "ksdevice=" which has a number of options that are useful but it depends on what exactly he's trying to achieve. The other method i thought of is to not setup network in main kickstart but instead do it with netconfig in %post where you can use the --hwaddr switch (and determine the target interface by it's mac address). Or perhaps, since anaconda does a good job of setting up interfaces for you, so just configure them up as you would like ignoring the fact that they may be pointing to the wrong NIC and then change /etc/modprobe.conf and /etc/sysconfig/network-interfaces/ifcfg-ethX files in %post to suite your needs (again determine the target via mac address) The first 3 bytes of the mac addresses are the manufacture code so you should be able to work our a list for your hardware using http://coffer.com/mac_find/ ________________________________ From: kickstart-list-bounces at redhat.com on behalf of Callahan, Tom Sent: Wed 30/05/2007 18:58 To: Discussion list about Kickstart; centos at centos.org Subject: Re: custom kernel for Centos 5 kickstart? I remember there being an option to define which MAC addr is defined as eth0, etc. This was something you could define on the kickstart command line.... Check out the documentation for anaconda. On 5/30/07 1:49 PM, "Guolin Cheng" wrote: Hi, Has any one tried to merge a custom kernel into vmlinuz/initrd.img for Centos 5 kickstart? Or similar Linux distribution? I am rebuilding a custom kernel from stock kernel source to build some hardware modules into static form, so that the hardware detection sequence can be altered in a hard way. Quick examples are: tg3 drivers will be loaded before e1000, 3w-9xxx be loaded before 3w-9xxx, etc. Please advice. If the object could be achieved through adjustment to /etc/modules.conf or similar files, then it could be even more fantastic. ________________________________ _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list ------------------------------- Tom Callahan TDS Engineering 410-229-1361 Tel 410-588-7605 Mobile 410-229-1512 Fax callahant at TESSCO.com TESSCO Your Total Source for Wireless http://www.tessco.com Network Infrastructure Equipment | Mobile Devices & Accessories | Installation, Test, Equipment & Supplies This e-mail is the property of Quadriga Worldwide Ltd, intended for the addressee only and confidential. Any dissemination, copying or distribution of this message or any attachments is strictly prohibited. If you have received this message in error, please notify us immediately by replying to the message and deleting it from your computer. Messages sent to and from Quadriga may be monitored. Quadriga cannot guarantee any message delivery method is secure or error-free. Information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. We do not accept responsibility for any errors or omissions in this message and/or attachment that arise as a result of transmission. You should carry out your own virus checks before opening any attachment. Any views or opinions presented are solely those of the author and do not necessarily represent those of Quadriga. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From CallahanT at tessco.com Wed May 30 23:42:54 2007 From: CallahanT at tessco.com (Callahan, Tom) Date: Wed, 30 May 2007 19:42:54 -0400 Subject: custom kernel for Centos 5 kickstart? Message-ID: <6982E319B698F143AEDB8F3D8398045702888052@EXCHANGE-1.tessco.com> I added a failover enabled qlogic fiberchannel driver to our custom kickstart CD, the initrd is just a loop file you can pull apart and recreate ================================ Sent from my wireless handheld device. Tom Callahan -----Original Message----- From: kickstart-list-bounces at redhat.com To: Discussion list about Kickstart ; centos at centos.org Sent: Wed May 30 19:34:10 2007 Subject: RE: custom kernel for Centos 5 kickstart? Hi, Thanks a lot for your great suggestions. I've tried ksdevice options as well. It does work for some cases. But the main problems. We have tens different network/harddrive combinations. And so it turns a simple issue into a unmanageable complex. Originally we use a custom booting kernel plus in-house sysimager solution to handle it, custom kernel fix the detection sequence problem and sysimager dumps a few tarball/images to hard drives. If we can find a way to use a custom kernel to replace default Kickstart kernel+initrd.img. Then we can fix the problem very elegantly. Any ideas on hacking Kickstart initrd.img? --Guolin -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Byrne, Nicholas Sent: Wednesday, May 30, 2007 3:40 PM To: Discussion list about Kickstart; centos at centos.org Subject: RE: custom kernel for Centos 5 kickstart? possibly "ksdevice=" which has a number of options that are useful but it depends on what exactly he's trying to achieve. The other method i thought of is to not setup network in main kickstart but instead do it with netconfig in %post where you can use the --hwaddr switch (and determine the target interface by it's mac address). Or perhaps, since anaconda does a good job of setting up interfaces for you, so just configure them up as you would like ignoring the fact that they may be pointing to the wrong NIC and then change /etc/modprobe.conf and /etc/sysconfig/network-interfaces/ifcfg-ethX files in %post to suite your needs (again determine the target via mac address) The first 3 bytes of the mac addresses are the manufacture code so you should be able to work our a list for your hardware using http://coffer.com/mac_find/ ________________________________ From: kickstart-list-bounces at redhat.com on behalf of Callahan, Tom Sent: Wed 30/05/2007 18:58 To: Discussion list about Kickstart; centos at centos.org Subject: Re: custom kernel for Centos 5 kickstart? I remember there being an option to define which MAC addr is defined as eth0, etc. This was something you could define on the kickstart command line.... Check out the documentation for anaconda. On 5/30/07 1:49 PM, "Guolin Cheng" wrote: Hi, Has any one tried to merge a custom kernel into vmlinuz/initrd.img for Centos 5 kickstart? Or similar Linux distribution? I am rebuilding a custom kernel from stock kernel source to build some hardware modules into static form, so that the hardware detection sequence can be altered in a hard way. Quick examples are: tg3 drivers will be loaded before e1000, 3w-9xxx be loaded before 3w-9xxx, etc. Please advice. If the object could be achieved through adjustment to /etc/modules.conf or similar files, then it could be even more fantastic. ________________________________ _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list ------------------------------- Tom Callahan TDS Engineering 410-229-1361 Tel 410-588-7605 Mobile 410-229-1512 Fax callahant at TESSCO.com TESSCO Your Total Source for Wireless http://www.tessco.com Network Infrastructure Equipment | Mobile Devices & Accessories | Installation, Test, Equipment & Supplies This e-mail is the property of Quadriga Worldwide Ltd, intended for the addressee only and confidential. Any dissemination, copying or distribution of this message or any attachments is strictly prohibited. If you have received this message in error, please notify us immediately by replying to the message and deleting it from your computer. Messages sent to and from Quadriga may be monitored. Quadriga cannot guarantee any message delivery method is secure or error-free. Information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. We do not accept responsibility for any errors or omissions in this message and/or attachment that arise as a result of transmission. You should carry out your own virus checks before opening any attachment. Any views or opinions presented are solely those of the author and do not necessarily represent those of Quadriga. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From guolin at alexa.com Wed May 30 23:44:30 2007 From: guolin at alexa.com (Guolin Cheng) Date: Wed, 30 May 2007 16:44:30 -0700 Subject: How the Centos 5.0's kickstart initrd.img is build?? Message-ID: Hi, Any one know how the Centos 5 kickstart initrd.img was built? I'll be very appreciated if any light can be shed on this topic. The basic problem to be solved here is: how to build a custom vmlinuz+initrd.img for Centos 5 kickstart with my own custom built kernel? I am not sure mkinitrd is the right tool for this purpose? If do, which are the modules I should include with '-preload' options? And if there are any extra files/steps I should include/follow for the initrd.img file? Not sure how the stage2.img is loaded by kernel after initrd.img is loaded and run. Please help. --Guolin -------------- next part -------------- An HTML attachment was scrubbed... URL: From guolin at alexa.com Thu May 31 00:33:24 2007 From: guolin at alexa.com (Guolin Cheng) Date: Wed, 30 May 2007 17:33:24 -0700 Subject: Very different layout -- RE: How the Centos 5.0's kickstart initrd.img is build?? In-Reply-To: Message-ID: Hi, I've tried run mkinitrd to create an initrd.img file for Centos 5 Kickstart, but it turns out that the default initrd.img is quite different from the initrd.img file created with mkinitrd. Any one know how to create a Kickstart initrd.img with custom kernel? Or the hacking steps on a existing Centos 5 kickstart initrd.img? root at Centos5TestBox aaa]# ls -alF original mkinitrd/ mkinitrd/: total 40 drwxr-xr-x 9 root root 4096 May 30 17:22 ./ drwxr-xr-x 4 root root 4096 May 30 17:29 ../ drwx------ 2 root root 4096 May 30 17:22 bin/ drwx------ 3 root root 4096 May 30 17:22 dev/ drwx------ 2 root root 4096 May 30 17:22 etc/ -rwx------ 1 root root 1833 May 30 17:22 init* drwx------ 2 root root 4096 May 30 17:22 lib/ drwx------ 2 root root 4096 May 30 17:22 proc/ lrwxrwxrwx 1 root root 3 May 30 17:22 sbin -> bin/ drwx------ 2 root root 4096 May 30 17:22 sys/ drwx------ 2 root root 4096 May 30 17:22 sysroot/ original: total 52 drwxr-xr-x 11 root root 4096 May 30 17:21 ./ drwxr-xr-x 4 root root 4096 May 30 17:29 ../ -rw-r--r-- 1 root root 62 May 30 17:21 .buildstamp -rw-r--r-- 1 root root 123 May 30 17:21 .profile lrwxrwxrwx 1 root root 4 May 30 17:21 bin -> sbin/ drwxr-xr-x 2 root root 4096 May 30 17:21 dev/ drwxr-xr-x 3 root root 4096 May 30 17:21 etc/ lrwxrwxrwx 1 root root 10 May 30 17:21 init -> /sbin/init* drwxr-xr-x 2 root root 4096 May 30 17:21 modules/ drwxr-xr-x 2 root root 4096 May 30 17:21 proc/ drwxr-xr-x 2 root root 4096 May 30 17:21 sbin/ drwxr-xr-x 2 root root 4096 May 30 17:21 selinux/ drwxr-xr-x 2 root root 4096 May 30 17:21 sys/ drwxr-xr-x 2 root root 4096 May 30 17:21 tmp/ drwxr-xr-x 6 root root 4096 May 30 17:21 var/ root at hello11 aaa]# --Guolin ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Guolin Cheng Sent: Wednesday, May 30, 2007 4:45 PM To: centos at centos.org; Discussion list about Kickstart Subject: How the Centos 5.0's kickstart initrd.img is build?? Hi, Any one know how the Centos 5 kickstart initrd.img was built? I'll be very appreciated if any light can be shed on this topic. The basic problem to be solved here is: how to build a custom vmlinuz+initrd.img for Centos 5 kickstart with my own custom built kernel? I am not sure mkinitrd is the right tool for this purpose? If do, which are the modules I should include with '-preload' options? And if there are any extra files/steps I should include/follow for the initrd.img file? Not sure how the stage2.img is loaded by kernel after initrd.img is loaded and run. Please help. --Guolin -------------- next part -------------- An HTML attachment was scrubbed... URL: From jce at zot.com Thu May 31 04:20:36 2007 From: jce at zot.com (Chris Edillon) Date: Thu, 31 May 2007 00:20:36 -0400 Subject: custom kernel for Centos 5 kickstart? In-Reply-To: References: Message-ID: <465E4D14.10407@zot.com> Guolin Cheng wrote: > I've tried ksdevice options as well. It does work for some cases. But > the main problems. We have tens different network/harddrive > combinations. And so it turns a simple issue into a unmanageable > complex. Originally we use a custom booting kernel plus in-house > sysimager solution to handle it, custom kernel fix the detection > sequence problem and sysimager dumps a few tarball/images to hard > drives. > could you use the "noprobe" kernel parameter at boot time to tell anaconda not to probe for network and storage devices, then use the "device" command in your kickstart config file to order the devices as you want? chris From jce at zot.com Thu May 31 04:31:08 2007 From: jce at zot.com (Chris Edillon) Date: Thu, 31 May 2007 00:31:08 -0400 Subject: Very different layout -- RE: How the Centos 5.0's kickstart initrd.img is build?? In-Reply-To: References: Message-ID: <465E4F8C.30209@zot.com> Guolin Cheng wrote: > I?ve tried run mkinitrd to create an initrd.img file for Centos 5 > Kickstart, but it turns out that the default initrd.img is quite > different from the initrd.img file created with mkinitrd. Any one know > how to create a Kickstart initrd.img with custom kernel? Or the hacking > steps on a existing Centos 5 kickstart initrd.img? > yeah, all mkinitrd does from what i remember is look at your existing machine's kernel module settings in /etc/modprobe.conf and create an initrd.img with those settings, so you can boot your machine with the proper drivers for your storage and network devices. the install initrd.img has much more functionality, primarily centered around device discovery and kickstarting. instead of trying to create a new install initrd.img, it's much easier to modify the one that comes with the OS. i posted something to this list a few weeks ago detailing how to do so with a RHEL3/4 box: https://www.redhat.com/archives/kickstart-list/2007-May/msg00081.html it should be much the same for RHEL5/CentOS5 unless they've changed the format of an initrd.img, but that should give you a good idea of where to start. as far as a custom install kernel, i haven't tried this before. i imagine that as long as it calls the init program from the initrd.img and all of the kernel modules you put in the initrd.img are compiled against the custom kernel, you could use any kernel you want. chris From debian at herakles.homelinux.org Thu May 31 22:44:06 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Fri, 01 Jun 2007 06:44:06 +0800 Subject: Very different layout -- RE: How the Centos 5.0's kickstart initrd.img is build?? In-Reply-To: References: Message-ID: <465F4FB6.9070808@herakles.homelinux.org> Guolin Cheng wrote: > Hi, > > > > I've tried run mkinitrd to create an initrd.img file for Centos 5 > Kickstart, but it turns out that the default initrd.img is quite > different from the initrd.img file created with mkinitrd. Any one know > how to create a Kickstart initrd.img with custom kernel? Or the hacking > steps on a existing Centos 5 kickstart initrd.img? > > > > root at Centos5TestBox aaa]# ls -alF original mkinitrd/ > > mkinitrd/: > > total 40 > > drwxr-xr-x 9 root root 4096 May 30 17:22 ./ > > drwxr-xr-x 4 root root 4096 May 30 17:29 ../ > > drwx------ 2 root root 4096 May 30 17:22 bin/ > > drwx------ 3 root root 4096 May 30 17:22 dev/ > > drwx------ 2 root root 4096 May 30 17:22 etc/ > > -rwx------ 1 root root 1833 May 30 17:22 init* > > drwx------ 2 root root 4096 May 30 17:22 lib/ > > drwx------ 2 root root 4096 May 30 17:22 proc/ > > lrwxrwxrwx 1 root root 3 May 30 17:22 sbin -> bin/ > > drwx------ 2 root root 4096 May 30 17:22 sys/ > > drwx------ 2 root root 4096 May 30 17:22 sysroot/ > > > > original: > > total 52 > > drwxr-xr-x 11 root root 4096 May 30 17:21 ./ > > drwxr-xr-x 4 root root 4096 May 30 17:29 ../ > > -rw-r--r-- 1 root root 62 May 30 17:21 .buildstamp > > -rw-r--r-- 1 root root 123 May 30 17:21 .profile > > lrwxrwxrwx 1 root root 4 May 30 17:21 bin -> sbin/ > > drwxr-xr-x 2 root root 4096 May 30 17:21 dev/ > > drwxr-xr-x 3 root root 4096 May 30 17:21 etc/ > > lrwxrwxrwx 1 root root 10 May 30 17:21 init -> /sbin/init* > > drwxr-xr-x 2 root root 4096 May 30 17:21 modules/ > > drwxr-xr-x 2 root root 4096 May 30 17:21 proc/ > > drwxr-xr-x 2 root root 4096 May 30 17:21 sbin/ > > drwxr-xr-x 2 root root 4096 May 30 17:21 selinux/ > > drwxr-xr-x 2 root root 4096 May 30 17:21 sys/ > > drwxr-xr-x 2 root root 4096 May 30 17:21 tmp/ > > drwxr-xr-x 6 root root 4096 May 30 17:21 var/ > > root at hello11 aaa]# > > > > --Guolin > > > > ________________________________ > > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Guolin Cheng > Sent: Wednesday, May 30, 2007 4:45 PM > To: centos at centos.org; Discussion list about Kickstart > Subject: How the Centos 5.0's kickstart initrd.img is build?? > > > > Hi, > > Any one know how the Centos 5 kickstart initrd.img was built? I'll be > very appreciated if any light can be shed on this topic. The basic > problem to be solved here is: how to build a custom vmlinuz+initrd.img > for Centos 5 kickstart with my own custom built kernel? > > I am not sure mkinitrd is the right tool for this purpose? If do, which > are the modules I should include with '-preload' options? And if there > are any extra files/steps I should include/follow for the initrd.img > file? Not sure how the stage2.img is loaded by kernel after initrd.img > is loaded and run. Please help. Last time I created an initrd for EL, I hacked on an existing one by ungzipping it, then unpacking with cpio. To put it back together, I used cpio and didn't bother compressing it. Worked fine, should work here. I forget what I was kickstarting, may well have been an EL5 beta. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list