From rigg0022 at umn.edu Wed Jan 2 22:13:08 2008 From: rigg0022 at umn.edu (Riggs, Ben) Date: Wed, 02 Jan 2008 16:13:08 -0600 Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? Message-ID: <477C0C74.5070206@umn.edu> We have new motherboards with onboard NICs which aren't recognized by the fedora 7 installer initrd/kernel. Naturally, not having a NIC makes using network drivers impossible. Anyone have the skinny, or a link? Thanks, Ben -- Systems Staff School of Mathematics University of Minnesota From Chip.Shabazian at bankofamerica.com Wed Jan 2 22:17:55 2008 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Wed, 02 Jan 2008 14:17:55 -0800 Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? In-Reply-To: <477C0C74.5070206@umn.edu> Message-ID: Only two ways I know of: 1) Lots of work to rebuild with your nic drivers 2) put the drivers on the CD/Flash that you are using for the install and use dd to load them If you are using PXEBoot, you've gotta go with #1 -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Riggs, Ben Sent: Wednesday, January 02, 2008 2:13 PM To: Discussion list about Kickstart Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? We have new motherboards with onboard NICs which aren't recognized by the fedora 7 installer initrd/kernel. Naturally, not having a NIC makes using network drivers impossible. Anyone have the skinny, or a link? Thanks, Ben -- Systems Staff School of Mathematics University of Minnesota _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From rigg0022 at umn.edu Wed Jan 2 22:26:29 2008 From: rigg0022 at umn.edu (Riggs, Ben) Date: Wed, 02 Jan 2008 16:26:29 -0600 Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? In-Reply-To: References: Message-ID: <477C0F95.2090705@umn.edu> Being an undergrad with no classes for the next couple weeks and plenty of time at work, I'm looking to find out what's behind door #1. Ben Shabazian, Chip wrote: > Only two ways I know of: > > 1) Lots of work to rebuild with your nic drivers > 2) put the drivers on the CD/Flash that you are using for the install > and use dd to load them > > If you are using PXEBoot, you've gotta go with #1 > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Riggs, Ben > Sent: Wednesday, January 02, 2008 2:13 PM > To: Discussion list about Kickstart > Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? > > We have new motherboards with onboard NICs which aren't recognized by > the fedora 7 installer initrd/kernel. Naturally, not having a NIC makes > using network drivers impossible. Anyone have the skinny, or a link? > > Thanks, > Ben > -- > Systems Staff > School of Mathematics > University of Minnesota From timborn at alcatel-lucent.com Wed Jan 2 22:34:25 2008 From: timborn at alcatel-lucent.com (Tim Born) Date: Wed, 02 Jan 2008 16:34:25 -0600 Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? In-Reply-To: <477C0C74.5070206@umn.edu> References: <477C0C74.5070206@umn.edu> Message-ID: <477C1171.3070506@alcatel-lucent.com> Riggs, Ben wrote: > We have new motherboards with onboard NICs which aren't recognized by > the fedora 7 installer initrd/kernel. Naturally, not having a NIC > makes using network drivers impossible. Anyone have the skinny, or a > link? > > Thanks, > Ben > -- > Systems Staff > School of Mathematics > University of Minnesota > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list Not tried it, so your milage may vary ... I think you want to find or setup a machine that has the same kernel as the one you are trying to install then use mkinitrd to create a custom initrd including your drivers: /sbin/mkinitrd --with=$yourdriver custom.img `uname -r` Once that's built, swap it in for the initrd on your kickstart or burn a custom ISO. good luck -tim -- Tim Born Alcatel-Lucent Ventures (ALV) timborn at alcatel-lucent.com +1 630/979-3118 -- "Not a single open-source software program has ever been sued for patent infringement, much less be found to infringe, while proprietary software, like Windows, is sued and found guilty of patent infringement quite frequently." -Dan Ravicher, author of the OSRM study and the executive director of the Public Patent Foundation From michael at kmaclub.com Wed Jan 2 22:44:49 2008 From: michael at kmaclub.com (michael at kmaclub.com) Date: Wed, 02 Jan 2008 15:44:49 -0700 Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? In-Reply-To: <477C1171.3070506@alcatel-lucent.com> References: <477C0C74.5070206@umn.edu> <477C1171.3070506@alcatel-lucent.com> Message-ID: <477C13E1.7020001@kmaclub.com> Tim Born wrote: > Riggs, Ben wrote: > >> We have new motherboards with onboard NICs which aren't recognized by >> the fedora 7 installer initrd/kernel. Naturally, not having a NIC >> makes using network drivers impossible. Anyone have the skinny, or a >> link? >> >> Thanks, >> Ben >> -- >> Systems Staff >> School of Mathematics >> University of Minnesota >> >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list > > Not tried it, so your milage may vary ... > > I think you want to find or setup a machine that has the same kernel as > the one you are trying to install > then use mkinitrd to create a custom initrd including your drivers: > > /sbin/mkinitrd --with=$yourdriver custom.img `uname -r` > > Once that's built, swap it in for the initrd on your kickstart or burn a > custom ISO. > I used to do this in the RH7/RH8 days so things may have changed. Best I recall I would add the kernel to my build machine that matches what is on the boot media and boot from it. Build the network modules for the new driver. Mount the boot media Extract the initrd Add the new drivers to the extracted initrd Update the pciids to include the new hardware ( if required ) Create a new initrd Update boot media with the new initrd Michael From koreesmith at gmail.com Wed Jan 2 22:50:12 2008 From: koreesmith at gmail.com (Koree A. Smith) Date: Wed, 2 Jan 2008 16:50:12 -0600 Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? In-Reply-To: <477C13E1.7020001@kmaclub.com> References: <477C0C74.5070206@umn.edu> <477C1171.3070506@alcatel-lucent.com> <477C13E1.7020001@kmaclub.com> Message-ID: <676847c60801021450l4e246b8bwbb9d44c97cc0524a@mail.gmail.com> This is something I did a long time ago to add some broadcom drivers that were newer than what RHEL 3 and 4 had to offer at the time. I believe this is the document I used to figure out how to do this: http://lists.us.dell.com/pipermail/linux-poweredge/2003-August/009343.html It is for RH7.3, but worked on RHEL 3 and 4. I would guess that it would be somewhat similar for RHEL 5, although I have not done this on RHEL 5 as of yet. Thanks, Koree -- Koree A. Smith, RHCE -- koreesmith at gmail.com koree at ameth.org On Jan 2, 2008 4:44 PM, michael at kmaclub.com wrote: > Tim Born wrote: > > Riggs, Ben wrote: > > > >> We have new motherboards with onboard NICs which aren't recognized by > >> the fedora 7 installer initrd/kernel. Naturally, not having a NIC > >> makes using network drivers impossible. Anyone have the skinny, or a > >> link? > >> > >> Thanks, > >> Ben > >> -- > >> Systems Staff > >> School of Mathematics > >> University of Minnesota > >> > >> _______________________________________________ > >> Kickstart-list mailing list > >> Kickstart-list at redhat.com > >> https://www.redhat.com/mailman/listinfo/kickstart-list > > > > Not tried it, so your milage may vary ... > > > > I think you want to find or setup a machine that has the same kernel as > > the one you are trying to install > > then use mkinitrd to create a custom initrd including your drivers: > > > > /sbin/mkinitrd --with=$yourdriver custom.img `uname -r` > > > > Once that's built, swap it in for the initrd on your kickstart or burn a > > custom ISO. > > > > > I used to do this in the RH7/RH8 days so things may have changed. > > Best I recall I would add the kernel to my build machine that matches > what is on the boot media and boot from it. > > Build the network modules for the new driver. > > Mount the boot media > Extract the initrd > Add the new drivers to the extracted initrd > Update the pciids to include the new hardware ( if required ) > Create a new initrd > Update boot media with the new initrd > > Michael > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mdehaan at redhat.com Fri Jan 4 21:19:53 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 04 Jan 2008 16:19:53 -0500 Subject: Reminder: kickstart-list has an IRC channel Message-ID: <477EA2F9.6090003@redhat.com> #kickstart on irc.freenode.net If you're on IRC already, join up! --Michael From Cameron.Kennedy at bovislendlease.com Mon Jan 7 13:46:28 2008 From: Cameron.Kennedy at bovislendlease.com (Kennedy, Cameron) Date: Mon, 7 Jan 2008 08:46:28 -0500 Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? In-Reply-To: References: <477C0C74.5070206@umn.edu> Message-ID: <544663BCC5B1E34698C8150370A668F30730B415@USATL01ME253.amer.lendlease.com> If I use the 2nd option below, what is the syntax to have dd look in a place other than a floppy for the drivers? In other words, I have server with USB and Cd-rom only, and no usb floppy available, but I need to have the raid drivers installed at system build time. I tried using dd but only get the option for it to search a floppy for the drivers. I am using the bootimg from a rhel4.5 cd and other than the lack of raid drivers, I can get the server built using kickstart via ftp option. If this is not possible, I can try option 1. Thanks for any ideas. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip Sent: Wednesday, January 02, 2008 5:18 PM To: Discussion list about Kickstart Subject: RE: How to rebuild an Anaconda initrd/kernel with updated drivers? Only two ways I know of: 1) Lots of work to rebuild with your nic drivers 2) put the drivers on the CD/Flash that you are using for the install and use dd to load them If you are using PXEBoot, you've gotta go with #1 -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Riggs, Ben Sent: Wednesday, January 02, 2008 2:13 PM To: Discussion list about Kickstart Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? We have new motherboards with onboard NICs which aren't recognized by the fedora 7 installer initrd/kernel. Naturally, not having a NIC makes using network drivers impossible. Anyone have the skinny, or a link? Thanks, Ben -- Systems Staff School of Mathematics University of Minnesota _______________________________________________ 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 ----------------------------------------- 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 Pablo.Iranzo at redhat.com Mon Jan 7 14:20:04 2008 From: Pablo.Iranzo at redhat.com (Pablo Iranzo =?ISO-8859-1?Q?G=F3mez?=) Date: Mon, 07 Jan 2008 15:20:04 +0100 Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? In-Reply-To: <544663BCC5B1E34698C8150370A668F30730B415@USATL01ME253.amer.lendlease.com> References: <477C0C74.5070206@umn.edu> <544663BCC5B1E34698C8150370A668F30730B415@USATL01ME253.amer.lendlease.com> Message-ID: <1199715604.10657.10.camel@iranzo.users.redhat.com> Hi Try with this "driverdisk --source=ftp://path/to/dd.img driverdisk --source=http://path/to/dd.img driverdisk --source=nfs:host:/path/to/img " syntax, check first if it works with your particular release at the documentation, if so, putting it at your kickstarting, would help. Regards Pablo El lun, 07-01-2008 a las 08:46 -0500, Kennedy, Cameron escribi?: > If I use the 2nd option below, what is the syntax to have dd look in a > place other than a floppy for the drivers? > In other words, I have server with USB and Cd-rom only, and no usb > floppy available, but I need to have the raid drivers installed at > system build time. I tried using dd but only get the option for it to > search a floppy for the drivers. I am using the bootimg from a rhel4.5 > cd and other than the lack of raid drivers, I can get the server built > using kickstart via ftp option. > > If this is not possible, I can try option 1. > Thanks for any ideas. > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip > Sent: Wednesday, January 02, 2008 5:18 PM > To: Discussion list about Kickstart > Subject: RE: How to rebuild an Anaconda initrd/kernel with updated > drivers? > > Only two ways I know of: > > 1) Lots of work to rebuild with your nic drivers > 2) put the drivers on the CD/Flash that you are using for the install > and use dd to load them > > If you are using PXEBoot, you've gotta go with #1 > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Riggs, Ben > Sent: Wednesday, January 02, 2008 2:13 PM > To: Discussion list about Kickstart > Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? > > We have new motherboards with onboard NICs which aren't recognized by > the fedora 7 installer initrd/kernel. Naturally, not having a NIC makes > using network drivers impossible. Anyone have the skinny, or a link? > > Thanks, > Ben > -- > Systems Staff > School of Mathematics > University of Minnesota > > _______________________________________________ > 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 > > ----------------------------------------- > 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 -- Pablo Iranzo G?mez (Pablo.Iranzo at redhat.com) RHCE/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 debian at herakles.homelinux.org Mon Jan 7 23:33:40 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Tue, 08 Jan 2008 08:33:40 +0900 Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? In-Reply-To: <477C0F95.2090705@umn.edu> References: <477C0F95.2090705@umn.edu> Message-ID: <4782B6D4.1080400@herakles.homelinux.org> Riggs, Ben wrote: > Being an undergrad with no classes for the next couple weeks and plenty > of time at work, I'm looking to find out what's behind door #1. I think it goes like this. I'm assuming pxe. cat | gunzip | cpio Inspect the tree you've created Pop the driver module in place (there may be a further cpio or similar archive you need to open). Repack if necessary read linuxrc to see where it goes. Change whatever seems good find | cpio | gzip > test boot. repeat as necessary. If you're booting from CD, you can do the above, but you need to unroll the ISO first: mount -o loop ... /mnt/iso find /mnt/iso -name TRANS.TBL -exec rm -vf {} \; Find the imbedded boot floppy and mount that. Do as above. You may need to remove the initrd from the floppy before putting it back. mkisofs cdrecord Note, I have assumed your driver's built to match the install kernel. This is necessary. Wonder what to do for the next ten days. > > Ben > > Shabazian, Chip wrote: >> Only two ways I know of: >> >> 1) Lots of work to rebuild with your nic drivers >> 2) put the drivers on the CD/Flash that you are using for the install >> and use dd to load them >> >> If you are using PXEBoot, you've gotta go with #1 >> >> -----Original Message----- >> From: kickstart-list-bounces at redhat.com >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Riggs, Ben >> Sent: Wednesday, January 02, 2008 2:13 PM >> To: Discussion list about Kickstart >> Subject: How to rebuild an Anaconda initrd/kernel with updated drivers? >> >> We have new motherboards with onboard NICs which aren't recognized by >> the fedora 7 installer initrd/kernel. Naturally, not having a NIC makes >> using network drivers impossible. Anyone have the skinny, or a link? >> >> Thanks, >> Ben -- 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 rvandolson at esri.com Wed Jan 9 22:29:21 2008 From: rvandolson at esri.com (Ray Van Dolson) Date: Wed, 9 Jan 2008 14:29:21 -0800 Subject: Managing many kickstart configurations. Message-ID: <20080109222921.GA2098@esri.com> Hey all, how do some of you who have many different kickstart configurations manage this? I have similar configurations for different minor versions of RH, but under the same major version -- and the same for CentOS. Much of the config files are the same save for a few poritions. I'm considering just writing a small CGI front-end that nicely pulls in a "common" config and then more specific items. Sure this can kind of be done via the %include keyword, but somehow doing this has never seemed as elegant. Anyone have anything particularly cool they're doing to manage large numbers of kickstart configs? I seem to remember stumbling across a web-based tool at some point but can't find it now. Ray From secroft at micron.com Wed Jan 9 22:36:33 2008 From: secroft at micron.com (secroft at micron.com) Date: Wed, 9 Jan 2008 15:36:33 -0700 Subject: Managing many kickstart configurations. In-Reply-To: <20080109222921.GA2098@esri.com> Message-ID: <3BA8753665DB594C8C35CFA8801AD52201EFFECA@ntxboimbx32.micron.com> Ray, Look at the cobbler project http://cobbler.et.redhat.com as that should meet your needs. Scott -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ray Van Dolson Sent: Wednesday, January 09, 2008 3:29 PM To: kickstart-list at redhat.com Subject: Managing many kickstart configurations. Hey all, how do some of you who have many different kickstart configurations manage this? I have similar configurations for different minor versions of RH, but under the same major version -- and the same for CentOS. Much of the config files are the same save for a few poritions. I'm considering just writing a small CGI front-end that nicely pulls in a "common" config and then more specific items. Sure this can kind of be done via the %include keyword, but somehow doing this has never seemed as elegant. Anyone have anything particularly cool they're doing to manage large numbers of kickstart configs? I seem to remember stumbling across a web-based tool at some point but can't find it now. Ray _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From rvandolson at esri.com Wed Jan 9 22:38:15 2008 From: rvandolson at esri.com (Ray Van Dolson) Date: Wed, 9 Jan 2008 14:38:15 -0800 Subject: Managing many kickstart configurations. In-Reply-To: <3BA8753665DB594C8C35CFA8801AD52201EFFECA@ntxboimbx32.micron.com> References: <20080109222921.GA2098@esri.com> <3BA8753665DB594C8C35CFA8801AD52201EFFECA@ntxboimbx32.micron.com> Message-ID: <20080109223814.GA2206@esri.com> Thanks guys, will check it out. I actually had this bookmarked but went right over it. :) From mdehaan at redhat.com Thu Jan 10 16:27:59 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 10 Jan 2008 11:27:59 -0500 Subject: Managing many kickstart configurations. In-Reply-To: <20080109222921.GA2098@esri.com> References: <20080109222921.GA2098@esri.com> Message-ID: <4786478F.8030607@redhat.com> Ray Van Dolson wrote: > Hey all, how do some of you who have many different kickstart > configurations manage this? I have similar configurations for > different minor versions of RH, but under the same major version -- and > the same for CentOS. Much of the config files are the same save for a > few poritions. > Cobbler has a fairly strong kickstart templating engine for doing this: See http://cobbler.et.redhat.com --Michael From mdehaan at redhat.com Thu Jan 10 16:29:00 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 10 Jan 2008 11:29:00 -0500 Subject: Managing many kickstart configurations. In-Reply-To: <3BA8753665DB594C8C35CFA8801AD52201EFFECA@ntxboimbx32.micron.com> References: <3BA8753665DB594C8C35CFA8801AD52201EFFECA@ntxboimbx32.micron.com> Message-ID: <478647CC.6030801@redhat.com> secroft at micron.com wrote: > Ray, > Look at the cobbler project http://cobbler.et.redhat.com as that > should meet your needs. > > Scott > Sorry for the additional post, didn't see the reply... Ray, yes, check it out ... if you have any questions, feel free to stop by #cobbler on irc.freenode.net or post on the mailing list. --Michael > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ray Van Dolson > Sent: Wednesday, January 09, 2008 3:29 PM > To: kickstart-list at redhat.com > Subject: Managing many kickstart configurations. > > Hey all, how do some of you who have many different kickstart > configurations manage this? I have similar configurations for different > minor versions of RH, but under the same major version -- and the same > for CentOS. Much of the config files are the same save for a few > poritions. > > I'm considering just writing a small CGI front-end that nicely pulls in > a "common" config and then more specific items. > > Sure this can kind of be done via the %include keyword, but somehow > doing this has never seemed as elegant. > > Anyone have anything particularly cool they're doing to manage large > numbers of kickstart configs? I seem to remember stumbling across a > web-based tool at some point but can't find it now. > > Ray > > _______________________________________________ > 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 squid at oranged.to Fri Jan 11 15:05:38 2008 From: squid at oranged.to (Jimmy Stewpot) Date: Fri, 11 Jan 2008 15:05:38 +0000 Subject: swap sizing in a kickstart script Message-ID: <478785C2.20307@oranged.to> Hello, Is there an automated way in the kickstart file for the system to automatically create the "redhat recommended" swap size? Regards, Jimmy From Pablo.Iranzo at redhat.com Fri Jan 11 15:11:01 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Fri, 11 Jan 2008 16:11:01 +0100 (CET) Subject: swap sizing in a kickstart script In-Reply-To: <478785C2.20307@oranged.to> References: <478785C2.20307@oranged.to> Message-ID: Sure, create a %pre section that obtains memory available on system, then create some conditionals to size it adecuately and wrote a "part" sentence with the recomended size to a /tmp/swappart file... then, %include /tmp/swappart with the rest of your partitions in order to have it included for kickstarting :) Regards Pablo -- 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?? On Fri, 11 Jan 2008, Jimmy Stewpot wrote: > Hello, > > Is there an automated way in the kickstart file for the system to > automatically create the "redhat recommended" swap size? > > Regards, > > Jimmy > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From Gerrard.Geldenhuis at datacash.com Fri Jan 11 15:19:47 2008 From: Gerrard.Geldenhuis at datacash.com (Gerrard Geldenhuis) Date: Fri, 11 Jan 2008 15:19:47 -0000 Subject: swap sizing in a kickstart script In-Reply-To: <478785C2.20307@oranged.to> References: <478785C2.20307@oranged.to> Message-ID: Hi Jimmy, I am not sure if there is a variable that you can use to calculate swap size based on memory. In our situation we know all the machines have the same amount of memory and allocate a static size. part swap --size=1000 --maxsize=2000 I am not aware of a "Redhat recommended" swap size. The convention used to be to have swap double the memory, but these days I don't know how appropriate that is. I don't fancy an >8Gb swap file... Regards > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Jimmy Stewpot > Sent: 11 January 2008 15:06 > To: kickstart-list at redhat.com > Subject: swap sizing in a kickstart script > > Hello, > > Is there an automated way in the kickstart file for the system to > automatically create the "redhat recommended" swap size? > > Regards, > > Jimmy From rainer at ultra-secure.de Fri Jan 11 15:26:40 2008 From: rainer at ultra-secure.de (Rainer Duffner) Date: Fri, 11 Jan 2008 16:26:40 +0100 Subject: swap sizing in a kickstart script In-Reply-To: References: <478785C2.20307@oranged.to> Message-ID: <47878AB0.8030701@ultra-secure.de> Gerrard Geldenhuis wrote: > Hi Jimmy, > I am not sure if there is a variable that you can use to calculate swap > size based on memory. > > In our situation we know all the machines have the same amount of memory > and allocate a static size. > > part swap --size=1000 --maxsize=2000 > > I am not aware of a "Redhat recommended" swap size. The convention used > to be to have swap double the memory, but these days I don't know how > appropriate that is. I don't fancy an >8Gb swap file... > > Swap should be at least a bit more than your maximum anticipated RAM-size. So you can make crash-dumps easily. If your box can take 32 GB RAM (AMD-servers of late - latest Intel can even take more), this might be problematic ;-) cheers, Rainer From squid at oranged.to Fri Jan 11 15:38:42 2008 From: squid at oranged.to (Jimmy Stewpot) Date: Fri, 11 Jan 2008 15:38:42 +0000 Subject: swap sizing in a kickstart script In-Reply-To: <478785C2.20307@oranged.to> References: <478785C2.20307@oranged.to> Message-ID: <47878D82.9000902@oranged.to> Hi All, Jimmy Stewpot wrote: > Hello, > > Is there an automated way in the kickstart file for the system to > automatically create the "redhat recommended" swap size? > > Regards, > > Jimmy > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list It seems that there is a --recommended variable in the code which will do it. After seeing that I found it in the redhat documentation http://www.redhat.com/docs/manuals/enterprise/RHEL-5-manual/en-US/RHEL510/Installation_Guide/s1-kickstart2-options.html Search for --recommended Regards From Gerrard.Geldenhuis at datacash.com Fri Jan 11 15:50:42 2008 From: Gerrard.Geldenhuis at datacash.com (Gerrard Geldenhuis) Date: Fri, 11 Jan 2008 15:50:42 -0000 Subject: swap sizing in a kickstart script In-Reply-To: <47878D82.9000902@oranged.to> References: <478785C2.20307@oranged.to> <47878D82.9000902@oranged.to> Message-ID: Hi I don't completely understand it but you have to set your overcommit ratio in the kernel to match you swap space Percentage of physical memory size to include in overcommit calculations. Memory allocation limit = swapspace + physmem * (overcommit_ratio / 100) swapspace = total size of all swap areas physmem = size of physical memory in system cat /proc/sys/vm/overcommit_ratio Unless I have the wrong end of the stick, then please correct me. I understood overcommit ratio to be the amount of swap space expected as a ratio of the amount of ram but after reading up on it again I don't think I understand it anymore. Apologies for complicating your answer. Regards > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Jimmy Stewpot > Sent: 11 January 2008 15:39 > To: squid at oranged.to; Discussion list about Kickstart > Subject: Re: swap sizing in a kickstart script > > Hi All, > > Jimmy Stewpot wrote: > > Hello, > > > > Is there an automated way in the kickstart file for the system to > > automatically create the "redhat recommended" swap size? > > > > Regards, > > > > Jimmy > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > It seems that there is a --recommended variable in the code which will > do it. After seeing that I found it in the redhat documentation > > http://www.redhat.com/docs/manuals/enterprise/RHEL-5-manual/en- > US/RHEL510/Installation_Guide/s1-kickstart2-options.html > > Search for --recommended > > Regard From Gerrard.Geldenhuis at datacash.com Fri Jan 11 16:32:09 2008 From: Gerrard.Geldenhuis at datacash.com (Gerrard Geldenhuis) Date: Fri, 11 Jan 2008 16:32:09 -0000 Subject: swap sizing in a kickstart script In-Reply-To: References: <478785C2.20307@oranged.to> <47878D82.9000902@oranged.to> Message-ID: Just read this which clarified it for me: BTW, with overcommit_memory=2, CommitLimit = (MemTotal * (overcommit_ratio / 100)) + SwapFree With the default of overcommit_ratio = 50, this simplifies to: CommitLimit = (MemTotal / 2) + SwapFree Unless you're setting overcommit_ratio to 100, this means that with 640MB of RAM and no swap, user programs are allowed to request 320MB of memory at most. I hope this helps... > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Gerrard Geldenhuis > Sent: 11 January 2008 15:51 > To: squid at oranged.to; Discussion list about Kickstart > Subject: RE: swap sizing in a kickstart script > > Hi > I don't completely understand it but you have to set your overcommit > ratio in the kernel to match you swap space > > Percentage of physical memory size to include in overcommit > calculations. > > Memory allocation limit = swapspace + physmem * (overcommit_ratio / 100) > > swapspace = total size of all swap areas > physmem = size of physical memory in system > > cat /proc/sys/vm/overcommit_ratio > > Unless I have the wrong end of the stick, then please correct me. > > I understood overcommit ratio to be the amount of swap space expected as > a ratio of the amount of ram but after reading up on it again I don't > think I understand it anymore. > > Apologies for complicating your answer. > > Regards > > > -----Original Message----- > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > > bounces at redhat.com] On Behalf Of Jimmy Stewpot > > Sent: 11 January 2008 15:39 > > To: squid at oranged.to; Discussion list about Kickstart > > Subject: Re: swap sizing in a kickstart script > > > > Hi All, > > > > Jimmy Stewpot wrote: > > > Hello, > > > > > > Is there an automated way in the kickstart file for the system to > > > automatically create the "redhat recommended" swap size? > > > > > > Regards, > > > > > > Jimmy > > > > > > _______________________________________________ > > > Kickstart-list mailing list > > > Kickstart-list at redhat.com > > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > It seems that there is a --recommended variable in the code which will > > do it. After seeing that I found it in the redhat documentation > > > > http://www.redhat.com/docs/manuals/enterprise/RHEL-5-manual/en- > > US/RHEL510/Installation_Guide/s1-kickstart2-options.html > > > > Search for --recommended > > > > Regard > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From douglas at hubler.us Mon Jan 14 19:34:34 2008 From: douglas at hubler.us (Douglas Hubler) Date: Mon, 14 Jan 2008 19:34:34 +0000 (UTC) Subject: error using kickstart file Message-ID: Hello All, I'm getting this error from anaconda 11.3.0.50 doing a kickstart install from cdrom or the http on the F8 release built using pungi (a tool that assembles packages and files to create an ISO installer) Stack Trace File "/usr/bin/anaconda", line 917, in kickstart.processKickstartFile(anaconda, opts.ksfile) ... (omitted) File "/usr/lib/python2.5/site-packages/pykickstart/commands/method.py", line 81, in parse (opts, extra) = op.parse_args(args=args) AttributeError: 'NoneType' object has no attribute 'parse_args' install exited abnormally [1/1] I originally posted here http://thread.gmane.org/gmane.linux.redhat.fedora.buildsystem/1158 And to anaconda list... http://thread.gmane.org/gmane.linux.redhat.anaconda.devel/5616 And to re-spin list... no link access * another person got it too. * someone suggested I post to re-spin project for updates but I did not hear back from anyone there. I think access to the source requires authorization * I would happily build my own updates.img file from the latest anaconda code if someone thinks that may work and can point me to the code I get this error with every incarnation of kickstart files. Here's the simplest one I have ----------------------------- install cdrom lang en_US.UTF-8 keyboard us %packages @core anaconda-runtime iscsi-initiator-utils memtest86+ vnc-server %end From lfarkas at bppiac.hu Wed Jan 16 11:20:22 2008 From: lfarkas at bppiac.hu (Farkas Levente) Date: Wed, 16 Jan 2008 12:20:22 +0100 Subject: zerombr Message-ID: <478DE876.2060807@bppiac.hu> hi, do i have to add a yes parameter to zerombr in kickstart file on rhel5 and fedora 8? there are different docs and different options. thanks. -- Levente "Si vis pacem para bellum!" From Gerrard.Geldenhuis at datacash.com Wed Jan 16 11:37:23 2008 From: Gerrard.Geldenhuis at datacash.com (Gerrard Geldenhuis) Date: Wed, 16 Jan 2008 11:37:23 -0000 Subject: zerombr In-Reply-To: <478DE876.2060807@bppiac.hu> References: <478DE876.2060807@bppiac.hu> Message-ID: Hi In rhel5.1 we use zerombr yes so yes. ;-) Don't know about fedora. Experiment see what happens... Regards > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Farkas Levente > Sent: 16 January 2008 11:20 > To: Discussion list about Kickstart > Subject: zerombr > > hi, > do i have to add a yes parameter to zerombr in kickstart file on rhel5 > and fedora 8? there are different docs and different options. > thanks. > > -- > Levente "Si vis pacem para bellum!" > From lfarkas at bppiac.hu Wed Jan 16 13:28:37 2008 From: lfarkas at bppiac.hu (Farkas Levente) Date: Wed, 16 Jan 2008 14:28:37 +0100 Subject: delete lvm groups in kickstart Message-ID: <478E0685.1070404@bppiac.hu> hi, in kickstart one can use clearpart to remove old partition, but there is no command to remove old lvm volgroup or logvols. and if there are volgroup or logvols already on the system where you want to install through kickstart the install process failed with a python stack trace since eg. VolGroup00 already exist and can't be created. currently there is no other way to delete only in %pre scripts but imho it'd be an essential feature. yours. -- Levente "Si vis pacem para bellum!" From Gerrard.Geldenhuis at datacash.com Wed Jan 16 13:33:17 2008 From: Gerrard.Geldenhuis at datacash.com (Gerrard Geldenhuis) Date: Wed, 16 Jan 2008 13:33:17 -0000 Subject: delete lvm groups in kickstart In-Reply-To: <478E0685.1070404@bppiac.hu> References: <478E0685.1070404@bppiac.hu> Message-ID: Hi I am assuming you don't want to use clearpart --all? Regards > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Farkas Levente > Sent: 16 January 2008 13:29 > To: Discussion list about Kickstart > Subject: delete lvm groups in kickstart > > hi, > in kickstart one can use clearpart to remove old partition, but there is > no command to remove old lvm volgroup or logvols. and if there are > volgroup or logvols already on the system where you want to install > through kickstart the install process failed with a python stack trace > since eg. VolGroup00 already exist and can't be created. > currently there is no other way to delete only in %pre scripts but imho > it'd be an essential feature. > yours. From lfarkas at bppiac.hu Wed Jan 16 13:37:39 2008 From: lfarkas at bppiac.hu (Farkas Levente) Date: Wed, 16 Jan 2008 14:37:39 +0100 Subject: where should i have to use %end Message-ID: <478E08A3.1060805@bppiac.hu> hi, where should i have to use %end in kickstart file? after each section or just pre and post or ...? i can't find any documentation about it but python kickstart parser in revisor complain about it that %end missing and this kickstart file will be depricated. thanks. -- Levente "Si vis pacem para bellum!" From lfarkas at bppiac.hu Wed Jan 16 13:44:16 2008 From: lfarkas at bppiac.hu (Farkas Levente) Date: Wed, 16 Jan 2008 14:44:16 +0100 Subject: delete lvm groups in kickstart In-Reply-To: References: <478E0685.1070404@bppiac.hu> Message-ID: <478E0A30.7020605@bppiac.hu> Gerrard Geldenhuis wrote: > Hi > I am assuming you don't want to use > clearpart --all? i use clearpart, BUT only on the system disk, ie: clearpart --initlabel --all --drives=sda while there are other data disk in the system and lvm groups are system wide not disk specific. there are some system where i would like to keep data on the data disks (those do not cointans volgroups) so i can't clearpart those disk. but if there is not data on them there may can be volgroup on them. but if i'd be able to lvremove VolGroup00 in kickstart then it'd be good for me. > Regards > >> -----Original Message----- >> From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- >> bounces at redhat.com] On Behalf Of Farkas Levente >> Sent: 16 January 2008 13:29 >> To: Discussion list about Kickstart >> Subject: delete lvm groups in kickstart >> >> hi, >> in kickstart one can use clearpart to remove old partition, but there > is >> no command to remove old lvm volgroup or logvols. and if there are >> volgroup or logvols already on the system where you want to install >> through kickstart the install process failed with a python stack trace >> since eg. VolGroup00 already exist and can't be created. >> currently there is no other way to delete only in %pre scripts but > imho >> it'd be an essential feature. >> yours. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Levente "Si vis pacem para bellum!" From Gerrard.Geldenhuis at datacash.com Wed Jan 16 14:02:39 2008 From: Gerrard.Geldenhuis at datacash.com (Gerrard Geldenhuis) Date: Wed, 16 Jan 2008 14:02:39 -0000 Subject: where should i have to use %end In-Reply-To: <478E08A3.1060805@bppiac.hu> References: <478E08A3.1060805@bppiac.hu> Message-ID: Hi I have checked my kickstart files on Satellite and they don't contain %end keywords. I doubt if it is needed, it might be a deprecated option... Regards > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Farkas Levente > Sent: 16 January 2008 13:38 > To: Discussion list about Kickstart > Subject: where should i have to use %end > > hi, > where should i have to use %end in kickstart file? after each section or > just pre and post or ...? > i can't find any documentation about it but python kickstart parser in > revisor complain about it that %end missing and this kickstart file will > be depricated. > thanks. From lfarkas at bppiac.hu Wed Jan 16 14:13:18 2008 From: lfarkas at bppiac.hu (Farkas Levente) Date: Wed, 16 Jan 2008 15:13:18 +0100 Subject: where should i have to use %end In-Reply-To: References: <478E08A3.1060805@bppiac.hu> Message-ID: <478E10FE.1020807@bppiac.hu> hmm: -------------------------------------------------- /usr/lib/python2.5/site-packages/pykickstart/parser.py:623: DeprecationWarning: Script does not end with %end. This syntax has been deprecated. It may be removed from future releases, which will result in a fatal error from kickstart. Please modify your kickstart file to use this updated syntax. warnings.warn(_("%s does not end with %%end. This syntax has been deprecated. It may be removed from future releases, which will result in a fatal error from kickstart. Please modify your kickstart file to use this updated syntax.") % _("Script"), DeprecationWarning) -------------------------------------------------- Gerrard Geldenhuis wrote: > Hi > I have checked my kickstart files on Satellite and they don't contain > %end keywords. I doubt if it is needed, it might be a deprecated > option... > > Regards > >> -----Original Message----- >> From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- >> bounces at redhat.com] On Behalf Of Farkas Levente >> Sent: 16 January 2008 13:38 >> To: Discussion list about Kickstart >> Subject: where should i have to use %end >> >> hi, >> where should i have to use %end in kickstart file? after each section > or >> just pre and post or ...? >> i can't find any documentation about it but python kickstart parser in >> revisor complain about it that %end missing and this kickstart file > will >> be depricated. >> thanks. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Levente "Si vis pacem para bellum!" From Pablo.Iranzo at redhat.com Wed Jan 16 14:26:50 2008 From: Pablo.Iranzo at redhat.com (Pablo Iranzo =?ISO-8859-1?Q?G=F3mez?=) Date: Wed, 16 Jan 2008 15:26:50 +0100 Subject: delete lvm groups in kickstart In-Reply-To: <478E0685.1070404@bppiac.hu> References: <478E0685.1070404@bppiac.hu> Message-ID: <1200493610.4319.30.camel@iranzo.users.redhat.com> I faced a similar problem, so what I do is to dd hardrive ;) It empties it so I can recreate pv,vg and lv's on it ;) Regards Pablo El mi?, 16-01-2008 a las 14:28 +0100, Farkas Levente escribi?: > hi, > in kickstart one can use clearpart to remove old partition, but there is > no command to remove old lvm volgroup or logvols. and if there are > volgroup or logvols already on the system where you want to install > through kickstart the install process failed with a python stack trace > since eg. VolGroup00 already exist and can't be created. > currently there is no other way to delete only in %pre scripts but imho > it'd be an essential feature. > yours. > -- Pablo Iranzo G?mez (Pablo.Iranzo at redhat.com) RHCE/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 lfarkas at bppiac.hu Wed Jan 16 14:30:20 2008 From: lfarkas at bppiac.hu (Farkas Levente) Date: Wed, 16 Jan 2008 15:30:20 +0100 Subject: delete lvm groups in kickstart In-Reply-To: <1200493610.4319.30.camel@iranzo.users.redhat.com> References: <478E0685.1070404@bppiac.hu> <1200493610.4319.30.camel@iranzo.users.redhat.com> Message-ID: <478E14FC.5020005@bppiac.hu> imho it's still better: %pre lvm lvremove VolGroup00 but it'd be a kickstart option... Pablo Iranzo G?mez wrote: > I faced a similar problem, so what I do is to dd hardrive ;) > > It empties it so I can recreate pv,vg and lv's on it ;) > > Regards > Pablo > > El mi?, 16-01-2008 a las 14:28 +0100, Farkas Levente escribi?: >> hi, >> in kickstart one can use clearpart to remove old partition, but there is >> no command to remove old lvm volgroup or logvols. and if there are >> volgroup or logvols already on the system where you want to install >> through kickstart the install process failed with a python stack trace >> since eg. VolGroup00 already exist and can't be created. >> currently there is no other way to delete only in %pre scripts but imho >> it'd be an essential feature. >> yours. >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list -- Levente "Si vis pacem para bellum!" From Pablo.Iranzo at redhat.com Wed Jan 16 14:35:09 2008 From: Pablo.Iranzo at redhat.com (Pablo Iranzo =?ISO-8859-1?Q?G=F3mez?=) Date: Wed, 16 Jan 2008 15:35:09 +0100 Subject: delete lvm groups in kickstart In-Reply-To: <478E14FC.5020005@bppiac.hu> References: <478E0685.1070404@bppiac.hu> <1200493610.4319.30.camel@iranzo.users.redhat.com> <478E14FC.5020005@bppiac.hu> Message-ID: <1200494109.4319.32.camel@iranzo.users.redhat.com> That makes you rely on knowing already existing vg names or autodetect them using another script ;) But definitely can be interesting to have that command for anaconda :) Regards Pablo El mi?, 16-01-2008 a las 15:30 +0100, Farkas Levente escribi?: > imho it's still better: > %pre > lvm lvremove VolGroup00 > but it'd be a kickstart option... > > Pablo Iranzo G?mez wrote: > > I faced a similar problem, so what I do is to dd hardrive ;) > > > > It empties it so I can recreate pv,vg and lv's on it ;) > > > > Regards > > Pablo > > > > El mi?, 16-01-2008 a las 14:28 +0100, Farkas Levente escribi?: > >> hi, > >> in kickstart one can use clearpart to remove old partition, but there is > >> no command to remove old lvm volgroup or logvols. and if there are > >> volgroup or logvols already on the system where you want to install > >> through kickstart the install process failed with a python stack trace > >> since eg. VolGroup00 already exist and can't be created. > >> currently there is no other way to delete only in %pre scripts but imho > >> it'd be an essential feature. > >> yours. > >> > >> > >> ------------------------------------------------------------------------ > >> > >> _______________________________________________ > >> 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/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 clumens at redhat.com Wed Jan 16 15:42:15 2008 From: clumens at redhat.com (Chris Lumens) Date: Wed, 16 Jan 2008 10:42:15 -0500 Subject: where should i have to use %end In-Reply-To: <478E08A3.1060805@bppiac.hu> References: <478E08A3.1060805@bppiac.hu> Message-ID: <20080116154215.GC31981@dhcp83-45.boston.redhat.com> > where should i have to use %end in kickstart file? after each section or > just pre and post or ...? > i can't find any documentation about it but python kickstart parser in > revisor complain about it that %end missing and this kickstart file will > be depricated. After any %pre/%post/%traceback scripts and after %packages. It's not yet required that you have them, and most likely won't be required until after RHEL6 so you have plenty of time to move to adding %end. - Chris From lfarkas at bppiac.hu Wed Jan 16 17:08:18 2008 From: lfarkas at bppiac.hu (Farkas Levente) Date: Wed, 16 Jan 2008 18:08:18 +0100 Subject: where should i have to use %end In-Reply-To: <20080116154215.GC31981@dhcp83-45.boston.redhat.com> References: <478E08A3.1060805@bppiac.hu> <20080116154215.GC31981@dhcp83-45.boston.redhat.com> Message-ID: <478E3A02.5030902@bppiac.hu> Chris Lumens wrote: >> where should i have to use %end in kickstart file? after each section or >> just pre and post or ...? >> i can't find any documentation about it but python kickstart parser in >> revisor complain about it that %end missing and this kickstart file will >> be depricated. > > After any %pre/%post/%traceback scripts and after %packages. what's traceback? > It's not yet required that you have them, and most likely won't be > required until after RHEL6 so you have plenty of time to move to adding > %end. > > - Chris > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Levente "Si vis pacem para bellum!" From mdehaan at redhat.com Wed Jan 16 17:14:34 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Wed, 16 Jan 2008 12:14:34 -0500 Subject: where should i have to use %end In-Reply-To: <20080116154215.GC31981@dhcp83-45.boston.redhat.com> References: <478E08A3.1060805@bppiac.hu> <20080116154215.GC31981@dhcp83-45.boston.redhat.com> Message-ID: <478E3B7A.9080407@redhat.com> Chris Lumens wrote: >> where should i have to use %end in kickstart file? after each section or >> just pre and post or ...? >> i can't find any documentation about it but python kickstart parser in >> revisor complain about it that %end missing and this kickstart file will >> be depricated. >> > > After any %pre/%post/%traceback scripts and after %packages. > > It's not yet required that you have them, and most likely won't be > required until after RHEL6 so you have plenty of time to move to adding > %end. > > - Chris > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > Does %end break any older distros if they are sharing common kickstart templates? --Michael From clumens at redhat.com Wed Jan 16 17:52:45 2008 From: clumens at redhat.com (Chris Lumens) Date: Wed, 16 Jan 2008 12:52:45 -0500 Subject: where should i have to use %end In-Reply-To: <478E3A02.5030902@bppiac.hu> References: <478E08A3.1060805@bppiac.hu> <20080116154215.GC31981@dhcp83-45.boston.redhat.com> <478E3A02.5030902@bppiac.hu> Message-ID: <20080116175245.GD31981@dhcp83-45.boston.redhat.com> > > After any %pre/%post/%traceback scripts and after %packages. > > what's traceback? Another kind of kickstart script that almost no one uses or knows about. It executes if a traceback occurs (only in stage2 - there's nothing we can do in the loader). - Chris From gary at mlbassoc.com Thu Jan 17 21:36:30 2008 From: gary at mlbassoc.com (Gary Thomas) Date: Thu, 17 Jan 2008 14:36:30 -0700 Subject: No selinux whatsoever Message-ID: <478FCA5E.3030007@mlbassoc.com> I'm trying to use anaconda+kickstart to load up a deeply embedded platform. This device will never need nor use selinux, so I want to figure out how to keep it from ever being installed, whatsoever. How do I make this happen in the kickstart file? Note: this is such a resource limited platform that simply installing the "selinux-policy-targetted" RPM takes around 5 hours! Hence my desire to never even try. -- ------------------------------------------------------------ Gary Thomas | Consulting for the MLB Associates | Embedded world ------------------------------------------------------------ From debian at herakles.homelinux.org Fri Jan 18 00:11:57 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Fri, 18 Jan 2008 09:11:57 +0900 Subject: No selinux whatsoever In-Reply-To: <478FCA5E.3030007@mlbassoc.com> References: <478FCA5E.3030007@mlbassoc.com> Message-ID: <478FEECD.7040207@herakles.homelinux.org> Gary Thomas wrote: > I'm trying to use anaconda+kickstart to load up a deeply > embedded platform. This device will never need nor use > selinux, so I want to figure out how to keep it from > ever being installed, whatsoever. You're using the wrong Linux distro; selinux is built into the kernel. [root at numbat ~]# grep SELINUX /boot/config-2.6.18-8.1.15.el5 | head -1 CONFIG_SECURITY_SELINUX=y [root at numbat ~]# > > How do I make this happen in the kickstart file? > > Note: this is such a resource limited platform that simply > installing the "selinux-policy-targetted" RPM takes around > 5 hours! Hence my desire to never even try. If you want prebuilt binaries I suggest Debian. Note that Debian's moving to apparmour, but I don't know what its implications are. -- 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 Pablo.Iranzo at redhat.com Fri Jan 18 00:59:02 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Fri, 18 Jan 2008 01:59:02 +0100 (CET) Subject: No selinux whatsoever In-Reply-To: <478FEECD.7040207@herakles.homelinux.org> References: <478FCA5E.3030007@mlbassoc.com> <478FEECD.7040207@herakles.homelinux.org> Message-ID: On Fri, 18 Jan 2008, John Summerfield wrote: > Gary Thomas wrote: > > I'm trying to use anaconda+kickstart to load up a deeply > > embedded platform. This device will never need nor use > > selinux, so I want to figure out how to keep it from > > ever being installed, whatsoever. > > You're using the wrong Linux distro; selinux is built into the kernel. > [root at numbat ~]# grep SELINUX /boot/config-2.6.18-8.1.15.el5 | head -1 > CONFIG_SECURITY_SELINUX=y > [root at numbat ~]# One thing is to get the support into kernel, and another one to have all userspace binaries, libraries and so installed > > How do I make this happen in the kickstart file? > > > > Note: this is such a resource limited platform that simply > > installing the "selinux-policy-targetted" RPM takes around > > 5 hours! Hence my desire to never even try. > What you need is a minimal installation, so two choices: "%packages --nobase --excludedocs" and then specify by hand what you really need or install on a big brother system, then remove everything not needed on your appliance and rsync fs to your device... install grub on storage device and try if it works or not ;) > If you want prebuilt binaries I suggest Debian. Note that Debian's > moving to apparmour, but I don't know what its implications are. Well, apparmour has raised several complaints about the approach to security, but in this case, the problem is about CPU muscle to install in an embedded device, not about SELinux nor AppArmor Regards Pablo From debian at herakles.homelinux.org Fri Jan 18 01:14:32 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Fri, 18 Jan 2008 10:14:32 +0900 Subject: No selinux whatsoever In-Reply-To: References: <478FCA5E.3030007@mlbassoc.com> <478FEECD.7040207@herakles.homelinux.org> Message-ID: <478FFD78.7020608@herakles.homelinux.org> Pablo Iranzo G?mez wrote: > One thing is to get the support into kernel, and another one to have all > userspace binaries, libraries and so installed Read the subject line, it says "No selinux whatsoever." -- 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 Pablo.Iranzo at redhat.com Fri Jan 18 01:17:20 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Fri, 18 Jan 2008 02:17:20 +0100 (CET) Subject: No selinux whatsoever In-Reply-To: <478FFD78.7020608@herakles.homelinux.org> References: <478FCA5E.3030007@mlbassoc.com> <478FEECD.7040207@herakles.homelinux.org> <478FFD78.7020608@herakles.homelinux.org> Message-ID: >From his note: "Note: this is such a resource limited platform that simply installing the "selinux-policy-targetted" RPM takes around 5 hours! Hence my desire to never even try." ;) -- 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? On Fri, 18 Jan 2008, John Summerfield wrote: > Pablo Iranzo G?mez wrote: > > > One thing is to get the support into kernel, and another one to have all > > userspace binaries, libraries and so installed > > Read the subject line, it says "No selinux whatsoever." > > -- > > Cheers > John > > -- spambait > 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu > -- Advice > http://webfoot.com/advice/email.top.php > http://www.catb.org/~esr/faqs/smart-questions.html > http://support.microsoft.com/kb/555375 > > You cannot reply off-list:-) > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From debian at herakles.homelinux.org Fri Jan 18 01:27:10 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Fri, 18 Jan 2008 10:27:10 +0900 Subject: No selinux whatsoever In-Reply-To: References: <478FCA5E.3030007@mlbassoc.com> <478FEECD.7040207@herakles.homelinux.org> <478FFD78.7020608@herakles.homelinux.org> Message-ID: <4790006E.6090607@herakles.homelinux.org> Pablo Iranzo G?mez wrote: >>From his note: > > "Note: this is such a resource limited platform that simply > installing the "selinux-policy-targetted" RPM takes around > 5 hours! Hence my desire to never even try." and? -- 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 Pablo.Iranzo at redhat.com Fri Jan 18 01:34:05 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Fri, 18 Jan 2008 02:34:05 +0100 (CET) Subject: No selinux whatsoever In-Reply-To: <4790006E.6090607@herakles.homelinux.org> References: <478FCA5E.3030007@mlbassoc.com> <478FEECD.7040207@herakles.homelinux.org> <478FFD78.7020608@herakles.homelinux.org> <4790006E.6090607@herakles.homelinux.org> Message-ID: On Fri, 18 Jan 2008, John Summerfield wrote: > Pablo Iranzo G??mez wrote: > >>From his note: > > > > "Note: this is such a resource limited platform that simply > > installing the "selinux-policy-targetted" RPM takes around > > 5 hours! Hence my desire to never even try." > > and? And... as he's talking about installation overhead, I'm not pretty sure about how many megabytes would save a kernel without selinux support compared with a one with it... His email was about anaconda+kickstart not about kernel optimization... Regards Pablo From mdehaan at redhat.com Fri Jan 18 16:19:24 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Fri, 18 Jan 2008 11:19:24 -0500 Subject: No selinux whatsoever In-Reply-To: <478FCA5E.3030007@mlbassoc.com> References: <478FCA5E.3030007@mlbassoc.com> Message-ID: <4790D18C.3060403@redhat.com> Gary Thomas wrote: > I'm trying to use anaconda+kickstart to load up a deeply > embedded platform. This device will never need nor use > selinux, so I want to figure out how to keep it from > ever being installed, whatsoever. > > How do I make this happen in the kickstart file? > > Note: this is such a resource limited platform that simply > installing the "selinux-policy-targetted" RPM takes around > 5 hours! Hence my desire to never even try. > Just add "selinux disabled" in your kickstart and it will not be enabled and will not be doing anything. There isn't a lot of overhead in terms of extra storage to worry about AFAIK. The policy shouldn't be being applied if don't turn selinux on (either in enforcing mode or permissive). I could be wrong about this however, have you tried disabling SELinux in your kickstart for starters? --Michael From gary at mlbassoc.com Fri Jan 18 16:43:25 2008 From: gary at mlbassoc.com (Gary Thomas) Date: Fri, 18 Jan 2008 09:43:25 -0700 Subject: No selinux whatsoever In-Reply-To: <4790D18C.3060403@redhat.com> References: <478FCA5E.3030007@mlbassoc.com> <4790D18C.3060403@redhat.com> Message-ID: <4790D72D.1010700@mlbassoc.com> Michael DeHaan wrote: > Gary Thomas wrote: >> I'm trying to use anaconda+kickstart to load up a deeply >> embedded platform. This device will never need nor use >> selinux, so I want to figure out how to keep it from >> ever being installed, whatsoever. >> >> How do I make this happen in the kickstart file? >> >> Note: this is such a resource limited platform that simply >> installing the "selinux-policy-targetted" RPM takes around >> 5 hours! Hence my desire to never even try. >> > > Just add "selinux disabled" in your kickstart and it will not be enabled > and will not be doing anything. > > There isn't a lot of overhead in terms of extra storage to worry about > AFAIK. > > The policy shouldn't be being applied if don't turn selinux on (either > in enforcing mode or permissive). I could be wrong about this however, > have you tried disabling SELinux in your kickstart for starters? Yes. I have "selinux --disabled" in my kickstart and I start anaconda with "selinux=0" (don't believe the documentation on this one - trust the code). It still loads the selinux packages and the loader/anaconda still tries to do stuff with selinux, e.g. from my install log: 12:38:33 WARNING : Failed to create /etc/selinux/config: Read-only file system 12:38:33 WARNING : Failed to create /etc/selinux/targeted/contexts/customizable_types: Read-only file system etc. BTW, my embedded kernel also is tuned for no selinux support, so even if the packages are installed, nothing happens (they just get in the way IMO) I have found that I can simply remove "selinux-policy-targetted" in my kickstart packages and this makes things much better. %packages @base -selinux-policy-targeted %end -- ------------------------------------------------------------ Gary Thomas | Consulting for the MLB Associates | Embedded world ------------------------------------------------------------ From Chip.Shabazian at bankofamerica.com Fri Jan 18 16:55:29 2008 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 18 Jan 2008 08:55:29 -0800 Subject: No selinux whatsoever In-Reply-To: <4790D72D.1010700@mlbassoc.com> Message-ID: You would be amazed at how much stuff you can remove from your system and still have it work. Here is what I did some time ago to get a RHEL 3 build to under 200 packages. First, build a system as minimal as you know you can. Next, run: rpm -qa --qf "%{NAME}\n" > rpms To get a listing of all the rpm's installed on your system. I suggest removing things from this list such as kernel, yum, ssh, and rpm itself if you want a usuable system after you run the next step. Once you have this "sanitized" list of rpm's, simply remove everything you can: for i in `cat rpms`; do rpm -e $i; if [ "$?" -eq 0 ] ; then echo -e "$i has been removed" >> removed_rpms.txt ; fi; done Keep running the above until nothing more is removed. Reboot the system and make sure it still works. If it does, and at a minimimum you can install new packages, you can take the list of rpm's in the removed_rpms.txt file and add them to your %packages section with - so they don't get installed on new builds. All of the above could be scripted to just keep running, but I only had to do it the one time so I didn't bother doing anything more with it. Chip -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Gary Thomas Sent: Friday, January 18, 2008 8:43 AM To: Discussion list about Kickstart Subject: Re: No selinux whatsoever Michael DeHaan wrote: > Gary Thomas wrote: >> I'm trying to use anaconda+kickstart to load up a deeply embedded >> platform. This device will never need nor use selinux, so I want to >> figure out how to keep it from ever being installed, whatsoever. >> >> How do I make this happen in the kickstart file? >> >> Note: this is such a resource limited platform that simply installing >> the "selinux-policy-targetted" RPM takes around >> 5 hours! Hence my desire to never even try. >> > > Just add "selinux disabled" in your kickstart and it will not be > enabled and will not be doing anything. > > There isn't a lot of overhead in terms of extra storage to worry about > AFAIK. > > The policy shouldn't be being applied if don't turn selinux on (either > in enforcing mode or permissive). I could be wrong about this however, > have you tried disabling SELinux in your kickstart for starters? Yes. I have "selinux --disabled" in my kickstart and I start anaconda with "selinux=0" (don't believe the documentation on this one - trust the code). It still loads the selinux packages and the loader/anaconda still tries to do stuff with selinux, e.g. from my install log: 12:38:33 WARNING : Failed to create /etc/selinux/config: Read-only file system 12:38:33 WARNING : Failed to create /etc/selinux/targeted/contexts/customizable_types: Read-only file system etc. BTW, my embedded kernel also is tuned for no selinux support, so even if the packages are installed, nothing happens (they just get in the way IMO) I have found that I can simply remove "selinux-policy-targetted" in my kickstart packages and this makes things much better. %packages @base -selinux-policy-targeted %end -- ------------------------------------------------------------ Gary Thomas | Consulting for the MLB Associates | Embedded world ------------------------------------------------------------ _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Pablo.Iranzo at redhat.com Fri Jan 18 17:29:42 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Fri, 18 Jan 2008 18:29:42 +0100 (CET) Subject: No selinux whatsoever In-Reply-To: <4790D72D.1010700@mlbassoc.com> References: <478FCA5E.3030007@mlbassoc.com> <4790D18C.3060403@redhat.com> <4790D72D.1010700@mlbassoc.com> Message-ID: As suggested, try %packages --nobase --excludedocs This will slim fit your deployment, but be sure to specify which packages you want to install on system... Regards Pablo -- 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?? On Fri, 18 Jan 2008, Gary Thomas wrote: > Michael DeHaan wrote: > > Gary Thomas wrote: > >> I'm trying to use anaconda+kickstart to load up a deeply > >> embedded platform. This device will never need nor use > >> selinux, so I want to figure out how to keep it from > >> ever being installed, whatsoever. > >> > >> How do I make this happen in the kickstart file? > >> > >> Note: this is such a resource limited platform that simply > >> installing the "selinux-policy-targetted" RPM takes around > >> 5 hours! Hence my desire to never even try. > >> > > > > Just add "selinux disabled" in your kickstart and it will not be enabled > > and will not be doing anything. > > > > There isn't a lot of overhead in terms of extra storage to worry about > > AFAIK. > > > > The policy shouldn't be being applied if don't turn selinux on (either > > in enforcing mode or permissive). I could be wrong about this however, > > have you tried disabling SELinux in your kickstart for starters? > > Yes. I have "selinux --disabled" in my kickstart and I start > anaconda with "selinux=0" (don't believe the documentation on > this one - trust the code). It still loads the selinux packages > and the loader/anaconda still tries to do stuff with selinux, e.g. > from my install log: > 12:38:33 WARNING : Failed to create /etc/selinux/config: Read-only file system > 12:38:33 WARNING : Failed to create /etc/selinux/targeted/contexts/customizable_types: Read-only > file system > etc. > > BTW, my embedded kernel also is tuned for no selinux support, > so even if the packages are installed, nothing happens (they > just get in the way IMO) > > I have found that I can simply remove "selinux-policy-targetted" > in my kickstart packages and this makes things much better. > > %packages > @base > -selinux-policy-targeted > %end > > -- > ------------------------------------------------------------ > Gary Thomas | Consulting for the > MLB Associates | Embedded world > ------------------------------------------------------------ > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From Darrick.Waller at baesystems.com Mon Jan 21 19:47:30 2008 From: Darrick.Waller at baesystems.com (Waller, Darrick J (US SSA)) Date: Mon, 21 Jan 2008 13:47:30 -0600 Subject: Pre-patched kickstart tree Message-ID: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> As the subject states, I am trying to create an kickstart tree that allows me to install a system with current packages, instead of older packages from the downloaded ISO's that I have. I have novi installed, and am able to merge the two RPMS directories that I have using hard links, but when I try and kickstart a machine using this directory as the install tree, I get the following error: "That directory does not seem to contain a Red Hat Enterprise Linux installation tree." I've tried linking the 'RedHat/base' directory, thinking I needed the stage2 stuff, but I still get the same error. Any advice on this, or alternate methods of doing the same thing would be great. -Darrick -------------- next part -------------- An HTML attachment was scrubbed... URL: From jkeating at j2solutions.net Mon Jan 21 19:55:48 2008 From: jkeating at j2solutions.net (Jesse Keating) Date: Mon, 21 Jan 2008 14:55:48 -0500 Subject: Pre-patched kickstart tree In-Reply-To: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> References: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> Message-ID: <20080121145548.3d5652f1@j2solutions.net> On Mon, 21 Jan 2008 13:47:30 -0600 "Waller, Darrick J (US SSA)" wrote: > As the subject states, I am trying to create an kickstart tree that > allows me to install a system with current packages, instead of older > packages from the downloaded ISO's that I have. I have novi installed, > and am able to merge the two RPMS directories that I have using hard > links, but when I try and kickstart a machine using this directory as > the install tree, I get the following error: > > > > "That directory does not seem to contain a Red Hat Enterprise Linux > installation tree." > > > > I've tried linking the 'RedHat/base' directory, thinking I needed the > stage2 stuff, but I still get the same error. Any advice on this, or > alternate methods of doing the same thing would be great. If this is RHEL5, why not just add a repository in your ks file that has all the updates in it? Far easier than trying to reconstruct the install tree. -- Jesse Keating RHCE (jkeating.livejournal.com) Fedora Project (fedoraproject.org/wiki/JesseKeating) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From Darrick.Waller at baesystems.com Mon Jan 21 20:42:12 2008 From: Darrick.Waller at baesystems.com (Waller, Darrick J (US SSA)) Date: Mon, 21 Jan 2008 14:42:12 -0600 Subject: Pre-patched kickstart tree References: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> <20080121145548.3d5652f1@j2solutions.net> Message-ID: <04799F26A23174449975FB1913E3A02F155D78@gldms20030.goldlnk.rootlnka.net> Sorry, I am trying to get this working with RHEL3. I will look into adding repositories to the kickstart file for RHEL5, that sounds really handy. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Jesse Keating Sent: Monday, January 21, 2008 1:56 PM To: kickstart-list at redhat.com Subject: Re: Pre-patched kickstart tree On Mon, 21 Jan 2008 13:47:30 -0600 "Waller, Darrick J (US SSA)" wrote: > As the subject states, I am trying to create an kickstart tree that > allows me to install a system with current packages, instead of older > packages from the downloaded ISO's that I have. I have novi installed, > and am able to merge the two RPMS directories that I have using hard > links, but when I try and kickstart a machine using this directory as > the install tree, I get the following error: > > > > "That directory does not seem to contain a Red Hat Enterprise Linux > installation tree." > > > > I've tried linking the 'RedHat/base' directory, thinking I needed the > stage2 stuff, but I still get the same error. Any advice on this, or > alternate methods of doing the same thing would be great. If this is RHEL5, why not just add a repository in your ks file that has all the updates in it? Far easier than trying to reconstruct the install tree. -- Jesse Keating RHCE (jkeating.livejournal.com) Fedora Project (fedoraproject.org/wiki/JesseKeating) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) From mdehaan at redhat.com Mon Jan 21 20:42:06 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 21 Jan 2008 15:42:06 -0500 Subject: Pre-patched kickstart tree In-Reply-To: <04799F26A23174449975FB1913E3A02F155D78@gldms20030.goldlnk.rootlnka.net> References: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> <20080121145548.3d5652f1@j2solutions.net> <04799F26A23174449975FB1913E3A02F155D78@gldms20030.goldlnk.rootlnka.net> Message-ID: <4795039E.9080007@redhat.com> Waller, Darrick J (US SSA) wrote: > Sorry, I am trying to get this working with RHEL3. > > I will look into adding repositories to the kickstart file for RHEL5, > that sounds really handy. > FYI ... if you are using Cobbler for net installs, it can help set all that up for you including the repo mirroring and so forth, and will keep track of what repos are needed by what profiles. cobbler import --name=rhel5 --mirror=/mnt/mymounteddvd cobbler repo add --name=epel5-x86 --mirror=http://download.fedora.redhat.com/pub/epel/5/i386/ cobbler repo add --name=mystuff --mirror=/path/to/my/extra/stuff cobbler repo sync cobbler profile edit --name=nameofprofile --repos="epel5-x86 mystuff" However I believe on IRC you said you are doing media based stuff so including the updates tricks really won't help you too much. If you do have a network available at install time though, I'd recommend using it, as the yum integration with kickstart (EL5 and later, Fedora 6 and later) is quite nice. > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Jesse Keating > Sent: Monday, January 21, 2008 1:56 PM > To: kickstart-list at redhat.com > Subject: Re: Pre-patched kickstart tree > > On Mon, 21 Jan 2008 13:47:30 -0600 > "Waller, Darrick J (US SSA)" wrote: > > >> As the subject states, I am trying to create an kickstart tree that >> allows me to install a system with current packages, instead of older >> packages from the downloaded ISO's that I have. I have novi installed, >> and am able to merge the two RPMS directories that I have using hard >> links, but when I try and kickstart a machine using this directory as >> the install tree, I get the following error: >> >> >> >> "That directory does not seem to contain a Red Hat Enterprise Linux >> installation tree." >> >> >> >> I've tried linking the 'RedHat/base' directory, thinking I needed the >> stage2 stuff, but I still get the same error. Any advice on this, or >> alternate methods of doing the same thing would be great. >> > > > If this is RHEL5, why not just add a repository in your ks file that > has all the updates in it? Far easier than trying to reconstruct the > install tree. > > From debian at herakles.homelinux.org Mon Jan 21 22:42:08 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Tue, 22 Jan 2008 07:42:08 +0900 Subject: Pre-patched kickstart tree In-Reply-To: <4795039E.9080007@redhat.com> References: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> <20080121145548.3d5652f1@j2solutions.net> <04799F26A23174449975FB1913E3A02F155D78@gldms20030.goldlnk.rootlnka.net> <4795039E.9080007@redhat.com> Message-ID: <47951FC0.8060507@herakles.homelinux.org> Michael DeHaan wrote: > Waller, Darrick J (US SSA) wrote: >> Sorry, I am trying to get this working with RHEL3. >> >> I will look into adding repositories to the kickstart file for RHEL5, >> that sounds really handy. >> > >>> I've tried linking the 'RedHat/base' directory, thinking I needed the >>> stage2 stuff, but I still get the same error. Any advice on this, or >>> alternate methods of doing the same thing would be great. >>> >> >> >> If this is RHEL5, why not just add a repository in your ks file that >> has all the updates in it? Far easier than trying to reconstruct the >> install tree. I don't know why folk are giving EL5/Fedora answers when it's clear OP isn't using one of those. See where he says "I've tried linking the 'RedHat/base'." Back on RHL7.x I used to ren a self-written script that updated the system after install. I don't know the procedure with official RHEL; on a clone I'd use a little sed in %pre to configure up2date/yum and upgrade, maybe off a local mirror but if not, then via a caching proxy. Not perfectly neat, but pretty easy to do and wouldn't add much to the install time. -- 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 mdehaan at redhat.com Mon Jan 21 22:38:29 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 21 Jan 2008 17:38:29 -0500 Subject: Pre-patched kickstart tree In-Reply-To: <47951FC0.8060507@herakles.homelinux.org> References: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> <20080121145548.3d5652f1@j2solutions.net> <04799F26A23174449975FB1913E3A02F155D78@gldms20030.goldlnk.rootlnka.net> <4795039E.9080007@redhat.com> <47951FC0.8060507@herakles.homelinux.org> Message-ID: <47951EE5.5030100@redhat.com> >>> >>> >>> If this is RHEL5, why not just add a repository in your ks file that >>> has all the updates in it? Far easier than trying to reconstruct the >>> install tree. > > > > I don't know why folk are giving EL5/Fedora answers when it's clear OP > isn't using one of those. See where he says "I've tried linking the > 'RedHat/base'." Because upgrading to something more current has a few advantages. Not just updates, but being able to install your own RPMs at kickstart time too. Of course, nothing is stopping %post or firstboot stuff for older distros. --Michael From debian at herakles.homelinux.org Mon Jan 21 22:50:24 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Tue, 22 Jan 2008 07:50:24 +0900 Subject: Pre-patched kickstart tree In-Reply-To: <47951EE5.5030100@redhat.com> References: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> <20080121145548.3d5652f1@j2solutions.net> <04799F26A23174449975FB1913E3A02F155D78@gldms20030.goldlnk.rootlnka.net> <4795039E.9080007@redhat.com> <47951FC0.8060507@herakles.homelinux.org> <47951EE5.5030100@redhat.com> Message-ID: <479521B0.8000508@herakles.homelinux.org> Michael DeHaan wrote: > >>>> >>>> >>>> If this is RHEL5, why not just add a repository in your ks file that >>>> has all the updates in it? Far easier than trying to reconstruct the >>>> install tree. >> >> >> >> I don't know why folk are giving EL5/Fedora answers when it's clear OP >> isn't using one of those. See where he says "I've tried linking the >> 'RedHat/base'." > > Because upgrading to something more current has a few advantages. Not > just updates, but being able to install your own RPMs at kickstart time > too. it might, but it's clear that's not what the OP was doing, and there are perfectly valid reasons for not upgrading. I think we all need to take care to read the question, ascertain the facts and then provide a considered reply. -- 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 mdehaan at redhat.com Mon Jan 21 23:45:47 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Mon, 21 Jan 2008 18:45:47 -0500 Subject: Pre-patched kickstart tree In-Reply-To: <479521B0.8000508@herakles.homelinux.org> References: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> <20080121145548.3d5652f1@j2solutions.net> <04799F26A23174449975FB1913E3A02F155D78@gldms20030.goldlnk.rootlnka.net> <4795039E.9080007@redhat.com> <47951FC0.8060507@herakles.homelinux.org> <47951EE5.5030100@redhat.com> <479521B0.8000508@herakles.homelinux.org> Message-ID: <47952EAB.4040104@redhat.com> John Summerfield wrote: > Michael DeHaan wrote: >> >>>>> >>>>> >>>>> If this is RHEL5, why not just add a repository in your ks file that >>>>> has all the updates in it? Far easier than trying to reconstruct the >>>>> install tree. >>> >>> >>> >>> I don't know why folk are giving EL5/Fedora answers when it's clear >>> OP isn't using one of those. See where he says "I've tried linking >>> the 'RedHat/base'." >> >> Because upgrading to something more current has a few advantages. >> Not just updates, but being able to install your own RPMs at >> kickstart time too. > > it might, but it's clear that's not what the OP was doing, and there > are perfectly valid reasons for not upgrading. > > I think we all need to take care to read the question, ascertain the > facts and then provide a considered reply. > > I was talking to Darrick on IRC earlier and had a few more facts than you did :) ( Anyhow, one last quick plug ... if you like, join #kickstart on irc.freenode.net ... nice channel, but would be nice if it grew some ) From mail-lists at karan.org Tue Jan 22 00:55:09 2008 From: mail-lists at karan.org (Karanbir Singh) Date: Tue, 22 Jan 2008 00:55:09 +0000 Subject: Pre-patched kickstart tree In-Reply-To: <47951EE5.5030100@redhat.com> References: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> <20080121145548.3d5652f1@j2solutions.net> <04799F26A23174449975FB1913E3A02F155D78@gldms20030.goldlnk.rootlnka.net> <4795039E.9080007@redhat.com> <47951FC0.8060507@herakles.homelinux.org> <47951EE5.5030100@redhat.com> Message-ID: <47953EED.4050303@karan.org> Michael DeHaan wrote: >> I don't know why folk are giving EL5/Fedora answers when it's clear OP >> isn't using one of those. See where he says "I've tried linking the >> 'RedHat/base'." > > Because upgrading to something more current has a few advantages. Not > just updates, but being able to install your own RPMs at kickstart time > too. I must have missed something, since the anaconda published in the RHEL5 source dir on ftp.redhat.com is incapable of doing completely updated installs. -- Karanbir Singh : http://www.karan.org/ : 2522219 at icq From debian at herakles.homelinux.org Tue Jan 22 00:52:43 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Tue, 22 Jan 2008 09:52:43 +0900 Subject: Pre-patched kickstart tree In-Reply-To: <47952EAB.4040104@redhat.com> References: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> <20080121145548.3d5652f1@j2solutions.net> <04799F26A23174449975FB1913E3A02F155D78@gldms20030.goldlnk.rootlnka.net> <4795039E.9080007@redhat.com> <47951FC0.8060507@herakles.homelinux.org> <47951EE5.5030100@redhat.com> <479521B0.8000508@herakles.homelinux.org> <47952EAB.4040104@redhat.com> Message-ID: <47953E5B.1000300@herakles.homelinux.org> Michael DeHaan wrote: > John Summerfield wrote: >> Michael DeHaan wrote: >> I think we all need to take care to read the question, ascertain the >> facts and then provide a considered reply. >> >> > I was talking to Darrick on IRC earlier and had a few more facts than > you did :) then you should have alluded to them. so others can see what you're talking about. > > ( Anyhow, one last quick plug ... if you like, join #kickstart on > irc.freenode.net ... nice channel, but would be nice if it grew some ) I spend way to much time on these lists. -- 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 mdehaan at redhat.com Tue Jan 22 14:53:44 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Tue, 22 Jan 2008 09:53:44 -0500 Subject: Pre-patched kickstart tree In-Reply-To: <47953EED.4050303@karan.org> References: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> <20080121145548.3d5652f1@j2solutions.net> <04799F26A23174449975FB1913E3A02F155D78@gldms20030.goldlnk.rootlnka.net> <4795039E.9080007@redhat.com> <47951FC0.8060507@herakles.homelinux.org> <47951EE5.5030100@redhat.com> <47953EED.4050303@karan.org> Message-ID: <47960378.6010405@redhat.com> Karanbir Singh wrote: > Michael DeHaan wrote: >>> I don't know why folk are giving EL5/Fedora answers when it's clear >>> OP isn't using one of those. See where he says "I've tried linking >>> the 'RedHat/base'." >> >> Because upgrading to something more current has a few advantages. >> Not just updates, but being able to install your own RPMs at >> kickstart time too. > > I must have missed something, since the anaconda published in the > RHEL5 source dir on ftp.redhat.com is incapable of doing completely > updated installs. > > Add a "repo" entry into your kickstart and it works. You can import/copy media from a DVD to use as your tree for RHEL5. From mail-lists at karan.org Tue Jan 22 16:01:58 2008 From: mail-lists at karan.org (Karanbir Singh) Date: Tue, 22 Jan 2008 16:01:58 +0000 Subject: Pre-patched kickstart tree In-Reply-To: <47960378.6010405@redhat.com> References: <04799F26A23174449975FB1913E3A02F155D77@gldms20030.goldlnk.rootlnka.net> <20080121145548.3d5652f1@j2solutions.net> <04799F26A23174449975FB1913E3A02F155D78@gldms20030.goldlnk.rootlnka.net> <4795039E.9080007@redhat.com> <47951FC0.8060507@herakles.homelinux.org> <47951EE5.5030100@redhat.com> <47953EED.4050303@karan.org> <47960378.6010405@redhat.com> Message-ID: <47961376.6040900@karan.org> Michael DeHaan wrote: > Add a "repo" entry into your kickstart and it works. > You can import/copy media from a DVD to use as your tree for RHEL5. > have you actually tried this ? -- Karanbir Singh : http://www.karan.org/ : 2522219 at icq From accarlson at gmail.com Tue Jan 22 19:04:36 2008 From: accarlson at gmail.com (Augusto Castelan Carlson) Date: Tue, 22 Jan 2008 17:04:36 -0200 Subject: append multiple ks.cfg and create the CD Message-ID: Hi! Looking at the docs I sow that I have to add ks.cfg to the isolinux dir to be able to use a kickstart file from the CD and how to build the ISO image. Searching the archives I also found how to add multiple files: https://www.redhat.com/archives/kickstart-list/2005-September/msg00076.html I appended to the CD the kickstart files and some additional RPMs that I would like to install using the post install section of kickstart. I have created the iso and successfully booted the CD. But it does not find the kickstart file that I pointed in isolinux.cfg, and if I try to install interactively, it says that the media is not a Fedora 8 CD. --isolinux.cfg for one of the kickstart files label testserver kernel vmlinuz append initrd=initrd.img ks=cdrom:/testserver.cfg --I'm creating the iso with: mkisofs -J -R -v -T -o ../cd.iso -b isolinux/isolinux.bin -c isolinux/boot.cat -no-emul-boot -boot-load-size 4 -boot-info-table . Any help? Thanks in advance! Regards, -- Augusto Castelan Carlson From Pablo.Iranzo at redhat.com Tue Jan 22 19:41:28 2008 From: Pablo.Iranzo at redhat.com (=?UTF-8?Q?Pablo_Iranzo_G=C3=B3mez?=) Date: Tue, 22 Jan 2008 20:41:28 +0100 (CET) Subject: append multiple ks.cfg and create the CD In-Reply-To: References: Message-ID: Did you took special care on keeping .discinfo file? on root of the CD/DVD tree? -- 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?? On Tue, 22 Jan 2008, Augusto Castelan Carlson wrote: > Hi! > > Looking at the docs I sow that I have to add ks.cfg to the isolinux > dir to be able to use a kickstart file from the CD and how to build > the ISO image. > > Searching the archives I also found how to add multiple files: > https://www.redhat.com/archives/kickstart-list/2005-September/msg00076.html > > I appended to the CD the kickstart files and some additional RPMs that > I would like to install using the post install section of kickstart. > > I have created the iso and successfully booted the CD. But it does not > find the kickstart file that I pointed in isolinux.cfg, and if I try > to install interactively, it says that the media is not a Fedora 8 CD. > > --isolinux.cfg for one of the kickstart files > label testserver > kernel vmlinuz > append initrd=initrd.img ks=cdrom:/testserver.cfg > > --I'm creating the iso with: > mkisofs -J -R -v -T -o ../cd.iso -b isolinux/isolinux.bin -c > isolinux/boot.cat -no-emul-boot -boot-load-size 4 -boot-info-table . > > Any help? > > Thanks in advance! > > Regards, > -- > Augusto Castelan Carlson > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From debian at herakles.homelinux.org Wed Jan 23 01:08:06 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Wed, 23 Jan 2008 10:08:06 +0900 Subject: append multiple ks.cfg and create the CD In-Reply-To: References: Message-ID: <47969376.10503@herakles.homelinux.org> Augusto Castelan Carlson wrote: > Hi! > > Looking at the docs I sow that I have to add ks.cfg to the isolinux > dir to be able to use a kickstart file from the CD and how to build > the ISO image. That's not quite true, the ks file's location can be specified at the boot menu > > Searching the archives I also found how to add multiple files: > https://www.redhat.com/archives/kickstart-list/2005-September/msg00076.html > > I appended to the CD the kickstart files and some additional RPMs that > I would like to install using the post install section of kickstart. > > I have created the iso and successfully booted the CD. But it does not > find the kickstart file that I pointed in isolinux.cfg, and if I try > to install interactively, it says that the media is not a Fedora 8 CD. > > --isolinux.cfg for one of the kickstart files > label testserver > kernel vmlinuz > append initrd=initrd.img ks=cdrom:/testserver.cfg > > --I'm creating the iso with: > mkisofs -J -R -v -T -o ../cd.iso -b isolinux/isolinux.bin -c > isolinux/boot.cat -no-emul-boot -boot-load-size 4 -boot-info-table . > > Any help? Go through the procedure again, and write down what you do at each step. If it still does not work, then describe the steps you too. I am not about to assume that any extenal document describes what you did; if the document is correct but it does not work for you, then then that doc does not help us diagnose your problem. fwiw I would likely do something like this: mount -o loop,ro .... edit ... (show results as you did) mkisofs umount mount ls -la /mnt/iso If you can test under xen or kvm, that saves burning CDs. Next best is CD-RW or DVD+-RW. -- 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 sreenivas.gattu at nsn.com Wed Jan 23 11:03:25 2008 From: sreenivas.gattu at nsn.com (Gattu, Sreenivas (NSN - IN/India - MiniMD)) Date: Wed, 23 Jan 2008 16:33:25 +0530 Subject: Clarificatiion on the RHL KickStart file... Message-ID: Hello, We have a kickstart file(ks.cfg) with the following line : #Disk partitioning information %include /tmp/part-include When we install RHL using the above bootable CD, we get an error like : "no such file /tmp/partinclude" found. Could you please clarify the following : - During kickstart installation, where is the /tmp directory refered to? - How can i include the part-include file in my linux custom CD? Regards, Srinu -------------- next part -------------- An HTML attachment was scrubbed... URL: From Pablo.Iranzo at redhat.com Wed Jan 23 11:10:30 2008 From: Pablo.Iranzo at redhat.com (Pablo Iranzo =?ISO-8859-1?Q?G=F3mez?=) Date: Wed, 23 Jan 2008 12:10:30 +0100 Subject: Clarificatiion on the RHL KickStart file... In-Reply-To: References: Message-ID: <1201086630.15441.111.camel@iranzo.users.redhat.com> Are you creating it during %pre ? El mi?, 23-01-2008 a las 16:33 +0530, Gattu, Sreenivas (NSN - IN/India - MiniMD) escribi?: > Hello, > > We have a kickstart file(ks.cfg) with the following line : > > #Disk partitioning information > %include /tmp/part-include > > When we install RHL using the above bootable CD, we get an error > like : "no such file /tmp/partinclude" found. > > Could you please clarify the following : > > - During kickstart installation, where is the /tmp directory refered > to? > - How can i include the part-include file in my linux custom CD? > > Regards, > Srinu > > _______________________________________________ > 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/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 accarlson at gmail.com Wed Jan 23 11:21:40 2008 From: accarlson at gmail.com (Augusto Castelan Carlson) Date: Wed, 23 Jan 2008 09:21:40 -0200 Subject: append multiple ks.cfg and create the CD In-Reply-To: References: Message-ID: Thanks. Resolved with .discinfo and add the ks files in the cdrom root. Regards, Augusto On 1/22/08, Pablo Iranzo G?mez wrote: > > Did you took special care on keeping .discinfo file? on root of > the CD/DVD tree? > From Pablo.Iranzo at redhat.com Wed Jan 23 11:28:44 2008 From: Pablo.Iranzo at redhat.com (Pablo Iranzo =?ISO-8859-1?Q?G=F3mez?=) Date: Wed, 23 Jan 2008 12:28:44 +0100 Subject: append multiple ks.cfg and create the CD In-Reply-To: References: Message-ID: <1201087724.21443.2.camel@iranzo.users.redhat.com> Perfect! :) El mi?, 23-01-2008 a las 09:21 -0200, Augusto Castelan Carlson escribi?: > Thanks. > > Resolved with .discinfo and add the ks files in the cdrom root. > > Regards, > > Augusto > > On 1/22/08, Pablo Iranzo G?mez wrote: > > > > Did you took special care on keeping .discinfo file? on root of > > the CD/DVD tree? > > > > _______________________________________________ > 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/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 debian at herakles.homelinux.org Wed Jan 23 12:20:54 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Wed, 23 Jan 2008 21:20:54 +0900 Subject: Clarificatiion on the RHL KickStart file... In-Reply-To: References: Message-ID: <47973126.4050702@herakles.homelinux.org> Gattu, Sreenivas (NSN - IN/India - MiniMD) wrote: > Hello, > > We have a kickstart file(ks.cfg) with the following line : > > #Disk partitioning information > %include /tmp/part-include > > When we install RHL using the above bootable CD, we get an error like : > "no such file /tmp/partinclude" found. > > Could you please clarify the following : > > - During kickstart installation, where is the /tmp directory refered to? > - How can i include the part-include file in my linux custom CD? Just to be sure, what release of Red Hat Linux are you installing? -- 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 huffcslists at gmail.com Fri Jan 25 15:22:34 2008 From: huffcslists at gmail.com (Craig Huff) Date: Fri, 25 Jan 2008 09:22:34 -0600 Subject: Need help to install packages from external repository during CentOS install Message-ID: <8f2ba1e90801250722g1036d6f2te296b350c774eb24@mail.gmail.com> About a week ago it was suggested that I use a kickstart file to install packages from an external repository (ATrpms) during CentOS 5.1 installs, so I have been giving it a go, unsuccessfully. I finally got enough information from the failures to figure out the problem, but I don't know what the solution is. It appears that despite having this: network --device=eth0 --bootproto=dhcp in the command section, the network is not working during %post execution and probably not during the system installation phase, either. I have tried installing nvidia-graphics169.07 (and the related pieces) in several ways, none of which worked. The first approach was to add this: repo --name=atrpms --baseurl=http://129.102.1.25/pub/atrpms/el5-i386/atrpms/stable in the command section and this: nvidia-graphics169.07 in the %packages section, but this failed. Note that the IP was used after several attempts to use dl.atrpms.net failed and I then discovered in the Installation Manual that, since I was using DHCP, name resolution was not working during the installation (or %post phase, for that matter). The second approach was to keep the repo --name... statement in the command section and use this for a (simplified) %post script: %post ( #Note: using subshell for logging, but it never works. yum -y --disablerepo=extras install nvidia-graphics169.07 ) 1>/root/Anaconda_post.log 2>&1 Finally, I dropped the repo --name... statement from the command section and tried copying from the kickstart medium an atrpms.repo file into /etc/yum.repos.d that looked like this: [atrpms] name=ATrpms for RHEL and CentOS $releasever - $basearch baseurl=http://129.102.1.25/pub/atrpms/el$releasever-$basearch/atrpms/stable enabled=1 gpgcheck=0 however, even this failed. Then, using the screen accessed with (I think) ctrl-alt-F2 which offered a shell, I tried it manually and discovered that the networking wasn't configured and (once I su'd to root), ifconfig -a reported the devices had no IP addresses. So, what do I do differently to get a kickstart file that will successfully install RPMs from ATrpms? I *could* install with a static IP, which may fix this problem, but my understanding is that the installed system would *also* be configured with the static IP, which is not what I want. Any help appreciated. Craig. From rainer at ultra-secure.de Fri Jan 25 15:28:41 2008 From: rainer at ultra-secure.de (Rainer Duffner) Date: Fri, 25 Jan 2008 16:28:41 +0100 Subject: Adding groups at install-time? Message-ID: <479A0029.8000609@ultra-secure.de> Hi, via the "user" directive, it's possible to add users. But it doesn't seem to be possible to move the user in a to-be-created dedicated group, because there's no way to create that group and it's not done automatically. Is this only possible via post-install? cheers, Raine From debian at herakles.homelinux.org Sat Jan 26 00:34:34 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Sat, 26 Jan 2008 09:34:34 +0900 Subject: Adding groups at install-time? In-Reply-To: <479A0029.8000609@ultra-secure.de> References: <479A0029.8000609@ultra-secure.de> Message-ID: <479A801A.20104@herakles.homelinux.org> Rainer Duffner wrote: > Hi, > > via the "user" directive, it's possible to add users. > But it doesn't seem to be possible to move the user in a to-be-created > dedicated group, because there's no way to create that group and it's > not done automatically. > Is this only possible via post-install? %post { groupadd useradd } 'til done -- 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 Axel.Thimm at ATrpms.net Sat Jan 26 07:54:08 2008 From: Axel.Thimm at ATrpms.net (Axel Thimm) Date: Sat, 26 Jan 2008 09:54:08 +0200 Subject: Need help to install packages from external repository during CentOS install In-Reply-To: <8f2ba1e90801250722g1036d6f2te296b350c774eb24@mail.gmail.com> References: <8f2ba1e90801250722g1036d6f2te296b350c774eb24@mail.gmail.com> Message-ID: <20080126075408.GD7607@puariko.nirvana> Hi Craig, On Fri, Jan 25, 2008 at 09:22:34AM -0600, Craig Huff wrote: > About a week ago it was suggested that I use a kickstart file to > install packages from an external repository (ATrpms) during CentOS > 5.1 installs, so I have been giving it a go, unsuccessfully. I > finally got enough information from the failures to figure out the > problem, but I don't know what the solution is. It appears that > despite having this: > > network --device=eth0 --bootproto=dhcp > > in the command section, the network is not working during %post > execution and probably not during the system installation phase, > either. I have tried installing nvidia-graphics169.07 (and the > related pieces) in several ways, none of which worked. > %post > ( #Note: using subshell for logging, but it never works. > yum -y --disablerepo=extras install nvidia-graphics169.07 > ) 1>/root/Anaconda_post.log 2>&1 This is what I use for DHCP/PXE kickstart files that are compatible from FC5-F9 and RHEL4-5: | %post --interpreter /bin/sh | #! /bin/sh | | exec > /root/post-install.log 2>&1 | set -vx | | source /etc/sysconfig/network | hostname ${HOSTNAME} | hostname=`hostname -s` | installdir=/srv/domain.tld/install | installtree=$installdir/installtrees/$hostname | logdir=$installdir/logs/${hostname} | | mkdir -p /dev/shm; mount -t tmpfs none /dev/shm | mkdir -p $installdir | mount install.domain.tld:$installdir $installdir | | mkdir -p $logdir | if test -e $installtree/hostscript.sh ; then | $installtree/hostscript.sh install > $logdir/install-`date +%Y%m%d-%H%M%S`.log 2>&1 < /dev/null | fi | mv -f /root/anaconda-ks.cfg $logdir/ | rm -f /srv/domain.tld/install/tftpboot/pxelinux.cfg/AAAAAAAA | | umount $installdir | umount /dev/shm; rmdir /dev/shm It's a bit obfuscated to protect the customers' data (e.g. the domain.tld and AAAAAAAA bits). But it basically boils down to adding as much as is needed for proper netwroking support, nfs-mounting a central folder and calling a postinstall script there. In this script I copy over the *.repo files I need and call yum (actually that's a lie I'm copying channels and use smart, but it will work with yum as well). Using the above method gives you at the very least a cheap way to centrally monitor your installs. There is also cobbler, a project that does this in a more professional way: http://cobbler.et.redhat.com/ -- Axel.Thimm at ATrpms.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From huffcslists at gmail.com Mon Jan 28 00:25:30 2008 From: huffcslists at gmail.com (Craig Huff) Date: Sun, 27 Jan 2008 18:25:30 -0600 Subject: Need help to install packages from external repository during CentOS install In-Reply-To: <20080126075408.GD7607@puariko.nirvana> References: <8f2ba1e90801250722g1036d6f2te296b350c774eb24@mail.gmail.com> <20080126075408.GD7607@puariko.nirvana> Message-ID: <8f2ba1e90801271625m1f4f588bkb49e3d3729b2d7ac@mail.gmail.com> On Jan 26, 2008 1:54 AM, Axel Thimm wrote: > Hi Craig, > > This is what I use for DHCP/PXE kickstart files that are compatible > from FC5-F9 and RHEL4-5: > > | %post --interpreter /bin/sh > | #! /bin/sh > | > | exec > /root/post-install.log 2>&1 > | set -vx > | > | > FWIW, I finally found the roots of my problem getting a %post script to work on a CentOS 5.1 install. They were two fold: First, I made the mistake of editing my ks.cfg file on a windoze pc at one point which inserted cr/lf pairs at the ends of all the lines. It seems that this was okay in the command section and the %packages section, but the shell interpreting the %post script went nuts, thinking they were text tokens and complaining about them not being valid parameters to all the commands (as far as it went before aborting). To make matters worse, I wasn't seeing them when I edited with vi on my linux boxen because apparently the "enhanced with vim" vi editors automatically made allowances and treated it as a "dosfile" so it wasn't until it registered that ^M wasn't a linefeed character but a carriage return that I figured out what was amiss. Changed the file type in a vi edit session on the file and all was fixed automagically. Finally I was getting a log file and able to see what else was amiss. Then I had to deal with the second problem -- network access problems. It took me awhile to determine that while the documentation implied that networking would be available if I specified the networking parameters even for a CD based install, my DHCP based install would not be able to resolve names to IP addresses. I tried using nslookup to get IP addresses for the repositories and other links, but even when I got ones that would work, that didn't solve the problem. A call to ifconfig in my %post script told the sad story. No IP, no networking. Period. Axel's example (excerpted above) got me going and eventually I just ran "/etc/init.d/network start" at the start of my %post script. Once I got there, I had full network support, including name resolution, so back to name references in the network calls for repositories and wgets of GPG keys, etc. BTW, I am using a USB thumb drive since there's no spare disk or floppy drive to use for the kickstart file and I find that I have to boot the system into a linux session and reboot with "shutdown -r now" or else the install CD environment won't see the USB drive. Even when it does, it reports that it couldn't find the kickstart file and I have to hit return to say I *really* meant the path I gave and again to say "go look again" before it will find the kickstart file and take off. Hope this helps someone else spend less time scratching their head and more time making progress. Craig. From debian at herakles.homelinux.org Tue Jan 29 08:25:46 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Tue, 29 Jan 2008 17:25:46 +0900 Subject: kickstart proxy Message-ID: <479EE30A.8080705@herakles.homelinux.org> I'm sure that last time I did a manual install using Anaconda, that I was invited to specify information about a proxy server. Just for once, I want to do a ks install using a proxy, but I can't see from http://www.redhat.com/docs/manuals/enterprise/RHEL-5-manual/Installation_Guide-en-US/s1-kickstart2-options.html how to do it. The RHL 9 docs, which google found first, don't say either. How is it done? -- 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 jgranado at redhat.com Tue Jan 29 08:40:34 2008 From: jgranado at redhat.com (Joel Andres Granados) Date: Tue, 29 Jan 2008 09:40:34 +0100 Subject: kickstart proxy In-Reply-To: <479EE30A.8080705@herakles.homelinux.org> References: <479EE30A.8080705@herakles.homelinux.org> Message-ID: <479EE682.9090309@redhat.com> John Summerfield wrote: > I'm sure that last time I did a manual install using Anaconda, that I > was invited to specify information about a proxy server. > > Just for once, I want to do a ks install using a proxy, but I can't see > from > http://www.redhat.com/docs/manuals/enterprise/RHEL-5-manual/Installation_Guide-en-US/s1-kickstart2-options.html > how to do it. > > The RHL 9 docs, which google found first, don't say either. > > How is it done? > > > Had a quick look at the code and I would dare to say that it is only implemented in the interactive installs. I don't see it in pykickstart and dont see it in the args that loader receives. not 100% sure though, only 99% :) Regards -- Joel Andres Granados Red Hat / Brno, Czech Republic From debian at herakles.homelinux.org Tue Jan 29 11:44:42 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Tue, 29 Jan 2008 20:44:42 +0900 Subject: kickstart proxy In-Reply-To: <479EE682.9090309@redhat.com> References: <479EE30A.8080705@herakles.homelinux.org> <479EE682.9090309@redhat.com> Message-ID: <479F11AA.7080709@herakles.homelinux.org> Joel Andres Granados wrote: > John Summerfield wrote: >> I'm sure that last time I did a manual install using Anaconda, that I >> was invited to specify information about a proxy server. >> >> Just for once, I want to do a ks install using a proxy, but I can't >> see from >> http://www.redhat.com/docs/manuals/enterprise/RHEL-5-manual/Installation_Guide-en-US/s1-kickstart2-options.html >> how to do it. >> >> The RHL 9 docs, which google found first, don't say either. >> >> How is it done? >> >> >> > > Had a quick look at the code and I would dare to say that it is only > implemented > in the interactive installs. I don't see it in pykickstart and dont see > it in > the args that loader receives. not 100% sure though, only 99% :) grumble grumble. That's what I thought from grepping and groping around. Do you need a bz entry? I'm going to resurrect a trick I employed for Taroon beta; it's documented in the list's archives. -- 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 kajtzu at basen.net Tue Jan 29 16:04:32 2008 From: kajtzu at basen.net (Kaj Niemi) Date: Tue, 29 Jan 2008 18:04:32 +0200 Subject: kickstart proxy In-Reply-To: <479F11AA.7080709@herakles.homelinux.org> References: <479EE30A.8080705@herakles.homelinux.org> <479EE682.9090309@redhat.com> <479F11AA.7080709@herakles.homelinux.org> Message-ID: Hi, On Jan 29, 2008, at 13:44, John Summerfield wrote: >> Had a quick look at the code and I would dare to say that it is >> only implemented >> in the interactive installs. I don't see it in pykickstart and >> dont see it in >> the args that loader receives. not 100% sure though, only 99% :) > > I'm going to resurrect a trick I employed for Taroon beta; it's > documented in the list's archives. I've used squid in reverse proxy mode. It works great if you're able to set the url in the kickstart file somehow. HTH Kaj -- Kaj J. Niemi +358 45 63 12000 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3811 bytes Desc: not available URL: From arifsaha at yahoo.com Wed Jan 30 01:30:04 2008 From: arifsaha at yahoo.com (S P Arif Sahari Wibowo) Date: Tue, 29 Jan 2008 20:30:04 -0500 (EST) Subject: Kickstart Installation using Existing LVM Logical Volume Message-ID: Hi! I hope this issue is not Fedora specific, so I posting the question here. I am working on a Fedora 8 kickstart installation that should use existing LVM logical volume as home partition. However, it keep failing with an error like this: > Could not allocate requested partitions: > > Adding this partition would not leave enough disk space > for already allocated logical volumes in 2008AF_data.. > > Press 'OK' to exit the installer. * 2008AF_data is the name of volume group The kickstart code dealing with partition is quoted below: clearpart --all --drives=sda part /boot --fstype ext3 --size=128 part pv.1 --size=1536 --grow --ondisk=sda volgroup 2008A6_work pv.1 logvol swap --fstype swap --name=2008A6_swap --vgname=2008A6_work --size=512 logvol /var --fstype ext3 --name=2008A6_var --vgname=2008A6_work --size=512 logvol / --fstype ext3 --name=2008A6_root --vgname=2008A6_work --size=3904 part pv.2 --noformat --onpart sdb1 volgroup 2008AF_data pv.2 --noformat --useexisting logvol /home --noformat --useexisting --name=2008AF_home --vgname=2008AF_data --fstype ext3 * 2008AF_home is the name of logical volume to be reused I already tried without --useexisting, with estimated size (not accurate, but I added --grow), without --fstype, etc. same error. Any idea what's wrong? Thanks! -- (stephan paul) Arif Sahari Wibowo _____ _____ _____ _____ /____ /____/ /____/ /____ _____/ / / / _____/ http://www.arifsaha.com/ From accarlson at gmail.com Wed Jan 30 13:37:41 2008 From: accarlson at gmail.com (Augusto Castelan Carlson) Date: Wed, 30 Jan 2008 11:37:41 -0200 Subject: Guest install without DHCP Message-ID: Hello! My Dom0 has the ip address 10.1.121.201/24 and is not connected to the network, because I do not want to be able to reach DHCP server as I will not have one in the production network. When I try to install a guest with kickstart, it tries to reach a DHCP server. Off course the dhcp timed out. I set the ip address 10.1.121.210/24 manually and I can see these messages in the console: DHCPv4 eth0 - TIMED OUT ethe0 is not a wireless adapter file location:fpt://10.1.121.201/guest.cfg trasnferring fpt://10.1.121.201/guest.cfg to a fd setting up kickstart results of url ks, url ftp:/10.1.121.201// setting language to en_US.UTF-8 need to set up networking going to pick interface going to do getNetConfig dnsservers is 10.1.3.1 ethe0 is not a wireless adapter It starts DHCPDISCOVER again. I can ping the muanually configured IP. Kickstart is configured to setup an static ip address to the guest, 10.1.121.202/24. network --device eth0 --bootproto static --ip 10.1.121.202 --netmask 255.255.255.0 --gateway 10.1.121.254 --nameserver 10.1.3.1 Any suggestion? Thanks in advance. Regards, -- Augusto Castelan Carlson From debian at herakles.homelinux.org Tue Jan 29 23:12:17 2008 From: debian at herakles.homelinux.org (John Summerfield) Date: Wed, 30 Jan 2008 08:12:17 +0900 Subject: kickstart proxy In-Reply-To: References: <479EE30A.8080705@herakles.homelinux.org> <479EE682.9090309@redhat.com> <479F11AA.7080709@herakles.homelinux.org> Message-ID: <479FB2D1.8030402@herakles.homelinux.org> Kaj Niemi wrote: > Hi, > > On Jan 29, 2008, at 13:44, John Summerfield wrote: > >>> Had a quick look at the code and I would dare to say that it is only >>> implemented >>> in the interactive installs. I don't see it in pykickstart and dont >>> see it in >>> the args that loader receives. not 100% sure though, only 99% :) >> >> I'm going to resurrect a trick I employed for Taroon beta; it's >> documented in the list's archives. > > > I've used squid in reverse proxy mode. It works great if you're able to > set the url in the kickstart file somehow. I set a DNAT rule on an intermediate box. Ordinarily I install from a local system on my LAN, so I've never needed a proxy. This time, the target system's on the wrong side of a slow wireless link and the Internet at large is faster. However, I wanted to cache for reruns. I can now report the proxy settings don't come up with an interactive kickstart. With taroon, the problem was a little different, Anaconda was using a duff host: header, so I had to interpose a local Apache to fix it. I don't recall a problem with proxy then, but perhaps I was doing an interactive not-ks install off the RH servers, or perhaps I was using transparent proxy. -- 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 Gerrard.Geldenhuis at datacash.com Wed Jan 30 14:28:29 2008 From: Gerrard.Geldenhuis at datacash.com (Gerrard Geldenhuis) Date: Wed, 30 Jan 2008 14:28:29 -0000 Subject: hostname lookup ... Message-ID: Hi I have written a script but it does not work as expected. The script is a post script with nochroot. I tested the dnsdomainname command in the console after my script failed and saw that it returns localdomain. Hostname and uname returns localhost.localdomain. We want the time servers to be different depending on in which DC the server is build... Can anyone suggest a better method of doing what I am trying to achieve here? The script: # Kickstart PRE Script # Set time to ensure SSL RHN Activation can complete successfully echo "starting time sync" date /mnt/sysimage/usr/sbin/ntpdate -b ntp.$(/bin/dnsdomainname) /mnt/sysimage/sbin/hwclock --systohc date echo "time sync complete" #backup old file cp /mnt/sysimage/etc/ntp.conf /mnt/sysimage/tmp/ntp.conf.old #create new config file cat > /tmp/sysimage/etc/ntp.conf <<'EOF' # /etc/ntp.conf, configuration for ntpd # ntpd will use syslog() if logfile is not defined #logfile /var/log/ntpd driftfile /var/lib/ntp/ntp.drift statsdir /var/log/ntpstats/ statistics loopstats peerstats clockstats filegen loopstats file loopstats type day enable filegen peerstats file peerstats type day enable filegen clockstats file clockstats type day enable EOF echo "server ntp.$(/mnt/sysimage/bin/dnsdomainname)" >> /tmp/ntp.conf echo "server ntp2.$(/mnt/sysimage/bin/dnsdomainname)" >> /tmp/ntp.conf Regards -------------- next part -------------- An HTML attachment was scrubbed... URL: From mats.o.karlsson at gmail.com Wed Jan 30 14:41:37 2008 From: mats.o.karlsson at gmail.com (Mats Karlsson) Date: Wed, 30 Jan 2008 15:41:37 +0100 Subject: hostname lookup ... In-Reply-To: References: Message-ID: <32d5c55a0801300641i2ad9a0c1h8f9caa43deb79f7@mail.gmail.com> On Jan 30, 2008 3:28 PM, Gerrard Geldenhuis wrote: > Hi > > I have written a script but it does not work as expected. > > The script is a post script with nochroot. > > I tested the dnsdomainname command in the console after my script failed > and saw that it returns localdomain. > > Hostname and uname returns localhost.localdomain. > > We want the time servers to be different depending on in which DC the > server is build? > > Can anyone suggest a better method of doing what I am trying to achieve > here? > > The script: > > # Kickstart PRE Script > > # Set time to ensure SSL RHN Activation can complete successfully > > echo "starting time sync" > > date > > /mnt/sysimage/usr/sbin/ntpdate -b ntp.$(/bin/dnsdomainname) > > /mnt/sysimage/sbin/hwclock --systohc > > date > > echo "time sync complete" > > > #backup old file > > cp /mnt/sysimage/etc/ntp.conf /mnt/sysimage/tmp/ntp.conf.old > > > > #create new config file > > cat > /tmp/sysimage/etc/ntp.conf <<'EOF' > > # /etc/ntp.conf, configuration for ntpd > > # ntpd will use syslog() if logfile is not defined #logfile /var/log/ntpd > > > > driftfile /var/lib/ntp/ntp.drift > > statsdir /var/log/ntpstats/ > > > > statistics loopstats peerstats clockstats filegen loopstats file loopstats > type day enable filegen peerstats file peerstats type day enable filegen > clockstats file clockstats type day enable > > EOF > > > > echo "server ntp.$(/mnt/sysimage/bin/dnsdomainname)" >> /tmp/ntp.conf > > echo "server ntp2.$(/mnt/sysimage/bin/dnsdomainname)" >> /tmp/ntp.conf > > > > Regards > Have you tried to start the network service ? service network start /Mats -------------- next part -------------- An HTML attachment was scrubbed... URL: From email at jasonkohles.com Wed Jan 30 16:10:39 2008 From: email at jasonkohles.com (Jason Kohles) Date: Wed, 30 Jan 2008 11:10:39 -0500 Subject: hostname lookup ... In-Reply-To: References: Message-ID: On Jan 30, 2008, at 9:28 AM, Gerrard Geldenhuis wrote: > Hi > > I have written a script but it does not work as expected. > > The script is a post script with nochroot. > > I tested the dnsdomainname command in the console after my script > failed and saw that it returns localdomain. > > Hostname and uname returns localhost.localdomain. > > We want the time servers to be different depending on in which DC > the server is build? > > Can anyone suggest a better method of doing what I am trying to > achieve here? > dnsdomainname is going to use /etc/hosts to determine what the domain name is, since you are running outside the chroot, it's going to look at the hosts file that is built into the installer root image, not the one that actually contains your hostname. You aren't doing anything here that requires you to be outside the chroot, so your best option is probably just to remove the --nochroot and remove /mnt/sysimage from all the paths. If you have things that do need to be done outside the chroot, you can either split them into two separate %post sections, or run dnsdomainname inside the chroot... %post --nochroot DOMAIN=$( chroot /mnt/sysimage /bin/dnsdomainname ) echo "starting time sync" The other option is not to use dnsdomainname at all, and just source / etc/sysconfig/network and extract the domainname from the hostname you find there... %post --nochroot . /mnt/sysimage/etc/sysconfig/network # to set $HOSTNAME export DOMAIN=${HOSTNAME#*.} -- Jason Kohles, RHCA RHCDS RHCE email at jasonkohles.com - http://www.jasonkohles.com/ "A witty saying proves nothing." -- Voltaire -------------- next part -------------- An HTML attachment was scrubbed... URL: From mdehaan at redhat.com Thu Jan 31 15:35:04 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 31 Jan 2008 10:35:04 -0500 Subject: Guest install without DHCP In-Reply-To: References: Message-ID: <47A1EAA8.2030303@redhat.com> Augusto Castelan Carlson wrote: > Hello! > > My Dom0 has the ip address 10.1.121.201/24 and is not connected to the > network, because I do not want to be able to reach DHCP server as I > will not have one in the production network. > When I try to install a guest with kickstart, it tries to reach a DHCP > server. Off course the dhcp timed out. I set the ip address > 10.1.121.210/24 manually and I can see these messages in the console: > > DHCPv4 eth0 - TIMED OUT > ethe0 is not a wireless adapter > file location:fpt://10.1.121.201/guest.cfg > trasnferring fpt://10.1.121.201/guest.cfg to a fd > setting up kickstart > results of url ks, url ftp:/10.1.121.201// > setting language to en_US.UTF-8 > need to set up networking > going to pick interface > going to do getNetConfig > dnsservers is 10.1.3.1 > ethe0 is not a wireless adapter > > It starts DHCPDISCOVER again. I can ping the muanually configured IP. > > Kickstart is configured to setup an static ip address to the guest, > 10.1.121.202/24. > > network --device eth0 --bootproto static --ip 10.1.121.202 --netmask > 255.255.255.0 --gateway 10.1.121.254 --nameserver 10.1.3.1 > > Any suggestion? > > Thanks in advance. > > Regards, > While kickstart is configured for no DHCP, you also need to indicate your network settings in the kernel arguments to prevent it from DHCP'ing earlier to fetch the kickstart. From mdehaan at redhat.com Thu Jan 31 15:37:40 2008 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 31 Jan 2008 10:37:40 -0500 Subject: Kickstart Installation using Existing LVM Logical Volume In-Reply-To: References: Message-ID: <47A1EB44.8080207@redhat.com> S P Arif Sahari Wibowo wrote: > Hi! > > I hope this issue is not Fedora specific, so I posting the question > here. I am working on a Fedora 8 kickstart installation that should > use existing LVM logical volume as home partition. However, it keep > failing with an error like this: > >> Could not allocate requested partitions: >> >> Adding this partition would not leave enough disk space >> for already allocated logical volumes in 2008AF_data.. >> >> Press 'OK' to exit the installer. > > * 2008AF_data is the name of volume group > > The kickstart code dealing with partition is quoted below: > > clearpart --all --drives=sda > part /boot --fstype ext3 --size=128 > part pv.1 --size=1536 --grow --ondisk=sda > volgroup 2008A6_work pv.1 > logvol swap --fstype swap --name=2008A6_swap --vgname=2008A6_work > --size=512 > logvol /var --fstype ext3 --name=2008A6_var --vgname=2008A6_work > --size=512 > logvol / --fstype ext3 --name=2008A6_root --vgname=2008A6_work > --size=3904 > part pv.2 --noformat --onpart sdb1 > volgroup 2008AF_data pv.2 --noformat --useexisting > logvol /home --noformat --useexisting --name=2008AF_home > --vgname=2008AF_data --fstype ext3 > > * 2008AF_home is the name of logical volume to be reused > > I already tried without --useexisting, with estimated size (not > accurate, but I added --grow), without --fstype, etc. same error. > > Any idea what's wrong? > > Thanks! > Have you looked at generating a sample kickstart using system-config-kickstart and using something derived from that one? --Michael From klaus.steden at thomson.net Thu Jan 31 19:23:27 2008 From: klaus.steden at thomson.net (Klaus Steden) Date: Thu, 31 Jan 2008 11:23:27 -0800 Subject: Weird issue trying to boot Dell 2450 ... Message-ID: Hi there, I've encountered a strange issue trying to bootstrap an old Dell 2450. It will boot from floppy, it will boot from a Dell provisioning CD, but it won't boot from a CentOS disc (neither the LiveCD nor a CentOS 4.5 ISO), and it won't boot its own hard drive. I'm working around it using a boot floppy (!! how vintage) to bootstrap the OS into loading, but I'm baffled as to why it won't load itself by any other means. When Windows 2003 was installed, it booted off the hard drive fine, so I know it -can- do it ... but I have no clue why it won't load Linux on its own. Has anyone ever had an experience like this with a machine, Dell or otherwise? It's running Fedora Core 6, so it's a fairly current OS, and has no issues when the OS is actually loaded. I'm okay with it booting off the floppy but that seems like kind of a ghetto solution, so I'd rather not have to if I don't have to. thanks, Klaus