From clumens at redhat.com Mon Aug 4 14:23:30 2008 From: clumens at redhat.com (Chris Lumens) Date: Mon, 4 Aug 2008 10:23:30 -0400 Subject: [PATCH] Typo fix In-Reply-To: <48886A45.6010105@redhat.com> References: <48886A45.6010105@redhat.com> Message-ID: <20080804142330.GB3902@localhost.localdomain> > diff --git a/pykickstart/commands/repo.py b/pykickstart/commands/repo.py > index 031e6d3..aaf542a 100644 > --- a/pykickstart/commands/repo.py > +++ b/pykickstart/commands/repo.py > @@ -136,7 +136,7 @@ class F8_Repo(FC6_Repo): > > def methodToRepo(self): > if not self.handler.method.url: > - raise KickstartError, > formatErrorMsg(self.handler.method.lineno, msg=_("Method must be a url to > be added to the repo lsit.")) > + raise KickstartError, > formatErrorMsg(self.handler.method.lineno, msg=_("Method must be a url to > be added to the repo list.")) > reponame = "ks-method-url" > repourl = self.handler.method.url > rd = self.handler.RepoData(name=reponame, baseurl=repourl) Applied, thanks. - Chris From chris at wavestore.com Mon Aug 4 14:41:17 2008 From: chris at wavestore.com (Christopher Mocock) Date: Mon, 04 Aug 2008 15:41:17 +0100 Subject: Disk controller detection order Message-ID: <4897150D.5000606@wavestore.com> Hi, Using revisor, I've composed a respin which uses my custom kernel package. Most of the systems I'm installing to have an IDE drive of 40-80GB and a bunch of SATA disks, either directly connected to the motherboard or via a 3Ware RAID card. My kickstart has something like the following: zerombr clearpart --all --drives=sda part /var/log --fstype ext3 --size=120 --ondisk=sda part / --fstype ext3 --size=10000 --ondisk=sda part swap --size=128 --grow --maxsize=256 --ondisk=sda So everything should get installed to sda, which I would hope would always be the first IDE disk. However, I'm finding that if I install to a system which has a 3Ware RAID card, the OS gets installed to the first SATA disk on the 3Ware. The %post script in my custom kernel rpms finds the UUID of /dev/sda1 by running /lib/udev/vol_id -u /dev/sda1 and then puts it as the root= parameter in grub.conf. When I then try to boot from my new installation, the system boots from the IDE disk, since that is configured as the first boot device in BIOS. Any ideas as to how I can reliably persuade anaconda that /dev/sda is the primary master IDE device rather than picking up the 3Ware SATA drives first? I noticed there's a --onbiosdisk option, but the docs don't make clear what this parameter should be. Thanks, -- Chris From terryb at gg.specsavers.com Mon Aug 4 15:35:26 2008 From: terryb at gg.specsavers.com (Terry Black) Date: Mon, 4 Aug 2008 16:35:26 +0100 Subject: Disk controller detection order In-Reply-To: <4897150D.5000606@wavestore.com> Message-ID: <1262125576.260191217864126831.JavaMail.root@ajax.gg.specsavers.com> IDE drives should be hda not sda? Terry ----- Original Message ----- From: Christopher Mocock Sent: Mon, 4/8/2008 15:41 To: kickstart-list at redhat.com Subject: Disk controller detection order Hi, Using revisor, I've composed a respin which uses my custom kernel package. Most of the systems I'm installing to have an IDE drive of 40-80GB and a bunch of SATA disks, either directly connected to the motherboard or via a 3Ware RAID card. My kickstart has something like the following: zerombr clearpart --all --drives=sda part /var/log --fstype ext3 --size=120 --ondisk=sda part / --fstype ext3 --size=10000 --ondisk=sda part swap --size=128 --grow --maxsize=256 --ondisk=sda So everything should get installed to sda, which I would hope would always be the first IDE disk. However, I'm finding that if I install to a system which has a 3Ware RAID card, the OS gets installed to the first SATA disk on the 3Ware. The %post script in my custom kernel rpms finds the UUID of /dev/sda1 by running /lib/udev/vol_id -u /dev/sda1 and then puts it as the root= parameter in grub.conf. When I then try to boot from my new installation, the system boots from the IDE disk, since that is configured as the first boot device in BIOS. Any ideas as to how I can reliably persuade anaconda that /dev/sda is the primary master IDE device rather than picking up the 3Ware SATA drives first? I noticed there's a --onbiosdisk option, but the docs don't make clear what this parameter should be. Thanks, -- Chris _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list ----------------------------------------- NOTICE: This message contains privileged and confidential information intended only for the addressee. If you have received this message in error you must not disseminate, copy or take action on it; please notify postmaster at specsavers.com Opinions expressed in this message are those of the sender and do not necessarily represent those of Specsavers. Although this e-mail and any attachments are believed to be virus free, e-mail communications are not 100% secure and Specsavers makes no warranty that this message is secure or virus free. All references to Specsavers mean Specsavers Optical Group Limited or any member of the Specsavers Group of Companies. Nothing in this transmission shall or shall be deemed to constitute an offer or acceptance of an offer or otherwise have the effect of forming a contract by electronic communication. Your name and address may be stored to facilitate communications. From chris at wavestore.com Mon Aug 4 15:51:05 2008 From: chris at wavestore.com (Christopher Mocock) Date: Mon, 04 Aug 2008 16:51:05 +0100 Subject: Disk controller detection order In-Reply-To: <1262125576.260191217864126831.JavaMail.root@ajax.gg.specsavers.com> References: <1262125576.260191217864126831.JavaMail.root@ajax.gg.specsavers.com> Message-ID: <48972569.8070909@wavestore.com> Terry Black wrote: > IDE drives should be hda not sda? > Terry I should have mentioned I'm using Fedora 9 and I believe all devices are sda onwards, are they not? -- Chris From klaus.steden at thomson.net Mon Aug 4 19:00:22 2008 From: klaus.steden at thomson.net (Klaus Steden) Date: Mon, 04 Aug 2008 12:00:22 -0700 Subject: Disk controller detection order In-Reply-To: <48972569.8070909@wavestore.com> Message-ID: Hello Chris, Unless the Linux kernel's disk device handling code has changed recently, IDE/ATA drives are still hd[a-z]; sd[a-z] are for SCSI, FC, SATA, iSCSI, etc. disks. cheers, Klaus On 8/4/08 8:51 AM, "Christopher Mocock" did etch on stone tablets: > Terry Black wrote: >> IDE drives should be hda not sda? >> Terry > > I should have mentioned I'm using Fedora 9 and I believe all devices are > sda onwards, are they not? > From clumens at redhat.com Mon Aug 4 19:20:01 2008 From: clumens at redhat.com (Chris Lumens) Date: Mon, 4 Aug 2008 15:20:01 -0400 Subject: Disk controller detection order In-Reply-To: References: <48972569.8070909@wavestore.com> Message-ID: <20080804192001.GC3902@localhost.localdomain> > Unless the Linux kernel's disk device handling code has changed recently, > IDE/ATA drives are still hd[a-z]; sd[a-z] are for SCSI, FC, SATA, iSCSI, > etc. disks. It has. libata makes everything be sd* now. - Chris From klaus.steden at thomson.net Mon Aug 4 19:30:41 2008 From: klaus.steden at thomson.net (Klaus Steden) Date: Mon, 04 Aug 2008 12:30:41 -0700 Subject: Disk controller detection order In-Reply-To: <20080804192001.GC3902@localhost.localdomain> Message-ID: Cool, thanks. When did this get migrated into the Enterprise branch? 5.0? Klaus On 8/4/08 12:20 PM, "Chris Lumens" did etch on stone tablets: >> Unless the Linux kernel's disk device handling code has changed recently, >> IDE/ATA drives are still hd[a-z]; sd[a-z] are for SCSI, FC, SATA, iSCSI, >> etc. disks. > > It has. libata makes everything be sd* now. > > - Chris > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From pbdlists at pinboard.com Mon Aug 4 20:38:36 2008 From: pbdlists at pinboard.com (pbdlists at pinboard.com) Date: Mon, 4 Aug 2008 22:38:36 +0200 Subject: Disk controller detection order In-Reply-To: <4897150D.5000606@wavestore.com> References: <4897150D.5000606@wavestore.com> Message-ID: <20080804203836.GA20366@pinboard.com> Hi Chris, I have an old machine with SCSI disks on an adaptec controller and recently added three SATA disks with a promise SATA controller add-on card. With this machine I have the same issue. The machine boots off the SCSI disk and sees this as sda, while during the installation it assigns sda/sdb/sdc to the three SATA disks and sdd to the SCSI disk. I believe that somehow controller detection in the installation kernel and the installed kernel is not in the same order. Rather than taking everything apart and trying to build a new installation initrd or whatever is needed I changed the kicktart file for this machine to install to sdd, which will be sda again after the installation ||bootloader --location=mbr --driveorder=sdd || ||clearpart --all --drives=sdd ||part /boot --ondisk=sdd --size=120 --asprimary --fstype ext3 ||part pv.00 --ondisk=sdd --size=8192 --grow Maybe this approach works for you as well if you know which machines need this hack. Cheers, Kurt On Mon, Aug 04, 2008 at 03:41:17PM +0100, Christopher Mocock wrote: > Hi, > > Using revisor, I've composed a respin which uses my custom kernel > package. Most of the systems I'm installing to have an IDE drive of > 40-80GB and a bunch of SATA disks, either directly connected to the > motherboard or via a 3Ware RAID card. > > My kickstart has something like the following: > > zerombr > clearpart --all --drives=sda > part /var/log --fstype ext3 --size=120 --ondisk=sda > part / --fstype ext3 --size=10000 --ondisk=sda > part swap --size=128 --grow --maxsize=256 --ondisk=sda > > So everything should get installed to sda, which I would hope would > always be the first IDE disk. However, I'm finding that if I install to > a system which has a 3Ware RAID card, the OS gets installed to the first > SATA disk on the 3Ware. The %post script in my custom kernel rpms finds > the UUID of /dev/sda1 by running /lib/udev/vol_id -u /dev/sda1 and then > puts it as the root= parameter in grub.conf. > When I then try to boot from my new installation, the system boots from > the IDE disk, since that is configured as the first boot device in BIOS. > > Any ideas as to how I can reliably persuade anaconda that /dev/sda is > the primary master IDE device rather than picking up the 3Ware SATA > drives first? > > I noticed there's a --onbiosdisk option, but the docs don't make clear > what this parameter should be. > > Thanks, > > -- > Chris > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From chris at wavestore.com Tue Aug 5 08:37:25 2008 From: chris at wavestore.com (Christopher Mocock) Date: Tue, 05 Aug 2008 09:37:25 +0100 Subject: Disk controller detection order In-Reply-To: References: Message-ID: <48981145.1030603@wavestore.com> Klaus Steden wrote: > > Cool, thanks. When did this get migrated into the Enterprise branch? 5.0? I don't use RHEL so can't answer definitively but I don't think this change has hit RHEL yet. It appeared in Fedora 7 as I recall so it's probably not too far off for RHEL. -- Chris From chris at wavestore.com Tue Aug 5 10:47:53 2008 From: chris at wavestore.com (Christopher Mocock) Date: Tue, 05 Aug 2008 11:47:53 +0100 Subject: Disk controller detection order In-Reply-To: <20080804203836.GA20366@pinboard.com> References: <4897150D.5000606@wavestore.com> <20080804203836.GA20366@pinboard.com> Message-ID: <48982FD9.7020403@wavestore.com> pbdlists at pinboard.com wrote: > Hi Chris, > > I have an old machine with SCSI disks on an adaptec controller and > recently added three SATA disks with a promise SATA controller add-on > card. With this machine I have the same issue. The machine boots off > the SCSI disk and sees this as sda, while during the installation it > assigns sda/sdb/sdc to the three SATA disks and sdd to the SCSI disk. > > I believe that somehow controller detection in the installation kernel > and the installed kernel is not in the same order. Interesting. I can't really do the hack you mention because my custom spin has to go on lots of machines with different numbers of drives so I can't make any such assumptions. Thanks for the info anyway. The only workaround I can think of is disconnecting all the drives from the RAID card, then doing the installation, then replacing the drives. Far from ideal. Not really sure where to go from here. -- Chris From buchholz at easystreet.net Tue Aug 5 16:15:25 2008 From: buchholz at easystreet.net (Don Buchholz) Date: Tue, 05 Aug 2008 09:15:25 -0700 Subject: Disk controller detection order In-Reply-To: <48982FD9.7020403@wavestore.com> References: <4897150D.5000606@wavestore.com> <20080804203836.GA20366@pinboard.com> <48982FD9.7020403@wavestore.com> Message-ID: <48987C9D.2090907@easystreet.net> Christopher Mocock wrote: > pbdlists at pinboard.com wrote: >> Hi Chris, >> >> I have an old machine with SCSI disks on an adaptec controller and >> recently added three SATA disks with a promise SATA controller add-on >> card. With this machine I have the same issue. The machine boots off >> the SCSI disk and sees this as sda, while during the installation it >> assigns sda/sdb/sdc to the three SATA disks and sdd to the SCSI disk. >> >> I believe that somehow controller detection in the installation kernel >> and the installed kernel is not in the same order. > > Interesting. I can't really do the hack you mention because my custom > spin has to go on lots of machines with different numbers of drives so I > can't make any such assumptions. Thanks for the info anyway. > > The only workaround I can think of is disconnecting all the drives from > the RAID card, then doing the installation, then replacing the drives. > Far from ideal. > > Not really sure where to go from here. > > Here's a shot in the dark ... would UDev custom rules solve your problem? I can't say I've spent any serious time doing udev customization, we were looking at hotplug handling, and I ran across these pages while researching that topic. If I read it correctly, you could insert your own rules to override the default naming. However, since you state your environment has serious challenges with non-uniform hardware configurations, this might just be another dead end. Of course, the UDev management might just kick in too late during a normal system boot ... My notes w/ some URLs are below. - Don ================================= -- From: http://www.wlug.org.nz/UDev ... udev is the UserSpace-based replacement for DevFs, an obsolete Linux Kernel Module that creates entries for available devices in /dev. udev is available in Kernel 2.5 and newer and uses HotPlug and the SysFs FileSystem. The home page for udev is at: http://www.kernel.org/pub/linux/utils/kernel/hotplug/udev.html See also: * Red Hat Magazine - Configuring Devices with udev -- http://www.redhat.com/magazine/002dec04/features/udev/ * Writing udev rules -- http://www.reactivated.net/udevrules.php * Gentoo Customizing udev HOWTO -- http://gentoo-wiki.com/HOWTO_Customizing_UDEV ... From rmercer at harris.com Tue Aug 5 17:11:01 2008 From: rmercer at harris.com (Rodney Mercer) Date: Tue, 05 Aug 2008 13:11:01 -0400 Subject: Kickstart-list Digest, Vol 54, Issue 2 In-Reply-To: <20080805160020.B8CDC61AAA7@hormel.redhat.com> References: <20080805160020.B8CDC61AAA7@hormel.redhat.com> Message-ID: <1217956261.11867.15.camel@ratbert.evn.harris.com> On Tue, 2008-08-05 at 12:00 -0400, Christopher Mocock wrote: > Interesting. I can't really do the hack you mention because my > custom > spin has to go on lots of machines with different numbers of > drives so I > can't make any such assumptions. Thanks for the info anyway. > > The only workaround I can think of is disconnecting all the > drives from > the RAID card, then doing the installation, then replacing the > drives. > Far from ideal. > > Not really sure where to go from here. > > > -- > Chris > Chris, Possibly, you can use the "set $(list-harddrives)" in your kickstart config %pre section to determine number of disks and their capacities. >From that, you may be able to create some logic to create a partitioning information file that could be included for the disk setup you desire based on each machines disk arrangement. Check out the following post for ideas and google "set $(list-harddrives)" for many more. Regards, Rodney. From luis.f.correia at gmail.com Wed Aug 6 16:17:59 2008 From: luis.f.correia at gmail.com (Luis Correia) Date: Wed, 6 Aug 2008 17:17:59 +0100 Subject: anaconda screens Message-ID: Hi, I'm about to do some funky %pre stuff in python, to decide whetner or not to install the machine and need some user interaction. We're only doing it in text mode, no need for fancy graphics. How do I from the %pre script write stuff to the screen, get input, maybe a button or two ? I've rummaged through the anaconda code, but it all seems a bit strange to me. Thanks for all the help you could provide, Luis Correia From jgranado at redhat.com Thu Aug 7 09:23:43 2008 From: jgranado at redhat.com (Joel Andres Granados) Date: Thu, 07 Aug 2008 11:23:43 +0200 Subject: anaconda screens In-Reply-To: References: Message-ID: <489ABF1F.5030504@redhat.com> Luis Correia wrote: > Hi, > > I'm about to do some funky %pre stuff in python, to decide whetner or > not to install the machine and need some user interaction. > We're only doing it in text mode, no need for fancy graphics. > > How do I from the %pre script write stuff to the screen, get input, > maybe a button or two ? > > I've rummaged through the anaconda code, but it all seems a bit strange to me. > > Thanks for all the help you could provide, > Luis Correia > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list Took a quick look at the code and it seems that it executes whatever is in pre with a redirect to tty3, including the input. This suggests to me that you can basically do a bash read (or however you want to prompt the user) and it will prompt you on the vt3. It wont work if you are installing through serial console though. Didn't actually test this. Hope it helps. -- Joel Andres Granados Red Hat / Brno, Czech Republic From luis.f.correia at gmail.com Thu Aug 7 09:37:20 2008 From: luis.f.correia at gmail.com (Luis Correia) Date: Thu, 7 Aug 2008 10:37:20 +0100 Subject: anaconda screens In-Reply-To: <489ABF1F.5030504@redhat.com> References: <489ABF1F.5030504@redhat.com> Message-ID: Hi Joel, On Thu, Aug 7, 2008 at 10:23, Joel Andres Granados wrote: > Luis Correia wrote: >> >> Hi, >> >> I'm about to do some funky %pre stuff in python, to decide whetner or >> not to install the machine and need some user interaction. >> We're only doing it in text mode, no need for fancy graphics. >> >> How do I from the %pre script write stuff to the screen, get input, >> maybe a button or two ? >> >> I've rummaged through the anaconda code, but it all seems a bit strange to >> me. >> >> Thanks for all the help you could provide, >> Luis Correia >> >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list > > Took a quick look at the code and it seems that it executes whatever is in > pre with a redirect to tty3, including the input. This suggests to me that > you can basically do a bash read (or however you want to prompt the user) > and it will prompt you on the vt3. It wont work if you are installing > through serial console though. Didn't actually test this. > > Hope it helps. So, that means that, in order to interact with the user in the main tty, I must modify Anaconda, right? If yes, then I could probably use the 'updates' method... > -- > Joel Andres Granados > Red Hat / Brno, Czech Republic Luis Correia From Pablo.Iranzo at redhat.com Thu Aug 7 09:39:34 2008 From: Pablo.Iranzo at redhat.com (Pablo Iranzo =?ISO-8859-1?Q?G=F3mez?=) Date: Thu, 07 Aug 2008 11:39:34 +0200 Subject: anaconda screens In-Reply-To: References: <489ABF1F.5030504@redhat.com> Message-ID: <1218101974.23784.109.camel@iranzo.users.redhat.com> No Luis, in your %pre script you can make a switch to vt3 -> chvt 3 do whatever echo and reads you want, and return back to anaconda with chvt 1 Regards Pablo El jue, 07-08-2008 a las 10:37 +0100, Luis Correia escribi?: > Hi Joel, > > On Thu, Aug 7, 2008 at 10:23, Joel Andres Granados wrote: > > Luis Correia wrote: > >> > >> Hi, > >> > >> I'm about to do some funky %pre stuff in python, to decide whetner or > >> not to install the machine and need some user interaction. > >> We're only doing it in text mode, no need for fancy graphics. > >> > >> How do I from the %pre script write stuff to the screen, get input, > >> maybe a button or two ? > >> > >> I've rummaged through the anaconda code, but it all seems a bit strange to > >> me. > >> > >> Thanks for all the help you could provide, > >> Luis Correia > >> > >> _______________________________________________ > >> Kickstart-list mailing list > >> Kickstart-list at redhat.com > >> https://www.redhat.com/mailman/listinfo/kickstart-list > > > > Took a quick look at the code and it seems that it executes whatever is in > > pre with a redirect to tty3, including the input. This suggests to me that > > you can basically do a bash read (or however you want to prompt the user) > > and it will prompt you on the vt3. It wont work if you are installing > > through serial console though. Didn't actually test this. > > > > Hope it helps. > > So, that means that, in order to interact with the user in the main > tty, I must modify Anaconda, right? > If yes, then I could probably use the 'updates' method... > > > -- > > Joel Andres Granados > > Red Hat / Brno, Czech Republic > > Luis Correia > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Pablo Iranzo G?mez (Pablo.Iranzo at redhat.com) RHCE/RHCSP/RHCSS Global Profesional Services Consultant Spain Phone: +34 645 01 01 49 (CET/CEST) GnuPG KeyID: 0xFAD3CF0D -- Inscrita en el Reg. Mercantil de Madrid ? C.I.F. B-82 65 79 41 Directores: Michael Cunningham, Charlie Peters y David Owens Direcci?n Registrada: Red Hat S.L., C/ Velazquez 63, Madrid 28001, Espa?a Direcci?n contacto: C/Jose Bardasano Baos, 9, Edif. Gorbea 3, Planta 3?D, 28016 Madrid, Spain -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Esta parte del mensaje est? firmada digitalmente URL: From Pablo.Iranzo at redhat.com Thu Aug 7 09:41:06 2008 From: Pablo.Iranzo at redhat.com (Pablo Iranzo =?ISO-8859-1?Q?G=F3mez?=) Date: Thu, 07 Aug 2008 11:41:06 +0200 Subject: anaconda screens In-Reply-To: <1218101974.23784.109.camel@iranzo.users.redhat.com> References: <489ABF1F.5030504@redhat.com> <1218101974.23784.109.camel@iranzo.users.redhat.com> Message-ID: <1218102066.23784.111.camel@iranzo.users.redhat.com> Of course, on vt3 you can 'simulate' anaconda screens ;). If you really need them on vt1 will need to modify anaconda, but IMHO that would make it harder to debug and mantain in the future. Regards Pablo El jue, 07-08-2008 a las 11:39 +0200, Pablo Iranzo G?mez escribi?: > No Luis, in your %pre script you can make a switch to vt3 -> chvt 3 > do whatever echo and reads you want, and return back to anaconda with > chvt 1 > > > Regards > Pablo > > > El jue, 07-08-2008 a las 10:37 +0100, Luis Correia escribi?: > > Hi Joel, > > > > On Thu, Aug 7, 2008 at 10:23, Joel Andres Granados wrote: > > > Luis Correia wrote: > > >> > > >> Hi, > > >> > > >> I'm about to do some funky %pre stuff in python, to decide whetner or > > >> not to install the machine and need some user interaction. > > >> We're only doing it in text mode, no need for fancy graphics. > > >> > > >> How do I from the %pre script write stuff to the screen, get input, > > >> maybe a button or two ? > > >> > > >> I've rummaged through the anaconda code, but it all seems a bit strange to > > >> me. > > >> > > >> Thanks for all the help you could provide, > > >> Luis Correia > > >> > > >> _______________________________________________ > > >> Kickstart-list mailing list > > >> Kickstart-list at redhat.com > > >> https://www.redhat.com/mailman/listinfo/kickstart-list > > > > > > Took a quick look at the code and it seems that it executes whatever is in > > > pre with a redirect to tty3, including the input. This suggests to me that > > > you can basically do a bash read (or however you want to prompt the user) > > > and it will prompt you on the vt3. It wont work if you are installing > > > through serial console though. Didn't actually test this. > > > > > > Hope it helps. > > > > So, that means that, in order to interact with the user in the main > > tty, I must modify Anaconda, right? > > If yes, then I could probably use the 'updates' method... > > > > > -- > > > Joel Andres Granados > > > Red Hat / Brno, Czech Republic > > > > Luis Correia > > > > _______________________________________________ > > 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 -- Pablo Iranzo G?mez (Pablo.Iranzo at redhat.com) RHCE/RHCSP/RHCSS Global Profesional Services Consultant Spain Phone: +34 645 01 01 49 (CET/CEST) GnuPG KeyID: 0xFAD3CF0D -- Inscrita en el Reg. Mercantil de Madrid ? C.I.F. B-82 65 79 41 Directores: Michael Cunningham, Charlie Peters y David Owens Direcci?n Registrada: Red Hat S.L., C/ Velazquez 63, Madrid 28001, Espa?a Direcci?n contacto: C/Jose Bardasano Baos, 9, Edif. Gorbea 3, Planta 3?D, 28016 Madrid, Spain -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Esta parte del mensaje est? firmada digitalmente URL: From luis.f.correia at gmail.com Thu Aug 7 09:46:44 2008 From: luis.f.correia at gmail.com (Luis Correia) Date: Thu, 7 Aug 2008 10:46:44 +0100 Subject: anaconda screens In-Reply-To: <1218102066.23784.111.camel@iranzo.users.redhat.com> References: <489ABF1F.5030504@redhat.com> <1218101974.23784.109.camel@iranzo.users.redhat.com> <1218102066.23784.111.camel@iranzo.users.redhat.com> Message-ID: Hi! On Thu, Aug 7, 2008 at 10:41, Pablo Iranzo G?mez wrote: > Of course, on vt3 you can 'simulate' anaconda screens ;). > > If you really need them on vt1 will need to modify anaconda, but IMHO > that would make it harder to debug and mantain in the future. > Well, the user wouldn't see anything if it isn't on vt1... I can try to show some stuff on vt3, with the 'emulated anaconda screens' but that is really just for debug, I couldn't go with that for a production env... Luis > Regards > Pablo > > El jue, 07-08-2008 a las 11:39 +0200, Pablo Iranzo G?mez escribi?: >> No Luis, in your %pre script you can make a switch to vt3 -> chvt 3 >> do whatever echo and reads you want, and return back to anaconda with >> chvt 1 >> >> >> Regards >> Pablo >> >> >> El jue, 07-08-2008 a las 10:37 +0100, Luis Correia escribi?: >> > Hi Joel, >> > >> > On Thu, Aug 7, 2008 at 10:23, Joel Andres Granados wrote: >> > > Luis Correia wrote: >> > >> >> > >> Hi, >> > >> >> > >> I'm about to do some funky %pre stuff in python, to decide whetner or >> > >> not to install the machine and need some user interaction. >> > >> We're only doing it in text mode, no need for fancy graphics. >> > >> >> > >> How do I from the %pre script write stuff to the screen, get input, >> > >> maybe a button or two ? >> > >> >> > >> I've rummaged through the anaconda code, but it all seems a bit strange to >> > >> me. >> > >> >> > >> Thanks for all the help you could provide, >> > >> Luis Correia >> > >> >> > >> _______________________________________________ >> > >> Kickstart-list mailing list >> > >> Kickstart-list at redhat.com >> > >> https://www.redhat.com/mailman/listinfo/kickstart-list >> > > >> > > Took a quick look at the code and it seems that it executes whatever is in >> > > pre with a redirect to tty3, including the input. This suggests to me that >> > > you can basically do a bash read (or however you want to prompt the user) >> > > and it will prompt you on the vt3. It wont work if you are installing >> > > through serial console though. Didn't actually test this. >> > > >> > > Hope it helps. >> > >> > So, that means that, in order to interact with the user in the main >> > tty, I must modify Anaconda, right? >> > If yes, then I could probably use the 'updates' method... >> > >> > > -- >> > > Joel Andres Granados >> > > Red Hat / Brno, Czech Republic >> > >> > Luis Correia >> > From Pablo.Iranzo at redhat.com Thu Aug 7 09:52:05 2008 From: Pablo.Iranzo at redhat.com (Pablo Iranzo =?ISO-8859-1?Q?G=F3mez?=) Date: Thu, 07 Aug 2008 11:52:05 +0200 Subject: anaconda screens In-Reply-To: References: <489ABF1F.5030504@redhat.com> <1218101974.23784.109.camel@iranzo.users.redhat.com> <1218102066.23784.111.camel@iranzo.users.redhat.com> Message-ID: <1218102725.23784.113.camel@iranzo.users.redhat.com> Luis If you use "chvt 3", the vt shown on your display will be vt3. ?Why the user will not be seeing it? Regards Pablo El jue, 07-08-2008 a las 10:46 +0100, Luis Correia escribi?: > Hi! > > On Thu, Aug 7, 2008 at 10:41, Pablo Iranzo G?mez > wrote: > > Of course, on vt3 you can 'simulate' anaconda screens ;). > > > > If you really need them on vt1 will need to modify anaconda, but IMHO > > that would make it harder to debug and mantain in the future. > > > > Well, the user wouldn't see anything if it isn't on vt1... > > I can try to show some stuff on vt3, with the 'emulated anaconda > screens' but that is really just for debug, I couldn't go with that > for a production env... > > Luis > > > > Regards > > Pablo > > > > El jue, 07-08-2008 a las 11:39 +0200, Pablo Iranzo G?mez escribi?: > >> No Luis, in your %pre script you can make a switch to vt3 -> chvt 3 > >> do whatever echo and reads you want, and return back to anaconda with > >> chvt 1 > >> > >> > >> Regards > >> Pablo > >> > >> > >> El jue, 07-08-2008 a las 10:37 +0100, Luis Correia escribi?: > >> > Hi Joel, > >> > > >> > On Thu, Aug 7, 2008 at 10:23, Joel Andres Granados wrote: > >> > > Luis Correia wrote: > >> > >> > >> > >> Hi, > >> > >> > >> > >> I'm about to do some funky %pre stuff in python, to decide whetner or > >> > >> not to install the machine and need some user interaction. > >> > >> We're only doing it in text mode, no need for fancy graphics. > >> > >> > >> > >> How do I from the %pre script write stuff to the screen, get input, > >> > >> maybe a button or two ? > >> > >> > >> > >> I've rummaged through the anaconda code, but it all seems a bit strange to > >> > >> me. > >> > >> > >> > >> Thanks for all the help you could provide, > >> > >> Luis Correia > >> > >> > >> > >> _______________________________________________ > >> > >> Kickstart-list mailing list > >> > >> Kickstart-list at redhat.com > >> > >> https://www.redhat.com/mailman/listinfo/kickstart-list > >> > > > >> > > Took a quick look at the code and it seems that it executes whatever is in > >> > > pre with a redirect to tty3, including the input. This suggests to me that > >> > > you can basically do a bash read (or however you want to prompt the user) > >> > > and it will prompt you on the vt3. It wont work if you are installing > >> > > through serial console though. Didn't actually test this. > >> > > > >> > > Hope it helps. > >> > > >> > So, that means that, in order to interact with the user in the main > >> > tty, I must modify Anaconda, right? > >> > If yes, then I could probably use the 'updates' method... > >> > > >> > > -- > >> > > Joel Andres Granados > >> > > Red Hat / Brno, Czech Republic > >> > > >> > Luis Correia > >> > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Pablo Iranzo G?mez (Pablo.Iranzo at redhat.com) RHCE/RHCSP/RHCSS Global Profesional Services Consultant Spain Phone: +34 645 01 01 49 (CET/CEST) GnuPG KeyID: 0xFAD3CF0D -- Inscrita en el Reg. Mercantil de Madrid ? C.I.F. B-82 65 79 41 Directores: Michael Cunningham, Charlie Peters y David Owens Direcci?n Registrada: Red Hat S.L., C/ Velazquez 63, Madrid 28001, Espa?a Direcci?n contacto: C/Jose Bardasano Baos, 9, Edif. Gorbea 3, Planta 3?D, 28016 Madrid, Spain -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Esta parte del mensaje est? firmada digitalmente URL: From luis.f.correia at gmail.com Thu Aug 7 09:55:40 2008 From: luis.f.correia at gmail.com (Luis Correia) Date: Thu, 7 Aug 2008 10:55:40 +0100 Subject: anaconda screens In-Reply-To: <1218102725.23784.113.camel@iranzo.users.redhat.com> References: <489ABF1F.5030504@redhat.com> <1218101974.23784.109.camel@iranzo.users.redhat.com> <1218102066.23784.111.camel@iranzo.users.redhat.com> <1218102725.23784.113.camel@iranzo.users.redhat.com> Message-ID: Hi Pablo, On Thu, Aug 7, 2008 at 10:52, Pablo Iranzo G?mez wrote: > Luis > If you use "chvt 3", the vt shown on your display will be vt3. > > ?Why the user will not be seeing it? Because I don't get stuff right the first time i read it .. Let me run some tests and I'll get back to you if need be. Gracias! Luis > > Regards > Pablo > > El jue, 07-08-2008 a las 10:46 +0100, Luis Correia escribi?: >> Hi! >> >> On Thu, Aug 7, 2008 at 10:41, Pablo Iranzo G?mez >> wrote: >> > Of course, on vt3 you can 'simulate' anaconda screens ;). >> > >> > If you really need them on vt1 will need to modify anaconda, but IMHO >> > that would make it harder to debug and mantain in the future. >> > >> >> Well, the user wouldn't see anything if it isn't on vt1... >> >> I can try to show some stuff on vt3, with the 'emulated anaconda >> screens' but that is really just for debug, I couldn't go with that >> for a production env... >> >> Luis >> >> >> > Regards >> > Pablo >> > >> > El jue, 07-08-2008 a las 11:39 +0200, Pablo Iranzo G?mez escribi?: >> >> No Luis, in your %pre script you can make a switch to vt3 -> chvt 3 >> >> do whatever echo and reads you want, and return back to anaconda with >> >> chvt 1 >> >> >> >> >> >> Regards >> >> Pablo >> >> >> >> >> >> El jue, 07-08-2008 a las 10:37 +0100, Luis Correia escribi?: >> >> > Hi Joel, >> >> > >> >> > On Thu, Aug 7, 2008 at 10:23, Joel Andres Granados wrote: >> >> > > Luis Correia wrote: >> >> > >> >> >> > >> Hi, >> >> > >> >> >> > >> I'm about to do some funky %pre stuff in python, to decide whetner or >> >> > >> not to install the machine and need some user interaction. >> >> > >> We're only doing it in text mode, no need for fancy graphics. >> >> > >> >> >> > >> How do I from the %pre script write stuff to the screen, get input, >> >> > >> maybe a button or two ? >> >> > >> >> >> > >> I've rummaged through the anaconda code, but it all seems a bit strange to >> >> > >> me. >> >> > >> >> >> > >> Thanks for all the help you could provide, >> >> > >> Luis Correia >> >> > >> >> >> > >> _______________________________________________ >> >> > >> Kickstart-list mailing list >> >> > >> Kickstart-list at redhat.com >> >> > >> https://www.redhat.com/mailman/listinfo/kickstart-list >> >> > > >> >> > > Took a quick look at the code and it seems that it executes whatever is in >> >> > > pre with a redirect to tty3, including the input. This suggests to me that >> >> > > you can basically do a bash read (or however you want to prompt the user) >> >> > > and it will prompt you on the vt3. It wont work if you are installing >> >> > > through serial console though. Didn't actually test this. >> >> > > >> >> > > Hope it helps. >> >> > >> >> > So, that means that, in order to interact with the user in the main >> >> > tty, I must modify Anaconda, right? >> >> > If yes, then I could probably use the 'updates' method... >> >> > >> >> > > -- >> >> > > Joel Andres Granados >> >> > > Red Hat / Brno, Czech Republic >> >> > >> >> > Luis Correia >> >> > >> >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list > -- > > Pablo Iranzo G?mez (Pablo.Iranzo at redhat.com) > RHCE/RHCSP/RHCSS Global Profesional Services Consultant Spain > Phone: +34 645 01 01 49 (CET/CEST) > GnuPG KeyID: 0xFAD3CF0D > > -- > Inscrita en el Reg. Mercantil de Madrid ? C.I.F. B-82 65 79 41 > Directores: Michael Cunningham, Charlie Peters y David Owens > Direcci?n Registrada: Red Hat S.L., C/ Velazquez 63, Madrid 28001, Espa?a > Direcci?n contacto: C/Jose Bardasano Baos, 9, Edif. Gorbea 3, Planta 3?D, 28016 Madrid, Spain > > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From debian at herakles.homelinux.org Thu Aug 7 13:38:42 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Thu, 07 Aug 2008 21:38:42 +0800 Subject: anaconda screens In-Reply-To: References: Message-ID: <489AFAE2.201@herakles.homelinux.org> Luis Correia wrote: > Hi, > > I'm about to do some funky %pre stuff in python, to decide whetner or > not to install the machine and need some user interaction. > We're only doing it in text mode, no need for fancy graphics. > > How do I from the %pre script write stuff to the screen, get input, > maybe a button or two ? > > I've rummaged through the anaconda code, but it all seems a bit strange to me. > > Thanks for all the help you could provide, > Luis Correia There's another, better list, for those who would hack on Anaconda. I'm sure you will get better advice there. -- 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 jwildebo at redhat.com Fri Aug 8 07:35:53 2008 From: jwildebo at redhat.com (Jan Wildeboer) Date: Fri, 08 Aug 2008 09:35:53 +0200 Subject: anaconda screens In-Reply-To: <489AFAE2.201@herakles.homelinux.org> References: <489AFAE2.201@herakles.homelinux.org> Message-ID: <489BF759.6090403@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 | Luis Correia wrote: |> Hi, |> |> I'm about to do some funky %pre stuff in python, to decide whetner or |> not to install the machine and need some user interaction. |> We're only doing it in text mode, no need for fancy graphics. |> |> How do I from the %pre script write stuff to the screen, get input, |> maybe a button or two ? |> |> I've rummaged through the anaconda code, but it all seems a bit |> strange to me. |> |> Thanks for all the help you could provide, |> Luis Correia For starters, use the internet search engine of your choice with: ~ chvt anaconda ;-) Jan -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Red Hat - http://enigmail.mozdev.org iD8DBQFIm/dZb2GgeF3rr7ARAv/RAJ9IR7M2xFD9/9nptcRLBpMakIRa+wCeKHHl No3PMQglxnrRRUOTHpaYYdE= =nsr3 -----END PGP SIGNATURE----- From luis.f.correia at gmail.com Fri Aug 8 07:56:40 2008 From: luis.f.correia at gmail.com (Luis Correia) Date: Fri, 8 Aug 2008 08:56:40 +0100 Subject: anaconda screens In-Reply-To: <489BF759.6090403@redhat.com> References: <489AFAE2.201@herakles.homelinux.org> <489BF759.6090403@redhat.com> Message-ID: Hi Jan, On Fri, Aug 8, 2008 at 08:35, Jan Wildeboer wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > | Luis Correia wrote: > |> Hi, > |> > |> I'm about to do some funky %pre stuff in python, to decide whetner or > |> not to install the machine and need some user interaction. > |> We're only doing it in text mode, no need for fancy graphics. > |> > |> How do I from the %pre script write stuff to the screen, get input, > |> maybe a button or two ? > |> > |> I've rummaged through the anaconda code, but it all seems a bit > |> strange to me. > |> > |> Thanks for all the help you could provide, > |> Luis Correia > > For starters, use the internet search engine of your choice with: > > ~ chvt anaconda > > ;-) > > Jan Meanwhile I've managed to get exactly what I wanted. The hint from Pablo was invaluable, as I've now done a much more simplified mock up of the Anaconda screens and I'm currently working on the interface to the user. It is actually very simple and easy to do, once you know what you want. Thank you all for the help. p.s. Jan, you might not remember me, but we spoke for a while in the Lisbon Open source conference, i'm the one working with CentOS :). Luis Correia From asmith11 at cox.net Wed Aug 13 15:58:34 2008 From: asmith11 at cox.net (asmith11 at cox.net) Date: Wed, 13 Aug 2008 11:58:34 -0400 Subject: Disable LVM via Kickstart File Message-ID: <20080813115834.ADMVB.97754.imail@eastrmwml10> Folks, Hope this isn't a repeat, couldn't get a clear answer via Google. Question: is there a way to instruct Anaconda not to use LVM for the system installation in a kickstart file? We would prefer that our system disks not be wrapped with an LVM layer. Thanks! -Andy Smith asmith11 at cox.net From dan at half-asleep.com Wed Aug 13 17:06:33 2008 From: dan at half-asleep.com (Daniel Segall) Date: Wed, 13 Aug 2008 13:06:33 -0400 Subject: Disable LVM via Kickstart File In-Reply-To: <20080813115834.ADMVB.97754.imail@eastrmwml10> References: <20080813115834.ADMVB.97754.imail@eastrmwml10> Message-ID: <48A31499.2060304@half-asleep.com> If you set your partitioning scheme in the kickstart without lvm, then no lvm will be used. Example: clearpart --all --initlabel part /boot --size 100 --fstype ext3 part / --size 5120 --fstype ext3 part /home --size 5120 --fstype ext3 part swap --size 2048 --fstype swap -Dan asmith11 at cox.net wrote: > Folks, > > Hope this isn't a repeat, couldn't get a clear answer via Google. > > Question: is there a way to instruct Anaconda not to use LVM for the system installation in a kickstart file? We would prefer that our system disks not be wrapped with an LVM layer. > > Thanks! > > -Andy Smith > asmith11 at cox.net > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From patchu1 at yahoo.com Wed Aug 13 21:40:44 2008 From: patchu1 at yahoo.com (Pat) Date: Wed, 13 Aug 2008 14:40:44 -0700 (PDT) Subject: Possible to log %post when using chroot? Message-ID: <384989.40676.qm@web32506.mail.mud.yahoo.com> Quote from the site http://kbase.redhat.com/faq/FAQ_42_10490.shtm: ============ %post nochroot --log=/tmp/post.log //post scripts and commands go here The logfiles will be saved in the temporary installation environment. In order to make them available when the installation is complete, they need to be copied to the installation environment in the %post section as show below. Note that this can not be done in a chrooted environment; the "nochroot" option must be used. ============ so when executing a chrooted %post script, it is not possible to log the activity? Is that right? For example, I have a chrooted post section like this: %post --log=/root/post.log //do stuff I tried that but the commands and resulting error messages (if any) were not logged with one exception. Within that post section, I did create a run once script which runs after the first reboot. The output from that script when it ran after the first reboot was recorded to the post.log which I don't quite understand. I'm glad I can see the output however I would like to see the results of the other post commands as well which were done after the OS install and before the first reboot. BTW I'm using CentOS 5.0 and Redhat 5.2. Thanks for any help! From Pablo.Iranzo at redhat.com Wed Aug 13 21:50:27 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Wed, 13 Aug 2008 23:50:27 +0200 (CEST) Subject: Possible to log %post when using chroot? In-Reply-To: <384989.40676.qm@web32506.mail.mud.yahoo.com> References: <384989.40676.qm@web32506.mail.mud.yahoo.com> Message-ID: Today there was a thread on cobbler-list: https://fedorahosted.org/pipermail/cobbler/2008-August/000492.html Not using --log, but in a 'functional' way. Regards Pablo On Wed, 13 Aug 2008, Pat wrote: > Quote from the site http://kbase.redhat.com/faq/FAQ_42_10490.shtm: > ============ > %post nochroot --log=/tmp/post.log > //post scripts and commands go here > > The logfiles will be saved in the temporary installation environment. In order to make them available when the installation is complete, they need to be copied to the installation environment in the %post section as show below. Note that this can not be done in a chrooted environment; the "nochroot" option must be used. > ============ > > so when executing a chrooted %post script, it is not possible to log the activity? Is that right? > > For example, I have a chrooted post section like this: > > %post --log=/root/post.log > //do stuff > > I tried that but the commands and resulting error messages (if any) were not logged with one exception. Within that post section, I did create a run once script which runs after the first reboot. The output from that script when it ran after the first reboot was recorded to the post.log which I don't quite understand. I'm glad I can see the output however I would like to see the results of the other post commands as well which were done after the OS install and before the first reboot. > > BTW I'm using CentOS 5.0 and Redhat 5.2. Thanks for any help! > > > > > > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -- Pablo Iranzo G?mez (http://Alufis35.uv.es/~iranzo/) (PGPKey Available on http://www.uv.es/~iranzop/PGPKey.pgp) -- Postulado de Boling sobre la Ley de Murphy: Si se encuentra bien, no se preocupe. Se le pasar? From ady.wicaksono at gmail.com Thu Aug 14 10:00:40 2008 From: ady.wicaksono at gmail.com (Ady Wicaksono) Date: Thu, 14 Aug 2008 18:00:40 +0800 Subject: change newt text on installer Message-ID: <2743ae110808140300x2ffd69f6p7e81c5520342445b@mail.gmail.com> If we use text installer for installing RHEL or CENTOS, we can see in the top left window text "Welcome to CentOS ... blablabla" How can we change this text? Thanks -- Regards, Ady Wicaksono Email: ady.wicaksono at gmail.com http://adywicaksono.wordpress.com/ From atodorov at redhat.com Thu Aug 14 10:17:05 2008 From: atodorov at redhat.com (Alexander Todorov) Date: Thu, 14 Aug 2008 13:17:05 +0300 Subject: change newt text on installer In-Reply-To: <2743ae110808140300x2ffd69f6p7e81c5520342445b@mail.gmail.com> References: <2743ae110808140300x2ffd69f6p7e81c5520342445b@mail.gmail.com> Message-ID: <48A40621.8090308@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Ady Wicaksono wrote: | If we use text installer for installing RHEL or CENTOS, we can see in | the top left window text "Welcome to CentOS ... blablabla" | | How can we change this text? | IIRC this comes from anaconda and is not related to kickstart. You'll have to rebuild your installation image with the text you want (this is the product name - - RHEL, Fedora, etc.). Check pungi or revisor projects for tools and instructions how to build a custom install image. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Red Hat - http://enigmail.mozdev.org iD8DBQFIpAYghmd3WOiFct4RCpqZAJ9ljzK+Sradj6GyFdcgL4dabTqF1QCgtx48 9po/unqhsblKq9RQoaJ0iRg= =Jw1Z -----END PGP SIGNATURE----- From stan.hearn at nscorp.com Thu Aug 14 12:51:00 2008 From: stan.hearn at nscorp.com (Hearn, Stan J.) Date: Thu, 14 Aug 2008 08:51:00 -0400 Subject: Possible to log %post when using chroot? In-Reply-To: <384989.40676.qm@web32506.mail.mud.yahoo.com> Message-ID: -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Pat For example, I have a chrooted post section like this: %post --log=/root/post.log //do stuff I tried that but the commands and resulting error messages (if any) were not logged with one exception. Within that post section, I did create a run once script which runs after the first reboot. The output from that script when it ran after the first reboot was recorded to the post.log which I don't quite understand. I'm glad I can see the output however I would like to see the results of the other post commands as well which were done after the OS install and before the first reboot. --------------------------- Remember that /root/post.log must exist before the chroot is created. And the command is relative to the non-chrooted file structure. The chroot will be /tmp/sysimage (at least for RHEL). Most people I find put the log in /tmp %post --log=/tmp/post.log ...commands here.... %post nochroot mv /tmp/post.log /mnt/sysimage/var/log/post.log Does that make sense? If you want to put it in root's home, you should use the file name of /tmp/sysimage/root/post.log. You wouldn't need to move it afterward. Regards, Stan Hearn From pgroven at 2wire.com Thu Aug 14 20:27:44 2008 From: pgroven at 2wire.com (Phillip Groven) Date: Thu, 14 Aug 2008 13:27:44 -0700 Subject: dhcpd.conf for kickstart Message-ID: <299E8BB0D5475E4FABB162E7DCA1BBE902E2873D@phxexch01.corp.2wire.com> When I use the first dhcpd.conf my kickstart works fine When I use the 2nd one I get PXE-E53: No boot filename received I am trying to get the dhcp server to assign a specific IP address based on the mac address and download the exact ks.cfg for that server. What am I doing wrong? WORKING DHCPD.CONF ddns-update-style int erim; ignore client-updates; subnet 10.28.200.0 netmask 255.255.255.0 { allow booting; allow bootp; option routers 10.28.200.228; option subnet-mask 255.255.255.0; option domain-name-servers 10.28.200.250; range dynamic-bootp 10.28.200.10 10.28.200.20; default-lease-time 21600; max-lease-time 43200; # bootp server and filename next-server 10.28.200.250; filename "pxelinux.0"; } NONWORKING DHDPD.CONF ddns-update-style interim; ignore client-updates; subnet 10.28.200.0 netmask 255.255.255.0 { allow booting; allow bootp; option routers 10.28.200.228; option subnet-mask 255.255.255.0; option domain-name-servers 10.28.200.250; range dynamic-bootp 10.28.200.1 10.28.200.252; default-lease-time 21600; max-lease-time 43200; group { option vendor-encapsulated-options ff; next-server 10.28.200.250; filename "pxelinux.0"; host prx53-1 { hardware ethernet 00:17:A4:BB:BF:7C; fixed-address 10.28.200.233; option host-name "prx53-1.domain.com"; } host w53-4 { hardware ethernet 00:E0:CC:21:C1:22; fixed-address 10.28.200.8; option host-name "w53-4.domain.com"; } host mon28-1 { hardware ethernet 00:14:DD:EF:C0:7A; fixed-address 10.28.200.245; option host-name "mon28-1.c28.domain.com"; filename "pxelinux.0"; next-server 10.28.200.250; } host w28-1 { hardware ethernet 00:14:FF:3D:3B:3A; fixed-address 10.28.200.1; option host-name "w28-1.domain.com"; } host w28-5 { hardware ethernet 00:14:DD:4b:3b:b2; fixed-address 10.28.200.5; filename "pxelinux.0"; option host-name "w28-5.domain.com"; next-server 10.28.200.250; } host w28-6 { hardware ethernet 00:11:AB:C3:4F:93; fixed-address 10.28.200.6; filename "pxelinux.0"; option host-name "w28-6.domain.com"; next-server 10.28.200.250; } -------------- next part -------------- An HTML attachment was scrubbed... URL: From Moray.Henderson at ict.om.org Fri Aug 15 09:23:20 2008 From: Moray.Henderson at ict.om.org (Moray Henderson (ICT)) Date: Fri, 15 Aug 2008 09:23:20 -0000 Subject: change newt text on installer In-Reply-To: <2743ae110808140300x2ffd69f6p7e81c5520342445b@mail.gmail.com> References: <2743ae110808140300x2ffd69f6p7e81c5520342445b@mail.gmail.com> Message-ID: >From: Ady Wicaksono [mailto:ady.wicaksono at gmail.com] > >If we use text installer for installing RHEL or CENTOS, we can see in >the top left window text "Welcome to CentOS ... blablabla" > >How can we change this text? Yes, this is Anaconda. I tried to get pungi to build CentOS 5.2, but it required versions of other packages that only appear in the later Fedoras, so I went back to the old-fashioned way of doing it. The command you need is buildinstall, from the anaconda-runtime package, which rebuilds the boot images for the CD. It's... not well documented. Note that it overwrites some of the files in the isolinux directory, so if you have customised those make sure you have backups. Come to think of it, make sure you have backups of your build root anyway. The commands you need will be something like PATH=/usr/lib/anaconda-runtime:$PATH buildinstall --comps \ --version \ --product \ --release \ --prodpath \ It's either the product or release name that gets displayed by the text mode installs. Moray. "To err is human. To purr, feline" http://members.aol.com/edgwddirk From kanarip at kanarip.com Fri Aug 15 09:25:51 2008 From: kanarip at kanarip.com (Jeroen van Meeuwen) Date: Fri, 15 Aug 2008 11:25:51 +0200 Subject: change newt text on installer In-Reply-To: References: <2743ae110808140300x2ffd69f6p7e81c5520342445b@mail.gmail.com> Message-ID: <48A54B9F.4020705@kanarip.com> Moray Henderson (ICT) wrote: >> From: Ady Wicaksono [mailto:ady.wicaksono at gmail.com] >> >> If we use text installer for installing RHEL or CENTOS, we can see in >> the top left window text "Welcome to CentOS ... blablabla" >> >> How can we change this text? > > Yes, this is Anaconda. I tried to get pungi to build CentOS 5.2, but it required versions of other packages that only appear in the later Fedoras, so I went back to the old-fashioned way of doing it. > May I recommend you try Revisor? > The command you need is buildinstall, from the anaconda-runtime package, which rebuilds the boot images for the CD. It's... not well documented. Note that it overwrites some of the files in the isolinux directory, so if you have customised those make sure you have backups. Come to think of it, make sure you have backups of your build root anyway. The commands you need will be something like > > PATH=/usr/lib/anaconda-runtime:$PATH > buildinstall --comps \ > --version \ > --product \ > --release \ > --prodpath \ > > > It's either the product or release name that gets displayed by the text mode installs. > IIRC, it's the Product name that is used for the text displayed described in this thread. Kind regards, Jeroen van Meeuwen -kanarip From Joe_Wulf at yahoo.com Fri Aug 15 13:30:37 2008 From: Joe_Wulf at yahoo.com (Joe_Wulf) Date: Fri, 15 Aug 2008 09:30:37 -0400 Subject: dhcpd.conf for kickstart In-Reply-To: <299E8BB0D5475E4FABB162E7DCA1BBE902E2873D@phxexch01.corp.2wire.com> References: <299E8BB0D5475E4FABB162E7DCA1BBE902E2873D@phxexch01.corp.2wire.com> Message-ID: Phillip, {resent--previous size was too big, so I chopped all but my reply} Here is what I have... which works for me, to do things the way you are asking for... I've tailored this to match your addresses, though. So this should work directly for you. Now, I've not 'tested' your config, simply adapted your addresses to mine. There might be better/easier ways to do this, too.... I'd welcome feedback. I've looked for a good resource or three that would serve as good documentation for DHCP. Hope this helps you. --------------------------------------------------------------------------------- ------------------------- ignore unknown-clients; authoritative; ddns-updates off; ddns-update-style interim; ignore client-updates; allow bootp; allow booting; allow duplicates; subnet 10.28.200.0 netmask 255.255.255.0 { # --- default gateway option routers 10.28.200.228; option subnet-mask 255.255.255.0; option broadcast-address 10.28.200.255; option domain-name-servers 10.28.200.250; option time-offset -18000; # Eastern Standard Time default-lease-time 21600; max-lease-time 43200; host prx53-1 { hardware ethernet 00:17:A4:BB:BF:7C; fixed-address 10.28.200.233; option host-name "prx53-1.domain.com"; next-server 10.28.200.250; filename "pxelinux.0"; } host w53-4 { hardware ethernet 00:E0:CC:21:C1:22; fixed-address 10.28.200.8; option host-name "w53-4.domain.com"; next-server 10.28.200.250; filename "pxelinux.0"; } host mon28-1 { hardware ethernet 00:14:DD:EF:C0:7A; fixed-address 10.28.200.245; option host-name "mon28-1.c28.domain.com"; next-server 10.28.200.250; filename "pxelinux.0"; } host w28-1 { hardware ethernet 00:14:FF:3D:3B:3A; fixed-address 10.28.200.1; option host-name "w28-1.domain.com"; next-server 10.28.200.250; filename "pxelinux.0"; } host w28-5 { hardware ethernet 00:14:DD:4b:3b:b2; fixed-address 10.28.200.5; option host-name "w28-5.domain.com"; next-server 10.28.200.250; filename "pxelinux.0"; } host w28-6 { hardware ethernet 00:11:AB:C3:4F:93; fixed-address 10.28.200.6; option host-name "w28-6.domain.com"; next-server 10.28.200.250; filename "pxelinux.0"; } } --------------------------------------------------------------------------------- ------------------------- R, -Joe Wulf, CISSP, USN(RET) Senior IA Engineer ProSync Technology Group, LLC www.prosync.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From patchu1 at yahoo.com Fri Aug 15 19:32:14 2008 From: patchu1 at yahoo.com (Pat) Date: Fri, 15 Aug 2008 12:32:14 -0700 (PDT) Subject: Possible to log %post when using chroot? In-Reply-To: Message-ID: <345923.69800.qm@web32505.mail.mud.yahoo.com> --- On Thu, 8/14/08, Hearn, Stan J. wrote: > > Remember that /root/post.log must exist before the chroot > is created. > And the command is relative to the non-chrooted file > structure. > > The chroot will be /tmp/sysimage (at least for RHEL). > > Most people I find put the log in /tmp > > %post --log=/tmp/post.log > ...commands here.... > > %post nochroot > mv /tmp/post.log /mnt/sysimage/var/log/post.log > > Does that make sense? If you want to put it in root's > home, you should > use the file name of /tmp/sysimage/root/post.log. You > wouldn't need to > move it afterward. > > Regards, > Stan Hearn Thanks for the explanation Stan, I understand it now. BTW, for RHEL 5.2, the chroot is /mnt/sysimage as /tmp/sysimage does not exist. Pat From chwilk at rice.edu Sun Aug 17 13:02:51 2008 From: chwilk at rice.edu (Chandler Wilkerson) Date: Sun, 17 Aug 2008 08:02:51 -0500 Subject: dhcpd.conf for kickstart In-Reply-To: <299E8BB0D5475E4FABB162E7DCA1BBE902E2873D@phxexch01.corp.2wire.com> References: <299E8BB0D5475E4FABB162E7DCA1BBE902E2873D@phxexch01.corp.2wire.com> Message-ID: <4922F538-ED06-49C4-9884-71E250271551@rice.edu> The problem you're running into with the second file is that your dynamic range overlaps your static host entries. Dynamic is used as a fallback when your host doesn't match any of the static rules. If you shrink the dynamic range, your statically defined hosts should start getting static IPs and PXE booting. -- Chandler Wilkerson Research Computing Rice University On Aug 14, 2008, at 3:27 PM, Phillip Groven wrote: > When I use the first dhcpd.conf my kickstart works fine > When I use the 2nd one I get > > PXE-E53: No boot filename received > I am trying to get the dhcp server to assign a specific IP address > based on the mac address and download the exact ks.cfg for that > server. > What am I doing wrong? > > WORKING DHCPD.CONF > ddns-update-style int > erim; > ignore client-updates; > > subnet 10.28.200.0 netmask 255.255.255.0 { > > allow booting; > allow bootp; > option routers 10.28.200.228; > option subnet-mask 255.255.255.0; > option domain-name-servers 10.28.200.250; > > range dynamic-bootp 10.28.200.10 10.28.200.20; > default-lease-time 21600; > max-lease-time 43200; > > # bootp server and filename > next-server 10.28.200.250; > filename "pxelinux.0"; > } > > > NONWORKING DHDPD.CONF > > ddns-update-style interim; > ignore client-updates; > > subnet 10.28.200.0 netmask 255.255.255.0 { > > allow booting; > allow bootp; > option routers 10.28.200.228; > option subnet-mask 255.255.255.0; > option domain-name-servers 10.28.200.250; > > range dynamic-bootp 10.28.200.1 10.28.200.252; > default-lease-time 21600; > max-lease-time 43200; > group { > option vendor-encapsulated-options ff; > next-server 10.28.200.250; > filename "pxelinux.0"; > host prx53-1 { > hardware ethernet 00:17:A4:BB:BF:7C; > fixed-address 10.28.200.233; > option host-name "prx53-1.domain.com"; > } > host w53-4 { > hardware ethernet 00:E0:CC:21:C1:22; > fixed-address 10.28.200.8; > option host-name "w53-4.domain.com"; > } > host mon28-1 { > hardware ethernet 00:14:DD:EF:C0:7A; > fixed-address 10.28.200.245; > option host-name "mon28-1.c28.domain.com"; > filename "pxelinux.0"; > next-server 10.28.200.250; > } > host w28-1 { > hardware ethernet 00:14:FF:3D:3B:3A; > fixed-address 10.28.200.1; > option host-name "w28-1.domain.com"; > } > host w28-5 { > hardware ethernet 00:14:DD:4b:3b:b2; > fixed-address 10.28.200.5; > filename "pxelinux.0"; > option host-name "w28-5.domain.com"; > next-server 10.28.200.250; > } > host w28-6 { > hardware ethernet 00:11:AB:C3:4F:93; > fixed-address 10.28.200.6; > filename "pxelinux.0"; > option host-name "w28-6.domain.com"; > next-server 10.28.200.250; > > } > !DSPAM:3857,48a49561325032724218994! > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > !DSPAM:3857,48a49561325032724218994! -------------- next part -------------- An HTML attachment was scrubbed... URL: From eric at trueblade.com Mon Aug 18 07:19:12 2008 From: eric at trueblade.com (Eric Smith) Date: Mon, 18 Aug 2008 03:19:12 -0400 Subject: anaconda screens In-Reply-To: References: Message-ID: <48A92270.3080503@trueblade.com> Luis Correia wrote: > Hi, > > I'm about to do some funky %pre stuff in python, to decide whetner or > not to install the machine and need some user interaction. > We're only doing it in text mode, no need for fancy graphics. > > How do I from the %pre script write stuff to the screen, get input, > maybe a button or two ? > > I've rummaged through the anaconda code, but it all seems a bit strange to me. Sorry for the late response, I've been on vacation. I know your question has been largely answered, but I thought I'd copy my response to a similar question from earlier this year. The answers I've seen to you question involve switching tty's, but there's also the issue of how to do the UI. The discussion was about using snack, which is (or was, at least; I'm still on FC2) the text based UI manager used by anaconda: ------------------------------- I use code like this: set_tty(1) # change to tty1, we're called by # kickstart with stdout as tty3 user_args = get_user_input() set_tty(3) # restore Where: def set_tty(n): f = open('/dev/tty%d' % n, 'a') os.dup2(f.fileno(), sys.stdin.fileno()) os.dup2(f.fileno(), sys.stdout.fileno()) os.dup2(f.fileno(), sys.stderr.fileno()) and get_user_input() uses snack. ------------------------------- You might want to check out snack (aka newt); there are a few examples at http://www.wanware.com/tsgdocs/snack.html. Eric. From mdehaan at redhat.com Mon Aug 18 17:15:49 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 18 Aug 2008 13:15:49 -0400 Subject: dhcpd.conf for kickstart In-Reply-To: <299E8BB0D5475E4FABB162E7DCA1BBE902E2873D@phxexch01.corp.2wire.com> References: <299E8BB0D5475E4FABB162E7DCA1BBE902E2873D@phxexch01.corp.2wire.com> Message-ID: <48A9AE45.2080709@redhat.com> Phillip Groven wrote: > > *When I use the first dhcpd.conf my kickstart works fine* > > *When I use the 2nd one I get * > > * * > > *PXE-E53: No boot filename received* > > *I am trying to get the dhcp server to assign a specific IP address > based on the mac address and download the exact ks.cfg for that server.* > > *What am I doing wrong?* > > * * > > *WORKING DHCPD.CONF* > > *ddns-update-style int* > > *erim;* > > *ignore client-updates;* > > * * > > *subnet 10.28.200.0 netmask 255.255.255.0 {* > > * * > > * allow booting;* > > * allow bootp;* > > * option routers 10.28.200.228;* > > * option subnet-mask 255.255.255.0;* > > * option domain-name-servers 10.28.200.250;* > > * * > > * range dynamic-bootp 10.28.200.10 10.28.200.20;* > > * default-lease-time 21600;* > > * max-lease-time 43200;* > > * * > > * # bootp server and filename* > > * next-server 10.28.200.250;* > > * filename "pxelinux.0";* > > *}* > > * * > > * * > > *NONWORKING DHDPD.CONF* > > * * > > *ddns-update-style interim;* > > *ignore client-updates;* > > * * > > *subnet 10.28.200.0 netmask 255.255.255.0 {* > > * * > > * allow booting;* > > * allow bootp;* > > * option routers 10.28.200.228;* > > * option subnet-mask 255.255.255.0;* > > * option domain-name-servers 10.28.200.250;* > > * * > > * range dynamic-bootp 10.28.200.1 10.28.200.252;* > > * default-lease-time 21600;* > > * max-lease-time 43200;* > > * group {* > > * option vendor-encapsulated-options ff;* > > * next-server 10.28.200.250;* > > * filename "pxelinux.0";* > > * host prx53-1 {* > > * hardware ethernet 00:17:A4:BB:BF:7C;* > > * fixed-address 10.28.200.233;* > > * option host-name "prx53-1.domain.com";* > > * *} > > * host w53-4 {* > > * hardware ethernet 00:E0:CC:21:C1:22;* > > * *fixed-address 10.28.200.8; > > * option host-name "w53-4.domain.com";* > > * }* > > * host mon28-1 {* > > * hardware ethernet 00:14:DD:EF:C0:7A;* > > * fixed-address 10.28.200.245;* > > * option host-name "mon28-1.c28.domain.com";* > > * filename "pxelinux.0";* > > * next-server 10.28.200.250;* > > * }* > > * host w28-1 {* > > * hardware ethernet 00:14:FF:3D:3B:3A;* > > * fixed-address 10.28.200.1;* > > * option host-name "w28-1.domain.com";* > > * }* > > * host w28-5 {* > > * hardware ethernet 00:14:DD:4b:3b:b2;* > > * fixed-address 10.28.200.5;* > > * filename "pxelinux.0";* > > * option host-name "w28-5.domain.com";* > > * next-server 10.28.200.250;* > > * }* > > * host w28-6 {* > > * hardware ethernet 00:11:AB:C3:4F:93;* > > * fixed-address 10.28.200.6;* > > * filename "pxelinux.0";* > > * option host-name "w28-6.domain.com";* > > * next-server 10.28.200.250;* > > * * > > *}* > > ------------------------------------------------------------------------ > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list Cobbler can take care of generating these things for you to help prevent problems like this... http://fedorahosted.org/cobbler (I believe Fedora Hosted is still down presently ... check back tomorrow if so... and read the "ManageDhcp" article under User Docs) In any case, "Non Working" means many things and it's helpful to define what "non-working" means to you as someone is unlikely to be copying and pasteing your DHCP config in and trying it. --Michael From nicholas.byrne at quadriga.com Fri Aug 22 13:14:57 2008 From: nicholas.byrne at quadriga.com (Nick Byrne) Date: Fri, 22 Aug 2008 14:14:57 +0100 Subject: change newt text on installer In-Reply-To: <48A54B9F.4020705@kanarip.com> References: <2743ae110808140300x2ffd69f6p7e81c5520342445b@mail.gmail.com> <48A54B9F.4020705@kanarip.com> Message-ID: <48AEBBD1.30907@quadriga.com> Change the first word on 2nd line of initrd.img's ".buildstamp" file. Check /usr/lib/anaconda/product.py on an installed system for more details on where it is derived from. Regards Nick Jeroen van Meeuwen wrote: > Moray Henderson (ICT) wrote: >>> From: Ady Wicaksono [mailto:ady.wicaksono at gmail.com] >>> >>> If we use text installer for installing RHEL or CENTOS, we can see in >>> the top left window text "Welcome to CentOS ... blablabla" >>> >>> How can we change this text? >> >> Yes, this is Anaconda. I tried to get pungi to build CentOS 5.2, but >> it required versions of other packages that only appear in the later >> Fedoras, so I went back to the old-fashioned way of doing it. >> > > May I recommend you try Revisor? > >> The command you need is buildinstall, from the anaconda-runtime >> package, which rebuilds the boot images for the CD. It's... not well >> documented. Note that it overwrites some of the files in the >> isolinux directory, so if you have customised those make sure you >> have backups. Come to think of it, make sure you have backups of >> your build root anyway. The commands you need will be something like >> >> PATH=/usr/lib/anaconda-runtime:$PATH >> buildinstall --comps \ >> --version \ >> --product \ >> --release \ >> --prodpath \ >> >> >> It's either the product or release name that gets displayed by the >> text mode installs. >> > > IIRC, it's the Product name that is used for the text displayed > described in this thread. > > Kind regards, > > Jeroen van Meeuwen > -kanarip > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list This e-mail is the property of Quadriga Worldwide Ltd, intended for the addressee only and confidential. Any dissemination, copying or distribution of this message or any attachments is strictly prohibited. If you have received this message in error, please notify us immediately by replying to the message and deleting it from your computer. Messages sent to and from Quadriga may be monitored. Quadriga cannot guarantee any message delivery method is secure or error-free. Information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. We do not accept responsibility for any errors or omissions in this message and/or attachment that arise as a result of transmission. You should carry out your own virus checks before opening any attachment. Any views or opinions presented are solely those of the author and do not necessarily represent those of Quadriga. From Cameron.Kennedy at bovislendlease.com Wed Aug 27 16:33:12 2008 From: Cameron.Kennedy at bovislendlease.com (Kennedy, Cameron) Date: Wed, 27 Aug 2008 12:33:12 -0400 Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot Message-ID: <59B63035E89DEB4C99AB5ECC9C115BB701E1A17F@USATL01ME253.amer.lendlease.com> Hi all, Just starting to play around with RHEL5.2 servers in our environment and am coming up with a strange kickstart issue. I have narrowed it down to the following scenario: Build a HP 385 server, using a RHEL5.2 DVD cut from a downloaded ISO, and the kickstart file located on a USB stick. After the initial build, the server reboots, the DVD and USB stick are manually removed and right where the GRUB screen should be displayed, I get a screen that looks like a kernel panic screen, though starts with the line "Illegal OPCODE" and then a bunch of hex output. If I build the server without the USB stick but using the same kickstart file (copied to the DVD) it works fine. Thinking that the USB stick (/dev/sda) was being incorporated into the build process (LVM), I added the line: "ignoredisk --drives=sda,sdb,sdc,sdd,sde,sdf" to the kickstart script. This did not fix the issue. I then added the line: "part /boot --fstype ext3 --size=200 --ondrive=cciss/c0d0" forcing /boot to be created on the internal raid disk, which also failed the fix the issue. I did watch the screen during the last build process and I know that /boot is created on /dev/cciss/c0d0 (HP raid) and the partition information on the USB stick appears to be unchanged (i.e. not formatted). So I am at a loss as to what else I could check. I do know the exact same kickstart script, server, usb stick combination builds a RHEL4.6 server without any problems. Has anyone else seen this or have any ideas? Thanks. ----------------------------------------- This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Greg.Caetano at hp.com Wed Aug 27 16:38:39 2008 From: Greg.Caetano at hp.com (Caetano, Greg) Date: Wed, 27 Aug 2008 16:38:39 +0000 Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot In-Reply-To: <59B63035E89DEB4C99AB5ECC9C115BB701E1A17F@USATL01ME253.amer.lendlease.com> References: <59B63035E89DEB4C99AB5ECC9C115BB701E1A17F@USATL01ME253.amer.lendlease.com> Message-ID: I would think part of the bootloader is being placed on the USB stick. What do you have for the bootloader statement in kickstart? is there a driveorder parameter on the directive? Greg Caetano Hewlett-Packard Company ESS Software Platform & Business Enablement Solutions Engineering Chicago, IL greg.caetano at hp.com Red Hat Certified Engineer ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Kennedy, Cameron Sent: Wednesday, August 27, 2008 11:33 AM To: Kickstart-list at redhat.com Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot Hi all, Just starting to play around with RHEL5.2 servers in our environment and am coming up with a strange kickstart issue. I have narrowed it down to the following scenario: Build a HP 385 server, using a RHEL5.2 DVD cut from a downloaded ISO, and the kickstart file located on a USB stick. After the initial build, the server reboots, the DVD and USB stick are manually removed and right where the GRUB screen should be displayed, I get a screen that looks like a kernel panic screen, though starts with the line "Illegal OPCODE" and then a bunch of hex output. If I build the server without the USB stick but using the same kickstart file (copied to the DVD) it works fine. Thinking that the USB stick (/dev/sda) was being incorporated into the build process (LVM), I added the line: "ignoredisk --drives=sda,sdb,sdc,sdd,sde,sdf" to the kickstart script. This did not fix the issue. I then added the line: "part /boot --fstype ext3 --size=200 --ondrive=cciss/c0d0" forcing /boot to be created on the internal raid disk, which also failed the fix the issue. I did watch the screen during the last build process and I know that /boot is created on /dev/cciss/c0d0 (HP raid) and the partition information on the USB stick appears to be unchanged (i.e. not formatted). So I am at a loss as to what else I could check. I do know the exact same kickstart script, server, usb stick combination builds a RHEL4.6 server without any problems. Has anyone else seen this or have any ideas? Thanks. ________________________________ "This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country." -------------- next part -------------- An HTML attachment was scrubbed... URL: From Cameron.Kennedy at bovislendlease.com Wed Aug 27 16:51:42 2008 From: Cameron.Kennedy at bovislendlease.com (Kennedy, Cameron) Date: Wed, 27 Aug 2008 12:51:42 -0400 Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot In-Reply-To: References: <59B63035E89DEB4C99AB5ECC9C115BB701E1A17F@USATL01ME253.amer.lendlease.com> Message-ID: <59B63035E89DEB4C99AB5ECC9C115BB701E1A19D@USATL01ME253.amer.lendlease.com> Thanks for the reply. I have this line for the bootloader: "bootloader --location=mbr --ondrive=cciss/c0d0" I added this at the same time as I modified the "part /boot" line shown below. But, this didn't seem to fix the issue. From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Caetano, Greg Sent: Wednesday, August 27, 2008 12:39 PM To: Discussion list about Kickstart Subject: RE: RHEL5.2, kickstart, USB, "OPCODE" error after reboot I would think part of the bootloader is being placed on the USB stick. What do you have for the bootloader statement in kickstart? is there a driveorder parameter on the directive? Greg Caetano Hewlett-Packard Company ESS Software Platform & Business Enablement Solutions Engineering Chicago, IL greg.caetano at hp.com Red Hat Certified Engineer ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Kennedy, Cameron Sent: Wednesday, August 27, 2008 11:33 AM To: Kickstart-list at redhat.com Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot Hi all, Just starting to play around with RHEL5.2 servers in our environment and am coming up with a strange kickstart issue. I have narrowed it down to the following scenario: Build a HP 385 server, using a RHEL5.2 DVD cut from a downloaded ISO, and the kickstart file located on a USB stick. After the initial build, the server reboots, the DVD and USB stick are manually removed and right where the GRUB screen should be displayed, I get a screen that looks like a kernel panic screen, though starts with the line "Illegal OPCODE" and then a bunch of hex output. If I build the server without the USB stick but using the same kickstart file (copied to the DVD) it works fine. Thinking that the USB stick (/dev/sda) was being incorporated into the build process (LVM), I added the line: "ignoredisk --drives=sda,sdb,sdc,sdd,sde,sdf" to the kickstart script. This did not fix the issue. I then added the line: "part /boot --fstype ext3 --size=200 --ondrive=cciss/c0d0" forcing /boot to be created on the internal raid disk, which also failed the fix the issue. I did watch the screen during the last build process and I know that /boot is created on /dev/cciss/c0d0 (HP raid) and the partition information on the USB stick appears to be unchanged (i.e. not formatted). So I am at a loss as to what else I could check. I do know the exact same kickstart script, server, usb stick combination builds a RHEL4.6 server without any problems. Has anyone else seen this or have any ideas? Thanks. ________________________________ "This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country." ----------------------------------------- This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country. -------------- next part -------------- An HTML attachment was scrubbed... URL: From mpotla at cisco.com Wed Aug 27 17:42:57 2008 From: mpotla at cisco.com (Murali Potla (mpotla)) Date: Wed, 27 Aug 2008 10:42:57 -0700 Subject: (no subject) In-Reply-To: <59B63035E89DEB4C99AB5ECC9C115BB701E1A17F@USATL01ME253.amer.lendlease.com> Message-ID: <5547014632ED654F971D7E1E0C2E0C3E065E3639@xmb-sjc-215.amer.cisco.com> Hi all, How can we specify the architecture of a package within kickstart file. I want to install x86 architecture version of libgcc thru kickstart process. How to specify this package in ks file In some cases on 64 bit platforms we need to install 32 bit packages also for backward compatibility for some applications. Thanks, Murali From Cameron.Kennedy at bovislendlease.com Wed Aug 27 17:55:52 2008 From: Cameron.Kennedy at bovislendlease.com (Kennedy, Cameron) Date: Wed, 27 Aug 2008 13:55:52 -0400 Subject: (no subject) In-Reply-To: <5547014632ED654F971D7E1E0C2E0C3E065E3639@xmb-sjc-215.amer.cisco.com> References: <59B63035E89DEB4C99AB5ECC9C115BB701E1A17F@USATL01ME253.amer.lendlease.com> <5547014632ED654F971D7E1E0C2E0C3E065E3639@xmb-sjc-215.amer.cisco.com> Message-ID: <59B63035E89DEB4C99AB5ECC9C115BB701E1A20C@USATL01ME253.amer.lendlease.com> Personally, I use the %post section to install these type of packages (like Oracle 32bit packages). One idea you could try, though I have not tested this, is specifying the package with an .i386 extension in your packages selection list. e.g. libgcc.i386 -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Murali Potla (mpotla) Sent: Wednesday, August 27, 2008 1:43 PM To: Discussion list about Kickstart Subject: (no subject) Hi all, How can we specify the architecture of a package within kickstart file. I want to install x86 architecture version of libgcc thru kickstart process. How to specify this package in ks file In some cases on 64 bit platforms we need to install 32 bit packages also for backward compatibility for some applications. Thanks, Murali _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list ----------------------------------------- This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country. From Tim.Mooney at ndsu.edu Wed Aug 27 18:26:49 2008 From: Tim.Mooney at ndsu.edu (Tim Mooney) Date: Wed, 27 Aug 2008 13:26:49 -0500 (CDT) Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot In-Reply-To: <59B63035E89DEB4C99AB5ECC9C115BB701E1A17F@USATL01ME253.amer.lendlease.com> References: <59B63035E89DEB4C99AB5ECC9C115BB701E1A17F@USATL01ME253.amer.lendlease.com> Message-ID: In regard to: RHEL5.2, kickstart, USB, "OPCODE" error after reboot,...: > Build a HP 385 server, using a RHEL5.2 DVD cut from a downloaded ISO, > and the kickstart file located on a USB stick. After the initial build, > the server reboots, the DVD and USB stick are manually removed and right > where the GRUB screen should be displayed, I get a screen that looks > like a kernel panic screen, though starts with the line "Illegal OPCODE" > and then a bunch of hex output. > > If I build the server without the USB stick but using the same kickstart > file (copied to the DVD) it works fine. Would loading the kickstart file via http work for you? Just put it on some web server and use linux ks=http://your.host/your/url/ks.cfg Tim -- Tim Mooney Tim.Mooney at ndsu.edu Enterprise Computing & Infrastructure 701-231-1076 (Voice) Room 242-J6, IACC Building 701-231-8541 (Fax) North Dakota State University, Fargo, ND 58105-5164 From Cameron.Kennedy at bovislendlease.com Wed Aug 27 18:35:28 2008 From: Cameron.Kennedy at bovislendlease.com (Kennedy, Cameron) Date: Wed, 27 Aug 2008 14:35:28 -0400 Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot In-Reply-To: References: <59B63035E89DEB4C99AB5ECC9C115BB701E1A17F@USATL01ME253.amer.lendlease.com> Message-ID: <59B63035E89DEB4C99AB5ECC9C115BB701E1A290@USATL01ME253.amer.lendlease.com> For the particular build I was try to do, yes I could have used an FTP, NFS, web server - I opted to just redo a DVD with KS file which worked. However, I was hoping to see if anyone else had the problem (and fix :-) ). RHEL5 is new for our environment but as I start to migrate the RHEL4 servers to RHEL5 this will become an issue in terms of my existing offsite DR process which relies on the ability to use a USB/kickstart process as the network will not be available at the time I am building the servers. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Tim Mooney Sent: Wednesday, August 27, 2008 2:27 PM To: Discussion list about Kickstart Subject: Re: RHEL5.2, kickstart, USB, "OPCODE" error after reboot In regard to: RHEL5.2, kickstart, USB, "OPCODE" error after reboot,...: > Build a HP 385 server, using a RHEL5.2 DVD cut from a downloaded ISO, > and the kickstart file located on a USB stick. After the initial build, > the server reboots, the DVD and USB stick are manually removed and right > where the GRUB screen should be displayed, I get a screen that looks > like a kernel panic screen, though starts with the line "Illegal OPCODE" > and then a bunch of hex output. > > If I build the server without the USB stick but using the same kickstart > file (copied to the DVD) it works fine. Would loading the kickstart file via http work for you? Just put it on some web server and use linux ks=http://your.host/your/url/ks.cfg Tim -- Tim Mooney Tim.Mooney at ndsu.edu Enterprise Computing & Infrastructure 701-231-1076 (Voice) Room 242-J6, IACC Building 701-231-8541 (Fax) North Dakota State University, Fargo, ND 58105-5164 _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list ----------------------------------------- This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country. From Greg.Caetano at hp.com Wed Aug 27 18:39:31 2008 From: Greg.Caetano at hp.com (Caetano, Greg) Date: Wed, 27 Aug 2008 18:39:31 +0000 Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot In-Reply-To: <59B63035E89DEB4C99AB5ECC9C115BB701E1A19D@USATL01ME253.amer.lendlease.com> References: <59B63035E89DEB4C99AB5ECC9C115BB701E1A17F@USATL01ME253.amer.lendlease.com> <59B63035E89DEB4C99AB5ECC9C115BB701E1A19D@USATL01ME253.amer.lendlease.com> Message-ID: Cameron Recheck the parameters for the bootloader directive. I believe it is --driveorder and not --ondrive Greg Caetano Hewlett-Packard Company ESS Software Platform & Business Enablement Solutions Engineering Chicago, IL greg.caetano at hp.com Red Hat Certified Engineer ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Kennedy, Cameron Sent: Wednesday, August 27, 2008 11:52 AM To: Discussion list about Kickstart Subject: RE: RHEL5.2, kickstart, USB, "OPCODE" error after reboot Thanks for the reply. I have this line for the bootloader: "bootloader --location=mbr --ondrive=cciss/c0d0" I added this at the same time as I modified the "part /boot" line shown below. But, this didn't seem to fix the issue. From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Caetano, Greg Sent: Wednesday, August 27, 2008 12:39 PM To: Discussion list about Kickstart Subject: RE: RHEL5.2, kickstart, USB, "OPCODE" error after reboot I would think part of the bootloader is being placed on the USB stick. What do you have for the bootloader statement in kickstart? is there a driveorder parameter on the directive? Greg Caetano Hewlett-Packard Company ESS Software Platform & Business Enablement Solutions Engineering Chicago, IL greg.caetano at hp.com Red Hat Certified Engineer ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Kennedy, Cameron Sent: Wednesday, August 27, 2008 11:33 AM To: Kickstart-list at redhat.com Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot Hi all, Just starting to play around with RHEL5.2 servers in our environment and am coming up with a strange kickstart issue. I have narrowed it down to the following scenario: Build a HP 385 server, using a RHEL5.2 DVD cut from a downloaded ISO, and the kickstart file located on a USB stick. After the initial build, the server reboots, the DVD and USB stick are manually removed and right where the GRUB screen should be displayed, I get a screen that looks like a kernel panic screen, though starts with the line "Illegal OPCODE" and then a bunch of hex output. If I build the server without the USB stick but using the same kickstart file (copied to the DVD) it works fine. Thinking that the USB stick (/dev/sda) was being incorporated into the build process (LVM), I added the line: "ignoredisk --drives=sda,sdb,sdc,sdd,sde,sdf" to the kickstart script. This did not fix the issue. I then added the line: "part /boot --fstype ext3 --size=200 --ondrive=cciss/c0d0" forcing /boot to be created on the internal raid disk, which also failed the fix the issue. I did watch the screen during the last build process and I know that /boot is created on /dev/cciss/c0d0 (HP raid) and the partition information on the USB stick appears to be unchanged (i.e. not formatted). So I am at a loss as to what else I could check. I do know the exact same kickstart script, server, usb stick combination builds a RHEL4.6 server without any problems. Has anyone else seen this or have any ideas? Thanks. ________________________________ "This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country." ________________________________ "This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country." -------------- next part -------------- An HTML attachment was scrubbed... URL: From mpotla at cisco.com Wed Aug 27 18:43:19 2008 From: mpotla at cisco.com (Murali Potla (mpotla)) Date: Wed, 27 Aug 2008 11:43:19 -0700 Subject: How to specify the package architecure within package list in kickstart file In-Reply-To: <59B63035E89DEB4C99AB5ECC9C115BB701E1A20C@USATL01ME253.amer.lendlease.com> Message-ID: <5547014632ED654F971D7E1E0C2E0C3E065E36B8@xmb-sjc-215.amer.cisco.com> I have tried libgcc.i386 It didn't work. Install bombed out Sorry for missing subject line in my first email Thanks, Murali -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Kennedy, Cameron Sent: Wednesday, August 27, 2008 10:56 AM To: Discussion list about Kickstart Subject: RE: (no subject) Personally, I use the %post section to install these type of packages (like Oracle 32bit packages). One idea you could try, though I have not tested this, is specifying the package with an .i386 extension in your packages selection list. e.g. libgcc.i386 -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Murali Potla (mpotla) Sent: Wednesday, August 27, 2008 1:43 PM To: Discussion list about Kickstart Subject: (no subject) Hi all, How can we specify the architecture of a package within kickstart file. I want to install x86 architecture version of libgcc thru kickstart process. How to specify this package in ks file In some cases on 64 bit platforms we need to install 32 bit packages also for backward compatibility for some applications. Thanks, Murali _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list ----------------------------------------- This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Gordon.Keegan at FMR.COM Wed Aug 27 19:00:11 2008 From: Gordon.Keegan at FMR.COM (Keegan, Gordon) Date: Wed, 27 Aug 2008 15:00:11 -0400 Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot References: <59B63035E89DEB4C99AB5ECC9C115BB701E1A17F@USATL01ME253.amer.lendlease.com> Message-ID: <942A778F73A72141889C1F53A8EAA1C50907C2E7@MSGMMKCLM2WIN.DMN1.FMR.COM> Cameron, We've seen similar behavior on a couple of our HP servers. I think there may be a bug in anaconda somewhere. If you can boot to into a rescue mode, check the /boot/grub/device.map file and see if "(hd0)" maps to "dev/cciss/c0d0" like it's supposed to, or if it maps to "/dev/sd##" instead. The couple of times that we've seen it, we have been able to edit the device.map file then re-run the grub-install and it booted up normally. Gordon _____ From: Kennedy, Cameron [mailto:Cameron.Kennedy at bovislendlease.com] Sent: Wednesday, August 27, 2008 12:33 PM To: Kickstart-list at redhat.com Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot Hi all, Just starting to play around with RHEL5.2 servers in our environment and am coming up with a strange kickstart issue. I have narrowed it down to the following scenario: Build a HP 385 server, using a RHEL5.2 DVD cut from a downloaded ISO, and the kickstart file located on a USB stick. After the initial build, the server reboots, the DVD and USB stick are manually removed and right where the GRUB screen should be displayed, I get a screen that looks like a kernel panic screen, though starts with the line "Illegal OPCODE" and then a bunch of hex output. If I build the server without the USB stick but using the same kickstart file (copied to the DVD) it works fine. Thinking that the USB stick (/dev/sda) was being incorporated into the build process (LVM), I added the line: "ignoredisk --drives=sda,sdb,sdc,sdd,sde,sdf" to the kickstart script. This did not fix the issue. I then added the line: "part /boot --fstype ext3 --size=200 --ondrive=cciss/c0d0" forcing /boot to be created on the internal raid disk, which also failed the fix the issue. I did watch the screen during the last build process and I know that /boot is created on /dev/cciss/c0d0 (HP raid) and the partition information on the USB stick appears to be unchanged (i.e. not formatted). So I am at a loss as to what else I could check. I do know the exact same kickstart script, server, usb stick combination builds a RHEL4.6 server without any problems. Has anyone else seen this or have any ideas? Thanks. _____ "This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country." -------------- next part -------------- An HTML attachment was scrubbed... URL: From klaus.steden at thomson.net Wed Aug 27 21:07:55 2008 From: klaus.steden at thomson.net (Klaus Steden) Date: Wed, 27 Aug 2008 14:07:55 -0700 Subject: How to specify the package architecure within package list in kickstart file In-Reply-To: <5547014632ED654F971D7E1E0C2E0C3E065E36B8@xmb-sjc-215.amer.cisco.com> Message-ID: Murali, That's the expected syntax for the '%packages' section ... -libgcc.i386 libgcc.x86_64 etc. What was the error message when it bombed out? It's not necessarily related. Klaus On 8/27/08 11:43 AM, "Murali Potla (mpotla)" did etch on stone tablets: > I have tried libgcc.i386 > > It didn't work. Install bombed out > > Sorry for missing subject line in my first email > > Thanks, > Murali > > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Kennedy, Cameron > Sent: Wednesday, August 27, 2008 10:56 AM > To: Discussion list about Kickstart > Subject: RE: (no subject) > > Personally, I use the %post section to install these type of packages > (like Oracle 32bit packages). > One idea you could try, though I have not tested this, is specifying the > package with an .i386 extension in your packages selection list. > > e.g. libgcc.i386 > > > > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Murali Potla > (mpotla) > Sent: Wednesday, August 27, 2008 1:43 PM > To: Discussion list about Kickstart > Subject: (no subject) > > Hi all, > > How can we specify the architecture of a package within kickstart file. > > I want to install x86 architecture version of libgcc thru kickstart > process. How to specify this package in ks file > > In some cases on 64 bit platforms we need to install 32 bit packages > also for backward compatibility for some applications. > > Thanks, > Murali > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > > ----------------------------------------- This email (including any > attachments) is confidential. If you are not the intended recipient you > must not copy, use, disclose, distribute or rely on the information > contained in it. If you have received this email in error, please > notify the sender immediately by reply email and delete the email from > your system. Confidentiality and legal privilege attached to this > communication are not waived or lost by reason of mistaken delivery to > you. Lend Lease does not guarantee that this email or the attachment(s) > are unaffected by computer virus, corruption or other defects. Lend > Lease may monitor incoming and outgoing emails for compliance with its > Email Policy. Please note that our servers may not be located in your > country. > > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From Cameron.Kennedy at bovislendlease.com Thu Aug 28 12:07:51 2008 From: Cameron.Kennedy at bovislendlease.com (Kennedy, Cameron) Date: Thu, 28 Aug 2008 08:07:51 -0400 Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot In-Reply-To: <942A778F73A72141889C1F53A8EAA1C50907C2E7@MSGMMKCLM2WIN.DMN1.FMR.COM> References: <59B63035E89DEB4C99AB5ECC9C115BB701E1A17F@USATL01ME253.amer.lendlease.com> <942A778F73A72141889C1F53A8EAA1C50907C2E7@MSGMMKCLM2WIN.DMN1.FMR.COM> Message-ID: <59B63035E89DEB4C99AB5ECC9C115BB701E1A5E8@USATL01ME253.amer.lendlease.com> Thanks Gordon and Greg for your advice. I was able to test both options and found them to work. Going forward "-driveorder" will have to be the option I use to get a consistent result, though, as this is primarily for DR planning purposes and I can't be guaranteed that I will have HP hardware (though we generally do in tests) I might also have to use some type of %pre script to determine if this is a HP server or something else and adjust the "-driveorder" parameter accordingly. It is odd that the "ignoredisk" option doesn't seem to take effect. Thanks From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Keegan, Gordon Sent: Wednesday, August 27, 2008 3:00 PM To: Discussion list about Kickstart Subject: RE: RHEL5.2, kickstart, USB, "OPCODE" error after reboot Cameron, We've seen similar behavior on a couple of our HP servers. I think there may be a bug in anaconda somewhere. If you can boot to into a rescue mode, check the /boot/grub/device.map file and see if "(hd0)" maps to "dev/cciss/c0d0" like it's supposed to, or if it maps to "/dev/sd##" instead. The couple of times that we've seen it, we have been able to edit the device.map file then re-run the grub-install and it booted up normally. Gordon ________________________________ From: Kennedy, Cameron [mailto:Cameron.Kennedy at bovislendlease.com] Sent: Wednesday, August 27, 2008 12:33 PM To: Kickstart-list at redhat.com Subject: RHEL5.2, kickstart, USB, "OPCODE" error after reboot Hi all, Just starting to play around with RHEL5.2 servers in our environment and am coming up with a strange kickstart issue. I have narrowed it down to the following scenario: Build a HP 385 server, using a RHEL5.2 DVD cut from a downloaded ISO, and the kickstart file located on a USB stick. After the initial build, the server reboots, the DVD and USB stick are manually removed and right where the GRUB screen should be displayed, I get a screen that looks like a kernel panic screen, though starts with the line "Illegal OPCODE" and then a bunch of hex output. If I build the server without the USB stick but using the same kickstart file (copied to the DVD) it works fine. Thinking that the USB stick (/dev/sda) was being incorporated into the build process (LVM), I added the line: "ignoredisk --drives=sda,sdb,sdc,sdd,sde,sdf" to the kickstart script. This did not fix the issue. I then added the line: "part /boot --fstype ext3 --size=200 --ondrive=cciss/c0d0" forcing /boot to be created on the internal raid disk, which also failed the fix the issue. I did watch the screen during the last build process and I know that /boot is created on /dev/cciss/c0d0 (HP raid) and the partition information on the USB stick appears to be unchanged (i.e. not formatted). So I am at a loss as to what else I could check. I do know the exact same kickstart script, server, usb stick combination builds a RHEL4.6 server without any problems. Has anyone else seen this or have any ideas? Thanks. ________________________________ "This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country." ----------------------------------------- This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country. -------------- next part -------------- An HTML attachment was scrubbed... URL: