From tom at ng23.net Wed Jul 1 17:02:30 2009 From: tom at ng23.net (Tom Brown) Date: Wed, 01 Jul 2009 18:02:30 +0100 Subject: Kickstart not rebooting on completion In-Reply-To: <4A495D92.6020303@yahoo.com> References: <06A6610D4F464D4EBEAFBF2C5F86911E010D6AF0@exchange2.columbia.tresys.com> <4A40EFEC.9070809@rice.edu> <06A6610D4F464D4EBEAFBF2C5F86911E01242990@exchange2.columbia.tresys.com> <4A495D92.6020303@yahoo.com> Message-ID: <4A4B96A6.3010200@ng23.net> Matthew Milliss wrote: > I'm using kickstart to install Fedora 11 and I'm having an issue with > the system not rebooting when the install has finished. The install is > running in text mode and it just stops on the "Running post install" > screen. If I force a reboot the install has been successful, but I > find myself wasting time waiting around to make sure the install has > finished before forcing a reboot. My kickstart file is below. Any help > would be greatly appreciated. > > # Kickstart file automatically generated by anaconda - with some > additions by Matt. > #version=F11 > install > text > cdrom > lang en_AU.UTF-8 > keyboard us > network --device=eth0 --bootproto=dhcp --hostname=snotrod > xconfig --startxonboot > rootpw --iscrypted > firewall --disabled > authconfig --enableshadow --passalgo=sha512 --enablefingerprint > selinux --permissive > timezone --utc Australia/Sydney > bootloader --location=mbr --driveorder=sda --append="quiet reboot=bios" > autopart > clearpart --all > reboot --eject > > repo --name=msl --baseurl=http://hannibal/tyroyummy/msl > > %packages > @admin-tools > @base > @core > @editors > @fonts > @gnome-desktop > @graphical-internet > @graphics > @hardware-support > @input-methods > @office > @online-docs > @printing > @sound-and-video > @base-x > xfsprogs > mtools > gpgme > pax > gnupg2 > gvfs-obexftp > ImageMagick > dcraw > netpbm-progs > hdparm > m17n-db-assamese > gok > m17n-db-tamil > m17n-db-gujarati > iok > m17n-db-hindi > m17n-db-punjabi > m17n-db-oriya > m17n-db-telugu > m17n-db-bengali > m17n-db-malayalam > m17n-contrib-sinhala > m17n-db-kannada > gdm > nvu > binutils > rdesktop > wireshark > wireshark-gnome > subversion > vnc > minicom > thunderbird > rpm-build > kernel-headers > createrepo > kernel-devel > yelp > jedit > vsftpd > AdobeReader_enu > flash-plugin > nxclient > nxnode > nxserver > ntp > %end > > %post > #!/bin/bash > env > /root/postks.log > mkdir /mnt/pub > mkdir /mnt/yum > mount -o ro,nolock -t nfs hannibal:/tyroyummy /mnt/yum > #mount -o ro,nolock -t nfs 192.168.100.58:/home/pub/ /mnt/pub > cp /etc/fstab /etc/fstab.bk > echo "avalon-2:/home/pub /mnt/pub nfs auto,rw,intr > 0 0" >> /etc/fstab > echo "hannibal:/tyroyummy /mnt/yum nfs auto,rw,intr > 0 0" >> /etc/fstab > cd /mnt/yum/SOE_Fedora11/install > ./install.sh > /var/log/msl_install_output.log > cd / > umount /mnt/yum > %end > > what does alt+f2 etc show you its waiting for From jimi at sngx.net Wed Jul 1 21:49:40 2009 From: jimi at sngx.net (James Cammarata) Date: Wed, 01 Jul 2009 16:49:40 -0500 Subject: Problem with kickstarts on rhel4u6 Message-ID: <63af20d2ce0440873880186136a12b23@sngx.net> Hi all, I'm having a problem with kickstarts on with RHEL4u6. The kickstart I'm using is essentially the default one provided with cobbler, with a few extras added in. The problem manifests as follows: 1) Kickstart begins, completes phase 1 (or what I call phase 1 anyway) 2) After it prints "...starting text mode", the install pauses. Switching to VTY 3, I see that it is sitting at a parted prompt. 3) Typing quit at the parted prompt causes the kickstart to proceed. Has anyone seen this behaviour before? Is this something specific to RHEL4u6? I don't think the issue lies with the kickstart file, as I use essentially the same template (especially the partitioning snippets) for RHEL5 without issue. Thanks, James C. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From co55-sy1t at dea.spamcon.org Thu Jul 2 06:19:00 2009 From: co55-sy1t at dea.spamcon.org (whitivery) Date: Wed, 01 Jul 2009 23:19:00 -0700 Subject: Anaconda early startup failure - traceback Message-ID: We have a working Kickstart server for CentOS 4.4, building targets running CentOS 4.4. I built a new server with CentOS 5.3, to build CentOS 5.3 targets, adapting for the changes I noticed as I went along. The new server boots and runs but won't build a target unit. When I PXE-boot a target unit, I get the boot menu and pick an entry, it starts the early part fine, loads stage2.img OK, shows a couple screens then a "Retrieving installation information" dialog/progress bar, and just when/after that gets to 100%, it halts with a traceback. I can't tell from the info shown what's wrong. I simplified the kickstart file a lot, removing custom pre/post stuff, and it still halts at the same point. I have placed files and information from /tmp here: anaconda.log: anacdump.txt: ks.cfg: syslog: ls -l /tmp: From kpowell at redhat.com Thu Jul 2 14:54:34 2009 From: kpowell at redhat.com (Kyle Powell) Date: Thu, 02 Jul 2009 10:54:34 -0400 Subject: Anaconda early startup failure - traceback In-Reply-To: References: Message-ID: <4A4CCA2A.5080602@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 whitivery wrote: > We have a working Kickstart server for CentOS 4.4, building > targets running CentOS 4.4. > > I built a new server with CentOS 5.3, to build CentOS 5.3 > targets, adapting for the changes I noticed as I went along. The > new server boots and runs but won't build a target unit. > > When I PXE-boot a target unit, I get the boot menu and pick an > entry, it starts the early part fine, loads stage2.img OK, shows > a couple screens then a "Retrieving installation information" > dialog/progress bar, and just when/after that gets to 100%, it > halts with a traceback. Looks like you need more RAM: http://www.centos.org/modules/newbb/viewtopic.php?topic_id=19367&forum=37 Same thing with Fedora: http://bugzilla.redhat.com/show_bug.cgi?id=499585 Minimum RAM requirement for RHEL 5 is 512MB. I assume that applies to CentOS as well: http://www.redhat.com/rhel/compare/ - -- Kyle Powell | Red Hat | Senior Consultant, RHCE -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFKTMoq7pTtanQdBU4RAlbuAJ0YM/q+6FlIz57HXoxOtdDmxh1j/wCfT1wk P2epw5YPq5DZexMyg74mf4k= =ijZD -----END PGP SIGNATURE----- From jimi at sngx.net Thu Jul 2 15:20:51 2009 From: jimi at sngx.net (James Cammarata) Date: Thu, 02 Jul 2009 10:20:51 -0500 Subject: Anaconda early startup failure - traceback In-Reply-To: <4A4CCA2A.5080602@redhat.com> References: <4A4CCA2A.5080602@redhat.com> Message-ID: <63ed34d13bf23ec4eac27a14215f3150@sngx.net> On Thu, 02 Jul 2009 10:54:34 -0400, Kyle Powell wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > whitivery wrote: >> We have a working Kickstart server for CentOS 4.4, building >> targets running CentOS 4.4. >> >> I built a new server with CentOS 5.3, to build CentOS 5.3 >> targets, adapting for the changes I noticed as I went along. The >> new server boots and runs but won't build a target unit. >> >> When I PXE-boot a target unit, I get the boot menu and pick an >> entry, it starts the early part fine, loads stage2.img OK, shows >> a couple screens then a "Retrieving installation information" >> dialog/progress bar, and just when/after that gets to 100%, it >> halts with a traceback. > > Looks like you need more RAM: > http://www.centos.org/modules/newbb/viewtopic.php?topic_id=19367&forum=37 > > Same thing with Fedora: > http://bugzilla.redhat.com/show_bug.cgi?id=499585 > > Minimum RAM requirement for RHEL 5 is 512MB. I assume that applies to > CentOS as > well: > http://www.redhat.com/rhel/compare/ Yes I've seen this same error on systems with insufficient ram (usually VM's). The key is that it fails on the lspci call, which is exactly where all of mine would fail. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From bernasconi.pablo at gmail.com Thu Jul 2 16:08:58 2009 From: bernasconi.pablo at gmail.com (Pablo Bernasconi) Date: Thu, 2 Jul 2009 13:08:58 -0300 Subject: Kickstart not rebooting on completion Message-ID: <272d5daf0907020908t7b50cf44r2fd519a95817bd19@mail.gmail.com> Hi Tom, Try this: # Kickstart file automatically generated by anaconda - with some additions by Matt. #version=F11 install text cdrom lang en_AU.UTF-8 keyboard us network --device=eth0 --bootproto=dhcp --hostname=snotrod xconfig --startxonboot rootpw --iscrypted firewall --disabled authconfig --enableshadow --passalgo=sha512 --enablefingerprint selinux --permissive timezone --utc Australia/Sydney bootloader --location=mbr --driveorder=sda --append="quiet reboot=bios" autopart clearpart --all *#reboot --eject reboot* -------------- next part -------------- An HTML attachment was scrubbed... URL: From co55-sy1t at dea.spamcon.org Fri Jul 3 08:41:48 2009 From: co55-sy1t at dea.spamcon.org (whitivery) Date: Fri, 03 Jul 2009 01:41:48 -0700 Subject: Anaconda early startup failure - traceback References: <4A4CCA2A.5080602@redhat.com> <63ed34d13bf23ec4eac27a14215f3150@sngx.net> Message-ID: <0tgr45t1emneprkctp6rjbpc4kad2cjeup@4ax.com> James Cammarata wrote: > >On Thu, 02 Jul 2009 10:54:34 -0400, Kyle Powell wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> whitivery wrote: >>> We have a working Kickstart server for CentOS 4.4, building >>> targets running CentOS 4.4. >>> >>> I built a new server with CentOS 5.3, to build CentOS 5.3 >>> targets, adapting for the changes I noticed as I went along. The >>> new server boots and runs but won't build a target unit. >>> >>> When I PXE-boot a target unit, I get the boot menu and pick an >>> entry, it starts the early part fine, loads stage2.img OK, shows >>> a couple screens then a "Retrieving installation information" >>> dialog/progress bar, and just when/after that gets to 100%, it >>> halts with a traceback. >> >> Looks like you need more RAM: >> http://www.centos.org/modules/newbb/viewtopic.php?topic_id=19367&forum=37 >> >> Same thing with Fedora: >> http://bugzilla.redhat.com/show_bug.cgi?id=499585 >> >> Minimum RAM requirement for RHEL 5 is 512MB. I assume that applies to >> CentOS as >> well: >> http://www.redhat.com/rhel/compare/ > >Yes I've seen this same error on systems with insufficient ram (usually >VM's). The key is that it fails on the lspci call, which is exactly where >all of mine would fail. Thanks to you and Kyle for the answers - I tried another box with 512M RAM and it worked fine. Seems nuts that hundreds of millions of bytes of memory can't even start a text mode install. With the box that worked, I still see some warnings in the anaconda.log like the following - do I need some additional kickstart settings or something to avoid these?: I don't need DNS hostname lookup so why is it trying?: 21:00:01 ERROR : no DNS servers, can't look up hostname 21:00:01 ERROR : no DNS servers, can't look up hostname I specified "url" so why these messages?: 21:00:01 ERROR : got to setupCdrom without a CD device 21:00:22 WARNING : no floppy devices found but we'll try fd0 anyway I specified "text" so why does it say "not available" - it's available, I just prefer text install: 21:00:25 WARNING : Graphical installation not available... Starting text mode. What does this mean?: 21:00:28 WARNING : Unable to find temp path, going to use ramfs path Why these?: 21:00:32 WARNING : step installtype does not exist 21:00:32 WARNING : step complete does not exist 21:00:32 WARNING : step complete does not exist 21:00:32 WARNING : step complete does not exist 21:00:32 WARNING : step complete does not exist 21:00:32 WARNING : step complete does not exist 21:00:32 WARNING : step complete does not exist 21:00:32 WARNING : step complete does not exist 21:00:32 WARNING : step complete does not exist 21:00:32 WARNING : step complete does not exist 21:00:32 WARNING : step complete does not exist 21:00:32 WARNING : step complete does not exist 21:00:32 WARNING : step complete does not exist 21:00:32 WARNING : step complete does not exist These seem to be bugs: 21:00:36 WARNING : /usr/lib/python2.4/site-packages/snack.py:250: DeprecationWarning: integer argument expected, got float self.w = _snack.scale(width, total) 21:00:36 WARNING : /usr/lib/python2.4/site-packages/snack.py:247: DeprecationWarning: integer argument expected, got float self.w.scaleSet(amount) 21:00:48 WARNING : /usr/lib/python2.4/site-packages/yum/repos.py:68: RuntimeWarning: YumProgress.progressbar called when popped self.callback.progressbar(num, len(repos), repo.id) In addition, there are numerous "WARNING: doesn't exist" scattered in the log near the end of the install. From loudredz71 at yahoo.com Mon Jul 6 23:00:48 2009 From: loudredz71 at yahoo.com (Jonathan Horne) Date: Mon, 6 Jul 2009 16:00:48 -0700 (PDT) Subject: disabling a driver during kickstart Message-ID: <114299.27171.qm@web45314.mail.sp1.yahoo.com> Greetings, Can someone tell me how to get kickstart for RHEL5x/CentOS5x to ignore multipath/SAN during system install? the other day multipath kicked in automatically during an install and i overwrote data on a LUN and the next thing i know all my LVMs now live out on the SAN. The only way i know of to get around this is to either pull the fiber before the install or get in the switch fabric and de-zone the port. it just seems logical that there would be a way to not destroy your LUNs during a scripted kickstart install. Can anyone shed light on this for me, or point me in the right direction? thanks, Jonathan Horne From larry.brigman at gmail.com Mon Jul 6 23:14:03 2009 From: larry.brigman at gmail.com (Larry Brigman) Date: Mon, 6 Jul 2009 16:14:03 -0700 Subject: disabling a driver during kickstart In-Reply-To: <114299.27171.qm@web45314.mail.sp1.yahoo.com> References: <114299.27171.qm@web45314.mail.sp1.yahoo.com> Message-ID: On Mon, Jul 6, 2009 at 4:00 PM, Jonathan Horne wrote: > > Greetings, > > Can someone tell me how to get kickstart for RHEL5x/CentOS5x to ignore multipath/SAN during system install? ?the other day multipath kicked in automatically during an install and i overwrote data on a LUN and the next thing i know all my LVMs now live out on the SAN. ?The only way i know of to get around this is to either pull the fiber before the install or get in the switch fabric and de-zone the port. > > it just seems logical that there would be a way to not destroy your LUNs during a scripted kickstart install. ?Can anyone shed light on this for me, or point me in the right direction? > We have something similar where we re-install a system but want to keep the contents of a configured software array. We use the %pre script to write out the partitioning file that is included. Our scripts have a lot of other details in them but here are the basic lines that should get you started. %pre #!/bin/sh outfile=/tmp/part-include all_dev_names=$(cd /sys/block; echo sd* ) cat <$outfile clearpart --drives=${install_dev} --all --initlabel ignoredisk --drives=$(echo "${all_dev_names/$install_dev/}" | tr ' ' ,) part /boot --fstype="ext2" ${install_dev:+--ondisk=$install_dev} --size=$boot_size part pv.01 --size=0 --grow ${install_dev:+--ondisk=$install_dev} volgroup VG00 pv.01 logvol / --vgname=VG00 --size=$root_size --name=root1 --fstype=ext3 --grow logvol /alt --vgname=VG00 --size=$root_size --name=root2 --fstype=ext3 --grow logvol /logging --vgname=VG00 --size=$log_size --name=log --fstype=ext3 --grow bootloader --append="crashkernel=128M at 16M" --location=mbr ${install_dev:+--driveorder=$install_dev} EOF From kpowell at redhat.com Tue Jul 7 03:35:08 2009 From: kpowell at redhat.com (Kyle Powell) Date: Mon, 06 Jul 2009 23:35:08 -0400 Subject: disabling a driver during kickstart In-Reply-To: <114299.27171.qm@web45314.mail.sp1.yahoo.com> References: <114299.27171.qm@web45314.mail.sp1.yahoo.com> Message-ID: <4A52C26C.6020208@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Jonathan Horne wrote: > Greetings, > > Can someone tell me how to get kickstart for RHEL5x/CentOS5x to ignore multipath/SAN during system install? the other day multipath kicked in automatically during an install and i overwrote data on a LUN and the next thing i know all my LVMs now live out on the SAN. The only way i know of to get around this is to either pull the fiber before the install or get in the switch fabric and de-zone the port. > > it just seems logical that there would be a way to not destroy your LUNs during a scripted kickstart install. Can anyone shed light on this for me, or point me in the right direction? > > thanks, > Jonathan Horne Hi Jonathan, The "latefcload" boot option has been around since RHEL 3 U6. http://www.redhat.com/docs/manuals/enterprise/RHEL-3-Manual/release-notes/as-x86/RELEASE-NOTES-U6-x86-en.html "Systems using storage attached to Fibre Channel host bus adapters (FC HBAs) may encounter issues during installation due to PCI device load ordering. To address this issue, a new boot option called latefcload was developed to delay the loading of FC HBAs until other storage devices have been loaded." The other, older solution I've seen used is specifying the "nostorage" boot option and then adding a "device scsi " to the kickstart file. If it works for you, I'd recommend the first option. Far more portable. - -- Kyle Powell | Red Hat | Senior Consultant, RHCE -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFKUsJs7pTtanQdBU4RAscrAJ4r5LD5JIcG2Cph3yWUsJFKF67R6QCfTiBQ RDS3XwBTJAAysoAJgyW+ymo= =oZB4 -----END PGP SIGNATURE----- From Floydsmith at aol.com Tue Jul 7 07:50:43 2009 From: Floydsmith at aol.com (Floydsmith at aol.com) Date: Tue, 7 Jul 2009 03:50:43 EDT Subject: core 11 - can't find dir /mnt/runtime/etc wjth harddrive type install media Message-ID: I downloaded fedora core 11 yesterday (070609) and put the image in partition /dev/sda4 and dir /070609/images/install.img. This image kickstart installs when burned onto a DVD. But when the same kickstart file is used except for line: harddrive --dir=/070609/images/install.img --partition=sda4 then VT3 says that it found the stage2 install.img file but then immediately freezses up. There is no messages at all in the VT3 console but back in the VT1 there is an error box saying that dir /mnt/runtime/etc can't be found. Any help greatly appreciated in advance. Floyd, ************** Looking for love this summer? Find it now on AOL Personals. (http://personals.aol.com/?ncid=emlcntuslove00000003) -------------- next part -------------- An HTML attachment was scrubbed... URL: From linux.mether at gmail.com Tue Jul 7 09:55:32 2009 From: linux.mether at gmail.com (Linux Geek) Date: Tue, 7 Jul 2009 15:25:32 +0530 Subject: PCI Devices ( NIC ) Enumerated on RHEL5 ( latest ) Message-ID: Hello, I am trying to install RHEL5.3 on a system with multiple Network interfaces. While installation the network device names are not read in correct/expected order. I set the ksdevice as eth0 ( using ksdevice=eth0 at boot prompt ) however the installer tries to acquire IP for eth2 device. I have tried using "pci=bfsort" with now luck which I believe is due its applicability on few systems. Disabling devices in BIOS din't help either. Is there any way to get this wroking ? Thanks. -------------- next part -------------- An HTML attachment was scrubbed... URL: From hhoffman at ip-solutions.net Tue Jul 7 10:56:32 2009 From: hhoffman at ip-solutions.net (Harry Hoffman) Date: Tue, 07 Jul 2009 06:56:32 -0400 Subject: PCI Devices ( NIC ) Enumerated on RHEL5 ( latest ) In-Reply-To: References: Message-ID: <4A5329E0.7010107@ip-solutions.net> Hi LG, You might be able to setup the ethernet devices via a bootparam. Check out the man page for bootparam and search for Ethernet Devices "man 7 bootparam" HTH, Harry Linux Geek wrote: > Hello, > > I am trying to install RHEL5.3 on a system with multiple Network > interfaces. While installation the network device names are not read in > correct/expected order. I set the ksdevice as eth0 ( using ksdevice=eth0 > at boot prompt ) however the installer tries to acquire IP for eth2 > device. I have tried using "pci=bfsort" with now luck which I believe is > due its applicability on few systems. > > Disabling devices in BIOS din't help either. > > Is there any way to get this wroking ? > > Thanks. > > > ------------------------------------------------------------------------ > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From jimi at sngx.net Tue Jul 7 11:35:07 2009 From: jimi at sngx.net (James Cammarata) Date: Tue, 07 Jul 2009 12:35:07 +0100 Subject: PCI Devices ( NIC ) Enumerated on RHEL5 ( latest ) In-Reply-To: <4A5329E0.7010107@ip-solutions.net> References: <4A5329E0.7010107@ip-solutions.net> Message-ID: <5dab52d4c3f7c6579a9bdfa960315be0@sngx.net> On Tue, 07 Jul 2009 06:56:32 -0400, Harry Hoffman wrote: > Hi LG, > > You might be able to setup the ethernet devices via a bootparam. > > Check out the man page for bootparam and search for Ethernet Devices > "man 7 bootparam" > > HTH, > Harry > > Linux Geek wrote: >> Hello, >> >> I am trying to install RHEL5.3 on a system with multiple Network >> interfaces. While installation the network device names are not read in >> correct/expected order. I set the ksdevice as eth0 ( using ksdevice=eth0 >> at boot prompt ) however the installer tries to acquire IP for eth2 >> device. I have tried using "pci=bfsort" with now luck which I believe is >> due its applicability on few systems. >> >> Disabling devices in BIOS din't help either. >> >> Is there any way to get this wroking ? >> >> Thanks. By default, cobbler uses ksdevice=bootif, and uses the extra parameter "IPAPPEND 2" in the pxelinux file (you must do this or it won't work). What this does is it tells Anaconda to use the ethernet device that was used to PXE boot. You must also be using a syslinux package of a certain version but with RHEL5.3 that shouldn't be a concern (on RHEL4 it can be). -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From larry.brigman at gmail.com Tue Jul 7 15:26:09 2009 From: larry.brigman at gmail.com (Larry Brigman) Date: Tue, 7 Jul 2009 08:26:09 -0700 Subject: PCI Devices ( NIC ) Enumerated on RHEL5 ( latest ) In-Reply-To: <5dab52d4c3f7c6579a9bdfa960315be0@sngx.net> References: <4A5329E0.7010107@ip-solutions.net> <5dab52d4c3f7c6579a9bdfa960315be0@sngx.net> Message-ID: On Tue, Jul 7, 2009 at 4:35 AM, James Cammarata wrote: > > On Tue, 07 Jul 2009 06:56:32 -0400, Harry Hoffman > wrote: >> Hi LG, >> >> You might be able to setup the ethernet devices via a bootparam. >> >> Check out the man page for bootparam and search for Ethernet Devices >> "man 7 bootparam" >> >> HTH, >> Harry >> >> Linux Geek wrote: >>> Hello, >>> >>> I am trying to install RHEL5.3 on a system with multiple Network >>> interfaces. While installation the network device names are not read in >>> correct/expected order. I set the ksdevice as eth0 ( using ksdevice=eth0 > >>> at boot prompt ) however the installer tries to acquire IP for eth2 >>> device. I have tried using "pci=bfsort" with now luck which I believe is > >>> due its applicability on few systems. >>> >>> Disabling devices in BIOS din't help either. >>> >>> Is there any way to get this wroking ? >>> >>> Thanks. > > By default, cobbler uses ksdevice=bootif, and uses the extra parameter > "IPAPPEND 2" in the pxelinux file (you must do this or it won't work). > What this does is it tells Anaconda to use the ethernet device that was > used to PXE boot. ?You must also be using a syslinux package of a certain > version but with RHEL5.3 that shouldn't be a concern (on RHEL4 it can be). > I also found this a while back. It worked for me. It is interesting that none of this is documented in the kickstart documentation. options. From clumens at redhat.com Tue Jul 7 18:51:37 2009 From: clumens at redhat.com (Chris Lumens) Date: Tue, 7 Jul 2009 14:51:37 -0400 Subject: PCI Devices ( NIC ) Enumerated on RHEL5 ( latest ) In-Reply-To: References: <4A5329E0.7010107@ip-solutions.net> <5dab52d4c3f7c6579a9bdfa960315be0@sngx.net> Message-ID: <20090707185137.GD5239@localhost.localdomain> > > By default, cobbler uses ksdevice=bootif, and uses the extra parameter > > "IPAPPEND 2" in the pxelinux file (you must do this or it won't work). > > What this does is it tells Anaconda to use the ethernet device that was > > used to PXE boot. ?You must also be using a syslinux package of a certain > > version but with RHEL5.3 that shouldn't be a concern (on RHEL4 it can be). > > > > I also found this a while back. It worked for me. > > It is interesting that none of this is documented in the kickstart > documentation. > options. This is documented on the anaconda options page of the wiki: https://fedoraproject.org/wiki/Anaconda_Boot_Options The kickstart page on the wiki should refer to it, though. I believe the docs writing guys will be referring to these pages for writing the documentation in RHEL6 and later. - Chris From jimi at sngx.net Wed Jul 8 19:05:09 2009 From: jimi at sngx.net (James Cammarata) Date: Wed, 08 Jul 2009 20:05:09 +0100 Subject: Problem with kickstarts on rhel4u6 In-Reply-To: <63af20d2ce0440873880186136a12b23@sngx.net> References: <63af20d2ce0440873880186136a12b23@sngx.net> Message-ID: <98f27823b429efb85c26f92cc4a8d0ef@sngx.net> On Wed, 01 Jul 2009 16:49:40 -0500, James Cammarata wrote: > Hi all, I'm having a problem with kickstarts on with RHEL4u6. The > kickstart I'm using is essentially the default one provided with cobbler, > with a few extras added in. The problem manifests as follows: > > 1) Kickstart begins, completes phase 1 (or what I call phase 1 anyway) > 2) After it prints "...starting text mode", the install pauses. Switching > to VTY 3, I see that it is sitting at a parted prompt. > 3) Typing quit at the parted prompt causes the kickstart to proceed. > > Has anyone seen this behaviour before? Is this something specific to > RHEL4u6? I don't think the issue lies with the kickstart file, as I use > essentially the same template (especially the partitioning snippets) for > RHEL5 without issue. > > Thanks, > > James C. One week, zero replies... I don't think I've even gotten emails from others on this list. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From co55-sy1t at dea.spamcon.org Thu Jul 9 06:00:47 2009 From: co55-sy1t at dea.spamcon.org (whitivery) Date: Wed, 08 Jul 2009 23:00:47 -0700 Subject: Anaconda early startup failure - traceback References: <4A4CCA2A.5080602@redhat.com> <63ed34d13bf23ec4eac27a14215f3150@sngx.net> <0tgr45t1emneprkctp6rjbpc4kad2cjeup@4ax.com> Message-ID: whitivery wrote: >James Cammarata wrote: > >> >>On Thu, 02 Jul 2009 10:54:34 -0400, Kyle Powell wrote: >>> -----BEGIN PGP SIGNED MESSAGE----- >>> Hash: SHA1 >>> >>> whitivery wrote: >>>> We have a working Kickstart server for CentOS 4.4, building >>>> targets running CentOS 4.4. >>>> >>>> I built a new server with CentOS 5.3, to build CentOS 5.3 >>>> targets, adapting for the changes I noticed as I went along. The >>>> new server boots and runs but won't build a target unit. >>>> >>>> When I PXE-boot a target unit, I get the boot menu and pick an >>>> entry, it starts the early part fine, loads stage2.img OK, shows >>>> a couple screens then a "Retrieving installation information" >>>> dialog/progress bar, and just when/after that gets to 100%, it >>>> halts with a traceback. >>> >>> Looks like you need more RAM: >>> http://www.centos.org/modules/newbb/viewtopic.php?topic_id=19367&forum=37 >>> >>> Same thing with Fedora: >>> http://bugzilla.redhat.com/show_bug.cgi?id=499585 >>> >>> Minimum RAM requirement for RHEL 5 is 512MB. I assume that applies to >>> CentOS as >>> well: >>> http://www.redhat.com/rhel/compare/ >> >>Yes I've seen this same error on systems with insufficient ram (usually >>VM's). The key is that it fails on the lspci call, which is exactly where >>all of mine would fail. > >Thanks to you and Kyle for the answers - I tried another box with >512M RAM and it worked fine. Seems nuts that hundreds of >millions of bytes of memory can't even start a text mode install. > >With the box that worked, I still see some warnings in the >anaconda.log like the following - do I need some additional >kickstart settings or something to avoid these?: > > I don't need DNS hostname lookup so why is it trying?: >21:00:01 ERROR : no DNS servers, can't look up hostname >21:00:01 ERROR : no DNS servers, can't look up hostname > > I specified "url" so why these messages?: >21:00:01 ERROR : got to setupCdrom without a CD device >21:00:22 WARNING : no floppy devices found but we'll try fd0 anyway > > I specified "text" so why does it say "not available" - it's > available, I just prefer text install: >21:00:25 WARNING : Graphical installation not available... Starting text >mode. > > What does this mean?: >21:00:28 WARNING : Unable to find temp path, going to use ramfs path > > Why these?: >21:00:32 WARNING : step installtype does not exist >21:00:32 WARNING : step complete does not exist >21:00:32 WARNING : step complete does not exist >21:00:32 WARNING : step complete does not exist >21:00:32 WARNING : step complete does not exist >21:00:32 WARNING : step complete does not exist >21:00:32 WARNING : step complete does not exist >21:00:32 WARNING : step complete does not exist >21:00:32 WARNING : step complete does not exist >21:00:32 WARNING : step complete does not exist >21:00:32 WARNING : step complete does not exist >21:00:32 WARNING : step complete does not exist >21:00:32 WARNING : step complete does not exist >21:00:32 WARNING : step complete does not exist > > These seem to be bugs: >21:00:36 WARNING : /usr/lib/python2.4/site-packages/snack.py:250: >DeprecationWarning: integer argument expected, got float > self.w = _snack.scale(width, total) > >21:00:36 WARNING : /usr/lib/python2.4/site-packages/snack.py:247: >DeprecationWarning: integer argument expected, got float > self.w.scaleSet(amount) > >21:00:48 WARNING : /usr/lib/python2.4/site-packages/yum/repos.py:68: >RuntimeWarning: YumProgress.progressbar called when popped > self.callback.progressbar(num, len(repos), repo.id) > > In addition, there are numerous "WARNING: doesn't > exist" scattered in the log near the end of the install. Following up to my post: another problem is that %end still doesn't work - using it to end the packages, pre, or post sections, it causes a fatal failure, kickstart stops. I expected this with my old CentOS 4.4 kickstart/anaconda, but not with the much newer CentOS 5.3 one - the web documentation implies that it should work. From co55-sy1t at dea.spamcon.org Thu Jul 9 07:13:58 2009 From: co55-sy1t at dea.spamcon.org (whitivery) Date: Thu, 09 Jul 2009 00:13:58 -0700 Subject: Anaconda early startup failure - traceback References: <4A4CCA2A.5080602@redhat.com> <63ed34d13bf23ec4eac27a14215f3150@sngx.net> <0tgr45t1emneprkctp6rjbpc4kad2cjeup@4ax.com> Message-ID: <326b555ngub0ioqpmumrtd5k6idju3gc1g@4ax.com> whitivery wrote: >whitivery wrote: > >>James Cammarata wrote: >> >>> >>>On Thu, 02 Jul 2009 10:54:34 -0400, Kyle Powell wrote: >>>> -----BEGIN PGP SIGNED MESSAGE----- >>>> Hash: SHA1 >>>> >>>> whitivery wrote: >>>>> We have a working Kickstart server for CentOS 4.4, building >>>>> targets running CentOS 4.4. >>>>> >>>>> I built a new server with CentOS 5.3, to build CentOS 5.3 >>>>> targets, adapting for the changes I noticed as I went along. The >>>>> new server boots and runs but won't build a target unit. >>>>> >>>>> When I PXE-boot a target unit, I get the boot menu and pick an >>>>> entry, it starts the early part fine, loads stage2.img OK, shows >>>>> a couple screens then a "Retrieving installation information" >>>>> dialog/progress bar, and just when/after that gets to 100%, it >>>>> halts with a traceback. >>>> >>>> Looks like you need more RAM: >>>> http://www.centos.org/modules/newbb/viewtopic.php?topic_id=19367&forum=37 >>>> >>>> Same thing with Fedora: >>>> http://bugzilla.redhat.com/show_bug.cgi?id=499585 >>>> >>>> Minimum RAM requirement for RHEL 5 is 512MB. I assume that applies to >>>> CentOS as >>>> well: >>>> http://www.redhat.com/rhel/compare/ >>> >>>Yes I've seen this same error on systems with insufficient ram (usually >>>VM's). The key is that it fails on the lspci call, which is exactly where >>>all of mine would fail. >> >>Thanks to you and Kyle for the answers - I tried another box with >>512M RAM and it worked fine. Seems nuts that hundreds of >>millions of bytes of memory can't even start a text mode install. >> >>With the box that worked, I still see some warnings in the >>anaconda.log like the following - do I need some additional >>kickstart settings or something to avoid these?: >> >> I don't need DNS hostname lookup so why is it trying?: >>21:00:01 ERROR : no DNS servers, can't look up hostname >>21:00:01 ERROR : no DNS servers, can't look up hostname >> >> I specified "url" so why these messages?: >>21:00:01 ERROR : got to setupCdrom without a CD device >>21:00:22 WARNING : no floppy devices found but we'll try fd0 anyway >> >> I specified "text" so why does it say "not available" - it's >> available, I just prefer text install: >>21:00:25 WARNING : Graphical installation not available... Starting text >>mode. >> >> What does this mean?: >>21:00:28 WARNING : Unable to find temp path, going to use ramfs path >> >> Why these?: >>21:00:32 WARNING : step installtype does not exist >>21:00:32 WARNING : step complete does not exist >>21:00:32 WARNING : step complete does not exist >>21:00:32 WARNING : step complete does not exist >>21:00:32 WARNING : step complete does not exist >>21:00:32 WARNING : step complete does not exist >>21:00:32 WARNING : step complete does not exist >>21:00:32 WARNING : step complete does not exist >>21:00:32 WARNING : step complete does not exist >>21:00:32 WARNING : step complete does not exist >>21:00:32 WARNING : step complete does not exist >>21:00:32 WARNING : step complete does not exist >>21:00:32 WARNING : step complete does not exist >>21:00:32 WARNING : step complete does not exist >> >> These seem to be bugs: >>21:00:36 WARNING : /usr/lib/python2.4/site-packages/snack.py:250: >>DeprecationWarning: integer argument expected, got float >> self.w = _snack.scale(width, total) >> >>21:00:36 WARNING : /usr/lib/python2.4/site-packages/snack.py:247: >>DeprecationWarning: integer argument expected, got float >> self.w.scaleSet(amount) >> >>21:00:48 WARNING : /usr/lib/python2.4/site-packages/yum/repos.py:68: >>RuntimeWarning: YumProgress.progressbar called when popped >> self.callback.progressbar(num, len(repos), repo.id) >> >> In addition, there are numerous "WARNING: doesn't >> exist" scattered in the log near the end of the install. > >Following up to my post: another problem is that %end still >doesn't work - using it to end the packages, pre, or post >sections, it causes a fatal failure, kickstart stops. > >I expected this with my old CentOS 4.4 kickstart/anaconda, but >not with the much newer CentOS 5.3 one - the web documentation >implies that it should work. Yet another follow-up problem/question: in the old CentOS 4.4 the --erroronfail didn't work. I thought it was supposed to work in this version (CentOS 5.3), but trying it in both the pre and post sections, it doesn't seem to do anything. If a command returns non-zero, or a bad (non-existent) command is run in either section, in the latter case there's a complaint in tty3 but in either case no effect on the install, it keeps going. Maybe this is how they are supposed to work, the documentation doesn't really say exactly what happens or what is considered a failure. From pgroven at 2wire.com Thu Jul 9 16:27:17 2009 From: pgroven at 2wire.com (Phillip Groven) Date: Thu, 9 Jul 2009 09:27:17 -0700 Subject: Kickstart not rebooting on completion In-Reply-To: <4A4B96A6.3010200@ng23.net> References: <06A6610D4F464D4EBEAFBF2C5F86911E010D6AF0@exchange2.columbia.tresys.com> <4A40EFEC.9070809@rice.edu> <06A6610D4F464D4EBEAFBF2C5F86911E01242990@exchange2.columbia.tresys.com><4A495D92.6020303@yahoo.com> <4A4B96A6.3010200@ng23.net> Message-ID: <299E8BB0D5475E4FABB162E7DCA1BBE903A8D2A6@phxexch01.corp.2wire.com> I use the following ######################################################################## #### # tell big daddy we are all done cat /root/install.log /root/anaconda-ks.cfg | mail -s 'ks install' myemail at domain.com shutdown -r now If I use reboot instead of shutdown -r now then Ananconda complains but almost always reboots anyway. -----Original Message----- From: Tom Brown [mailto:tom at ng23.net] Sent: Wednesday, July 01, 2009 10:03 AM To: Discussion list about Kickstart Subject: Re: Kickstart not rebooting on completion Matthew Milliss wrote: > I'm using kickstart to install Fedora 11 and I'm having an issue with > the system not rebooting when the install has finished. The install is > running in text mode and it just stops on the "Running post install" > screen. If I force a reboot the install has been successful, but I > find myself wasting time waiting around to make sure the install has > finished before forcing a reboot. My kickstart file is below. Any help > would be greatly appreciated. > > # Kickstart file automatically generated by anaconda - with some > additions by Matt. > #version=F11 > install > text > cdrom > lang en_AU.UTF-8 > keyboard us > network --device=eth0 --bootproto=dhcp --hostname=snotrod > xconfig --startxonboot > rootpw --iscrypted > firewall --disabled > authconfig --enableshadow --passalgo=sha512 --enablefingerprint > selinux --permissive > timezone --utc Australia/Sydney > bootloader --location=mbr --driveorder=sda --append="quiet reboot=bios" > autopart > clearpart --all > reboot --eject > > repo --name=msl --baseurl=http://hannibal/tyroyummy/msl > > %packages > @admin-tools > @base > @core > @editors > @fonts > @gnome-desktop > @graphical-internet > @graphics > @hardware-support > @input-methods > @office > @online-docs > @printing > @sound-and-video > @base-x > xfsprogs > mtools > gpgme > pax > gnupg2 > gvfs-obexftp > ImageMagick > dcraw > netpbm-progs > hdparm > m17n-db-assamese > gok > m17n-db-tamil > m17n-db-gujarati > iok > m17n-db-hindi > m17n-db-punjabi > m17n-db-oriya > m17n-db-telugu > m17n-db-bengali > m17n-db-malayalam > m17n-contrib-sinhala > m17n-db-kannada > gdm > nvu > binutils > rdesktop > wireshark > wireshark-gnome > subversion > vnc > minicom > thunderbird > rpm-build > kernel-headers > createrepo > kernel-devel > yelp > jedit > vsftpd > AdobeReader_enu > flash-plugin > nxclient > nxnode > nxserver > ntp > %end > > %post > #!/bin/bash > env > /root/postks.log > mkdir /mnt/pub > mkdir /mnt/yum > mount -o ro,nolock -t nfs hannibal:/tyroyummy /mnt/yum > #mount -o ro,nolock -t nfs 192.168.100.58:/home/pub/ /mnt/pub > cp /etc/fstab /etc/fstab.bk > echo "avalon-2:/home/pub /mnt/pub nfs auto,rw,intr > 0 0" >> /etc/fstab > echo "hannibal:/tyroyummy /mnt/yum nfs auto,rw,intr > 0 0" >> /etc/fstab > cd /mnt/yum/SOE_Fedora11/install > ./install.sh > /var/log/msl_install_output.log > cd / > umount /mnt/yum > %end > > what does alt+f2 etc show you its waiting for From srs at redhat.com Fri Jul 10 21:41:18 2009 From: srs at redhat.com (Srinivas Satyavarpu) Date: Fri, 10 Jul 2009 14:41:18 -0700 Subject: kickstart question... Message-ID: <4A57B57E.5030500@redhat.com> Hello One of my customer is wanting to set up a root disk mirroring using software raid. He wants to use satellite server / kickstart profile to achieve the same. I believe one can do a combination of %pre script as well as partitioning disk section in the kickstart file to achieve root disk mirroring. But my customer wants to keep the kickstart profile fairly generic and hence cannot contain specific disk names like /dev/sda or /dev/hda because the kickstart profile has to work on multiple systems. Can some one please suggest how one can achieve the above using some sort of %pre script. Any help would be much appreciated. Thx Srini -- http://www.redhat.com/promo/summit/2009/ ::::::::::::::::::::::::::::::::::::::: Srini Satyavarpu Solutions Architect Red Hat, Inc. mobile: 408.464.8553 email: srs at redhat.com ::::::::::::::::::::::::::::::::::::::: From chip.shabazian at bankofamerica.com Fri Jul 10 22:23:11 2009 From: chip.shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 10 Jul 2009 15:23:11 -0700 Subject: kickstart question... In-Reply-To: <4A57B57E.5030500@redhat.com> References: <4A57B57E.5030500@redhat.com> Message-ID: I've got some useful tips on how to do this in a presentation I did two years ago (and will be doing at Red Hat Summit in case anyone wants to come) http://www.shabazian.com/lw2007.pdf -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Srinivas Satyavarpu Sent: Friday, July 10, 2009 2:41 PM To: kickstart-list at redhat.com Subject: kickstart question... Hello One of my customer is wanting to set up a root disk mirroring using software raid. He wants to use satellite server / kickstart profile to achieve the same. I believe one can do a combination of %pre script as well as partitioning disk section in the kickstart file to achieve root disk mirroring. But my customer wants to keep the kickstart profile fairly generic and hence cannot contain specific disk names like /dev/sda or /dev/hda because the kickstart profile has to work on multiple systems. Can some one please suggest how one can achieve the above using some sort of %pre script. Any help would be much appreciated. Thx Srini -- http://www.redhat.com/promo/summit/2009/ ::::::::::::::::::::::::::::::::::::::: Srini Satyavarpu Solutions Architect Red Hat, Inc. mobile: 408.464.8553 email: srs at redhat.com ::::::::::::::::::::::::::::::::::::::: _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From invite+kjdmu3pd3~1_ at facebookmail.com Sat Jul 11 12:05:11 2009 From: invite+kjdmu3pd3~1_ at facebookmail.com (NNoman Yousuf) Date: Sat, 11 Jul 2009 05:05:11 -0700 Subject: Check out my photos on Facebook Message-ID: <43aced5980be68fa6e07552c36ea2283@localhost.localdomain> Hi kickstart-list at redhat.com, I set up a Facebook profile where I can post my pictures, videos and events and I want to add you as a friend so you can see it. First, you need to join Facebook! Once you join, you can also create your own profile. Thanks, NNoman To sign up for Facebook, follow the link below: http://www.facebook.com/p.php?i=522073481&k=Y51X63USUX3AUCD1QB63YQWPVTKD&r kickstart-list at redhat.com was invited to join Facebook by NNoman Yousuf. If you do not wish to receive this type of email from Facebook in the future, please click on the link below to unsubscribe. http://www.facebook.com/o.php?k=b89dec&u=100000085635187&mid=c27d25G5af31594f073G0G8 Facebook's offices are located at 1601 S. California Ave., Palo Alto, CA 94304. -------------- next part -------------- An HTML attachment was scrubbed... URL: From hardik at statementofpurpose.com Fri Jul 17 12:10:48 2009 From: hardik at statementofpurpose.com (Hardik Modi) Date: Fri, 17 Jul 2009 08:10:48 -0400 Subject: Kickstart upgrade package management Message-ID: Hello All, I've been a satisfied user of kickstart for custom fresh installs for some time now. The machines I'm installing to are often not connected to the Internet and I'm looking at using the upgrade function to have my machines step up from CentOS 4.4 to 5.3. With a simple upgrade kickstart (only the essential lines), I'm able to run through an upgrade cycle. This takes care of most of the packages that I expect to see upgraded. The reason I'm writing to the list is there are a few packages that I'd rather the upgrader didn't touch and that I'd then fix up in the post-install. Is there a direct way of asserting this 'ignore package' command? The guides that I find online are quite clear that the %packages section is ignored in an upgrade kickstart. Thanks! Hardik. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Joel.Schuweiler at schmoozecom.com Wed Jul 22 16:22:41 2009 From: Joel.Schuweiler at schmoozecom.com (Joel Schuweiler) Date: Wed, 22 Jul 2009 11:22:41 -0500 Subject: Repo on a cd Message-ID: I have a few custom packages I need to install. I would like to install them in the %packages section, not post. Is it possible to specify a repo directive like the following: repo -name=customrpms -url=./customrepos thoughts? -Joel -------------- next part -------------- An HTML attachment was scrubbed... URL: From jason at rampaginggeek.com Wed Jul 22 22:55:56 2009 From: jason at rampaginggeek.com (Jason Edgecombe) Date: Wed, 22 Jul 2009 18:55:56 -0400 Subject: Repo on a cd In-Reply-To: References: Message-ID: <4A6798FC.1090500@rampaginggeek.com> Joel Schuweiler wrote: > I have a few custom packages I need to install. I would like to install them in the %packages section, not post. > > Is it possible to specify a repo directive like the following: > > repo -name=customrpms -url=./customrepos > > thoughts? > yes, but it should be the following, assuming you're installing from CD: repo -name=customrpms -url=/mnt/source/customrepos The install media is mounted as /mnt/source or something similar. Jason From Joel.Schuweiler at schmoozecom.com Fri Jul 24 13:51:43 2009 From: Joel.Schuweiler at schmoozecom.com (Joel Schuweiler) Date: Fri, 24 Jul 2009 08:51:43 -0500 Subject: Repo on a cd In-Reply-To: <4A6798FC.1090500@rampaginggeek.com> Message-ID: Jason, In anaconda.log it complains that it can't retrieve repository metadata when doing this. Any thoughts? Anyway to get it to be more verbose so I can see where exactly it's looking? Thanks, joel -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Jason Edgecombe Sent: Wednesday, July 22, 2009 5:56 PM To: Discussion list about Kickstart Subject: Re: Repo on a cd Joel Schuweiler wrote: > I have a few custom packages I need to install. I would like to install them in the %packages section, not post. > > Is it possible to specify a repo directive like the following: > > repo -name=customrpms -url=./customrepos > > thoughts? > yes, but it should be the following, assuming you're installing from CD: repo -name=customrpms -url=/mnt/source/customrepos The install media is mounted as /mnt/source or something similar. Jason _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From mdehaan at redhat.com Mon Jul 27 15:09:09 2009 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 27 Jul 2009 11:09:09 -0400 Subject: Repo on a cd In-Reply-To: References: Message-ID: <4A6DC315.8080002@redhat.com> On 07/24/2009 09:51 AM, Joel Schuweiler wrote: > Jason, > > In anaconda.log it complains that it can't retrieve repository metadata when doing this. > > Any thoughts? Anyway to get it to be more verbose so I can see where exactly it's looking? > > Thanks, > joel > Did you run createrepo on the directory in question? --Michael > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Jason Edgecombe > Sent: Wednesday, July 22, 2009 5:56 PM > To: Discussion list about Kickstart > Subject: Re: Repo on a cd > > Joel Schuweiler wrote: > >> I have a few custom packages I need to install. I would like to install them in the %packages section, not post. >> >> Is it possible to specify a repo directive like the following: >> >> repo -name=customrpms -url=./customrepos >> >> thoughts? >> >> > yes, but it should be the following, assuming you're installing from CD: > > repo -name=customrpms -url=/mnt/source/customrepos > > > The install media is mounted as /mnt/source or something similar. > > Jason > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Joel.Schuweiler at schmoozecom.com Mon Jul 27 16:47:05 2009 From: Joel.Schuweiler at schmoozecom.com (Joel Schuweiler) Date: Mon, 27 Jul 2009 11:47:05 -0500 Subject: Repo on a cd In-Reply-To: <4A6DC315.8080002@redhat.com> References: <4A6DC315.8080002@redhat.com> Message-ID: <8F21FFC2-362D-4F19-A898-A5CFE7647261@schmoozecom.com> Ended up still being a bad repo line On Jul 27, 2009, at 9:07 AM, "Michael DeHaan" > wrote: On 07/24/2009 09:51 AM, Joel Schuweiler wrote: Jason, In anaconda.log it complains that it can't retrieve repository metadata when doing this. Any thoughts? Anyway to get it to be more verbose so I can see where exactly it's looking? Thanks, joel Did you run createrepo on the directory in question? --Michael -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Jason Edgecombe Sent: Wednesday, July 22, 2009 5:56 PM To: Discussion list about Kickstart Subject: Re: Repo on a cd Joel Schuweiler wrote: I have a few custom packages I need to install. I would like to install them in the %packages section, not post. Is it possible to specify a repo directive like the following: repo -name=customrpms -url=./customrepos thoughts? yes, but it should be the following, assuming you're installing from CD: repo -name=customrpms -url=/mnt/source/customrepos The install media is mounted as /mnt/source or something similar. Jason _______________________________________________ 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 -------------- next part -------------- An HTML attachment was scrubbed... URL: From mdehaan at redhat.com Mon Jul 27 16:51:19 2009 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 27 Jul 2009 12:51:19 -0400 Subject: Repo on a cd In-Reply-To: <8F21FFC2-362D-4F19-A898-A5CFE7647261@schmoozecom.com> References: <4A6DC315.8080002@redhat.com> <8F21FFC2-362D-4F19-A898-A5CFE7647261@schmoozecom.com> Message-ID: <4A6DDB07.9020606@redhat.com> On 07/27/2009 12:47 PM, Joel Schuweiler wrote: > Ended up still being a bad repo line I am not sure I follow.... your problem is resolved, or you are still trying to get the "repo" line correct? --Michael > > > > On Jul 27, 2009, at 9:07 AM, "Michael DeHaan" > wrote: > >> On 07/24/2009 09:51 AM, Joel Schuweiler wrote: >>> Jason, >>> >>> In anaconda.log it complains that it can't retrieve repository metadata when doing this. >>> >>> Any thoughts? Anyway to get it to be more verbose so I can see where exactly it's looking? >>> >>> Thanks, >>> joel >>> >> >> >> Did you run createrepo on the directory in question? >> >> --Michael >> >> >>> -----Original Message----- >>> From:kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Jason Edgecombe >>> Sent: Wednesday, July 22, 2009 5:56 PM >>> To: Discussion list about Kickstart >>> Subject: Re: Repo on a cd >>> >>> Joel Schuweiler wrote: >>> >>>> I have a few custom packages I need to install. I would like to install them in the %packages section, not post. >>>> >>>> Is it possible to specify a repo directive like the following: >>>> >>>> repo -name=customrpms -url=./customrepos >>>> >>>> thoughts? >>>> >>>> >>> yes, but it should be the following, assuming you're installing from CD: >>> >>> repo -name=customrpms -url=/mnt/source/customrepos >>> >>> >>> The install media is mounted as /mnt/source or something similar. >>> >>> Jason >>> >>> _______________________________________________ >>> Kickstart-list mailing list >>> Kickstart-list at redhat.com >>> https://www.redhat.com/mailman/listinfo/kickstart-list >>> >>> _______________________________________________ >>> Kickstart-list mailing list >>> Kickstart-list at redhat.com >>> https://www.redhat.com/mailman/listinfo/kickstart-list >>> >> >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list > ------------------------------------------------------------------------ > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From erlingre at gmail.com Wed Jul 29 10:05:25 2009 From: erlingre at gmail.com (Erling Ringen Elvsrud) Date: Wed, 29 Jul 2009 12:05:25 +0200 Subject: Manual network config, fetch tcp/ip setup from stage1 in stage2 Message-ID: <664c5a070907290305u758485fel32d54975efae43d9@mail.gmail.com> Hello list, I have these constraints for kickstarting: - no dhcp - must use cd-boot - must work in many different network-setups Currently I have managed to get a setup where have to supply network configuration twice. First in stage1 of anaconda to fetch stage2 from RH Satellite Server and then in stage2 for configuring the system that is kickstarted. Do you know if there is a way to fetch the tcp/ip config from stage 1 in stage 2 to avoid repeating the configuration? like in %prescript? Thanks, Erling From kpowell at redhat.com Wed Jul 29 15:53:14 2009 From: kpowell at redhat.com (Kyle Powell) Date: Wed, 29 Jul 2009 11:53:14 -0400 Subject: Manual network config, fetch tcp/ip setup from stage1 in stage2 In-Reply-To: <664c5a070907290305u758485fel32d54975efae43d9@mail.gmail.com> References: <664c5a070907290305u758485fel32d54975efae43d9@mail.gmail.com> Message-ID: <4A70706A.6090300@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Erling Ringen Elvsrud wrote: > Do you know if there is a way to fetch the tcp/ip config from stage 1 > in stage 2 to avoid > repeating the configuration? like in %prescript? Hi Erling, Absolutely, and you're on the right track. You can %include a file generated with a %pre script in your kickstart. Simply have a pre script that generates a file containing the network line (network --device ethx --bootproto static [--noipv6] --ip x.x.x.x --netmask x.x.x.x --gateway x.x.x.x --nameserver x.x.x.x[,x.x.x.x] [--hostname %s]) then use a %include statement to add the contents of the file to the kickstart. You may be able to simply parse the contents of /tmp/netinfo or the output of ifconfig, or you may need to resort to python (isys.getIPAddress). Incidentally, you could also avoid having to enter the info for stage 1 by passing it as command line parameters (ip=, netmask=, etc...). If you do this you could parse /proc/cmdline for the info. - -- Kyle Powell | Red Hat | Senior Consultant, RHCE -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFKcHBq7pTtanQdBU4RAgRHAJoDG0aWogdd+slNWxxNyezHy7+7DwCfWlM0 /i3qKr+wLmpLZBZ2V5/H6sI= =owwX -----END PGP SIGNATURE----- From Joel.Schuweiler at schmoozecom.com Wed Jul 29 17:01:34 2009 From: Joel.Schuweiler at schmoozecom.com (Joel Schuweiler) Date: Wed, 29 Jul 2009 12:01:34 -0500 Subject: Hide package selection screen Message-ID: Due to the nature of my kickstart, I pre-define lots of things, but I still want to provide the user with a way to change some things, as a result I do an interactive install. However, I do not want the user to be able to pick packages. Is there a way to prevent the user from being shown this step? Thanks, Joel -------------- next part -------------- An HTML attachment was scrubbed... URL: From clumens at redhat.com Wed Jul 29 20:13:12 2009 From: clumens at redhat.com (Chris Lumens) Date: Wed, 29 Jul 2009 16:13:12 -0400 Subject: Hide package selection screen In-Reply-To: References: Message-ID: <20090729201312.GW9030@localhost.localdomain> > Due to the nature of my kickstart, I pre-define lots of things, but I > still want to provide the user with a way to change some things, as a > result I do an interactive install. However, I do not want the user to > be able to pick packages. Is there a way to prevent the user from > being shown this step? No, an interactive kickstart install means you stop on everything. There's no way to get finer grained control over that. - Chris From Joel.Schuweiler at schmoozecom.com Wed Jul 29 20:06:17 2009 From: Joel.Schuweiler at schmoozecom.com (Joel Schuweiler) Date: Wed, 29 Jul 2009 15:06:17 -0500 Subject: Hide package selection screen In-Reply-To: <20090729201312.GW9030@localhost.localdomain> References: <20090729201312.GW9030@localhost.localdomain> Message-ID: I can see in anaconda how easy it would be to get rid of this screen... it's fairly simple logic where it decides to display it or not... less hackish though, where does it pull in the groups it displays? Is it out of one of the xml files and I could just modify that? -Joel -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Chris Lumens Sent: Wednesday, July 29, 2009 3:13 PM To: kickstart-list at redhat.com Subject: Re: Hide package selection screen > Due to the nature of my kickstart, I pre-define lots of things, but I > still want to provide the user with a way to change some things, as a > result I do an interactive install. However, I do not want the user to > be able to pick packages. Is there a way to prevent the user from > being shown this step? No, an interactive kickstart install means you stop on everything. There's no way to get finer grained control over that. - Chris _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From clumens at redhat.com Wed Jul 29 21:18:59 2009 From: clumens at redhat.com (Chris Lumens) Date: Wed, 29 Jul 2009 17:18:59 -0400 Subject: Hide package selection screen In-Reply-To: References: <20090729201312.GW9030@localhost.localdomain> Message-ID: <20090729211859.GY9030@localhost.localdomain> > I can see in anaconda how easy it would be to get rid of this > screen... it's fairly simple logic where it decides to display it or > not... less hackish though, where does it pull in the groups it > displays? Is it out of one of the xml files and I could just modify > that? The comps.xml file is where it's mostly driven from, though the installclasses are involved here too. - Chris From erlingre at gmail.com Thu Jul 30 10:03:09 2009 From: erlingre at gmail.com (Erling Ringen Elvsrud) Date: Thu, 30 Jul 2009 12:03:09 +0200 Subject: Manual network config, fetch tcp/ip setup from stage1 in stage2 In-Reply-To: <4A70706A.6090300@redhat.com> References: <664c5a070907290305u758485fel32d54975efae43d9@mail.gmail.com> <4A70706A.6090300@redhat.com> Message-ID: <664c5a070907300303v62aeee09j71d984beaea60874@mail.gmail.com> On Wed, Jul 29, 2009 at 5:53 PM, Kyle Powell wrote: [...] > Absolutely, and you're on the right track. You can %include a file generated > with a %pre script in your kickstart. Simply have a pre script that generates a > file containing the network line (network --device ethx --bootproto static > [--noipv6] --ip x.x.x.x --netmask x.x.x.x --gateway x.x.x.x --nameserver > x.x.x.x[,x.x.x.x] [--hostname %s]) then use a %include statement to add the > contents of the file to the kickstart. You may be able to simply parse the > contents of /tmp/netinfo or the output of ifconfig, or you may need to resort to > python (isys.getIPAddress). Incidentally, you could also avoid having to enter > the info for stage 1 by passing it as command line parameters (ip=, netmask=, > etc...). If you do this you could parse /proc/cmdline for the info. Thanks for your reply. I did not mention that I kickstart via Satellite Server in the first post, but it might be relevant. I have retried a kickstart and captured screenshots to make it easier to understand what I have done. Here is a quick and dirty description of the process: http://www.reisesakte.info/anaconda/ The problem is that when I kickstart via Satellite Server the network configuration is repeated twice even before Anaconda reaches stage2 as far as I understand. So %prescript is not executed early enough. It is probably common to use DHCP/PXE which would avoid the multiple configurations before stage2 I suppose, but at my site it is quite complex to get such a setup and I prefer to use CD-boot if possible. Note that the install session documented works, but is a bit cumbersome. I hope to be able to polish the install routines more so my users can perform installation themselves. For instance VMware administrators. Supplying ip config on the kernel options might work, but I would prefer nice dialog interfaces. Any suggestions for how to best kickstart systems from satellite server from a custom boot-iso are welcome. Thanks, Erling From ebrown at lanl.gov Thu Jul 30 15:23:26 2009 From: ebrown at lanl.gov (Ed Brown) Date: Thu, 30 Jul 2009 09:23:26 -0600 Subject: Manual network config, fetch tcp/ip setup from stage1 in stage2 In-Reply-To: <664c5a070907300303v62aeee09j71d984beaea60874@mail.gmail.com> References: <664c5a070907290305u758485fel32d54975efae43d9@mail.gmail.com> <4A70706A.6090300@redhat.com> <664c5a070907300303v62aeee09j71d984beaea60874@mail.gmail.com> Message-ID: <4A71BAEE.605@lanl.gov> Erling Ringen Elvsrud wrote: > Supplying ip config on the kernel options might work, > but I would prefer nice dialog interfaces. Anaconda and kickstart have improved a lot in the 7 or 8 years I've been using them, but this question of how to allow for just the static ip configuration at kickstart time continues to come up over and over and over. Currently, your best bet is to forget about having a nice dialog interface, and go with the kernel options at the boot prompt. As you've seen, the dialogs are just plain broken for this purpose, and can result in an incomplete configuration afterward (no hostname in /etc/sysconfig/network and only a localhost entry in /etc/hosts). Since you are already creating your own boot iso and targets therein, you can include the parameters that are unlikely to vary. Assuming you know the dns server and netmask and possibly the gateway, the user may only have to append "ip=" to the target name, which is even easier than going through the dialogs. E.g.: label one kernel vmlinuz append initrd=initrd.img text nofb gateway=1.2.3.1 dns=1.2.3.4 netmask=255.255.255.0 ks=url_of_ks_file method=url_of_satellite label two kernel vmlinuz append initrd=initrd.img text nofb dns=1.2.3.1 netmask=255.255.255.0 ks=url_of_ks_file method=url_of_satellite Using target one, the user only needs to enter the ip, with target two, she'd have to enter ip and gateway. I haven't found it necessary to include dhcptimeout=0 using this method. You can also provide an example or reminder in your boot.msg file to help your users know what to do at the boot prompt: ----------------------------------------------------------------- ... Please enter one of the following: one ip=1.2.3.??? OR: two ip=1.2.???.??? gateway=1.2.???.252 Or specify custom parameters as in this full syntax example: linux ks= method= ip=1.2.xxx.yyy gateway=1.2.xxx.252 dns=1.2.3.1 netmask=255.255.255.0 ----------------------------------------------------------------- In your kickstart file, use a network line like: network --bootproto=static --noipv6 and no other pre/post hackery is needed. hth, Ed From CallahanT at tessco.com Thu Jul 30 15:39:19 2009 From: CallahanT at tessco.com (Callahan, Tom) Date: Thu, 30 Jul 2009 11:39:19 -0400 Subject: Manual network config, fetch tcp/ip setup from stage1 in stage2 In-Reply-To: <664c5a070907300303v62aeee09j71d984beaea60874@mail.gmail.com> Message-ID: Nice dialog interfaces to me, don't mean kickstart. I view kickstart as a set it and forget it install method, as I think most people use it for. If you want install dialogs, why not just boot the DVD and do the install the old fashioned way? -- Tom Callahan Technology, Development and Services (TDS) 410-229-1361 Tel 410-229-1511 Fax callahant at tessco.com Visit TESSCO.com TESSCO Your Total Source? for Everything Wireless Network Infrastructure Equipment | Mobile Devices & Accessories | Installation, Test, Equipment & Supplies ?You may never know what results come of your action, but if you do nothing there will be no result? --Mohandas Gandhi On 7/30/09 6:03 AM, "Erling Ringen Elvsrud" wrote: > On Wed, Jul 29, 2009 at 5:53 PM, Kyle Powell wrote: > [...] >> Absolutely, and you're on the right track. You can %include a file generated >> with a %pre script in your kickstart. Simply have a pre script that generates >> a >> file containing the network line (network --device ethx --bootproto static >> [--noipv6] --ip x.x.x.x --netmask x.x.x.x --gateway x.x.x.x --nameserver >> x.x.x.x[,x.x.x.x] [--hostname %s]) then use a %include statement to add the >> contents of the file to the kickstart. You may be able to simply parse the >> contents of /tmp/netinfo or the output of ifconfig, or you may need to resort >> to >> python (isys.getIPAddress). Incidentally, you could also avoid having to >> enter >> the info for stage 1 by passing it as command line parameters (ip=, netmask=, >> etc...). If you do this you could parse /proc/cmdline for the info. > > Thanks for your reply. I did not mention that I kickstart via > Satellite Server in the first post, but it might be relevant. I have > retried a kickstart and captured screenshots to make it easier to > understand what I have done. > > Here is a quick and dirty description of the process: > > http://www.reisesakte.info/anaconda/ > > The problem is that when I kickstart via Satellite Server the network > configuration is repeated > twice even before Anaconda reaches stage2 as far as I understand. So > %prescript is not > executed early enough. It is probably common to use DHCP/PXE which > would avoid the multiple configurations before stage2 I suppose, but > at my site it is quite complex to get such a setup and I prefer to use > CD-boot if possible. > > Note that the install session documented works, but is a bit > cumbersome. I hope to be able to polish the install routines more so > my users can perform installation themselves. For instance VMware > administrators. Supplying ip config on the kernel options might work, > but I would prefer nice dialog interfaces. > > Any suggestions for how to best kickstart systems from satellite > server from a custom boot-iso are welcome. > > Thanks, > > Erling > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3483 bytes Desc: not available URL: From mdehaan at redhat.com Thu Jul 30 15:41:41 2009 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 30 Jul 2009 11:41:41 -0400 Subject: Manual network config, fetch tcp/ip setup from stage1 in stage2 In-Reply-To: <664c5a070907300303v62aeee09j71d984beaea60874@mail.gmail.com> References: <664c5a070907290305u758485fel32d54975efae43d9@mail.gmail.com> <4A70706A.6090300@redhat.com> <664c5a070907300303v62aeee09j71d984beaea60874@mail.gmail.com> Message-ID: <4A71BF35.9070002@redhat.com> On 07/30/2009 06:03 AM, Erling Ringen Elvsrud wrote: > On Wed, Jul 29, 2009 at 5:53 PM, Kyle Powell wrote: > [...] > >> Absolutely, and you're on the right track. You can %include a file generated >> with a %pre script in your kickstart. Simply have a pre script that generates a >> file containing the network line (network --device ethx --bootproto static >> [--noipv6] --ip x.x.x.x --netmask x.x.x.x --gateway x.x.x.x --nameserver >> x.x.x.x[,x.x.x.x] [--hostname %s]) then use a %include statement to add the >> contents of the file to the kickstart. You may be able to simply parse the >> contents of /tmp/netinfo or the output of ifconfig, or you may need to resort to >> python (isys.getIPAddress). Incidentally, you could also avoid having to enter >> the info for stage 1 by passing it as command line parameters (ip=, netmask=, >> etc...). If you do this you could parse /proc/cmdline for the info. >> > > Thanks for your reply. I did not mention that I kickstart via > Satellite Server in the first post, but it might be relevant. I have > retried a kickstart and captured screenshots to make it easier to > understand what I have done. > > Here is a quick and dirty description of the process: > > http://www.reisesakte.info/anaconda/ > > The problem is that when I kickstart via Satellite Server the network > configuration is repeated > twice even before Anaconda reaches stage2 as far as I understand. So > %prescript is not > executed early enough. It is probably common to use DHCP/PXE which > would avoid the multiple configurations before stage2 I suppose, but > at my site it is quite complex to get such a setup and I prefer to use > CD-boot if possible. > > Note that the install session documented works, but is a bit > cumbersome. I hope to be able to polish the install routines more so > my users can perform installation themselves. For instance VMware > administrators. Supplying ip config on the kernel options might work, > but I would prefer nice dialog interfaces. > > Any suggestions for how to best kickstart systems from satellite > server from a custom boot-iso are welcome. > > Thanks, > > Erling > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > Using "cobbler buildiso" to create a boot CD with all of your systems on it would work for the above. You just have to pick what system you are installing from the menu at boot time and the install would be fully-automatic from there. This requires cobbler, though Satellite 5.3 will have it embedded (making this a supported feature). Until then you'll have to stand up a cobbler box and just have those systems register to Satellite in %post (by calling rhnreg_ks). We've also done some work in the past doing similar with a live CD (so you don't have to reburn the ISO when you add systems), but this needs some attention to get it working again. Regardless, this is nice because the kickstarts are sourced from the boot server, so the only thing that is really on the CD (burnt in) are the install kernels/initrds and kernel arguments. So you can change the kickstarts without reburning the CDs. (Now if you have a /lot/ of systems, this is where we should really revive the PXE-simulating live CD, because then we can look up the network info via the MAC address.) --Michael -------------- next part -------------- An HTML attachment was scrubbed... URL: From erlingre at gmail.com Fri Jul 31 06:28:47 2009 From: erlingre at gmail.com (Erling Ringen Elvsrud) Date: Fri, 31 Jul 2009 08:28:47 +0200 Subject: Manual network config, fetch tcp/ip setup from stage1 in stage2 In-Reply-To: References: <664c5a070907300303v62aeee09j71d984beaea60874@mail.gmail.com> Message-ID: <664c5a070907302328l4ba046cbqc47c327674d35785@mail.gmail.com> On 7/30/09, Callahan, Tom wrote: > Nice dialog interfaces to me, don't mean kickstart. I view kickstart as a > set it and forget it install method, as I think most people use it for. > > If you want install dialogs, why not just boot the DVD and do the install > the old fashioned way? That is not an option in my case as it is too slow, requires more interaction, is less flexible and is not scriptable. The scripting features of kickstart is really useful. For instance I use the %post section to install vmware-tools if the system is running on vmware (I use dmidecode to detect this). Using kickstart even with some manual interaction also ensures more consistency in the installations so in my opinion it is a really good idea even for just a few systems. Best regards, Erling Ringen Elvsrud