From alexander_rau at yahoo.com Sat Oct 1 04:18:30 2005 From: alexander_rau at yahoo.com (Alexander Rau) Date: Fri, 30 Sep 2005 21:18:30 -0700 (PDT) Subject: RHEL 4 DVD ? Message-ID: <20051001041830.95114.qmail@web52102.mail.yahoo.com> Hey: I followed the guide below on the nahant-list to create an installation DVD. Just wondering if anybody had any idea how I could add a folder with custom RPM to be installed in the %post section.... Thanks AR ================== mount -o ro,loop RHEL4-i386-WS-disc1.iso /mnt/tmp1 # etc. for discs 2,3,4 cp -a /mnt/tmp1/isolinux /mnt/tmp1/.discinfo . chmod +w isolinux/isolinux.bin {Edit ./.discinfo file to list four CDs "1,2,3,4" instead of just "1"} mkisofs -v -pad -o RHEL4-i386-DVD.iso \ -V "RHEL4-i386-DVD.2005-02-16" \ -b isolinux/isolinux.bin -c isolinux/boot.cat \ -no-emul-boot -boot-load-size 4 -boot-info-table \ -r -R -T -m TRANS.TBL \ -x /mnt/tmp1/.discinfo \ -x /mnt/tmp1/isolinux \ -graft-points /mnt/tmp1 .discinfo=.discinfo isolinux/=isolinux \ RedHat=/mnt/tmp2/RedHat \ RedHat=/mnt/tmp3/RedHat \ RedHat=/mnt/tmp4/RedHat /usr/lib/anaconda-runtime/implantisomd5 RHEL4-i386-DVD.iso From dan at half-asleep.com Mon Oct 3 12:36:57 2005 From: dan at half-asleep.com (Daniel Segall) Date: Mon, 3 Oct 2005 08:36:57 -0400 (EDT) Subject: Multiple ks.cfg's on a single boot media? In-Reply-To: <888f25240509300819j6ed473f2qdc10e3b5501da944@mail.gmail.com> References: <34648.192.80.55.74.1128092532.squirrel@webmail.half-asleep.com> <888f25240509300819j6ed473f2qdc10e3b5501da944@mail.gmail.com> Message-ID: <60882.192.80.55.74.1128343017.squirrel@webmail.half-asleep.com> I can't seem to make this work right. My CD boots up correctly, and I get my menu (boot.msg). However, whenever I try to select any of my images, I just get an error stating it could not find vmlinuz-whatever. My isolinux.cfg looks like this: label Rescue kernel vmlinuz-RHES-4-U1 append initrd=initrd-RHES-4-U1.img rescue ks=cdrom:/rescue.ks.cfg text devfs=nomount ramdisk=10240 vga=0xF01 label RHES-3 kernel vmlinuz-RHES-3-U5 append initrd=initrd-RHES-3-U5.img ks=cdrom:/RHES-3-U5-ks.cfg label RHES-4 kernel vmlinuz-RHES-4-U1 append initrd=initrd-RHES-4-U1.img ks=cdrom:/kickstart/RHES-4-U1-ks.cfg label FC-4 kernel vmlinuz-FC-4 append initrd=initrd-FC-4.img ks=cdrom:/FC-4-ks.cfg I started with the boot.iso from FC4, and added all of my vmlinuz images from each version, and my various ks.cfg's. I made a test option, using the vmlinuz image that was on the original FC4 boot.iso with the following options, and it works as expected. label test kernel vmlinuz append initrd=initrd.img ks=cdrom:/FC-4-ks.cfg Any ideas why it can't find anything but vmlinuz? Thanks, -Dan On Fri, September 30, 2005 11:19 am, Robert Wehner said: > On 9/30/05, Daniel Segall wrote: >> Hey Guys, >> >> I am experimenting with moving my pxe based kickstart system to a >> bootable >> media/network install. Currently I have my kickstart server on a private >> network, with everything centrally located on that server. I then >> connect >> servers to that network and boot them into pxe, where my menu of images >> will pop up. This works great, but it's a pain in the ass because I have >> to physically connect all of the systems to my network in order to kick >> them. >> >> What I would like to do is setup a cd containing all of my boot images, >> kickstart configs, post install scripts, etc. Then just do http installs >> over the public network. I know how to create the boot iso with >> everything, but i'm not sure how to specify the individual ks.cfg's for >> each image in the isolinux.cfg. With a single config, I could just use >> ks=cdrom. Is it possible to use a path with this? > > It is possible. I update the isolinux.cfg file to append extra stanzas > for each ks.cfg on the CD. Something like: > > label webserver > kernel vmlinuz > append initrd=initrd.img ks=cdrom:/webserver.ks.cfg > label database > kernel vmlinuz > append initrd=initrd.img ks=cdrom:/database.ks.cfg > > At the boot prompt the user just types in 'database' or 'webserver' to > get the correct config. they type nothing if they want an interactive > install. > > -- > ---- > Robert Wehner > rwehner at gmail.com > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > From dan at half-asleep.com Mon Oct 3 12:42:52 2005 From: dan at half-asleep.com (Daniel Segall) Date: Mon, 3 Oct 2005 08:42:52 -0400 (EDT) Subject: RHEL 4 DVD ? In-Reply-To: <20051001041830.95114.qmail@web52102.mail.yahoo.com> References: <20051001041830.95114.qmail@web52102.mail.yahoo.com> Message-ID: <44534.192.80.55.74.1128343372.squirrel@webmail.half-asleep.com> This is just a guess, as I have never tried making a DVD... Couldn't you simply copy the iso's to disk, and create a new directory under say disk 4, then proceed with the steps you listed? Using your local copy's rather than the mounted images of course. -Dan On Sat, October 1, 2005 12:18 am, Alexander Rau said: > Hey: > > I followed the guide below on the nahant-list to > create an installation DVD. > > Just wondering if anybody had any idea how I could add > a folder with custom RPM to be installed in the %post > section.... > > Thanks > > AR > > ================== > mount -o ro,loop RHEL4-i386-WS-disc1.iso /mnt/tmp1 # > etc. for discs 2,3,4 > > > cp -a /mnt/tmp1/isolinux /mnt/tmp1/.discinfo . > > chmod +w isolinux/isolinux.bin > > {Edit ./.discinfo file to list four CDs "1,2,3,4" > instead of just "1"} > > mkisofs -v -pad -o RHEL4-i386-DVD.iso \ > -V "RHEL4-i386-DVD.2005-02-16" \ > -b isolinux/isolinux.bin -c isolinux/boot.cat \ > -no-emul-boot -boot-load-size 4 -boot-info-table \ > -r -R -T -m TRANS.TBL \ > -x /mnt/tmp1/.discinfo \ > -x /mnt/tmp1/isolinux \ > -graft-points /mnt/tmp1 .discinfo=.discinfo > isolinux/=isolinux \ > RedHat=/mnt/tmp2/RedHat \ > RedHat=/mnt/tmp3/RedHat \ > RedHat=/mnt/tmp4/RedHat > > > /usr/lib/anaconda-runtime/implantisomd5 > RHEL4-i386-DVD.iso > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > From bda20 at cam.ac.uk Mon Oct 3 12:56:05 2005 From: bda20 at cam.ac.uk (Ben) Date: Mon, 3 Oct 2005 13:56:05 +0100 (BST) Subject: Multiple ks.cfg's on a single boot media? In-Reply-To: <60882.192.80.55.74.1128343017.squirrel@webmail.half-asleep.com> References: <34648.192.80.55.74.1128092532.squirrel@webmail.half-asleep.com> <888f25240509300819j6ed473f2qdc10e3b5501da944@mail.gmail.com> <60882.192.80.55.74.1128343017.squirrel@webmail.half-asleep.com> Message-ID: On Mon, 3 Oct 2005, Daniel Segall wrote: > I can't seem to make this work right. My CD boots up correctly, and I get > my menu (boot.msg). However, whenever I try to select any of my images, I > just get an error stating it could not find vmlinuz-whatever. My > isolinux.cfg looks like this: > > [description] > > Any ideas why it can't find anything but vmlinuz? I had this problem. I solved it by taking all the hyphens ("-" characters) out of the vmlinuz filenames. I.e. vmlinuz4 rather than vmlinuz-4. Just to be on the safe side I did it with the initrd.img files as well. Ben -- Unix Support, MISD, University of Cambridge, England Plugger of wire, typer of keyboard, imparter of Clue Life Is Short. It's All Good. From dan at half-asleep.com Mon Oct 3 13:01:52 2005 From: dan at half-asleep.com (Daniel Segall) Date: Mon, 3 Oct 2005 09:01:52 -0400 (EDT) Subject: Multiple ks.cfg's on a single boot media? In-Reply-To: References: <34648.192.80.55.74.1128092532.squirrel@webmail.half-asleep.com> <888f25240509300819j6ed473f2qdc10e3b5501da944@mail.gmail.com> <60882.192.80.55.74.1128343017.squirrel@webmail.half-asleep.com> Message-ID: <59096.192.80.55.74.1128344512.squirrel@webmail.half-asleep.com> Thanks! I was kind of thinking this was the problem. -Dan On Mon, October 3, 2005 8:56 am, Ben said: > On Mon, 3 Oct 2005, Daniel Segall wrote: > >> I can't seem to make this work right. My CD boots up correctly, and I >> get >> my menu (boot.msg). However, whenever I try to select any of my images, >> I >> just get an error stating it could not find vmlinuz-whatever. My >> isolinux.cfg looks like this: >> >> [description] >> >> Any ideas why it can't find anything but vmlinuz? > > I had this problem. I solved it by taking all the hyphens ("-" > characters) > out of the vmlinuz filenames. I.e. vmlinuz4 rather than vmlinuz-4. Just > to > be on the safe side I did it with the initrd.img files as well. > > Ben > -- > Unix Support, MISD, University of Cambridge, England > Plugger of wire, typer of keyboard, imparter of Clue > Life Is Short. It's All Good. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > From alexander_rau at yahoo.com Mon Oct 3 13:24:00 2005 From: alexander_rau at yahoo.com (Alexander Rau) Date: Mon, 3 Oct 2005 06:24:00 -0700 (PDT) Subject: RHEL 4 DVD ? In-Reply-To: <44534.192.80.55.74.1128343372.squirrel@webmail.half-asleep.com> Message-ID: <20051003132400.19512.qmail@web52102.mail.yahoo.com> Thanks I actually just copied disc one to a local directory, made the necassary changes and added my cusotm RPM to it.... Works great....no more switching CD's.... AR --- Daniel Segall wrote: > This is just a guess, as I have never tried making a > DVD... > > Couldn't you simply copy the iso's to disk, and > create a new directory > under say disk 4, then proceed with the steps you > listed? Using your local > copy's rather than the mounted images of course. > > -Dan > > > On Sat, October 1, 2005 12:18 am, Alexander Rau > said: > > Hey: > > > > I followed the guide below on the nahant-list to > > create an installation DVD. > > > > Just wondering if anybody had any idea how I could > add > > a folder with custom RPM to be installed in the > %post > > section.... > > > > Thanks > > > > AR > > > > ================== > > mount -o ro,loop RHEL4-i386-WS-disc1.iso /mnt/tmp1 > # > > etc. for discs 2,3,4 > > > > > > cp -a /mnt/tmp1/isolinux /mnt/tmp1/.discinfo . > > > > chmod +w isolinux/isolinux.bin > > > > {Edit ./.discinfo file to list four CDs "1,2,3,4" > > instead of just "1"} > > > > mkisofs -v -pad -o RHEL4-i386-DVD.iso \ > > -V "RHEL4-i386-DVD.2005-02-16" \ > > -b isolinux/isolinux.bin -c isolinux/boot.cat \ > > -no-emul-boot -boot-load-size 4 -boot-info-table > \ > > -r -R -T -m TRANS.TBL \ > > -x /mnt/tmp1/.discinfo \ > > -x /mnt/tmp1/isolinux \ > > -graft-points /mnt/tmp1 .discinfo=.discinfo > > isolinux/=isolinux \ > > RedHat=/mnt/tmp2/RedHat \ > > RedHat=/mnt/tmp3/RedHat \ > > RedHat=/mnt/tmp4/RedHat > > > > > > /usr/lib/anaconda-runtime/implantisomd5 > > RHEL4-i386-DVD.iso > > > > > > _______________________________________________ > > 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 dan at half-asleep.com Mon Oct 3 13:35:05 2005 From: dan at half-asleep.com (Daniel Segall) Date: Mon, 3 Oct 2005 09:35:05 -0400 (EDT) Subject: Multiple ks.cfg's on a single boot media? In-Reply-To: <59096.192.80.55.74.1128344512.squirrel@webmail.half-asleep.com> References: <34648.192.80.55.74.1128092532.squirrel@webmail.half-asleep.com> <888f25240509300819j6ed473f2qdc10e3b5501da944@mail.gmail.com> <60882.192.80.55.74.1128343017.squirrel@webmail.half-asleep.com> <59096.192.80.55.74.1128344512.squirrel@webmail.half-asleep.com> Message-ID: <49525.192.80.55.74.1128346505.squirrel@webmail.half-asleep.com> I spoke too soon. It still gives me the same error after renaming all of my images and editing the isolinux.cfg to match: boot: Rescue Could not find kernel image: vmlinuzRHES4U1 boot: FC-4 Could not find kernel image: vmlinuzFC4 Tried making the names lowercase as well, just to be sure, and I still get no love. Any other thoughts? Thanks, -Dan On Mon, October 3, 2005 9:01 am, Daniel Segall said: > Thanks! > I was kind of thinking this was the problem. > > -Dan > > > On Mon, October 3, 2005 8:56 am, Ben said: >> On Mon, 3 Oct 2005, Daniel Segall wrote: >> >>> I can't seem to make this work right. My CD boots up correctly, and I >>> get >>> my menu (boot.msg). However, whenever I try to select any of my images, >>> I >>> just get an error stating it could not find vmlinuz-whatever. My >>> isolinux.cfg looks like this: >>> >>> [description] >>> >>> Any ideas why it can't find anything but vmlinuz? >> >> I had this problem. I solved it by taking all the hyphens ("-" >> characters) >> out of the vmlinuz filenames. I.e. vmlinuz4 rather than vmlinuz-4. >> Just >> to >> be on the safe side I did it with the initrd.img files as well. >> >> Ben >> -- >> Unix Support, MISD, University of Cambridge, England >> Plugger of wire, typer of keyboard, imparter of Clue >> Life Is Short. It's All Good. >> >> _______________________________________________ >> 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 brilong at cisco.com Mon Oct 3 13:42:08 2005 From: brilong at cisco.com (Brian Long) Date: Mon, 03 Oct 2005 09:42:08 -0400 Subject: Multiple ks.cfg's on a single boot media? In-Reply-To: <49525.192.80.55.74.1128346505.squirrel@webmail.half-asleep.com> References: <34648.192.80.55.74.1128092532.squirrel@webmail.half-asleep.com> <888f25240509300819j6ed473f2qdc10e3b5501da944@mail.gmail.com> <60882.192.80.55.74.1128343017.squirrel@webmail.half-asleep.com> <59096.192.80.55.74.1128344512.squirrel@webmail.half-asleep.com> <49525.192.80.55.74.1128346505.squirrel@webmail.half-asleep.com> Message-ID: <1128346929.4008.17.camel@brilong-lnx.cisco.com> On Mon, 2005-10-03 at 09:35 -0400, Daniel Segall wrote: > I spoke too soon. > It still gives me the same error after renaming all of my images and > editing the isolinux.cfg to match: isolinux may have an 8-character limitation. Try reducing all kernel names to something <= 8 characters. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From wpilorz at bdk.pl Mon Oct 3 14:00:16 2005 From: wpilorz at bdk.pl (Wojtek.Pilorz) Date: Mon, 3 Oct 2005 16:00:16 +0200 (CEST) Subject: Multiple ks.cfg's on a single boot media? In-Reply-To: <49525.192.80.55.74.1128346505.squirrel@webmail.half-asleep.com> Message-ID: On Mon, 3 Oct 2005, Daniel Segall wrote: > Date: Mon, 3 Oct 2005 09:35:05 -0400 (EDT) > From: Daniel Segall > Reply-To: Discussion list about Kickstart > To: Discussion list about Kickstart > Subject: Re: Multiple ks.cfg's on a single boot media? > > I spoke too soon. > It still gives me the same error after renaming all of my images and > editing the isolinux.cfg to match: > > boot: Rescue > Could not find kernel image: vmlinuzRHES4U1 > boot: FC-4 > Could not find kernel image: vmlinuzFC4 > > Tried making the names lowercase as well, just to be sure, and I still get > no love. > > Any other thoughts? > > Thanks, > -Dan Please see a quote from http://syslinux.zytor.com/iso.php: Note that ISOLINUX only uses the "plain" ISO 9660 filenames, i.e. it does not support Rock Ridge or Joliet filenames. It can still be used on a disk which uses Rock Ridge and/or Joliet extensions, of course. Under Linux, you can verify the plain filenames by mounting with the "-o norock,nojoliet" option to the mount command. Note, however, that ISOLINUX does support "long" (level 2) ISO 9660 plain filenames, so if compatibility with short-names-only operating systems like MS-DOS is not an issue, you can use the "-l" or "-iso-level 2" option to mkisofs to generate long (up to 31 characters) plain filenames. ISOLINUX does not support discontiguous files, interleaved mode, or logical block and sector sizes other than 2048. This should normally not be a problem. I think this explains your problems. Best regards, Wojtek From dan at half-asleep.com Mon Oct 3 15:29:02 2005 From: dan at half-asleep.com (Daniel Segall) Date: Mon, 3 Oct 2005 11:29:02 -0400 (EDT) Subject: Multiple ks.cfg's on a single boot media? In-Reply-To: <1128346929.4008.17.camel@brilong-lnx.cisco.com> References: <34648.192.80.55.74.1128092532.squirrel@webmail.half-asleep.com> <888f25240509300819j6ed473f2qdc10e3b5501da944@mail.gmail.com> <60882.192.80.55.74.1128343017.squirrel@webmail.half-asleep.com> <59096.192.80.55.74.1128344512.squirrel@webmail.half-asleep.com> <49525.192.80.55.74.1128346505.squirrel@webmail.half-asleep.com> <1128346929.4008.17.camel@brilong-lnx.cisco.com> Message-ID: <41049.192.80.55.74.1128353342.squirrel@webmail.half-asleep.com> That was it. It applies to all filenames, including initrd images, and kickstart config files. Thanks! -Dan On Mon, October 3, 2005 9:42 am, Brian Long said: > On Mon, 2005-10-03 at 09:35 -0400, Daniel Segall wrote: >> I spoke too soon. >> It still gives me the same error after renaming all of my images and >> editing the isolinux.cfg to match: > > isolinux may have an 8-character limitation. Try reducing all kernel > names to something <= 8 characters. > > /Brian/ > > -- > Brian Long | | | > IT Data Center Systems | .|||. .|||. > Cisco Linux Developer | ..:|||||||:...:|||||||:.. > Phone: (919) 392-7363 | C i s c o S y s t e m s > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > From AjitabhPandey at Gmail.com Mon Oct 3 22:06:20 2005 From: AjitabhPandey at Gmail.com (Ajitabh Pandey) Date: Mon, 3 Oct 2005 23:06:20 +0100 Subject: Grub not getting installed after kickstart installation Message-ID: <20051003220620.GA10480@bluedeb> Hi, I am trying to make a kickstart CD to install RHEL3 Update 3 on an HP DL585 server. The install is taking place on disks controlled by internal RAID controller (HP Smart array 5i). The install seems to work like breeze. At the end of installation, there are some issues - (1.) All the post install scripts which tries to mount the cd-rom to copy files from CD to hardisk are failing. Possibly some silly mistake. Will be glad if someone points those out. (2.) Although the grub was being installed as a part of the install process, when trying to boot from hardisk I was just staring at a message - "Attempting to boot from Hard Disk (C:)". I tried installing grub through the post-install section of kickstart but that also showed me just the word: GRUB (3.) When I boot from the Cd in rescue mode and try to do a chroot into /mnt/sysimage for doing grub-install i find that most core-utils binaries like ls, rm, cp etc are not installed. While I clearly remember seeing core-utils being installed during installation and the rpm is present on Cd as well as mentioned in the kickstart file. What's happening. On the other hand if I do a minimum install from first two CDs of RHEL3 Update 3 the system is perfectly bootable and works fine. Can someone point out what am I doing wrong? I am attaching my kickstart file for reference. Cheers. -- Ajitabh Pandey http://www.ajitabhpandey.info/ ICQ - 150615062 Registered Linux User - 240748 GnuPG Key ID - 35CF8CC4 Key fingerprint = E1A8 657D BE0C 4747 52EC 10C4 1AC2 C124 35CF 8CC4 ----------------------------------- Q: How was Thomas J. Watson buried? A: 9 edge down. -------------- next part -------------- # Name - wwllnx1001.ks # Description - Kickstart configuration file for wwllnx1001.bds.tv # Author - Ajitabh Pandey # History - Ver 0.1 on 27th September 2005 # # System language lang en_GB # Language modules to install langsupport da_DK da_DK da nl_NL nl_NL nl fr_FR fr_FR fr de_DE de_DE de it_IT it_IT it no_NO no_NO no pt_PT pt_PT pt es_ES es_ES es sv_SE sv_SE sv en_GB en --default=en_GB # System keyboard keyboard uk # System mouse mouse msintellips/2 # X configuration xconfig --card "ATI Mach64" --videoram 8128 --hsync 31.5-37.9 --vsync 50.0-61.0 --resolution 800x600 --depth 24 --startxonboot --defaultdesktop gnome # Network configuration network --device eth0 --bootproto dhcp network --device eth1 --bootproto dhcp # Firewall firewall --disabled # Sets up the authentication options for the system authconfig --enableshadow --enablemd5 # Sytem timezone timezone Europe/London # Root password rootpw --iscrypted $1$HSr5n.Fp$sLq58u9VFqGBavm7aQ5Cf1 # Reboot after installation # reboot # Install Red Hat Linux instead of upgrade install # Use CDROM installation media cdrom # Clear the Master Boot Record zerombr yes # Partition clearing information clearpart --all # Bootloader location bootloader --location=mbr # Partition details # /boot partition part /boot --fstype "ext3" --size=128 --ondisk=cciss/c0d0 # swap partitions part swap --size=2048 --ondisk=cciss/c0d0 part swap --size=2048 --ondisk=cciss/c0d0 part swap --size=2048 --ondisk=cciss/c0d0 part swap --size=2048 --ondisk=cciss/c0d0 part swap --size=2048 --ondisk=cciss/c0d0 part swap --size=2048 --ondisk=cciss/c0d0 part swap --size=2048 --ondisk=cciss/c0d0 part swap --size=2048 --ondisk=cciss/c0d0 # LVM physical volume which can grow in size part pv.100000 --size=1 --grow --ondisk=cciss/c0d0 # Volume Group vg_root volgroup vg_root pv.100000 # Logical Volumes under Volume Group vg_root logvol /apps --fstype ext3 --name=lv_apps --vgname=vg_root --size=2048 logvol /var/www --fstype ext3 --name=lv_www --vgname=vg_root --size=16384 logvol /var --fstype ext3 --name=lv_var --vgname=vg_root --size=2048 logvol / --fstype ext3 --name=lv_root --vgname=vg_root --size=2048 logvol /tmp --fstype ext3 --name=lv_tmp --vgname=vg_root --size=2048 logvol /var/ftp --fstype ext3 --name=lv_ftp --vgname=vg_root --size=1 --grow logvol /usr --fstype ext3 --name=lv_usr --vgname=vg_root --size=16384 # Package install information %packages --resolvedeps libgcc setup basesystem tzdata glibc bzip2-libs cracklib e2fsprogs ethtool gdbm glib2 hdparm laus-libs libacl losetup mingetty bash MAKEDEV net-tools poptk rootfiles shadow-utils newt hotplug cracklib-dicts file cpio findutils grep ash krb5-libs ncurses less openssl psmisc python rhpl dev authconfig sysklogd tar util-linux vim-minimal initscripts cyrus-sasl-md5 kernel-smp libuser usermode redhat-config-mouse rpm-libs mailcap redhat-menus dump dos2unix eject hesiod krbafs attr libjpeg libtool-libs lsof bzip2 htmlview nc pam_passwdqc parted pax fbset logrotate pspell rdist rsh schedutils lockdev ntsysv slocate symlinks tftp unix2dos wireless-tools zip fontconfig libtiff binutils elfutils eal3-certification jwhois krbafs-utils libgcj m4 mgetty mtools man mtr nscd authd bind-utils libwvstreams pdksh psacct bc lftp libxml2-python rhnlib ppp stunnel syslinux talk tcsh time utempter apmd cyrus-sasl-plain dhclient ipsec-tools iptables kernel-pcmcia-cs autofs nss_ldap openssh-clients openssh-server portmap prelink redhat-config-securitylevel-tui sendmail tcpdump vixie-cron XFree86-Mesa-libGL cups ypbind rpm-python up2date redhat-release hwdata redhat-logos filesystem termcap glibc-common beecrypt chkconfig db4 elfutils-libelf expat glib gmp iputils libattr libtermcap lvm mktemp iproute mount pcre raidtools setserial slang usbutils words zlib info ed gawk coreutils grub modutils gpm gzip procps readline pyxf86config sed pam kudzu SysVinit mkinitrd vim-common which cyrus-sasl kernel openldap passwd kbd rpm eal3-certification-doc man-pages rmt specspo dosfstools finger jfsutils lha acl libstdc++ lslk mailx crontabs mt-st ncompress pam_smb patch perl perl-Filter procmail rdate rpmdb-redhat rsync setarch netconfig setuptool star tcp_wrappers traceroute unzip XFree86-libs-data freetype libpng libxml2 diffutils at groff krb5-workstation laus logwatch make irda-utils aspell minicom nano nss_db bind-libs cups-libs pam_krb5 pinfo pyOpenSSL ftp gettext python-optik jpackage-utils sharutils sudo sysreport mkbootdisk telnet tmpwatch wget cyrus-sasl-gssapi devlabel diskdumputils isdn4k-utils iptables-ipv6 kernel-utils gnupg openssh netdump pciutils nfs-utils quota rp-pppoe mdadm vconfig wvdial XFree86-libs xinetd redhat-lsb yp-tools redhat-config-network-tui up2date-update comps %post # Installing grub to the MBR at the end of install: /sbin/grub --batch --device-map=/dev/null </etc/hosts # Copy files from CDROM /bin/mkdir /mnt/cdrom /bin/mount /dev/hda /mnt/cdrom /bin/cp /mnt/cdrom/inittab /etc /bin/tar -xvzf /mnt/cdrom/netback-up.tar.gz # Unmount CDROM and eject the CD /bin/umount /dev/hda /usr/bin/eject /dev/hda -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From Chip.Shabazian at bankofamerica.com Mon Oct 3 22:27:45 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Mon, 03 Oct 2005 15:27:45 -0700 Subject: Grub not getting installed after kickstart installation Message-ID: Are you installing from a local CD, or via Virtual Media over the iLO? Try removing everything from your %post and make /boot it's own NATIVE partition, not LVM -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ajitabh Pandey Sent: Monday, October 03, 2005 3:06 PM To: kickstart-list at redhat.com Subject: Grub not getting installed after kickstart installation Hi, I am trying to make a kickstart CD to install RHEL3 Update 3 on an HP DL585 server. The install is taking place on disks controlled by internal RAID controller (HP Smart array 5i). The install seems to work like breeze. At the end of installation, there are some issues - (1.) All the post install scripts which tries to mount the cd-rom to copy files from CD to hardisk are failing. Possibly some silly mistake. Will be glad if someone points those out. (2.) Although the grub was being installed as a part of the install process, when trying to boot from hardisk I was just staring at a message - "Attempting to boot from Hard Disk (C:)". I tried installing grub through the post-install section of kickstart but that also showed me just the word: GRUB (3.) When I boot from the Cd in rescue mode and try to do a chroot into /mnt/sysimage for doing grub-install i find that most core-utils binaries like ls, rm, cp etc are not installed. While I clearly remember seeing core-utils being installed during installation and the rpm is present on Cd as well as mentioned in the kickstart file. What's happening. On the other hand if I do a minimum install from first two CDs of RHEL3 Update 3 the system is perfectly bootable and works fine. Can someone point out what am I doing wrong? I am attaching my kickstart file for reference. Cheers. -- Ajitabh Pandey http://www.ajitabhpandey.info/ ICQ - 150615062 Registered Linux User - 240748 GnuPG Key ID - 35CF8CC4 Key fingerprint = E1A8 657D BE0C 4747 52EC 10C4 1AC2 C124 35CF 8CC4 ----------------------------------- Q: How was Thomas J. Watson buried? A: 9 edge down. From sjbrowne at bluebottle.com Tue Oct 4 00:14:24 2005 From: sjbrowne at bluebottle.com (Stuart J. Browne) Date: Tue, 4 Oct 2005 10:14:24 +1000 Subject: Grub not getting installed after kickstart installation References: <20051003220620.GA10480@bluedeb> Message-ID: <200510040014.j940ETdR029999@bluebottle-fe4.bluebottle.com> Ajitabh, In your %post section, you have these two lines: # /bin/echo -e "root (hd0,0) \n install /grub/stage1 d (hd0) /grub/stage2 p \ (hd0,0)/grub/grub.conf \n quit" | /sbin/grub --batch You've only commented out the first part of it, not the second. It'd come up with an error 'command not found: hd0,0', or some such. Comment out that 2nd line, and try it again. Now, as for 'ls', 'mv' etc. not being mounted when you chroot in emergency mode, is somewhat odd. Are you sure it's not just that /bin/ isn't in the emergency sesions path? You've actually looked in /bin, not found them, double checked to ensure that the 'cureutils' package is installed? Stuart J. Browne ----- Original Message ----- From: "Ajitabh Pandey" To: Sent: Tuesday, October 04, 2005 8:06 AM Subject: Grub not getting installed after kickstart installation Hi, I am trying to make a kickstart CD to install RHEL3 Update 3 on an HP DL585 server. The install is taking place on disks controlled by internal RAID controller (HP Smart array 5i). The install seems to work like breeze. At the end of installation, there are some issues - (1.) All the post install scripts which tries to mount the cd-rom to copy files from CD to hardisk are failing. Possibly some silly mistake. Will be glad if someone points those out. (2.) Although the grub was being installed as a part of the install process, when trying to boot from hardisk I was just staring at a message - "Attempting to boot from Hard Disk (C:)". I tried installing grub through the post-install section of kickstart but that also showed me just the word: GRUB (3.) When I boot from the Cd in rescue mode and try to do a chroot into /mnt/sysimage for doing grub-install i find that most core-utils binaries like ls, rm, cp etc are not installed. While I clearly remember seeing core-utils being installed during installation and the rpm is present on Cd as well as mentioned in the kickstart file. What's happening. On the other hand if I do a minimum install from first two CDs of RHEL3 Update 3 the system is perfectly bootable and works fine. Can someone point out what am I doing wrong? I am attaching my kickstart file for reference. Cheers. -- Ajitabh Pandey http://www.ajitabhpandey.info/ ICQ - 150615062 Registered Linux User - 240748 GnuPG Key ID - 35CF8CC4 Key fingerprint = E1A8 657D BE0C 4747 52EC 10C4 1AC2 C124 35CF 8CC4 ----------------------------------- Q: How was Thomas J. Watson buried? A: 9 edge down. -------------------------------------------------------------------------------- > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From ajitabhpandey at gmail.com Tue Oct 4 09:09:44 2005 From: ajitabhpandey at gmail.com (Ajitabh Pandey) Date: Tue, 4 Oct 2005 10:09:44 +0100 Subject: Grub not getting installed after kickstart installation In-Reply-To: References: Message-ID: On 10/3/05, Shabazian, Chip wrote: > > Are you installing from a local CD, or via Virtual Media over the iLO? I am installing from a local CD. Try removing everything from your %post and make /boot it's own NATIVE > partition, not LVM /boot is primary partition and not a part of LVM (check the ks file posted with the original post). Also I added those %post commands later, it was not working anyway so I thought I will try installing Grub through %post. Installation always went smooth but when I try to boot from the hard disk, nothing is coming up. So it looks like a problem with grub getting installed on the MBR. Cheers. -- Regards. Ajitabh Pandey http://www.ajitabhpandey.info/ ICQ - 150615062 Registered Linux User - 240748 GnuPG Key ID - 35CF8CC4 Key fingerprint = E1A8 657D BE0C 4747 52EC 10C4 1AC2 C124 35CF 8CC4 -------------- next part -------------- An HTML attachment was scrubbed... URL: From ajitabhpandey at gmail.com Tue Oct 4 09:16:07 2005 From: ajitabhpandey at gmail.com (Ajitabh Pandey) Date: Tue, 4 Oct 2005 10:16:07 +0100 Subject: Grub not getting installed after kickstart installation In-Reply-To: <200510040014.j940ETdR029999@bluebottle-fe4.bluebottle.com> References: <20051003220620.GA10480@bluedeb> <200510040014.j940ETdR029999@bluebottle-fe4.bluebottle.com> Message-ID: Hi, On 10/4/05, Stuart J. Browne wrote: > > Ajitabh, > > In your %post section, you have these two lines: > # /bin/echo -e "root (hd0,0) \n install /grub/stage1 d (hd0) /grub/stage2 > p > \ > (hd0,0)/grub/grub.conf \n quit" | /sbin/grub --batch > > You've only commented out the first part of it, not the second. It'd come > up with an error 'command not found: hd0,0', or some such. Yeah, I realised that after I posted to the list. I had removed those lines entirely in my next attempt. My point was first there should not be a need of installing grub (am I really correct in this ??) through post install section. Second even if I am installing it using the grub --batch command (lines prior to the one you pointed out), why is it still not working. Comment out that 2nd line, and try it again. Tried and still the same results. Now, as for 'ls', 'mv' etc. not being mounted when you chroot in emergency > mode, is somewhat odd. Are you sure it's not just that /bin/ isn't in the > emergency sesions path? You've actually looked in /bin, not found them, > double checked to ensure that the 'cureutils' package is installed? I am sure that the coreutils package is installed as there is no error while installing it. I checked for the presence of these commands when I chrooted to /mnt/sysimage and then tried to run grub-install. All i got from grub-install was - "could not find rm....." Then I came out of chroot and using the ls from the cd checked for the presentce of crusial core--utils files in /mnt/sysimage/bin and couldn't find them. -- Regards. Ajitabh Pandey http://www.ajitabhpandey.info/ ICQ - 150615062 Registered Linux User - 240748 GnuPG Key ID - 35CF8CC4 Key fingerprint = E1A8 657D BE0C 4747 52EC 10C4 1AC2 C124 35CF 8CC4 -------------- next part -------------- An HTML attachment was scrubbed... URL: From brilong at cisco.com Tue Oct 4 14:06:06 2005 From: brilong at cisco.com (Brian Long) Date: Tue, 04 Oct 2005 10:06:06 -0400 Subject: Grub not getting installed after kickstart installation In-Reply-To: <20051003220620.GA10480@bluedeb> References: <20051003220620.GA10480@bluedeb> Message-ID: <1128434766.4049.28.camel@brilong-lnx.cisco.com> On Mon, 2005-10-03 at 23:06 +0100, Ajitabh Pandey wrote: > Hi, > > I am trying to make a kickstart CD to install RHEL3 Update 3 on an HP DL585 > server. The install is taking place on disks controlled by internal RAID > controller (HP Smart array 5i). The install seems to work like breeze. At the > end of installation, there are some issues - Hello, Is there any other storage (i.e. SAN) connected to this host? If so, I wonder if grub is installed to /dev/sda MBR instead of /dev/cciss/c0d0? If you have alternate storage, you could use "bootloader --location=mbr --driveorder=cciss/c0d0,sda". This will ensure GRUB installs to c0d0. Just an idea. We have a DL585 running RHEL 3 Update 5 with zero GRUB issues. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From Chip.Shabazian at bankofamerica.com Tue Oct 4 16:28:06 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Tue, 04 Oct 2005 09:28:06 -0700 Subject: Grub not getting installed after kickstart installation Message-ID: Ok, so this installed ONCE? I have run into this issue with the DL servers (and the CCISS controller). This happens when you have previously setup disks using LVM on a cciss controller. If you delete and recreate the partitions in the CCISS Controller ROM BIOS setup, it will work again, one more time (that's how I originally figured it out). OR, add -initlabel to your clearpart command in the kickstart file: clearpart --all -initlabel Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ajitabh Pandey Sent: Tuesday, October 04, 2005 2:10 AM To: kickstart-list at redhat.com Subject: Re: Grub not getting installed after kickstart installation On 10/3/05, Shabazian, Chip wrote: Are you installing from a local CD, or via Virtual Media over the iLO? I am installing from a local CD. Try removing everything from your %post and make /boot it's own NATIVE partition, not LVM /boot is primary partition and not a part of LVM (check the ks file posted with the original post). Also I added those %post commands later, it was not working anyway so I thought I will try installing Grub through %post. Installation always went smooth but when I try to boot from the hard disk, nothing is coming up. So it looks like a problem with grub getting installed on the MBR. Cheers. -- Regards. Ajitabh Pandey http://www.ajitabhpandey.info/ ICQ - 150615062 Registered Linux User - 240748 GnuPG Key ID - 35CF8CC4 Key fingerprint = E1A8 657D BE0C 4747 52EC 10C4 1AC2 C124 35CF 8CC4 -------------- next part -------------- An HTML attachment was scrubbed... URL: From brilong at cisco.com Wed Oct 5 13:55:03 2005 From: brilong at cisco.com (Brian Long) Date: Wed, 05 Oct 2005 09:55:03 -0400 Subject: Core includes Dialup?!? Message-ID: <1128520503.4108.1.camel@brilong-lnx.cisco.com> What is the Red Hat reason for including the Dialup comps group in Core? This means any basic install has crap like wvdial, ppp, etc. I explicitly have to exclude a lot of these packages that shouldn't be on any profiles except mobile laptops. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From Matt.Fahrner at COAT.COM Wed Oct 5 14:03:32 2005 From: Matt.Fahrner at COAT.COM (Matt Fahrner) Date: Wed, 05 Oct 2005 10:03:32 -0400 Subject: Core includes Dialup?!? In-Reply-To: <1128520503.4108.1.camel@brilong-lnx.cisco.com> References: <1128520503.4108.1.camel@brilong-lnx.cisco.com> Message-ID: <4343DD34.7040001@COAT.COM> I agree its inclusion is dubious, but I do know a lot of people who have workstations, not just laptops, that use PPP etc... - Matt Brian Long wrote: > What is the Red Hat reason for including the Dialup comps group in Core? > This means any basic install has crap like wvdial, ppp, etc. I > explicitly have to exclude a lot of these packages that shouldn't be on > any profiles except mobile laptops. > > /Brian/ -- --------------------------------------------------------------------- Matt Fahrner 2 South Park St. Manager of Networking Willis House Burlington Coat Factory Warehouse Lebanon, N.H. 03766 TEL: (603) 448-4100 x5150 USA FAX: (603) 443-6190 Matt.Fahrner at COAT.COM --------------------------------------------------------------------- From dan at half-asleep.com Wed Oct 5 14:22:30 2005 From: dan at half-asleep.com (Daniel Segall) Date: Wed, 5 Oct 2005 10:22:30 -0400 (EDT) Subject: Core includes Dialup?!? In-Reply-To: <4343DD34.7040001@COAT.COM> References: <1128520503.4108.1.camel@brilong-lnx.cisco.com> <4343DD34.7040001@COAT.COM> Message-ID: <33839.192.80.55.74.1128522150.squirrel@webmail.half-asleep.com> There are a lot of inclusions that I find to be completely outdated, or just plain unnecessary. Regardless of how many people may or may not need them, these kinds of things should be much more modular than they are. When I think CORE, I think the most basic packages required to make the system run. Everything else should be elsewhere. -Dan On Wed, October 5, 2005 10:03 am, Matt Fahrner said: > I agree its inclusion is dubious, but I do know a lot of people who have > workstations, not just laptops, that use PPP etc... > > - Matt > > Brian Long wrote: >> What is the Red Hat reason for including the Dialup comps group in Core? >> This means any basic install has crap like wvdial, ppp, etc. I >> explicitly have to exclude a lot of these packages that shouldn't be on >> any profiles except mobile laptops. >> >> /Brian/ > > -- > > --------------------------------------------------------------------- > Matt Fahrner 2 South Park St. > Manager of Networking Willis House > Burlington Coat Factory Warehouse Lebanon, N.H. 03766 > TEL: (603) 448-4100 x5150 USA > FAX: (603) 443-6190 Matt.Fahrner at COAT.COM > --------------------------------------------------------------------- > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > From mshuler at rackspace.com Wed Oct 5 14:25:20 2005 From: mshuler at rackspace.com (Michael Shuler) Date: Wed, 05 Oct 2005 09:25:20 -0500 Subject: Core includes Dialup?!? In-Reply-To: <1128520503.4108.1.camel@brilong-lnx.cisco.com> References: <1128520503.4108.1.camel@brilong-lnx.cisco.com> Message-ID: <4343E250.3060001@rackspace.com> Brian Long wrote: > What is the Red Hat reason for including the Dialup comps group in Core? > This means any basic install has crap like wvdial, ppp, etc. I > explicitly have to exclude a lot of these packages that shouldn't be on > any profiles except mobile laptops. Just to clarify my agreement - ppp, etc., is only one example. A "Core" install, in my opinion, should be the bare minimum for a networked server with remote SSH access. That's it. We maintain minimal kickstarts with around a 500M footprint, and it does require a large amount of hacking to get it down to a proper core of needed packages, in spite of the "mandatory" and "required" dictations. Kind Regards, Michael From brilong at cisco.com Wed Oct 5 15:12:02 2005 From: brilong at cisco.com (Brian Long) Date: Wed, 05 Oct 2005 11:12:02 -0400 Subject: Core includes Dialup?!? In-Reply-To: <4343E250.3060001@rackspace.com> References: <1128520503.4108.1.camel@brilong-lnx.cisco.com> <4343E250.3060001@rackspace.com> Message-ID: <1128525122.4108.7.camel@brilong-lnx.cisco.com> On Wed, 2005-10-05 at 09:25 -0500, Michael Shuler wrote: > Brian Long wrote: > > What is the Red Hat reason for including the Dialup comps group in Core? > > This means any basic install has crap like wvdial, ppp, etc. I > > explicitly have to exclude a lot of these packages that shouldn't be on > > any profiles except mobile laptops. > > Just to clarify my agreement - ppp, etc., is only one example. A "Core" > install, in my opinion, should be the bare minimum for a networked > server with remote SSH access. That's it. > > We maintain minimal kickstarts with around a 500M footprint, and it does > require a large amount of hacking to get it down to a proper core of > needed packages, in spite of the "mandatory" and "required" dictations. Yup, Red Hat includes redhat-lsb, which requires a ton of non-essential stuff, to ensure LSB compatibility out-of-the-box. However, I don't believe redhat-lsb requires ppp, pppoe or wvdial. :) /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From tdiehl at rogueind.com Wed Oct 5 18:38:02 2005 From: tdiehl at rogueind.com (Tom Diehl) Date: Wed, 5 Oct 2005 14:38:02 -0400 (EDT) Subject: Core includes Dialup?!? In-Reply-To: <1128525122.4108.7.camel@brilong-lnx.cisco.com> References: <1128520503.4108.1.camel@brilong-lnx.cisco.com> <4343E250.3060001@rackspace.com> <1128525122.4108.7.camel@brilong-lnx.cisco.com> Message-ID: On Wed, 5 Oct 2005, Brian Long wrote: > On Wed, 2005-10-05 at 09:25 -0500, Michael Shuler wrote: > > Brian Long wrote: > > > What is the Red Hat reason for including the Dialup comps group in Core? > > > This means any basic install has crap like wvdial, ppp, etc. I > > > explicitly have to exclude a lot of these packages that shouldn't be on > > > any profiles except mobile laptops. > > > > Just to clarify my agreement - ppp, etc., is only one example. A "Core" > > install, in my opinion, should be the bare minimum for a networked > > server with remote SSH access. That's it. > > > > We maintain minimal kickstarts with around a 500M footprint, and it does > > require a large amount of hacking to get it down to a proper core of > > needed packages, in spite of the "mandatory" and "required" dictations. > > Yup, Red Hat includes redhat-lsb, which requires a ton of non-essential > stuff, to ensure LSB compatibility out-of-the-box. However, I don't > believe redhat-lsb requires ppp, pppoe or wvdial. :) I think it is part of the "Red Hat Linux for dummies" install philosophy that has been around since the RHL 6.x days. IIRC the thought was that it was easier for people like you and me to take it out then it is to get a noob to put it in. IOW it cuts down on support calls. :-) Having said that I agree a base install is waaaaaaaaaaaaaaay to fat. Especially since the target audience is a corporate It department. Regards, Tom Diehl tdiehl at rogueind.com Spamtrap address mtd123 at rogueind.com From clumens at redhat.com Thu Oct 6 21:43:10 2005 From: clumens at redhat.com (Chris Lumens) Date: Thu, 6 Oct 2005 17:43:10 -0400 (EDT) Subject: new kickstart work Message-ID: As anyone who's been following Rawhide knows, kickstart has been undergoing a whole lot of work recently. This work has been a complete rewrite of the kickstart file parser and data representation in anaconda due to a couple realizations: (1) anaconda and s-c-kickstart use completely different code bases. The biggest problem here is that s-c-kickstart is falling farther and farther out of sync with anaconda as to what is supported and what's not. (2) There are some long-standing bugs in kickstart that cannot be fixed because the current code is sort of tough to work with. One example of this is post scripts in include files all get executed in the chroot environment, regardless of what you want. (3) We don't have any validation tools right now besides writing a kickstart file and cramming it through anaconda to see if it works or not. (4) Even within anaconda, there are multiple kickstart file parsers. (5) s-c-kickstart doesn't understand all the options that kickstart supports so if you take an existing file and run it through the program, what you get out will have lots of data stripped (if it even reads in the file at all). Okay, enough rationale. My main idea was to create a single new codebase that handles all the reading and writing of kickstart files and to make that code generic enough to where it's not a pain to adapt it to your uses. Therefore, I present the new pykickstart package in Rawhide. It's been in the works within anaconda for a couple weeks now but I have finally broken it out into its own package so you can use it for your own purposes. As always, the documentation is a bit light at the moment but that will be coming soon. The code is pretty well commented so it should be reasonably easy to follow, but here's an outline of the core: - KickstartData is a class containing a variety of attributes that represents the kickstart file. Wherever possible, things are represented in some real type instead of just a string. There are a variety of other Kickstart*Data classes found in data.py. These are meant to be put in the lists in KickstartData. - KickstartParser is the central class. It is essentially a state machine that transitions on seeing the various sections of the kickstart file. Functions are run upon state transitions, and these functions may be overridden in a subclass of KickstartParser if you need specialized behavior. The core function of readKickstart should be generic enough to never need overriding, unless you are doing something really crazy. If it's not flexible enough, this is probably a bug on my part. - KickstartHandlers deal with all the commands found in the command section of your kickstart file. Each command has its own handling function that deals with the legal options for the command and sets values in KickstartData appropriately. Again, you may subclass KickstartHandlers to do whatever you would like. If you want to make your own subclass handlers, it is a good idea to call the superclass handler first to set KickstartData. Then, do whatever you would like. - We are using all sorts of features from OptionParser, which I am not going to go into here. However, one of the nice features is the ability to mark options as deprecated. - KickstartWriter takes a KickstartData object and spits out a file. It similarly has handlers for each command, though I haven't done nearly as much with it as I have with the front end. So real quick: how would you go about using all this stuff? I'm glad you asked. from pykickstart.parser import * from pykickstart.data import * ksdata = KickstartData() kshandlers = KickstartHandlers(ksdata) ksparser = KickstartParser(ksdata, kshandlers) ksparser.readKickstart("ks.cfg") There you have it. You've now got ksdata populated and ready for use. Oh, you might also want to handle the various exceptions that can be thrown. How would you go about extending pykickstart for your own purposes? I'm going to save that for another long email. Or, you can look at kickstart.py in anaconda and see what's going on. It's really not too difficult. I think this is going to result in being able to make a lot of new stuff around kickstart pretty easily. My first goal is getting s-c-kickstart fixed up to use pykickstart (ugh, this is lots of work). My second goal is making a ksvalidator. That should be really simple. Oh, and a warning. This is all still in a lot of flux and before FC5, I might totally rewrite parts of the API. Nothing is stable yet and shouldn't be taken as such. Feel free to experiment with it, but understand I might get a crazy idea and redo something. So, thoughts? - Chris From alexander_rau at yahoo.com Fri Oct 7 03:03:48 2005 From: alexander_rau at yahoo.com (Alexander Rau) Date: Thu, 6 Oct 2005 20:03:48 -0700 (PDT) Subject: RHEL 4 DVD ? Message-ID: <20051007030348.81985.qmail@web52113.mail.yahoo.com> The DVD is working great. However when browsing the finished DVD the file names are all in the 8.3 file format. Any ideas how I can change this? I tried the -J option with mkisofs but it gives me an error message (soemthing about a doublicate boot.cat, I think). Thanks AR From sjbrowne at bluebottle.com Fri Oct 7 03:17:34 2005 From: sjbrowne at bluebottle.com (Stuart J. Browne) Date: Fri, 7 Oct 2005 13:17:34 +1000 Subject: RHEL 4 DVD ? References: <20051007030348.81985.qmail@web52113.mail.yahoo.com> Message-ID: <06eb01c5caed$aa728920$6501a8c0@bluebottle.com> You might want to try -R (RockRidge/Unix), not -J (Juliet/MS Windows), and possibly add a -T (Translation Table). ----- Original Message ----- From: "Alexander Rau" To: Sent: Friday, October 07, 2005 1:03 PM Subject: Re: RHEL 4 DVD ? > The DVD is working great. However when browsing the > finished DVD the file names are all in the 8.3 file > format. Any ideas how I can change this? I tried the > -J option with mkisofs but it gives me an error > message (soemthing about a doublicate boot.cat, I > think). > > Thanks > > AR > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From clumens at redhat.com Fri Oct 7 13:53:48 2005 From: clumens at redhat.com (Chris Lumens) Date: Fri, 7 Oct 2005 09:53:48 -0400 (EDT) Subject: new kickstart work In-Reply-To: <1128635979.30435.4.camel@ignacio.lan> References: <1128635979.30435.4.camel@ignacio.lan> Message-ID: >> from pykickstart.parser import * >> from pykickstart.data import * >> >> ksdata = KickstartData() >> kshandlers = KickstartHandlers(ksdata) >> ksparser = KickstartParser(ksdata, kshandlers) >> ksparser.readKickstart("ks.cfg") > > Do we care about ksdata and kshandlers after the fact? If you care about the data in the kickstart file, then you do care about ksdata. It will be populated as an effect of the call to readKickstart. I can't think of a good reason off-hand why you would care about kshandlers, though. - Chris From clumens at redhat.com Thu Oct 6 21:43:10 2005 From: clumens at redhat.com (Chris Lumens) Date: Thu, 6 Oct 2005 17:43:10 -0400 (EDT) Subject: new kickstart work Message-ID: As anyone who's been following Rawhide knows, kickstart has been undergoing a whole lot of work recently. This work has been a complete rewrite of the kickstart file parser and data representation in anaconda due to a couple realizations: (1) anaconda and s-c-kickstart use completely different code bases. The biggest problem here is that s-c-kickstart is falling farther and farther out of sync with anaconda as to what is supported and what's not. (2) There are some long-standing bugs in kickstart that cannot be fixed because the current code is sort of tough to work with. One example of this is post scripts in include files all get executed in the chroot environment, regardless of what you want. (3) We don't have any validation tools right now besides writing a kickstart file and cramming it through anaconda to see if it works or not. (4) Even within anaconda, there are multiple kickstart file parsers. (5) s-c-kickstart doesn't understand all the options that kickstart supports so if you take an existing file and run it through the program, what you get out will have lots of data stripped (if it even reads in the file at all). Okay, enough rationale. My main idea was to create a single new codebase that handles all the reading and writing of kickstart files and to make that code generic enough to where it's not a pain to adapt it to your uses. Therefore, I present the new pykickstart package in Rawhide. It's been in the works within anaconda for a couple weeks now but I have finally broken it out into its own package so you can use it for your own purposes. As always, the documentation is a bit light at the moment but that will be coming soon. The code is pretty well commented so it should be reasonably easy to follow, but here's an outline of the core: - KickstartData is a class containing a variety of attributes that represents the kickstart file. Wherever possible, things are represented in some real type instead of just a string. There are a variety of other Kickstart*Data classes found in data.py. These are meant to be put in the lists in KickstartData. - KickstartParser is the central class. It is essentially a state machine that transitions on seeing the various sections of the kickstart file. Functions are run upon state transitions, and these functions may be overridden in a subclass of KickstartParser if you need specialized behavior. The core function of readKickstart should be generic enough to never need overriding, unless you are doing something really crazy. If it's not flexible enough, this is probably a bug on my part. - KickstartHandlers deal with all the commands found in the command section of your kickstart file. Each command has its own handling function that deals with the legal options for the command and sets values in KickstartData appropriately. Again, you may subclass KickstartHandlers to do whatever you would like. If you want to make your own subclass handlers, it is a good idea to call the superclass handler first to set KickstartData. Then, do whatever you would like. - We are using all sorts of features from OptionParser, which I am not going to go into here. However, one of the nice features is the ability to mark options as deprecated. - KickstartWriter takes a KickstartData object and spits out a file. It similarly has handlers for each command, though I haven't done nearly as much with it as I have with the front end. So real quick: how would you go about using all this stuff? I'm glad you asked. from pykickstart.parser import * from pykickstart.data import * ksdata = KickstartData() kshandlers = KickstartHandlers(ksdata) ksparser = KickstartParser(ksdata, kshandlers) ksparser.readKickstart("ks.cfg") There you have it. You've now got ksdata populated and ready for use. Oh, you might also want to handle the various exceptions that can be thrown. How would you go about extending pykickstart for your own purposes? I'm going to save that for another long email. Or, you can look at kickstart.py in anaconda and see what's going on. It's really not too difficult. I think this is going to result in being able to make a lot of new stuff around kickstart pretty easily. My first goal is getting s-c-kickstart fixed up to use pykickstart (ugh, this is lots of work). My second goal is making a ksvalidator. That should be really simple. Oh, and a warning. This is all still in a lot of flux and before FC5, I might totally rewrite parts of the API. Nothing is stable yet and shouldn't be taken as such. Feel free to experiment with it, but understand I might get a crazy idea and redo something. So, thoughts? - Chris From Elizabeth.Brosch at thomson.com Fri Oct 7 19:30:31 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Fri, 7 Oct 2005 15:30:31 -0400 Subject: Kickstart fails Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4DA6@tshuspaphimbx01.ERF.THOMSON.COM> I have been trying to get kickstart working but have had no success. Here are the last few lines of the kickstart process before it dies: *getting kickstart file *sending dhcp request through device eth0 *waiting for link * 5 seconds * running dhcp for eth0 * pump told us: No DHCP reply received * eth0 isn't a wireless adaptor I have checked every possible configuration with regards to dhcp, pxe, tftp, but cannot find anything wrong. Has anyone else seen this type of behavior? Thanks in advance. Liz Brosch TSH System Services Philadelphia, PA 19104 office: (215) 386-0100 x1144 cell: (267) 784-9166 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Fri Oct 7 19:43:03 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 07 Oct 2005 12:43:03 -0700 Subject: Kickstart fails Message-ID: Will this server get a dhcp address if you boot from the boot.iso (or disk 1 with askmethod) and select http, ftp or nfs as your install path? If so, you are probably running into a switch timing issue that I have to deal with every day. Try adding an ethtool option to the end of your command line: if 100MB Full Duplex: ksdevice=eth0 eth0_ethtool="autoneg=off speed=100 duplex=full" If 1000MB Full: ksdevice=eth0 eth0_ethtool="autoneg=on speed=1000 duplex=full" Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 12:31 PM To: kickstart-list at redhat.com Subject: Kickstart fails I have been trying to get kickstart working but have had no success. Here are the last few lines of the kickstart process before it dies: *getting kickstart file *sending dhcp request through device eth0 *waiting for link * 5 seconds * running dhcp for eth0 * pump told us: No DHCP reply received * eth0 isn't a wireless adaptor I have checked every possible configuration with regards to dhcp, pxe, tftp, but cannot find anything wrong. Has anyone else seen this type of behavior? Thanks in advance. Liz Brosch TSH System Services Philadelphia, PA 19104 office: (215) 386-0100 x1144 cell: (267) 784-9166 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Elizabeth.Brosch at thomson.com Fri Oct 7 19:56:24 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Fri, 7 Oct 2005 15:56:24 -0400 Subject: Kickstart fails Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4DA9@tshuspaphimbx01.ERF.THOMSON.COM> Thanks for the reply Chip but it didn't work. I received the same error. Liz ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip Sent: Friday, October 07, 2005 3:43 PM To: Discussion list about Kickstart Subject: RE: Kickstart fails Will this server get a dhcp address if you boot from the boot.iso (or disk 1 with askmethod) and select http, ftp or nfs as your install path? If so, you are probably running into a switch timing issue that I have to deal with every day. Try adding an ethtool option to the end of your command line: if 100MB Full Duplex: ksdevice=eth0 eth0_ethtool="autoneg=off speed=100 duplex=full" If 1000MB Full: ksdevice=eth0 eth0_ethtool="autoneg=on speed=1000 duplex=full" Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 12:31 PM To: kickstart-list at redhat.com Subject: Kickstart fails I have been trying to get kickstart working but have had no success. Here are the last few lines of the kickstart process before it dies: *getting kickstart file *sending dhcp request through device eth0 *waiting for link * 5 seconds * running dhcp for eth0 * pump told us: No DHCP reply received * eth0 isn't a wireless adaptor I have checked every possible configuration with regards to dhcp, pxe, tftp, but cannot find anything wrong. Has anyone else seen this type of behavior? Thanks in advance. Liz Brosch TSH System Services Philadelphia, PA 19104 office: (215) 386-0100 x1144 cell: (267) 784-9166 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Fri Oct 7 19:58:31 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 07 Oct 2005 12:58:31 -0700 Subject: Kickstart fails Message-ID: Do you get an address when you boot from disk 1 or the boot.iso? ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 12:56 PM To: kickstart-list at redhat.com Subject: RE: Kickstart fails Thanks for the reply Chip but it didn't work. I received the same error. Liz ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip Sent: Friday, October 07, 2005 3:43 PM To: Discussion list about Kickstart Subject: RE: Kickstart fails Will this server get a dhcp address if you boot from the boot.iso (or disk 1 with askmethod) and select http, ftp or nfs as your install path? If so, you are probably running into a switch timing issue that I have to deal with every day. Try adding an ethtool option to the end of your command line: if 100MB Full Duplex: ksdevice=eth0 eth0_ethtool="autoneg=off speed=100 duplex=full" If 1000MB Full: ksdevice=eth0 eth0_ethtool="autoneg=on speed=1000 duplex=full" Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 12:31 PM To: kickstart-list at redhat.com Subject: Kickstart fails I have been trying to get kickstart working but have had no success. Here are the last few lines of the kickstart process before it dies: *getting kickstart file *sending dhcp request through device eth0 *waiting for link * 5 seconds * running dhcp for eth0 * pump told us: No DHCP reply received * eth0 isn't a wireless adaptor I have checked every possible configuration with regards to dhcp, pxe, tftp, but cannot find anything wrong. Has anyone else seen this type of behavior? Thanks in advance. Liz Brosch TSH System Services Philadelphia, PA 19104 office: (215) 386-0100 x1144 cell: (267) 784-9166 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Elizabeth.Brosch at thomson.com Fri Oct 7 20:19:05 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Fri, 7 Oct 2005 16:19:05 -0400 Subject: Kickstart fails Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4DAA@tshuspaphimbx01.ERF.THOMSON.COM> The server receives an address during PXE: output from syslog: Oct 7 16:10:24 edcappsrhat7 dhcpd: DHCPDISCOVER from 00:14:c2:3c:a9:d2 via eth0 Oct 7 16:10:24 edcappsrhat7 dhcpd: DHCPOFFER on 10.xxx.xxx.xxx to 00:14:c2:3c:a9:d2 via eth0 Oct 7 16:10:28 edcappsrhat7 dhcpd: DHCPREQUEST for 10.xxx.xxx.xxx (10.xxx.xxx.xxx) from 00:14:c2:3c:a9:d2 via eth0 Oct 7 16:10:28 edcappsrhat7 dhcpd: DHCPACK on 10.xxx.xxx.xxx to 00:14:c2:3c:a9:d2 via eth0 Oct 7 20:10:28 edcappsrhat7 in.tftpd[9102]: RRQ from 10.xxx.xxx.xxx filename pxelinux.0 Oct 7 20:10:28 edcappsrhat7 in.tftpd[9102]: tftp: client does not accept options Oct 7 20:10:28 edcappsrhat7 in.tftpd[9103]: RRQ from 10.xxx.xxx.xxx filename pxelinux.0 Oct 7 20:10:28 edcappsrhat7 in.tftpd[9104]: RRQ from 10.xxx.xxx.xxx filename pxelinux.cfg/01-00-14-c2-3c-a9-d2 Oct 7 20:10:28 edcappsrhat7 in.tftpd[9104]: sending NAK (1, File not found) to 10.xxx.xxx.xxx Oct 7 20:10:28 edcappsrhat7 in.tftpd[9105]: RRQ from 10.xxx.xxx.xxx filename pxelinux.cfg/0xxxxxx Oct 7 20:12:03 edcappsrhat7 in.tftpd[9224]: RRQ from 10.xxx.xxx.xxx filename vmlinuz Oct 7 20:12:03 edcappsrhat7 in.tftpd[9225]: RRQ from 10.xxx.xxx.xxx filename initrd.img Then it starts the process and dies with the message I posted earlier. It just hangs at the: Welcome to Red Hat Enterprise Linux Configure TCP/IP screen Thanks. Liz Brosch TSH System Services Philadelphia, PA 19104 office: (215) 386-0100 x1144 cell: (267) 784-9166 ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip Sent: Friday, October 07, 2005 3:59 PM To: Discussion list about Kickstart Subject: RE: Kickstart fails Do you get an address when you boot from disk 1 or the boot.iso? ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 12:56 PM To: kickstart-list at redhat.com Subject: RE: Kickstart fails Thanks for the reply Chip but it didn't work. I received the same error. Liz ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip Sent: Friday, October 07, 2005 3:43 PM To: Discussion list about Kickstart Subject: RE: Kickstart fails Will this server get a dhcp address if you boot from the boot.iso (or disk 1 with askmethod) and select http, ftp or nfs as your install path? If so, you are probably running into a switch timing issue that I have to deal with every day. Try adding an ethtool option to the end of your command line: if 100MB Full Duplex: ksdevice=eth0 eth0_ethtool="autoneg=off speed=100 duplex=full" If 1000MB Full: ksdevice=eth0 eth0_ethtool="autoneg=on speed=1000 duplex=full" Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 12:31 PM To: kickstart-list at redhat.com Subject: Kickstart fails I have been trying to get kickstart working but have had no success. Here are the last few lines of the kickstart process before it dies: *getting kickstart file *sending dhcp request through device eth0 *waiting for link * 5 seconds * running dhcp for eth0 * pump told us: No DHCP reply received * eth0 isn't a wireless adaptor I have checked every possible configuration with regards to dhcp, pxe, tftp, but cannot find anything wrong. Has anyone else seen this type of behavior? Thanks in advance. Liz Brosch TSH System Services Philadelphia, PA 19104 office: (215) 386-0100 x1144 cell: (267) 784-9166 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image002.jpg Type: image/jpeg Size: 8564 bytes Desc: image002.jpg URL: From Chip.Shabazian at bankofamerica.com Fri Oct 7 20:34:20 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 07 Oct 2005 13:34:20 -0700 Subject: Kickstart fails Message-ID: Ok, you are getting your initial pxeboot IP, it successfully sends the kernel and initrd via tftp. At this point, the NIC recycles itself, and the negotiation between NIC and switch starts all over. This looks so far like the exact problem we experience, with the same HP DL servers. What does your pxeboot config file contain? Oct 7 20:10:28 edcappsrhat7 in.tftpd[9105]: RRQ from 10.xxx.xxx.xxx filename pxelinux.cfg/0xxxxxx That is where you need to put the ethtool option if you are pxebooting. Can you boot the server with linux rescue, and see what kind of link you have (100,1000,half/full duplex) Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 1:19 PM To: kickstart-list at redhat.com Subject: RE: Kickstart fails The server receives an address during PXE: output from syslog: Oct 7 16:10:24 edcappsrhat7 dhcpd: DHCPDISCOVER from 00:14:c2:3c:a9:d2 via eth0 Oct 7 16:10:24 edcappsrhat7 dhcpd: DHCPOFFER on 10.xxx.xxx.xxx to 00:14:c2:3c:a9:d2 via eth0 Oct 7 16:10:28 edcappsrhat7 dhcpd: DHCPREQUEST for 10.xxx.xxx.xxx (10.xxx.xxx.xxx) from 00:14:c2:3c:a9:d2 via eth0 Oct 7 16:10:28 edcappsrhat7 dhcpd: DHCPACK on 10.xxx.xxx.xxx to 00:14:c2:3c:a9:d2 via eth0 Oct 7 20:10:28 edcappsrhat7 in.tftpd[9102]: RRQ from 10.xxx.xxx.xxx filename pxelinux.0 Oct 7 20:10:28 edcappsrhat7 in.tftpd[9102]: tftp: client does not accept options Oct 7 20:10:28 edcappsrhat7 in.tftpd[9103]: RRQ from 10.xxx.xxx.xxx filename pxelinux.0 Oct 7 20:10:28 edcappsrhat7 in.tftpd[9104]: RRQ from 10.xxx.xxx.xxx filename pxelinux.cfg/01-00-14-c2-3c-a9-d2 Oct 7 20:10:28 edcappsrhat7 in.tftpd[9104]: sending NAK (1, File not found) to 10.xxx.xxx.xxx Oct 7 20:10:28 edcappsrhat7 in.tftpd[9105]: RRQ from 10.xxx.xxx.xxx filename pxelinux.cfg/0xxxxxx Oct 7 20:12:03 edcappsrhat7 in.tftpd[9224]: RRQ from 10.xxx.xxx.xxx filename vmlinuz Oct 7 20:12:03 edcappsrhat7 in.tftpd[9225]: RRQ from 10.xxx.xxx.xxx filename initrd.img Then it starts the process and dies with the message I posted earlier. It just hangs at the: Welcome to Red Hat Enterprise Linux Configure TCP/IP screen Thanks. Liz Brosch TSH System Services Philadelphia, PA 19104 office: (215) 386-0100 x1144 cell: (267) 784-9166 ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip Sent: Friday, October 07, 2005 3:59 PM To: Discussion list about Kickstart Subject: RE: Kickstart fails Do you get an address when you boot from disk 1 or the boot.iso? ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 12:56 PM To: kickstart-list at redhat.com Subject: RE: Kickstart fails Thanks for the reply Chip but it didn't work. I received the same error. Liz ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip Sent: Friday, October 07, 2005 3:43 PM To: Discussion list about Kickstart Subject: RE: Kickstart fails Will this server get a dhcp address if you boot from the boot.iso (or disk 1 with askmethod) and select http, ftp or nfs as your install path? If so, you are probably running into a switch timing issue that I have to deal with every day. Try adding an ethtool option to the end of your command line: if 100MB Full Duplex: ksdevice=eth0 eth0_ethtool="autoneg=off speed=100 duplex=full" If 1000MB Full: ksdevice=eth0 eth0_ethtool="autoneg=on speed=1000 duplex=full" Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 12:31 PM To: kickstart-list at redhat.com Subject: Kickstart fails I have been trying to get kickstart working but have had no success. Here are the last few lines of the kickstart process before it dies: *getting kickstart file *sending dhcp request through device eth0 *waiting for link * 5 seconds * running dhcp for eth0 * pump told us: No DHCP reply received * eth0 isn't a wireless adaptor I have checked every possible configuration with regards to dhcp, pxe, tftp, but cannot find anything wrong. Has anyone else seen this type of behavior? Thanks in advance. Liz Brosch TSH System Services Philadelphia, PA 19104 office: (215) 386-0100 x1144 cell: (267) 784-9166 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image002.jpg Type: image/jpeg Size: 8564 bytes Desc: image002.jpg URL: From Elizabeth.Brosch at thomson.com Fri Oct 7 20:44:24 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Fri, 7 Oct 2005 16:44:24 -0400 Subject: Kickstart fails Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4DAC@tshuspaphimbx01.ERF.THOMSON.COM> Chip: Here is the contents of my pxeboot config file: default linux prompt 1 label linux kernel vmlinuz append ksdevice=eth0 eth0_ethtool="autoneg=on speed=1000 duplex=full" load_ramdisk=1 initrd=initrd.img network ks=nfs:10.xxx.xxx.xxx:/tftpboot/ks.cfg ip=dhcp Sorry but how can I boot this client into linux rescue? Liz -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Fri Oct 7 20:59:24 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 07 Oct 2005 13:59:24 -0700 Subject: Kickstart fails Message-ID: Boot from disk 1, and at the boot: prompt, type linux rescue Here is my (modified to remove host and domain names) pxeboot config file: default hostname label hostname kernel RHELAS3U1/vmlinuz append initrd=RHELAS3U1/initrd.img nofb ramdisk_size=10000 ks=nfs:kickstart_server.domainname.com:/kickstart/configs/hostname ksdevice=eth0 eth0_ethtool="autoneg=on speed=1000 duplex=full" So try this for yours: default linux label linux kernel vmlinuz append initrd=initrd.img nofb ramdisk_size=10000 ks=nfs:10.xxx.xxx.xxx:/tftpboot/ks.cfg ksdevice=eth0 eth0_ethtool="autoneg=on speed=1000 duplex=full" Based on the fact that there are no paths in front of your kernel and initrd, I assume that they are in the root of your tftpboot directory, or the same place as pxelinux.0 ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 1:44 PM To: kickstart-list at redhat.com Subject: RE: Kickstart fails Chip: Here is the contents of my pxeboot config file: default linux prompt 1 label linux kernel vmlinuz append ksdevice=eth0 eth0_ethtool="autoneg=on speed=1000 duplex=full" load_ramdisk=1 initrd=initrd.img network ks=nfs:10.xxx.xxx.xxx:/tftpboot/ks.cfg ip=dhcp Sorry but how can I boot this client into linux rescue? Liz -------------- next part -------------- An HTML attachment was scrubbed... URL: From Elizabeth.Brosch at thomson.com Fri Oct 7 21:11:44 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Fri, 7 Oct 2005 17:11:44 -0400 Subject: Kickstart fails Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4DB1@tshuspaphimbx01.ERF.THOMSON.COM> Chip: I see that there was a discussion on this list regarding my same problem. It is from the 2005-May Archive and the thread is "Network Kickstart Weirdness". I have already checked with my network team and I have portfast enabled so that shouldn't be the issue. The only thing left is a possible bug with Anaconda. Have you ever rebuilt that? -------------- next part -------------- An HTML attachment was scrubbed... URL: From dgrubb3 at comcast.net Fri Oct 7 21:21:35 2005 From: dgrubb3 at comcast.net (dgrubb3 at comcast.net) Date: Fri, 07 Oct 2005 21:21:35 +0000 Subject: Kickstart fails Message-ID: <100720052121.22379.4346E6DF0003F1BE0000576B2205886014CC0D0D9A9D090B@comcast.net> Liz, If you've got a network switch (CISCO 37xx, 65xx, etc) you may want to check that the spantree portfast is enabled for the port to which you are connected. I've had lots of trouble with this in the past. The spanning tree path calculation in the switch takes up to 50 seconds to make itself happy. The installer does not have that much patience. Enabling portfast reduces this time considerably. Dave > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list -------------- next part -------------- An embedded message was scrubbed... From: Subject: Kickstart fails Date: Fri, 7 Oct 2005 19:33:13 +0000 Size: 7733 URL: From Elizabeth.Brosch at thomson.com Fri Oct 7 21:29:30 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Fri, 7 Oct 2005 17:29:30 -0400 Subject: Kickstart fails Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4DB3@tshuspaphimbx01.ERF.THOMSON.COM> Thanks for reply. I confirmed with network team and they said that portfast was enabled. I think I'm down to two options: Anaconda isn't using my eth0 nic (at present the only one connected) or rebuilding Anaconda to hard code a 30 second sleep. I see some information about that at http://www.redhat.com/archives/kickstart-list/2005-March/msg00087.html. Kickstart is becoming very difficult! From Chip.Shabazian at bankofamerica.com Fri Oct 7 21:38:25 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 07 Oct 2005 14:38:25 -0700 Subject: Kickstart fails Message-ID: Boot into rescue, and ensure that you can get an IP address, then run ethtool to see what the link is. We fix this problem (which occures globally for us) with the ethtool statement. I've never seen a situation where that HASN'T worked when it's due to the portfast issue. Chip -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 2:30 PM To: kickstart-list at redhat.com Subject: RE: Kickstart fails Thanks for reply. I confirmed with network team and they said that portfast was enabled. I think I'm down to two options: Anaconda isn't using my eth0 nic (at present the only one connected) or rebuilding Anaconda to hard code a 30 second sleep. I see some information about that at http://www.redhat.com/archives/kickstart-list/2005-March/msg00087.html. Kickstart is becoming very difficult! _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From kernel at linuxace.com Fri Oct 7 21:39:04 2005 From: kernel at linuxace.com (Phil Oester) Date: Fri, 7 Oct 2005 14:39:04 -0700 Subject: Kickstart fails In-Reply-To: <2213CAA79DA9094AB52517FDADEF07FA021E4DB3@tshuspaphimbx01.ERF.THOMSON.COM> References: <2213CAA79DA9094AB52517FDADEF07FA021E4DB3@tshuspaphimbx01.ERF.THOMSON.COM> Message-ID: <20051007213904.GA6100@linuxace.com> On Fri, Oct 07, 2005 at 05:29:30PM -0400, Elizabeth.Brosch at thomson.com wrote: > Thanks for reply. I confirmed with network team and they said that > portfast was enabled. I think I'm down to two options: > Anaconda isn't using my eth0 nic (at present the only one connected) Provide output from 'lspci | grep Eth', as well as which nic you would like to be eth0. Phil From Elizabeth.Brosch at thomson.com Fri Oct 7 21:53:04 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Fri, 7 Oct 2005 17:53:04 -0400 Subject: Kickstart fails Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4DB7@tshuspaphimbx01.ERF.THOMSON.COM> Chip - what command do I use to see what the NIC is configured at? From Elizabeth.Brosch at thomson.com Fri Oct 7 21:54:38 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Fri, 7 Oct 2005 17:54:38 -0400 Subject: Kickstart fails Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4DB8@tshuspaphimbx01.ERF.THOMSON.COM> Interesting enough, I do obtain an IP via dhcp when running linux rescue. When I display ifconfig it is eth3. The output from syslog on my dhcp server shows that it is eth0! From Chip.Shabazian at bankofamerica.com Fri Oct 7 21:56:04 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 07 Oct 2005 14:56:04 -0700 Subject: Kickstart fails Message-ID: ethtool eth0 This will give you output that looks like this: Settings for eth0: Supported ports: [ MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Advertised auto-negotiation: Yes Speed: 100Mb/s Duplex: Half Port: Twisted Pair PHYAD: 1 Transceiver: internal Auto-negotiation: on Supports Wake-on: g Wake-on: d Current message level: 0x000000ff (255) Link detected: yes -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 2:53 PM To: kickstart-list at redhat.com Subject: RE: Kickstart fails Chip - what command do I use to see what the NIC is configured at? _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Elizabeth.Brosch at thomson.com Fri Oct 7 21:57:57 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Fri, 7 Oct 2005 17:57:57 -0400 Subject: Kickstart fails Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4DB9@tshuspaphimbx01.ERF.THOMSON.COM> ethtool is not found. I'm running this on the client I just booted into linux rescue. From Chip.Shabazian at bankofamerica.com Fri Oct 7 22:02:40 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 07 Oct 2005 15:02:40 -0700 Subject: Kickstart fails Message-ID: Ok, looks like you are running into a bus enumeration problem (Installing RHEL 4?). Also, you obviously you have another NIC installed in the server besides the two built in. You can change the pxeboot config and your kickstart file to use eth3, that should fix the switch issue. I don't know if you can use a hardware= statement to tell the server WHICH NIC is eth0, I haven't ever tried that. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 2:55 PM To: kickstart-list at redhat.com Subject: RE: Kickstart fails Interesting enough, I do obtain an IP via dhcp when running linux rescue. When I display ifconfig it is eth3. The output from syslog on my dhcp server shows that it is eth0! _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Elizabeth.Brosch at thomson.com Fri Oct 7 22:04:38 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Fri, 7 Oct 2005 18:04:38 -0400 Subject: Kickstart fails Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4DBA@tshuspaphimbx01.ERF.THOMSON.COM> I just changed the pxeconfig file - ksdevice eth3. I finally received an IP addr and a mountd request! But then the installer goes looking for eth0 network info and fails there. From Chip.Shabazian at bankofamerica.com Fri Oct 7 22:05:56 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 07 Oct 2005 15:05:56 -0700 Subject: Kickstart fails Message-ID: You need to change it in your kickstart ks.cfg file as well -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 3:05 PM To: kickstart-list at redhat.com Subject: RE: Kickstart fails I just changed the pxeconfig file - ksdevice eth3. I finally received an IP addr and a mountd request! But then the installer goes looking for eth0 network info and fails there. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Elizabeth.Brosch at thomson.com Fri Oct 7 22:07:28 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Fri, 7 Oct 2005 18:07:28 -0400 Subject: Kickstart fails Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4DBB@tshuspaphimbx01.ERF.THOMSON.COM> I just caught that and am giving it another shot. I can't thank you enough for taking the time to contact me. I have been at these for the last couple of weeks. I'll let you know how it works out. Again, thanks From Elizabeth.Brosch at thomson.com Fri Oct 7 22:22:39 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Fri, 7 Oct 2005 18:22:39 -0400 Subject: Kickstart fails Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4DBC@tshuspaphimbx01.ERF.THOMSON.COM> Success!!!!!!!! I have installed my client! From Chip.Shabazian at bankofamerica.com Fri Oct 7 22:25:50 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 07 Oct 2005 15:25:50 -0700 Subject: Kickstart fails Message-ID: Congratulations ;-D -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Friday, October 07, 2005 3:23 PM To: kickstart-list at redhat.com Subject: RE: Kickstart fails Success!!!!!!!! I have installed my client! _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From jkeating at j2solutions.net Fri Oct 7 22:30:35 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Fri, 07 Oct 2005 15:30:35 -0700 Subject: new kickstart work In-Reply-To: References: Message-ID: <1128724235.7991.95.camel@prometheus.gamehouse.com> On Thu, 2005-10-06 at 17:43 -0400, Chris Lumens wrote: > Oh, and a warning. This is all still in a lot of flux and before FC5, I > might totally rewrite parts of the API. Nothing is stable yet and > shouldn't be taken as such. Feel free to experiment with it, but > understand I might get a crazy idea and redo something. Are there any planned changes to the syntax of ks.cfg files? I may have to go and do a bunch of contract work to port files I created at another company to the new format..... (: -- Jesse Keating RHCE (http://geek.j2solutions.net) Fedora Legacy Team (http://www.fedoralegacy.org) GPG Public Key (http://geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From tdiehl at rogueind.com Sat Oct 8 04:16:20 2005 From: tdiehl at rogueind.com (Tom Diehl) Date: Sat, 8 Oct 2005 00:16:20 -0400 (EDT) Subject: Kickstart fails In-Reply-To: <2213CAA79DA9094AB52517FDADEF07FA021E4DBA@tshuspaphimbx01.ERF.THOMSON.COM> References: <2213CAA79DA9094AB52517FDADEF07FA021E4DBA@tshuspaphimbx01.ERF.THOMSON.COM> Message-ID: On Fri, 7 Oct 2005 Elizabeth.Brosch at thomson.com wrote: > I just changed the pxeconfig file - ksdevice eth3. I finally received > an IP addr and a mountd request! But then the installer goes looking > for eth0 network info and fails there. Save yourself a bunch pf pain in the future use ksdevice=link in the pxe config file or on the command line. That way it will use whatever nic has a link. Worst case is you boot once and the device in the ks.cfg file is wrong but anaconda will tell you that, so you can fix the cfg file. HTH, Tom Diehl tdiehl at rogueind.com Spamtrap address mtd123 at rogueind.com From brilong at cisco.com Mon Oct 10 14:14:14 2005 From: brilong at cisco.com (Brian Long) Date: Mon, 10 Oct 2005 10:14:14 -0400 Subject: Kickstart fails In-Reply-To: References: <2213CAA79DA9094AB52517FDADEF07FA021E4DBA@tshuspaphimbx01.ERF.THOMSON.COM> Message-ID: <1128953654.3969.13.camel@brilong-lnx> On Sat, 2005-10-08 at 00:16 -0400, Tom Diehl wrote: > On Fri, 7 Oct 2005 Elizabeth.Brosch at thomson.com wrote: > > > I just changed the pxeconfig file - ksdevice eth3. I finally received > > an IP addr and a mountd request! But then the installer goes looking > > for eth0 network info and fails there. > > Save yourself a bunch pf pain in the future use ksdevice=link in the pxe config > file or on the command line. That way it will use whatever nic has a link. > Worst case is you boot once and the device in the ks.cfg file is wrong but > anaconda will tell you that, so you can fix the cfg file. We've started using ksdevice= on our new PxeLinux-based installations. We have servers with public and private networks, so ksdevice=link will not always work (it might try to kickstart on a private network with no DHCP server, no router, etc). /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From clumens at redhat.com Mon Oct 10 14:23:57 2005 From: clumens at redhat.com (Chris Lumens) Date: Mon, 10 Oct 2005 10:23:57 -0400 (EDT) Subject: new kickstart work In-Reply-To: <1128724235.7991.95.camel@prometheus.gamehouse.com> References: <1128724235.7991.95.camel@prometheus.gamehouse.com> Message-ID: >> Oh, and a warning. This is all still in a lot of flux and before FC5, I >> might totally rewrite parts of the API. Nothing is stable yet and >> shouldn't be taken as such. Feel free to experiment with it, but >> understand I might get a crazy idea and redo something. > > Are there any planned changes to the syntax of ks.cfg files? I may have > to go and do a bunch of contract work to port files I created at another > company to the new format..... (: We've talked about this a little bit, but nothing serious. We haven't even gotten so far as to decide what sort of format it'd be. Too much other stuff has changed in Fedora for this release to even really thing about changing the ks format. - Chris From tdiehl at rogueind.com Mon Oct 10 16:21:02 2005 From: tdiehl at rogueind.com (Tom Diehl) Date: Mon, 10 Oct 2005 12:21:02 -0400 (EDT) Subject: Kickstart fails In-Reply-To: <1128953654.3969.13.camel@brilong-lnx> References: <2213CAA79DA9094AB52517FDADEF07FA021E4DBA@tshuspaphimbx01.ERF.THOMSON.COM> <1128953654.3969.13.camel@brilong-lnx> Message-ID: On Mon, 10 Oct 2005, Brian Long wrote: > On Sat, 2005-10-08 at 00:16 -0400, Tom Diehl wrote: > > On Fri, 7 Oct 2005 Elizabeth.Brosch at thomson.com wrote: > > > > > I just changed the pxeconfig file - ksdevice eth3. I finally received > > > an IP addr and a mountd request! But then the installer goes looking > > > for eth0 network info and fails there. > > > > Save yourself a bunch pf pain in the future use ksdevice=link in the pxe config > > file or on the command line. That way it will use whatever nic has a link. > > Worst case is you boot once and the device in the ks.cfg file is wrong but > > anaconda will tell you that, so you can fix the cfg file. > > We've started using ksdevice= on our new PxeLinux-based > installations. We have servers with public and private networks, so > ksdevice=link will not always work (it might try to kickstart on a > private network with no DHCP server, no router, etc). Thanks, I just learned something. I did not know it would take a mac address. :-) That makes it a good day!! Regards, Tom Diehl tdiehl at rogueind.com Spamtrap address mtd123 at rogueind.com From joe at eshu.net Wed Oct 12 00:51:26 2005 From: joe at eshu.net (Joe Christy) Date: Tue, 11 Oct 2005 17:51:26 -0700 Subject: Neccessary metadata for NFS kickstart of base+updates(+extras)? Message-ID: <434C5E0E.5070809@eshu.net> While I've found various (dated) descriptions of how to build iso's for FC4 w/ updates replacing their respective originals, I haven't found corresponding documentation for an NFS based kickstart. If I'm _not_ using iso's, just a Fedora/{base,RPMS} tree with latest update rpms, what meta-data do I need to generate so that I can do a kickstart install? Is hdlist still necessary, since there are, strictly speaking no install discs? ... and what If I want to supplement my install with some rpms from Extras? Do I need only add the packages into comps.xml and proceed as before? Any pointers would be welcome. -- ============================= Joe Christy ============================== ------------------ http://public.xdi.org/=joe.christy ------------------ == If I can save you any time, give it to me, I'll keep it with mine. == From kernel at linuxace.com Wed Oct 12 01:04:30 2005 From: kernel at linuxace.com (Phil Oester) Date: Tue, 11 Oct 2005 18:04:30 -0700 Subject: Neccessary metadata for NFS kickstart of base+updates(+extras)? In-Reply-To: <434C5E0E.5070809@eshu.net> References: <434C5E0E.5070809@eshu.net> Message-ID: <20051012010430.GA12653@linuxace.com> On Tue, Oct 11, 2005 at 05:51:26PM -0700, Joe Christy wrote: > While I've found various (dated) descriptions of how to build iso's for > FC4 w/ updates replacing their respective originals, I haven't found > corresponding documentation for an NFS based kickstart. > > If I'm _not_ using iso's, just a Fedora/{base,RPMS} tree with latest > update rpms, what meta-data do I need to generate so that I can do a > kickstart install? Is hdlist still necessary, since there are, strictly > speaking no install discs? A simple genhdlist: /usr/lib/anaconda-runtime/genhdlist --productpath Fedora /foo/kickstart will regenerate hdlist (yes, it is needed). > ... and what If I want to supplement my install with some rpms from > Extras? Do I need only add the packages into comps.xml and proceed as > before? Not familiar with this one Phil From tibbs at math.uh.edu Wed Oct 12 02:25:22 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Tue, 11 Oct 2005 21:25:22 -0500 Subject: Neccessary metadata for NFS kickstart of base+updates(+extras)? In-Reply-To: <434C5E0E.5070809@eshu.net> (Joe Christy's message of "Tue, 11 Oct 2005 17:51:26 -0700") References: <434C5E0E.5070809@eshu.net> Message-ID: >>>>> "JC" == Joe Christy writes: JC> [...] I haven't found corresponding documentation for an NFS based JC> kickstart. Lots of info should be in the archive of this list. JC> If I'm _not_ using iso's, just a Fedora/{base,RPMS} tree with JC> latest update rpms, what meta-data do I need to generate so that I JC> can do a kickstart install? Is hdlist still necessary, since there JC> are, strictly speaking no install discs? You need hdlist. You may also need to run pkgorder. I use this tool to keep things updated: http://www.math.uh.edu/~tibbs/do_update You just feed it the locations of your install tree and your updates mirror and it will apply the updates and do the other bits necessary to make an installable tree. JC> ... and what If I want to supplement my install with some rpms JC> from Extras? Do I need only add the packages into comps.xml and JC> proceed as before? You could copy them into the tree, edit comps if you want, and do the genhdlist/pkgorder thing (or run the above tool with "--genhdlist"). I prefer to keep the base install as pristine as possible and install as much as possible via yum either in %post or after the system boots. This makes moving between major OS versions a bit simpler. - J< From klaus.steden at thomson.net Wed Oct 12 02:45:08 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Tue, 11 Oct 2005 22:45:08 -0400 Subject: Neccessary metadata for NFS kickstart of base+updates(+extras)? In-Reply-To: References: <434C5E0E.5070809@eshu.net> Message-ID: <20051012024508.GN1808@thomson.net> > >>>>> "JC" == Joe Christy writes: > > JC> [...] I haven't found corresponding documentation for an NFS based > JC> kickstart. > > Lots of info should be in the archive of this list. > > JC> If I'm _not_ using iso's, just a Fedora/{base,RPMS} tree with > JC> latest update rpms, what meta-data do I need to generate so that I > JC> can do a kickstart install? Is hdlist still necessary, since there > JC> are, strictly speaking no install discs? > > You need hdlist. You may also need to run pkgorder. I use this tool > to keep things updated: http://www.math.uh.edu/~tibbs/do_update > > You just feed it the locations of your install tree and your updates > mirror and it will apply the updates and do the other bits necessary > to make an installable tree. > > JC> ... and what If I want to supplement my install with some rpms > JC> from Extras? Do I need only add the packages into comps.xml and > JC> proceed as before? > > You could copy them into the tree, edit comps if you want, and do the > genhdlist/pkgorder thing (or run the above tool with "--genhdlist"). > > I prefer to keep the base install as pristine as possible and install > as much as possible via yum either in %post or after the system boots. > This makes moving between major OS versions a bit simpler. > The script that Jason mentions has been added to this page of the Fedora Wiki (with a few minor tweaks): http://fedoraproject.org/wiki/AnacondaGenHdList Klaus From list at akwebhosting.net Thu Oct 13 02:25:25 2005 From: list at akwebhosting.net (Gary P) Date: Wed, 12 Oct 2005 18:25:25 -0800 Subject: Flash Drive Message-ID: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> Hi Everyone I have a 512 MB flash drive and trying to make it bootable so I can run my ks.cfg files. I have been at this for about 3 days now and heres everything I have done so far, all without any success 1. I tried the dd with bootdisk.img and that doesnt work, bascially turns the 512 MB pariton into a 1.44 MB floppy 2. I paritoned the drive going the route of FAT and using syslinux, when i try this, i get the SYSLinux line but it just sits there, doesnt load or boot All the readme and how'tos i found so far on the net dont work. has anyone gotten a flash stick to bootup and actually load?? if so, please i am begging you to give me some guidance and howto so i can get this stick to bootup -------------- next part -------------- An HTML attachment was scrubbed... URL: From tibbs at math.uh.edu Thu Oct 13 02:37:47 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Wed, 12 Oct 2005 21:37:47 -0500 Subject: Flash Drive In-Reply-To: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> (Gary P.'s message of "Wed, 12 Oct 2005 18:25:25 -0800") References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> Message-ID: >>>>> "GP" == Gary P writes: GP> All the readme and how'tos i found so far on the net dont GP> work. has anyone gotten a flash stick to bootup and actually GP> load?? Sure, I did it just by dd'ing diskboot.img to a flash drive I had lying around. I found it a little small so I just looked at the script that generates it (/usr/lib/anaconda-runtime/mk-images.i386) and made a larger image with room for more kernels. Here's the basics: mkdir -p $TOPDESTPATH/images/fobpath dd if=/dev/zero of=$TOPDESTPATH/images/diskboot.img bs=1M count=6 2>/dev/null mkdosfs $TOPDESTPATH/images/diskboot.img $SYSLINUX $TOPDESTPATH/images/diskboot.img mount -o loop $TOPDESTPATH/images/diskboot.img $TOPDESTPATH/images/fobpath cp -r $TOPDESTPATH/isolinux/* $TOPDESTPATH/images/fobpath/ grep -v local $TOPDESTPATH/images/fobpath/isolinux.cfg > $TOPDESTPATH/images/fobpath/syslinux.cfg rm -f $TOPDESTPATH/images/fobpath/isolinux.cfg umount -f $TOPDESTPATH/images/fobpath rmdir $TOPDESTPATH/images/fobpath change the count=6 on the second line to approximate the size of your flask drive if you want to fill it. I used 126 for a 128MB drive I had around. But honestly, if diskboot.img didn't work for you then I think you must have some other issue. The image worked for me on pretty much every USB-bootable machine I have access to. You aren't dd'ing the image to a partition on the flash drive, are you? It should be something like: dd if=diskboot.img of=/dev/sda not dd if=diskboot.img of=/dev/sda1 - J< From error27 at gmail.com Thu Oct 13 02:39:46 2005 From: error27 at gmail.com (Dan Carpenter) Date: Wed, 12 Oct 2005 19:39:46 -0700 Subject: Flash Drive In-Reply-To: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> Message-ID: 2005/10/12, Gary P : > I have a 512 MB flash drive and trying to make it bootable so I can run my > ks.cfg files. Could you boot from the CDROM and use ks=hd:sda/ks.cfg regards, dan carpenter From list at akwebhosting.net Thu Oct 13 02:39:11 2005 From: list at akwebhosting.net (Gary P) Date: Wed, 12 Oct 2005 18:39:11 -0800 Subject: Flash Drive References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> Message-ID: <001b01c5cf9f$4c20e0e0$6501a8c0@akwebhosting> ----- Original Message ----- From: "Jason L Tibbitts III" To: "Discussion list about Kickstart" Sent: Wednesday, October 12, 2005 6:37 PM Subject: Re: Flash Drive >>>>>> "GP" == Gary P writes: > > GP> All the readme and how'tos i found so far on the net dont > GP> work. has anyone gotten a flash stick to bootup and actually > GP> load?? > > Sure, I did it just by dd'ing diskboot.img to a flash drive I had > lying around. I found it a little small so I just looked at the > script that generates it (/usr/lib/anaconda-runtime/mk-images.i386) > and made a larger image with room for more kernels. > > Here's the basics: > > mkdir -p $TOPDESTPATH/images/fobpath > dd if=/dev/zero of=$TOPDESTPATH/images/diskboot.img bs=1M count=6 > 2>/dev/null > mkdosfs $TOPDESTPATH/images/diskboot.img > $SYSLINUX $TOPDESTPATH/images/diskboot.img > mount -o loop $TOPDESTPATH/images/diskboot.img $TOPDESTPATH/images/fobpath > cp -r $TOPDESTPATH/isolinux/* $TOPDESTPATH/images/fobpath/ > grep -v local $TOPDESTPATH/images/fobpath/isolinux.cfg > > $TOPDESTPATH/images/fobpath/syslinux.cfg > rm -f $TOPDESTPATH/images/fobpath/isolinux.cfg > umount -f $TOPDESTPATH/images/fobpath > rmdir $TOPDESTPATH/images/fobpath > > change the count=6 on the second line to approximate the size of your > flask drive if you want to fill it. I used 126 for a 128MB drive I > had around. > > But honestly, if diskboot.img didn't work for you then I think you > must have some other issue. The image worked for me on pretty much > every USB-bootable machine I have access to. > > You aren't dd'ing the image to a partition on the flash drive, are > you? It should be something like: > > dd if=diskboot.img of=/dev/sda > > not > > dd if=diskboot.img of=/dev/sda1 > > - J< > > thanks, well the problem i got stuck at with DD was the fact that it kept formatting it to 1.44MB so thats as far as i went when i went that route i will read over your email closely and give it a try, thanks! Gary From list at akwebhosting.net Thu Oct 13 02:40:10 2005 From: list at akwebhosting.net (Gary P) Date: Wed, 12 Oct 2005 18:40:10 -0800 Subject: Flash Drive References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> Message-ID: <002001c5cf9f$6ecd38a0$6501a8c0@akwebhosting> ----- Original Message ----- From: "Dan Carpenter" To: "Discussion list about Kickstart" Sent: Wednesday, October 12, 2005 6:39 PM Subject: Re: Flash Drive 2005/10/12, Gary P : > I have a 512 MB flash drive and trying to make it bootable so I can run my > ks.cfg files. Could you boot from the CDROM and use ks=hd:sda/ks.cfg regards, dan carpenter Yes thats what we are doing right now, but we have a NOC full of about 50 machines, and non have a cdrom, so its a hassle for our techs to track down our external rom, hook it up, find the proper cd etc i would rather just send this down there and bang, no cds to deal with From tibbs at math.uh.edu Thu Oct 13 02:53:29 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Wed, 12 Oct 2005 21:53:29 -0500 Subject: Flash Drive In-Reply-To: <001b01c5cf9f$4c20e0e0$6501a8c0@akwebhosting> (Gary P.'s message of "Wed, 12 Oct 2005 18:39:11 -0800") References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> <001b01c5cf9f$4c20e0e0$6501a8c0@akwebhosting> Message-ID: >>>>> "GP" == Gary P writes: GP> well the problem i got stuck at with DD was the fact that GP> it kept formatting it to 1.44MB I'm not really sure what's up here. diskboot.img should be about 6MB; there's no formatting involved in copying it byte-for-byte to the flash drive. - J< From list at akwebhosting.net Thu Oct 13 03:01:28 2005 From: list at akwebhosting.net (Gary P) Date: Wed, 12 Oct 2005 19:01:28 -0800 Subject: Flash Drive References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> Message-ID: <002e01c5cfa2$68ebe9b0$6501a8c0@akwebhosting> ----- Original Message ----- From: "Jason L Tibbitts III" To: "Discussion list about Kickstart" Sent: Wednesday, October 12, 2005 6:37 PM Subject: Re: Flash Drive >>>>>> "GP" == Gary P writes: > > GP> All the readme and how'tos i found so far on the net dont > GP> work. has anyone gotten a flash stick to bootup and actually > GP> load?? > > Sure, I did it just by dd'ing diskboot.img to a flash drive I had > lying around. I found it a little small so I just looked at the > script that generates it (/usr/lib/anaconda-runtime/mk-images.i386) > and made a larger image with room for more kernels. > > Here's the basics: > > mkdir -p $TOPDESTPATH/images/fobpath > dd if=/dev/zero of=$TOPDESTPATH/images/diskboot.img bs=1M count=6 > 2>/dev/null > mkdosfs $TOPDESTPATH/images/diskboot.img > $SYSLINUX $TOPDESTPATH/images/diskboot.img > mount -o loop $TOPDESTPATH/images/diskboot.img $TOPDESTPATH/images/fobpath > cp -r $TOPDESTPATH/isolinux/* $TOPDESTPATH/images/fobpath/ > grep -v local $TOPDESTPATH/images/fobpath/isolinux.cfg > > $TOPDESTPATH/images/fobpath/syslinux.cfg > rm -f $TOPDESTPATH/images/fobpath/isolinux.cfg > umount -f $TOPDESTPATH/images/fobpath > rmdir $TOPDESTPATH/images/fobpath > > change the count=6 on the second line to approximate the size of your > flask drive if you want to fill it. I used 126 for a 128MB drive I > had around. > > But honestly, if diskboot.img didn't work for you then I think you > must have some other issue. The image worked for me on pretty much > every USB-bootable machine I have access to. > > You aren't dd'ing the image to a partition on the flash drive, are > you? It should be something like: > > dd if=diskboot.img of=/dev/sda > > not > > dd if=diskboot.img of=/dev/sda1 > > - J< > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > > Ok, did what you said (dd if=/dev/sda of=bootdisk.img bs=1M count=510) and on reboot i am getting boot failed From list at akwebhosting.net Thu Oct 13 03:07:19 2005 From: list at akwebhosting.net (Gary P) Date: Wed, 12 Oct 2005 19:07:19 -0800 Subject: Flash Drive References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting><001b01c5cf9f$4c20e0e0$6501a8c0@akwebhosting> Message-ID: <003b01c5cfa3$3a13eb50$6501a8c0@akwebhosting> ----- Original Message ----- From: "Jason L Tibbitts III" To: "Discussion list about Kickstart" Sent: Wednesday, October 12, 2005 6:53 PM Subject: Re: Flash Drive >>>>>> "GP" == Gary P writes: > > GP> well the problem i got stuck at with DD was the fact that > GP> it kept formatting it to 1.44MB > > I'm not really sure what's up here. diskboot.img should be about 6MB; > there's no formatting involved in copying it byte-for-byte to the > flash drive. > when i run your script, heres what i get [root at localhost home]# sh script.sh mkdosfs 2.8 (28 Feb 2001) script.sh: line 4: /images/diskboot.img: Permission denied cp: cannot stat `/isolinux/*': No such file or directory grep: /images/fobpath/isolinux.cfg: No such file or directory heres whats there [root at localhost images]# ls -l total 536080 -rw-r--r-- 1 root root 548405248 Oct 13 11:07 diskboot.img From tibbs at math.uh.edu Thu Oct 13 03:11:37 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Wed, 12 Oct 2005 22:11:37 -0500 Subject: Flash Drive In-Reply-To: <002e01c5cfa2$68ebe9b0$6501a8c0@akwebhosting> (Gary P.'s message of "Wed, 12 Oct 2005 19:01:28 -0800") References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> <002e01c5cfa2$68ebe9b0$6501a8c0@akwebhosting> Message-ID: >>>>> "GP" == Gary P writes: GP> Ok, did what you said (dd if=/dev/sda of=bootdisk.img bs=1M GP> count=510) and on reboot i am getting boot failed That's backwards; you copied 510MB from the flash drive to the file bootdisk.img. Assuming your drive really is /dev/sda, it should be nothing more than: dd if=diskboot.img of=/dev/sda where diskboot.img is the file in the "images" directory of your Fedora/Red Hat/Centos/whatever media. - J< From tibbs at math.uh.edu Thu Oct 13 03:32:14 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Wed, 12 Oct 2005 22:32:14 -0500 Subject: Flash Drive In-Reply-To: <003b01c5cfa3$3a13eb50$6501a8c0@akwebhosting> (Gary P.'s message of "Wed, 12 Oct 2005 19:07:19 -0800") References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> <001b01c5cf9f$4c20e0e0$6501a8c0@akwebhosting> <003b01c5cfa3$3a13eb50$6501a8c0@akwebhosting> Message-ID: >>>>> "GP" == Gary P writes: GP> when i run your script, heres what i get That wasn't really a script, it was just some information on how you might go about making your own diskboot.img if you wanted to do it by hand. I included it without knowing the complete nature of your problem or your experience level. You should just take the distro-provided diskboot.img and put it on the drive. Nothing more. Get into the details if you want to hack further. - J< From list at akwebhosting.net Thu Oct 13 03:53:32 2005 From: list at akwebhosting.net (Gary P) Date: Wed, 12 Oct 2005 19:53:32 -0800 Subject: Flash Drive References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting><001b01c5cf9f$4c20e0e0$6501a8c0@akwebhosting><003b01c5cfa3$3a13eb50$6501a8c0@akwebhosting> Message-ID: <004801c5cfa9$aede3cf0$6501a8c0@akwebhosting> ----- Original Message ----- From: "Jason L Tibbitts III" To: "Discussion list about Kickstart" Sent: Wednesday, October 12, 2005 7:32 PM Subject: Re: Flash Drive >>>>>> "GP" == Gary P writes: > > GP> when i run your script, heres what i get > > That wasn't really a script, it was just some information on how you > might go about making your own diskboot.img if you wanted to do it by > hand. I included it without knowing the complete nature of your > problem or your experience level. > > You should just take the distro-provided diskboot.img and put it on > the drive. Nothing more. Get into the details if you want to hack > further. > > - J< > > yah after i sent that, i figured that much out now i am stuck (again) when i run dd if=/home/diskboot.img of=/dev/sda bs=1M count=512 496+0 records in 496+0 records out it just sits there, i can see the light on the flash card blinking but this has been going on for about 20-30 minutes now.... From list at akwebhosting.net Thu Oct 13 04:18:08 2005 From: list at akwebhosting.net (Gary P) Date: Wed, 12 Oct 2005 20:18:08 -0800 Subject: Flash Drive References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting><001b01c5cf9f$4c20e0e0$6501a8c0@akwebhosting><003b01c5cfa3$3a13eb50$6501a8c0@akwebhosting> Message-ID: <005301c5cfad$1eb77ac0$6501a8c0@akwebhosting> ----- Original Message ----- From: "Jason L Tibbitts III" To: "Discussion list about Kickstart" Sent: Wednesday, October 12, 2005 7:32 PM Subject: Re: Flash Drive >>>>>> "GP" == Gary P writes: > > GP> when i run your script, heres what i get > > That wasn't really a script, it was just some information on how you > might go about making your own diskboot.img if you wanted to do it by > hand. I included it without knowing the complete nature of your > problem or your experience level. > > You should just take the distro-provided diskboot.img and put it on > the drive. Nothing more. Get into the details if you want to hack > further. > After running the dd command (and it finaly finished, heres the contents of the stick) do i still have to run something like syslinux on the device?? [root at localhost stick]# dir boot.cat general.msg ldlinux.sys param.msg snake.msg syslinux.cfg vmlinuz.txt boot.msg initrd.img options.msg rescue.msg splash.lss vmlinuz From tibbs at math.uh.edu Thu Oct 13 04:22:11 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Wed, 12 Oct 2005 23:22:11 -0500 Subject: Flash Drive In-Reply-To: <005301c5cfad$1eb77ac0$6501a8c0@akwebhosting> (Gary P.'s message of "Wed, 12 Oct 2005 20:18:08 -0800") References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> <001b01c5cf9f$4c20e0e0$6501a8c0@akwebhosting> <003b01c5cfa3$3a13eb50$6501a8c0@akwebhosting> <005301c5cfad$1eb77ac0$6501a8c0@akwebhosting> Message-ID: >>>>> "GP" == Gary P writes: GP> After running the dd command (and it finaly finished, heres the GP> contents of the stick) do i still have to run something like GP> syslinux on the device?? It was already run on the image, so there shouldn't be any reason to run it again. - J< From list at akwebhosting.net Thu Oct 13 04:47:36 2005 From: list at akwebhosting.net (Gary P) Date: Wed, 12 Oct 2005 20:47:36 -0800 Subject: Flash Drive References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting><001b01c5cf9f$4c20e0e0$6501a8c0@akwebhosting><003b01c5cfa3$3a13eb50$6501a8c0@akwebhosting><005301c5cfad$1eb77ac0$6501a8c0@akwebhosting> Message-ID: <005801c5cfb1$3c6a2f00$6501a8c0@akwebhosting> ----- Original Message ----- From: "Jason L Tibbitts III" To: "Discussion list about Kickstart" Sent: Wednesday, October 12, 2005 8:22 PM Subject: Re: Flash Drive >>>>>> "GP" == Gary P writes: > > GP> After running the dd command (and it finaly finished, heres the > GP> contents of the stick) do i still have to run something like > GP> syslinux on the device?? > > It was already run on the image, so there shouldn't be any reason to > run it again. > > - J< > bascially back to square one....i've honestly tried everything and cant figure it out I've done everything you stated and other readmes and sometimes i get 'bootup failed' or like right now, it just sits there likes its trying to load syslinux but doesnt go anywhere From tibbs at math.uh.edu Thu Oct 13 05:00:43 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Thu, 13 Oct 2005 00:00:43 -0500 Subject: Flash Drive In-Reply-To: <005801c5cfb1$3c6a2f00$6501a8c0@akwebhosting> (Gary P.'s message of "Wed, 12 Oct 2005 20:47:36 -0800") References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> <001b01c5cf9f$4c20e0e0$6501a8c0@akwebhosting> <003b01c5cfa3$3a13eb50$6501a8c0@akwebhosting> <005301c5cfad$1eb77ac0$6501a8c0@akwebhosting> <005801c5cfb1$3c6a2f00$6501a8c0@akwebhosting> Message-ID: >>>>> "GP" == Gary P writes: GP> bascially back to square one....i've honestly tried everything and GP> cant figure it out I've done everything you stated and other GP> readmes and sometimes i get 'bootup failed' or like right now, it GP> just sits there likes its trying to load syslinux but doesnt go GP> anywhere On multiple computers, known to boot other USB devices? Some machines just don't handle booting from USB very well. I know that a simple dd of diskboot.img produces a flash device which will boot on several machines I have around. Likewise, the same device won't boot on a couple of machines; these don't seem to properly handle a USB floppy, either. - J< From error27 at gmail.com Thu Oct 13 05:06:08 2005 From: error27 at gmail.com (Dan Carpenter) Date: Wed, 12 Oct 2005 22:06:08 -0700 Subject: Flash Drive In-Reply-To: <002001c5cf9f$6ecd38a0$6501a8c0@akwebhosting> References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> <002001c5cf9f$6ecd38a0$6501a8c0@akwebhosting> Message-ID: 2005/10/12, Gary P : > Yes thats what we are doing right now, but we have a NOC full of about 50 > machines, and non have a cdrom, so its a hassle for our techs to track down > our external rom, hook it up, find the proper cd etc > > i would rather just send this down there and bang, no cds to deal with > I just used the cdrom as an example. You could do the same thing with an NFS install etc. What kind of media are you using for the install? You could put the kickstart file on the nfs server instead even if you wanted to. regards, dan carpenter From list at akwebhosting.net Thu Oct 13 05:14:09 2005 From: list at akwebhosting.net (Gary P) Date: Wed, 12 Oct 2005 21:14:09 -0800 Subject: Flash Drive References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting><001b01c5cf9f$4c20e0e0$6501a8c0@akwebhosting><003b01c5cfa3$3a13eb50$6501a8c0@akwebhosting><005301c5cfad$1eb77ac0$6501a8c0@akwebhosting><005801c5cfb1$3c6a2f00$6501a8c0@akwebhosting> Message-ID: <005f01c5cfb4$f2690940$6501a8c0@akwebhosting> ----- Original Message ----- From: "Jason L Tibbitts III" To: "Discussion list about Kickstart" Sent: Wednesday, October 12, 2005 9:00 PM Subject: Re: Flash Drive >>>>>> "GP" == Gary P writes: > > GP> bascially back to square one....i've honestly tried everything and > GP> cant figure it out I've done everything you stated and other > GP> readmes and sometimes i get 'bootup failed' or like right now, it > GP> just sits there likes its trying to load syslinux but doesnt go > GP> anywhere > > On multiple computers, known to boot other USB devices? Some machines > just don't handle booting from USB very well. I know that a simple dd > of diskboot.img produces a flash device which will boot on several > machines I have around. Likewise, the same device won't boot on a > couple of machines; these don't seem to properly handle a USB floppy, > either. > > - J< > well these are pretty new machines and BIOS even sees the flash stick and its selectable as one of the pull downs of the HDD prioritys From list at akwebhosting.net Thu Oct 13 05:15:47 2005 From: list at akwebhosting.net (Gary P) Date: Wed, 12 Oct 2005 21:15:47 -0800 Subject: Flash Drive References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting><002001c5cf9f$6ecd38a0$6501a8c0@akwebhosting> Message-ID: <006401c5cfb5$2bf76170$6501a8c0@akwebhosting> ----- Original Message ----- From: "Dan Carpenter" To: "Discussion list about Kickstart" Sent: Wednesday, October 12, 2005 9:06 PM Subject: Re: Flash Drive 2005/10/12, Gary P : > Yes thats what we are doing right now, but we have a NOC full of about 50 > machines, and non have a cdrom, so its a hassle for our techs to track > down > our external rom, hook it up, find the proper cd etc > > i would rather just send this down there and bang, no cds to deal with > I just used the cdrom as an example. You could do the same thing with an NFS install etc. What kind of media are you using for the install? You could put the kickstart file on the nfs server instead even if you wanted to. regards, dan carpenter I have looked at NFS and it looks viable but im not to linux savy to go into something like that From joe at swelltech.com Thu Oct 13 07:11:49 2005 From: joe at swelltech.com (Joe Cooper) Date: Thu, 13 Oct 2005 02:11:49 -0500 Subject: Flash Drive In-Reply-To: <004801c5cfa9$aede3cf0$6501a8c0@akwebhosting> References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting><001b01c5cf9f$4c20e0e0$6501a8c0@akwebhosting><003b01c5cfa3$3a13eb50$6501a8c0@akwebhosting> <004801c5cfa9$aede3cf0$6501a8c0@akwebhosting> Message-ID: <434E08B5.6070403@swelltech.com> Gary P wrote: > > ----- Original Message ----- From: "Jason L Tibbitts III" > > To: "Discussion list about Kickstart" > Sent: Wednesday, October 12, 2005 7:32 PM > Subject: Re: Flash Drive > > >>>>>>> "GP" == Gary P writes: >> >> >> GP> when i run your script, heres what i get >> >> That wasn't really a script, it was just some information on how you >> might go about making your own diskboot.img if you wanted to do it by >> hand. I included it without knowing the complete nature of your >> problem or your experience level. >> >> You should just take the distro-provided diskboot.img and put it on >> the drive. Nothing more. Get into the details if you want to hack >> further. >> >> - J< >> >> > yah after i sent that, i figured that much out > > now i am stuck (again) when i run dd if=/home/diskboot.img of=/dev/sda > bs=1M count=512 > 496+0 records in > 496+0 records out > > it just sits there, i can see the light on the flash card blinking but > this has been going on for about 20-30 minutes now.... That isn't the command that has been recommended to you. You've added "bs=1M count=512"...this is going to corrupt the filesystem. I don't know if it will prevent booting, but it isn't what you want to do. Use the command that was recommended: dd if=/home/diskboot.img of=/dev/sda Nothing more. This is not the time for improvising. (BTW-Make sure your earlier dd in the opposite direction hasn't overwritten your diskboot.img. That would also lead to brokenness of some degree.) If that doesn't boot, then your systems won't boot from this particular USB device. That is stage one in your troubleshooting adventure. You're trying to tackle too many things at once. Getting a bigger image is for after you know these servers will boot from the device you want to boot from, and the command you've given is not the way to accomplish it. Another aside...I use PXE booting when I have a slew of servers to install. No messing about with physical media...but you do have to setup a few things on a local server. From cleve.cornelius at kroger.com Thu Oct 13 08:00:41 2005 From: cleve.cornelius at kroger.com (cleve.cornelius at kroger.com) Date: Thu, 13 Oct 2005 04:00:41 -0400 Subject: Cleve Cornelius/MIS/CORP/KrogerCo is out of the office. Message-ID: I will be out of the office starting 10/13/2005 and will not return until 10/17/2005. If this is an urgent matter please contact the NOC and they will page the on call engineer to assist you. From phr at doc.ic.ac.uk Thu Oct 13 10:53:33 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Thu, 13 Oct 2005 11:53:33 +0100 (BST) Subject: Flash Drive In-Reply-To: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> References: <001101c5cf9d$5f7a41b0$6501a8c0@akwebhosting> Message-ID: On Wed, 12 Oct 2005, Gary P wrote: >I have a 512 MB flash drive and trying to make it bootable so I can run >my ks.cfg files. [snip] >All the readme and how'tos i found so far on the net dont work. Did you check the syslinux docs on this subject: http://syslinux.zytor.com/usbkey.php Also: http://d-i.pascal.at/ explains how to install Debian, but the steps would be very similar for Redhat. Cheers, Phil From sdouche at gmail.com Wed Oct 19 10:35:43 2005 From: sdouche at gmail.com (Sebastien Douche) Date: Wed, 19 Oct 2005 12:35:43 +0200 Subject: [urgent] Custom CD doesn't recognize stage2 Message-ID: <5e1183fa0510190335i1b786a26vb4dd71a320ace049@mail.gmail.com> hi, I have a problem to create a custom CD of Red Hat ES 3 on IBM zSeries (x86-64) : The bootdisk.img load, I choose the entry in my custom menu, vmlinuz load, installer reads the kickstart file and error appears : red hat cd not found. Same error If I insert RedHat CD 1 after my custom CD. What's wrong ? I have all files RedHat/base directory (with custom hdlist*). discinfo file is also present (no modification). I use this directive : mkisofs -r -T -J -V "Master" -b images/bootdisk.img -O ~/master.iso Any ideas ? -- S?bastien Douche From cleve.cornelius at kroger.com Wed Oct 19 19:53:15 2005 From: cleve.cornelius at kroger.com (cleve.cornelius at kroger.com) Date: Wed, 19 Oct 2005 15:53:15 -0400 Subject: Cleve Cornelius/MIS/CORP/KrogerCo is out of the office. Message-ID: I will be out of the office starting Wed 10/19/2005 and will not return until Mon 10/24/2005. If this is an urgent matter please contact the NOC and they will page the on call engineer to assist you. From clumens at redhat.com Thu Oct 20 18:12:08 2005 From: clumens at redhat.com (Chris Lumens) Date: Thu, 20 Oct 2005 14:12:08 -0400 (EDT) Subject: more on the new kickstart stuff Message-ID: I've added a bunch of documentation to the new pykickstart package now, complete with examples of how you can extend it to do what you want. Let me know if there are any deficiencies, if anything is unclear, etc. and I'll work on it. Next thing is to write a validator and make these error messages much more useful. - Chris From aboojy at gmail.com Sun Oct 23 23:02:06 2005 From: aboojy at gmail.com (abuji) Date: Sun, 23 Oct 2005 18:02:06 -0500 Subject: genhdlist Message-ID: <811ba0c10510231602i36c42f80xb97469ab92058c26@mail.gmail.com> All, We are testing RH4.0 update 2 build off a kickstart server (at Rh3.0level), configured to support RH4.0 client builds. Incidentally after running "/usr/lib/anaconda-runtime/genhdlist /images/rh40u2" (/images/rh40u2 is the root dir where RedHat resides) the build fails. The client boots fien, begins the build but quits. The genhdlist command completes without error and hdlist and hdlist2 files are created fine. Even without adding any rpms to the RPMS dir, the build fails Was wondering what is the correct syntax for genhdlist. There is no man pages or info on it. Thanks Abraham -------------- next part -------------- An HTML attachment was scrubbed... URL: From phr at doc.ic.ac.uk Mon Oct 24 00:17:34 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Mon, 24 Oct 2005 01:17:34 +0100 (BST) Subject: genhdlist In-Reply-To: <811ba0c10510231602i36c42f80xb97469ab92058c26@mail.gmail.com> References: <811ba0c10510231602i36c42f80xb97469ab92058c26@mail.gmail.com> Message-ID: Please don't post HTML to mailing lists. On Sun, 23 Oct 2005, abuji wrote: > We are testing RH4.0 update 2 build off a kickstart server (at >Rh3.0level), configured to support RH4.0 client builds. > Incidentally after running "/usr/lib/anaconda-runtime/genhdlist >/images/rh40u2" (/images/rh40u2 is the root dir where RedHat resides) the >build fails. The client boots fien, begins the build but quits. > The genhdlist command completes without error and hdlist and hdlist2 files >are created fine. Even without adding any rpms to the RPMS dir, the build >fails Fails how? Are any error messages seen on the other virtual consoles? (Alt+F2, F3 etc.) If anaconda is still behaving strangely w.r.t. package ordering, try section 9 of: http://rau.homedns.org/twiki/bin/view/Anaconda/CustomizingAnaconda (you can ignore the rest of that page, as you're not rebuilding anaconda as a whole.) Cheers, Phil From arifsaha at yahoo.com Mon Oct 24 20:58:15 2005 From: arifsaha at yahoo.com (S P Arif Sahari Wibowo) Date: Mon, 24 Oct 2005 15:58:15 -0500 (CDT) Subject: Passing additional options to mkfs Message-ID: Hi! Are there any way to pass additional / file-system-specific options for mkfs in kickstart? For instance, I would like to define the block size and the frequency of inode for one partition. If it cannot be done from the "part" statement, will it work if I put commands to reformat / tune the partition in the "pre" section? Thanks! -- Stephan Paul Arif Sahari Wibowo _____ _____ _____ _____ /____ /____/ /____/ /____ _____/ / / / _____/ http://www.arifsaha.com/ From Chip.Shabazian at bankofamerica.com Mon Oct 24 21:04:42 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Mon, 24 Oct 2005 14:04:42 -0700 Subject: Passing additional options to mkfs Message-ID: You can setup your partition as you want in %pre, then use: part --noformat --onpart=/dev/specific_partition in the command section -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of S P Arif Sahari Wibowo Sent: Monday, October 24, 2005 1:58 PM To: kickstart-list at redhat.com Subject: Passing additional options to mkfs Hi! Are there any way to pass additional / file-system-specific options for mkfs in kickstart? For instance, I would like to define the block size and the frequency of inode for one partition. If it cannot be done from the "part" statement, will it work if I put commands to reformat / tune the partition in the "pre" section? Thanks! -- Stephan Paul Arif Sahari Wibowo _____ _____ _____ _____ /____ /____/ /____/ /____ _____/ / / / _____/ http://www.arifsaha.com/ _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From arifsaha at yahoo.com Mon Oct 24 21:16:23 2005 From: arifsaha at yahoo.com (S P Arif Sahari Wibowo) Date: Mon, 24 Oct 2005 16:16:23 -0500 (CDT) Subject: Passing additional options to mkfs In-Reply-To: References: Message-ID: On Mon, 24 Oct 2005, Shabazian, Chip wrote: > You can setup your partition as you want in %pre, then use: > part --noformat --onpart=/dev/specific_partition > in the command section Sounds good. I was not sure whether "%pre" is executed before or after kickstart doing its formating. What is the recommended tools to create partition while in "%pre"? sfdisk? Thanks! -- Stephan Paul Arif Sahari Wibowo _____ _____ _____ _____ /____ /____/ /____/ /____ _____/ / / / _____/ http://www.arifsaha.com/ From klaus.steden at thomson.net Mon Oct 24 21:19:29 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Mon, 24 Oct 2005 17:19:29 -0400 Subject: Passing additional options to mkfs In-Reply-To: References: Message-ID: <20051024211929.GG33316@thomson.net> > On Mon, 24 Oct 2005, Shabazian, Chip wrote: > >You can setup your partition as you want in %pre, then use: > >part --noformat --onpart=/dev/specific_partition > >in the command section > > Sounds good. I was not sure whether "%pre" is executed before or > after kickstart doing its formating. > > What is the recommended tools to create partition while in > "%pre"? sfdisk? > '%pre' runs before anything else in Kickstart (i.e. before disks are formatted), '%post' runs after everything else. I think by now both fdisk and sfdisk are available in the bootstrap environment. hth, Klaus From brilong at cisco.com Tue Oct 25 15:35:31 2005 From: brilong at cisco.com (Brian Long) Date: Tue, 25 Oct 2005 11:35:31 -0400 Subject: Ignoring SAN devices during kickstart? Message-ID: <1130254531.4050.49.camel@brilong-lnx> Hello, We running into interactive kickstart issues when specifying specific "clearpart" and "bootloader" directives and other devices (i.e. /dev/sda) are read-only EMC SAN devices. Anaconda consistently says "The partition table on device sda was unreadable." and asks if I want to initialize this drive. Since this is the EMC Volume Logix LUN, it's read-only and needs to be disregarded. What are folks doing when they re-kickstart a machine with SAN LUNs presented? On our HP Proliants, I've started specifying drive order cciss/c0d0 so Anaconda doesn't tell GRUB to install to /dev/sda. Any other tips would be appreciated. If I have 30 LUNs presented to a machine and some of those LUNs are formatted with Sun labels, Anaconda asks me if I want to format them. I'd rather tell Anaconda to just proceed with loading to cciss/c0d0 and disregard these LUNs. In our RHEL 3 customized distro, we actually removed the qla2x00 drivers from the boot images to avoid overwriting SAN data. Now we're trying to use the untouched RHEL 4 boot images and running into all these issues. Thanks for any help! /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From Robert.McKenzie at csr.com Tue Oct 25 15:57:00 2005 From: Robert.McKenzie at csr.com (Robert McKenzie) Date: Tue, 25 Oct 2005 16:57:00 +0100 Subject: Ignoring SAN devices during kickstart? Message-ID: <140F44C0EA24954FA7744AA7D60A62007E381C@exchange03> We have the same problem with out Dell EMC SAN. The only way I've gotten around it was to disconnect the fibre from the machine being kicked started. On first reboot plug it back in. It's a headache but seems to be the only reliable way to get around it. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Brian Long Sent: 25 October 2005 16:36 To: kickstart-list at redhat.com Subject: Ignoring SAN devices during kickstart? Hello, We running into interactive kickstart issues when specifying specific "clearpart" and "bootloader" directives and other devices (i.e. /dev/sda) are read-only EMC SAN devices. Anaconda consistently says "The partition table on device sda was unreadable." and asks if I want to initialize this drive. Since this is the EMC Volume Logix LUN, it's read-only and needs to be disregarded. What are folks doing when they re-kickstart a machine with SAN LUNs presented? On our HP Proliants, I've started specifying drive order cciss/c0d0 so Anaconda doesn't tell GRUB to install to /dev/sda. Any other tips would be appreciated. If I have 30 LUNs presented to a machine and some of those LUNs are formatted with Sun labels, Anaconda asks me if I want to format them. I'd rather tell Anaconda to just proceed with loading to cciss/c0d0 and disregard these LUNs. In our RHEL 3 customized distro, we actually removed the qla2x00 drivers from the boot images to avoid overwriting SAN data. Now we're trying to use the untouched RHEL 4 boot images and running into all these issues. Thanks for any help! /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list To report this email as spam click https://www.mailcontrol.com/sr/wQw0zmjPoHdJTZGyOCrrhg== TCv0mf2rI!8A5HVV5Qz5YSGfnUssFB9X752PnfZ+ZPYwd4ixF065tcsqUnXEt4TCGE!Koqps hAlU4I+QGQhvcWu00+UlDnkj+hFhVRR10F2dKeedcDpZfkg0gYa4wF3SN+E31NqKvM0Lkn2P D8NFvzUBZv6ueekOKrI5QEcfQxZelNDjQizYcNb . From brilong at cisco.com Tue Oct 25 17:18:26 2005 From: brilong at cisco.com (Brian Long) Date: Tue, 25 Oct 2005 13:18:26 -0400 Subject: Ignoring SAN devices during kickstart? In-Reply-To: <140F44C0EA24954FA7744AA7D60A62007E381C@exchange03> References: <140F44C0EA24954FA7744AA7D60A62007E381C@exchange03> Message-ID: <1130260706.4050.56.camel@brilong-lnx> On Tue, 2005-10-25 at 16:57 +0100, Robert McKenzie wrote: > We have the same problem with out Dell EMC SAN. The only way I've > gotten around it was to disconnect the fibre from the machine being > kicked started. On first reboot plug it back in. It's a headache but > seems to be the only reliable way to get around it. That is not scalable. When you manage hundreds of remote servers, you cannot get someone to walk into a locked-down data center, unplug the SAN, etc. We need to find a scalable way to fix this. If there is some magic I can include on the kernel command-line to tell it to not load qla2300 or only load qla2300 when local storage is not found, that would be ideal. /Brian/ P.S. Please don't top-post. > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Brian Long > Sent: 25 October 2005 16:36 > To: kickstart-list at redhat.com > Subject: Ignoring SAN devices during kickstart? > > > Hello, > > We running into interactive kickstart issues when specifying specific > "clearpart" and "bootloader" directives and other devices (i.e. > /dev/sda) are read-only EMC SAN devices. Anaconda consistently says > "The partition table on device sda was unreadable." and asks if I want > to initialize this drive. Since this is the EMC Volume Logix LUN, it's > read-only and needs to be disregarded. > > What are folks doing when they re-kickstart a machine with SAN LUNs > presented? On our HP Proliants, I've started specifying drive order > cciss/c0d0 so Anaconda doesn't tell GRUB to install to /dev/sda. Any > other tips would be appreciated. > > If I have 30 LUNs presented to a machine and some of those LUNs are > formatted with Sun labels, Anaconda asks me if I want to format them. > I'd rather tell Anaconda to just proceed with loading to cciss/c0d0 and > disregard these LUNs. > > In our RHEL 3 customized distro, we actually removed the qla2x00 drivers > from the boot images to avoid overwriting SAN data. Now we're trying to > use the untouched RHEL 4 boot images and running into all these issues. > > Thanks for any help! > > /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From alexander_rau at yahoo.com Tue Oct 25 17:21:27 2005 From: alexander_rau at yahoo.com (Alexander Rau) Date: Tue, 25 Oct 2005 10:21:27 -0700 (PDT) Subject: Ignoring SAN devices during kickstart? In-Reply-To: <1130260706.4050.56.camel@brilong-lnx> Message-ID: <20051025172127.42966.qmail@web52111.mail.yahoo.com> I did this with the noprobe option when installing and then loading the rest of the necessary drivers manually except the qla2300. This was for a manual install though... Not sure if this works in a kickstarted environment.... AR From Chip.Shabazian at bankofamerica.com Tue Oct 25 17:31:25 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Tue, 25 Oct 2005 10:31:25 -0700 Subject: Ignoring SAN devices during kickstart? Message-ID: What version and update are you using? There is a --loadfclast command that can be used on RHEL 3 after update 4 I believe. Not sure what update you need for RHEL 4. If anyone has any better ideas, I would love to hear them as we run into this all the time. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Alexander Rau Sent: Tuesday, October 25, 2005 10:21 AM To: Discussion list about Kickstart Subject: RE: Ignoring SAN devices during kickstart? I did this with the noprobe option when installing and then loading the rest of the necessary drivers manually except the qla2300. This was for a manual install though... Not sure if this works in a kickstarted environment.... AR _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From brilong at cisco.com Tue Oct 25 17:52:46 2005 From: brilong at cisco.com (Brian Long) Date: Tue, 25 Oct 2005 13:52:46 -0400 Subject: Ignoring SAN devices during kickstart? In-Reply-To: References: Message-ID: <1130262766.4050.60.camel@brilong-lnx> On Tue, 2005-10-25 at 10:31 -0700, Shabazian, Chip wrote: > What version and update are you using? There is a --loadfclast command > that can be used on RHEL 3 after update 4 I believe. Not sure what > update you need for RHEL 4. > > If anyone has any better ideas, I would love to hear them as we run into > this all the time. Uggh, more top-posting :) latefcload was only implemented in U6 to my knowledge. Also, this just means qla2300 will be loaded after mptbase or whatever local SCSI disks you have. This is fine when you want to make sure /dev/sda is the local SCSI disk; however with HP Proliants, the RAID controller disks show up as /dev/cciss/c0d0, c0d0, etc. /dev/sda is the first SAN LUN and I want Anaconda to just completely disregard it. I wonder if Alexander's noprobe option along with "device scsi cciss" or something in ks.cfg might work. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From kernel at linuxace.com Tue Oct 25 17:57:25 2005 From: kernel at linuxace.com (Phil Oester) Date: Tue, 25 Oct 2005 10:57:25 -0700 Subject: Ignoring SAN devices during kickstart? In-Reply-To: <1130262766.4050.60.camel@brilong-lnx> References: <1130262766.4050.60.camel@brilong-lnx> Message-ID: <20051025175725.GA7985@linuxace.com> On Tue, Oct 25, 2005 at 01:52:46PM -0400, Brian Long wrote: > latefcload was only implemented in U6 to my knowledge. Also, this just > means qla2300 will be loaded after mptbase or whatever local SCSI disks > you have. This is fine when you want to make sure /dev/sda is the local > SCSI disk; however with HP Proliants, the RAID controller disks show up > as /dev/cciss/c0d0, c0d0, etc. /dev/sda is the first SAN LUN and I want > Anaconda to just completely disregard it. > > I wonder if Alexander's noprobe option along with "device scsi cciss" or > something in ks.cfg might work. You'd be better off with 'nostorage' on the command line, so that anaconda will still probe other devices such as network. Phil From smah at vmware.com Thu Oct 27 00:33:14 2005 From: smah at vmware.com (Stephen Mah) Date: Wed, 26 Oct 2005 17:33:14 -0700 Subject: swap partition on 2nd slice Message-ID: <4360204A.9080802@vmware.com> Is there a way to make kickstart force the swap partition to the second slice? steve From sjbrowne at bluebottle.com Thu Oct 27 01:00:19 2005 From: sjbrowne at bluebottle.com (Stuart J. Browne) Date: Thu, 27 Oct 2005 11:00:19 +1000 Subject: swap partition on 2nd slice References: <4360204A.9080802@vmware.com> Message-ID: <0d7001c5da91$cea12eb0$271aa8c0@bluebottle.com> ----- Original Message ----- From: "Stephen Mah" To: Sent: Thursday, October 27, 2005 10:33 AM Subject: swap partition on 2nd slice > Is there a way to make kickstart force the swap partition to the second > slice? Look at --onpart i.e. part swap --onpart=/dev/hda2 bkx From dan at half-asleep.com Thu Oct 27 01:13:13 2005 From: dan at half-asleep.com (Daniel Segall) Date: Wed, 26 Oct 2005 21:13:13 -0400 Subject: swap partition on 2nd slice In-Reply-To: <0d7001c5da91$cea12eb0$271aa8c0@bluebottle.com> References: <4360204A.9080802@vmware.com> <0d7001c5da91$cea12eb0$271aa8c0@bluebottle.com> Message-ID: <436029A9.3010303@half-asleep.com> Stuart J. Browne wrote: > ----- Original Message ----- From: "Stephen Mah" > To: > Sent: Thursday, October 27, 2005 10:33 AM > Subject: swap partition on 2nd slice > > >> Is there a way to make kickstart force the swap partition to the >> second slice? > > > Look at --onpart > > i.e. > > part swap --onpart=/dev/hda2 > > bkx You should also note that --onpart expects the partition to be created prior to install. -Dan From LAWRENCE.SORRILLO at HAMPTONU.EDU Thu Oct 27 17:34:03 2005 From: LAWRENCE.SORRILLO at HAMPTONU.EDU (LAWRENCE.SORRILLO at HAMPTONU.EDU) Date: Thu, 27 Oct 2005 13:34:03 -0400 Subject: nfs kickstart network install Message-ID: <9B4DE93284CA884FA41C11102AEB741E0410C3B2@HUMX.infotech.hamptonu.edu> I want to perform an Kickstart network installation via nfs. I place the following line immediately following the "install" keyword in the Kickstart file. install nfs --server=137.198.41.40 --dir=/admin/linux/scientific/304/i386 However, the system still asks me to interactively specify the nfs server and intallation directory tree. What I am doing wrong here? I don't want to be asked for the NFS server specifics, I want the nfs parameters in the kickstart file to be used instead. Thanks, -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Daniel Segall Sent: Wednesday, October 26, 2005 8:13 PM To: Discussion list about Kickstart Subject: Re: swap partition on 2nd slice Stuart J. Browne wrote: > ----- Original Message ----- From: "Stephen Mah" > To: > Sent: Thursday, October 27, 2005 10:33 AM > Subject: swap partition on 2nd slice > > >> Is there a way to make kickstart force the swap partition to the >> second slice? > > > Look at --onpart > > i.e. > > part swap --onpart=/dev/hda2 > > bkx You should also note that --onpart expects the partition to be created prior to install. -Dan _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Elizabeth.Brosch at thomson.com Thu Oct 27 17:38:56 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Thu, 27 Oct 2005 13:38:56 -0400 Subject: nfs kickstart network install Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4E79@tshuspaphimbx01.ERF.THOMSON.COM> This may not even be an issue but I have my entry as: nfs --server 10.xxx.xxx.xxx --dir /usr/osdist/rhas4 notice I do not have "=". From LAWRENCE.SORRILLO at HAMPTONU.EDU Thu Oct 27 17:42:28 2005 From: LAWRENCE.SORRILLO at HAMPTONU.EDU (LAWRENCE.SORRILLO at HAMPTONU.EDU) Date: Thu, 27 Oct 2005 13:42:28 -0400 Subject: nfs kickstart network install Message-ID: <9B4DE93284CA884FA41C11102AEB741E0410C3B3@HUMX.infotech.hamptonu.edu> Does this nfs option have to be in a particular place in the file? -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Thursday, October 27, 2005 12:39 PM To: kickstart-list at redhat.com Subject: RE: nfs kickstart network install This may not even be an issue but I have my entry as: nfs --server 10.xxx.xxx.xxx --dir /usr/osdist/rhas4 notice I do not have "=". _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Chip.Shabazian at bankofamerica.com Thu Oct 27 17:44:57 2005 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Thu, 27 Oct 2005 10:44:57 -0700 Subject: nfs kickstart network install Message-ID: I too don't use the = although the documentation says you can.... Is the ONLY interactive prompt you are getting for the nfs source? Specifically, are you getting network information prompts as well? How are you loading your kickstart? NFS? HTTP? Floppy? Embeded? The nfs option does NOT have to be in any particular place, as long as it's in the command section. In our kickstart file, it's actually the first uncommented line so that admins don't have to look past the first 10 lines of the kickstart file if they wish to make changes. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Thursday, October 27, 2005 10:39 AM To: kickstart-list at redhat.com Subject: RE: nfs kickstart network install This may not even be an issue but I have my entry as: nfs --server 10.xxx.xxx.xxx --dir /usr/osdist/rhas4 notice I do not have "=". _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Elizabeth.Brosch at thomson.com Thu Oct 27 17:46:28 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Thu, 27 Oct 2005 13:46:28 -0400 Subject: nfs kickstart network install Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E4E7B@tshuspaphimbx01.ERF.THOMSON.COM> Here are the first few lines of my ks.cfg file: nfs entry up to the point of the "install" command: # Kickstart file for Red Hat Enterprise Linux AS 4 # # Set some basic configuration lang en_US langsupport en_US keyboard us mouse genericps/2 --emulthree # System timezone timezone America/New_York # where to find the distribution ISOs nfs --server 10.xxx.xxx.xxx --dir /usr/osdist/rhas4 # Set root password rootpw --iscrypted $ # bootloader bootloader --location=mbr --append="rhgb quiet" # do an install install # Partition configuration ..... From LAWRENCE.SORRILLO at HAMPTONU.EDU Thu Oct 27 18:06:41 2005 From: LAWRENCE.SORRILLO at HAMPTONU.EDU (LAWRENCE.SORRILLO at HAMPTONU.EDU) Date: Thu, 27 Oct 2005 14:06:41 -0400 Subject: nfs kickstart network install Message-ID: <9B4DE93284CA884FA41C11102AEB741E0410C3B4@HUMX.infotech.hamptonu.edu> I boot from floppy Then at the boot prompt: linux ks=http://x.x.x.x/dirs/test.fs.cfg dd (when I do not supply a driver disk the system forces me to manually choose one as it complains that it is unable to find devices of the type needed for this installation.) when asked, I insert the drvnet.img floppy to chose the network card driver. After I choose the a driver the web server logs shows where the test.ks.cfg file is accessed. The client install still asks for the nfs server though, but all the other options as far as partitioning and package options are read from the Kickstart file correctly. Its interesting that I just realized that the NIC driver I choose at installtion time is not the one that the system uses when it is completely installed. It seems then the root of the problem is correctly supplying the correct NIC driver for the system. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip Sent: Thursday, October 27, 2005 12:45 PM To: Discussion list about Kickstart Subject: RE: nfs kickstart network install I too don't use the = although the documentation says you can.... Is the ONLY interactive prompt you are getting for the nfs source? Specifically, are you getting network information prompts as well? How are you loading your kickstart? NFS? HTTP? Floppy? Embeded? The nfs option does NOT have to be in any particular place, as long as it's in the command section. In our kickstart file, it's actually the first uncommented line so that admins don't have to look past the first 10 lines of the kickstart file if they wish to make changes. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Elizabeth.Brosch at thomson.com Sent: Thursday, October 27, 2005 10:39 AM To: kickstart-list at redhat.com Subject: RE: nfs kickstart network install This may not even be an issue but I have my entry as: nfs --server 10.xxx.xxx.xxx --dir /usr/osdist/rhas4 notice I do not have "=". _______________________________________________ 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 s.davison at computer.org Thu Oct 27 18:37:18 2005 From: s.davison at computer.org (s.davison at computer.org) Date: Thu, 27 Oct 2005 18:37:18 +0000 Subject: nfs kickstart network install Message-ID: What you are seeing sounds similar to a problem we encountered. We would get a dialog box titled "Error" with the message "That directory could not be mounted from the server." If you acknowledge the error box by pressing the space bar, you then get another dialog box titled "NFS Setup". That box has keyin fields for the server IP address and directory. The fields should contain the values from the nfs line of the kickstart file. In our case, we could use the tab key to toggle to the OK and dismiss the box, not changing either the IP address or the directory path. Anaconda would try again. Sometimes we had to go through those steps several times (up to five, worst case), but eventually anaconda would find the server and connect to it. I don't know if that's the problem you're seeing -- the key is whether you see the same values in the "NFS Setup" dialog as you specified on the nfs line in the kickstart file. If so, I suspect you are having the same problem we did. If not, I suspect a syntax problem in the kickstart file. The problem I have described above is one that is well known. It was discussed at some length on this list a few months ago. It is a bug in anaconda. Some high-speed ethernet switches go through a fairly lengthy initialization procedure when an ethernet card first starts up. Anaconda times out too soon, and gives up on trying to make the connection. You can be sure that's the problem if the connection is successfully made after you retry as described above. We dealt with the problem by hacking anaconda -- applying a patch that was provided by a contributor to this list. If that's the problem I can probably find the posting and refer you to it. Stowe Davison >-----Original Message----- >From: LAWRENCE.SORRILLO at HAMPTONU.EDU [mailto:LAWRENCE.SORRILLO at HAMPTONU.EDU] >Sent: Thursday, October 27, 2005 02:06 PM >To: kickstart-list at redhat.com >Subject: RE: nfs kickstart network install > >I boot from floppy >Then at the boot prompt: > >linux ks=http://x.x.x.x/dirs/test.fs.cfg dd (when I do not supply a >driver disk the system forces me to manually choose one as it complains >that it is unable to find devices of the type needed for this >installation.) > >when asked, I insert the drvnet.img floppy to chose the network card >driver. >After I choose the a driver the web server logs shows where the >test.ks.cfg file is accessed. > >The client install still asks for the nfs server though, but all the >other options as far as partitioning and package options are read from >the Kickstart file correctly. > >Its interesting that I just realized that the NIC driver I choose at >installtion time is not the one that the system uses when it is >completely installed. > >It seems then the root of the problem is correctly supplying the correct >NIC driver for the system. > >-----Original Message----- >From: kickstart-list-bounces at redhat.com >[mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip >Sent: Thursday, October 27, 2005 12:45 PM >To: Discussion list about Kickstart >Subject: RE: nfs kickstart network install > >I too don't use the = although the documentation says you can.... > >Is the ONLY interactive prompt you are getting for the nfs source? >Specifically, are you getting network information prompts as well? >How are you loading your kickstart? NFS? HTTP? Floppy? Embeded? > >The nfs option does NOT have to be in any particular place, as long as >it's in the command section. In our kickstart file, it's actually the >first uncommented line so that admins don't have to look past the first >10 lines of the kickstart file if they wish to make changes. > >-----Original Message----- >From: kickstart-list-bounces at redhat.com >[mailto:kickstart-list-bounces at redhat.com] On Behalf Of >Elizabeth.Brosch at thomson.com >Sent: Thursday, October 27, 2005 10:39 AM >To: kickstart-list at redhat.com >Subject: RE: nfs kickstart network install > >This may not even be an issue but I have my entry as: > >nfs --server 10.xxx.xxx.xxx --dir /usr/osdist/rhas4 > >notice I do not have "=". > > > >_______________________________________________ >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 poken77 at hotmail.com Sat Oct 29 15:49:30 2005 From: poken77 at hotmail.com (kenneth graham) Date: Sat, 29 Oct 2005 15:49:30 +0000 Subject: bootable CD or diskette for FC3, how can i do it? Message-ID: Hi guys, I am a kind of new using Linux, and I got a big task to do as soon as possible, i have been reading a lot about the kickstart but so far i can not get it working so if you have a few minute, i will really appreciated !! so, so far I have a good installation of FC3 and the ks.cfg file that anaconda generated during my installation, so I have also 5 servers in the netwotk with debian :( and I must change them to FC3 with the same package and everything like the one i have ready now. the installation must be with ks=http://server/directory/ (via http) 1. could you explaind to me STEP by STEP how can i do a good bootable CD and be able to do a kickstart installation using a http server to take the packages and/or the ks.cfg file ? (i have tried it twice, but still not working, i thing i am missing something :) ) 2. do i need to copy the 3 CDs of FC3 in the http server? 3. I hace the DHCP server installed , do i really need it? 4. is it possible to do the kickstart installation with a diskette? if yes, how? thanks in advance for you help, guys ...I really need this, my work depend on it and i am tired of using windows as server or desktop, i am moving to linux thanks bye From phr at doc.ic.ac.uk Sun Oct 30 01:20:23 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Sun, 30 Oct 2005 01:20:23 +0000 (GMT) Subject: bootable CD or diskette for FC3, how can i do it? In-Reply-To: References: Message-ID: On Sat, 29 Oct 2005, kenneth graham wrote: >1. could you explaind to me STEP by STEP how can i do a good bootable CD and >be able to do a kickstart installation using a http server to take the >packages and/or the ks.cfg file ? (i have tried it twice, but still not >working, i thing i am missing something :) ) What did you try? What happened? What error messages did anaconda display (check Alt+F2, Alt+F3 etc. virtual consoles) >2. do i need to copy the 3 CDs of FC3 in the http server? (4 CDs.) Yes. In theory, if you don't specify any packages from, say, CD2 then nothing will complain that those RPMs are missing. The README file in the base of the Fedora tree explains how to copy the CDs over one another. >3. I hace the DHCP server installed , do i really need it? Not if you specify static IP addresses everywhere. >4. is it possible to do the kickstart installation with a diskette? if yes, >how? It's possible to put the kickstart file on a floppy, but you can't boot from one as the install images have grown too large in recent releases. An alternative I'd recommend looking into if you'll be doing a lot of installs is network booting, where all the configuration can live on the server and there's no need to create custom floppies or CDs per host or per subnet. Cheers, Phil From poken77 at hotmail.com Sun Oct 30 22:02:06 2005 From: poken77 at hotmail.com (kenneth graham) Date: Sun, 30 Oct 2005 22:02:06 +0000 Subject: bootable CD or diskette for FC3, how can i do it? In-Reply-To: Message-ID: thanks Phil for your awnsers and your time actually what i tried is to put a small kernel in a CD and from it, changing the BIOS , but that was a bad idea, there were panic errors etc...i dont really remember , and i dont have the CDs anymore!! can i boot form CD 1 of FC3 and when I get the "boot> " wirte : boot> linux ks=http://some.server/ks.cfg ip=1.2.3.4 netmask=255.255.255.128 gateway=1.2.3.1 dns=1.2.3.2 and after take the CD out, because i suppose that he will take the packages that I specified inside the ks.cfg from the http server? i have many servers to do the same, so I think i will put everything in the http server, with different ks.cfg do i need to have all of the packages in the same directory? thanks again, have a good day > >1. could you explaind to me STEP by STEP how can i do a good bootable CD >and > >be able to do a kickstart installation using a http server to take the > >packages and/or the ks.cfg file ? (i have tried it twice, but still not > >working, i thing i am missing something :) ) > >What did you try? What happened? What error messages did anaconda >display (check Alt+F2, Alt+F3 etc. virtual consoles) > > >2. do i need to copy the 3 CDs of FC3 in the http server? > >(4 CDs.) Yes. In theory, if you don't specify any packages from, say, >CD2 then nothing will complain that those RPMs are missing. The README >file in the base of the Fedora tree explains how to copy the CDs over >one another. > > >3. I hace the DHCP server installed , do i really need it? > >Not if you specify static IP addresses everywhere. > > >4. is it possible to do the kickstart installation with a diskette? if >yes, > >how? > >It's possible to put the kickstart file on a floppy, but you can't boot >from one as the install images have grown too large in recent releases. > >An alternative I'd recommend looking into if you'll be doing a lot of >installs is network booting, where all the configuration can live on the >server and there's no need to create custom floppies or CDs per host or >per subnet. > > >Cheers, >Phil > >_______________________________________________ >Kickstart-list mailing list >Kickstart-list at redhat.com >https://www.redhat.com/mailman/listinfo/kickstart-list From sjbrowne at bluebottle.com Sun Oct 30 22:36:12 2005 From: sjbrowne at bluebottle.com (Stuart J. Browne) Date: Mon, 31 Oct 2005 09:36:12 +1100 Subject: bootable CD or diskette for FC3, how can i do it? References: Message-ID: <099701c5dda3$a8629d50$271aa8c0@bluebottle.com> > can i boot form CD 1 of FC3 and when I get the "boot> " wirte : > > boot> linux ks=http://some.server/ks.cfg ip=1.2.3.4 > netmask=255.255.255.128 gateway=1.2.3.1 dns=1.2.3.2 You said you have a DHCP server, why not use it? anyway.. > and after take the CD out, because i suppose that he will take the > packages that I specified inside the ks.cfg from the http server? I can't remember if it tries to access the boot media again, but it will take the packages from the source defined in the ks.cfg file, where ever it came from, i.e. I used to use a HTTP ks, with NFS packages. > i have many servers to do the same, so I think i will put everything in > the http server, with different ks.cfg *nod* good. > do i need to have all of the packages in the same directory? There are two ways. Yes, one is to put all the packages in the Fedora/RPMS/ structure (as it appears on the CD's), so all the RPM's end up in the same directory. The other involves using ISO files, but that isn't available for all file-transfer methods. Reading the documentation on setting up a KS server would help: http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/custom-guide/ch-kickstart2.html (part 8+9) and http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/install-guide/s1-begininstall-net.html. >> >4. is it possible to do the kickstart installation with a diskette? if >>yes, how? >> >>It's possible to put the kickstart file on a floppy, but you can't boot >>from one as the install images have grown too large in recent releases. What I did was to create a custom bootable CD with update network and IO drivers on it for my systems, but I was using dynamically generated ks configurations, and DHCP in order to automate the entire process. >>An alternative I'd recommend looking into if you'll be doing a lot of >>installs is network booting, where all the configuration can live on the >>server and there's no need to create custom floppies or CDs per host or >>per subnet. bkx From poken77 at hotmail.com Mon Oct 31 02:31:52 2005 From: poken77 at hotmail.com (kenneth graham) Date: Mon, 31 Oct 2005 02:31:52 +0000 Subject: bootable CD or diskette for FC3, how can i do it? In-Reply-To: <099701c5dda3$a8629d50$271aa8c0@bluebottle.com> Message-ID: hi Stuart ... may I know how you did the your custom CD bootable ? but please give me some details of HOW-TO like for dummies :) thanks everybody for your time Kenneth >From: "Stuart J. Browne" >Reply-To: Discussion list about Kickstart >To: "Discussion list about Kickstart" >Subject: Re: bootable CD or diskette for FC3, how can i do it? >Date: Mon, 31 Oct 2005 09:36:12 +1100 > >>can i boot form CD 1 of FC3 and when I get the "boot> " wirte : >> >>boot> linux ks=http://some.server/ks.cfg ip=1.2.3.4 >>netmask=255.255.255.128 gateway=1.2.3.1 dns=1.2.3.2 > >You said you have a DHCP server, why not use it? anyway.. > >>and after take the CD out, because i suppose that he will take the >>packages that I specified inside the ks.cfg from the http server? > >I can't remember if it tries to access the boot media again, but it will >take the packages from the source defined in the ks.cfg file, where ever it >came from, i.e. I used to use a HTTP ks, with NFS packages. > >>i have many servers to do the same, so I think i will put everything in >>the http server, with different ks.cfg > >*nod* good. > >>do i need to have all of the packages in the same directory? > >There are two ways. Yes, one is to put all the packages in the >Fedora/RPMS/ structure (as it appears on the CD's), so all the RPM's end up >in the same directory. The other involves using ISO files, but that isn't >available for all file-transfer methods. > >Reading the documentation on setting up a KS server would help: > >http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/custom-guide/ch-kickstart2.html >(part 8+9) and >http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/install-guide/s1-begininstall-net.html. > >>> >4. is it possible to do the kickstart installation with a diskette? if >>>yes, how? >>> >>>It's possible to put the kickstart file on a floppy, but you can't boot >>>from one as the install images have grown too large in recent releases. > >What I did was to create a custom bootable CD with update network and IO >drivers on it for my systems, but I was using dynamically generated ks >configurations, and DHCP in order to automate the entire process. > >>>An alternative I'd recommend looking into if you'll be doing a lot of >>>installs is network booting, where all the configuration can live on the >>>server and there's no need to create custom floppies or CDs per host or >>>per subnet. > >bkx > >_______________________________________________ >Kickstart-list mailing list >Kickstart-list at redhat.com >https://www.redhat.com/mailman/listinfo/kickstart-list From sjbrowne at bluebottle.com Mon Oct 31 02:56:13 2005 From: sjbrowne at bluebottle.com (Stuart J. Browne) Date: Mon, 31 Oct 2005 13:56:13 +1100 Subject: bootable CD or diskette for FC3, how can i do it? References: Message-ID: <008e01c5ddc6$a91b02f0$271aa8c0@bluebottle.com> > may I know how you did the your custom CD bootable ? but please give me > some details of > HOW-TO like for dummies :) Umm, this was a while ago at a different company, so this'll be from memory.. so please bear with me ;) ... ... ... FC4 is easier than what I did.. ;) They provide an ISO image so you can burn your own bootable CD. /images/boot.iso is your friend ;) If you want to customize it, you should be able to loop-mount it, and make modifications. If you want to fully customize it, the isolinux instructions (http://syslinux.zytor.com/iso.php) will help. But yea, 'cd /tmp;mkdir boot;mount -oloop boot.iso boot/', remove some of the msg files that aren't needed, customize 'isolinux.cfg', throw a 'ks.cfg' in to the fray, and away you go. bkx > thanks everybody for your time > Kenneth > >>>can i boot form CD 1 of FC3 and when I get the "boot> " wirte : >>> >>>boot> linux ks=http://some.server/ks.cfg ip=1.2.3.4 >>>netmask=255.255.255.128 gateway=1.2.3.1 dns=1.2.3.2 >> >>You said you have a DHCP server, why not use it? anyway.. >> >>>and after take the CD out, because i suppose that he will take the >>>packages that I specified inside the ks.cfg from the http server? >> >>I can't remember if it tries to access the boot media again, but it will >>take the packages from the source defined in the ks.cfg file, where ever >>it came from, i.e. I used to use a HTTP ks, with NFS packages. >> >>>i have many servers to do the same, so I think i will put everything in >>>the http server, with different ks.cfg >> >>*nod* good. >> >>>do i need to have all of the packages in the same directory? >> >>There are two ways. Yes, one is to put all the packages in the >>Fedora/RPMS/ structure (as it appears on the CD's), so all the RPM's end >>up in the same directory. The other involves using ISO files, but that >>isn't available for all file-transfer methods. >> >>Reading the documentation on setting up a KS server would help: >> >>http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/custom-guide/ch-kickstart2.html >>(part 8+9) and >>http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/install-guide/s1-begininstall-net.html. >> >>>> >4. is it possible to do the kickstart installation with a diskette? if >>>>yes, how? >>>> >>>>It's possible to put the kickstart file on a floppy, but you can't boot >>>>from one as the install images have grown too large in recent releases. >> >>What I did was to create a custom bootable CD with update network and IO >>drivers on it for my systems, but I was using dynamically generated ks >>configurations, and DHCP in order to automate the entire process. >> >>>>An alternative I'd recommend looking into if you'll be doing a lot of >>>>installs is network booting, where all the configuration can live on the >>>>server and there's no need to create custom floppies or CDs per host or >>>>per subnet. >> >>bkx >> >>_______________________________________________ >>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 atagarsi at cgg.com Mon Oct 31 10:41:12 2005 From: atagarsi at cgg.com (tagarsi, arun) Date: Mon, 31 Oct 2005 10:41:12 -0000 Subject: bootable CD or diskette for FC3, how can i do it? Message-ID: <867A31BCCABE274D9C6C3CF547F55665280488@uk02ex01.int.cgg.com> Hello Kenneth, One other way is to use PXE boot and kickstart, this is probably more useful if you have lots of nodes to build. Just reading it may help you decide what you want to do. http://www.stanford.edu/~alfw/PXE-Kickstart/PXE-Kickstart.html This link is a good place to start I think. Thanks Arun -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com]On Behalf Of kenneth graham Sent: Monday, October 31, 2005 2:32 AM To: kickstart-list at redhat.com Subject: Re: bootable CD or diskette for FC3, how can i do it? hi Stuart ... may I know how you did the your custom CD bootable ? but please give me some details of HOW-TO like for dummies :) thanks everybody for your time Kenneth >From: "Stuart J. Browne" >Reply-To: Discussion list about Kickstart >To: "Discussion list about Kickstart" >Subject: Re: bootable CD or diskette for FC3, how can i do it? >Date: Mon, 31 Oct 2005 09:36:12 +1100 > >>can i boot form CD 1 of FC3 and when I get the "boot> " wirte : >> >>boot> linux ks=http://some.server/ks.cfg ip=1.2.3.4 >>netmask=255.255.255.128 gateway=1.2.3.1 dns=1.2.3.2 > >You said you have a DHCP server, why not use it? anyway.. > >>and after take the CD out, because i suppose that he will take the >>packages that I specified inside the ks.cfg from the http server? > >I can't remember if it tries to access the boot media again, but it will >take the packages from the source defined in the ks.cfg file, where ever it >came from, i.e. I used to use a HTTP ks, with NFS packages. > >>i have many servers to do the same, so I think i will put everything in >>the http server, with different ks.cfg > >*nod* good. > >>do i need to have all of the packages in the same directory? > >There are two ways. Yes, one is to put all the packages in the >Fedora/RPMS/ structure (as it appears on the CD's), so all the RPM's end up >in the same directory. The other involves using ISO files, but that isn't >available for all file-transfer methods. > >Reading the documentation on setting up a KS server would help: > >http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/custom-guide/ch-kickstart2.html >(part 8+9) and >http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/install-guide/s1-begininstall-net.html. > >>> >4. is it possible to do the kickstart installation with a diskette? if >>>yes, how? >>> >>>It's possible to put the kickstart file on a floppy, but you can't boot >>>from one as the install images have grown too large in recent releases. > >What I did was to create a custom bootable CD with update network and IO >drivers on it for my systems, but I was using dynamically generated ks >configurations, and DHCP in order to automate the entire process. > >>>An alternative I'd recommend looking into if you'll be doing a lot of >>>installs is network booting, where all the configuration can live on the >>>server and there's no need to create custom floppies or CDs per host or >>>per subnet. > >bkx > >_______________________________________________ >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 LLavender at kmg.com Mon Oct 31 14:00:11 2005 From: LLavender at kmg.com (Lavender, Leo) Date: Mon, 31 Oct 2005 08:00:11 -0600 Subject: bootable CD or diskette for FC3, how can i do it? Message-ID: <044BBA9ACDF3D948BC41F0DD8DFADF314F2472@houexg3.kmg.com> (Pardon me If I say something dumb, I am a new listener on this list.) Since you already have Linux partitions on the servers you can just put initrd.img, and vmlinuz from your new distribution CD images/pxeboot directory on to one of the existing partitions; modify grub.conf to point to that partition with a ks=nfs:ipnum:/whatever on end of the kernel line; and reboot. This can be done from an nfs mounted script which also modifies ks files for upgrading a herd of boxes without visiting them. Leo -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of tagarsi, arun Sent: Monday, October 31, 2005 4:41 AM To: Discussion list about Kickstart Subject: RE: bootable CD or diskette for FC3, how can i do it? Hello Kenneth, One other way is to use PXE boot and kickstart, this is probably more useful if you have lots of nodes to build. Just reading it may help you decide what you want to do. http://www.stanford.edu/~alfw/PXE-Kickstart/PXE-Kickstart.html This link is a good place to start I think. Thanks Arun -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com]On Behalf Of kenneth graham Sent: Monday, October 31, 2005 2:32 AM To: kickstart-list at redhat.com Subject: Re: bootable CD or diskette for FC3, how can i do it? hi Stuart ... may I know how you did the your custom CD bootable ? but please give me some details of HOW-TO like for dummies :) thanks everybody for your time Kenneth >From: "Stuart J. Browne" >Reply-To: Discussion list about Kickstart >To: "Discussion list about Kickstart" >Subject: Re: bootable CD or diskette for FC3, how can i do it? >Date: Mon, 31 Oct 2005 09:36:12 +1100 > >>can i boot form CD 1 of FC3 and when I get the "boot> " wirte : >> >>boot> linux ks=http://some.server/ks.cfg ip=1.2.3.4 >>netmask=255.255.255.128 gateway=1.2.3.1 dns=1.2.3.2 > >You said you have a DHCP server, why not use it? anyway.. > >>and after take the CD out, because i suppose that he will take the >>packages that I specified inside the ks.cfg from the http server? > >I can't remember if it tries to access the boot media again, but it >will >take the packages from the source defined in the ks.cfg file, where ever it >came from, i.e. I used to use a HTTP ks, with NFS packages. > >>i have many servers to do the same, so I think i will put everything >>in >>the http server, with different ks.cfg > >*nod* good. > >>do i need to have all of the packages in the same directory? > >There are two ways. Yes, one is to put all the packages in the >Fedora/RPMS/ structure (as it appears on the CD's), so all the RPM's end up >in the same directory. The other involves using ISO files, but that isn't >available for all file-transfer methods. > >Reading the documentation on setting up a KS server would help: > >http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/custom-guide/ch-k >ickstart2.html >(part 8+9) and >http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/install-guide/s1- begininstall-net.html. > >>> >4. is it possible to do the kickstart installation with a diskette? >>> >if >>>yes, how? >>> >>>It's possible to put the kickstart file on a floppy, but you can't >>>boot from one as the install images have grown too large in recent >>>releases. > >What I did was to create a custom bootable CD with update network and >IO >drivers on it for my systems, but I was using dynamically generated ks >configurations, and DHCP in order to automate the entire process. > >>>An alternative I'd recommend looking into if you'll be doing a lot of >>>installs is network booting, where all the configuration can live on >>>the server and there's no need to create custom floppies or CDs per >>>host or per subnet. > >bkx > >_______________________________________________ >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