From xxxxx at xxxxxxxx.xxx Fri Jun 1 10:47:10 2007 From: xxxxx at xxxxxxxx.xxx (xxxxxxxxxxxx) Date: Fri, 1 Jun 2007 12:47:10 +0200 Subject: Excluding RPMs fails? Message-ID: <18E77A5AD48C4AFB9DEB89632B2789E0@WORKSTATION> Hi, I am currently using anaconda 11.2.0.63, And I've built a custom distro. Now, all works great except for one thing. The target system is a i586 system with no CDROM drive (no hookup possible), and the building system (workstation) is an i686. My distro has both i586/686/x86_64 RPMs, in case I want to install it on 686/64b systems. I use the following code in my kickstart (example): kernel.i586 -kernel.i686 -kernel.x86_64 This way, the i586 kernel DOES get installed, the other 2 will not be. At least, that's how it should be... But, instead, the i686 system I use to install the drives on install the i686 kernel... not the i586... Any idea? Thanks in advance for your time and reply! From rfpike at fedex.com Fri Jun 1 17:27:50 2007 From: rfpike at fedex.com (Robbie Pike) Date: Fri, 1 Jun 2007 12:27:50 -0500 Subject: Kickstart issue Message-ID: I have an issue re-kickstarting RHEL 4 servers when I need to redploy them. I have HP Linux servers that the CCISS raid controller (came out of Compaq). The disk presented by this raid controller should up as /dev/cciss/c#d#. If I have the systems connected to SAN storage then they show up as /dev/sd[a, b,c, etc.]. If I try to kickstart a HP server that has SAN devices already zoned to it the kickstart process fails with "Input/output error during write on /dev/sda" error. Does anyone know what could be the problem and how to address. I'm would like to eleminate having to disconnecting the fiber connections or unzoning luns. Here's my partition section of the ks.cfg zerombr yes clearpart --drives=cciss/c0d0,cciss/c0d1 --all part swap --start=1 --end=800 --ondisk cciss/c0d0 --asprimary part / --fstype ext3 --start=801 --end=1057 --ondisk cciss/c0d0 --asprimary part pv.00 --size 1 --grow --asprimary --ondisk cciss/c0d0 volgroup vg00 pv.00 logvol /usr --vgname=vg00 --size=2024 --name=lvol1 logvol /var --vgname=vg00 --size=4096 --name=lvol2 logvol /home --vgname=vg00 --size=1024 --name=lvol3 logvol /opt --vgname=vg00 --size=4096 --name=lvol4 Thanks for any help, -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Fri Jun 1 17:44:32 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 01 Jun 2007 10:44:32 -0700 Subject: Kickstart issue In-Reply-To: Message-ID: This is a problem for us as well. The most reliable solution I've been given (but haven't had time to try) is to bust open your initrd and change the entry for your HBA to "ignore" in the /modules/pcitable file. There are a number of options that are SUPPOSED to work, like latefcload, but it hasn't worked for us. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Robbie Pike Sent: Friday, June 01, 2007 10:28 AM To: kickstart-list at redhat.com Subject: Kickstart issue I have an issue re-kickstarting RHEL 4 servers when I need to redploy them. I have HP Linux servers that the CCISS raid controller (came out of Compaq). The disk presented by this raid controller should up as /dev/cciss/c#d#. If I have the systems connected to SAN storage then they show up as /dev/sd[a, b,c, etc.]. If I try to kickstart a HP server that has SAN devices already zoned to it the kickstart process fails with "Input/output error during write on /dev/sda" error. Does anyone know what could be the problem and how to address. I'm would like to eleminate having to disconnecting the fiber connections or unzoning luns. Here's my partition section of the ks.cfg zerombr yes clearpart --drives=cciss/c0d0,cciss/c0d1 --all part swap --start=1 --end=800 --ondisk cciss/c0d0 --asprimary part / --fstype ext3 --start=801 --end=1057 --ondisk cciss/c0d0 --asprimary part pv.00 --size 1 --grow --asprimary --ondisk cciss/c0d0 volgroup vg00 pv.00 logvol /usr --vgname=vg00 --size=2024 --name=lvol1 logvol /var --vgname=vg00 --size=4096 --name=lvol2 logvol /home --vgname=vg00 --size=1024 --name=lvol3 logvol /opt --vgname=vg00 --size=4096 --name=lvol4 Thanks for any help, -------------- next part -------------- An HTML attachment was scrubbed... URL: From debian at herakles.homelinux.org Fri Jun 1 22:49:59 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Sat, 02 Jun 2007 06:49:59 +0800 Subject: Excluding RPMs fails? In-Reply-To: <18E77A5AD48C4AFB9DEB89632B2789E0@WORKSTATION> References: <18E77A5AD48C4AFB9DEB89632B2789E0@WORKSTATION> Message-ID: <4660A297.9060904@herakles.homelinux.org> xxxxxx xxxxx wrote: > Hi, > > I am currently using anaconda 11.2.0.63, And I've built a custom distro. > > Now, all works great except for one thing. > The target system is a i586 system with no CDROM drive (no hookup possible), > and the building system (workstation) is an i686. > > My distro has both i586/686/x86_64 RPMs, in case I want to install it on > 686/64b systems. > > I use the following code in my kickstart (example): > kernel.i586 > -kernel.i686 > -kernel.x86_64 > > This way, the i586 kernel DOES get installed, the other 2 will not be. At > least, that's how it should be... > But, instead, the i686 system I use to install the drives on install the > i686 kernel... not the i586... > > Any idea? Yeah. Anaconda knows you're being stoopid and that you really should use a 686 kernel. If it on %post by installing the kernel you want. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From jason at rampaginggeek.com Fri Jun 1 23:31:21 2007 From: jason at rampaginggeek.com (Jason Edgecombe) Date: Fri, 01 Jun 2007 19:31:21 -0400 Subject: Excluding RPMs fails? In-Reply-To: <18E77A5AD48C4AFB9DEB89632B2789E0@WORKSTATION> References: <18E77A5AD48C4AFB9DEB89632B2789E0@WORKSTATION> Message-ID: <4660AC49.5090809@rampaginggeek.com> xxxxxx xxxxx wrote: > Hi, > > I am currently using anaconda 11.2.0.63, And I've built a custom distro. > > Now, all works great except for one thing. > The target system is a i586 system with no CDROM drive (no hookup possible), > and the building system (workstation) is an i686. > > My distro has both i586/686/x86_64 RPMs, in case I want to install it on > 686/64b systems. > > I use the following code in my kickstart (example): > kernel.i586 > -kernel.i686 > -kernel.x86_64 > > This way, the i586 kernel DOES get installed, the other 2 will not be. At > least, that's how it should be... > But, instead, the i686 system I use to install the drives on install the > i686 kernel... not the i586... > > Any idea? > > Thanks in advance for your time and reply! > Is there any reason you can't do a network-based install using PXE, USB or grub boot from the hard drive? Is the target system networked? Jason From xxxxx at xxxxxxxx.xxx Sat Jun 2 07:11:35 2007 From: xxxxx at xxxxxxxx.xxx (xxxxxx xxxxx) Date: Sat, 2 Jun 2007 09:11:35 +0200 Subject: Excluding RPMs fails? Message-ID: Best regards, RAQTweak.com ------------------------------------------- raqtweak.com - cobalt experts http://www.raqtweak.com/ TOLL FREE: +1-888-678-1555 Well, the target system is a Cobalt RAQ. They do not have a VGA card or CDROM drive... They can boot from an NFS mount however... using DHCP... Is there a way kickstart can handle that? Otherwise, what are my options...? From xxxxx at xxxxxxxx.xxx Sat Jun 2 07:12:49 2007 From: xxxxx at xxxxxxxx.xxx (xxxxxx xxxxx) Date: Sat, 2 Jun 2007 09:12:49 +0200 Subject: Excluding RPMs fails? Message-ID: <5E2725B672BA439B88C0140D330F08E5@WORKSTATION> > Is there any reason you can't do a network-based install using PXE, USB or > grub boot from the hard drive? Is the target system networked? Well, the target system is a Cobalt RAQ. They do not have a VGA card or CDROM drive... They can boot from an NFS mount however... using DHCP... Is there a way kickstart can handle that? Otherwise, what are my options...? From jason at rampaginggeek.com Sat Jun 2 12:45:05 2007 From: jason at rampaginggeek.com (Jason Edgecombe) Date: Sat, 02 Jun 2007 08:45:05 -0400 Subject: Excluding RPMs fails? In-Reply-To: <5E2725B672BA439B88C0140D330F08E5@WORKSTATION> References: <5E2725B672BA439B88C0140D330F08E5@WORKSTATION> Message-ID: <46616651.8070204@rampaginggeek.com> xxxxxx xxxxx wrote: >> Is there any reason you can't do a network-based install using PXE, USB or >> > > >> grub boot from the hard drive? Is the target system networked? >> > > Well, the target system is a Cobalt RAQ. They do not have a VGA card or > CDROM drive... They can boot from an NFS mount however... using DHCP... Is > there a way kickstart can handle that? > > Otherwise, what are my options...? > hmmm, If you can get them to PXE boot, then that would be best. Otherwise, if you have to make the Raq somehow boot the vmlinuz and initrd files from the isolinux folder on the first cdrom. For example, I do remote hands-off installs without PXE on older linux boxes by copying the vmlinuz and initrd files from the cd to the /boot partition and then I copy a grub.conf file to /boot which has the new vmlinuz and initrd as the default boot option and then I reboot. The grub file includes the ks option to pull my kickstart file from the web, which installs from an NFS server. in your case, I would recommend the vnc option in your kickstart file so you can watch the install remotely. https://www.redhat.com/docs/manuals/enterprise/RHEL-5-manual/Installation_Guide-en-US/s1-kickstart2-options.html The main goal is just to get the Raq to somehow boot the vmlinuz & initrd that is copied from the cdrom. This could be via PXE or you can copy the files to the hard drive. If you have an ext2/3 partition on the Raq hard drive and can copy those two files and install grub, then you should be good to go. Does that make sense? Jason From debian at herakles.homelinux.org Sat Jun 2 23:27:09 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Sun, 03 Jun 2007 07:27:09 +0800 Subject: Excluding RPMs fails? In-Reply-To: References: Message-ID: <4661FCCD.5060907@herakles.homelinux.org> xxxxxx xxxxx wrote: > Best regards, > RAQTweak.com > > ------------------------------------------- > raqtweak.com - cobalt experts > http://www.raqtweak.com/ > TOLL FREE: +1-888-678-1555 > > > > > Well, the target system is a Cobalt RAQ. They do not have a VGA card or > CDROM drive... They can boot from an NFS mount however... using DHCP... Is > there a way kickstart can handle that? DHCP can provide information such as IP address, a server (name or IP), a boot file name. Presumably, your RAQ supports PXE or some other means to use this to "boot from an NFS mount." You haven't mentioned what you are trying to install, but RH documents for RHL/RHEL describe how to do network installs. Read it, interpret as needed for your environment. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From briandlong at gmail.com Tue Jun 5 00:55:49 2007 From: briandlong at gmail.com (Brian Long) Date: Mon, 4 Jun 2007 20:55:49 -0400 Subject: Kickstart issue In-Reply-To: References: Message-ID: <1c59d60b0706041755q44e4008bpac3329683a9072ee@mail.gmail.com> On 6/1/07, Shabazian, Chip wrote: > > > This is a problem for us as well. The most reliable solution I've been > given (but haven't had time to try) is to bust open your initrd and change > the entry for your HBA to "ignore" in the /modules/pcitable file. > > There are a number of options that are SUPPOSED to work, like latefcload, > but it hasn't worked for us. latefcload was added in a fairly recent update for RHEL 3, but I don't know if it's in RHEL 4. "nostorage" is another option only for RHEL 4 (and probably RHEL 5, but I have not tested it). On my previous team, we specified "nostorage" for all servers and then ks.cfg contained "device scsi cciss" for HP. This made sure we didn't overwrite any SAN storage. /Brian/ From jce at zot.com Tue Jun 5 03:34:28 2007 From: jce at zot.com (Chris Edillon) Date: Mon, 04 Jun 2007 23:34:28 -0400 Subject: Kickstart issue In-Reply-To: <1c59d60b0706041755q44e4008bpac3329683a9072ee@mail.gmail.com> References: <1c59d60b0706041755q44e4008bpac3329683a9072ee@mail.gmail.com> Message-ID: <4664D9C4.3020302@zot.com> Brian Long wrote: > On 6/1/07, Shabazian, Chip wrote: >> >> This is a problem for us as well. The most reliable solution I've been >> given (but haven't had time to try) is to bust open your initrd and >> change >> the entry for your HBA to "ignore" in the /modules/pcitable file. >> >> There are a number of options that are SUPPOSED to work, like latefcload, >> but it hasn't worked for us. > > latefcload was added in a fairly recent update for RHEL 3, but I don't > know if it's in RHEL 4. "nostorage" is another option only for RHEL 4 > (and probably RHEL 5, but I have not tested it). On my previous team, > we specified "nostorage" for all servers and then ks.cfg contained > "device scsi cciss" for HP. This made sure we didn't overwrite any > SAN storage. > latefcload seems to be there at least since RHEL4u3, and it's there in RHEL5. we couldn't make much use of it, though, except in conjunction with ignoredisk in RHEL5. the stubs seem to be there for ignoredisk in the anaconda that comes with RHEL4, but it is silently ignored. using noprobe/nostorage hasn't been an option for us either, since we have a single kickstart for all of our machines instead of one per hardware type. haven't taken the time to see about writing a %pre script which would look at the available storage devices from a known list and generate the proper device line for the ks.cfg. besides, it makes sense to let anaconda probe the hardware since it's pretty good at doing so accurately. what i'd *really* like to see is the ability to use the device command to ignore certain devices instead of specify, i.e. "device scsi qla2xxx ignore" which would load all of the probed scsi drivers except for qla2xxx. any anaconda developers out there listening...? chris From jentrena at mailext.com Tue Jun 5 10:38:00 2007 From: jentrena at mailext.com (Entrena Perez, Julio [HP]) Date: Tue, 5 Jun 2007 12:38:00 +0200 Subject: Kickstart issue In-Reply-To: <4664D9C4.3020302@zot.com> References: <1c59d60b0706041755q44e4008bpac3329683a9072ee@mail.gmail.com> <4664D9C4.3020302@zot.com> Message-ID: <80FDDD0100BF5F42BF6391EAB5C8D36301DC8FA3@sbext231.mailext.com> Hi Chris, The easiest solution is unplugging the fiber cables from the HBAs during the installation. In case that this is not an option for you, you may want to take a look at http://kbase.redhat.com/faq/FAQ_79_8927.shtm Hope this helps, give us some feedback. Best, Juli -----Mensaje original----- De: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] En nombre de Chris Edillon Enviado el: martes, 05 de junio de 2007 5:34 Para: Discussion list about Kickstart Asunto: Re: Kickstart issue Brian Long wrote: > On 6/1/07, Shabazian, Chip wrote: >> >> This is a problem for us as well. The most reliable solution I've been >> given (but haven't had time to try) is to bust open your initrd and >> change >> the entry for your HBA to "ignore" in the /modules/pcitable file. >> >> There are a number of options that are SUPPOSED to work, like latefcload, >> but it hasn't worked for us. > > latefcload was added in a fairly recent update for RHEL 3, but I don't > know if it's in RHEL 4. "nostorage" is another option only for RHEL 4 > (and probably RHEL 5, but I have not tested it). On my previous team, > we specified "nostorage" for all servers and then ks.cfg contained > "device scsi cciss" for HP. This made sure we didn't overwrite any > SAN storage. > latefcload seems to be there at least since RHEL4u3, and it's there in RHEL5. we couldn't make much use of it, though, except in conjunction with ignoredisk in RHEL5. the stubs seem to be there for ignoredisk in the anaconda that comes with RHEL4, but it is silently ignored. using noprobe/nostorage hasn't been an option for us either, since we have a single kickstart for all of our machines instead of one per hardware type. haven't taken the time to see about writing a %pre script which would look at the available storage devices from a known list and generate the proper device line for the ks.cfg. besides, it makes sense to let anaconda probe the hardware since it's pretty good at doing so accurately. what i'd *really* like to see is the ability to use the device command to ignore certain devices instead of specify, i.e. "device scsi qla2xxx ignore" which would load all of the probed scsi drivers except for qla2xxx. any anaconda developers out there listening...? chris _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From alessiof at gmail.com Mon Jun 11 10:24:38 2007 From: alessiof at gmail.com (Alessio Focardi) Date: Mon, 11 Jun 2007 12:24:38 +0200 Subject: FC7 kickstart: CDROM non found Message-ID: <8bc972b70706110324j7cd4dd65q37fd301de2c0d476@mail.gmail.com> Hi, I moved my personalized install from FC6 to F7 but i'm getting into troubles with the ks.cfg file. My iso works flawless if run from a cd drive that's connected to the primary ide but does not work using a cdrom on the secondary controller or from an usb cdrom. Iso boots then i get the request to input manually the position of the kickstart file. Log shows getting kickstart file from first CDROM no CDROM found I also have to say that the FC6 iso is installing (with the kickstart) from primary, secondary and usb cdroms on the same machine. Here is my relevant isolinux.cfg label linux kernel vmlinuz append ks=cdrom:/ks.cfg initrd=initrd.img skipddc quiet Probably the problem depends from the "device waltrough" of anaconda, but thats my guess. Any help will be greatly appreciated! Tnx for the support! -------------- next part -------------- An HTML attachment was scrubbed... URL: From debian at herakles.homelinux.org Mon Jun 11 11:11:04 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Mon, 11 Jun 2007 19:11:04 +0800 Subject: FC7 kickstart: CDROM non found In-Reply-To: <8bc972b70706110324j7cd4dd65q37fd301de2c0d476@mail.gmail.com> References: <8bc972b70706110324j7cd4dd65q37fd301de2c0d476@mail.gmail.com> Message-ID: <466D2DC8.5010904@herakles.homelinux.org> Alessio Focardi wrote: > Hi, > > I moved my personalized install from FC6 to F7 but i'm getting into > troubles > with the ks.cfg file. > > My iso works flawless if run from a cd drive that's connected to the > primary ide > but does not work using a cdrom on the secondary controller or from an > usb cdrom. > > Iso boots then i get the request to input manually the position of the > kickstart > file. > > Log shows > > getting kickstart file from first CDROM > no CDROM found > > I also have to say that the FC6 iso is installing (with the kickstart) from > primary, secondary and usb cdroms on the same machine. Does a manual install of F7 work? > > Here is my relevant isolinux.cfg > > label linux > kernel vmlinuz > append ks=cdrom:/ks.cfg initrd=initrd.img skipddc quiet > > Probably the problem depends from the "device waltrough" of anaconda, > but thats > my guess. > > Any help will be greatly appreciated! > > Tnx for the support! > > > ------------------------------------------------------------------------ > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From alessiof at gmail.com Tue Jun 12 07:43:14 2007 From: alessiof at gmail.com (Alessio Focardi) Date: Tue, 12 Jun 2007 09:43:14 +0200 Subject: FC7 kickstart: CDROM non found In-Reply-To: <466D2DC8.5010904@herakles.homelinux.org> References: <8bc972b70706110324j7cd4dd65q37fd301de2c0d476@mail.gmail.com> <466D2DC8.5010904@herakles.homelinux.org> Message-ID: <8bc972b70706120043o4e0ee9d5v841541c50f8f9f65@mail.gmail.com> On 6/11/07, John Summerfield wrote: > > Alessio Focardi wrote: > > Hi, > > > > I moved my personalized install from FC6 to F7 but i'm getting into > > troubles > > with the ks.cfg file. > > > > My iso works flawless if run from a cd drive that's connected to the > > primary ide > > but does not work using a cdrom on the secondary controller or from an > > usb cdrom. > > Does a manual install of F7 work? No, I just noticed that! So the problem is not anaconda, or kickstart ... probably they removed some ide driver from the installation iso. Anyone has some hint about how i can add support for my chipset (via700) to the fc7 iso ? Shall I try a "mashup" of FC6 and F7 iso's ? Tnx for the support! -------------- next part -------------- An HTML attachment was scrubbed... URL: From debian at herakles.homelinux.org Tue Jun 12 12:37:41 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Tue, 12 Jun 2007 20:37:41 +0800 Subject: FC7 kickstart: CDROM non found In-Reply-To: <8bc972b70706120043o4e0ee9d5v841541c50f8f9f65@mail.gmail.com> References: <8bc972b70706110324j7cd4dd65q37fd301de2c0d476@mail.gmail.com> <466D2DC8.5010904@herakles.homelinux.org> <8bc972b70706120043o4e0ee9d5v841541c50f8f9f65@mail.gmail.com> Message-ID: <466E9395.3050408@herakles.homelinux.org> Alessio Focardi wrote: > On 6/11/07, John Summerfield wrote: >> >> Alessio Focardi wrote: >> > Hi, >> > >> > I moved my personalized install from FC6 to F7 but i'm getting into >> > troubles >> > with the ks.cfg file. >> > >> > My iso works flawless if run from a cd drive that's connected to the >> > primary ide >> > but does not work using a cdrom on the secondary controller or from an >> > usb cdrom. >> >> Does a manual install of F7 work? > > > No, I just noticed that! > > So the problem is not anaconda, or kickstart ... probably they removed some > ide driver from the installation iso. I wouldn't rule out Anaconda, but you're definitely in the Wrong Place now. Probably the Fedora Users' list is a better forum. > > Anyone has some hint about how i can add support for my chipset (via700) to > the fc7 iso ? > > Shall I try a "mashup" of FC6 and F7 iso's ? No. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From Jan.Michael at cern.ch Wed Jun 13 15:15:11 2007 From: Jan.Michael at cern.ch (Jan Michael) Date: Wed, 13 Jun 2007 17:15:11 +0200 Subject: lvm, creation of physical volumes Message-ID: <72BAE107-D5B6-4338-A243-F1C30D0D0002@cern.ch> Hello kickstart mailing list, my questions regards the creation of physical volumes via lvm. I specified the following disk layout in my kickstart file: 26 clearpart --all --drives cciss/c0d0 --initlabel 27 28 part /boot --size 100 --ondisk cciss/c0d0 --fstype ext3 29 part / --size 7168 --ondisk cciss/c0d0 --fstype ext3 30 part swap --size 2048 --ondisk cciss/c0d0 --fstype swap 31 part pv.00 --size 1024 --ondisk cciss/c0d0 --fstype ext3 --grow This works well and will successfully create the following partition table: Disk /dev/cciss/c0d0: 73.3 GB, 73372631040 bytes 255 heads, 63 sectors/track, 8920 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Device Boot Start End Blocks Id System /dev/cciss/c0d0p1 * 1 13 104391 83 Linux /dev/cciss/c0d0p2 14 927 7341705 83 Linux /dev/cciss/c0d0p3 928 1188 2096482+ 82 Linux swap /dev/cciss/c0d0p4 1189 8920 62107290 5 Extended /dev/cciss/c0d0p5 1189 8920 62107258+ 8e Linux LVM So far so good. But then there is no physical volume created. The output of 'pvs' and 'lvmdiskscan' shows me nothing about my lv. Therefore it is necessary to know when anaconda creates the physical volume (pv), like I can do with 'pvcreate'. Will it be created with the above shown line in the kickstart file and something else is wrong or does the pv will not created until I issue something like that: volgroup VolGroup00 pv.00 Thank you for your help. Cheers, Jan -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 4028 bytes Desc: not available URL: From Greg.Caetano at hp.com Wed Jun 13 15:56:17 2007 From: Greg.Caetano at hp.com (Caetano, Greg) Date: Wed, 13 Jun 2007 11:56:17 -0400 Subject: lvm, creation of physical volumes In-Reply-To: <72BAE107-D5B6-4338-A243-F1C30D0D0002@cern.ch> References: <72BAE107-D5B6-4338-A243-F1C30D0D0002@cern.ch> Message-ID: Here is an example of anaconda creating the LV as reported in anaconda-ks.cfg after an interactive installation: # The following is the partition information you requested # Note that any partitions you deleted are not expressed # here so unless you clear all partitions first, this is # not guaranteed to work #clearpart --all --drives=cciss/c0d0,cciss/c0d1 #part /boot --fstype ext3 --size=100 --ondisk=cciss/c0d0 #part pv.4 --size=0 --grow --ondisk=cciss/c0d1 #part pv.3 --size=0 --grow --ondisk=cciss/c0d0 #volgroup VolGroup01 --pesize=32768 pv.4 pv.3 #logvol swap --fstype swap --name=LogVol01 --vgname=VolGroup01 --size=1000 --grow --maxsize=1984 #logvol / --fstype ext3 --name=LogVol00 --vgname=VolGroup01 --size=1024 --grow Greg Caetano HP TSG Linux Solutions Alliances Engineering Chicago, IL greg.caetano at hp.com Red Hat Certified Engineer RHCE#803004972711193 -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Jan Michael Sent: Wednesday, June 13, 2007 10:15 AM To: kickstart-list at redhat.com Subject: lvm, creation of physical volumes Hello kickstart mailing list, my questions regards the creation of physical volumes via lvm. I specified the following disk layout in my kickstart file: 26 clearpart --all --drives cciss/c0d0 --initlabel 27 28 part /boot --size 100 --ondisk cciss/c0d0 --fstype ext3 29 part / --size 7168 --ondisk cciss/c0d0 --fstype ext3 30 part swap --size 2048 --ondisk cciss/c0d0 --fstype swap 31 part pv.00 --size 1024 --ondisk cciss/c0d0 --fstype ext3 --grow This works well and will successfully create the following partition table: Disk /dev/cciss/c0d0: 73.3 GB, 73372631040 bytes 255 heads, 63 sectors/track, 8920 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Device Boot Start End Blocks Id System /dev/cciss/c0d0p1 * 1 13 104391 83 Linux /dev/cciss/c0d0p2 14 927 7341705 83 Linux /dev/cciss/c0d0p3 928 1188 2096482+ 82 Linux swap /dev/cciss/c0d0p4 1189 8920 62107290 5 Extended /dev/cciss/c0d0p5 1189 8920 62107258+ 8e Linux LVM So far so good. But then there is no physical volume created. The output of 'pvs' and 'lvmdiskscan' shows me nothing about my lv. Therefore it is necessary to know when anaconda creates the physical volume (pv), like I can do with 'pvcreate'. Will it be created with the above shown line in the kickstart file and something else is wrong or does the pv will not created until I issue something like that: volgroup VolGroup00 pv.00 Thank you for your help. Cheers, Jan From srobson at cadence.com Wed Jun 13 16:11:44 2007 From: srobson at cadence.com (Steve Robson) Date: Wed, 13 Jun 2007 17:11:44 +0100 Subject: lvm, creation of physical volumes In-Reply-To: <20070613160015.998BB7362F@hormel.redhat.com> References: <20070613160015.998BB7362F@hormel.redhat.com> Message-ID: <46701740.9090803@cadence.com> > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Jan Michael > Sent: Wednesday, June 13, 2007 10:15 AM > To: kickstart-list at redhat.com > Subject: lvm, creation of physical volumes > > Hello kickstart mailing list, > > my questions regards the creation of physical volumes via lvm. > I specified the following disk layout in my kickstart file: > > 26 clearpart --all --drives cciss/c0d0 --initlabel > 27 > 28 part /boot --size 100 --ondisk > cciss/c0d0 --fstype ext3 > 29 part / --size 7168 --ondisk > cciss/c0d0 --fstype ext3 > 30 part swap --size 2048 --ondisk > cciss/c0d0 --fstype swap > 31 part pv.00 --size 1024 --ondisk > cciss/c0d0 --fstype ext3 --grow > > This works well and will successfully create the following partition > table: > > Disk /dev/cciss/c0d0: 73.3 GB, 73372631040 bytes > 255 heads, 63 sectors/track, 8920 cylinders Units = cylinders of 16065 * > 512 = 8225280 bytes > > Device Boot Start End Blocks Id System > /dev/cciss/c0d0p1 * 1 13 104391 83 Linux > /dev/cciss/c0d0p2 14 927 7341705 83 Linux > /dev/cciss/c0d0p3 928 1188 2096482+ 82 Linux > swap > /dev/cciss/c0d0p4 1189 8920 62107290 5 Extended > /dev/cciss/c0d0p5 1189 8920 62107258+ 8e Linux > LVM > > So far so good. But then there is no physical volume created. The output > of 'pvs' and 'lvmdiskscan' shows me nothing about my lv. Hi Jan, As Greg Caetano showed, you need to define at least one logical volume. Here's an excerpt from my ks.cfg file, which successfully sets up LVM using three physical HDDs. You'll get the picture: zerombr yes clearpart --all ## Note: 255GB FS size limit in LVM part /boot --fstype ext3 --size 100 --asprimary --ondisk=sda part pv.00 --size 1 --grow --ondisk=sda part pv.01 --size 1 --grow --ondisk=sdb part pv.02 --size 1 --grow --ondisk=sdc volgroup vg00 pv.00 pv.01 pv.02 logvol / --fstype ext3 --vgname=vg00 --size=9000 --name=lroot logvol /tmp --fstype ext3 --vgname=vg00 --size=50000 --name=ltmp logvol /vol0 --fstype ext3 --vgname=vg00 --size=100000 --name=lvol0 logvol /vol1 --fstype ext3 --vgname=vg00 --size=100000 --name=lvol1 logvol swap --vgname=vg00 --size=32768 --name=lswap -- Regards, Steve IT Support - UNIX/Linux Cadence Design Systems Bagshot Road Bracknell BERKSHIRE RG12 0PH UK From Jan.Michael at cern.ch Wed Jun 13 16:41:21 2007 From: Jan.Michael at cern.ch (Jan Michael) Date: Wed, 13 Jun 2007 18:41:21 +0200 Subject: lvm, creation of physical volumes In-Reply-To: <46701740.9090803@cadence.com> References: <20070613160015.998BB7362F@hormel.redhat.com> <46701740.9090803@cadence.com> Message-ID: <1FBDF0DA-C15A-4D4C-9F26-53E105C9E207@cern.ch> Hi Greg, hi Steve, hi list, I'm sorry. It seems to be that I put concern in wrong words. I don't want anaconda/kickstart to set up the whole lvm environment for the system. Anaconda should simply create the partition and initialize this partition to make it ready for lvm usage. The reason why it is not necessary to create a volume group is, that this will be done by another component later in the installation process. I would like to know, if the part command takes care of the creation of the partition and its initialization (i.e. with pvcreate) or if it is not possible to seperate this step from the volume group creation. >> 31 part pv.00 --size 1024 -- >> ondisk cciss/c0d0 --fstype ext3 Cheers, Jan On 13.06.2007, at 18:11, Steve Robson wrote: >> -----Original Message----- >> From: kickstart-list-bounces at redhat.com >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Jan Michael >> Sent: Wednesday, June 13, 2007 10:15 AM >> To: kickstart-list at redhat.com >> Subject: lvm, creation of physical volumes >> Hello kickstart mailing list, >> my questions regards the creation of physical volumes via lvm. >> I specified the following disk layout in my kickstart file: >> 26 clearpart --all --drives cciss/c0d0 --initlabel >> 27 >> 28 part /boot --size 100 -- >> ondisk cciss/c0d0 --fstype ext3 >> 29 part / --size 7168 -- >> ondisk cciss/c0d0 --fstype ext3 >> 30 part swap --size 2048 -- >> ondisk cciss/c0d0 --fstype swap >> 31 part pv.00 --size 1024 -- >> ondisk cciss/c0d0 --fstype ext3 --grow >> This works well and will successfully create the following partition >> table: >> Disk /dev/cciss/c0d0: 73.3 GB, 73372631040 bytes >> 255 heads, 63 sectors/track, 8920 cylinders Units = cylinders of >> 16065 * >> 512 = 8225280 bytes >> Device Boot Start End Blocks Id >> System >> /dev/cciss/c0d0p1 * 1 13 104391 83 Linux >> /dev/cciss/c0d0p2 14 927 7341705 83 Linux >> /dev/cciss/c0d0p3 928 1188 2096482+ 82 >> Linux swap >> /dev/cciss/c0d0p4 1189 8920 62107290 5 >> Extended >> /dev/cciss/c0d0p5 1189 8920 62107258+ 8e >> Linux LVM >> So far so good. But then there is no physical volume created. The >> output >> of 'pvs' and 'lvmdiskscan' shows me nothing about my lv. > > Hi Jan, > > As Greg Caetano showed, you need to define at least one logical > volume. Here's an excerpt from my ks.cfg file, which successfully > sets up LVM using three physical HDDs. You'll get the picture: > > zerombr yes > clearpart --all > ## Note: 255GB FS size limit in LVM > part /boot --fstype ext3 --size 100 --asprimary --ondisk=sda > part pv.00 --size 1 --grow --ondisk=sda > part pv.01 --size 1 --grow --ondisk=sdb > part pv.02 --size 1 --grow --ondisk=sdc > volgroup vg00 pv.00 pv.01 pv.02 > logvol / --fstype ext3 --vgname=vg00 --size=9000 --name=lroot > logvol /tmp --fstype ext3 --vgname=vg00 --size=50000 --name=ltmp > logvol /vol0 --fstype ext3 --vgname=vg00 --size=100000 --name=lvol0 > logvol /vol1 --fstype ext3 --vgname=vg00 --size=100000 --name=lvol1 > logvol swap --vgname=vg00 --size=32768 --name=lswap > > -- > Regards, > Steve > > IT Support - UNIX/Linux Cadence Design Systems > Bagshot Road > Bracknell BERKSHIRE > RG12 0PH UK > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 4028 bytes Desc: not available URL: From tv at solnet.ch Thu Jun 14 00:49:09 2007 From: tv at solnet.ch (Thomas Vogt) Date: Thu, 14 Jun 2007 02:49:09 +0200 Subject: kickstart with fedora7 and arping checks Message-ID: <46709085.6070006@solnet.ch> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Hi Since I use fedora 4 with kickstart I always had to remove some arping checks in the network scripts of fedora because of my arp proxy in front of all fedora boxes. Now I try to install fedora 7 via kickstart. Is there a way to disable any arping check in fedora without always removing some parts in the network script code with %post commands? In Fedora7 I have another issue. I guess I removed all arping checks but I still have some issues to set an IP address to eth0. If I add the ip via ifconfig and route add command it works but with /etc/rc.d/init.d/network start,stop,restart the script starts the network without showing any errors but it doesn't set any ip address. It's all static and the ifcfg-eth0 looks ok. Is there a way to enable some debug flags for scripts? Regards, Thomas -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iD8DBQFGcJCFGCwkYTI5tyARCF5bAJ96NsI0lT4OBxB65Y1Z529jBgPhigCghCVg CAtnwe8dfjwYrq/LiHUKFg8= =eHeF -----END PGP SIGNATURE----- From wpilorz at bdk.pl Thu Jun 14 06:20:59 2007 From: wpilorz at bdk.pl (Wojtek.Pilorz) Date: Thu, 14 Jun 2007 08:20:59 +0200 (CEST) Subject: kickstart with fedora7 and arping checks In-Reply-To: <46709085.6070006@solnet.ch> Message-ID: On Thu, 14 Jun 2007, Thomas Vogt wrote: > Hash: SHA256 > > Hi > > Since I use fedora 4 with kickstart I always had to remove some arping > checks in the network scripts of fedora because of my arp proxy in front > of all fedora boxes. > > Now I try to install fedora 7 via kickstart. Is there a way to disable > any arping check in fedora without always removing some parts in the > network script code with %post commands? > > In Fedora7 I have another issue. I guess I removed all arping checks but > I still have some issues to set an IP address to eth0. If I add the ip > via ifconfig and route add command it works but with > /etc/rc.d/init.d/network start,stop,restart the script starts the > network without showing any errors but it doesn't set any ip address. If you use static ip address then you might havae issue with arping. You might also run the script through bash explicitely - if that would work, it would most probably mean some SELinux issue. > It's all static and the ifcfg-eth0 looks ok. Is there a way to enable > some debug flags for scripts? bash -vx /etc/rc.d/.... Use it only after verifying that bash /etc/rc.d/.... does not work (see above). > > Regards, > Thomas From Matt.Fahrner at coat.com Thu Jun 14 15:53:33 2007 From: Matt.Fahrner at coat.com (Matt Fahrner) Date: Thu, 14 Jun 2007 11:53:33 -0400 Subject: NFS versus HTTP kickstart Message-ID: <4671647D.3060404@coat.com> We currently use the NFS based kickstart for our installations but are trying to move to HTTP. However, in both our Fedora and RedHat 9 tests, while most things seem to be the same, some things don't seem to behave symmetrically. For instance, and our biggest problem, is with Fedora Core 6 where with an NFS install it correctly probes and sets the graphics card in "/etc/xorg.conf" (to Intel "i810"), but with an HTTP install it probes it and sets the wrong one ("vesa"). We verified it's not an issue with the "%post" having a dependency on the NFS install directory, plus none of our home grown code/RPMs touches "xorg.conf". You'd think it would behave the same regardless. Anyone with a possible explanation (or solution)? Thanks in advance, - Matt -- --------------------------------------------------------------------- Matt Fahrner 2 South Park St. Chief Systems Architect Willis House Burlington Coat Factory Warehouse Lebanon, N.H. 03766 Tel: (603) 448-4100 x5150 USA Fax: (603) 443-6190 Matt.Fahrner at COAT.COM --------------------------------------------------------------------- From gwaugh at frontstreetnetworks.com Thu Jun 14 15:17:14 2007 From: gwaugh at frontstreetnetworks.com (Gerald Waugh) Date: Thu, 14 Jun 2007 10:17:14 -0500 Subject: FC7 kickstart: CDROM non found In-Reply-To: <8bc972b70706110324j7cd4dd65q37fd301de2c0d476@mail.gmail.com> Message-ID: <027401c7ae97$1abe1cb0$0601a8c0@systemax> Hi, We have problems similar on some servers, when using usb cdrom. We wait for the pause when the boot xcreen reads "Loading * Watch the display, when pause at "Loading /sbin/loader" * Quickly remove and reinsert the USB cable to the CD ROM Drive * You should see an info box "Loading USB storage driver" * If you see "Select Language" then you missed it. * Power down and start over.. - Gerald Waugh http://frontstreetnetworks.com Front Street Networks 4604 Fern Ave Shreveport, LA 71105-3118 V: 318-862-3575/3577 F: 318-862-3576 -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Alessio Focardi Sent: Monday, June 11, 2007 5:25 AM To: kickstart-list at redhat.com Subject: FC7 kickstart: CDROM non found Hi, I moved my personalized install from FC6 to F7 but i'm getting into troubles with the ks.cfg file. My iso works flawless if run from a cd drive that's connected to the primary ide but does not work using a cdrom on the secondary controller or from an usb cdrom. Iso boots then i get the request to input manually the position of the kickstart file. Log shows getting kickstart file from first CDROM no CDROM found I also have to say that the FC6 iso is installing (with the kickstart) from primary, secondary and usb cdroms on the same machine. Here is my relevant isolinux.cfg label linux kernel vmlinuz append ks=cdrom:/ks.cfg initrd=initrd.img skipddc quiet Probably the problem depends from the "device waltrough" of anaconda, but thats my guess. Any help will be greatly appreciated! Tnx for the support! -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ayoung at vigilos.com Thu Jun 14 17:44:02 2007 From: ayoung at vigilos.com (Alasdair Young) Date: Thu, 14 Jun 2007 10:44:02 -0700 Subject: Any idea how to avoid formatting my USB drive? Message-ID: <46717E62.7040608@vigilos.com> Hi. I am trying to set up an automatic install of FC6 using kickstart. I have a bootable USB stick with a syslinux.cfg file that points to a kickstart file located on a central server (via http). Everything works well and is completely automated, however I cannot seem to work out how to stop the install from wanting to format my USB stick. I have tried removing "--all" from the clearpart command below, or even just changing it to "--linux" without any luck.I've searched google, but I can't find any discussion of the issue. Is there any way to specify a drive that should NOT be formatted, or any other way to work around this problem? Kind regards, - alasdair I have the following kickstart file: install url --url http://pxe/fedora/6/i386/core/os skipx text lang en_US.UTF-8 keyboard us network --device eth0 --bootproto static --ip=10.22.22.1 --netmask=255.255.255.0 network --device eth1 --bootproto dhcp rootpw --iscrypted (**whatever**) firewall --enabled --port=22:tcp authconfig --enableshadow --enablemd5 selinux --disabled timezone --utc GMT bootloader --location=mbr --driveorder=hda clearpart --all --drives=hda part /boot --fstype ext3 --size=100 --ondisk=hda part pv.7 --size=0 --grow --ondisk=hda volgroup VolGroup00 --pesize=32768 pv.7 logvol / --fstype ext3 --name=LogVol00 --vgname=VolGroup00 --size=1024 --grow logvol swap --fstype swap --name=LogVol01 --vgname=VolGroup00 --size=1000 --grow --maxsize=1984 %packages @base %post // various things here that are irrelevant. From Chip.Shabazian at bankofamerica.com Thu Jun 14 17:48:47 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Thu, 14 Jun 2007 10:48:47 -0700 Subject: Any idea how to avoid formatting my USB drive? In-Reply-To: <46717E62.7040608@vigilos.com> Message-ID: You simply tell kickstart to ignore the drive in the kickstart command section. Assuming your usb drive is sda: ignoredrisk drives=sda -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Alasdair Young Sent: Thursday, June 14, 2007 10:44 AM To: kickstart-list at redhat.com Subject: Any idea how to avoid formatting my USB drive? Hi. I am trying to set up an automatic install of FC6 using kickstart. I have a bootable USB stick with a syslinux.cfg file that points to a kickstart file located on a central server (via http). Everything works well and is completely automated, however I cannot seem to work out how to stop the install from wanting to format my USB stick. I have tried removing "--all" from the clearpart command below, or even just changing it to "--linux" without any luck.I've searched google, but I can't find any discussion of the issue. Is there any way to specify a drive that should NOT be formatted, or any other way to work around this problem? Kind regards, - alasdair I have the following kickstart file: install url --url http://pxe/fedora/6/i386/core/os skipx text lang en_US.UTF-8 keyboard us network --device eth0 --bootproto static --ip=10.22.22.1 --netmask=255.255.255.0 network --device eth1 --bootproto dhcp rootpw --iscrypted (**whatever**) firewall --enabled --port=22:tcp authconfig --enableshadow --enablemd5 selinux --disabled timezone --utc GMT bootloader --location=mbr --driveorder=hda clearpart --all --drives=hda part /boot --fstype ext3 --size=100 --ondisk=hda part pv.7 --size=0 --grow --ondisk=hda volgroup VolGroup00 --pesize=32768 pv.7 logvol / --fstype ext3 --name=LogVol00 --vgname=VolGroup00 --size=1024 --grow logvol swap --fstype swap --name=LogVol01 --vgname=VolGroup00 --size=1000 --grow --maxsize=1984 %packages @base %post // various things here that are irrelevant. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From ayoung at vigilos.com Thu Jun 14 18:14:25 2007 From: ayoung at vigilos.com (Alasdair Young) Date: Thu, 14 Jun 2007 11:14:25 -0700 Subject: Any idea how to avoid formatting my USB drive? In-Reply-To: References: Message-ID: <46718581.7000209@vigilos.com> Thanks for this :) In case anyone has the same problem, the exact command is: ignoredisk --drives=sda This solved my problem. Thank you once again! - alasdair Shabazian, Chip wrote: > You simply tell kickstart to ignore the drive in the kickstart command > section. Assuming your usb drive is sda: > > ignoredrisk drives=sda > > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Alasdair Young > Sent: Thursday, June 14, 2007 10:44 AM > To: kickstart-list at redhat.com > Subject: Any idea how to avoid formatting my USB drive? > > Hi. > > I am trying to set up an automatic install of FC6 using kickstart. I > have a bootable USB stick with a syslinux.cfg file that points to a > kickstart file located on a central server (via http). > > Everything works well and is completely automated, however I cannot seem > to work out how to stop the install from wanting to format my USB stick. > > I have tried removing "--all" from the clearpart command below, or even > just changing it to "--linux" without any luck.I've searched google, but > I can't find any discussion of the issue. > > Is there any way to specify a drive that should NOT be formatted, or any > other way to work around this problem? > > Kind regards, > > - alasdair > > > > I have the following kickstart file: > > install > url --url http://pxe/fedora/6/i386/core/os skipx text > > lang en_US.UTF-8 > keyboard us > network --device eth0 --bootproto static --ip=10.22.22.1 > --netmask=255.255.255.0 network --device eth1 --bootproto dhcp rootpw > --iscrypted (**whatever**) firewall --enabled --port=22:tcp authconfig > --enableshadow --enablemd5 selinux --disabled timezone --utc GMT > bootloader --location=mbr --driveorder=hda > > clearpart --all --drives=hda > part /boot --fstype ext3 --size=100 --ondisk=hda part pv.7 --size=0 > --grow --ondisk=hda volgroup VolGroup00 --pesize=32768 pv.7 logvol / > --fstype ext3 --name=LogVol00 --vgname=VolGroup00 --size=1024 --grow > logvol swap --fstype swap --name=LogVol01 --vgname=VolGroup00 > --size=1000 --grow --maxsize=1984 > > %packages > @base > > %post > // various things here that are irrelevant. > > _______________________________________________ > 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 Chip.Shabazian at bankofamerica.com Tue Jun 19 16:35:29 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Tue, 19 Jun 2007 09:35:29 -0700 Subject: Linuxworld Kickstart Tips and Tricks In-Reply-To: Message-ID: Howdy, As I mentioned a while back, I'm doing a "Kickstart Tips and Tricks" seminar at Linuxworld this summer, and I need to finish (ok, START and finish) my presentation this weekend. I've gotten a few good tips from you all, so this is my final call for help: I will of course make the training materials I develop available to all when it's done. Here is what I'm looking for: * If you were attending this training, what would you like to see covered * Although you may be well versed in kickstart, what should I cover for those not so experienced * What other technologies do you use in your provisioning of systems * What "best practices" have you learned * What else should I be asking for and presenting * And of course, all those wonderful tips and tricks that we have all learned over the years Anything you send may be printed in the workbook that accompanies the tutorial, so I need your approval to print it. If you would like to be credited, please provide me with how you want it to appear. Here is a brief outline that I have so far, both content AND topics are appreciated: What is kickstart Why use kickstart Kickstart sections boot: options Command Section %include is your friend Yummy builds %Pre %Pre Magic %Post To chroot or not to chroot %Post magic Life after %post Troubleshooting Common problems Tools to make life easier Cobbler Revisor Novi mRepo Kickweb On the fly ks.cfg Other provisioning technologies Best Practices Resources Thanks Please email me directly at chip.shabazian at bankofamerica.com Thanks everyone! Chip Shabazian _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Klaus.Steden at thomson.net Tue Jun 19 18:29:13 2007 From: Klaus.Steden at thomson.net (Steden Klaus) Date: Tue, 19 Jun 2007 14:29:13 -0400 Subject: Linuxworld Kickstart Tips and Tricks References: Message-ID: One of the things I've found with Kickstart is that the anaconda-ks.cfg file that gets copied during last-mile installation isn't quite as complete as I'd like as an administrator. One of the things I do in %post is copy the working ks.cfg (found at /tmp/ks.cfg by default, I think) to a file on disk so I have a more complete picture of the system's install information. I did do some work building an on-the-fly Kickstart generator, although there's a page on the Anaconda Wiki about it, so I'll spare everyone the gory details. :-) Cheers, Klaus > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Shabazian, Chip > Sent: Tuesday, June 19, 2007 9:35 AM > To: Discussion list about Kickstart > Subject: Linuxworld Kickstart Tips and Tricks > > Howdy, > > As I mentioned a while back, I'm doing a "Kickstart Tips and Tricks" > seminar at Linuxworld this summer, and I need to finish (ok, START and > finish) my presentation this weekend. I've gotten a few good tips from > you all, so this is my final call for help: > > I will of course make the training materials I develop available to all > when it's done. > > Here is what I'm looking for: > * If you were attending this training, what would you like to see > covered > * Although you may be well versed in kickstart, what should I cover for > those not so experienced > * What other technologies do you use in your provisioning of systems > * What "best practices" have you learned > * What else should I be asking for and presenting > * And of course, all those wonderful tips and tricks that we have all > learned over the years > > Anything you send may be printed in the workbook that accompanies the > tutorial, so I need your approval to print it. If you would like to be > credited, please provide me with how you want it to appear. > > Here is a brief outline that I have so far, both content AND topics are > appreciated: > > What is kickstart > Why use kickstart > Kickstart sections > boot: options > Command Section > %include is your friend > Yummy builds > %Pre > %Pre Magic > %Post > To chroot or not to chroot > %Post magic > Life after %post > Troubleshooting > Common problems > Tools to make life easier > Cobbler > Revisor > Novi > mRepo > Kickweb > On the fly ks.cfg > Other provisioning technologies > Best Practices > Resources > Thanks > > Please email me directly at chip.shabazian at bankofamerica.com > > Thanks everyone! > > Chip Shabazian > > _______________________________________________ > 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 CallahanT at tessco.com Tue Jun 19 18:31:16 2007 From: CallahanT at tessco.com (Callahan, Tom) Date: Tue, 19 Jun 2007 14:31:16 -0400 Subject: Linuxworld Kickstart Tips and Tricks In-Reply-To: Message-ID: I also like that idea, but I would warn to make sure the password sections are removed when copying to the system. A simple sed or awk one-liner should be able to accomplish this... Thanks, Tom Callahan On 6/19/07 2:29 PM, "Steden Klaus" wrote: > > > One of the things I've found with Kickstart is that the anaconda-ks.cfg > file that gets copied during last-mile installation isn't quite as > complete as I'd like as an administrator. One of the things I do in > %post is copy the working ks.cfg (found at /tmp/ks.cfg by default, I > think) to a file on disk so I have a more complete picture of the > system's install information. > > I did do some work building an on-the-fly Kickstart generator, although > there's a page on the Anaconda Wiki about it, so I'll spare everyone the > gory details. :-) > > Cheers, > Klaus > >> > -----Original Message----- >> > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- >> > bounces at redhat.com] On Behalf Of Shabazian, Chip >> > Sent: Tuesday, June 19, 2007 9:35 AM >> > To: Discussion list about Kickstart >> > Subject: Linuxworld Kickstart Tips and Tricks >> > >> > Howdy, >> > >> > As I mentioned a while back, I'm doing a "Kickstart Tips and Tricks" >> > seminar at Linuxworld this summer, and I need to finish (ok, START and >> > finish) my presentation this weekend. I've gotten a few good tips > from >> > you all, so this is my final call for help: >> > >> > I will of course make the training materials I develop available to > all >> > when it's done. >> > >> > Here is what I'm looking for: >> > * If you were attending this training, what would you like to see >> > covered >> > * Although you may be well versed in kickstart, what should I cover > for >> > those not so experienced >> > * What other technologies do you use in your provisioning of systems >> > * What "best practices" have you learned >> > * What else should I be asking for and presenting >> > * And of course, all those wonderful tips and tricks that we have all >> > learned over the years >> > >> > Anything you send may be printed in the workbook that accompanies the >> > tutorial, so I need your approval to print it. If you would like to > be >> > credited, please provide me with how you want it to appear. >> > >> > Here is a brief outline that I have so far, both content AND topics > are >> > appreciated: >> > >> > What is kickstart >> > Why use kickstart >> > Kickstart sections >> > boot: options >> > Command Section >> > %include is your friend >> > Yummy builds >> > %Pre >> > %Pre Magic >> > %Post >> > To chroot or not to chroot >> > %Post magic >> > Life after %post >> > Troubleshooting >> > Common problems >> > Tools to make life easier >> > Cobbler >> > Revisor >> > Novi >> > mRepo >> > Kickweb >> > On the fly ks.cfg >> > Other provisioning technologies >> > Best Practices >> > Resources >> > Thanks >> > >> > Please email me directly at chip.shabazian at bankofamerica.com >> > >> > Thanks everyone! >> > >> > Chip Shabazian >> > >> > _______________________________________________ >> > Kickstart-list mailing list >> > Kickstart-list at redhat.com >> > https://www.redhat.com/mailman/listinfo/kickstart-list >> > >> > _______________________________________________ >> > Kickstart-list mailing list >> > Kickstart-list at redhat.com >> > https://www.redhat.com/mailman/listinfo/kickstart-list > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > ------------------------------- Tom Callahan TDS Engineering 410-229-1361 Tel 410-588-7605 Mobile 410-229-1512 Fax callahant at TESSCO.com TESSCO Your Total Source for Wireless http://www.tessco.com Network Infrastructure Equipment | Mobile Devices & Accessories | Installation, Test, Equipment & Supplies -------------- next part -------------- An HTML attachment was scrubbed... URL: From Klaus.Steden at thomson.net Tue Jun 19 20:07:24 2007 From: Klaus.Steden at thomson.net (Steden Klaus) Date: Tue, 19 Jun 2007 16:07:24 -0400 Subject: Linuxworld Kickstart Tips and Tricks References: Message-ID: Not to belabour the point (since it's a good one), but that's kind of a waste, since the anaconda-ks.cfg that gets copied in contains a copy of the root password as well (encrypted, but still). It's saved mode 0400, but still, it seems like extra effort when the system doesn't help you out to begin with. :-) Cheers, Klaus _____ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Callahan, Tom Sent: Tuesday, June 19, 2007 11:31 AM To: Discussion list about Kickstart Subject: Re: Linuxworld Kickstart Tips and Tricks I also like that idea, but I would warn to make sure the password sections are removed when copying to the system. A simple sed or awk one-liner should be able to accomplish this... Thanks, Tom Callahan On 6/19/07 2:29 PM, "Steden Klaus" wrote: One of the things I've found with Kickstart is that the anaconda-ks.cfg file that gets copied during last-mile installation isn't quite as complete as I'd like as an administrator. One of the things I do in %post is copy the working ks.cfg (found at /tmp/ks.cfg by default, I think) to a file on disk so I have a more complete picture of the system's install information. I did do some work building an on-the-fly Kickstart generator, although there's a page on the Anaconda Wiki about it, so I'll spare everyone the gory details. :-) Cheers, Klaus > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Shabazian, Chip > Sent: Tuesday, June 19, 2007 9:35 AM > To: Discussion list about Kickstart > Subject: Linuxworld Kickstart Tips and Tricks > > Howdy, > > As I mentioned a while back, I'm doing a "Kickstart Tips and Tricks" > seminar at Linuxworld this summer, and I need to finish (ok, START and > finish) my presentation this weekend. I've gotten a few good tips from > you all, so this is my final call for help: > > I will of course make the training materials I develop available to all > when it's done. > > Here is what I'm looking for: > * If you were attending this training, what would you like to see > covered > * Although you may be well versed in kickstart, what should I cover for > those not so experienced > * What other technologies do you use in your provisioning of systems > * What "best practices" have you learned > * What else should I be asking for and presenting > * And of course, all those wonderful tips and tricks that we have all > learned over the years > > Anything you send may be printed in the workbook that accompanies the > tutorial, so I need your approval to print it. If you would like to be > credited, please provide me with how you want it to appear. > > Here is a brief outline that I have so far, both content AND topics are > appreciated: > > What is kickstart > Why use kickstart > Kickstart sections > boot: options > Command Section > %include is your friend > Yummy builds > %Pre > %Pre Magic > %Post > To chroot or not to chroot > %Post magic > Life after %post > Troubleshooting > Common problems > Tools to make life easier > Cobbler > Revisor > Novi > mRepo > Kickweb > On the fly ks.cfg > Other provisioning technologies > Best Practices > Resources > Thanks > > Please email me directly at chip.shabazian at bankofamerica.com > > Thanks everyone! > > Chip Shabazian > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list ------------------------------- Tom Callahan TDS Engineering 410-229-1361 Tel 410-588-7605 Mobile 410-229-1512 Fax callahant at TESSCO.com TESSCO Your Total Source for Wireless http://www.tessco.com Network Infrastructure Equipment | Mobile Devices & Accessories | Installation, Test, Equipment & Supplies -------------- next part -------------- An HTML attachment was scrubbed... URL: From bill at wiliweld.com Tue Jun 19 20:38:32 2007 From: bill at wiliweld.com (Bill-Schoolcraft) Date: Tue, 19 Jun 2007 13:38:32 -0700 (PDT) Subject: Linuxworld Kickstart (seminar fees?) In-Reply-To: References: Message-ID: <20070619133628.B90570@corten11-bsd.billschoolcraft.com> At Tue, 19 Jun 2007 it looks like Shabazian, Chip composed: > Howdy, > > As I mentioned a while back, I'm doing a "Kickstart Tips and Tricks" > seminar at Linuxworld this summer, and I need to finish (ok, START and > finish) my presentation this weekend. I've gotten a few good tips from > you all, so this is my final call for help: I usually get an expo pass (free) for LW, and was wondering the cost of this seminar this year? Thanks -- Bill Schoolcraft <*> http://wiliweld.com "Health nuts are going to feel stupid someday, lying in hospitals dying of nothing." -- Redd Foxx From Chip.Shabazian at bankofamerica.com Tue Jun 19 20:59:12 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Tue, 19 Jun 2007 13:59:12 -0700 Subject: Linuxworld Kickstart (seminar fees?) In-Reply-To: <20070619133628.B90570@corten11-bsd.billschoolcraft.com> Message-ID: I'm not sure of the costs, but anyone on this list can get 20% off any registration costs as one of my guests. Simply register by 8/5 and use priority code SPK20. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Bill-Schoolcraft Sent: Tuesday, June 19, 2007 1:39 PM To: Discussion list about Kickstart Subject: Re: Linuxworld Kickstart (seminar fees?) At Tue, 19 Jun 2007 it looks like Shabazian, Chip composed: > Howdy, > > As I mentioned a while back, I'm doing a "Kickstart Tips and Tricks" > seminar at Linuxworld this summer, and I need to finish (ok, START and > finish) my presentation this weekend. I've gotten a few good tips > from you all, so this is my final call for help: I usually get an expo pass (free) for LW, and was wondering the cost of this seminar this year? Thanks -- Bill Schoolcraft <*> http://wiliweld.com "Health nuts are going to feel stupid someday, lying in hospitals dying of nothing." -- Redd Foxx _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From debian at herakles.homelinux.org Tue Jun 19 22:40:26 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Wed, 20 Jun 2007 06:40:26 +0800 Subject: Linuxworld Kickstart Tips and Tricks In-Reply-To: References: Message-ID: <46785B5A.2070709@herakles.homelinux.org> Shabazian, Chip wrote: > Howdy, > > As I mentioned a while back, I'm doing a "Kickstart Tips and Tricks" > seminar at Linuxworld this summer, and I need to finish (ok, START and > finish) my presentation this weekend. I've gotten a few good tips from > you all, so this is my final call for help: > > I will of course make the training materials I develop available to all > when it's done. > > Here is what I'm looking for: > * If you were attending this training, what would you like to see > covered > * Although you may be well versed in kickstart, what should I cover for > those not so experienced Provoke the imagination. Here's a stanza from one of my PXE boot entries: label centos4k kernel /linux-install/CentOS/4.3/vmlinuz append initrd=/linux-install/CentOS/4.3/initrd.img vga=794 ramdisk_size=8192 ks=http://rhel.demo.lan/ ks/ks?OS=CentOS&REL=4&PACK=server ksdevice=link We're running a CGI script to create a ks file. Parameters such as OS, REL, PACK are interpreted by the CGI script to modify the package selection. A bit like install classes embedded in comps. > * What other technologies do you use in your provisioning of systems > * What "best practices" have you learned > * What else should I be asking for and presenting > * And of course, all those wonderful tips and tricks that we have all > learned over the years > > Anything you send may be printed in the workbook that accompanies the > tutorial, so I need your approval to print it. If you would like to be > credited, please provide me with how you want it to appear. That's cool. My actual CGI script is a motley assortment of small shell scripts with not much checking - a bit of an embarrassment really, but the main thing is the idea. > > Here is a brief outline that I have so far, both content AND topics are > appreciated: > > What is kickstart > Why use kickstart > Kickstart sections > boot: options > Command Section > %include is your friend > Yummy builds > %Pre I have been known to fdisk < Z here. > %Pre Magic > %Post > To chroot or not to chroot I regularly do both. It's a good place to run sed over a few config files where the vendor hasn't a clue. [ /etc/configfile~ ] || cp /etc/configfile /etc/configfile~ sed -r \ -e .... \ -e .... # etc The initial cp aids in testing. Reruns always have the original input. And it documents what's changed. > %Post magic > Life after %post Ah yes. I was doing the "first boot" trick at RHL 7.3, > Troubleshooting > Common problems > Tools to make life easier > Cobbler > Revisor > Novi > mRepo > Kickweb > On the fly ks.cfg Point Clark Networks (CLarkConnect) modifies Anaconda to create a ks file and then go. Tedious, I think, but that's their solution. I'd rather hack Anaconda and run it beforehand, then pick up the ks file from network, floppy or USB. For ClarkConnect, they could hardcode the floppy and USB locations and test them. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From debian at herakles.homelinux.org Tue Jun 19 22:43:00 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Wed, 20 Jun 2007 06:43:00 +0800 Subject: Linuxworld Kickstart Tips and Tricks In-Reply-To: References: Message-ID: <46785BF4.4060802@herakles.homelinux.org> Steden Klaus wrote: > One of the things I've found with Kickstart is that the anaconda-ks.cfg > file that gets copied during last-mile installation isn't quite as > complete as I'd like as an administrator. One of the things I do in > %post is copy the working ks.cfg (found at /tmp/ks.cfg by default, I > think) to a file on disk so I have a more complete picture of the > system's install information. I agree with that one. It's the only way to capture the scripts. And, once it's in /root, it's documentation what was done to the system. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From debian at herakles.homelinux.org Tue Jun 19 22:45:18 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Wed, 20 Jun 2007 06:45:18 +0800 Subject: Linuxworld Kickstart Tips and Tricks In-Reply-To: References: Message-ID: <46785C7E.60902@herakles.homelinux.org> Callahan, Tom wrote: > I also like that idea, but I would warn to make sure the password sections > are removed when copying to the system. A simple sed or awk one-liner should > be able to accomplish this... > Call me simple, but if it's in /root I don't see a problem, unless one works for a Really Paranoid Organisation, an those should be running passwd -l root asap. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From jeff.yana at rttusa.com Wed Jun 20 01:38:55 2007 From: jeff.yana at rttusa.com (Jeff Yana) Date: Tue, 19 Jun 2007 18:38:55 -0700 Subject: Kickstart Install Under Fedora Core 5 Fails Message-ID: Dear List- I have a fully functional Kickstart infrastructure in place for FC6. To date, I have performed no fewer than 25 kickstart installs (using FC6). Due to serious application compatibility issues, I am forced to take these machines back to an earlier version of Fedora ( version 5). Currently, the client systems successfully boot (using PXEboot). Some point after copying down the kickstart file, at the point where it attempts to mount the stage2.img file, it fails. I have tried using ISOs in addition to the contents of the ISOs within the installation tree, but neither work. Just to be sure that there is nothing wrong with my setup, I tried to install using the older FC6 ISOs and with those the client installs succeed. I have verified that the installation tree is available over NFS, so I am sure the problem is not with NFS. Below are the error messages I receive: VT1 "The directory does not seem to contain a Fedora Core Installation Tree." VT3 (Truncated and paraphrased, may contain errors) (.snip.) INFO: going to do nfsGetSetup INFO: mounting nfs path 192.168.X.X:/some/nfs/path WARNING: unable to access /mnt/source/images/stage2.img INFO: mntloop loop7 on /tmp/loopimage as /mnt/source/FC-5-x86_64-DVD.iso fd is 20 ERROR: open file to loop mount /tmp/loopimage/images/stage2.img failed INFO: unmounting loopback /mnt/runtime loop0 ERROR: LOOP_CLR_FD failed for /mnt/runtime loop0 (no such device or address) (.snip.) Does any one on this list have any suggestions? Thanks in advance. Jeff Yana -------------- next part -------------- An HTML attachment was scrubbed... URL: From debian at herakles.homelinux.org Wed Jun 20 02:12:12 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Wed, 20 Jun 2007 10:12:12 +0800 Subject: Kickstart Install Under Fedora Core 5 Fails In-Reply-To: References: Message-ID: <46788CFC.5050909@herakles.homelinux.org> Jeff Yana wrote: > Dear List- > > I have a fully functional Kickstart infrastructure in place for FC6. To date, I have performed no fewer than 25 kickstart installs (using FC6). Due to serious application compatibility issues, I am forced to take these machines back to an earlier version of Fedora ( version 5). Currently, the client systems successfully boot (using PXEboot). Some point after copying down the kickstart file, at the point where it attempts to mount the stage2.img file, it fails. I have tried using ISOs in addition to the contents of the ISOs within the installation tree, but neither work. Just to be sure that there is nothing wrong with my setup, I tried to install using the older FC6 ISOs and with those the client installs succeed. I have verified that the installation tree is available over NFS, so I am sure the problem is not with NFS. > > Below are the error messages I receive: > > VT1 > "The directory does not seem to contain a Fedora Core Installation Tree." > > VT3 > (Truncated and paraphrased, may contain errors) > (.snip.) > INFO: going to do nfsGetSetup > INFO: mounting nfs path 192.168.X.X:/some/nfs/path > WARNING: unable to access /mnt/source/images/stage2.img > INFO: mntloop loop7 on /tmp/loopimage as /mnt/source/FC-5-x86_64-DVD.iso fd is 20 > ERROR: open file to loop mount /tmp/loopimage/images/stage2.img failed > INFO: unmounting loopback /mnt/runtime loop0 > ERROR: LOOP_CLR_FD failed for /mnt/runtime loop0 (no such device or address) > (.snip.) > > Does any one on this list have any suggestions? > > Thanks in advance. > > Jeff Yana And the logs on ther server say? fwiw I used to use nfs, but now it's always http. The logs are better. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From Chip.Shabazian at bankofamerica.com Wed Jun 20 07:05:52 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Wed, 20 Jun 2007 00:05:52 -0700 Subject: Kickstart Install Under Fedora Core 5 Fails In-Reply-To: <46788CFC.5050909@herakles.homelinux.org> Message-ID: Not only are the logs better, but http is faster... -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of John Summerfield Sent: Tuesday, June 19, 2007 7:12 PM To: Discussion list about Kickstart Subject: Re: Kickstart Install Under Fedora Core 5 Fails Jeff Yana wrote: > Dear List- > > I have a fully functional Kickstart infrastructure in place for FC6. To date, I have performed no fewer than 25 kickstart installs (using FC6). Due to serious application compatibility issues, I am forced to take these machines back to an earlier version of Fedora ( version 5). Currently, the client systems successfully boot (using PXEboot). Some point after copying down the kickstart file, at the point where it attempts to mount the stage2.img file, it fails. I have tried using ISOs in addition to the contents of the ISOs within the installation tree, but neither work. Just to be sure that there is nothing wrong with my setup, I tried to install using the older FC6 ISOs and with those the client installs succeed. I have verified that the installation tree is available over NFS, so I am sure the problem is not with NFS. > > Below are the error messages I receive: > > VT1 > "The directory does not seem to contain a Fedora Core Installation Tree." > > VT3 > (Truncated and paraphrased, may contain errors) > (.snip.) > INFO: going to do nfsGetSetup > INFO: mounting nfs path 192.168.X.X:/some/nfs/path > WARNING: unable to access /mnt/source/images/stage2.img > INFO: mntloop loop7 on /tmp/loopimage as /mnt/source/FC-5-x86_64-DVD.iso fd is 20 > ERROR: open file to loop mount /tmp/loopimage/images/stage2.img failed > INFO: unmounting loopback /mnt/runtime loop0 > ERROR: LOOP_CLR_FD failed for /mnt/runtime loop0 (no such device or address) > (.snip.) > > Does any one on this list have any suggestions? > > Thanks in advance. > > Jeff Yana And the logs on ther server say? fwiw I used to use nfs, but now it's always http. The logs are better. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From srobson at cadence.com Wed Jun 20 08:28:24 2007 From: srobson at cadence.com (Steve Robson) Date: Wed, 20 Jun 2007 09:28:24 +0100 Subject: Linuxworld Kickstart Tips and Tricks In-Reply-To: <20070619200829.57BC07414E@hormel.redhat.com> References: <20070619200829.57BC07414E@hormel.redhat.com> Message-ID: <4678E528.7020109@cadence.com> > From: "Shabazian, Chip" > > Howdy, > > As I mentioned a while back, I'm doing a "Kickstart Tips and Tricks" > seminar at Linuxworld this summer, and I need to finish (ok, START and > finish) my presentation this weekend. I've gotten a few good tips from > you all, so this is my final call for help: > * What else should I be asking for and presenting Hi Chip, One aspect of kickstart that I've not been able to solve satisfactorily is to have the process detect existing LVM partitions, retaining all of them and reformatting some of them. I had to resort to using a ks.cfg file which had no partitioning information in it at all, which allowed anaconda to drop out to disk druid. I could then tell it what I wanted and the install then carried on. If you have a solution to this which would allow a 100% unattanded install, I for one would appreciate it. -- Regards, Steve IT Support - UNIX/Linux Cadence Design Systems Bagshot Road Bracknell BERKSHIRE RG12 0PH UK From lfarkas at bppiac.hu Thu Jun 21 09:52:18 2007 From: lfarkas at bppiac.hu (Farkas Levente) Date: Thu, 21 Jun 2007 11:52:18 +0200 Subject: driverdisk and include Message-ID: <467A4A52.3030103@bppiac.hu> hi, i'd like to put together a few kickstart file. there are common part of them like common packages and directives. so i put them in a separate file and include them. but i want to do this from a pxe boot. i already find how can i retrive include file from the %pre script to be able to include them. but my problem is that i need to add a driverdisk to each kickstart file since our server use r1000 network card which is not in the current anaconda kernel. so my question - what stage are run during a kickstart pxe boot? - what information should have to be available in each stage? - can i put the driverdisk --source=ftp://.../r1000-2.6.18_8.img line into a common included file or should it have to be in the main file? - is it documented anywhere or just i'm not able to find it? thank you for your help in advance. -- Levente "Si vis pacem para bellum!" From lfarkas at bppiac.hu Thu Jun 21 14:24:36 2007 From: lfarkas at bppiac.hu (Farkas Levente) Date: Thu, 21 Jun 2007 16:24:36 +0200 Subject: how to create initrd for pxe kickstart boot Message-ID: <467A8A24.1050805@bppiac.hu> hi, i need to create a new inird which include r1000 ethernet driver for oxe boot. i read everywhere that i've to: --------------------- mv initrd.imginitrd.img.gz gunzip initrd.img.gz mkdir init mount -o loop initrd.img init ---------------------- but these last step always gives me: mount: you must specify the filesystem type and i can't find any kind of fs type which would be good. although with gunzip initrd.img | cpio -i --make-directories seems to working, but even if i don't modify it and find . -print | cpio -o -c |gzip -9 >initrd.img.new the result is different form the original one. so what is the proper procedure to recreate the initrd file? thank you for your help in advance. -- Levente "Si vis pacem para bellum!" From nicholas.byrne at quadriga.com Thu Jun 21 16:08:35 2007 From: nicholas.byrne at quadriga.com (Nicholas Byrne) Date: Thu, 21 Jun 2007 17:08:35 +0100 Subject: how to create initrd for pxe kickstart boot In-Reply-To: <467A8A24.1050805@bppiac.hu> References: <467A8A24.1050805@bppiac.hu> Message-ID: <467AA283.7030503@quadriga.com> what distribution are you using? It might be a cpio archive or in cramfs filesystem - start here http://wiki.openvz.org/Modifying_initrd_image But once you've unzipped it, type "file init" to see what it is Farkas Levente wrote: > hi, > i need to create a new inird which include r1000 ethernet driver for oxe > boot. i read everywhere that i've to: > --------------------- > mv initrd.imginitrd.img.gz > gunzip initrd.img.gz > mkdir init > mount -o loop initrd.img init > ---------------------- > but these last step always gives me: > mount: you must specify the filesystem type > and i can't find any kind of fs type which would be good. > although with > gunzip initrd.img | cpio -i --make-directories > seems to working, but even if i don't modify it and > find . -print | cpio -o -c |gzip -9 >initrd.img.new > the result is different form the original one. > so what is the proper procedure to recreate the initrd file? > thank you for your help in advance. > > This e-mail is the property of Quadriga Worldwide Ltd, intended for the addressee only and confidential. Any dissemination, copying or distribution of this message or any attachments is strictly prohibited. If you have received this message in error, please notify us immediately by replying to the message and deleting it from your computer. Messages sent to and from Quadriga may be monitored. Quadriga cannot guarantee any message delivery method is secure or error-free. Information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. We do not accept responsibility for any errors or omissions in this message and/or attachment that arise as a result of transmission. You should carry out your own virus checks before opening any attachment. Any views or opinions presented are solely those of the author and do not necessarily represent those of Quadriga. From Chip.Shabazian at bankofamerica.com Thu Jun 21 16:37:54 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Thu, 21 Jun 2007 09:37:54 -0700 Subject: driverdisk and include In-Reply-To: <467A4A52.3030103@bppiac.hu> Message-ID: You can add the dd line to boot: line, just make sure you don't exceed 255 characters total with all your other options included. Chip Shabazian VP ; Sr. Consultant - System Engineering Bank of America E&O OS Services -- Linux e: chip.shabazian at bankofamerica.com p: 925 692-7000 -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Farkas Levente Sent: Thursday, June 21, 2007 2:52 AM To: kickstart-list at redhat.com Subject: driverdisk and include hi, i'd like to put together a few kickstart file. there are common part of them like common packages and directives. so i put them in a separate file and include them. but i want to do this from a pxe boot. i already find how can i retrive include file from the %pre script to be able to include them. but my problem is that i need to add a driverdisk to each kickstart file since our server use r1000 network card which is not in the current anaconda kernel. so my question - what stage are run during a kickstart pxe boot? - what information should have to be available in each stage? - can i put the driverdisk --source=ftp://.../r1000-2.6.18_8.img line into a common included file or should it have to be in the main file? - is it documented anywhere or just i'm not able to find it? thank you for your help in advance. -- Levente "Si vis pacem para bellum!" _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From mdehaan at redhat.com Thu Jun 21 19:40:26 2007 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 21 Jun 2007 15:40:26 -0400 Subject: how to create initrd for pxe kickstart boot In-Reply-To: <467AA283.7030503@quadriga.com> References: <467A8A24.1050805@bppiac.hu> <467AA283.7030503@quadriga.com> Message-ID: <467AD42A.6070604@redhat.com> Nicholas Byrne wrote: > what distribution are you using? It might be a cpio archive or in cramfs > filesystem - start here http://wiki.openvz.org/Modifying_initrd_image > But once you've unzipped it, type "file init" to see what it is > If you want to read source, koan (http://cobbler.et.redhat.com/) has some code to inject ks files into initrd's and covers both types already. http://git.et.redhat.com/?p=koan.git;a=summary > Farkas Levente wrote: > >> hi, >> i need to create a new inird which include r1000 ethernet driver for oxe >> boot. i read everywhere that i've to: >> --------------------- >> mv initrd.imginitrd.img.gz >> gunzip initrd.img.gz >> mkdir init >> mount -o loop initrd.img init >> ---------------------- >> but these last step always gives me: >> mount: you must specify the filesystem type >> and i can't find any kind of fs type which would be good. >> although with >> gunzip initrd.img | cpio -i --make-directories >> seems to working, but even if i don't modify it and >> find . -print | cpio -o -c |gzip -9 >initrd.img.new >> the result is different form the original one. >> so what is the proper procedure to recreate the initrd file? >> thank you for your help in advance. >> >> >> > > > > This e-mail is the property of Quadriga Worldwide Ltd, intended for the addressee only and confidential. Any dissemination, copying or distribution of this message or any attachments is strictly prohibited. > > If you have received this message in error, please notify us immediately by replying to the message and deleting it from your computer. > > Messages sent to and from Quadriga may be monitored. > > Quadriga cannot guarantee any message delivery method is secure or error-free. Information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. > > We do not accept responsibility for any errors or omissions in this message and/or attachment that arise as a result of transmission. > > You should carry out your own virus checks before opening any attachment. > > Any views or opinions presented are solely those of the author and do not necessarily represent those of Quadriga. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From mdehaan at redhat.com Thu Jun 21 19:43:01 2007 From: mdehaan at redhat.com (Michael DeHaan) Date: Thu, 21 Jun 2007 15:43:01 -0400 Subject: how to create initrd for pxe kickstart boot In-Reply-To: <467AD42A.6070604@redhat.com> References: <467A8A24.1050805@bppiac.hu> <467AA283.7030503@quadriga.com> <467AD42A.6070604@redhat.com> Message-ID: <467AD4C5.2050404@redhat.com> Michael DeHaan wrote: > Nicholas Byrne wrote: >> what distribution are you using? It might be a cpio archive or in cramfs >> filesystem - start here http://wiki.openvz.org/Modifying_initrd_image >> But once you've unzipped it, type "file init" to see what it is >> > > If you want to read source, koan (http://cobbler.et.redhat.com/) has > some code to inject ks files into initrd's and covers both types already. > > http://git.et.redhat.com/?p=koan.git;a=summary Sorry, I should be more specific. Look at line 399 http://git.et.redhat.com/?p=koan.git;a=blob;f=koan/app.py;h=b57354758d55417af9534b70f6fc58962a007137;hb=1399f4ae6dd173896923b278e1540a60293f0ef7 Note: GPL'd. From lfarkas at bppiac.hu Thu Jun 21 20:47:06 2007 From: lfarkas at bppiac.hu (Farkas Levente) Date: Thu, 21 Jun 2007 22:47:06 +0200 Subject: how to create initrd for pxe kickstart boot In-Reply-To: <467AA283.7030503@quadriga.com> References: <467A8A24.1050805@bppiac.hu> <467AA283.7030503@quadriga.com> Message-ID: <467AE3CA.9000305@bppiac.hu> Nicholas Byrne wrote: > what distribution are you using? It might be a cpio archive or in cramfs > filesystem - start here http://wiki.openvz.org/Modifying_initrd_image > But once you've unzipped it, type "file init" to see what it is centos 5.0 > Farkas Levente wrote: >> hi, >> i need to create a new inird which include r1000 ethernet driver for oxe >> boot. i read everywhere that i've to: >> --------------------- >> mv initrd.imginitrd.img.gz >> gunzip initrd.img.gz >> mkdir init >> mount -o loop initrd.img init >> ---------------------- >> but these last step always gives me: >> mount: you must specify the filesystem type >> and i can't find any kind of fs type which would be good. >> although with >> gunzip initrd.img | cpio -i --make-directories >> seems to working, but even if i don't modify it and >> find . -print | cpio -o -c |gzip -9 >initrd.img.new >> the result is different form the original one. >> so what is the proper procedure to recreate the initrd file? >> thank you for your help in advance. -- Levente "Si vis pacem para bellum!" From debian at herakles.homelinux.org Thu Jun 21 22:47:53 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Fri, 22 Jun 2007 06:47:53 +0800 Subject: driverdisk and include In-Reply-To: References: Message-ID: <467B0019.1080709@herakles.homelinux.org> Shabazian, Chip wrote: > You can add the dd line to boot: line, just make sure you don't exceed > 255 characters total with all your other options included. > > hi, > i'd like to put together a few kickstart file. there are common part of > them like common packages and directives. so i put them in a separate > file and include them. but i want to do this from a pxe boot. i already > find how can i retrive include file from the %pre script to be able to > include them. but my problem is that i need to add a driverdisk to each > kickstart file since our server use r1000 network card which is not in > the current anaconda kernel. so my question > - what stage are run during a kickstart pxe boot? > - what information should have to be available in each stage? > - can i put the > driverdisk --source=ftp://.../r1000-2.6.18_8.img > line into a common included file or should it have to be in the main > file? > - is it documented anywhere or just i'm not able to find it? > thank you for your help in advance. > I'm sure it's documented, and I'm sure I've read it. Did you read the RHEL documentation? Fedora? I prefer http installs, I think I get better logs. Check your after you do a kickstart and see what files it's looking for. It might already be looking for a driver disk. Read the recent (still-running?) hints and tricks thread for my favourite technique for creating ks files. It's another reason to prefer http. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From lfarkas at bppiac.hu Thu Jun 21 22:56:39 2007 From: lfarkas at bppiac.hu (Farkas Levente) Date: Fri, 22 Jun 2007 00:56:39 +0200 Subject: driverdisk and include In-Reply-To: References: Message-ID: <467B0227.700@bppiac.hu> but i need network card driver disk so i cant add dd=ftp:// since there is no ftp since there is no network in the kernel!!!! Shabazian, Chip wrote: > You can add the dd line to boot: line, just make sure you don't exceed > 255 characters total with all your other options included. > > > Chip Shabazian > VP ; Sr. Consultant - System Engineering > Bank of America > E&O OS Services -- Linux > e: chip.shabazian at bankofamerica.com > p: 925 692-7000 > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Farkas Levente > Sent: Thursday, June 21, 2007 2:52 AM > To: kickstart-list at redhat.com > Subject: driverdisk and include > > hi, > i'd like to put together a few kickstart file. there are common part of > them like common packages and directives. so i put them in a separate > file and include them. but i want to do this from a pxe boot. i already > find how can i retrive include file from the %pre script to be able to > include them. but my problem is that i need to add a driverdisk to each > kickstart file since our server use r1000 network card which is not in > the current anaconda kernel. so my question > - what stage are run during a kickstart pxe boot? > - what information should have to be available in each stage? > - can i put the > driverdisk --source=ftp://.../r1000-2.6.18_8.img > line into a common included file or should it have to be in the main > file? > - is it documented anywhere or just i'm not able to find it? > thank you for your help in advance. > -- Levente "Si vis pacem para bellum!" From debian at herakles.homelinux.org Thu Jun 21 23:05:16 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Fri, 22 Jun 2007 07:05:16 +0800 Subject: driverdisk and include In-Reply-To: <467B0227.700@bppiac.hu> References: <467B0227.700@bppiac.hu> Message-ID: <467B042C.3060306@herakles.homelinux.org> Farkas Levente wrote: > but i need network card driver disk so i cant add dd=ftp:// since there > is no ftp since there is no network in the kernel!!!! > > Oh. The driver disk isn't the solution to your problem. http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/sysadmin-guide/s1-kickstart2-options.html doesn't say it, but I recall reading the driver disk's intended for disk and such. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From Chip.Shabazian at bankofamerica.com Thu Jun 21 23:14:23 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Thu, 21 Jun 2007 16:14:23 -0700 Subject: driverdisk and include In-Reply-To: <467B0227.700@bppiac.hu> Message-ID: That would be a problem.... sorry. I've got the same problem busting open the pxeboot initrd. The instructions you have found work for the regular initrd from the boot.iso image. You will need to use the instructions Michael send earlier from Koan and use cpio for the pxeboot initrd. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Farkas Levente Sent: Thursday, June 21, 2007 3:57 PM To: Discussion list about Kickstart Subject: Re: driverdisk and include but i need network card driver disk so i cant add dd=ftp:// since there is no ftp since there is no network in the kernel!!!! Shabazian, Chip wrote: > You can add the dd line to boot: line, just make sure you don't exceed > 255 characters total with all your other options included. > > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Farkas Levente > Sent: Thursday, June 21, 2007 2:52 AM > To: kickstart-list at redhat.com > Subject: driverdisk and include > > hi, > i'd like to put together a few kickstart file. there are common part > of them like common packages and directives. so i put them in a > separate file and include them. but i want to do this from a pxe boot. > i already find how can i retrive include file from the %pre script to > be able to include them. but my problem is that i need to add a > driverdisk to each kickstart file since our server use r1000 network > card which is not in the current anaconda kernel. so my question > - what stage are run during a kickstart pxe boot? > - what information should have to be available in each stage? > - can i put the > driverdisk --source=ftp://.../r1000-2.6.18_8.img > line into a common included file or should it have to be in the main > file? > - is it documented anywhere or just i'm not able to find it? > thank you for your help in advance. > -- Levente "Si vis pacem para bellum!" _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From jce at zot.com Tue Jun 26 05:31:59 2007 From: jce at zot.com (Chris Edillon) Date: Tue, 26 Jun 2007 01:31:59 -0400 Subject: how to create initrd for pxe kickstart boot In-Reply-To: <467AA283.7030503@quadriga.com> References: <467A8A24.1050805@bppiac.hu> <467AA283.7030503@quadriga.com> Message-ID: <4680A4CF.1010806@zot.com> Nicholas Byrne wrote: > what distribution are you using? It might be a cpio archive or in cramfs > filesystem - start here http://wiki.openvz.org/Modifying_initrd_image unfortunately that won't help with a PXEboot initrd.img, which is very different under the hood. this may help: http://www.redhat.com/archives/kickstart-list/2007-May/msg00081.html it covers modifying an existing driver. adding a new driver won't be much different, except you'll also need proper information for the module-info and pcitable files for your new driver. it looks like you already found this link from your message: https://listman.redhat.com/archives/kickstart-list/2003-March/msg00183.html which should help with modifying the pcitable. > But once you've unzipped it, type "file init" to see what it is > nicholas is right, you should see if the initrd.img file you are starting with is a compressed filesystem image. if it's a cpio archive, you can start by unarchiving it to a directory and hacking away. if it's cramfs, you'll need to mount it, copy the contents to a writable directory (cramfs is read-only), modify, and use mkcramfs against the writable directory to create a new image. something like: mkdir /tmp/mnt mount -o loop cramfs.img /mnt rsync -avW /mnt/ /tmp/mnt mkcramfs /tmp/mnt /tmp/cramfs-new.img hope this helps, chris From lgopinat at redhat.com Tue Jun 26 13:28:36 2007 From: lgopinat at redhat.com (Liju Gopinath) Date: Tue, 26 Jun 2007 18:58:36 +0530 Subject: kickstart fails on rhel 4.3 complaining it cannot read root label Message-ID: <46811484.1030309@redhat.com> Hello, One of the customers query. kickstart fails on rhel 4.3 complaining it cannot read root label. This is on a dell 6850. Attached scsi disk sda at scsi6, channel 2, id 0, lun 0 Loading jbd.ko module Loading ext3.ko module Creating root device mkrootdev: labelKernel panic - not syncing: Attempted to kill init! / not found Mo u [c (B [?1;6;7l > [?25h [0;37;40m [2J [5;1H Press the spacebar to pause... KEY MAPPING FOR CONSOLE REDIRECTION: Use the <0> key sequence for Use the <@> key sequence for Use the key sequence for Use the key sequence for Use the key sequence for Kickstart file: --------------- install nfs --server 10.252.4.94 --dir /spare/RedHatAS4/ reboot zerombr yes lang en_US.UTF-8 langsupport --default=en_US.UTF-8 en_US.UTF-8 keyboard us # Intentionally configuring eth0 so that install will be non-interactive -CC #network --device eth0 --onboot no --bootproto dhcp #network --device eth1 --onboot no --bootproto dhcp #network --device eth2 --onboot no --bootproto dhcp #network --device eth3 --onboot no --bootproto dhcp network --device eth4 --onboot no --bootproto dhcp #network --device eth5 --onboot no --bootproto dhcp rootpw --iscrypted $1$UBaqGDAc$Uvm1lPaGmFtcbfdIYD5NI0 firewall --disabled selinux --disabled authconfig --enableshadow --enablemd5 timezone Europe/London bootloader --location=mbr clearpart --all part /boot --fstype ext3 --size=200 --asprimary part /u00 --fstype ext3 --size=40960 part /tmp --fstype ext3 --size=20480 part / --fstype ext3 --size=16384 --asprimary part /var --fstype ext3 --size=8192 part /opt --fstype ext3 --size=8192 part swap --size=4096 --asprimary part /spare --fstype ext3 --size=100 --grow %packages @ compat-arch-development @ admin-tools @ system-tools @ text-internet @ dialup @ compat-arch-support @ legacy-software-development @ mail-server @ server-cfg @ development-tools e2fsprogs kernel-smp-devel kernel-devel kernel-smp %post Thanks for any update. Liju From Matt.Fahrner at coat.com Tue Jun 26 14:58:21 2007 From: Matt.Fahrner at coat.com (Matt Fahrner) Date: Tue, 26 Jun 2007 10:58:21 -0400 Subject: Loading device firmware in kickstart Message-ID: <4681298D.2010001@coat.com> Hi all, We've built a kickstart image (initrd.img) with custom modules to support the Intel PRO/Wireless 2200BG wireless card (it uses the "ipw2200" module). As we've had experience with this before, getting the module itself to load wasn't too hard, however in order for the card to work it needs to load a firmware image as well. We believe that part is failing. What I've tried to do is put the firmware files in the same spot under the "initrd" image as on a normal multi-user system, namely "/lib/firmware". However as best as we can tell this isn't working. We are going to try "/etc/firmware" as well, however I thought someone might have had some experience with this before, so I am trying this list. Does someone have some advice here? Any help would be appreciated. Thanks, - Matt -- --------------------------------------------------------------------- Matt Fahrner 2 South Park St. Chief Systems Architect Willis House Burlington Coat Factory Warehouse Lebanon, N.H. 03766 Tel: (603) 448-4100 x5150 USA Fax: (603) 443-6190 Matt.Fahrner at COAT.COM --------------------------------------------------------------------- From Greg.Caetano at hp.com Tue Jun 26 15:24:56 2007 From: Greg.Caetano at hp.com (Caetano, Greg) Date: Tue, 26 Jun 2007 11:24:56 -0400 Subject: Loading device firmware in kickstart In-Reply-To: <4681298D.2010001@coat.com> References: <4681298D.2010001@coat.com> Message-ID: Matt: >From the ipw2200 install document: MAKE SURE THAT THE FOLLOWING CAPABILITIES ARE ENABLED: ~~~~~~~~~~~~~~~~~~~~~~~~~~~ #define CONFIG_FW_LOADER 1 ~~~~~~~~~~~~~~~~~~~~~~~~~~~ ipw2200 loads firmware via the Linux firmware hotplug capability (see later section on firmware loading). In 2.6.x, this is enabled via menuconfig: Device Drivers -> Generic Driver Options -> Hotplug firmware loading support Greg Caetano HP TSG Linux Solutions Alliances Engineering Chicago, IL greg.caetano at hp.com Red Hat Certified Engineer RHCE#803004972711193 -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Matt Fahrner Sent: Tuesday, June 26, 2007 9:58 AM To: kickstart-list at redhat.com Subject: Loading device firmware in kickstart Hi all, We've built a kickstart image (initrd.img) with custom modules to support the Intel PRO/Wireless 2200BG wireless card (it uses the "ipw2200" module). As we've had experience with this before, getting the module itself to load wasn't too hard, however in order for the card to work it needs to load a firmware image as well. We believe that part is failing. What I've tried to do is put the firmware files in the same spot under the "initrd" image as on a normal multi-user system, namely "/lib/firmware". However as best as we can tell this isn't working. We are going to try "/etc/firmware" as well, however I thought someone might have had some experience with this before, so I am trying this list. Does someone have some advice here? Any help would be appreciated. Thanks, - Matt -- --------------------------------------------------------------------- Matt Fahrner 2 South Park St. Chief Systems Architect Willis House Burlington Coat Factory Warehouse Lebanon, N.H. 03766 Tel: (603) 448-4100 x5150 USA Fax: (603) 443-6190 Matt.Fahrner at COAT.COM --------------------------------------------------------------------- _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Matt.Fahrner at coat.com Tue Jun 26 15:46:13 2007 From: Matt.Fahrner at coat.com (Matt Fahrner) Date: Tue, 26 Jun 2007 11:46:13 -0400 Subject: Loading device firmware in kickstart In-Reply-To: References: <4681298D.2010001@coat.com> Message-ID: <468134C5.9060103@coat.com> Thanks, unfortunately I was hoping to stick with a generic kernel because though I know how to build a standard kernel, I'm not sure how to build what used to be known as a "BOOT" kernel. It looks worse than this though, from what I can see I need "udev" user-space hotplug support to load firmware. That means among other things getting "udevd" to start at boot. I highly doubt any of that is encoded in the standard kickstart image and may be a bear to get in. I can't seem to find any examples of someone using "udevd" in kickstart either... Fun, fun, fun. Thanks again, - Matt Caetano, Greg wrote: > Matt: > >>From the ipw2200 install document: > > MAKE SURE THAT THE FOLLOWING CAPABILITIES ARE ENABLED: > ~~~~~~~~~~~~~~~~~~~~~~~~~~~ > #define CONFIG_FW_LOADER 1 > ~~~~~~~~~~~~~~~~~~~~~~~~~~~ > ipw2200 loads firmware via the Linux firmware hotplug capability (see > later > section on firmware loading). In 2.6.x, this is enabled via menuconfig: > > Device Drivers -> > Generic Driver Options -> > Hotplug firmware loading support > > > Greg Caetano > HP TSG Linux Solutions Alliances Engineering > Chicago, IL > greg.caetano at hp.com > Red Hat Certified Engineer > RHCE#803004972711193 > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Matt Fahrner > Sent: Tuesday, June 26, 2007 9:58 AM > To: kickstart-list at redhat.com > Subject: Loading device firmware in kickstart > > Hi all, > > We've built a kickstart image (initrd.img) with custom modules to > support the Intel PRO/Wireless 2200BG wireless card (it uses the > "ipw2200" module). As we've had experience with this before, getting the > module itself to load wasn't too hard, however in order for the card to > work it needs to load a firmware image as well. We believe that part is > failing. > > What I've tried to do is put the firmware files in the same spot under > the "initrd" image as on a normal multi-user system, namely > "/lib/firmware". However as best as we can tell this isn't working. > > We are going to try "/etc/firmware" as well, however I thought someone > might have had some experience with this before, so I am trying this > list. Does someone have some advice here? > > Any help would be appreciated. > > Thanks, > > - Matt > > -- > --------------------------------------------------------------------- > Matt Fahrner 2 South Park St. > Chief Systems Architect Willis House > Burlington Coat Factory Warehouse Lebanon, N.H. 03766 > Tel: (603) 448-4100 x5150 USA > Fax: (603) 443-6190 Matt.Fahrner at COAT.COM > --------------------------------------------------------------------- > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- --------------------------------------------------------------------- Matt Fahrner 2 South Park St. Chief Systems Architect Willis House Burlington Coat Factory Warehouse Lebanon, N.H. 03766 Tel: (603) 448-4100 x5150 USA Fax: (603) 443-6190 Matt.Fahrner at COAT.COM --------------------------------------------------------------------- From debian at herakles.homelinux.org Tue Jun 26 23:47:05 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Wed, 27 Jun 2007 07:47:05 +0800 Subject: Loading device firmware in kickstart In-Reply-To: <4681298D.2010001@coat.com> References: <4681298D.2010001@coat.com> Message-ID: <4681A579.1090508@herakles.homelinux.org> Matt Fahrner wrote: > Hi all, > > We've built a kickstart image (initrd.img) with custom modules to > support the Intel PRO/Wireless 2200BG wireless card (it uses the > "ipw2200" module). As we've had experience with this before, getting the > module itself to load wasn't too hard, however in order for the card to > work it needs to load a firmware image as well. We believe that part is > failing. > > What I've tried to do is put the firmware files in the same spot under > the "initrd" image as on a normal multi-user system, namely > "/lib/firmware". However as best as we can tell this isn't working. > > We are going to try "/etc/firmware" as well, however I thought someone > might have had some experience with this before, so I am trying this > list. Does someone have some advice here? > > Any help would be appreciated. If you want this working during the install, I am not surprised what you do doesn't work. I'm assuming you are booting local media, USB disk, CD or similar. What I would do is explore what the existing scripts do for an installed system, and write and equivalent script to run in %pre. It's a couple of years or more since I looked at this, but as I recall it's fairly simple: echo something>somehere to say you want firmware loaded. cat firmware >somewhereelse echo somethingelse >somewhere to say firmware loading's done. The trick is in finding out what to plop where. You don't really need all that general-purpose hardware detection stuff. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From r.tiwari at nihilent.com Wed Jun 27 05:47:47 2007 From: r.tiwari at nihilent.com (Rajesh Tiwari) Date: Wed, 27 Jun 2007 11:17:47 +0530 Subject: kickstart comps.xml file issue. Message-ID: <1145971D427D7F43B0DBDAE22221C4B50329A1D7@nipns00a210.fps.nihilent.com> Hi, I've created custom iso using some of the rpms provided by CentOS-4. In order to include new rpms in kickstart cfg file, i need to modify comps.xml file. I created Two new groups with my custome rpms and removed rest of the contents of comps.xml file. But it didn't work. However if i put my new group section at top of comps.xml file then it works but while installation process i see a lot of message like <> rpms not found. Can anyone please suggest me how to mdify comps.xml file or if it is now required. Thanks and Regards, Rajesh -**************Nihilent*************** " *** All information contained in this communication is confidential, proprietary, privileged and is intended for the addressees only. If you have received this E-mail in error please notify mail administrator by telephone on +91-20-39846100 or E-mail the sender by replying to this message, and then delete this E-mail and other copies of it from your computer system. Any unauthorized dissemination, publication, transfer or use of the contents of this communication, with or without modifications is punishable under the relevant law. Nihilent has scanned this mail with current virus checking technologies. However, Nihilent makes no representations or warranties to the effect that this communication is virus-free. Nihilent reserves the right to monitor all E-mail communications through its Corporate Network. *** " *************************************************************************- From Klaus.Steden at thomson.net Wed Jun 27 05:56:27 2007 From: Klaus.Steden at thomson.net (Steden Klaus) Date: Wed, 27 Jun 2007 01:56:27 -0400 Subject: kickstart comps.xml file issue. Message-ID: For CentOS 4 you also need to modify the hdlist file to complete dependency information. There is a script in python on the Fedora Wiki that will do that for you easily. Hth, Klaus ----- Original Message ----- From: kickstart-list-bounces at redhat.com To: kickstart-list at redhat.com Sent: Tue Jun 26 22:47:47 2007 Subject: kickstart comps.xml file issue. Hi, I've created custom iso using some of the rpms provided by CentOS-4. In order to include new rpms in kickstart cfg file, i need to modify comps.xml file. I created Two new groups with my custome rpms and removed rest of the contents of comps.xml file. But it didn't work. However if i put my new group section at top of comps.xml file then it works but while installation process i see a lot of message like <> rpms not found. Can anyone please suggest me how to mdify comps.xml file or if it is now required. Thanks and Regards, Rajesh -**************Nihilent*************** " *** All information contained in this communication is confidential, proprietary, privileged and is intended for the addressees only. If you have received this E-mail in error please notify mail administrator by telephone on +91-20-39846100 or E-mail the sender by replying to this message, and then delete this E-mail and other copies of it from your computer system. Any unauthorized dissemination, publication, transfer or use of the contents of this communication, with or without modifications is punishable under the relevant law. Nihilent has scanned this mail with current virus checking technologies. However, Nihilent makes no representations or warranties to the effect that this communication is virus-free. Nihilent reserves the right to monitor all E-mail communications through its Corporate Network. *** " *************************************************************************- _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From muksyed at stanford.edu Wed Jun 27 06:56:21 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Tue, 26 Jun 2007 23:56:21 -0700 Subject: Kickstart RAID1 help Message-ID: <007e01c7b888$4565c8f0$cd1c42ab@stanford.edu> Hi I am trying to get Kickstart working with RAID1. The regular Kickstart works fine (without RAID1). Googl'ing for these errors didn't help. Any help is appreciated. Thanks much. -Mukarram Syed. This is my RAID1 setup: #Partition clearing information clearpart --all --initlabel ############################################################# # RAID1 Partitioning: part raid.01 --size 100 --ondisk sda part raid.02 --size 100 --ondisk sdb # / part raid.03 --size 25000 --ondisk sda part raid.04 --size 25000 --ondisk sdb # Swap part raid.05 --recommended --ondisk sda part raid.06 --recommended --ondisk sdb # /u01 part raid.07 --size 18000 --ondisk sda part raid.08 --size 18000 --ondisk sdb # /x01 part raid.09 --size 4000 --ondisk sda part raid.10 --size 4000 --ondisk sdb # /s01 part raid.11 --size 15000 --ondisk sda part raid.12 --size 15000 --ondisk sdb #Disk partitioning information raid /boot --fstype ext3 --level 1 --device md0 raid.01 raid.02 --asprimary raid / --fstype ext3 --level 1 --device md1 raid.03 raid.04 --asprimary raid swap --fstype swap --level 1 --device md2 raid.05 raid.06 --asprimary raid /u01 --fstype ext3 --level 1 --device md4 raid.07 raid.08 raid /x01 --fstype ext3 --level 1 --device md5 raid.09 raid.10 raid /s01 --fstype ext3 --level 1 --device md6 raid.11 raid.12 ############################################################# This is the error I am getting on the console: Probing for monitor type: Unknown monitor Probing for mouse type: No - mouse Graphical installation not available... Starting text mode. Traceback (most recent call last): File "/usr/bin/anaconda", line 1130, in ? instClass.setInstallData(id) File "/usr/lib/anaconda/kickstart.py", line 1439, in setInstallData raise KickstartError, e kickstart.KickstartValueError: partition command requires a size specification install exited abnormally sending termination signals...done sending kill signals...done disabling swap... unmounting filesystems... /mnt/runtime done disabling /dev/loop0 /proc/bus/usb done /proc done /dev/pts done /sys done /tmp/ramfs done /selinux done you may safely reboot your system -------------- next part -------------- An HTML attachment was scrubbed... URL: From srobson at cadence.com Wed Jun 27 08:24:36 2007 From: srobson at cadence.com (Steve Robson) Date: Wed, 27 Jun 2007 09:24:36 +0100 Subject: Kickstart RAID1 help In-Reply-To: <20070627065635.54E3973BFC@hormel.redhat.com> References: <20070627065635.54E3973BFC@hormel.redhat.com> Message-ID: <46821EC4.6050604@cadence.com> > Subject: Kickstart RAID1 help > From: "Mukarram Syed" > Date: Tue, 26 Jun 2007 23:56:21 -0700 > To: > > To: > Hi > > I am trying to get Kickstart working with RAID1. The regular Kickstart > works fine (without RAID1). Googl?ing for these errors didn?t help. > > Any help is appreciated. > > Thanks much. > -Mukarram Syed. > > This is my RAID1 setup: > > #Partition clearing information > > clearpart --all --initlabel > ############################################################# > > # RAID1 Partitioning: > part raid.01 --size 100 --ondisk sda > part raid.02 --size 100 --ondisk sdb > # / > part raid.03 --size 25000 --ondisk sda > part raid.04 --size 25000 --ondisk sdb > # Swap > part raid.05 --recommended --ondisk sda > part raid.06 --recommended --ondisk sdb > # /u01 > part raid.07 --size 18000 --ondisk sda > part raid.08 --size 18000 --ondisk sdb > # /x01 > part raid.09 --size 4000 --ondisk sda > part raid.10 --size 4000 --ondisk sdb > # /s01 > part raid.11 --size 15000 --ondisk sda > part raid.12 --size 15000 --ondisk sdb > #Disk partitioning information > raid /boot --fstype ext3 --level 1 --device md0 raid.01 raid.02 --asprimary > raid / --fstype ext3 --level 1 --device md1 raid.03 raid.04 --asprimary > raid swap --fstype swap --level 1 --device md2 raid.05 raid.06 --asprimary > raid /u01 --fstype ext3 --level 1 --device md4 raid.07 raid.08 > raid /x01 --fstype ext3 --level 1 --device md5 raid.09 raid.10 > raid /s01 --fstype ext3 --level 1 --device md6 raid.11 raid.12 > ############################################################# > > This is the error I am getting on the console: > Probing for monitor type: Unknown monitor > Probing for mouse type: No - mouse > Graphical installation not available... Starting text mode. > Traceback (most recent call last): > *File "/usr/bin/anaconda", line 1130, in ?* > * instClass.setInstallData(id)* > * File "/usr/lib/anaconda/kickstart.py", line 1439, in setInstallData* > * raise KickstartError, e* > *kickstart.KickstartValueError: partition command requires a size > specification* > *install exited abnormally* > > sending termination signals...done > sending kill signals...done > disabling swap... > unmounting filesystems... > /mnt/runtime done > disabling /dev/loop0 > /proc/bus/usb done > /proc done > /dev/pts done > /sys done > /tmp/ramfs done > /selinux done > > you may safely reboot your system I think it's a syntax snafu. The sysadmin guide at http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/sysadmin-guide/s1-kickstart2-options.html shows that where you had: raid /s01 --fstype ext3 --level 1 --device md6 raid.11 raid.12 you want: raid /s01 --fstype ext3 --level=1 --device=md6 raid.11 raid.12 -- Regards, Steve IT Support - UNIX/Linux Cadence Design Systems Bagshot Road Bracknell BERKSHIRE RG12 0PH UK From email at jasonkohles.com Wed Jun 27 13:33:00 2007 From: email at jasonkohles.com (Jason Kohles) Date: Wed, 27 Jun 2007 09:33:00 -0400 Subject: Kickstart RAID1 help In-Reply-To: <007e01c7b888$4565c8f0$cd1c42ab@stanford.edu> References: <007e01c7b888$4565c8f0$cd1c42ab@stanford.edu> Message-ID: On Jun 27, 2007, at 2:56 AM, Mukarram Syed wrote: > Hi > > I am trying to get Kickstart working with RAID1. The regular > Kickstart works fine (without RAID1). > > Googl?ing for these errors didn?t help. > > You can't do this: > # Swap > > part raid.05 --recommended --ondisk sda > > part raid.06 --recommended --ondisk sdb > > --recommended only works when the mountpoint is 'swap', it won't work with raid mountpoints... In general I wouldn't recommend building a raid for swap anyway, it doesn't gain you much and the performance overhead for RAID will make a difference in this situation, since you want your memory to be as fast as possible. What I would do instead is... part swap --recommended --ondisk sda part swap --recommended --ondisk sdb This will give you twice as much swap without the performance overhead, and if one disk fails you can just 'swapoff' that partition and keep running with just one swap partition. -- Jason Kohles email at jasonkohles.com http://www.jasonkohles.com/ "A witty saying proves nothing." -- Voltaire From briandlong at gmail.com Wed Jun 27 13:38:59 2007 From: briandlong at gmail.com (Brian Long) Date: Wed, 27 Jun 2007 09:38:59 -0400 Subject: kickstart comps.xml file issue. In-Reply-To: <1145971D427D7F43B0DBDAE22221C4B50329A1D7@nipns00a210.fps.nihilent.com> References: <1145971D427D7F43B0DBDAE22221C4B50329A1D7@nipns00a210.fps.nihilent.com> Message-ID: <1c59d60b0706270638x1906ecb1nb5b1d8918676de18@mail.gmail.com> On 6/27/07, Rajesh Tiwari wrote: > Can anyone please suggest me how to mdify comps.xml file or if it is now required. Rajesh, Two groups, core and base, are required by Anaconda. If you don't define them, Anaconda will barf. Also, core and base should contain a minimal working system with closed dependencies. If you start removing RPMs from core and base, make sure they're not needed by anything. If you don't care about LSB compliance, removing the redhat-lsb RPM and anything you don't need which is required by redhat-lsb, that's a good start to making a smaller distro. /Brian/ From Matt.Fahrner at coat.com Wed Jun 27 14:25:46 2007 From: Matt.Fahrner at coat.com (Matt Fahrner) Date: Wed, 27 Jun 2007 10:25:46 -0400 Subject: Loading device firmware in kickstart In-Reply-To: <4681A579.1090508@herakles.homelinux.org> References: <4681298D.2010001@coat.com> <4681A579.1090508@herakles.homelinux.org> Message-ID: <4682736A.5080804@coat.com> John Summerfield wrote: > If you want this working during the install, I am not surprised what you > do doesn't work. Me neither really, but you gotta try the easy way first... > I'm assuming you are booting local media, USB disk, CD or similar. Actually a hard drive partition. Basically we have a "maintenance" partition that we place on each system that can be booted off to re-kickstart the system. This allows us both to reinstall a scragged system, but also do major upgrades that are too complicated to deal with at a multi-user level in an unattended environment. The later need, which is almost never used, will probably be mitigated somewhat by Fedora's capabilities of running OS version upgrades via yum. > What I would do is explore what the existing scripts do for an installed > system, and write and equivalent script to run in %pre. That's a good idea - thank you. Don't know how easy it would be to get the "udevd" junk to work there, but it's worth a try. > It's a couple of years or more since I looked at this, but as I recall > it's fairly simple: > echo something>somehere to say you want firmware loaded. > cat firmware >somewhereelse > echo somethingelse >somewhere to say firmware loading's done. Or try that, which apparently is what the "udevd" stuff ultimately does from my reading. > The trick is in finding out what to plop where. You don't really need > all that general-purpose hardware detection stuff. True, very true, though we have a (relatively small) variety of hardware we have to support so we do need some auto-detection. Since this seems like a bit of work, and Atheros cards are easy to come by, I'm going to punt and try using an Atheros based cards (which are an alternative we have to the Intel 2200) and inject "madwifi" drivers into the "initrd". The hard part there will be getting the modules to compile for the "2.6.18-1.2798.fc6" kernel, that or redo the "modules.cgz" with all new modules for a later kernel. Unfortunately I haven't done this for a while and FC6 is a bit of a learning curve. It used to be easier because you could just use the "BOOT" kernel config - I have no idea where this is pulled no (it appears perhaps to be a generic i586 kernel in "boot.iso" - at least generic i586 modules seem to work with it for versioning). Thanks for your help and good advice John, - Matt -- --------------------------------------------------------------------- Matt Fahrner 2 South Park St. Chief Systems Architect Willis House Burlington Coat Factory Warehouse Lebanon, N.H. 03766 Tel: (603) 448-4100 x5150 USA Fax: (603) 443-6190 Matt.Fahrner at COAT.COM --------------------------------------------------------------------- From muksyed at stanford.edu Wed Jun 27 15:39:25 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Wed, 27 Jun 2007 08:39:25 -0700 Subject: Kickstart RAID1 help In-Reply-To: Message-ID: <00a301c7b8d1$58219520$cd1c42ab@stanford.edu> Thanks Jason. This and also Steve pointed out a syntax error. A combination of both worked. Regards -Mukarram. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Jason Kohles Sent: Wednesday, June 27, 2007 6:33 AM To: Discussion list about Kickstart Subject: Re: Kickstart RAID1 help On Jun 27, 2007, at 2:56 AM, Mukarram Syed wrote: > Hi > > I am trying to get Kickstart working with RAID1. The regular > Kickstart works fine (without RAID1). > > Googl'ing for these errors didn't help. > > You can't do this: > # Swap > > part raid.05 --recommended --ondisk sda > > part raid.06 --recommended --ondisk sdb > > --recommended only works when the mountpoint is 'swap', it won't work with raid mountpoints... In general I wouldn't recommend building a raid for swap anyway, it doesn't gain you much and the performance overhead for RAID will make a difference in this situation, since you want your memory to be as fast as possible. What I would do instead is... part swap --recommended --ondisk sda part swap --recommended --ondisk sdb This will give you twice as much swap without the performance overhead, and if one disk fails you can just 'swapoff' that partition and keep running with just one swap partition. -- Jason Kohles email at jasonkohles.com http://www.jasonkohles.com/ "A witty saying proves nothing." -- Voltaire _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From muksyed at stanford.edu Wed Jun 27 15:39:47 2007 From: muksyed at stanford.edu (Mukarram Syed) Date: Wed, 27 Jun 2007 08:39:47 -0700 Subject: Kickstart RAID1 help In-Reply-To: <46821EC4.6050604@cadence.com> Message-ID: <00a401c7b8d1$6503c510$cd1c42ab@stanford.edu> Thanks Steve. It worked. -Mukarram. -----Original Message----- From: Steve Robson [mailto:srobson at cadence.com] Sent: Wednesday, June 27, 2007 1:25 AM To: kickstart-list at redhat.com Cc: muksyed at stanford.edu Subject: Re: Kickstart RAID1 help > Subject: Kickstart RAID1 help > From: "Mukarram Syed" > Date: Tue, 26 Jun 2007 23:56:21 -0700 > To: > > To: > Hi > > I am trying to get Kickstart working with RAID1. The regular Kickstart > works fine (without RAID1). Googl'ing for these errors didn't help. > > Any help is appreciated. > > Thanks much. > -Mukarram Syed. > > This is my RAID1 setup: > > #Partition clearing information > > clearpart --all --initlabel > ############################################################# > > # RAID1 Partitioning: > part raid.01 --size 100 --ondisk sda > part raid.02 --size 100 --ondisk sdb > # / > part raid.03 --size 25000 --ondisk sda > part raid.04 --size 25000 --ondisk sdb > # Swap > part raid.05 --recommended --ondisk sda > part raid.06 --recommended --ondisk sdb > # /u01 > part raid.07 --size 18000 --ondisk sda > part raid.08 --size 18000 --ondisk sdb > # /x01 > part raid.09 --size 4000 --ondisk sda > part raid.10 --size 4000 --ondisk sdb > # /s01 > part raid.11 --size 15000 --ondisk sda > part raid.12 --size 15000 --ondisk sdb > #Disk partitioning information > raid /boot --fstype ext3 --level 1 --device md0 raid.01 raid.02 --asprimary > raid / --fstype ext3 --level 1 --device md1 raid.03 raid.04 --asprimary > raid swap --fstype swap --level 1 --device md2 raid.05 raid.06 --asprimary > raid /u01 --fstype ext3 --level 1 --device md4 raid.07 raid.08 > raid /x01 --fstype ext3 --level 1 --device md5 raid.09 raid.10 > raid /s01 --fstype ext3 --level 1 --device md6 raid.11 raid.12 > ############################################################# > > This is the error I am getting on the console: > Probing for monitor type: Unknown monitor > Probing for mouse type: No - mouse > Graphical installation not available... Starting text mode. > Traceback (most recent call last): > *File "/usr/bin/anaconda", line 1130, in ?* > * instClass.setInstallData(id)* > * File "/usr/lib/anaconda/kickstart.py", line 1439, in setInstallData* > * raise KickstartError, e* > *kickstart.KickstartValueError: partition command requires a size > specification* > *install exited abnormally* > > sending termination signals...done > sending kill signals...done > disabling swap... > unmounting filesystems... > /mnt/runtime done > disabling /dev/loop0 > /proc/bus/usb done > /proc done > /dev/pts done > /sys done > /tmp/ramfs done > /selinux done > > you may safely reboot your system I think it's a syntax snafu. The sysadmin guide at http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/sysadmin-guide/s 1-kickstart2-options.html shows that where you had: raid /s01 --fstype ext3 --level 1 --device md6 raid.11 raid.12 you want: raid /s01 --fstype ext3 --level=1 --device=md6 raid.11 raid.12 -- Regards, Steve IT Support - UNIX/Linux Cadence Design Systems Bagshot Road Bracknell BERKSHIRE RG12 0PH UK From debian at herakles.homelinux.org Wed Jun 27 23:46:43 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Thu, 28 Jun 2007 07:46:43 +0800 Subject: Loading device firmware in kickstart In-Reply-To: <4682736A.5080804@coat.com> References: <4681298D.2010001@coat.com> <4681A579.1090508@herakles.homelinux.org> <4682736A.5080804@coat.com> Message-ID: <4682F6E3.7090201@herakles.homelinux.org> Matt Fahrner wrote: > John Summerfield wrote: >> If you want this working during the install, I am not surprised what >> you do doesn't work. > > Me neither really, but you gotta try the easy way first... > >> I'm assuming you are booting local media, USB disk, CD or similar. > > Actually a hard drive partition. Basically we have a "maintenance" > partition that we place on each system that can be booted off to > re-kickstart the system. This allows us both to reinstall a scragged > system, but also do major upgrades that are too complicated to deal with > at a multi-user level in an unattended environment. > > The later need, which is almost never used, will probably be mitigated > somewhat by Fedora's capabilities of running OS version upgrades via yum. > >> What I would do is explore what the existing scripts do for an >> installed system, and write and equivalent script to run in %pre. > > That's a good idea - thank you. Don't know how easy it would be to get > the "udevd" junk to work there, but it's worth a try. NO NO. I didn't say to get it working, that's a lot of bother. But to replace it with something simplistic, which is what I propose, you need to know what it does. > >> It's a couple of years or more since I looked at this, but as I recall >> it's fairly simple: >> echo something>somehere to say you want firmware loaded. >> cat firmware >somewhereelse >> echo somethingelse >somewhere to say firmware loading's done. > > Or try that, which apparently is what the "udevd" stuff ultimately does > from my reading. Exactly. > >> The trick is in finding out what to plop where. You don't really need >> all that general-purpose hardware detection stuff. > > True, very true, though we have a (relatively small) variety of hardware > we have to support so we do need some auto-detection. Probably, loading firmware for something that's not there will simply fail. > > Since this seems like a bit of work, and Atheros cards are easy to come > by, I'm going to punt and try using an Atheros based cards (which are an > alternative we have to the Intel 2200) and inject "madwifi" drivers into > the "initrd". The hard part there will be getting the modules to compile > for the "2.6.18-1.2798.fc6" kernel, that or redo the "modules.cgz" with > all new modules for a later kernel. Bear in mind that Fedora doesn't support atheros (or didn't last time I looked). The HAL stuff isn't exactly open source. For those wanting enterprise-grade Linux, look at Scientific Linux. > > Unfortunately I haven't done this for a while and FC6 is a bit of a > learning curve. It used to be easier because you could just use the > "BOOT" kernel config - I have no idea where this is pulled no (it > appears perhaps to be a generic i586 kernel in "boot.iso" - at least > generic i586 modules seem to work with it for versioning). > > Thanks for your help and good advice John, Also remember FC6 has a short life. If you particularly like that level of software, look at Scientific Linux which is built from RHEL (like CentOS), but has some non-free stuff added. Like proper java, and (I'm fairly sure) Madwifi. If I were installing new Fedora I'd look at Fedora 7. > > - Matt -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From alessiof at gmail.com Thu Jun 28 08:17:51 2007 From: alessiof at gmail.com (Alessio Focardi) Date: Thu, 28 Jun 2007 10:17:51 +0200 Subject: Repack a FC7 install to include an unsupported controller Message-ID: <8bc972b70706280117n658549d2ib62d9af2e4138d5b@mail.gmail.com> Hi, we normally use a kickstarted FC7 installation to install customers servers, recently we tried this method with a "mini pc" that has a VIA chipset (VIA CX700M) that doesn't seem to be supported: no hard disk is detected while installing. I suppose that the initrd of the install ISO should be updated, but it's not clear to me how this can be accomplished. Particullary my first question is about what's actually included in the standard initrd so I can generate a new one with the same stuff plus the VIA driver. Also: adding the driver to the install initrd will solve the problem or should I also package a new kernel rpm or else ? sorry if my request is off topic, and tnx for any help! Alessio Focardi Interconnessioni Italy -------------- next part -------------- An HTML attachment was scrubbed... URL: From debian at herakles.homelinux.org Thu Jun 28 23:17:13 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Fri, 29 Jun 2007 07:17:13 +0800 Subject: Repack a FC7 install to include an unsupported controller In-Reply-To: <8bc972b70706280117n658549d2ib62d9af2e4138d5b@mail.gmail.com> References: <8bc972b70706280117n658549d2ib62d9af2e4138d5b@mail.gmail.com> Message-ID: <46844179.20102@herakles.homelinux.org> Alessio Focardi wrote: > Hi, > > we normally use a kickstarted FC7 installation to install customers > servers, > recently we tried this method with a "mini pc" that has a VIA chipset (VIA > CX700M) that doesn't seem to be supported: no hard disk is detected while > installing. > > I suppose that the initrd of the install ISO should be updated, but it's > not > clear to me how this can be accomplished. I've never needed to do such, but I'd have thought that if Fedora 7 (note it-s _not_ Fedora Core) is going to run on it without extra drivers then it would install without this problem. It might just be that some fiddling with PCI tables to match driver with chipset is required. I suggest getting to VC2 when the install fails and try using modprobe to install likely drivers. via82cxxx is a candidate. You should, of course, confirm that the IDE controllers are actually turned on in the BIOS. You might also try the latest Knoppix CD: it seems to run on pretty much any x86 hardware. And also the latest Fedora development (aka rawhide) install media. > > Particullary my first question is about what's actually included in the > standard initrd so I can generate a new one with the same stuff plus the > VIA > driver. There's a fair bit out about inspecting the initrd. Google knows. > -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From jce at zot.com Fri Jun 29 07:29:06 2007 From: jce at zot.com (Chris Edillon) Date: Fri, 29 Jun 2007 03:29:06 -0400 Subject: Loading device firmware in kickstart In-Reply-To: <4682736A.5080804@coat.com> References: <4681298D.2010001@coat.com> <4681A579.1090508@herakles.homelinux.org> <4682736A.5080804@coat.com> Message-ID: <4684B4C2.2080408@zot.com> Matt Fahrner wrote: > Unfortunately I haven't done this for a while and FC6 is a bit of a > learning curve. It used to be easier because you could just use the > "BOOT" kernel config - I have no idea where this is pulled no (it > appears perhaps to be a generic i586 kernel in "boot.iso" - at least > generic i586 modules seem to work with it for versioning). > i believe that the BOOT kernel was used in RHEL3 and prior, and that starting with RHEL4 the pxeboot kernel is the standard RHEL4 single-processor kernel. chris