From debian at herakles.homelinux.org Wed Apr 1 02:12:48 2009 From: debian at herakles.homelinux.org (John Summerfield) Date: Wed, 01 Apr 2009 10:12:48 +0800 Subject: Scripting the url --url install method? In-Reply-To: <21F8C5D33731F44BB9DCC396A12F9F2110A32D15@WLVEXM01.corp.ad.diginsite.com> References: <21F8C5D33731F44BB9DCC396A12F9F2110A32D15@WLVEXM01.corp.ad.diginsite.com> Message-ID: <49D2CDA0.9010702@herakles.homelinux.org> Hajducko, Steven wrote: > Is it possible to use a %pre script in any way to customize the install method? I'm attempting to create a kickstart file that we can move around our network, however, some of our networks require different URL's to grab the ks file and to find the install media. > > What I have currently isn't working as anaconda doesn't seem to like that it can't find the install method before it runs the %pre section. > > Here is what I have currently - > > install > %include /tmp/url > > ... > > %pre > > # Append network hostname as some tiers don't have DNS - if we don't have DNS > # then the hostname becomes 'localhost' when using the network information from the > # boot line > HOSTNAME=$( cat /proc/cmdline | sed -e s'/ /\n/g' | grep '^hostname=' ) > echo "network --${HOSTNAME}" > /tmp/hostname > > # Get our architecture > ARCH=$(uname -i) > > # Figure out the kickstart URL from the boot line so we can find the install media > KSURL=$(cat /proc/cmdline | sed -e 's/ /\n/g' | grep '^ks=' | cut -d= -f2 | cut -d/ -f1,2,3) > echo "url --url=${KSURL}/yum/rhel5-server-${ARCH}/disc1" > /tmp/url > > ... > > We boot from the DVD and enter the network information on the linux: boot command, so ultimately our boot line looks something like: > > linux ks=http://yum.example.com:9300/ks/hp-rhel5.cfg hostname=test1.example.com ip=1.1.1.2 netmask=255.255.255.0 gateway=1.1.1.1 dns=1.1.1.3 > > I don't see any errors when viewing the logs - it just prompts me for the install method. > > I've also tried something like the following in hopes that when it re-read the kickstart file, the URL line existed and it would use that instead - unfortunately that didn't work either. > > install > %include /tmp/url > cdrom > Better, ks=http://kickstart.example.com/cgi/ks? ks could take into account the source IP address in framing its response. dhcpd can be configured to give different responses depending on who's asking: PXE can get one set of responses, anaconda another, debian-installer a third, Windows can get something completely different. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu -- Advice http://webfoot.com/advice/email.top.php http://www.catb.org/~esr/faqs/smart-questions.html http://support.microsoft.com/kb/555375 You cannot reply off-list:-) From steven.hajducko at digitalinsight.com Thu Apr 2 18:29:03 2009 From: steven.hajducko at digitalinsight.com (Hajducko, Steven) Date: Thu, 2 Apr 2009 11:29:03 -0700 Subject: Scripting the url --url install method? In-Reply-To: <49D2CDA0.9010702@herakles.homelinux.org> References: <21F8C5D33731F44BB9DCC396A12F9F2110A32D15@WLVEXM01.corp.ad.diginsite.com> <49D2CDA0.9010702@herakles.homelinux.org> Message-ID: <21F8C5D33731F44BB9DCC396A12F9F2110A3324A@WLVEXM01.corp.ad.diginsite.com> Yeah, I finally understood what that little note on the %pre section in the Kickstart manual meant - The pre-installation script section of kickstart cannot manage multiple install trees or source media. This information must be included for each created ks.cfg file, as the pre-installation script occurs during the second stage of the installation process. So I ended up creating a ks.php that generated the cfg based on the requesting IP and just return that. -- sh > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John > Summerfield > Sent: Tuesday, March 31, 2009 7:13 PM > To: Discussion list about Kickstart > Subject: Re: Scripting the url --url install method? > > Hajducko, Steven wrote: > > Is it possible to use a %pre script in any way to customize > the install method? I'm attempting to create a kickstart > file that we can move around our network, however, some of > our networks require different URL's to grab the ks file and > to find the install media. > > > > What I have currently isn't working as anaconda doesn't > seem to like that it can't find the install method before it > runs the %pre section. > > > > Here is what I have currently - > > > > install > > %include /tmp/url > > > > ... > > > > %pre > > > > # Append network hostname as some tiers don't have DNS - if > we don't > > have DNS # then the hostname becomes 'localhost' when using the > > network information from the # boot line HOSTNAME=$( cat > /proc/cmdline > > | sed -e s'/ /\n/g' | grep '^hostname=' ) echo "network > --${HOSTNAME}" > > > /tmp/hostname > > > > # Get our architecture > > ARCH=$(uname -i) > > > > # Figure out the kickstart URL from the boot line so we can > find the > > install media KSURL=$(cat /proc/cmdline | sed -e 's/ /\n/g' | grep > > '^ks=' | cut -d= -f2 | cut -d/ -f1,2,3) echo "url > > --url=${KSURL}/yum/rhel5-server-${ARCH}/disc1" > /tmp/url > > > > ... > > > > We boot from the DVD and enter the network information on > the linux: boot command, so ultimately our boot line looks > something like: > > > > linux ks=http://yum.example.com:9300/ks/hp-rhel5.cfg > > hostname=test1.example.com ip=1.1.1.2 netmask=255.255.255.0 > > gateway=1.1.1.1 dns=1.1.1.3 > > > > I don't see any errors when viewing the logs - it just > prompts me for the install method. > > > > I've also tried something like the following in hopes that > when it re-read the kickstart file, the URL line existed and > it would use that instead - unfortunately that didn't work either. > > > > install > > %include /tmp/url > > cdrom > > > > Better, > ks=http://kickstart.example.com/cgi/ks? ks > could take into account the source IP address in framing its response. > > dhcpd can be configured to give different responses depending on who's > asking: PXE can get one set of responses, anaconda another, > debian-installer a third, Windows can get something > completely different. > > > > -- > > Cheers > John > > -- spambait > 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu > -- Advice > http://webfoot.com/advice/email.top.php > http://www.catb.org/~esr/faqs/smart-questions.html > http://support.microsoft.com/kb/555375 > > You cannot reply off-list:-) > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From srs at redhat.com Mon Apr 6 21:53:00 2009 From: srs at redhat.com (Srinivas Satyavarpu) Date: Mon, 06 Apr 2009 14:53:00 -0700 Subject: How to blacklist a driver (say emulex, lpfc) during anaconda/kickstart/as or a kernel parameter during re-imaging a box Message-ID: <49DA79BC.3030201@redhat.com> Team Does any one know how to blacklist (do not load) a module/driver ( say emulex driver lpfc) i.e. not to install at the time of re-imaging a box. Some where in kickstart file or in the kernel parameter line or anaconda. Thx Srini -- http://www.redhat.com/promo/summit/2009/ Srinivas Satyavarpu[Srini] Solutions Architect Direct:650.254.4249 Red Hat, Inc. Cell: 408.464.8553 444, Castro Street,12th Floor Email: srs at redhat.com Mountain View,CA,94041 Web: www.redhat.com From jmaestri at soleranetworks.com Tue Apr 7 00:26:58 2009 From: jmaestri at soleranetworks.com (Jason Maestri) Date: Mon, 6 Apr 2009 17:26:58 -0700 Subject: Anaconda unable to mount stage2. Message-ID: Hello List, I'm seeing an error when trying to install off of a hard drive. I am currently installing with my kickstart file inside of my initrd, and my installation media is a hard drive connected to the system via USB. My kernel line (in grub) is as follows: "kernel vmlinuz ks=file:/kickstarts/install.ks stage2=hd:/dev/sdd1:/images/stage2.img". The error I am seeing indicates that the system was unable to mount the stage2 image because the loop device (/dev/loop0) it was attempting to use does not exist. Here is the actual text of the error: 16:04:21 INFO : mounting device sdd1 for hard drive install 16:05:51 INFO : Path to stage 2 image is /mnt/isodir/images/stage2.img 16:05:51 ERROR : failed to mount loopack device /dev/loop0 on /mnt/runtime as /tmp/install.img: /dev/loop0: No such file or directory 16:05:54 INFO : unmounting loopback /mnt/runtime /dev/loop0 16:05:54 ERROR : LOOP_CLR_FD failed for /mnt/runtime /dev/loop0: Bad file descriptor Does anyone know how to get around this? Am I doing this wrong? I appreciate your help. Thank you, Jason Maestri -- Jason C. Maestri Senior Software Engineer Solera Networks O (801) 623.5192 C (801) 867.7654 http://www.soleranetworks.com The information in this e-mail and in any attachments is confidential and may be privileged. If you are not the intended recipient, please destroy this message, delete any copies held on your systems and notify the sender immediately. You should not retain copy or use this e-mail for any purpose, nor disclose all or any part of its content to any other person. From mikejennings at gmail.com Wed Apr 8 14:37:32 2009 From: mikejennings at gmail.com (Mike Jennings) Date: Wed, 8 Apr 2009 09:37:32 -0500 Subject: Sending Module Options through Anaconda Message-ID: <006A9EA4-07A2-4C21-931E-0F6332FA9989@gmail.com> Hello, I'm trying to install EL4 and I need to pass the module option, e1000 eeprom_bad_csum_allow=1. I've tried driver options in the ks.cfg as well as passing the option on the initrd append line. Is there another way of doing this? Mike Jennings mikejennings at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jlaska at redhat.com Wed Apr 8 14:44:37 2009 From: jlaska at redhat.com (James Laska) Date: Wed, 08 Apr 2009 10:44:37 -0400 Subject: Sending Module Options through Anaconda In-Reply-To: <006A9EA4-07A2-4C21-931E-0F6332FA9989@gmail.com> References: <006A9EA4-07A2-4C21-931E-0F6332FA9989@gmail.com> Message-ID: <1239201877.3196.111.camel@flatline.devel.redhat.com> On Wed, 2009-04-08 at 09:37 -0500, Mike Jennings wrote: > Hello, > > > I'm trying to install EL4 and I need to pass the module option, > e1000 eeprom_bad_csum_allow=1. I've tried driver options in the ks.cfg > as well as passing the option on the initrd append line. Is there > another way of doing this? If doing a kickstart install, I have seen people blacklist a module from loading (see https://fedoraproject.org/wiki/Anaconda/Options) and then manually load modules as needed in %pre. Similarly, I've seen folks use noprobe, and then manually load *all* drivers in %pre. Thanks, James -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part URL: From tom at ng23.net Thu Apr 9 11:16:38 2009 From: tom at ng23.net (Tom Brown) Date: Thu, 09 Apr 2009 12:16:38 +0100 Subject: Adding user with rpm in %packages section Message-ID: <49DDD916.1020301@ng23.net> Hi I have an rpm that creates a user as part of the %pre in the rpm like so %pre getent group orca >/dev/null || groupadd -r orca getent passwd orca >/dev/null || useradd -r -g orca -d /var/orca -s /bin/bash -c "Orca Application User" orca This works fine for when the package is installed in the %post or even after installation. However when i put this package into the %packages section of my kickstart the package fails to install with Installing procallator - 1.0-6.noarch /var/tmp/rpm-tmp.18620: line 1: groupadd: command not found /var/tmp/rpm-tmp.18620: line 2: useradd: command not found error: %pre(procallator-1.0-6.noarch) scriptlet failed, exit status 127 error: install: %pre scriptlet failed (2), skipping procallator-1.0-6 Now i presume this is something to do with groupadd and useradd not being available in anaconda, if so how can i resolve this so that this can be installed in the %packages section or later on with yum without changing the package? thanks From mrose at n-able.com Thu Apr 9 13:03:05 2009 From: mrose at n-able.com (Matt Rose) Date: Thu, 9 Apr 2009 09:03:05 -0400 Subject: Adding user with rpm in %packages section In-Reply-To: <49DDD916.1020301@ng23.net> References: <49DDD916.1020301@ng23.net> Message-ID: <49DDF209.2010409@n-able.com> You need to make sure adduser and addgroup are in the Requires: line of the spec file. Possibly you need a PreReq: line in the .spec file, but according to Jeff Johnson, that's not clear. Try putting adduser/group in the Requires line first. Matt Tom Brown wrote: > Hi > > I have an rpm that creates a user as part of the %pre in the rpm like so > > %pre > getent group orca >/dev/null || groupadd -r orca > getent passwd orca >/dev/null || useradd -r -g orca -d /var/orca -s > /bin/bash -c "Orca Application User" orca > > This works fine for when the package is installed in the %post or even > after installation. > > However when i put this package into the %packages section of my > kickstart the package fails to install with > > Installing procallator - 1.0-6.noarch > /var/tmp/rpm-tmp.18620: line 1: groupadd: command not found > /var/tmp/rpm-tmp.18620: line 2: useradd: command not found > error: %pre(procallator-1.0-6.noarch) scriptlet failed, exit status 127 > error: install: %pre scriptlet failed (2), skipping procallator-1.0-6 > > Now i presume this is something to do with groupadd and useradd not > being available in anaconda, if so how can i resolve this so that this > can be installed in the %packages section or later on with yum without > changing the package? > > thanks > > _______________________________________________ > 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 tom at ng23.net Thu Apr 9 14:42:38 2009 From: tom at ng23.net (Tom Brown) Date: Thu, 09 Apr 2009 15:42:38 +0100 Subject: Adding user with rpm in %packages section In-Reply-To: <49DDF209.2010409@n-able.com> References: <49DDD916.1020301@ng23.net> <49DDF209.2010409@n-able.com> Message-ID: <49DE095E.5010706@ng23.net> > You need to make sure adduser and addgroup are in the Requires: line > of the spec file. Possibly you need a PreReq: line in the .spec file, > but according to Jeff Johnson, that's not clear. Try putting > adduser/group in the Requires line first. > that did not seem to help - Requires: useradd groupadd will keep digging thanks From mrose at n-able.com Thu Apr 9 14:48:33 2009 From: mrose at n-able.com (Matt Rose) Date: Thu, 9 Apr 2009 10:48:33 -0400 Subject: Adding user with rpm in %packages section In-Reply-To: <49DE095E.5010706@ng23.net> References: <49DDD916.1020301@ng23.net> <49DDF209.2010409@n-able.com> <49DE095E.5010706@ng23.net> Message-ID: <49DE0AC1.6020001@n-able.com> Whoops! you need to specify the full path, or the package name, so either Requires: /usr/sbin/useradd,/usr/sbin/groupadd or Requires: shadow-utils Matt Tom Brown wrote: >> You need to make sure adduser and addgroup are in the Requires: line >> of the spec file. Possibly you need a PreReq: line in the .spec file, >> but according to Jeff Johnson, that's not clear. Try putting >> adduser/group in the Requires line first. >> >> > > that did not seem to help - > > Requires: useradd groupadd > > will keep digging > > thanks > > _______________________________________________ > 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 clumens at redhat.com Thu Apr 9 14:53:43 2009 From: clumens at redhat.com (Chris Lumens) Date: Thu, 9 Apr 2009 10:53:43 -0400 Subject: Adding user with rpm in %packages section In-Reply-To: <49DE0AC1.6020001@n-able.com> References: <49DDD916.1020301@ng23.net> <49DDF209.2010409@n-able.com> <49DE095E.5010706@ng23.net> <49DE0AC1.6020001@n-able.com> Message-ID: <20090409145343.GA3563@localhost.localdomain> > Whoops! you need to specify the full path, or the package name, so either > > Requires: /usr/sbin/useradd,/usr/sbin/groupadd > > or > > Requires: shadow-utils You really want to do Requires(pre): shadow-utils to make sure the package gets sorted correctly. - Chris From tom at ng23.net Thu Apr 9 15:32:43 2009 From: tom at ng23.net (Tom Brown) Date: Thu, 09 Apr 2009 16:32:43 +0100 Subject: Adding user with rpm in %packages section In-Reply-To: <20090409145343.GA3563@localhost.localdomain> References: <49DDD916.1020301@ng23.net> <49DDF209.2010409@n-able.com> <49DE095E.5010706@ng23.net> <49DE0AC1.6020001@n-able.com> <20090409145343.GA3563@localhost.localdomain> Message-ID: <49DE151B.7030704@ng23.net> > You really want to do Requires(pre): shadow-utils to make sure the > package gets sorted correctly. > > many thanks - that resolved it From tibbs at math.uh.edu Thu Apr 9 15:53:04 2009 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Thu, 09 Apr 2009 10:53:04 -0500 Subject: Adding user with rpm in %packages section In-Reply-To: <20090409145343.GA3563@localhost.localdomain> (Chris Lumens's message of "Thu\, 9 Apr 2009 10\:53\:43 -0400") References: <49DDD916.1020301@ng23.net> <49DDF209.2010409@n-able.com> <49DE095E.5010706@ng23.net> <49DE0AC1.6020001@n-able.com> <20090409145343.GA3563@localhost.localdomain> Message-ID: >>>>> "CL" == Chris Lumens writes: CL> You really want to do Requires(pre): shadow-utils to make sure the CL> package gets sorted correctly. Of tangential relevance, perhaps, but the Fedora packaging guidelines on this subject are at http://fedoraproject.org/wiki/Packaging:UsersAndGroups - J< From patchu1 at yahoo.com Mon Apr 13 20:36:26 2009 From: patchu1 at yahoo.com (Pat) Date: Mon, 13 Apr 2009 13:36:26 -0700 (PDT) Subject: Problem with updated driverdisk not being used Message-ID: <452299.53376.qm@web32506.mail.mud.yahoo.com> Quick Summary: RHEL 5.2 is using the default driver rather than the updated driver I specified using the driverdisk option in my kickstart file. Detailed Description: I have a system with a 3ware 9650SE RAID card which I'm installing RHEL 5.2 (x86_64) on. The system will be booting from this 3ware card. RHEL 5.2 comes with a default driver that can automatically detect and use this card. The driver is 3w-9xxx Driver version: 2.26.02.008. I'm trying to use an updated driver from 3ware, 2.26.08.004, by specifying the driverdisk option in my kickstart file: driverdisk --source=http://192.168.0.119/3waredriverdisk.img BTW, I generated the img driver disk using the instructions from the page http://www.ruizs.org/archives/49: #Create a blank, 20MB image dd if=/dev/zero of=/root/driverdisk.img bs=1M count=2 #Format the image with ext2 mkfs -t ext2 -q /root/driverdisk.img #mount it and copy the files over mount -o loop /root/driverdisk.img /mnt/tmp cp /root/3ware/* /mnt/tmp/ umount /mnt/tmp However after the installation is complete, the default 3ware driver is still being used. I should note the rest of my kickstart directives were followed correctly. I have an idea on what is going wrong. The kickstart command I use is: linux ks=http://192.168.0.119/mykickstart.cfg Right after I hit return, RHEL loads up the default 3w-9xxx driver. It did not even go on to the network to read my kickstart file, I know this since I'm prompted to choose which ethernet port to use by anaconda since there are two network ports on the system, the default 3w-9xxx driver loads before that. So I'm guessing my driverdisk option in my kickstart is ignored since a driver is already loaded. Perhaps anaconda would not have loaded the default 3ware driver if I had used the linux dd command at the prompt rather than specifying it in the kickstart? Does anyone have suggestions on a workaround? I can't use the linux dd command since I don't want to swap disk everytime. Thank for any assistance! Regards, Pat From msivak at redhat.com Wed Apr 15 10:28:49 2009 From: msivak at redhat.com (Martin Sivak) Date: Wed, 15 Apr 2009 06:28:49 -0400 (EDT) Subject: Problem with updated driverdisk not being used In-Reply-To: <452299.53376.qm@web32506.mail.mud.yahoo.com> Message-ID: <690714351.755711239791329850.JavaMail.root@zmail06.collab.prod.int.phx2.redhat.com> Hello Pat, yes, your idea is right on. Anaconda needs to load the network driver to access your kickstart file and even to access the driver disc. And it is not possible to replace earlier loaded driver using driverdisc. If you need to use updated NIC driver, avoid using network before the driver disc gets loaded. Which leaves you the options of "linux dd" or using driverdisc autodetection (just set the driverdisc's label to "oemdrv"). Martin Sivak Anaconda team ----- "Pat" wrote: > Quick Summary: > > RHEL 5.2 is using the default driver rather than the updated driver I > specified using the driverdisk option in my kickstart file. > > > Detailed Description: > > I have a system with a 3ware 9650SE RAID card which I'm installing > RHEL 5.2 (x86_64) on. The system will be booting from this 3ware card. > RHEL 5.2 comes with a default driver that can automatically detect and > use this card. The driver is 3w-9xxx Driver version: 2.26.02.008. > > I'm trying to use an updated driver from 3ware, 2.26.08.004, by > specifying the driverdisk option in my kickstart file: > > driverdisk --source=http://192.168.0.119/3waredriverdisk.img > > BTW, I generated the img driver disk using the instructions from the > page http://www.ruizs.org/archives/49: > > #Create a blank, 20MB image > dd if=/dev/zero of=/root/driverdisk.img bs=1M count=2 > #Format the image with ext2 > mkfs -t ext2 -q /root/driverdisk.img > #mount it and copy the files over > mount -o loop /root/driverdisk.img /mnt/tmp > cp /root/3ware/* /mnt/tmp/ > umount /mnt/tmp > > > However after the installation is complete, the default 3ware driver > is still being used. I should note the rest of my kickstart directives > were followed correctly. > > I have an idea on what is going wrong. The kickstart command I use > is: > > linux ks=http://192.168.0.119/mykickstart.cfg > > Right after I hit return, RHEL loads up the default 3w-9xxx driver. It > did not even go on to the network to read my kickstart file, I know > this since I'm prompted to choose which ethernet port to use by > anaconda since there are two network ports on the system, the default > 3w-9xxx driver loads before that. So I'm guessing my driverdisk option > in my kickstart is ignored since a driver is already loaded. Perhaps > anaconda would not have loaded the default 3ware driver if I had used > the linux dd command at the prompt rather than specifying it in the > kickstart? > > Does anyone have suggestions on a workaround? I can't use the linux dd > command since I don't want to swap disk everytime. Thank for any > assistance! > > Regards, > Pat > > > > > > > > > > > > > > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From patchu1 at yahoo.com Wed Apr 15 23:02:14 2009 From: patchu1 at yahoo.com (Pat) Date: Wed, 15 Apr 2009 16:02:14 -0700 (PDT) Subject: Problem with updated driverdisk not being used Message-ID: <489438.33702.qm@web32507.mail.mud.yahoo.com> Hi Martin, Thanks for the reply. I confirmed that using "linux dd" with the driver disk does indeed cause RHEL to use the newer driver for the 3ware RAID card. About driverdisc autodetection, I install RHEL 5.2 from a DVD which is the only drive I utilize, I'm not quite sure I can avoid having to use "linux dd" and then swapping discs in order to load in the driverdisc even if it has the label oemdrv. I'm trying to keep user interaction and discs at a minimum for this installation so having to swap discs is not an option for me. If they can be acquired over the network, that is a different story. Another solution I tried was copying my kickstart file to the RHEL install DVD. I also copied the contents of the driverdisc to the root of the install DVD as well: modules.alias modules.cgz modules.dep modules.pcimap pci.ids pcitable rhdd My kickstart file then contains the following line: driverdisk hda --type=iso9660 So anaconda does not need to go out onto the network for my kickstart file or my driver disk. This setup worked for me before when RHEL 4.x did not come with a driver for this RAID card. In this case, RHEL 5.2 does come with a default driver but I'm trying to use a newer driver with the driver disk. However when I tried this, the installer took quite awhile at the driver disk portion but it did continue on. After the install was complete and I booted into the OS, I verified the default driver for the 3ware card was still being used. It appears that unless you specify "linux dd" at the boot prompt, Anaconda loads in the default drivers before the kickstart file is read in no matter where that kickstart file is, even if the kickstart file is local to the install disc. Is that correct? It's possible my install DVD with the embedded kickstart file and driver disk was somehow incorrect, for example if something changed from RHEL from 4.x to 5.x. If you think I may have left something something out when I created it, let me know. I was thinking of trying it again but copy the img file of the driver disk to the install DVD rather than the individual files. In this case I would use this line in my kickstart file: driverdisk --source=hda:/3waredriverdisk.img But again, if the default drivers load before the kickstart file is read, this change would not solve my problem. Thanks, Pat Thanks, Pat > On Apr 15, 2009, at 3:28 AM, Martin Sivak wrote: > > Hello Pat, > > yes, your idea is right on. Anaconda needs to load the > network driver to access your kickstart file and even to > access the driver disc. And it is not possible to replace > earlier loaded driver using driverdisc. > > If you need to use updated NIC driver, avoid using network > before the driver disc gets loaded. Which leaves you the > options of "linux dd" or using driverdisc autodetection > (just set the driverdisc's label to "oemdrv"). > > Martin Sivak > Anaconda team > > ----- "Pat" > wrote: > > Quick Summary: > > RHEL 5.2 is using the default driver rather than the > updated driver I > specified using the driverdisk option in my kickstart > file. > > > Detailed Description: > > I have a system with a 3ware 9650SE RAID card which I'm > installing > RHEL 5.2 (x86_64) on. The system will be booting from this > 3ware card. > RHEL 5.2 comes with a default driver that can automatically > detect and > use this card. The driver is 3w-9xxx Driver version: > 2.26.02.008. > > I'm trying to use an updated driver from 3ware, > 2.26.08.004, by > specifying the driverdisk option in my kickstart file: > > driverdisk --source=http://192.168.0.119/3waredriverdisk.img > > BTW, I generated the img driver disk using the instructions > from the > page http://www.ruizs.org/archives/49: > > #Create a blank, 20MB image > dd if=/dev/zero of=/root/driverdisk.img bs=1M count=2 > #Format the image with ext2 > mkfs -t ext2 -q /root/driverdisk.img > #mount it and copy the files over > mount -o loop /root/driverdisk.img /mnt/tmp > cp /root/3ware/* /mnt/tmp/ > umount /mnt/tmp > > > However after the installation is complete, the default > 3ware driver > is still being used. I should note the rest of my kickstart > directives > were followed correctly. > > I have an idea on what is going wrong. The kickstart > command I use > is: > > linux ks=http://192.168.0.119/mykickstart.cfg > > Right after I hit return, RHEL loads up the default 3w-9xxx > driver. It > did not even go on to the network to read my kickstart > file, I know > this since I'm prompted to choose which ethernet port to > use by > anaconda since there are two network ports on the system, > the default > 3w-9xxx driver loads before that. So I'm guessing my > driverdisk option > in my kickstart is ignored since a driver is already > loaded. Perhaps > anaconda would not have loaded the default 3ware driver if > I had used > the linux dd command at the prompt rather than specifying > it in the > kickstart? > > Does anyone have suggestions on a workaround? I can't use > the linux dd > command since I don't want to swap disk everytime. Thank > for any > assistance! > > Regards, > Pat > > > > > > > > > > > > > > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > > > > From Floydsmith at aol.com Thu Apr 23 13:02:18 2009 From: Floydsmith at aol.com (Floydsmith at aol.com) Date: Thu, 23 Apr 2009 09:02:18 EDT Subject: Out of memory process 564 (loader) killed custom install core 10 Message-ID: While trying to kickstart install core 10 (I have done this many times with 9, 8 and 7) from image located on local harddrive I get (in VT4): <6>124554 pages non-shared <3>Out of memory: kill process 564 (loader) <3>Killed orocess 564 (loader) I have 512Meg and the exact same image installs find from DVD. Any help greatly appreciated in advance. Floyd ************** Big savings on Dell XPS Laptops and Desktops! (http://pr.atwola.com/promoclk/100126575x1219799634x1201361008/aol?redir=http:%2F%2Fad.doubl eclick.net%2Fclk%3B214133440%3B36002254%3Bj) -------------- next part -------------- An HTML attachment was scrubbed... URL: From anthony.seward at ieee.org Tue Apr 28 20:58:44 2009 From: anthony.seward at ieee.org (Tony Seward) Date: Tue, 28 Apr 2009 14:58:44 -0600 Subject: Obtaining the tracebacks from anaconda Message-ID: <1240952324.4214.134.camel@florete.mza.com> I'm getting a traceback when installing using kickstart and I can't figure out how to get the tracebacks that are in bugzilla. Can anyone help? Thanks, Tony From clumens at redhat.com Tue Apr 28 21:59:43 2009 From: clumens at redhat.com (Chris Lumens) Date: Tue, 28 Apr 2009 17:59:43 -0400 Subject: Obtaining the tracebacks from anaconda In-Reply-To: <1240952324.4214.134.camel@florete.mza.com> References: <1240952324.4214.134.camel@florete.mza.com> Message-ID: <20090428215943.GF5296@localhost.localdomain> > I'm getting a traceback when installing using kickstart and I can't > figure out how to get the tracebacks that are in bugzilla. Can anyone > help? What do you mean "get the tracebacks that are in bugzilla"? Do you want to query bugzilla for tracebacks? Or, do you want the file that would be attached to a bug that's filed automatically? If the latter, it's /tmp/anacdump.txt. - Chris From anthony.seward at ieee.org Tue Apr 28 22:10:35 2009 From: anthony.seward at ieee.org (Tony Seward) Date: Tue, 28 Apr 2009 16:10:35 -0600 Subject: Obtaining the tracebacks from anaconda In-Reply-To: <20090428215943.GF5296@localhost.localdomain> References: <1240952324.4214.134.camel@florete.mza.com> <20090428215943.GF5296@localhost.localdomain> Message-ID: <1240956635.4214.209.camel@florete.mza.com> The latter. However, I can't access the filesystem from terminal #2 once the traceback occurs. Tony On Tue, 2009-04-28 at 17:59 -0400, Chris Lumens wrote: > > I'm getting a traceback when installing using kickstart and I can't > > figure out how to get the tracebacks that are in bugzilla. Can anyone > > help? > > What do you mean "get the tracebacks that are in bugzilla"? Do you want > to query bugzilla for tracebacks? Or, do you want the file that would > be attached to a bug that's filed automatically? If the latter, it's > /tmp/anacdump.txt. > > - Chris > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From clumens at redhat.com Tue Apr 28 22:15:35 2009 From: clumens at redhat.com (Chris Lumens) Date: Tue, 28 Apr 2009 18:15:35 -0400 Subject: Obtaining the tracebacks from anaconda In-Reply-To: <1240956635.4214.209.camel@florete.mza.com> References: <1240952324.4214.134.camel@florete.mza.com> <20090428215943.GF5296@localhost.localdomain> <1240956635.4214.209.camel@florete.mza.com> Message-ID: <20090428221535.GG5296@localhost.localdomain> > The latter. However, I can't access the filesystem from terminal #2 > once the traceback occurs. You could use a %traceback script in your kickstart file to scp it to another computer. - Chris From anthony.seward at ieee.org Tue Apr 28 22:41:07 2009 From: anthony.seward at ieee.org (Tony Seward) Date: Tue, 28 Apr 2009 16:41:07 -0600 Subject: Obtaining the tracebacks from anaconda In-Reply-To: <1240956635.4214.209.camel@florete.mza.com> References: <1240952324.4214.134.camel@florete.mza.com> <20090428215943.GF5296@localhost.localdomain> <1240956635.4214.209.camel@florete.mza.com> Message-ID: <1240958467.4214.213.camel@florete.mza.com> Also, I tried adding %traceback mount /dev/sdb1 /tmp/mnt cp /tmp/anacdump.txt /tmp/mnt umount /tmp/mnt sleep 600 to my kickstart file, but the system can only be rebooted and I don't get anacdump.txt on my USB drive (/dev/sdb1). Tony On Tue, 2009-04-28 at 16:10 -0600, Tony Seward wrote: > The latter. However, I can't access the filesystem from terminal #2 > once the traceback occurs. > > Tony > > On Tue, 2009-04-28 at 17:59 -0400, Chris Lumens wrote: > > > I'm getting a traceback when installing using kickstart and I can't > > > figure out how to get the tracebacks that are in bugzilla. Can anyone > > > help? > > > > What do you mean "get the tracebacks that are in bugzilla"? Do you want > > to query bugzilla for tracebacks? Or, do you want the file that would > > be attached to a bug that's filed automatically? If the latter, it's > > /tmp/anacdump.txt. > > > > - Chris > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > From anthony.seward at ieee.org Wed Apr 29 17:07:32 2009 From: anthony.seward at ieee.org (Tony Seward) Date: Wed, 29 Apr 2009 11:07:32 -0600 Subject: Obtaining the tracebacks from anaconda In-Reply-To: <20090428221535.GG5296@localhost.localdomain> References: <1240952324.4214.134.camel@florete.mza.com> <20090428215943.GF5296@localhost.localdomain> <1240956635.4214.209.camel@florete.mza.com> <20090428221535.GG5296@localhost.localdomain> Message-ID: <1241024852.4331.35.camel@florete.mza.com> I'm having problems getting networking to work and that is why I was trying to write the traceback to a USB stick. It does not appear as if the code in my %traceback script is called. %traceback mount /dev/sdb1 /tmp/mnt cp /tmp/anacdump.txt /tmp/mnt umount /tmp/mnt sleep 600 Tony On Tue, 2009-04-28 at 18:15 -0400, Chris Lumens wrote: > > The latter. However, I can't access the filesystem from terminal #2 > > once the traceback occurs. > > You could use a %traceback script in your kickstart file to scp it to > another computer. > > - Chris > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list >