From info at hostinthebox.net Mon May 2 21:22:20 2005 From: info at hostinthebox.net (dan) Date: Mon, 02 May 2005 14:22:20 -0700 Subject: Overriding the default install of @Core and @Base Message-ID: <42769A0C.5040605@hostinthebox.net> Hello, all - I had a question regarding the, from what I can see, forced install of the @Core and @Base groups. According to http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/custom-guide/s1-kickstart2-packageselection.html: "In most cases, it is only necessary to list the desired groups and not individual packages. Note that the Core and Base groups are always selected by default, so it is not necessary to specify them in the %packages section." That's not what I want. I'm a bit picky about my conf files, and want to take the files listed as "installed" in my /root/install.log file, and place them in my kickstart %packages section. There are some items in the @Base and @Core groups that I do not want/need installed, so I'd rather specify item by item explicitly what I WANT to install, not what I DON'T want to install. I want to make this an "inclusive" install, rather than an "exclusive" install, if that makes sense. If anyone here has battled with trying to make the perfect install situation, you know what I'm talking about ;) Any help regarding this would be greatly appreciated. Thanks! -dant From James.Oden at tekelec.com Mon May 2 21:25:13 2005 From: James.Oden at tekelec.com (Oden, James) Date: Mon, 2 May 2005 16:25:13 -0500 Subject: Overriding the default install of @Core and @Base Message-ID: <58292FA6B3EEFD49AEDAF6597E21E717010BED8E@DCEVS2.tekelec.com> > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of dan > Sent: Monday, May 02, 2005 5:22 PM > To: Discussion list about Kickstart > Subject: Overriding the default install of @Core and @Base > > Hello, all - > > I had a question regarding the, from what I can see, forced > install of > the @Core and @Base groups. > > According to > http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/custom-g > uide/s1-kickstart2-packageselection.html: > > "In most cases, it is only necessary to list the desired > groups and not > individual packages. Note that the Core and Base groups are always > selected by default, so it is not necessary to specify them in the > %packages section." > > That's not what I want. I rebuild the comps.xml file and put in @Base and @Core the packages I want. At that point my %packages section in the kickstart file is empty. Cheers...james From skvidal at phy.duke.edu Mon May 2 21:26:34 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Mon, 02 May 2005 17:26:34 -0400 Subject: Overriding the default install of @Core and @Base In-Reply-To: <42769A0C.5040605@hostinthebox.net> References: <42769A0C.5040605@hostinthebox.net> Message-ID: <1115069195.7569.19.camel@cutter> > If anyone here has battled with trying to make the perfect install > situation, you know what I'm talking about ;) Any help regarding this > would be greatly appreciated. > And has been answered here on MANY MANY occasions you can either: 1. rebuild the installer with a new comps.xml to your specifications (harder than you think it might be, esp to get the deps right) 2. remove the packages you don't want in %post using something that can automatically remove deps - like yum, apt, etc. -sv From info at hostinthebox.net Mon May 2 21:39:00 2005 From: info at hostinthebox.net (dan) Date: Mon, 02 May 2005 14:39:00 -0700 Subject: Overriding the default install of @Core and @Base (replying to list this time) In-Reply-To: <1115069195.7569.19.camel@cutter> References: <42769A0C.5040605@hostinthebox.net> <1115069195.7569.19.camel@cutter> Message-ID: <42769DF4.5070404@hostinthebox.net> seth vidal wrote: >>If anyone here has battled with trying to make the perfect install >>situation, you know what I'm talking about ;) Any help regarding this >>would be greatly appreciated. >> > > > And has been answered here on MANY MANY occasions you can either: > 1. rebuild the installer with a new comps.xml to your specifications > (harder than you think it might be, esp to get the deps right) > > 2. remove the packages you don't want in %post using something that can > automatically remove deps - like yum, apt, etc. > > -sv > I know, it does make me a terrible person for askin again. My apologies. Thanks for the tip, Oden and Seth. I'll be sure to do this. Thanks -dant From cmadams at hiwaay.net Mon May 2 21:41:11 2005 From: cmadams at hiwaay.net (Chris Adams) Date: Mon, 2 May 2005 16:41:11 -0500 Subject: Overriding the default install of @Core and @Base In-Reply-To: <1115069195.7569.19.camel@cutter> References: <42769A0C.5040605@hostinthebox.net> <1115069195.7569.19.camel@cutter> Message-ID: <20050502214111.GB1294729@hiwaay.net> Once upon a time, seth vidal said: > > If anyone here has battled with trying to make the perfect install > > situation, you know what I'm talking about ;) Any help regarding this > > would be greatly appreciated. > > And has been answered here on MANY MANY occasions you can either: 0. remove the packages you don't want in the %packages section with "-packagename" (e.g. "-system-config-mouse"). > 1. rebuild the installer with a new comps.xml to your specifications > (harder than you think it might be, esp to get the deps right) > > 2. remove the packages you don't want in %post using something that can > automatically remove deps - like yum, apt, etc. -- Chris Adams Systems and Network Administrator - HiWAAY Internet Services I don't speak for anybody but myself - that's enough trouble. From info at hostinthebox.net Mon May 2 21:44:08 2005 From: info at hostinthebox.net (dan) Date: Mon, 02 May 2005 14:44:08 -0700 Subject: Overriding the default install of @Core and @Base In-Reply-To: <20050502214111.GB1294729@hiwaay.net> References: <42769A0C.5040605@hostinthebox.net> <1115069195.7569.19.camel@cutter> <20050502214111.GB1294729@hiwaay.net> Message-ID: <42769F28.90003@hostinthebox.net> Chris Adams wrote: > Once upon a time, seth vidal said: > >>>If anyone here has battled with trying to make the perfect install >>>situation, you know what I'm talking about ;) Any help regarding this >>>would be greatly appreciated. >> >>And has been answered here on MANY MANY occasions you can either: > > > 0. remove the packages you don't want in the %packages section with > "-packagename" (e.g. "-system-config-mouse"). > > >>1. rebuild the installer with a new comps.xml to your specifications >>(harder than you think it might be, esp to get the deps right) >> >>2. remove the packages you don't want in %post using something that can >>automatically remove deps - like yum, apt, etc. > > Yes, I understand that much, but thanks. What I was initially trying to do was override the installation of @Base and @Core, not to selectively remove packages. But as James had first pointed out, it might be in my better interest to hack up a comps.xml and use that. This sounds like the cleanest approach. Thanks for all the feedback! -dant From marcr at ifae.es Tue May 3 09:38:12 2005 From: marcr at ifae.es (Marc Rodriguez) Date: Tue, 03 May 2005 11:38:12 +0200 Subject: build a new kernel In-Reply-To: <42769DF4.5070404@hostinthebox.net> References: <42769A0C.5040605@hostinthebox.net> <1115069195.7569.19.camel@cutter> <42769DF4.5070404@hostinthebox.net> Message-ID: <42774684.9070300@ifae.es> Hi Anybody know how to make a kernel for kickstart -- .----. _.'__ `. .--(.)(.) /#\ .' @ /###\ : , ##### `-..__.-' _.-\###/ Marc Rodriguez marc.rodriguez at pic.es Port d informacio Cientifica# Campus UAB (Edifici D) ## 08193 Bellaterra # (Barcelona) SPAIN Tel: +34 93 5813322 Fax: +34 93 5814110 ########## ########### ############################# From info at hostinthebox.net Tue May 3 18:37:47 2005 From: info at hostinthebox.net (dan) Date: Tue, 03 May 2005 11:37:47 -0700 Subject: Verbose Install log Message-ID: <4277C4FB.40004@hostinthebox.net> Hello, all - I was looking for a way to make an even more verbose install log than /root/install.log. That file is a good resource, but I want to know if the RPM installations encounter any errors or anything like that. I've noticed that one of my kickstart installs was so simplistic, other things started to break due to missing packages and such. I'd like to try to troubleshoot this from the inside out, and a very verbose install-time log would be wonderful. Thanks! -dant From info at hostinthebox.net Tue May 3 19:04:27 2005 From: info at hostinthebox.net (dan) Date: Tue, 03 May 2005 12:04:27 -0700 Subject: build a new kernel In-Reply-To: <42774684.9070300@ifae.es> References: <42769A0C.5040605@hostinthebox.net> <1115069195.7569.19.camel@cutter> <42769DF4.5070404@hostinthebox.net> <42774684.9070300@ifae.es> Message-ID: <4277CB3B.9070702@hostinthebox.net> Marc Rodriguez wrote: > Hi > > Anybody know how to make a kernel for kickstart > I'm not sure that the kernel has as much to do with a kickstart than Anaconda and friends do. The kernel does not preform the kickstart install. Perhaps Anaconda is what you're looking for some information on? Thanks -dant From regatta at gmail.com Tue May 3 22:44:12 2005 From: regatta at gmail.com (Regatta) Date: Wed, 04 May 2005 01:44:12 +0300 Subject: kickstart, nfs, CD boot Message-ID: <1115160252.26047.26.camel@localhost> Guys, Did anyway try to boot from a CD and then let anaconda get the kickstart from nfs server ? for RHEL 3 it's always work and we don't see any problem, but for RHEL 4 it never work (maybe I'm wrong, but I need someone to tell me) another question, does anyone know about any GUI that let you: 1. add your kickstart path and installation method on it as a profile 2. Add any machine information in the program then make the program easy boot the machine and kickstart it using PXE boot and dhcp ? From rreed at ops.sgp.arm.gov Tue May 3 20:50:00 2005 From: rreed at ops.sgp.arm.gov (Ronald Reed) Date: Tue, 03 May 2005 15:50:00 -0500 Subject: Device Parameter in Kickstart File In-Reply-To: References: <1114707471.2619.23.camel@ronnote> <42712472.8080802@hostinthebox.net> <1114711771.2619.35.camel@ronnote> Message-ID: <1115153400.2632.67.camel@ronnote> This is what I have found so far: If you need a device parameter in your kickstart file, and you need more than 1 module loaded, then separate them with a colon ":". Example: device scsi aacraid:megaraid:aic7xxx This should be documented somewhere, because I have searched for days and found nothing that mentioned this. I had to dig thru the anaconda source code to figure this out. Hopefully this post will help someone else too. Ron On Fri, 2005-04-29 at 20:58, Dan Carpenter wrote: > On 4/28/05, Ronald Reed wrote: > > Anyone know what section of the anaconda code read the device parameter > > so that I can find it in the source code? > > I'm new to this as well. Check out anaconda CVS. > cd anaconda/loader2/ > look for loadKickstartModule in kickstart.c and modules.c > > I don't completely understand you're setup, but maybe you could try > getting access to the install media and doing all the rest in the %pre and > %post scripts. I've never had to do this myself... > > regards, > dan > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- =========================== Ron Reed RedHat Certified Engineer SGP Computer Department Manager Unix Systems Administrator ARM SGP CART Site (580)388-4053 ron.reed at arm.gov From vectorz2 at gmail.com Wed May 4 00:24:57 2005 From: vectorz2 at gmail.com (John Tran) Date: Tue, 3 May 2005 17:24:57 -0700 Subject: Kickstart without DHCP server Message-ID: Hey everyone. Is there a way to do kickstart if I don't have a dhcp server on the network? I heard booting this may work, but it doesn't for me (from the boot prompt of the cd): linux ks=nfs:192.168.11.4:/kickstart/kick.cfg ksdevice=eth0 ip=192.168.11.5 netmask=255.255.255.0 gateway=192.168.11.1 Doesn't work, I was wondering if anyone else has any info to help me out. From regatta at gmail.com Wed May 4 05:59:59 2005 From: regatta at gmail.com (Regatta) Date: Wed, 04 May 2005 08:59:59 +0300 Subject: Kickstart without DHCP server In-Reply-To: References: Message-ID: <1115186399.26047.32.camel@localhost> Yes, It's work for Me, just remember to also put the dns= and make sure you but the correct IP On Tue, 2005-05-03 at 17:24 -0700, John Tran wrote: > Hey everyone. Is there a way to do kickstart if I don't have a dhcp > server on the network? > > I heard booting this may work, but it doesn't for me (from the boot > prompt of the cd): > > linux ks=nfs:192.168.11.4:/kickstart/kick.cfg ksdevice=eth0 > ip=192.168.11.5 netmask=255.255.255.0 gateway=192.168.11.1 > > Doesn't work, I was wondering if anyone else has any info to help me out. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From drkludge at cox.net Thu May 5 00:46:52 2005 From: drkludge at cox.net (Greg Morgan) Date: Wed, 04 May 2005 17:46:52 -0700 Subject: Device Parameter in Kickstart File In-Reply-To: <1115153400.2632.67.camel@ronnote> References: <1114707471.2619.23.camel@ronnote> <42712472.8080802@hostinthebox.net> <1114711771.2619.35.camel@ronnote> <1115153400.2632.67.camel@ronnote> Message-ID: <42796CFC.4090409@cox.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ronald Reed wrote: > This is what I have found so far: > > If you need a device parameter in your kickstart file, and you need more > than 1 module loaded, then separate them with a colon ":". Example: > > device scsi aacraid:megaraid:aic7xxx http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/custom-guide/s1-kickstart2-options.html I guess you have found an undocumented feature because it only talks about one driver here without the : separation syntax. This of course would lead to trying several device lines. At most anaconda may only be expecting a scsi and a eth line. Perhaps this was added after 9? I'll make a note of it for later. > > This should be documented somewhere, because I have searched for days > and found nothing that mentioned this. I had to dig thru the anaconda > source code to figure this out. Hopefully this post will help someone > else too. > > Ron > > On Fri, 2005-04-29 at 20:58, Dan Carpenter wrote: > >>On 4/28/05, Ronald Reed wrote: >> >>>Anyone know what section of the anaconda code read the device parameter >>>so that I can find it in the source code? >> >>I'm new to this as well. Check out anaconda CVS. >>cd anaconda/loader2/ >>look for loadKickstartModule in kickstart.c and modules.c >> >>I don't completely understand you're setup, but maybe you could try >>getting access to the install media and doing all the rest in the %pre and >>%post scripts. I've never had to do this myself... >> >>regards, >>dan -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFCeWz8xyxe5L6mr7IRArGwAJ0Rpgbid9S5vHWTSZ25WjcHp8rnBwCfUr8z A6njzjVVBs/tkGP7lRQ27rA= =2RXn -----END PGP SIGNATURE----- From drkludge at cox.net Thu May 5 00:52:47 2005 From: drkludge at cox.net (Greg Morgan) Date: Wed, 04 May 2005 17:52:47 -0700 Subject: kickstart, nfs, CD boot In-Reply-To: <1115160252.26047.26.camel@localhost> References: <1115160252.26047.26.camel@localhost> Message-ID: <42796E5F.2030103@cox.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Regatta wrote: > Guys, > > Did anyway try to boot from a CD and then let anaconda get the kickstart > from nfs server ? > > for RHEL 3 it's always work and we don't see any problem, but for RHEL 4 > it never work (maybe I'm wrong, but I need someone to tell me) Could your NFS server be a v2 NFS server? I am not sure if there is a place to say in Anaconda nfsvers=2. This idea may be a place to look. Greg -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFCeW5fxyxe5L6mr7IRAj0kAKCBZ8bCY4CYYT2OohCU9fx5RGYktQCgjUYx t4jtxJmtehJaW0TO5OJs4JU= =CAI4 -----END PGP SIGNATURE----- From rreed at ops.sgp.arm.gov Thu May 5 16:22:04 2005 From: rreed at ops.sgp.arm.gov (Ronald Reed) Date: Thu, 05 May 2005 11:22:04 -0500 Subject: Device Parameter in Kickstart File In-Reply-To: <42796CFC.4090409@cox.net> References: <1114707471.2619.23.camel@ronnote> <42712472.8080802@hostinthebox.net> <1114711771.2619.35.camel@ronnote> <1115153400.2632.67.camel@ronnote> <42796CFC.4090409@cox.net> Message-ID: <1115310124.2623.11.camel@ronnote> I tried multiple device lines in the kickstart, but anaconda would only read the first one. I don't know what would happen if one of the modules needed extra options. Ron On Wed, 2005-05-04 at 19:46, Greg Morgan wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Ronald Reed wrote: > > This is what I have found so far: > > > > If you need a device parameter in your kickstart file, and you need more > > than 1 module loaded, then separate them with a colon ":". Example: > > > > device scsi aacraid:megaraid:aic7xxx > > http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/custom-guide/s1-kickstart2-options.html > > I guess you have found an undocumented feature because it only talks > about one driver here without the : separation syntax. This of course > would lead to trying several device lines. At most anaconda may only be > expecting a scsi and a eth line. Perhaps this was added after 9? I'll > make a note of it for later. > > > > > This should be documented somewhere, because I have searched for days > > and found nothing that mentioned this. I had to dig thru the anaconda > > source code to figure this out. Hopefully this post will help someone > > else too. > > > > Ron > > > > On Fri, 2005-04-29 at 20:58, Dan Carpenter wrote: > > > >>On 4/28/05, Ronald Reed wrote: > >> > >>>Anyone know what section of the anaconda code read the device parameter > >>>so that I can find it in the source code? > >> > >>I'm new to this as well. Check out anaconda CVS. > >>cd anaconda/loader2/ > >>look for loadKickstartModule in kickstart.c and modules.c > >> > >>I don't completely understand you're setup, but maybe you could try > >>getting access to the install media and doing all the rest in the %pre and > >>%post scripts. I've never had to do this myself... > >> > >>regards, > >>dan > > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.2.4 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > > iD8DBQFCeWz8xyxe5L6mr7IRArGwAJ0Rpgbid9S5vHWTSZ25WjcHp8rnBwCfUr8z > A6njzjVVBs/tkGP7lRQ27rA= > =2RXn > -----END PGP SIGNATURE----- > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- =========================== Ron Reed RedHat Certified Engineer SGP Computer Department Manager Unix Systems Administrator ARM SGP CART Site (580)388-4053 ron.reed at arm.gov From llwebb at gotadsl.co.uk Thu May 5 16:40:36 2005 From: llwebb at gotadsl.co.uk (Lloyd) Date: Thu, 05 May 2005 17:40:36 +0100 Subject: PRE or POST NET config Message-ID: <427A4C84.2090204@gotadsl.co.uk> Hi, Would anyone care to share a script for a PRE or POST network configuration? I would like to prompt the person installing the system for their hostname, IP, netmask, default route and DNS settings if possible. Netconfig gets all screwed up on screen for some strange reason and my scripting skills are pretty low. Many thanks LW From jkeating at j2solutions.net Thu May 5 16:16:04 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Thu, 05 May 2005 09:16:04 -0700 Subject: PRE or POST NET config In-Reply-To: <427A4C84.2090204@gotadsl.co.uk> References: <427A4C84.2090204@gotadsl.co.uk> Message-ID: <1115309764.6029.7.camel@jkeating2.hq.pogolinux.com> On Thu, 2005-05-05 at 17:40 +0100, Lloyd wrote: > Hi, > > Would anyone care to share a script for a PRE or POST network > configuration? I would like to prompt the person installing the > system > for their hostname, IP, netmask, default route and DNS settings if > possible. Netconfig gets all screwed up on screen for some strange > reason and my scripting skills are pretty low. > Why not use .unconfigured for the first time the system boots up it will ask that prior to trying to start the network. -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From llwebb at gotadsl.co.uk Thu May 5 18:53:33 2005 From: llwebb at gotadsl.co.uk (Lloyd) Date: Thu, 05 May 2005 19:53:33 +0100 Subject: PRE or POST NET config In-Reply-To: <1115309764.6029.7.camel@jkeating2.hq.pogolinux.com> References: <427A4C84.2090204@gotadsl.co.uk> <1115309764.6029.7.camel@jkeating2.hq.pogolinux.com> Message-ID: <427A6BAD.10400@gotadsl.co.uk> Jesse Keating wrote: >On Thu, 2005-05-05 at 17:40 +0100, Lloyd wrote: > > >>Hi, >> >>Would anyone care to share a script for a PRE or POST network >>configuration? I would like to prompt the person installing the >>system >>for their hostname, IP, netmask, default route and DNS settings if >>possible. Netconfig gets all screwed up on screen for some strange >>reason and my scripting skills are pretty low. >> >> >> > >Why not use .unconfigured for the first time the system boots up it will >ask that prior to trying to start the network. > > > Hi Jesse, Sorry do you have an example? I havent come across this before. Cheers From teng at dataway.com Thu May 5 18:58:51 2005 From: teng at dataway.com (Tedman Eng) Date: Thu, 5 May 2005 11:58:51 -0700 Subject: PRE or POST NET config Message-ID: <37ED92F9890FAF4BB947613C66FF8B1AFC74E2@dw-mail.dataway.com> man sys-unconfig -----Original Message----- From: Lloyd [mailto:llwebb at gotadsl.co.uk] Sent: Thursday, May 05, 2005 11:54 AM To: Discussion list about Kickstart Subject: Re: PRE or POST NET config Jesse Keating wrote: >On Thu, 2005-05-05 at 17:40 +0100, Lloyd wrote: > > >>Hi, >> >>Would anyone care to share a script for a PRE or POST network >>configuration? I would like to prompt the person installing the >>system >>for their hostname, IP, netmask, default route and DNS settings if >>possible. Netconfig gets all screwed up on screen for some strange >>reason and my scripting skills are pretty low. >> >> >> > >Why not use .unconfigured for the first time the system boots up it will >ask that prior to trying to start the network. > > > Hi Jesse, Sorry do you have an example? I havent come across this before. Cheers _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From jkeating at j2solutions.net Thu May 5 19:25:46 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Thu, 05 May 2005 12:25:46 -0700 Subject: PRE or POST NET config In-Reply-To: <427A6BAD.10400@gotadsl.co.uk> References: <427A4C84.2090204@gotadsl.co.uk> <1115309764.6029.7.camel@jkeating2.hq.pogolinux.com> <427A6BAD.10400@gotadsl.co.uk> Message-ID: <1115321146.6029.19.camel@jkeating2.hq.pogolinux.com> On Thu, 2005-05-05 at 19:53 +0100, Lloyd wrote: > Hi Jesse, > > Sorry do you have an example? I havent come across this before. Take a peek at /etc/rc.sysinit It looks for a file on the root file system named '.unconfigured' and will run a bunch of config programs to setup the system per a user's desires. Great for OEM systems. -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From vectorz2 at gmail.com Fri May 6 16:03:08 2005 From: vectorz2 at gmail.com (John Tran) Date: Fri, 6 May 2005 09:03:08 -0700 Subject: How to add raid and nic driver to Kickstart beginning Message-ID: I read that I'm supposed to pass a 'dd' command at the linux boot prompt in order to add in 3rd party raid and nic drivers that aren't supported by the kernel. I am kickstarting RH Enterprise 3v1 and it doesn't support my Dell 220 Raid. So how do I go about automating it so that I don't have to insert a driver disk at the beginning of the kickstart for the raid and nic drivers? From rreed at ops.sgp.arm.gov Fri May 6 16:06:02 2005 From: rreed at ops.sgp.arm.gov (Ronald Reed) Date: Fri, 06 May 2005 11:06:02 -0500 Subject: How to add raid and nic driver to Kickstart beginning In-Reply-To: References: Message-ID: <1115395562.2623.25.camel@ronnote> What drivers do you need? The isolinux boot on the RHE CD should have most of the drivers. Ron On Fri, 2005-05-06 at 11:03, John Tran wrote: > I read that I'm supposed to pass a 'dd' command at the linux boot > prompt in order to add in 3rd party raid and nic drivers that aren't > supported by the kernel. > > I am kickstarting RH Enterprise 3v1 and it doesn't support my Dell 220 > Raid. So how do I go about automating it so that I don't have to > insert a driver disk at the beginning of the kickstart for the raid > and nic drivers? > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- =========================== Ron Reed RedHat Certified Engineer SGP Computer Department Manager Unix Systems Administrator ARM SGP CART Site (580)388-4053 ron.reed at arm.gov From dan at half-asleep.com Fri May 6 16:06:55 2005 From: dan at half-asleep.com (Daniel Segall) Date: Fri, 6 May 2005 12:06:55 -0400 (EDT) Subject: How to add raid and nic driver to Kickstart beginning In-Reply-To: References: Message-ID: <52581.192.80.55.73.1115395615.squirrel@webmail.half-asleep.com> I'd recommend trying RHES3 U4. I know that prior to U3 some of my Dell's SCSI drivers were not present yet. Just a thought that might save you the extra work of adding your drivers to the image. -Dan > I read that I'm supposed to pass a 'dd' command at the linux boot > prompt in order to add in 3rd party raid and nic drivers that aren't > supported by the kernel. > > I am kickstarting RH Enterprise 3v1 and it doesn't support my Dell 220 > Raid. So how do I go about automating it so that I don't have to > insert a driver disk at the beginning of the kickstart for the raid > and nic drivers? > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > From vectorz2 at gmail.com Fri May 6 16:19:04 2005 From: vectorz2 at gmail.com (John Tran) Date: Fri, 6 May 2005 09:19:04 -0700 Subject: How to add raid and nic driver to Kickstart beginning In-Reply-To: <52581.192.80.55.73.1115395615.squirrel@webmail.half-asleep.com> References: <52581.192.80.55.73.1115395615.squirrel@webmail.half-asleep.com> Message-ID: Yes, definitely. We didn't have any issues w/ U4, but the problem is the Dynamo application that we're using at this time they do not support it running on U4. The highest revision they support is U1. :( On 5/6/05, Daniel Segall wrote: > I'd recommend trying RHES3 U4. I know that prior to U3 some of my Dell's > SCSI drivers were not present yet. Just a thought that might save you the > extra work of adding your drivers to the image. > > -Dan > > > > I read that I'm supposed to pass a 'dd' command at the linux boot > > prompt in order to add in 3rd party raid and nic drivers that aren't > > supported by the kernel. > > > > I am kickstarting RH Enterprise 3v1 and it doesn't support my Dell 220 > > Raid. So how do I go about automating it so that I don't have to > > insert a driver disk at the beginning of the kickstart for the raid > > and nic drivers? > > > > _______________________________________________ > > 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 llwebb at gotadsl.co.uk Fri May 6 16:51:42 2005 From: llwebb at gotadsl.co.uk (Lloyd) Date: Fri, 06 May 2005 17:51:42 +0100 Subject: PRE or POST NET config In-Reply-To: <1115321146.6029.19.camel@jkeating2.hq.pogolinux.com> References: <427A4C84.2090204@gotadsl.co.uk> <1115309764.6029.7.camel@jkeating2.hq.pogolinux.com> <427A6BAD.10400@gotadsl.co.uk> <1115321146.6029.19.camel@jkeating2.hq.pogolinux.com> Message-ID: <427BA09E.8020202@gotadsl.co.uk> Jesse Keating wrote: >On Thu, 2005-05-05 at 19:53 +0100, Lloyd wrote: > > >>Hi Jesse, >> >>Sorry do you have an example? I havent come across this before. >> >> > >Take a peek at /etc/rc.sysinit It looks for a file on the root file >system named '.unconfigured' and will run a bunch of config programs to >setup the system per a user's desires. Great for OEM systems. > > > Tried it, it works a treat. Thanks v much From andrew.w.robinson at mms.gov Fri May 6 17:25:37 2005 From: andrew.w.robinson at mms.gov (Robinson, Andrew W.) Date: Fri, 6 May 2005 11:25:37 -0600 Subject: OT: pxe server - dhcp proxy question Message-ID: <379313C94B76D2119AB60008C7A402E409E463F4@IMSNOLAA> I am exploring the use of the proxy-dhcp server. The one that comes with Red Hat Enterprise Linux 3 Update 4 is pxe-0.1-36.i386.rpm. With a pxe server without the dhcp proxy, when a client pxe-boots, it displays the menu specified in the /tftpboot/linux-install/msgs/boot.msg file. When I install the dhcp proxy (pxe-0.1-36.i386.rpm), when the client pxe-boots, the messages displayed are those specified in the /etc/pxe.conf file. I would like to get the dhcp proxy to use the messages in the boot.msg file. I assume there is some way to specify that in the pxe.conf file. However, I do not understand the dhcp proxy or the pxe.conf file to know how to do that. Plus the documentation seems sparse. Can someone tell me how to accomplish my goal? Better yet, can someone direct me to a forum where I may be enlightened in this and other pxe-boot questions? Thanks! Andrew Robinson From dan at half-asleep.com Fri May 6 19:17:12 2005 From: dan at half-asleep.com (Daniel Segall) Date: Fri, 6 May 2005 15:17:12 -0400 (EDT) Subject: OT: pxe server - dhcp proxy question In-Reply-To: <379313C94B76D2119AB60008C7A402E409E463F4@IMSNOLAA> References: <379313C94B76D2119AB60008C7A402E409E463F4@IMSNOLAA> Message-ID: <56852.192.80.55.74.1115407032.squirrel@webmail.half-asleep.com> The redhat manual goes into a bit of detail on setting up a pxe/tftp/dhcp kickstart environment. Chapters 9, 10, 14, & 15 should help you out. It can be found at https://www.redhat.com/docs/manuals/enterprise/RHEL-3-Manual/sysadmin-guide/index.html -Dan > I am exploring the use of the proxy-dhcp server. The one that comes with > Red > Hat Enterprise Linux 3 Update 4 is pxe-0.1-36.i386.rpm. With a pxe server > without the dhcp proxy, when a client pxe-boots, it displays the menu > specified in the /tftpboot/linux-install/msgs/boot.msg file. When I > install > the dhcp proxy (pxe-0.1-36.i386.rpm), when the client pxe-boots, the > messages displayed are those specified in the /etc/pxe.conf file. I would > like to get the dhcp proxy to use the messages in the boot.msg file. I > assume there is some way to specify that in the pxe.conf file. However, I > do > not understand the dhcp proxy or the pxe.conf file to know how to do that. > Plus the documentation seems sparse. Can someone tell me how to accomplish > my goal? Better yet, can someone direct me to a forum where I may be > enlightened in this and other pxe-boot questions? > > Thanks! > > Andrew Robinson > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > From andrew.w.robinson at mms.gov Fri May 6 19:38:45 2005 From: andrew.w.robinson at mms.gov (Robinson, Andrew W.) Date: Fri, 6 May 2005 13:38:45 -0600 Subject: OT: pxe server - dhcp proxy question Message-ID: <379313C94B76D2119AB60008C7A402E409E463F5@IMSNOLAA> The redhat manual explains how to set up a pxe-boot environment _without_ a dhcp proxy. I'm trying to go the next step and add the dhcp proxy (btw, using the pxe-0.1-36.i386.rpm package provided by redhat. Just to make things interesting, the dchp proxy server is named pxe.) I'm looking for help in configuring the dhcp proxy server. Andrew > From: Daniel Segall [mailto:dan at half-asleep.com] > > The redhat manual goes into a bit of detail on setting up a > pxe/tftp/dhcp > kickstart environment. Chapters 9, 10, 14, & 15 should help > you out. It > can be found at > https://www.redhat.com/docs/manuals/enterprise/RHEL-3-Manual/s > ysadmin-guide/index.html > > -Dan > > > I am exploring the use of the proxy-dhcp server. The one > that comes with > > Red > > Hat Enterprise Linux 3 Update 4 is pxe-0.1-36.i386.rpm. > With a pxe server > > without the dhcp proxy, when a client pxe-boots, it > displays the menu > > specified in the /tftpboot/linux-install/msgs/boot.msg file. When I > > install > > the dhcp proxy (pxe-0.1-36.i386.rpm), when the client pxe-boots, the > > messages displayed are those specified in the /etc/pxe.conf > file. I would > > like to get the dhcp proxy to use the messages in the > boot.msg file. I > > assume there is some way to specify that in the pxe.conf > file. However, I > > do > > not understand the dhcp proxy or the pxe.conf file to know > how to do that. > > Plus the documentation seems sparse. Can someone tell me > how to accomplish > > my goal? Better yet, can someone direct me to a forum where I may be > > enlightened in this and other pxe-boot questions? > > > > Thanks! > > > > Andrew Robinson > > > > _______________________________________________ > > 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 dan at half-asleep.com Fri May 6 19:52:36 2005 From: dan at half-asleep.com (Daniel Segall) Date: Fri, 6 May 2005 15:52:36 -0400 (EDT) Subject: OT: pxe server - dhcp proxy question In-Reply-To: <379313C94B76D2119AB60008C7A402E409E463F5@IMSNOLAA> References: <379313C94B76D2119AB60008C7A402E409E463F5@IMSNOLAA> Message-ID: <36817.192.80.55.74.1115409156.squirrel@webmail.half-asleep.com> Here's how I do it... dhcpd.conf: ################################### ddns-update-style interim; ignore client-updates; subnet 10.1.1.0 netmask 255.255.255.0 { # --- default gateway option routers 10.1.1.1; option subnet-mask 255.255.255.0; # option nis-domain "domain.org"; # option domain-name "domain.org"; # option domain-name-servers 192.168.1.1; # option time-offset -18000; # Eastern Standard Time # option ntp-servers 192.168.1.1; # option netbios-name-servers 192.168.1.1; # --- Selects point-to-point node (default is hybrid). Don't change this unless # -- you understand Netbios very well # option netbios-node-type 2; range dynamic-bootp 10.1.1.200 10.1.1.250; default-lease-time 21600; max-lease-time 43200; # we want the nameserver to appear at a fixed address # host ns { # next-server marvin.redhat.com; # hardware ethernet 12:34:56:78:AB:CD; # fixed-address 207.175.42.254; # } filename "pxelinux.0"; next-server 10.1.1.1; } In the pxe.conf: ################################### [OurIpAddress] 10.1.1.1 [Mtftpd_Root_Directory] /tftpboot [UseDHCPPort] 0 [StartProxy] 1 [startBootService] 1 [MasterProxy] 1 [Discovery_BCast_Disabled] 0 [Discovery_MCast_Disabled] 0 [Prompt] 10,Press F8 to view menu ... /tftpboot/pxelinux.cfg/default ################################### # Default PXE Config # Local boot by default default local # Always prompt user prompt 1 # Display our boot message display pxeboot.msg # Boot automatically after 10 minute timeout 6000 label local localboot 0 label Rescue kernel vmlinuz-RHES-3 append initrd=initrd-RHES-3.img rescue ks=nfs:10.1.1.1:/kickstart/rescue.ks.cfg text devfs=nomount ramdisk=10240 vga=0xF01 label RHES-3-U4 kernel vmlinuz-RHES-3-U4 append initrd=initrd-RHES-3-U4.img ks=nfs:10.1.1.1:/kickstart/RHES-3-U4-ks.cfg label FC-3 kernel vmlinuz-FC-3 append initrd=initrd-FC-3.img ks=nfs:10.1.1.1:/kickstart/FC-3-ks.cfg ################################### That's really about it. -Dan > The redhat manual explains how to set up a pxe-boot environment _without_ > a > dhcp proxy. I'm trying to go the next step and add the dhcp proxy (btw, > using the pxe-0.1-36.i386.rpm package provided by redhat. Just to make > things interesting, the dchp proxy server is named pxe.) I'm looking for > help in configuring the dhcp proxy server. > > Andrew > >> From: Daniel Segall [mailto:dan at half-asleep.com] >> >> The redhat manual goes into a bit of detail on setting up a >> pxe/tftp/dhcp >> kickstart environment. Chapters 9, 10, 14, & 15 should help >> you out. It >> can be found at >> https://www.redhat.com/docs/manuals/enterprise/RHEL-3-Manual/s >> ysadmin-guide/index.html >> >> -Dan >> >> > I am exploring the use of the proxy-dhcp server. The one >> that comes with >> > Red >> > Hat Enterprise Linux 3 Update 4 is pxe-0.1-36.i386.rpm. >> With a pxe server >> > without the dhcp proxy, when a client pxe-boots, it >> displays the menu >> > specified in the /tftpboot/linux-install/msgs/boot.msg file. When I >> > install >> > the dhcp proxy (pxe-0.1-36.i386.rpm), when the client pxe-boots, the >> > messages displayed are those specified in the /etc/pxe.conf >> file. I would >> > like to get the dhcp proxy to use the messages in the >> boot.msg file. I >> > assume there is some way to specify that in the pxe.conf >> file. However, I >> > do >> > not understand the dhcp proxy or the pxe.conf file to know >> how to do that. >> > Plus the documentation seems sparse. Can someone tell me >> how to accomplish >> > my goal? Better yet, can someone direct me to a forum where I may be >> > enlightened in this and other pxe-boot questions? >> > >> > Thanks! >> > >> > Andrew Robinson >> > >> > _______________________________________________ >> > 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 > > From teng at dataway.com Fri May 6 21:31:35 2005 From: teng at dataway.com (Tedman Eng) Date: Fri, 6 May 2005 14:31:35 -0700 Subject: How to add raid and nic driver to Kickstart beginning Message-ID: <37ED92F9890FAF4BB947613C66FF8B1AFC74E6@dw-mail.dataway.com> I had the exact same issue (Dell Poweredge 2850), vendor support required RHEL3u1. You can load the driver disk from within the kickstart file (even via network! YAY!!) >From /usr/share/doc/anaconda-9.1.1/kickstart-docs.txt ----- driverdisk (optional) Driver diskettes can be used during kickstart installations. You need to copy the driver diskettes's contents to the root directory of a partition on the system's hard drive. Then you need to use the driverdisk command to tell the installation program where to look for the driver disk. driverdisk [--type=] Alternatively, a network location can be specified for the driver diskette: driverdisk --source=ftp://path/to/dd.img driverdisk --source=http://path/to/dd.img driverdisk --source=nfs:host:/path/to/img Partition containing the driver disk. --type= File system type (for example, vfat or ext2). ----- I believe the downloadable drivers from Dell come as dd images already, and instructs you to use dd or rawwrite to create the floppy. If you have the driver on floppy, use dd in reverse to create an image file from the floppy. > -----Original Message----- > From: John Tran [mailto:vectorz2 at gmail.com] > Sent: Friday, May 06, 2005 9:19 AM > To: Discussion list about Kickstart > Subject: Re: How to add raid and nic driver to Kickstart beginning > > > Yes, definitely. We didn't have any issues w/ U4, but the problem is > the Dynamo application that we're using at this time they do not > support it running on U4. The highest revision they support is U1. > :( > > On 5/6/05, Daniel Segall wrote: > > I'd recommend trying RHES3 U4. I know that prior to U3 some > of my Dell's > > SCSI drivers were not present yet. Just a thought that > might save you the > > extra work of adding your drivers to the image. > > > > -Dan > > > > > > > I read that I'm supposed to pass a 'dd' command at the linux boot > > > prompt in order to add in 3rd party raid and nic drivers > that aren't > > > supported by the kernel. > > > > > > I am kickstarting RH Enterprise 3v1 and it doesn't > support my Dell 220 > > > Raid. So how do I go about automating it so that I don't have to > > > insert a driver disk at the beginning of the kickstart > for the raid > > > and nic drivers? > > > > > > _______________________________________________ > > > 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 > From andrew.w.robinson at mms.gov Fri May 6 22:22:29 2005 From: andrew.w.robinson at mms.gov (Robinson, Andrew W.) Date: Fri, 6 May 2005 16:22:29 -0600 Subject: OT: pxe server - dhcp proxy question Message-ID: <379313C94B76D2119AB60008C7A402E409E463F7@IMSNOLAA> Dan, what's the rest of your /etc/pxe.conf file look like? For what you showed, mine is the same as yours. However, when I try to pxe boot, the server tries to serve /tftpboot/X86PC/UNDI/linux-install/linux.0 instead of /tftpboot/linux-install/pxelinux.0. Thanks! Andrew > From: Daniel Segall [mailto:dan at half-asleep.com] > > Here's how I do it... > > dhcpd.conf: > ################################### > > ddns-update-style interim; > ignore client-updates; > > subnet 10.1.1.0 netmask 255.255.255.0 { > > # --- default gateway > option routers 10.1.1.1; > option subnet-mask 255.255.255.0; > > # option nis-domain "domain.org"; > # option domain-name "domain.org"; > # option domain-name-servers 192.168.1.1; > > # option time-offset -18000; # Eastern > Standard Time > # option ntp-servers 192.168.1.1; > # option netbios-name-servers 192.168.1.1; > # --- Selects point-to-point node (default is hybrid). Don't > change this > unless > # -- you understand Netbios very well > # option netbios-node-type 2; > > range dynamic-bootp 10.1.1.200 10.1.1.250; > default-lease-time 21600; > max-lease-time 43200; > > # we want the nameserver to appear at a fixed address > # host ns { > # next-server marvin.redhat.com; > # hardware ethernet 12:34:56:78:AB:CD; > # fixed-address 207.175.42.254; > # } > filename "pxelinux.0"; > next-server 10.1.1.1; > } > > In the pxe.conf: > ################################### > > [OurIpAddress] > 10.1.1.1 > > [Mtftpd_Root_Directory] > /tftpboot > > [UseDHCPPort] > 0 > > [StartProxy] > 1 > > [startBootService] > 1 > > [MasterProxy] > 1 > > [Discovery_BCast_Disabled] > 0 > > [Discovery_MCast_Disabled] > 0 > > [Prompt] > 10,Press F8 to view menu ... > > > /tftpboot/pxelinux.cfg/default > ################################### > # Default PXE Config > > # Local boot by default > default local > > # Always prompt user > prompt 1 > > # Display our boot message > display pxeboot.msg > > # Boot automatically after 10 minute > timeout 6000 > > label local > localboot 0 > label Rescue > kernel vmlinuz-RHES-3 > append initrd=initrd-RHES-3.img rescue > ks=nfs:10.1.1.1:/kickstart/rescue.ks.cfg text devfs=nomount > ramdisk=10240 vga=0xF01 > label RHES-3-U4 > kernel vmlinuz-RHES-3-U4 > append initrd=initrd-RHES-3-U4.img > ks=nfs:10.1.1.1:/kickstart/RHES-3-U4-ks.cfg > label FC-3 > kernel vmlinuz-FC-3 > append initrd=initrd-FC-3.img > ks=nfs:10.1.1.1:/kickstart/FC-3-ks.cfg > > ################################### > > > That's really about it. > > -Dan > > > > > The redhat manual explains how to set up a pxe-boot > environment _without_ > > a > > dhcp proxy. I'm trying to go the next step and add the dhcp > proxy (btw, > > using the pxe-0.1-36.i386.rpm package provided by redhat. > Just to make > > things interesting, the dchp proxy server is named pxe.) > I'm looking for > > help in configuring the dhcp proxy server. > > > > Andrew > > > >> From: Daniel Segall [mailto:dan at half-asleep.com] > >> > >> The redhat manual goes into a bit of detail on setting up a > >> pxe/tftp/dhcp > >> kickstart environment. Chapters 9, 10, 14, & 15 should help > >> you out. It > >> can be found at > >> https://www.redhat.com/docs/manuals/enterprise/RHEL-3-Manual/s > >> ysadmin-guide/index.html > >> > >> -Dan > >> > >> > I am exploring the use of the proxy-dhcp server. The one > >> that comes with > >> > Red > >> > Hat Enterprise Linux 3 Update 4 is pxe-0.1-36.i386.rpm. > >> With a pxe server > >> > without the dhcp proxy, when a client pxe-boots, it > >> displays the menu > >> > specified in the /tftpboot/linux-install/msgs/boot.msg > file. When I > >> > install > >> > the dhcp proxy (pxe-0.1-36.i386.rpm), when the client > pxe-boots, the > >> > messages displayed are those specified in the /etc/pxe.conf > >> file. I would > >> > like to get the dhcp proxy to use the messages in the > >> boot.msg file. I > >> > assume there is some way to specify that in the pxe.conf > >> file. However, I > >> > do > >> > not understand the dhcp proxy or the pxe.conf file to know > >> how to do that. > >> > Plus the documentation seems sparse. Can someone tell me > >> how to accomplish > >> > my goal? Better yet, can someone direct me to a forum > where I may be > >> > enlightened in this and other pxe-boot questions? > >> > > >> > Thanks! > >> > > >> > Andrew Robinson > >> > > >> > _______________________________________________ > >> > Kickstart-list mailing list > >> > Kickstart-list at redhat.com > >> > https://www.redhat.com/mailman/listinfo/kickstart-list > >> > > >> > > >> > >> > >> _______________________________________________ > >> Kickstart-list mailing list > >> Kickstart-list at redhat.com > >> https://www.redhat.com/mailman/listinfo/kickstart-list > >> > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From dan at half-asleep.com Mon May 9 12:22:17 2005 From: dan at half-asleep.com (Daniel Segall) Date: Mon, 9 May 2005 08:22:17 -0400 (EDT) Subject: OT: pxe server - dhcp proxy question In-Reply-To: <379313C94B76D2119AB60008C7A402E409E463F7@IMSNOLAA> References: <379313C94B76D2119AB60008C7A402E409E463F7@IMSNOLAA> Message-ID: <59211.192.80.55.74.1115641337.squirrel@webmail.half-asleep.com> I actually have a symlink /linux-install to /tftpboot/X86PC/UNDI/linux-install/ My full pxe.conf is attached. -Dan > Dan, what's the rest of your /etc/pxe.conf file look like? For what you > showed, mine is the same as yours. However, when I try to pxe boot, the > server tries to serve /tftpboot/X86PC/UNDI/linux-install/linux.0 instead > of /tftpboot/linux-install/pxelinux.0. > > Thanks! > > Andrew > >> From: Daniel Segall [mailto:dan at half-asleep.com] >> >> Here's how I do it... >> >> dhcpd.conf: >> ################################### >> >> ddns-update-style interim; >> ignore client-updates; >> >> subnet 10.1.1.0 netmask 255.255.255.0 { >> >> # --- default gateway >> option routers 10.1.1.1; >> option subnet-mask 255.255.255.0; >> >> # option nis-domain "domain.org"; >> # option domain-name "domain.org"; >> # option domain-name-servers 192.168.1.1; >> >> # option time-offset -18000; # Eastern >> Standard Time >> # option ntp-servers 192.168.1.1; >> # option netbios-name-servers 192.168.1.1; >> # --- Selects point-to-point node (default is hybrid). Don't >> change this >> unless >> # -- you understand Netbios very well >> # option netbios-node-type 2; >> >> range dynamic-bootp 10.1.1.200 10.1.1.250; >> default-lease-time 21600; >> max-lease-time 43200; >> >> # we want the nameserver to appear at a fixed address >> # host ns { >> # next-server marvin.redhat.com; >> # hardware ethernet 12:34:56:78:AB:CD; >> # fixed-address 207.175.42.254; >> # } >> filename "pxelinux.0"; >> next-server 10.1.1.1; >> } >> >> In the pxe.conf: >> ################################### >> >> [OurIpAddress] >> 10.1.1.1 >> >> [Mtftpd_Root_Directory] >> /tftpboot >> >> [UseDHCPPort] >> 0 >> >> [StartProxy] >> 1 >> >> [startBootService] >> 1 >> >> [MasterProxy] >> 1 >> >> [Discovery_BCast_Disabled] >> 0 >> >> [Discovery_MCast_Disabled] >> 0 >> >> [Prompt] >> 10,Press F8 to view menu ... >> >> >> /tftpboot/pxelinux.cfg/default >> ################################### >> # Default PXE Config >> >> # Local boot by default >> default local >> >> # Always prompt user >> prompt 1 >> >> # Display our boot message >> display pxeboot.msg >> >> # Boot automatically after 10 minute >> timeout 6000 >> >> label local >> localboot 0 >> label Rescue >> kernel vmlinuz-RHES-3 >> append initrd=initrd-RHES-3.img rescue >> ks=nfs:10.1.1.1:/kickstart/rescue.ks.cfg text devfs=nomount >> ramdisk=10240 vga=0xF01 >> label RHES-3-U4 >> kernel vmlinuz-RHES-3-U4 >> append initrd=initrd-RHES-3-U4.img >> ks=nfs:10.1.1.1:/kickstart/RHES-3-U4-ks.cfg >> label FC-3 >> kernel vmlinuz-FC-3 >> append initrd=initrd-FC-3.img >> ks=nfs:10.1.1.1:/kickstart/FC-3-ks.cfg >> >> ################################### >> >> >> That's really about it. >> >> -Dan >> >> >> >> > The redhat manual explains how to set up a pxe-boot >> environment _without_ >> > a >> > dhcp proxy. I'm trying to go the next step and add the dhcp >> proxy (btw, >> > using the pxe-0.1-36.i386.rpm package provided by redhat. >> Just to make >> > things interesting, the dchp proxy server is named pxe.) >> I'm looking for >> > help in configuring the dhcp proxy server. >> > >> > Andrew >> > >> >> From: Daniel Segall [mailto:dan at half-asleep.com] >> >> >> >> The redhat manual goes into a bit of detail on setting up a >> >> pxe/tftp/dhcp >> >> kickstart environment. Chapters 9, 10, 14, & 15 should help >> >> you out. It >> >> can be found at >> >> https://www.redhat.com/docs/manuals/enterprise/RHEL-3-Manual/s >> >> ysadmin-guide/index.html >> >> >> >> -Dan >> >> >> >> > I am exploring the use of the proxy-dhcp server. The one >> >> that comes with >> >> > Red >> >> > Hat Enterprise Linux 3 Update 4 is pxe-0.1-36.i386.rpm. >> >> With a pxe server >> >> > without the dhcp proxy, when a client pxe-boots, it >> >> displays the menu >> >> > specified in the /tftpboot/linux-install/msgs/boot.msg >> file. When I >> >> > install >> >> > the dhcp proxy (pxe-0.1-36.i386.rpm), when the client >> pxe-boots, the >> >> > messages displayed are those specified in the /etc/pxe.conf >> >> file. I would >> >> > like to get the dhcp proxy to use the messages in the >> >> boot.msg file. I >> >> > assume there is some way to specify that in the pxe.conf >> >> file. However, I >> >> > do >> >> > not understand the dhcp proxy or the pxe.conf file to know >> >> how to do that. >> >> > Plus the documentation seems sparse. Can someone tell me >> >> how to accomplish >> >> > my goal? Better yet, can someone direct me to a forum >> where I may be >> >> > enlightened in this and other pxe-boot questions? >> >> > >> >> > Thanks! >> >> > >> >> > Andrew Robinson >> >> > >> >> > _______________________________________________ >> >> > Kickstart-list mailing list >> >> > Kickstart-list at redhat.com >> >> > https://www.redhat.com/mailman/listinfo/kickstart-list >> >> > >> >> > >> >> >> >> >> >> _______________________________________________ >> >> Kickstart-list mailing list >> >> Kickstart-list at redhat.com >> >> https://www.redhat.com/mailman/listinfo/kickstart-list >> >> >> > >> > _______________________________________________ >> > Kickstart-list mailing list >> > Kickstart-list at redhat.com >> > https://www.redhat.com/mailman/listinfo/kickstart-list >> > >> > >> >> >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list >> > > _______________________________________________ > 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: pxe.conf Type: application/octet-stream Size: 5357 bytes Desc: not available URL: From Eric.Doutreleau at int-evry.fr Tue May 10 15:22:09 2005 From: Eric.Doutreleau at int-evry.fr (Eric Doutreleau) Date: Tue, 10 May 2005 17:22:09 +0200 Subject: problems with genhdlist and pkgorder Message-ID: <1115738529.13414.6.camel@rezo.int-evry.fr> hi i m trying to kickstart install my FC3 machines in order to speed it up i would like to merge updates in the core distributions. then i copy the updates in the Fedora/RPMS directory i remove the old packages i ran /usr/lib/anaconda-runtime/pkgorder /var/ftp/pub/distrib/mci-12/ i386 Fedora | tee /var/ftp/pub/distrib/mci-12/pkgfile /var/ftp/pub/distrib/mci-12/ is my base directory i got a bunch of messages like that warning: ignore package name relation(s) [0] pango-gtkbeta-devel -> pango-gtkbeta warning: ignore package name relation(s) [1] XFree86 -> Mesa warning: ignore package name relation(s) [2] xorg-x11 -> Mesa warning: ignore package name relation(s) [3] compat-glibc -> db2 warning: ignore package name relation(s) [4] compat-glibc -> db1 warning: ignore package name relation(s) [5] pam -> initscripts warning: ignore package name relation(s) [6] initscripts -> sysklogd warning: ignore package name relation(s) [7] arts -> kdelibs-sound warning: ignore package name relation(s) [8] libgnomeprint15 -> gnome-print warning: ignore package name relation(s) [9] nautilus -> nautilus- mozilla warning: ignore package name relation(s) [10] tcl -> postgresql-tcl warning: ignore package name relation(s) [11] libtermcap -> bash warning: ignore package name relation(s) [12] modutils -> vixie-cron warning: ignore package name relation(s) [13] ypbind -> yp-tools warning: ignore package name relation(s) [14] ghostscript-fonts -> ghostscriptwarning: ignore package name relation(s) [15] usermode -> util-linux warning: ignore package name relation(s) [16] control-center -> xscreensaver warning: ignore package name relation(s) [17] kdemultimedia-arts -> kdemultimedia-libs warning: ignore package name relation(s) [18] initscripts -> util- linux warning: ignore package name relation(s) [19] XFree86-libs -> XFree86- Mesa-libGL warning: ignore package name relation(s) [20] xorg-x11-libs -> xorg- x11-Mesa-libGL warning: ignore package name relation(s) [21] mysql -> perl-DBD-MySQL warning: ignore package name relation(s) [22] ghostscript -> gimp- print warning: ignore package name relation(s) [23] bind -> bind-utils warning: ignore package name relation(s) [24] perl -> mod_perl warning: ignore package name relation(s) [25] perl -> perl-Filter warning: ignore package name relation(s) [26] coreutils -> pam warning: ignore package name relation(s) [27] perl -> mrtg warning: ignore package name relation(s) [28] perl-Date-Calc -> perl- Bit-Vector warning: ignore package name relation(s) [29] glibc-debug -> glibc- devel warning: ignore package name relation(s) [30] xinitrc -> XFree86 warning: ignore package name relation(s) [31] xinitrc -> xorg-x11 warning: ignore package name relation(s) [32] xemacs -> apel-xemacs warning: ignore package name relation(s) [33] gstreamer-tools -> gstreamer warning: ignore package name relation(s) [34] gimp -> gimp-print- plugin warning: ignore package name relation(s) [35] redhat-lsb -> redhat-lsb warning: ignore package name relation(s) [36] info -> ncurses warning: ignore package name relation(s) [37] aspell -> aspell-en warning: ignore package name relation(s) [38] dbus -> dbus-glib warning: ignore package name relation(s) [39] openjade -> docbook-dtds warning: ignore package name relation(s) [40] xemacs -> xemacs-sumo warning: ignore package name relation(s) [41] ncurses -> gpm warning: ignore package name relation(s) [42] cyrus-sasl -> openldap warning: ignore package name relation(s) [43] initscripts -> kernel warning: ignore package name relation(s) [44] initscripts -> kernel- smp warning: ignore package name relation(s) [45] rpm-libs -> rpm warning: ignore package name relation(s) [46] httpd -> httpd-suexec warning: ignore package name relation(s) [47] php -> php-pear warning: ignore package name relation(s) [48] gnome-python2 -> gnome- python2-bonobo warning: ignore package name relation(s) [49] openoffice.org-libs -> openoffice.org warning: ignore package name relation(s) [50] gtk+ -> gdk-pixbuf warning: ignore package name relation(s) [51] nautilus -> nautilus-cd- burner warning: LOOP: warning: kernel-utils-2.4-13.1.49_FC3 PreReq: /sbin/service warning: removing initscripts-7.93.7-1 "Requires: /sbin/nash" from tsort relations. warning: initscripts-7.93.7-1 Requires: /sbin/nash warning: mkinitrd-4.1.18-2 PreReq: dev warning: udev-039-10.FC3.7 PreReq: lvm2 warning: removing lvm2-2.00.25-1.01 "Requires: kernel >= 2.6" from tsort relations. warning: lvm2-2.00.25-1.01 Requires: kernel >= 2.6 warning: kernel-2.6.11-1.14_FC3 PreReq: kernel- utils warning: LOOP: warning: removing kernel-utils-2.4-13.1.49_FC3 "Requires: kudzu" from tsort relations. warning: kernel-utils-2.4-13.1.49_FC3 Requires: kudzu warning: removing kudzu-1.1.95-1 "Requires: hal >= 0.2.96" from tsort relations.warning: kudzu-1.1.95-1 Requires: hal >= 0.2.96 warning: removing hal-0.4.7-1.FC3 "Requires: kernel >= 2.6.10" from tsort relations. warning: hal-0.4.7-1.FC3 Requires: kernel >= 2.6.10warning: kernel-2.6.11-1.14_FC3 PreReq: kernel-utils then when i run genhdlist /usr/lib/anaconda-runtime/genhdlist -- fileorder /var/ftp/pub/distrib/mci-12/pkgfile --withnumbers -- productpath Fedora /var/ftp/pub/distrib/mci-12/ and when i launch the install it exit with the message installatio nexited abnormally just before begging installing the packages has someone succeed in updating a FC3 distribution? thanks in advance for any help From tibbs at math.uh.edu Tue May 10 15:36:37 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Tue, 10 May 2005 10:36:37 -0500 Subject: problems with genhdlist and pkgorder In-Reply-To: <1115738529.13414.6.camel@rezo.int-evry.fr> (Eric Doutreleau's message of "Tue, 10 May 2005 17:22:09 +0200") References: <1115738529.13414.6.camel@rezo.int-evry.fr> Message-ID: >>>>> "ED" == Eric Doutreleau writes: ED> has someone succeed in updating a FC3 distribution? Yes, I do it all the time. I've posted my script before, but here's a link to what I currently use on FC3-i386: http://www.math.uh.edu/~tibbs/do_update I should probably update the version and copyright information one day. - J< From koreesmith at gmail.com Wed May 11 13:44:33 2005 From: koreesmith at gmail.com (Koree A. Smith) Date: Wed, 11 May 2005 08:44:33 -0500 Subject: How to add raid and nic driver to Kickstart beginning In-Reply-To: <1115395562.2623.25.camel@ronnote> References: <1115395562.2623.25.camel@ronnote> Message-ID: <676847c6050511064443f8a3a9@mail.gmail.com> I would agree with Ronald. Update 4 took care of a lot of our Dell Server RAID and NIC card driver issues. Koree On 5/6/05, Ronald Reed wrote: > What drivers do you need? The isolinux boot on the RHE CD should have > most of the drivers. > > Ron > > On Fri, 2005-05-06 at 11:03, John Tran wrote: > > I read that I'm supposed to pass a 'dd' command at the linux boot > > prompt in order to add in 3rd party raid and nic drivers that aren't > > supported by the kernel. > > > > I am kickstarting RH Enterprise 3v1 and it doesn't support my Dell 220 > > Raid. So how do I go about automating it so that I don't have to > > insert a driver disk at the beginning of the kickstart for the raid > > and nic drivers? > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > -- > =========================== > Ron Reed > RedHat Certified Engineer > SGP Computer Department Manager > Unix Systems Administrator > ARM SGP CART Site > (580)388-4053 > ron.reed at arm.gov > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -- Koree A. Smith -- koreesmith at gmail.com From herrold at owlriver.com Wed May 11 17:39:31 2005 From: herrold at owlriver.com (R P Herrold) Date: Wed, 11 May 2005 13:39:31 -0400 (EDT) Subject: ks] Minimal Install Guidelines In-Reply-To: <426EB933.1040307@hostinthebox.net> References: <426EB933.1040307@hostinthebox.net> Message-ID: On Tue, 26 Apr 2005, dan wrote: > I've been working with a few documents that describe how to go about making a > minimalistic install, but they all seem kinda hoakey. I know you guys > probably have a lot of experience doing this sort of thing, so I'm hoping > that I might be able to get some answers from you all. from the better late than never department: I wrote a general tool: http://www.owlriver.com/tips/tiny-centos/ Because the of the ongoing questions, we sat down to develop and document a kickstart configuration file and script, to work generally for any anaconda installation mediated host, for this goal state -- This ks.cfg is suitable for doing such 'tiny' installations. It is of the customary type ks.cfg file, with some 'special sauce' in the %post stanza, to strip out undesired package. -- Russ Herrold From herrold at owlriver.com Wed May 11 17:41:17 2005 From: herrold at owlriver.com (R P Herrold) Date: Wed, 11 May 2005 13:41:17 -0400 (EDT) Subject: ks] Re: Minimal Install Guidelines In-Reply-To: <426ECAA2.3070303@hostinthebox.net> References: <426EB933.1040307@hostinthebox.net> <426EBEBC.5020508@hostinthebox.net> <426EC2C5.5040100@hostinthebox.net> <426ECAA2.3070303@hostinthebox.net> Message-ID: On Tue, 26 Apr 2005, dan wrote: > Well, I got it down to 479M. Yay for me. The one problem I see is > glibc-common is over 200M in itself. catching up -- under 350M with the tool referenced a moment ago. - r From info at hostinthebox.net Wed May 11 18:27:06 2005 From: info at hostinthebox.net (dan) Date: Wed, 11 May 2005 11:27:06 -0700 Subject: ks] Re: Minimal Install Guidelines In-Reply-To: References: <426EB933.1040307@hostinthebox.net> <426EBEBC.5020508@hostinthebox.net> <426EC2C5.5040100@hostinthebox.net> <426ECAA2.3070303@hostinthebox.net> Message-ID: <42824E7A.9020603@hostinthebox.net> R P Herrold wrote: > On Tue, 26 Apr 2005, dan wrote: > >> Well, I got it down to 479M. Yay for me. The one problem I see is >> glibc-common is over 200M in itself. > > > catching up -- under 350M with the tool referenced a moment ago. > > - r > Russ - Hey, I appreciate that. I'll take a day here and hack through it and see if I can use this. Thanks! -dant From kickstart at lordcow.org Fri May 13 15:18:36 2005 From: kickstart at lordcow.org (gareth) Date: Fri, 13 May 2005 17:18:36 +0200 Subject: boot disk Message-ID: <20050513151836.GA23885@lordcow.org> yo, i wanted to make a kickstart boot disk, the redhat manual said i first have to make a boot disk by imaging images/bootdisk.img (from the cd) to a formatted stiffy disk. however the only .img file under images is diskboot.img which's 6MB? From error27 at gmail.com Sat May 14 02:27:48 2005 From: error27 at gmail.com (Dan Carpenter) Date: Fri, 13 May 2005 19:27:48 -0700 Subject: boot disk In-Reply-To: <20050513151836.GA23885@lordcow.org> References: <20050513151836.GA23885@lordcow.org> Message-ID: You actually don't need to make a boot floppy to use kickstarts right? mkfs.vat /dev/fd0 mount /mnt/floppy/ cp ks.cfg /mnt/floppy umount /mnt/floppy Boot to the install CD and at the boot prompt type "linux ks=floppy". The bootdisk.img has a huge kernel and initrd so it's not going to fit on a floppy ever. To be honest I can't think why the docs even mention it at all. regards, dan carpenter From info at hostinthebox.net Sat May 14 02:29:37 2005 From: info at hostinthebox.net (Dan Trainor) Date: Fri, 13 May 2005 19:29:37 -0700 Subject: boot disk In-Reply-To: References: <20050513151836.GA23885@lordcow.org> Message-ID: <42856291.7030101@hostinthebox.net> I've been confused by this a few times, too. I just don't know ;) Thanks -dant Dan Carpenter wrote: > You actually don't need to make a boot floppy to use kickstarts right? > > mkfs.vat /dev/fd0 > mount /mnt/floppy/ > > cp ks.cfg /mnt/floppy > umount /mnt/floppy > > Boot to the install CD and at the boot prompt type "linux ks=floppy". > > The bootdisk.img has a huge kernel and initrd so it's not going to fit > on a floppy ever. To be honest I can't think why the docs even > mention it at all. > > regards, > dan carpenter > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > From marquez at us.es Mon May 16 08:30:38 2005 From: marquez at us.es (Antonio M. Marquez Cruz) Date: 16 May 2005 10:30:38 +0200 Subject: boot disk In-Reply-To: <20050513151836.GA23885@lordcow.org> References: <20050513151836.GA23885@lordcow.org> Message-ID: <1116232238.5095.135.camel@delfin.us.es> Hi, To use this image you need a pendrive USB device, but booting from a USB device depends on your motherboard allowing for this option. On Fri, 2005-05-13 at 17:18, gareth wrote: > yo, i wanted to make a kickstart boot disk, the redhat > manual said i first have to make a boot disk by imaging > images/bootdisk.img (from the cd) to a formatted stiffy disk. > however the only .img file under images is diskboot.img > which's 6MB? > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Dr. Antonio M. Marquez Dpt. of Physical Chemistry Departamento de Qu?mica F?sica University of Seville Universidad de Sevilla E-41012 Seville (SPAIN) 41012 Sevilla Phone: 34-95-4557177 (Ext. 213) Telf. 95-4557177 (Ext. 213) Fax: 34-95-4557174 Fax. 95-4557174 From jkeating at j2solutions.net Mon May 16 22:22:04 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Mon, 16 May 2005 15:22:04 -0700 Subject: boot disk In-Reply-To: <20050513151836.GA23885@lordcow.org> References: <20050513151836.GA23885@lordcow.org> Message-ID: <1116282124.5928.43.camel@jkeating2.hq.pogolinux.com> On Fri, 2005-05-13 at 17:18 +0200, gareth wrote: > yo, i wanted to make a kickstart boot disk, the redhat > manual said i first have to make a boot disk by imaging > images/bootdisk.img (from the cd) to a formatted stiffy disk. > however the only .img file under images is diskboot.img > which's 6MB? > Fedora Core and Red Hat Enterprise Linux (4) no longer support booting from floppy disks. If you want to boot from CD and use a kickstart config held on a floppy disk, that is fine. Just put the ks.cfg file in the root directory of your floppy, boot the CD with ks=floppy option. -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From bedouglas at earthlink.net Mon May 16 22:28:57 2005 From: bedouglas at earthlink.net (bruce) Date: Mon, 16 May 2005 15:28:57 -0700 Subject: boot disk In-Reply-To: <1116282124.5928.43.camel@jkeating2.hq.pogolinux.com> Message-ID: <15a401c55a66$a6546540$0301a8c0@Mesa.com> jesse, it is possible to boot with simply a floppy, if you have access to a network http/ftp server. there are (believe it or not) servers that only have floppy drives, with no hard drives... in fact, i had to essentially do a RH install using a floppy, and then do the network install to get to FC... -bruce -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com]On Behalf Of Jesse Keating Sent: Monday, May 16, 2005 3:22 PM To: kickstart-list at redhat.com Subject: Re: boot disk On Fri, 2005-05-13 at 17:18 +0200, gareth wrote: > yo, i wanted to make a kickstart boot disk, the redhat > manual said i first have to make a boot disk by imaging > images/bootdisk.img (from the cd) to a formatted stiffy disk. > however the only .img file under images is diskboot.img > which's 6MB? > Fedora Core and Red Hat Enterprise Linux (4) no longer support booting from floppy disks. If you want to boot from CD and use a kickstart config held on a floppy disk, that is fine. Just put the ks.cfg file in the root directory of your floppy, boot the CD with ks=floppy option. -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From jkeating at j2solutions.net Mon May 16 22:37:42 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Mon, 16 May 2005 15:37:42 -0700 Subject: boot disk In-Reply-To: <15a401c55a66$a6546540$0301a8c0@Mesa.com> References: <15a401c55a66$a6546540$0301a8c0@Mesa.com> Message-ID: <1116283062.5928.46.camel@jkeating2.hq.pogolinux.com> On Mon, 2005-05-16 at 15:28 -0700, bruce wrote: > jesse, > > it is possible to boot with simply a floppy, if you have access to a > network > http/ftp server. there are (believe it or not) servers that only have > floppy > drives, with no hard drives... > > in fact, i had to essentially do a RH install using a floppy, and then > do > the network install to get to FC... Well, sure. There are all kinds of ways to hack/boot a box and then bootstrap whatever into it. However Red Hat no longer produces floppy boot images as the initrd is just too big these days. Even with the smallest driver set it can be too big. -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From ashley.hodder at otago.ac.nz Mon May 16 23:27:33 2005 From: ashley.hodder at otago.ac.nz (Ashley Hodder) Date: Tue, 17 May 2005 11:27:33 +1200 Subject: ks from usb memory stick Message-ID: <1116286052.6236.348.camel@malibu.otago.ac.nz> Hi, I'm trying to get kickstart to take the ks.cfg from a usb memory stick (as my new HP EVO hasn't a floppy). Trying to load ES 4 from a CD.When loading after the ks= line the kernel loads usb-storage. I've tried all sorts of combos such as boot: linux ks=hd:sda1/ks.cfg boot: linux ks=usb1 can't find any reference on google etc Any pointers would be great.. Ashley Hodder ITS University of Otago New Zealand From error27 at gmail.com Tue May 17 00:42:44 2005 From: error27 at gmail.com (Dan Carpenter) Date: Mon, 16 May 2005 17:42:44 -0700 Subject: ks from usb memory stick In-Reply-To: <1116286052.6236.348.camel@malibu.otago.ac.nz> References: <1116286052.6236.348.camel@malibu.otago.ac.nz> Message-ID: > boot: linux ks=hd:sda1/ks.cfg That is the correct way. (Unless you have another scsi device then it's sdb1 or whatever). Make sure your USB device is being seen correctly. Switch to tty2 and try mount it manually etc. Make sure legacy USB is turned on the BIOS as well. regards, dan carpenter From drkludge at cox.net Tue May 17 04:11:49 2005 From: drkludge at cox.net (Greg Morgan) Date: Mon, 16 May 2005 21:11:49 -0700 Subject: ks from usb memory stick In-Reply-To: <1116286052.6236.348.camel@malibu.otago.ac.nz> References: <1116286052.6236.348.camel@malibu.otago.ac.nz> Message-ID: <42896F05.8070909@cox.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ashley Hodder wrote: > Hi, I'm trying to get kickstart to take the ks.cfg from a usb memory > stick (as my new HP EVO hasn't a floppy). Trying to load ES 4 from a > CD.When loading after the ks= line the kernel loads usb-storage. > > I've tried all sorts of combos such as > > boot: linux ks=hd:sda1/ks.cfg > boot: linux ks=usb1 > > can't find any reference on google etc > > Any pointers would be great.. Rationalization produced this write up https://www.redhat.com/archives/kickstart-list/2004-September/msg00018.html. It covers many of the possible issues that you can run into. For example, if other devices are present that you are not aware of, then you need to look at sdb1, etc. Greg Morgan -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFCiW8Fxyxe5L6mr7IRAnfnAJ9voWKqpjhQectkpvIH9l0jiQ7zvACffv7s cfIj54z8ECiExYpfixuYERM= =PWTK -----END PGP SIGNATURE----- From error27 at gmail.com Tue May 17 16:26:49 2005 From: error27 at gmail.com (Dan Carpenter) Date: Tue, 17 May 2005 12:26:49 -0400 Subject: boot disk In-Reply-To: <1116282124.5928.43.camel@jkeating2.hq.pogolinux.com> References: <20050513151836.GA23885@lordcow.org> <1116282124.5928.43.camel@jkeating2.hq.pogolinux.com> Message-ID: On 5/16/05, Jesse Keating wrote: > > Fedora Core and Red Hat Enterprise Linux (4) no longer support booting > from floppy disks. Since 1.4M is no longer the max size what is the new max size that we will allow the initrd to be? I'm asking this because there is a lot of masochistic code in the initrd and now that we don't use boot floppies, it may not be needed any more. It would be cool to put a small busy box shell in the initrd so that you could have a shell prompt right away. You could use this to modprobe drivers and type lspci and basically manually fix things the same as you do in other busy box boot environments. At work we have a hacked up initrd so that we can put driver disks directly into the initrd. A lot of systems don't have a floppy or cdrom and if it's a scsi driver you can't store it on the drive. These days I guess everyone would use USB, but that's slow and hands on where our solution is awesome and completely automated. There are probably other ways the initrd could be made more useful now that we don't have to deal with the 1.4M limit in the stock install. regards, dan carpenter From bertignac at gmail.com Wed May 18 08:12:53 2005 From: bertignac at gmail.com (L B) Date: Wed, 18 May 2005 10:12:53 +0200 Subject: NFS install and interactive asking of static ip Message-ID: Hi, I run RHAS3 update 4. I would like to : . Do an NFS install *not* in interactive mode. . Get my first IP with DHCP (text mode) to get the ks.cfg and the rpms. . Have a graphical screen asking me only for my static IP, hostname, DNS servers etc.. and nothing else. . Install redhat etc... If I don't put any network stuff in the ks.cfg, kickstart doesn't ask me networking info : >From the doc : "If the installation does require networking and network information is not provided in the kickstart file, the Red Hat Linux installation program assumes that the installation should be done over eth0 via a dynamic IP address (BOOTP/DHCP), and configures the final, installed system to determine its IP address dynamically. The network option configures networking information for kickstart installations via a network as well as for the installed system." Can I make kickstart ask me the networking configuration ? Thank you From SSeremeth at anacomp.com Wed May 18 15:51:52 2005 From: SSeremeth at anacomp.com (Seremeth, Stephen) Date: Wed, 18 May 2005 11:51:52 -0400 Subject: Argh. Anaconda and Sun's Java RPMs in ES 4 Message-ID: <66911C8E1F2DA24EB29DE289297A135E4856A3@usrd104> Maybe someone has some constructive thoughts on this one. I'm working out the kinks in a network install, and I had it working, then I added a few rpms, then all went to hell. Since it was several weeks between "had it working" and "hell" stages, I couldn't remember the nuances of what probably broke my install. So -- long story short: * pxe booting followed by unattended kickstart install * comps.xml stripped down to base packages * test rpm db created for making sure dependencies resolved * getfullcomps.py also run to make sure everything copasetic * After enabling portFast on the switch (why this is necessary I'm still not 100% sure since the initial PXE dhcp request works), install began working * Added j2sdk 1.4.2, jdk 1.5.0_02, and my own Sun Java System Webserver RPM (they don't distribute an rpm for this, thanks, Sun) * test rpm db created for making sure dependencies resolved * getfullcomps.py also run to make sure everything copasetic * Reran genhdlist/pkgorder/genhdlist with proper syntax * pkgorder spits out: "warning: LOOP:" followed by some garbage a few times -- but this is OK supposedly * when genhdlist the second time this happens: [root at myserver rh_es_4]# genhdlist --withnumbers --fileorder /dist/rh_es_4/custom/pkgfile --hdlist /dist/rh_es_4/custom/i386/RedHat/base/hdlist /dist/rh_es_4/custom/i386 WARNING: ordering not found for j2sdk-1_4_2-linux-i586.rpm WARNING: ordering not found for jdk-1_5_0_02-linux-i586.rpm Reboot on install-target and PXE boot off of install server. Anaconda: * loads drivers * gets an ip over dhcp * gets kickstart script off of NFS server * Formats drive per ks script * Gets to "Preparing RPM Transaction" and BOOM! "Installer exiting abnormally" * Only thing of note on virtual consoles is the presence of: "WARNING: not all packages in hdlist had order tag" The rpms install just fine if installed on a running ES 4 machine, just not via anaconda thus far. Any ideas? I am about to try without my custom rpm and the newer jdk rpm, because I had a customized installer I did with AS 2.1 and the older jdk installed just fine. Thanks for any help. Regards, Steve -------------- next part -------------- An HTML attachment was scrubbed... URL: From brilong at cisco.com Wed May 18 16:11:12 2005 From: brilong at cisco.com (Brian Long) Date: Wed, 18 May 2005 12:11:12 -0400 Subject: Argh. Anaconda and Sun's Java RPMs in ES 4 In-Reply-To: <66911C8E1F2DA24EB29DE289297A135E4856A3@usrd104> References: <66911C8E1F2DA24EB29DE289297A135E4856A3@usrd104> Message-ID: <1116432672.5709.15.camel@localhost.localdomain> > [root at myserver rh_es_4]# genhdlist --withnumbers -- > fileorder /dist/rh_es_4/custom/pkgfile -- > hdlist /dist/rh_es_4/custom/i386/RedHat/base/hdlist /dist/rh_es_4/custom/i386 > WARNING: ordering not found for j2sdk-1_4_2-linux-i586.rpm > WARNING: ordering not found for jdk-1_5_0_02-linux-i586.rpm Notice Sun's RPMs do not follow standard RPM naming conventions. You have to rename the RPMs to j2sdk-1_4_2-fcs.i586.rpm and jdk-1.5.0_02- fcs.i586.rpm (run an rpm -qpi on the RPM and you'll see the SRPM mentioned). You need the RPM name, version and release to match the name, version and release inside the RPM :) -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From SSeremeth at anacomp.com Wed May 18 19:11:44 2005 From: SSeremeth at anacomp.com (Seremeth, Stephen) Date: Wed, 18 May 2005 15:11:44 -0400 Subject: Argh. Anaconda and Sun's Java RPMs in ES 4 Message-ID: <66911C8E1F2DA24EB29DE289297A135E4856A6@usrd104> > > > [root at myserver rh_es_4]# genhdlist --withnumbers -- > > fileorder /dist/rh_es_4/custom/pkgfile -- > > hdlist /dist/rh_es_4/custom/i386/RedHat/base/hdlist > /dist/rh_es_4/custom/i386 > > WARNING: ordering not found for j2sdk-1_4_2-linux-i586.rpm > > WARNING: ordering not found for jdk-1_5_0_02-linux-i586.rpm > > Notice Sun's RPMs do not follow standard RPM naming conventions. You > have to rename the RPMs to j2sdk-1_4_2-fcs.i586.rpm and jdk-1.5.0_02- > fcs.i586.rpm (run an rpm -qpi on the RPM and you'll see the SRPM > mentioned). You need the RPM name, version and release to match the > name, version and release inside the RPM :) > Brian - You are a saviour. If I got paid double-time for the length of time it took me to find the source of issues like this one, I would be quite wealthy by now. Well, maybe not, but the bonus would be nice! Renaming Sun's crappy RPMs fixed the problem. I'm not sure why Anaconda is so finicky about this when they can be installed using rpm by itself, but at this point I don't much care about that -- it's clear Sun isn't following the standard (and that I should have caught this). I will write Sun to complain even though I'm sure they won't fix it. Maybe whomever is in charge of Anaconda development should consider this a feature request for a better error message than "rpm transaction failed"? More detail for those who care/whom I might save any hassle: jdk-1_5_0_02-linux-i586.rpm [root at myserver rh_es_4]# rpm -qip my_rpms/jdk-1_5_0_02-linux-i586.rpm Name : jdk Relocations: /usr/java Version : 1.5.0_02 Vendor: Sun Microsystems, Inc. Release : fcs Build Date: Fri 04 Mar 2005 08:18:03 AM EST Install Date: (not installed) Build Host: tiger-linux Group : Development/Tools Source RPM: jdk-1.5.0_02-fcs.src.rpm Size : 80307553 License: Sun Microsystems Binary Code License (BCL) Signature : (none) Packager : Java Software URL : http://java.sun.com/ Summary : Java(TM) 2 Platform Standard Edition Development Kit Description : The Java 2 Platform Standard Edition Development Kit (JDK) includes both the runtime environment (Java virtual machine, the Java platform classes and supporting files) and development tools (compilers, debuggers, tool libraries and other tools). The JDK is a development environment for building applications, applets and components that can be deployed with the Java 2 Platform Standard Edition Runtime Environment. ----------------------------------- Had Sun named this file correctly it would have been (and is now on my equipment): jdk-1.5.0_02-fcs.i586.rpm Following the standard convention which Brian so keenly noted: --..rpm The same goes for Sun's previous jdk RPM (they can't even seem to make up their mind about the name!): j2sdk-1_4_2-linux-i586.rpm [root at myserver rh_es_4]# rpm -qip my_rpms/j2sdk-1_4_2-linux-i586.rpm Name : j2sdk Relocations: (not relocatable) Version : 1.4.2 Vendor: Sun Microsystems Release : fcs Build Date: Fri 20 Jun 2003 05:52:19 AM EDT Install Date: (not installed) Build Host: localhost.localdomain Group : Development/Tools Source RPM: j2sdk-1.4.2-fcs.src.rpm Size : 65687911 License: 1994-2001 Sun Microsystems, Inc. Signature : (none) Packager : Java Software URL : http://java.sun.com/linux Summary : Java(TM) 2 Software Development Kit, Standard Edition Description : The Java 2 SDK, Standard Edition includes the Java Virtual Machine, core class libraries and tools used by programmers to develop Java software applets and applications. The SDK also provides the foundation for IDE (Integrated Development Environment) tools such as Sun's Forte for Java, Community Edition, the Java(TM) 2 Platform, Enterprise Edition (J2EE), Java-based application servers and more. The Java 2 Software Development Kit, SDK, is a development environment for building applications, applets, and components that can be deployed on the Java platform. The Java 2 SDK software includes tools useful for developing and testing programs written in the Java programming language and running on the Java platform. These tools are designed to be used from the command line. Except for appletviewer, these tools do not provide a graphical user interface. --------------------------------------------------- So that rpm is now: j2sdk-1.4.2-fcs.i586.rpm And now the install is working -- at least as far as rpm installation is concerned. And I'm on to my next problem... Which is unrelated and much less of a concern. Hope this thread saved someone some hassle. Best Regards, Steve From info at hostinthebox.net Wed May 18 19:44:22 2005 From: info at hostinthebox.net (dan) Date: Wed, 18 May 2005 12:44:22 -0700 Subject: Argh. Anaconda and Sun's Java RPMs in ES 4 In-Reply-To: <66911C8E1F2DA24EB29DE289297A135E4856A6@usrd104> References: <66911C8E1F2DA24EB29DE289297A135E4856A6@usrd104> Message-ID: <428B9B16.9000306@hostinthebox.net> Seremeth, Stephen wrote: >>>[root at myserver rh_es_4]# genhdlist --withnumbers -- >>>fileorder /dist/rh_es_4/custom/pkgfile -- >>>hdlist /dist/rh_es_4/custom/i386/RedHat/base/hdlist >> >>/dist/rh_es_4/custom/i386 >> >>>WARNING: ordering not found for j2sdk-1_4_2-linux-i586.rpm >>>WARNING: ordering not found for jdk-1_5_0_02-linux-i586.rpm >> >>Notice Sun's RPMs do not follow standard RPM naming conventions. You >>have to rename the RPMs to j2sdk-1_4_2-fcs.i586.rpm and jdk-1.5.0_02- >>fcs.i586.rpm (run an rpm -qpi on the RPM and you'll see the SRPM >>mentioned). You need the RPM name, version and release to match the >>name, version and release inside the RPM :) >> > > > Brian - > > You are a saviour. If I got paid double-time for the length of time it > took me to find the source of issues like this one, I would be quite > wealthy by now. Well, maybe not, but the bonus would be nice! > > Renaming Sun's crappy RPMs fixed the problem. I'm not sure why Anaconda > is so finicky about this when they can be installed using rpm by itself, > but at this point I don't much care about that -- it's clear Sun isn't > following the standard (and that I should have caught this). I will > write Sun to complain even though I'm sure they won't fix it. Maybe > whomever is in charge of Anaconda development should consider this a > feature request for a better error message than "rpm transaction > failed"? > > More detail for those who care/whom I might save any hassle: > jdk-1_5_0_02-linux-i586.rpm > > [root at myserver rh_es_4]# rpm -qip my_rpms/jdk-1_5_0_02-linux-i586.rpm > Name : jdk Relocations: /usr/java > Version : 1.5.0_02 Vendor: Sun > Microsystems, Inc. > Release : fcs Build Date: Fri 04 Mar 2005 > 08:18:03 AM EST > Install Date: (not installed) Build Host: tiger-linux > Group : Development/Tools Source RPM: > jdk-1.5.0_02-fcs.src.rpm > Size : 80307553 License: Sun Microsystems > Binary Code License (BCL) > Signature : (none) > Packager : Java Software > URL : http://java.sun.com/ > Summary : Java(TM) 2 Platform Standard Edition Development Kit > Description : > The Java 2 Platform Standard Edition Development Kit (JDK) includes both > the > runtime environment (Java virtual machine, the Java platform classes and > supporting files) and development tools (compilers, debuggers, tool > libraries > and other tools). > > The JDK is a development environment for building applications, applets > and > components that can be deployed with the Java 2 Platform Standard > Edition > Runtime Environment. > > ----------------------------------- > Had Sun named this file correctly it would have been (and is now on my > equipment): > jdk-1.5.0_02-fcs.i586.rpm > > Following the standard convention which Brian so keenly noted: > --..rpm > > The same goes for Sun's previous jdk RPM (they can't even seem to make > up their mind about the name!): > j2sdk-1_4_2-linux-i586.rpm > > [root at myserver rh_es_4]# rpm -qip my_rpms/j2sdk-1_4_2-linux-i586.rpm > Name : j2sdk Relocations: (not > relocatable) > Version : 1.4.2 Vendor: Sun Microsystems > Release : fcs Build Date: Fri 20 Jun 2003 > 05:52:19 AM EDT > Install Date: (not installed) Build Host: > localhost.localdomain > Group : Development/Tools Source RPM: > j2sdk-1.4.2-fcs.src.rpm > Size : 65687911 License: 1994-2001 Sun > Microsystems, Inc. > Signature : (none) > Packager : Java Software > URL : http://java.sun.com/linux > Summary : Java(TM) 2 Software Development Kit, Standard Edition > Description : > The Java 2 SDK, Standard Edition includes the Java > Virtual Machine, core class libraries and tools used > by programmers to develop Java software applets and > applications. The SDK also provides the foundation > for IDE (Integrated Development Environment) tools > such as Sun's Forte for Java, Community Edition, > the Java(TM) 2 Platform, Enterprise Edition (J2EE), > Java-based application servers and more. > The Java 2 Software Development Kit, SDK, is a > development environment for building applications, > applets, and components that can be deployed on > the Java platform. The Java 2 SDK software includes > tools useful for developing and testing programs > written in the Java programming language and running > on the Java platform. These tools are designed to > be used from the command line. Except for > appletviewer, these tools do not provide a > graphical user interface. > --------------------------------------------------- > So that rpm is now: > j2sdk-1.4.2-fcs.i586.rpm > > And now the install is working -- at least as far as rpm installation is > concerned. > > And I'm on to my next problem... Which is unrelated and much less of a > concern. Hope this thread saved someone some hassle. > > Best Regards, > > Steve > Steve - Thanks for the tip, I appreciate it. I'm going to start doing something similar very soon here, and this sure will make my job easier. And even if I did get overtime, you can't have it. Sorry. How about a case of beer, instead? Thanks! -dant From error27 at gmail.com Wed May 18 19:54:31 2005 From: error27 at gmail.com (Dan Carpenter) Date: Wed, 18 May 2005 15:54:31 -0400 Subject: NFS install and interactive asking of static ip In-Reply-To: References: Message-ID: On 5/18/05, L B wrote: > Hi, > > I run RHAS3 update 4. > > I would like to : > > . Do an NFS install *not* in interactive mode. > . Get my first IP with DHCP (text mode) to get the ks.cfg and the rpms. linux ip=dhcp ks=nfs:192.168.100.100:/home/path/to/ks.cfg If you have more than one ethernet device it asks you which to use. There is no way to make it test all the devices like you can with SuSE. You can specify the ethernet device but I forget the format for that because I never know which ethernet devices are going to be connected ahead of time. > . Have a graphical screen asking me only for my static IP, hostname, > DNS servers etc.. and nothing else. It used to ask you if it wasn't in the ks.cfg file... wierd. I would just automate this and put it in the post install. Perhaps something like a script that would parse /proc/cmdline and set the IP based on that. Or I would put a pause in the kickstart and create a small script that could be run from tty2. That is a hack. > . Install redhat etc... Create the kickstart (using magic). regards, dan carpenter From svenkat at india.ti.com Thu May 19 04:26:42 2005 From: svenkat at india.ti.com (Venkatachalam, Saravanakumar) Date: Thu, 19 May 2005 09:56:42 +0530 Subject: NFS install and interactive asking of static ip In-Reply-To: References: Message-ID: <428C1582.9050307@india.ti.com> Hi, Try providing this option network --device eth0 --bootproto static thanks, saravan L B wrote: >Hi, > >I run RHAS3 update 4. > >I would like to : > > . Do an NFS install *not* in interactive mode. > . Get my first IP with DHCP (text mode) to get the ks.cfg and the rpms. > . Have a graphical screen asking me only for my static IP, hostname, >DNS servers etc.. and nothing else. > . Install redhat etc... > >If I don't put any network stuff in the ks.cfg, kickstart doesn't ask >me networking info : >>From the doc : >"If the installation does require networking and network information >is not provided in the kickstart file, the Red Hat Linux installation >program assumes that the installation should be done over eth0 via a >dynamic IP address (BOOTP/DHCP), and configures the final, installed >system to determine its IP address dynamically. The network option >configures networking information for kickstart installations via a >network as well as for the installed system." > >Can I make kickstart ask me the networking configuration ? > >Thank you > >_______________________________________________ >Kickstart-list mailing list >Kickstart-list at redhat.com >https://www.redhat.com/mailman/listinfo/kickstart-list > > > From bertignac at gmail.com Thu May 19 07:44:12 2005 From: bertignac at gmail.com (L B) Date: Thu, 19 May 2005 09:44:12 +0200 Subject: NFS install and interactive asking of static ip In-Reply-To: <428C1582.9050307@india.ti.com> References: <428C1582.9050307@india.ti.com> Message-ID: On 5/19/05, Venkatachalam, Saravanakumar wrote: > Hi, > > Try providing this option > > network --device eth0 --bootproto static This doesn't work, ks didn't ask for the network configuration, it just misconfigured the IP and I had no network configuration at all after the install finished. From kickstart at lordcow.org Thu May 19 16:21:00 2005 From: kickstart at lordcow.org (gareth) Date: Thu, 19 May 2005 18:21:00 +0200 Subject: boot disk In-Reply-To: References: <20050513151836.GA23885@lordcow.org> Message-ID: <20050519162100.GA9971@lordcow.org> sweet that worked thanx On Fri 2005-05-13 (19:27), Dan Carpenter wrote: > You actually don't need to make a boot floppy to use kickstarts right? > > mkfs.vat /dev/fd0 > mount /mnt/floppy/ > > cp ks.cfg /mnt/floppy > umount /mnt/floppy > > Boot to the install CD and at the boot prompt type "linux ks=floppy". > > The bootdisk.img has a huge kernel and initrd so it's not going to fit > on a floppy ever. To be honest I can't think why the docs even > mention it at all. > > regards, > dan carpenter > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From klaus.steden at thomson.net Fri May 20 06:17:58 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Fri, 20 May 2005 02:17:58 -0400 Subject: problems with kickstart and automated partitioning Message-ID: <20050520061758.GP3570@thomson.net> Hello, I'm attempting to run automated KickStart install but I'm getting stymied by the partitioning phase. It works fine if I use two partitions + swap, but if I try to use three partitions + swap, I get errors from Anaconda (they vary depending on the options I specify in the 'part' command). Here is my current desired partition table: -- cut -- bootloader --location=mbr part / --fstype=ext3 --onpart=hda1 --size=500 part swap --fstype=swap --onpart=hda2 --noformat --recommended part /var --fstype=ext3 --ondisk=hda3 --size=1000 part /usr --fstype=ext3 --ondisk=hda4 --size=1000 --grow -- cut -- Here is the error I get from anaconda: -- cut -- Traceback (most recent call last): File "/usr/bin/anaconda", line 1122, in ? instClass.setInstallData(id) File "/usr/lib/anaconda/kickstart.py", line 1440, in setInstallData raise KickstartError, e kickstart.KickstartValueError: specified disk hda3 in partition command which does not exist install exited abnormally etc. -- cut -- Can someone clue me in, or help me debug my partition table? I know that /dev/hda[1-4] don't exist as device files when anaconda starts ... do they need to be there, or does it create them automagically? Any help is appreciated! thanks, Klaus I can't grok from the anaconda docs what exactly I'm doing wrong. The disk is pre-partitioned (using 'sfdisk') so I know for sure that all the partitions exist. From Christian.Rohrmeier at SCHERING.DE Fri May 20 08:26:17 2005 From: Christian.Rohrmeier at SCHERING.DE (Christian.Rohrmeier at SCHERING.DE) Date: Fri, 20 May 2005 10:26:17 +0200 Subject: Is a kickstart installation without DHCP (manual IP) possible? Message-ID: Hello Kickstarters, I am trying to perform a kickstart install in a LAN segment that doesn't have DHCP available. I am booting from the 1st RHEL CD. My boot: prompt options to RHEL(2.1/3/4) look like this: linux ks=http://some.server/some.script ip=1.2.3.4 netmask=255.255.255.128 gateway=1.2.3.1 dns=1.2.3.2 The result of this is that the a normal interactive install is started from CD; the kickstart script is ignored. I have tried to put the ks= at the back of the boot: prompt options, but that didn't help any. Also adding a ksdevice=eth0 option didn't help. How does one use a remote kickstart script when there is no dynamic IP available? Many thanks, Christian Rohrmeier _________________ Christian Rohrmeier Schering AG Corporate IT - Infrastructure and Services Computer Systems and Operations System Administration - Research and Development Tel +49 30 468 15794 Fax +49 30 468 95794 From klaus.steden at thomson.net Fri May 20 08:39:47 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Fri, 20 May 2005 04:39:47 -0400 Subject: Is a kickstart installation without DHCP (manual IP) possible? In-Reply-To: References: Message-ID: <20050520083946.GQ3570@thomson.net> > Hello Kickstarters, > > I am trying to perform a kickstart install in a LAN segment that doesn't > have DHCP available. I am booting from the 1st RHEL CD. > > My boot: prompt options to RHEL(2.1/3/4) look like this: > > linux ks=http://some.server/some.script ip=1.2.3.4 netmask=255.255.255.128 > gateway=1.2.3.1 dns=1.2.3.2 > > The result of this is that the a normal interactive install is started from > CD; the kickstart script is ignored. I have tried to put the ks= at the > back of the boot: prompt options, but that didn't help any. Also adding a > ksdevice=eth0 option didn't help. > > How does one use a remote kickstart script when there is no dynamic IP > available? > Christian, I ran into almost that exact problem very recently, and found the only way to convince anaconda to leave the CD alone was to put literally nothing on the CD (I'm using VMware to test, though, so I didn't make any coasters) - just the 'isolinux' directory and its' contents. After seeing that the CD isn't useable, anaconda will attempt to fetch the netstg2.img from somewhere, and installation proceeds acoordingly. hth, Klaus From Christian.Rohrmeier at SCHERING.DE Fri May 20 09:12:53 2005 From: Christian.Rohrmeier at SCHERING.DE (Christian.Rohrmeier at SCHERING.DE) Date: Fri, 20 May 2005 11:12:53 +0200 Subject: Is a kickstart installation without DHCP (manual IP) possible? Message-ID: Hi Klaus, such craziness!!! I will give it a shot. Thanks, Christian _________________ Christian Rohrmeier Schering AG Corporate IT - Infrastructure and Services Computer Systems and Operations System Administration - Research and Development Tel +49 30 468 15794 Fax +49 30 468 95794 Klaus Steden To: Discussion list about Kickstart Sent by: cc: kickstart-list-bounces Subject: Re: Is a kickstart installation without DHCP (manual IP) possible? @redhat.com 20.05.2005 10:39 Please respond to Discussion list about Kickstart > Hello Kickstarters, > > I am trying to perform a kickstart install in a LAN segment that doesn't > have DHCP available. I am booting from the 1st RHEL CD. > > My boot: prompt options to RHEL(2.1/3/4) look like this: > > linux ks=http://some.server/some.script ip=1.2.3.4 netmask=255.255.255.128 > gateway=1.2.3.1 dns=1.2.3.2 > > The result of this is that the a normal interactive install is started from > CD; the kickstart script is ignored. I have tried to put the ks= at the > back of the boot: prompt options, but that didn't help any. Also adding a > ksdevice=eth0 option didn't help. > > How does one use a remote kickstart script when there is no dynamic IP > available? > Christian, I ran into almost that exact problem very recently, and found the only way to convince anaconda to leave the CD alone was to put literally nothing on the CD (I'm using VMware to test, though, so I didn't make any coasters) - just the 'isolinux' directory and its' contents. After seeing that the CD isn't useable, anaconda will attempt to fetch the netstg2.img from somewhere, and installation proceeds acoordingly. hth, Klaus _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Christian.Rohrmeier at SCHERING.DE Fri May 20 10:19:31 2005 From: Christian.Rohrmeier at SCHERING.DE (Christian.Rohrmeier at SCHERING.DE) Date: Fri, 20 May 2005 12:19:31 +0200 Subject: Accessing boot: prompt options inside KS script Message-ID: Hi All, Is it possible to access the boot: prompt options in the %pre and %post scripts? As in, if on my bootprompt I have: linux ks=http://some.host/some.file -ip=1.2.3.4 that I can then use the value of "-ip=" in the kickstart script itself? Thanks, Christian _________________ Christian Rohrmeier Schering AG Corporate IT - Infrastructure and Services Computer Systems and Operations System Administration - Research and Development Tel +49 30 468 15794 Fax +49 30 468 95794 From klaus.steden at thomson.net Fri May 20 10:23:11 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Fri, 20 May 2005 06:23:11 -0400 Subject: Accessing boot: prompt options inside KS script In-Reply-To: References: Message-ID: <20050520102311.GR3570@thomson.net> > > Hi All, > > Is it possible to access the boot: prompt options in the %pre and %post > scripts? As in, if on my bootprompt I have: > > linux ks=http://some.host/some.file -ip=1.2.3.4 > > that I can then use the value of "-ip=" in the kickstart script itself? > Christian, You can access these parameters by reading /proc/cmdline. The contents of this file are read-only, and contain all the boot args except those that get used (and thus eaten) by the kernel (i.e. initrd= etc.). It'll take a little bit of perl, awk, python, sed, or shell to turn them into something useful, but they're there if you need them. hth, Klaus From phr at doc.ic.ac.uk Fri May 20 10:40:37 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Fri, 20 May 2005 11:40:37 +0100 (BST) Subject: problems with kickstart and automated partitioning In-Reply-To: <20050520061758.GP3570@thomson.net> References: <20050520061758.GP3570@thomson.net> Message-ID: On Fri, 20 May 2005, Klaus Steden wrote: >I'm attempting to run automated KickStart install but I'm getting stymied by >the partitioning phase. It works fine if I use two partitions + swap, but if >I try to use three partitions + swap, I get errors from Anaconda (they vary >depending on the options I specify in the 'part' command). > >Here is my current desired partition table: > >-- cut -- > > bootloader --location=mbr > part / --fstype=ext3 --onpart=hda1 --size=500 > part swap --fstype=swap --onpart=hda2 --noformat --recommended > part /var --fstype=ext3 --ondisk=hda3 --size=1000 > part /usr --fstype=ext3 --ondisk=hda4 --size=1000 --grow Not sure you can interchange onpart and ondisk like that... Cheers, Phil From klaus.steden at thomson.net Fri May 20 11:29:37 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Fri, 20 May 2005 07:29:37 -0400 Subject: problems with kickstart and automated partitioning In-Reply-To: References: <20050520061758.GP3570@thomson.net> Message-ID: <20050520112937.GS3570@thomson.net> > > > >Here is my current desired partition table: > > > >-- cut -- > > > > bootloader --location=mbr > > part / --fstype=ext3 --onpart=hda1 --size=500 > > part swap --fstype=swap --onpart=hda2 --noformat --recommended > > part /var --fstype=ext3 --ondisk=hda3 --size=1000 > > part /usr --fstype=ext3 --ondisk=hda4 --size=1000 --grow > > Not sure you can interchange onpart and ondisk like that... > It turns I was over-thinking the problem, and this worked: -- cut -- zerombr yes clearpart --all part / --fstype ext3 --size 500 part swap --recommended part /var --fstype ext3 --size 1000 part /usr --fstype ext3 --size 1000 --grow -- cut -- Klaus From jkeating at j2solutions.net Fri May 20 19:06:47 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Fri, 20 May 2005 12:06:47 -0700 Subject: problems with kickstart and automated partitioning In-Reply-To: <20050520061758.GP3570@thomson.net> References: <20050520061758.GP3570@thomson.net> Message-ID: <1116616007.5879.97.camel@jkeating2.hq.pogolinux.com> On Fri, 2005-05-20 at 02:17 -0400, Klaus Steden wrote: > bootloader --location=mbr > part / --fstype=ext3 --onpart=hda1 --size=500 > part swap --fstype=swap --onpart=hda2 --noformat --recommend onpart expects the partition to already be there. ondisk tells it which disk to make the partition on. -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From seberino at spawar.navy.mil Fri May 20 22:30:28 2005 From: seberino at spawar.navy.mil (seberino at spawar.navy.mil) Date: Fri, 20 May 2005 15:30:28 -0700 Subject: Help!!! Windows partitions getting clobbered!!!.... Message-ID: <20050520223028.GA27205@spawar.navy.mil> It appears 'clearpart' is the Kickstart switch to prevent Windows partitions from getting blasted when "kickstarting" Fedora Core 3. I may not be using clearpart correctly (if that is even the right way?) because I'm blowing away Windows partitions!! Any help would be greatly appreciated. Sincerely, Chris From tibbs at math.uh.edu Fri May 20 22:39:28 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Fri, 20 May 2005 17:39:28 -0500 Subject: Help!!! Windows partitions getting clobbered!!!.... In-Reply-To: <20050520223028.GA27205@spawar.navy.mil> (seberino@spawar.navy.mil's message of "Fri, 20 May 2005 15:30:28 -0700") References: <20050520223028.GA27205@spawar.navy.mil> Message-ID: >>>>> "s" == seberino writes: s> It appears 'clearpart' is the Kickstart switch to prevent Windows s> partitions from getting blasted when "kickstarting" Fedora Core 3. clearpart tells kickstart how to clear the partition table. s> I may not be using clearpart correctly (if that is even the right s> way?) because I'm blowing away Windows partitions!! Well, the docs seem pretty clear: clearpart (optional) Removes partitions from the system, prior to creation of new partitions. By default, no partitions are removed. --all Erases all partitions from the system. [...] --linux Erases all Linux partitions. --none (default) Do not remove any partitions. (from http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/sysadmin-guide/s1-kickstart2-options.html ) So "clearpart --linux" should do the trick to erase existing Linux partitions but keep others. - J< From joe at swelltech.com Fri May 20 22:42:36 2005 From: joe at swelltech.com (Joe Cooper) Date: Fri, 20 May 2005 17:42:36 -0500 Subject: Help!!! Windows partitions getting clobbered!!!.... In-Reply-To: <20050520223028.GA27205@spawar.navy.mil> References: <20050520223028.GA27205@spawar.navy.mil> Message-ID: <428E67DC.4000804@swelltech.com> clearpart --linux seberino at spawar.navy.mil wrote: > It appears 'clearpart' is the Kickstart switch to prevent > Windows partitions from getting blasted when "kickstarting" > Fedora Core 3. > > I may not be using clearpart correctly (if that is even > the right way?) because I'm blowing away Windows partitions!! > > Any help would be greatly appreciated. > > Sincerely, > > Chris > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From gressett at noao.edu Fri May 20 22:40:50 2005 From: gressett at noao.edu (Katrina Gressett) Date: Fri, 20 May 2005 15:40:50 -0700 (MST) Subject: Help!!! Windows partitions getting clobbered!!!.... In-Reply-To: <20050520223028.GA27205@spawar.navy.mil> Message-ID: Clearpart is supposed to do exactly what it says, clear partitions. Instead of doing 'clearpart --all' try 'clearpart --linux'. That should remove all the linux partions but keep any others. Good Luck, Katrina On Fri, 20 May 2005 seberino at spawar.navy.mil wrote: > It appears 'clearpart' is the Kickstart switch to prevent > Windows partitions from getting blasted when "kickstarting" > Fedora Core 3. > > I may not be using clearpart correctly (if that is even > the right way?) because I'm blowing away Windows partitions!! > > Any help would be greatly appreciated. > > Sincerely, > > Chris > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -- -- ___________________________________________________________ Katrina Gressett NSO/GONG -- Data Reduction Specialist gressett at noao.edu 520.318.8568 ___________________________________________________________ From teh at phy.anl.gov Fri May 20 22:57:44 2005 From: teh at phy.anl.gov (Ken Teh) Date: Fri, 20 May 2005 17:57:44 -0500 (CDT) Subject: Help!!! Windows partitions getting clobbered!!!.... In-Reply-To: References: Message-ID: I have a follow-up question: What does clearpart --all --drives=hda According to the docs, '--all' means 'Erases all partitions from the system' which would suggest the '--drives' is meaningless. Or does it really mean 'Erases all partitions from drive hda'. Ken On Fri, 20 May 2005, Katrina Gressett wrote: > Clearpart is supposed to do exactly what it says, clear partitions. > Instead of doing 'clearpart --all' try 'clearpart --linux'. That should > remove all the linux partions but keep any others. > > Good Luck, > Katrina > > On Fri, 20 May 2005 seberino at spawar.navy.mil wrote: > > > It appears 'clearpart' is the Kickstart switch to prevent > > Windows partitions from getting blasted when "kickstarting" > > Fedora Core 3. > > > > I may not be using clearpart correctly (if that is even > > the right way?) because I'm blowing away Windows partitions!! > > > > Any help would be greatly appreciated. > > > > Sincerely, > > > > Chris > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > From info at hostinthebox.net Fri May 20 23:04:33 2005 From: info at hostinthebox.net (dan) Date: Fri, 20 May 2005 16:04:33 -0700 Subject: Help!!! Windows partitions getting clobbered!!!.... In-Reply-To: References: Message-ID: <428E6D01.1050003@hostinthebox.net> Ken - As I understand it (and as would make syntactical sense) this would do exactly that. Clear all partitions on this specific drive. For reasons out of my control *grumble* work forbids it at this moment *grumble*, I'm not able to take a gander at the RH Kickstart guide at the moment - but I would imagine that this is correct. Hope that helps -dant Ken Teh wrote: > I have a follow-up question: What does > > clearpart --all --drives=hda > > According to the docs, '--all' means 'Erases all partitions from the system' > which would suggest the '--drives' is meaningless. Or does it really mean > 'Erases all partitions from drive hda'. > > Ken > > > On Fri, 20 May 2005, Katrina Gressett wrote: > > >>Clearpart is supposed to do exactly what it says, clear partitions. >>Instead of doing 'clearpart --all' try 'clearpart --linux'. That should >>remove all the linux partions but keep any others. >> >>Good Luck, >>Katrina >> >>On Fri, 20 May 2005 seberino at spawar.navy.mil wrote: >> >> >>>It appears 'clearpart' is the Kickstart switch to prevent >>>Windows partitions from getting blasted when "kickstarting" >>>Fedora Core 3. >>> >>>I may not be using clearpart correctly (if that is even >>>the right way?) because I'm blowing away Windows partitions!! >>> >>>Any help would be greatly appreciated. >>> >>>Sincerely, >>> >>>Chris >>> >>>_______________________________________________ >>>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 info at hostinthebox.net Fri May 20 23:07:00 2005 From: info at hostinthebox.net (dan) Date: Fri, 20 May 2005 16:07:00 -0700 Subject: Accessing boot: prompt options inside KS script In-Reply-To: <20050520102311.GR3570@thomson.net> References: <20050520102311.GR3570@thomson.net> Message-ID: <428E6D94.5050205@hostinthebox.net> Klaus Steden wrote: >>Hi All, >> >>Is it possible to access the boot: prompt options in the %pre and %post >>scripts? As in, if on my bootprompt I have: >> >>linux ks=http://some.host/some.file -ip=1.2.3.4 >> >>that I can then use the value of "-ip=" in the kickstart script itself? >> > > Christian, > > You can access these parameters by reading /proc/cmdline. > > The contents of this file are read-only, and contain all the boot args except > those that get used (and thus eaten) by the kernel (i.e. initrd= > etc.). > > It'll take a little bit of perl, awk, python, sed, or shell to turn them into > something useful, but they're there if you need them. > > hth, > Klaus > Klaus - That is awesmoe. I will most definately be using that. Thanks -dant From tibbs at math.uh.edu Fri May 20 23:24:04 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Fri, 20 May 2005 18:24:04 -0500 Subject: Help!!! Windows partitions getting clobbered!!!.... In-Reply-To: (Ken Teh's message of "Fri, 20 May 2005 17:57:44 -0500 (CDT)") References: Message-ID: >>>>> "KT" == Ken Teh writes: KT> According to the docs, '--all' means 'Erases all partitions from KT> the system' which would suggest the '--drives' is meaningless. Or KT> does it really mean 'Erases all partitions from drive hda'. The docs include an explicit example: --drives= Specifies which drives to clear partitions from. For example, the following clears all the partitions on the first two drives on the primary IDE controller: clearpart --drives=hda,hdb --all http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/sysadmin-guide/s1-kickstart2-options.html - J< From seberino at spawar.navy.mil Fri May 20 23:54:58 2005 From: seberino at spawar.navy.mil (seberino at spawar.navy.mil) Date: Fri, 20 May 2005 16:54:58 -0700 Subject: clearpart --linux --initlabel ok? In-Reply-To: <428E67DC.4000804@swelltech.com> References: <20050520223028.GA27205@spawar.navy.mil> <428E67DC.4000804@swelltech.com> Message-ID: <20050520235458.GA27494@spawar.navy.mil> Thanks for all the replies. My kickstart config file has clearpart --linux --initlabel Is there any reason the --initlabel would mess up Windoze partition? Chris From tibbs at math.uh.edu Sat May 21 00:08:32 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Fri, 20 May 2005 19:08:32 -0500 Subject: clearpart --linux --initlabel ok? In-Reply-To: <20050520235458.GA27494@spawar.navy.mil> (seberino@spawar.navy.mil's message of "Fri, 20 May 2005 16:54:58 -0700") References: <20050520223028.GA27205@spawar.navy.mil> <428E67DC.4000804@swelltech.com> <20050520235458.GA27494@spawar.navy.mil> Message-ID: >>>>> "s" == seberino writes: s> Is there any reason the --initlabel would mess up Windoze s> partition? I wouldn't assume that anything on the drive would survive having the partition table completely erased. - J< From seberino at spawar.navy.mil Sat May 21 00:12:42 2005 From: seberino at spawar.navy.mil (Christian Seberino) Date: Fri, 20 May 2005 17:12:42 -0700 Subject: clearpart --linux --initlabel ok? In-Reply-To: References: <20050520223028.GA27205@spawar.navy.mil> <428E67DC.4000804@swelltech.com> <20050520235458.GA27494@spawar.navy.mil> Message-ID: <1116634362.8757.78.camel@seberino.spawar.navy.mil> Joe Perhaps a dumb question but just so I understand.... Are you saying that --initlabel OVERRIDES the --linux part and basically says "pretend I said --all instead of --linux and blow away the drive anyways" ??? Chris On Fri, 2005-05-20 at 17:08, Jason L Tibbitts III wrote: > >>>>> "s" == seberino writes: > > s> Is there any reason the --initlabel would mess up Windoze > s> partition? > > I wouldn't assume that anything on the drive would survive having the > partition table completely erased. > > - J< > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From mlblandf at sedona.ch.intel.com Sat May 21 00:17:36 2005 From: mlblandf at sedona.ch.intel.com (Michael Blandford) Date: Fri, 20 May 2005 17:17:36 -0700 Subject: Accessing boot: prompt options inside KS script In-Reply-To: <428E6D94.5050205@hostinthebox.net> References: <20050520102311.GR3570@thomson.net> <428E6D94.5050205@hostinthebox.net> Message-ID: <428E7E20.7050303@sedona.ch.intel.com> dan wrote: > Klaus Steden wrote: > >>> Hi All, >>> >>> Is it possible to access the boot: prompt options in the %pre and %post >>> scripts? As in, if on my bootprompt I have: >>> >>> linux ks=http://some.host/some.file -ip=1.2.3.4 >>> >>> that I can then use the value of "-ip=" in the kickstart script itself? >>> >> >> Christian, >> >> You can access these parameters by reading /proc/cmdline. >> >> The contents of this file are read-only, and contain all the boot >> args except >> those that get used (and thus eaten) by the kernel (i.e. initrd= >> etc.). >> >> It'll take a little bit of perl, awk, python, sed, or shell to turn >> them into >> something useful, but they're there if you need them. >> >> hth, >> Klaus >> > > Klaus - > > That is awesmoe. I will most definately be using that. Something like this should work. eval export `cat /proc/cmdline | tr ' ' '\012' | grep =` This will set a variable for all values containing an = that get passed to the kernel. Michael Disclaimer: The content of this message is my personal opinion only and although I am an employee of Intel, the statements I make here in no way represent Intel's position on the issue, nor am I authorized to speak on behalf of Intel on this matter. From tibbs at math.uh.edu Sat May 21 00:23:39 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Fri, 20 May 2005 19:23:39 -0500 Subject: clearpart --linux --initlabel ok? In-Reply-To: <1116634362.8757.78.camel@seberino.spawar.navy.mil> (Christian Seberino's message of "Fri, 20 May 2005 17:12:42 -0700") References: <20050520223028.GA27205@spawar.navy.mil> <428E67DC.4000804@swelltech.com> <20050520235458.GA27494@spawar.navy.mil> <1116634362.8757.78.camel@seberino.spawar.navy.mil> Message-ID: >>>>> "CS" == Christian Seberino writes: CS> Are you saying that --initlabel OVERRIDES the CS> --linux part and basically says My understanding is that --initlabel instructs kickstart to completely reinitialize the partition table back to an empty state, which would render --linux somewhat irrelevant. I spent five minutes in the source attempting to verify this and was not successful. You can install the anaconda package and look in /usr/lib/anaconda yourself if you want a more definitive answer. - J< From seberino at spawar.navy.mil Sat May 21 00:28:26 2005 From: seberino at spawar.navy.mil (Christian Seberino) Date: Fri, 20 May 2005 17:28:26 -0700 Subject: clearpart --linux --initlabel ok? In-Reply-To: References: <20050520223028.GA27205@spawar.navy.mil> <428E67DC.4000804@swelltech.com> <20050520235458.GA27494@spawar.navy.mil> <1116634362.8757.78.camel@seberino.spawar.navy.mil> Message-ID: <1116635306.8757.87.camel@seberino.spawar.navy.mil> Jason I'm *very* thankful for your reply! You gave me hope and something to try instead of flailing around in the dark. :) Chris On Fri, 2005-05-20 at 17:23, Jason L Tibbitts III wrote: > >>>>> "CS" == Christian Seberino writes: > > CS> Are you saying that --initlabel OVERRIDES the > > CS> --linux part and basically says > > My understanding is that --initlabel instructs kickstart to completely > reinitialize the partition table back to an empty state, which would > render --linux somewhat irrelevant. I spent five minutes in the > source attempting to verify this and was not successful. You can > install the anaconda package and look in /usr/lib/anaconda yourself if > you want a more definitive answer. > > - J< > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From mikem.rtp at gmail.com Sat May 21 00:32:01 2005 From: mikem.rtp at gmail.com (Mike McLean) Date: Fri, 20 May 2005 20:32:01 -0400 Subject: clearpart --linux --initlabel ok? In-Reply-To: References: <20050520223028.GA27205@spawar.navy.mil> <428E67DC.4000804@swelltech.com> <20050520235458.GA27494@spawar.navy.mil> <1116634362.8757.78.camel@seberino.spawar.navy.mil> Message-ID: <4f50e068050520173228fadad9@mail.gmail.com> On 5/20/05, Jason L Tibbitts III wrote: > My understanding is that --initlabel instructs kickstart to completely > reinitialize the partition table back to an empty state, which would > render --linux somewhat irrelevant. That is correct. The --initlabel option is pretty extreme, and is not normally needed. From brandono at berkeley.edu Sat May 21 01:19:35 2005 From: brandono at berkeley.edu (Brandon Ooi) Date: Fri, 20 May 2005 18:19:35 -0700 Subject: Missing Operating System problem Message-ID: <428E8CA7.5060001@berkeley.edu> Hi, We're using Fedora Core 3 and after kickstarts, the machines say "Missing Operating System" on post. We have the following line in our kickstart which should work... bootloader --location=mbr ==append="selinux=0 rhgb quiet" Even more mysterious is that if we attempt to reimage the machine again (exact same procedure)... it works. As you can see, trying to trace this problem requires me to break out a new machines. Anybody have any ideas to remedy this? Brandon From error27 at gmail.com Sat May 21 02:54:50 2005 From: error27 at gmail.com (Dan Carpenter) Date: Fri, 20 May 2005 19:54:50 -0700 Subject: Missing Operating System problem In-Reply-To: <428E8CA7.5060001@berkeley.edu> References: <428E8CA7.5060001@berkeley.edu> Message-ID: > bootloader --location=mbr ==append="selinux=0 rhgb quiet" You have a typo. I'm sure it's correct in the actual kickstart, but I have to point it out or it will bother me like a sore tooth. The problem is really that GRUB sucks on the FC3 install CDs. Are you using the 64bit CPUs? If so then GRUB _really_ sucks. I haven't tried the FC4test releases but hopefully it's fixed. regards, dan carpenter From Christian.Rohrmeier at SCHERING.DE Sat May 21 16:04:50 2005 From: Christian.Rohrmeier at SCHERING.DE (Christian.Rohrmeier at SCHERING.DE) Date: Sat, 21 May 2005 18:04:50 +0200 Subject: clearpart --linux --initlabel ok? Message-ID: Hi All, I ran into this problem recently. I used "clearpart --linux". My admins found that after install the system didn't boot properly. It turns out the server had previously had a non-linux OS installed. Therefore I find that "--initlabel" is very important to guarantee that anaconda will be able to partition the drive, unless ofcourse you know for sure that there was only Linux on your server perviously. -Christian _________________ Christian Rohrmeier Schering AG Corporate IT - Infrastructure and Services Computer Systems and Operations System Administration - Research and Development Tel +49 30 468 15794 Fax +49 30 468 95794 Mike McLean Sent by: To: Discussion list about Kickstart kickstart-list-bounces cc: @redhat.com Subject: Re: clearpart --linux --initlabel ok? 21.05.2005 02:32 Please respond to Mike McLean; Please respond to Discussion list about Kickstart On 5/20/05, Jason L Tibbitts III wrote: > My understanding is that --initlabel instructs kickstart to completely > reinitialize the partition table back to an empty state, which would > render --linux somewhat irrelevant. That is correct. The --initlabel option is pretty extreme, and is not normally needed. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Christian.Rohrmeier at SCHERING.DE Sat May 21 16:11:37 2005 From: Christian.Rohrmeier at SCHERING.DE (Christian.Rohrmeier at SCHERING.DE) Date: Sat, 21 May 2005 18:11:37 +0200 Subject: Unwanted RHEL 2.1 "Add User ID" dialog duing kickstart install. Message-ID: Hello again, I notice that during a RHEL 2.1 kickstart install, unlike with RHEL 3 and 4, Anaconda asks the user for information on adding a non-root user. This is not very cool, because my usermanagement happens after installation. I don't want people being asked to add userIDs during install. Is there a way to turn off the "Add User ID" dialog? Thanks, -Christian _________________ Christian Rohrmeier Schering AG Corporate IT - Infrastructure and Services Computer Systems and Operations System Administration - Research and Development Tel +49 30 468 15794 Fax +49 30 468 95794 From Christian.Rohrmeier at SCHERING.DE Sat May 21 16:20:40 2005 From: Christian.Rohrmeier at SCHERING.DE (Christian.Rohrmeier at SCHERING.DE) Date: Sat, 21 May 2005 18:20:40 +0200 Subject: Accessing boot: prompt options inside KS script Message-ID: Hi Klaus, Thanks for the /proc/cmd tip. I extracted the commandline options like this: ip=`grep ip /proc/cmdline | sed 's/.*ip=//' |sed 's/ .*//'` The first grep is there to return a blank line incase the option is not present at all (else the second sed would return the portion of the line after the first whitespace even though ip= was not present). The two seds split up the work, the first returns the entire command line past ip=, and the second then retum the first part up to the first blank space. Since all the command line options are space delimted, this works just fine. Busybox doesn't include awk, otherwise it would have been easier to just use that. But awk is derived from sed, so sed does fine as well. -Christian _________________ Christian Rohrmeier Schering AG Corporate IT - Infrastructure and Services Computer Systems and Operations System Administration - Research and Development Tel +49 30 468 15794 Fax +49 30 468 95794 Klaus Steden To: Discussion list about Kickstart Sent by: cc: kickstart-list-bounces Subject: Re: Accessing boot: prompt options inside KS script @redhat.com 20.05.2005 12:23 Please respond to Discussion list about Kickstart > > Hi All, > > Is it possible to access the boot: prompt options in the %pre and %post > scripts? As in, if on my bootprompt I have: > > linux ks=http://some.host/some.file -ip=1.2.3.4 > > that I can then use the value of "-ip=" in the kickstart script itself? > Christian, You can access these parameters by reading /proc/cmdline. The contents of this file are read-only, and contain all the boot args except those that get used (and thus eaten) by the kernel (i.e. initrd= etc.). It'll take a little bit of perl, awk, python, sed, or shell to turn them into something useful, but they're there if you need them. hth, Klaus _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From jkeating at j2solutions.net Sat May 21 16:23:26 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Sat, 21 May 2005 09:23:26 -0700 Subject: clearpart --linux --initlabel ok? In-Reply-To: References: Message-ID: <1116692606.3532.2.camel@yoda.loki.me> On Sat, 2005-05-21 at 18:04 +0200, Christian.Rohrmeier at SCHERING.DE wrote: > > Hi All, > > I ran into this problem recently. I used "clearpart --linux". My > admins > found that after install the system didn't boot properly. It turns out > the > server had previously had a non-linux OS installed. Therefore I find > that > "--initlabel" is very important to guarantee that anaconda will be > able to > partition the drive, unless ofcourse you know for sure that there was > only > Linux on your server perviously. Be wary of this. initlabel will write a new partition label to your disk, and that in turn will remove ALL partitions. It will NOT honor --linux. -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From klaus.steden at thomson.net Sat May 21 19:17:22 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Sat, 21 May 2005 15:17:22 -0400 Subject: Accessing boot: prompt options inside KS script In-Reply-To: References: Message-ID: <20050521191722.GW3570@thomson.net> > Hi Klaus, > > Thanks for the /proc/cmd tip. > > I extracted the commandline options like this: > > ip=`grep ip /proc/cmdline | sed 's/.*ip=//' |sed 's/ .*//'` > > The first grep is there to return a blank line incase the option is not > present at all (else the second sed would return the portion of the line > after the first whitespace even though ip= was not present). > > The two seds split up the work, the first returns the entire command line > past ip=, and the second then retum the first part up to the first blank > space. Since all the command line options are space delimted, this works > just fine. > > Busybox doesn't include awk, otherwise it would have been easier to just > use that. But awk is derived from sed, so sed does fine as well. > Christian - here is the sort of syntax I use to do that (just in case anyone likes other ways, too) if grep -i -q "ip=[0-9]" /proc/cmdline then ip=`cat /proc/cmdline | sed 's/.*ip=\([^ ]*\).*/\1/'` fi The sed expression returns a copy of everything -after- "ip=" up to but not including the first blank space (so, the value you want). Klaus From phr at doc.ic.ac.uk Sun May 22 13:36:34 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Sun, 22 May 2005 14:36:34 +0100 (BST) Subject: Accessing boot: prompt options inside KS script In-Reply-To: References: Message-ID: On Sat, 21 May 2005 Christian.Rohrmeier at SCHERING.DE wrote: >I extracted the commandline options like this: >ip=`grep ip /proc/cmdline | sed 's/.*ip=//' |sed 's/ .*//'` Excessive sloping matchstick substitutions in sed make my eyes water - I find it conceptually easier to break at whitespace into separate lines: $ busybox tr -s ' ' '\n' < /proc/cmdline | busybox grep ^ip= One can of course remove the "busybox" wrapper in the kickstart environment itself, although it doesn't hurt. Cheers, Phil From mikem.rtp at gmail.com Sun May 22 18:49:53 2005 From: mikem.rtp at gmail.com (Mike McLean) Date: Sun, 22 May 2005 14:49:53 -0400 Subject: Accessing boot: prompt options inside KS script In-Reply-To: References: Message-ID: <4f50e06805052211495328853a@mail.gmail.com> > Excessive sloping matchstick substitutions in sed make my eyes water - I > find it conceptually easier to break at whitespace into separate lines: I just use bash itself. It's a little more typing, but more adaptable and readable. set -- $(cat /proc/cmdline) for arg in "$@"; do case $arg in ip=*) ip=${arg#ip=} ;; esac done Of course, you could use --interpreter /usr/bin/python, in which case, you'd have: for arg in open('/proc/cmdline').read().split(): if arg[0:3] == "ip=": ip = arg[3:] From andrew.w.robinson at mms.gov Mon May 23 18:35:00 2005 From: andrew.w.robinson at mms.gov (Robinson, Andrew W.) Date: Mon, 23 May 2005 12:35:00 -0600 Subject: Network kickstart weirdness Message-ID: <379313C94B76D2119AB60008C7A402E40CA4C2D1@IMSNOLAA> I think I have encountered this problem before, but cannot remember the solution. I am trying to perform a network kickstart on a Dell PE 2650 server. The OS is RHEL 3 U4. No matter how I try this, the client seems not to recognize that the kickstart file is present. It always asks for network information. The system successfully obtained an address through dhcp for the pxe-boot, but then it cannot seem to get an address to start the installation process. I am perplexed. I do not know if this is a network problem, a problem with the kickstart file, or maybe a problem with anaconda. Can anyone suggest any clues that might enable me to figure out what is wrong? Thanks! Andrew Robinson From SSeremeth at anacomp.com Mon May 23 18:40:31 2005 From: SSeremeth at anacomp.com (Seremeth, Stephen) Date: Mon, 23 May 2005 14:40:31 -0400 Subject: Network kickstart weirdness Message-ID: <66911C8E1F2DA24EB29DE289297A135E4856B7@usrd104> If you ask me, it's this bug: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=151872 But having said that, if you change the port on your switch to enable portFast, I predict you'll be all set. That's what we had to do. PXE was getting an IP just fine -- it was Kickstart/Anaconda that was choking in this regard. Regards, Steve P.S. We were plugged into a Cisco Catalyst 4000(?). > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > Robinson, Andrew W. > Sent: Monday, May 23, 2005 2:35 PM > To: 'Discussion list about Kickstart' > Subject: Network kickstart weirdness > > I think I have encountered this problem before, but cannot > remember the > solution. I am trying to perform a network kickstart on a Dell PE 2650 > server. The OS is RHEL 3 U4. No matter how I try this, the > client seems not > to recognize that the kickstart file is present. It always > asks for network > information. The system successfully obtained an address > through dhcp for > the pxe-boot, but then it cannot seem to get an address to start the > installation process. I am perplexed. I do not know if this > is a network > problem, a problem with the kickstart file, or maybe a problem with > anaconda. Can anyone suggest any clues that might enable me > to figure out > what is wrong? > > Thanks! > > Andrew Robinson > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From dan at half-asleep.com Mon May 23 18:44:36 2005 From: dan at half-asleep.com (Daniel Segall) Date: Mon, 23 May 2005 14:44:36 -0400 (EDT) Subject: Network kickstart weirdness In-Reply-To: <66911C8E1F2DA24EB29DE289297A135E4856B7@usrd104> References: <66911C8E1F2DA24EB29DE289297A135E4856B7@usrd104> Message-ID: <44992.192.80.55.74.1116873876.squirrel@webmail.half-asleep.com> It could also be the 2nd nic in the server. Anaconda is completely retarded when it comes to kickstarting a system with multiple nics. Try connecting both nics to your kickstart lan just in case it's swapping eth0 to eth1 or something. -Dan > If you ask me, it's this bug: > > https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=151872 > > But having said that, if you change the port on your switch to enable > portFast, I predict you'll be all set. > > That's what we had to do. PXE was getting an IP just fine -- it was > Kickstart/Anaconda that was choking in this regard. > > Regards, > > Steve > > P.S. We were plugged into a Cisco Catalyst 4000(?). > >> -----Original Message----- >> From: kickstart-list-bounces at redhat.com >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of >> Robinson, Andrew W. >> Sent: Monday, May 23, 2005 2:35 PM >> To: 'Discussion list about Kickstart' >> Subject: Network kickstart weirdness >> >> I think I have encountered this problem before, but cannot >> remember the >> solution. I am trying to perform a network kickstart on a Dell PE 2650 >> server. The OS is RHEL 3 U4. No matter how I try this, the >> client seems not >> to recognize that the kickstart file is present. It always >> asks for network >> information. The system successfully obtained an address >> through dhcp for >> the pxe-boot, but then it cannot seem to get an address to start the >> installation process. I am perplexed. I do not know if this >> is a network >> problem, a problem with the kickstart file, or maybe a problem with >> anaconda. Can anyone suggest any clues that might enable me >> to figure out >> what is wrong? >> >> Thanks! >> >> Andrew Robinson >> >> _______________________________________________ >> 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 andrew.w.robinson at mms.gov Mon May 23 20:37:42 2005 From: andrew.w.robinson at mms.gov (Robinson, Andrew W.) Date: Mon, 23 May 2005 14:37:42 -0600 Subject: Network kickstart weirdness Message-ID: <379313C94B76D2119AB60008C7A402E40CA4C2D2@IMSNOLAA> You were correct sir! portFast fixed the problem. I think I need to put up a sign in my cube that reads "portFast!". Thanks! Andrew > From: Seremeth, Stephen [mailto:SSeremeth at anacomp.com] > > If you ask me, it's this bug: > > https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=151872 > > But having said that, if you change the port on your switch to enable > portFast, I predict you'll be all set. > > That's what we had to do. PXE was getting an IP just fine -- it was > Kickstart/Anaconda that was choking in this regard. > > Regards, > > Steve > > P.S. We were plugged into a Cisco Catalyst 4000(?). > > > -----Original Message----- > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > > Robinson, Andrew W. > > Sent: Monday, May 23, 2005 2:35 PM > > To: 'Discussion list about Kickstart' > > Subject: Network kickstart weirdness > > > > I think I have encountered this problem before, but cannot > > remember the > > solution. I am trying to perform a network kickstart on a > Dell PE 2650 > > server. The OS is RHEL 3 U4. No matter how I try this, the > > client seems not > > to recognize that the kickstart file is present. It always > > asks for network > > information. The system successfully obtained an address > > through dhcp for > > the pxe-boot, but then it cannot seem to get an address to start the > > installation process. I am perplexed. I do not know if this > > is a network > > problem, a problem with the kickstart file, or maybe a problem with > > anaconda. Can anyone suggest any clues that might enable me > > to figure out > > what is wrong? > > > > Thanks! > > > > Andrew Robinson > > > > _______________________________________________ > > 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 vectorz2 at gmail.com Mon May 23 20:47:48 2005 From: vectorz2 at gmail.com (John Tran) Date: Mon, 23 May 2005 13:47:48 -0700 Subject: Network kickstart weirdness In-Reply-To: <66911C8E1F2DA24EB29DE289297A135E4856B7@usrd104> References: <66911C8E1F2DA24EB29DE289297A135E4856B7@usrd104> Message-ID: Do you know if there is any other way around this bug, except for disabling PortFast on the switch? I don't have access to the network gear, and I've been wrestling w/ this issue forever. On 5/23/05, Seremeth, Stephen wrote: > If you ask me, it's this bug: > > https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=151872 > > But having said that, if you change the port on your switch to enable > portFast, I predict you'll be all set. > > That's what we had to do. PXE was getting an IP just fine -- it was > Kickstart/Anaconda that was choking in this regard. > > Regards, > > Steve > > P.S. We were plugged into a Cisco Catalyst 4000(?). > > > -----Original Message----- > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of > > Robinson, Andrew W. > > Sent: Monday, May 23, 2005 2:35 PM > > To: 'Discussion list about Kickstart' > > Subject: Network kickstart weirdness > > > > I think I have encountered this problem before, but cannot > > remember the > > solution. I am trying to perform a network kickstart on a Dell PE 2650 > > server. The OS is RHEL 3 U4. No matter how I try this, the > > client seems not > > to recognize that the kickstart file is present. It always > > asks for network > > information. The system successfully obtained an address > > through dhcp for > > the pxe-boot, but then it cannot seem to get an address to start the > > installation process. I am perplexed. I do not know if this > > is a network > > problem, a problem with the kickstart file, or maybe a problem with > > anaconda. Can anyone suggest any clues that might enable me > > to figure out > > what is wrong? > > > > Thanks! > > > > Andrew Robinson > > > > _______________________________________________ > > 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 dpeterson7 at nextag.com Mon May 23 20:53:56 2005 From: dpeterson7 at nextag.com (Dan Peterson) Date: Mon, 23 May 2005 13:53:56 -0700 Subject: Network kickstart weirdness Message-ID: <452262D678B33E4E956977866B2C3FE2015448B1@corpmail2.corp.nextag.com> We ran into this as well. It seems to be a problem with the E1000 adaptor (which the 2650 and 1850 [and others] are sporting). Check out http://www.redhat.com/archives/kickstart-list/2005-March/msg00087.html We modified init.c as described in that note; rebuilt loader and put it back in the initrd.img. Problem solved. Not clean, not pretty, but effective! If you need more detailed steps let me know. Dan Peterson. -----Original Message----- From: Robinson, Andrew W. [mailto:andrew.w.robinson at mms.gov] Sent: Monday, May 23, 2005 11:35 AM To: Discussion list about Kickstart Subject: Network kickstart weirdness I think I have encountered this problem before, but cannot remember the solution. I am trying to perform a network kickstart on a Dell PE 2650 server. The OS is RHEL 3 U4. No matter how I try this, the client seems not to recognize that the kickstart file is present. It always asks for network information. The system successfully obtained an address through dhcp for the pxe-boot, but then it cannot seem to get an address to start the installation process. I am perplexed. I do not know if this is a network problem, a problem with the kickstart file, or maybe a problem with anaconda. Can anyone suggest any clues that might enable me to figure out what is wrong? Thanks! Andrew Robinson _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list CONFIDENTIALITY NOTICE ======================= This email message and any attachments are for the exclusive use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message along with any attachments, from your computer system. If you are the intended recipient, please be advised that the content of this message is subject to access, review and disclosure by the sender's Email System Administrator. From plarsen at famlarsen.homelinux.com Mon May 23 21:49:43 2005 From: plarsen at famlarsen.homelinux.com (Peter Larsen) Date: Mon, 23 May 2005 17:49:43 -0400 Subject: Network kickstart weirdness In-Reply-To: <452262D678B33E4E956977866B2C3FE2015448B1@corpmail2.corp.nextag.com> References: <452262D678B33E4E956977866B2C3FE2015448B1@corpmail2.corp.nextag.com> Message-ID: <42924FF7.6010104@famlarsen.homelinux.com> It was my big mystery too - until I realized that the devil was in our switch. The PXE seems to send a few "up/down" signals to our switch so it finds DHCP rather "fast" ... 2-4 seconds. However, once booted and the RH installer runs through the same steps, the switch is taking forever to reset/open the port. This results in a time-out, and kickstart asking for network information - it assumes something is wrong. My "solution" is no more than just hitting "ok" when being asked for network information, and on the second try our switch has finally gotten around to opening up again, and things work. I don't know if that helps you - it's not a fool proof situation; but in our cases we only use Kickstart to build a baseline system on new boxes. So it's no big deal to us. Regards Peter Larsen Dan Peterson wrote: > We ran into this as well. It seems to be a problem with the E1000 adaptor > (which the 2650 and 1850 [and others] are sporting). Check out > http://www.redhat.com/archives/kickstart-list/2005-March/msg00087.html > > We modified init.c as described in that note; rebuilt loader and put it back > in the initrd.img. Problem solved. Not clean, not pretty, but effective! > > If you need more detailed steps let me know. > > Dan Peterson. > > -----Original Message----- > From: Robinson, Andrew W. [mailto:andrew.w.robinson at mms.gov] > Sent: Monday, May 23, 2005 11:35 AM > To: Discussion list about Kickstart > Subject: Network kickstart weirdness > > I think I have encountered this problem before, but cannot remember the > solution. I am trying to perform a network kickstart on a Dell PE 2650 > server. The OS is RHEL 3 U4. No matter how I try this, the client seems not > to recognize that the kickstart file is present. It always asks for network > information. The system successfully obtained an address through dhcp for > the pxe-boot, but then it cannot seem to get an address to start the > installation process. I am perplexed. I do not know if this is a network > problem, a problem with the kickstart file, or maybe a problem with > anaconda. Can anyone suggest any clues that might enable me to figure out > what is wrong? > > Thanks! > > Andrew Robinson > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > CONFIDENTIALITY NOTICE > ======================= > This email message and any attachments are for the exclusive use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message along with any attachments, from your computer system. If you are the intended recipient, please be advised that the content of this message is subject to access, review and disclosure by the sender's Email System Administrator. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From vectorz2 at gmail.com Mon May 23 22:12:13 2005 From: vectorz2 at gmail.com (John Tran) Date: Mon, 23 May 2005 15:12:13 -0700 Subject: Network kickstart weirdness In-Reply-To: <452262D678B33E4E956977866B2C3FE2015448B1@corpmail2.corp.nextag.com> References: <452262D678B33E4E956977866B2C3FE2015448B1@corpmail2.corp.nextag.com> Message-ID: Dan, I could definitely use the detailed steps you mentioned. TIA! On 5/23/05, Dan Peterson wrote: > We ran into this as well. It seems to be a problem with the E1000 adaptor > (which the 2650 and 1850 [and others] are sporting). Check out > http://www.redhat.com/archives/kickstart-list/2005-March/msg00087.html > > We modified init.c as described in that note; rebuilt loader and put it back > in the initrd.img. Problem solved. Not clean, not pretty, but effective! > > If you need more detailed steps let me know. > > Dan Peterson. From gbromley at intstar.com Tue May 24 04:14:48 2005 From: gbromley at intstar.com (Gareth Bromley) Date: Tue, 24 May 2005 05:14:48 +0100 Subject: Network kickstart weirdness In-Reply-To: <452262D678B33E4E956977866B2C3FE2015448B1@corpmail2.corp.nextag.com> References: <452262D678B33E4E956977866B2C3FE2015448B1@corpmail2.corp.nextag.com> Message-ID: <4292AA38.3020008@intstar.com> Dan Peterson wrote: > We ran into this as well. It seems to be a problem with the E1000 adaptor > (which the 2650 and 1850 [and others] are sporting). Check out > http://www.redhat.com/archives/kickstart-list/2005-March/msg00087.html From mem doesn't just affect E1000's. Its actually related to the loader/kernel 'resetting' the NIC interface. The switch, in non-portfast mode, then goes through its steps of LISTEN, LEARNING, FORWARDING etc related to standard spanning tree behaviour to protect against bridging loops in the network. In standard spanning tree this can take upto 30 seconds, although variants do exist to reduce this time, which would disrupt requests for DHCP information and/or fetching of Kickstart files. > We modified init.c as described in that note; rebuilt loader and put it back > in the initrd.img. Problem solved. Not clean, not pretty, but effective! Just a hack to get round the STP timers. Maybe a cleaner option would not be to reset the interface if booted via PXE, although how would wouold detect this I don't know. HTH G From klaus.steden at thomson.net Tue May 24 21:28:23 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Tue, 24 May 2005 17:28:23 -0400 Subject: obnoxious/unnecessary package dependencies In-Reply-To: <4f50e06805052211495328853a@mail.gmail.com> References: <4f50e06805052211495328853a@mail.gmail.com> Message-ID: <20050524212823.GY3570@thomson.net> Okay, I've got my kickstart setup working now ... but I'm a little mystified as to why it's installing packages that are nowhere near what I've specified. This is the packages list I'm specifying: -- cut -- @ admin-tools @ base-x @ compat-arch-support @ legacy-network-server @ system-tools e2fsprogs grub kernel portmap yp-tools ypbind -am-utils -autofs -- cut -- ... but this is what gets installed: -- cut -- @ legacy-network-server @ icelandic-support @ german-support @ arabic-support @ danish-support @ catalan-support @ bengali-support @ spanish-support @ brazilian-support @ hebrew-support @ estonian-support @ french-support @ system-tools @ turkish-support @ portuguese-support @ finnish-support @ norwegian-support @ admin-tools @ romanian-support @ tamil-support @ japanese-support @ hindi-support @ punjabi-support @ gujarati-support @ base-x @ chinese-support @ korean-support @ welsh-support @ italian-support @ slovak-support @ slovenian-support @ ukrainian-support @ czech-support @ bulgarian-support @ hungarian-support @ swedish-support @ british-support @ polish-support @ dutch-support @ russian-support @ greek-support @ compat-arch-support kernel -autofs grub portmap yp-tools ypbind -am-utils e2fsprogs -- cut -- Nothing against the Swedish, Punjabi, Czechs, or Bulgarians, but I really don't need to support those locales, fonts, dictionaries, or desktops. I really don't want this stuff installed, and I'm sure I didn't ask for it. What gives? Is there a tidier way to get a stripped-down system without having to specify everything manually? thanks, Klaus From jkeating at j2solutions.net Tue May 24 22:00:58 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Tue, 24 May 2005 15:00:58 -0700 Subject: obnoxious/unnecessary package dependencies In-Reply-To: <20050524212823.GY3570@thomson.net> References: <4f50e06805052211495328853a@mail.gmail.com> <20050524212823.GY3570@thomson.net> Message-ID: <1116972058.5520.20.camel@jkeating2.hq.pogolinux.com> On Tue, 2005-05-24 at 17:28 -0400, Klaus Steden wrote: > Okay, I've got my kickstart setup working now ... but I'm a little > mystified > as to why it's installing packages that are nowhere near what I've > specified. > This is the packages list I'm specifying: Can you post your language line? -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From klaus.steden at thomson.net Tue May 24 22:20:12 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Tue, 24 May 2005 18:20:12 -0400 Subject: obnoxious/unnecessary package dependencies In-Reply-To: <1116972058.5520.20.camel@jkeating2.hq.pogolinux.com> References: <4f50e06805052211495328853a@mail.gmail.com> <20050524212823.GY3570@thomson.net> <1116972058.5520.20.camel@jkeating2.hq.pogolinux.com> Message-ID: <20050524222012.GZ3570@thomson.net> > On Tue, 2005-05-24 at 17:28 -0400, Klaus Steden wrote: > > Okay, I've got my kickstart setup working now ... but I'm a little > > mystified > > as to why it's installing packages that are nowhere near what I've > > specified. > > This is the packages list I'm specifying: > > Can you post your language line? > Sure. lang en_US.UTF-8 langsupport --default=en_US.UTF-8 keyboard us If you want the anaconda-ks.cfg that gets saved vs. the one that's used, I can provide those easily, too. Klaus From jkeating at j2solutions.net Tue May 24 23:24:36 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Tue, 24 May 2005 16:24:36 -0700 Subject: obnoxious/unnecessary package dependencies In-Reply-To: <20050524222012.GZ3570@thomson.net> References: <4f50e06805052211495328853a@mail.gmail.com> <20050524212823.GY3570@thomson.net> <1116972058.5520.20.camel@jkeating2.hq.pogolinux.com> <20050524222012.GZ3570@thomson.net> Message-ID: <1116977076.5703.0.camel@jkeating2.hq.pogolinux.com> On Tue, 2005-05-24 at 18:20 -0400, Klaus Steden wrote: > Sure. > > lang en_US.UTF-8 > langsupport --default=en_US.UTF-8 > keyboard us > > If you want the anaconda-ks.cfg that gets saved vs. the one that's > used, I can > provide those easily, too. Remove the langsupport line. Just the lang line will ensure that you get english. The lansupport line will pull in all other languages, and make english the default. Probably not what you wanted (: -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From klaus.steden at thomson.net Wed May 25 00:07:01 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Tue, 24 May 2005 20:07:01 -0400 Subject: obnoxious/unnecessary package dependencies In-Reply-To: <1116977076.5703.0.camel@jkeating2.hq.pogolinux.com> References: <4f50e06805052211495328853a@mail.gmail.com> <20050524212823.GY3570@thomson.net> <1116972058.5520.20.camel@jkeating2.hq.pogolinux.com> <20050524222012.GZ3570@thomson.net> <1116977076.5703.0.camel@jkeating2.hq.pogolinux.com> Message-ID: <20050525000701.GC3570@thomson.net> > > Sure. > > > > lang en_US.UTF-8 > > langsupport --default=en_US.UTF-8 > > keyboard us > > > > If you want the anaconda-ks.cfg that gets saved vs. the one that's > > used, I can > > provide those easily, too. > > Remove the langsupport line. Just the lang line will ensure that you > get english. The lansupport line will pull in all other languages, and > make english the default. Probably not what you wanted (: > Jesse - thanks for the info - is there an FAQ on the various package groupings? For example, I don't need anything more than basic X11 support (i.e. libXt.so, libICE.so, but no need for GNOME, KDE, xterm or other X11 apps), and definitely don't need any support for printers whatsoever. Is there a syntax for ignoring or not installing entire groups of packages? Klaus From dan at half-asleep.com Wed May 25 02:35:58 2005 From: dan at half-asleep.com (Daniel Segall) Date: Tue, 24 May 2005 22:35:58 -0400 Subject: obnoxious/unnecessary package dependencies In-Reply-To: <20050525000701.GC3570@thomson.net> References: <4f50e06805052211495328853a@mail.gmail.com> <20050524212823.GY3570@thomson.net> <1116972058.5520.20.camel@jkeating2.hq.pogolinux.com> <20050524222012.GZ3570@thomson.net> <1116977076.5703.0.camel@jkeating2.hq.pogolinux.com> <20050525000701.GC3570@thomson.net> Message-ID: <4293E48E.20908@half-asleep.com> If all you want is the basic x libs, you dont need to install X at all. What I recommend for a good starting point is just @ Server & @ Development Tools. If you look in base/comps.xml, you can see all of the package groups. -Dan Klaus Steden wrote: >>>Sure. >>> >>>lang en_US.UTF-8 >>>langsupport --default=en_US.UTF-8 >>>keyboard us >>> >>>If you want the anaconda-ks.cfg that gets saved vs. the one that's >>>used, I can >>>provide those easily, too. >> >>Remove the langsupport line. Just the lang line will ensure that you >>get english. The lansupport line will pull in all other languages, and >>make english the default. Probably not what you wanted (: >> > > Jesse - thanks for the info - is there an FAQ on the various package > groupings? For example, I don't need anything more than basic X11 support > (i.e. libXt.so, libICE.so, but no need for GNOME, KDE, xterm or other X11 > apps), and definitely don't need any support for printers whatsoever. > > Is there a syntax for ignoring or not installing entire groups of packages? > > Klaus > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From svenkat at india.ti.com Wed May 25 04:51:41 2005 From: svenkat at india.ti.com (Venkatachalam, Saravanakumar) Date: Wed, 25 May 2005 10:21:41 +0530 Subject: NFS install and interactive asking of static ip In-Reply-To: References: <428C1582.9050307@india.ti.com> Message-ID: <4294045D.1000704@india.ti.com> L B wrote: >On 5/19/05, Venkatachalam, Saravanakumar wrote: > > >>Hi, >> >>Try providing this option >> >>network --device eth0 --bootproto static >> >> > >This doesn't work, ks didn't ask for the network configuration, it >just misconfigured the IP and I had no network configuration at all >after the install finished. > > > This issue is still not fixed: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=30190 -------------- next part -------------- An HTML attachment was scrubbed... URL: From brilong at cisco.com Tue May 31 19:46:10 2005 From: brilong at cisco.com (Brian Long) Date: Tue, 31 May 2005 15:46:10 -0400 Subject: yum in kickstart %post and HP Insight Mgr Message-ID: <1117568770.3743.32.camel@localhost.localdomain> I'm trying to use yum in the %post section of kickstart. With well- written RPMs, this works great. However, I've come across the lovely HP Insight Manager RPMS (hpasm, hprsm, etc) which seem to run "service hpasm start" during the RPM postinstall (yuck). This wouldn't be too bad if the daemons actually disassociated from the running terminal. I notice yum install the group "yum groupinstall Insight-Manager" and yum exits, but the rest of the kickstart %post doesn't run. If I look at the hpasmd processes, they're tied to tty1 and owned by init :( If I hit Alt-F2 in kickstart and chroot /mnt/sysimage, service hpasm stop, the rest of the kickstart %post finishes. I also strace'd the /tmp/ks- script process and it's stuck at waitpid(-1, ) I cannot wait on a vendor escalation at this time, so I see one method of a workaround. Move the yum installs to a firstboot scenario instead of %post. I'd really like to find another way around this. Any ideas would be appreciated. Thanks. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From SSeremeth at anacomp.com Tue May 31 19:46:54 2005 From: SSeremeth at anacomp.com (Seremeth, Stephen) Date: Tue, 31 May 2005 15:46:54 -0400 Subject: Kickstart dies after post-install Message-ID: <66911C8E1F2DA24EB29DE289297A135E4856CC@usrd104> Hello - I get almost all the way through my kickstart process and I get a Traceback: ------------------------------------------------------------------------ ------ Traceback (most recent call last): File "/usr/bin/anaconda", line 1173, in ? intf.run(id, dispatch, configFileData) File "/usr/src/build/500926-i386/install//usr/lib/anaconda/text.py", line 480, in run exec s File "", line 1, in ? ImportError: cannot import name FinishedWindow Local variables in innermost frame: lang: en_US.UTF-8 nextWindow: None langname: English win: self: args: () classNames: ('FinishedWindow',) dispatch: oldlang: None s: from complete_text import FinishedWindow; nextWindow = FinishedWindow lastrc: 0 setupForInstall: progress_text.setupForInstall file: complete_text rc: 0 step: 0 foo: complete id: configFileData: {'TitleBar': 'pixmaps/anaconda_header.png', 'Title': 'Red Hat Linux Beta', 'Splashscreen': 'pixmaps/first.png', 'WelcomeScreen': 'pixmaps/spl ash.png'} ---------------------------------------------------------------------- If I reboot the box, it's usable and the install appears to have gone OK. More info: ---------------------------------------------------------------------- * moving (1) to step copylogs * Copying anaconda logs * moving (1) to step dopostaction * Running kickstart %post script(s) * All kickstart %post script(s) have been run * moving (1) to step methodcomplete * moving (1) to step complete ---------------------------------------------------------------------- There are some ugly errors in the anacdump.txt file, also: ------------------------------------------------------------------------ ---------------- <3>audit(1107976290.258:0): avc: denied { getattr } for pid=1886 exe=/sbin/portmap path=/dev/null dev=ram0 ino=40 scontext=user_u:system_r:portmap_t tcont ext=system_u:object_r:file_t tclass=chr_file <4>RPC: garbage, exit EIO <4>RPC: garbage, exit EIO <4>RPC: garbage, exit EIO <4>RPC: garbage, exit EIO <4>RPC: garbage, exit EIO <4>RPC: garbage, exit EIO <3>Buffer I/O error on device loop0, logical block 4105 <4>RPC: garbage, exit EIO <3>Buffer I/O error on device loop0, logical block 4105 <4>RPC: garbage, exit EIO <3>Buffer I/O error on device loop0, logical block 4105 <4>Error -3 while decompressing! <4>f881f42c(-16817932)->ff19e000(4096) <4>RPC: garbage, exit EIO <3>Buffer I/O error on device loop0, logical block 4105 <4>Error -3 while decompressing! <4>f881b100(-16817120)->ff19f000(4096) <4>RPC: garbage, exit EIO ------------------------------------------------------------------------ ---- Any thoughts? Thanks for any tips - Steve ___________________________ Steve Seremeth System Engineer sseremeth at anacomp.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From phr at doc.ic.ac.uk Tue May 31 20:02:27 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Tue, 31 May 2005 21:02:27 +0100 (BST) Subject: yum in kickstart %post and HP Insight Mgr In-Reply-To: <1117568770.3743.32.camel@localhost.localdomain> References: <1117568770.3743.32.camel@localhost.localdomain> Message-ID: On Tue, 31 May 2005, Brian Long wrote: >I'm trying to use yum in the %post section of kickstart. With well- >written RPMs, this works great. However, I've come across the lovely HP >Insight Manager RPMS (hpasm, hprsm, etc) which seem to run "service >hpasm start" during the RPM postinstall (yuck). This wouldn't be too >bad if the daemons actually disassociated from the running terminal. Don't let it have them: %post yum groupinstall Insight-Manager < /dev/null > /dev/null 2>&1 >I notice yum install the group "yum groupinstall Insight-Manager" and >yum exits, but the rest of the kickstart %post doesn't run. If I look >at the hpasmd processes, they're tied to tty1 and owned by init :( If >I hit Alt-F2 in kickstart and chroot /mnt/sysimage, service hpasm stop, >the rest of the kickstart %post finishes. I also strace'd the /tmp/ks- >script process and it's stuck at waitpid(-1, ) If that doesn't work, then lay a trap: %post ( while ! killall -q hpasm; do sleep 3; done & ) yum groupinstall Insight-Manager Hope at least one of these works... :) Cheers, Phil From phr at doc.ic.ac.uk Tue May 31 20:05:29 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Tue, 31 May 2005 21:05:29 +0100 (BST) Subject: Kickstart dies after post-install In-Reply-To: <66911C8E1F2DA24EB29DE289297A135E4856CC@usrd104> References: <66911C8E1F2DA24EB29DE289297A135E4856CC@usrd104> Message-ID: On Tue, 31 May 2005, Seremeth, Stephen wrote: ><4>RPC: garbage, exit EIO ><3>Buffer I/O error on device loop0, logical block 4105 ><4>RPC: garbage, exit EIO ><3>Buffer I/O error on device loop0, logical block 4105 ><4>RPC: garbage, exit EIO ><3>Buffer I/O error on device loop0, logical block 4105 ><4>Error -3 while decompressing! ><4>f881f42c(-16817932)->ff19e000(4096) ><4>RPC: garbage, exit EIO ><3>Buffer I/O error on device loop0, logical block 4105 ><4>Error -3 while decompressing! ><4>f881b100(-16817120)->ff19f000(4096) ><4>RPC: garbage, exit EIO >------------------------------------------------------------------------ > >Any thoughts? Ugh - looks like bad disk (or RAM for loop0?) Try checking hardware with DFT and memtest86+ ? Possibly the media you're installing from is bad. Cheers, Phil From brilong at cisco.com Tue May 31 20:49:34 2005 From: brilong at cisco.com (Brian Long) Date: Tue, 31 May 2005 16:49:34 -0400 Subject: yum in kickstart %post and HP Insight Mgr In-Reply-To: References: <1117568770.3743.32.camel@localhost.localdomain> Message-ID: <1117572574.3743.46.camel@localhost.localdomain> On Tue, 2005-05-31 at 21:02 +0100, Philip Rowlands wrote: > On Tue, 31 May 2005, Brian Long wrote: > > >I'm trying to use yum in the %post section of kickstart. With well- > >written RPMs, this works great. However, I've come across the lovely HP > >Insight Manager RPMS (hpasm, hprsm, etc) which seem to run "service > >hpasm start" during the RPM postinstall (yuck). This wouldn't be too > >bad if the daemons actually disassociated from the running terminal. > > Don't let it have them: > > %post > yum groupinstall Insight-Manager < /dev/null > /dev/null 2>&1 I tried a variation of this so I could keep things logged (in order to debug possible problems in the future): chvt 3 yum groupinstall Insight-Manager &1 chvt 1 hpasmd is still tied to tty1. If I kill the "tee" process, the %post finishes running just fine. Any ideas how to keep a log of all of yum's output and display it in parallel instead of using tee? I know about yum.log, but I was hoping to grab possible errors to the postinstall log. I guess I could run "chvt 3; (tail -f /var/log/postinstall &)" and then run the "yum >> /var/log/postinstall 2>&1". Any other ideas would be appreciated. Thanks for your help! /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From klaus.steden at thomson.net Tue May 31 21:05:43 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Tue, 31 May 2005 17:05:43 -0400 Subject: yum in kickstart %post and HP Insight Mgr In-Reply-To: <1117572574.3743.46.camel@localhost.localdomain> References: <1117568770.3743.32.camel@localhost.localdomain> <1117572574.3743.46.camel@localhost.localdomain> Message-ID: <20050531210543.GE16989@thomson.net> This might be a stupid suggestion, but would using 'nohup' be of any value? > > > [some preamble cut] > > > > Don't let it have them: > > > > %post > > yum groupinstall Insight-Manager < /dev/null > /dev/null 2>&1 > > I tried a variation of this so I could keep things logged (in order to > debug possible problems in the future): > > chvt 3 > yum groupinstall Insight-Manager &1 > chvt 1 > > hpasmd is still tied to tty1. If I kill the "tee" process, the %post > finishes running just fine. Any ideas how to keep a log of all of yum's > output and display it in parallel instead of using tee? I know about > yum.log, but I was hoping to grab possible errors to the postinstall > log. > > I guess I could run "chvt 3; (tail -f /var/log/postinstall &)" and then > run the "yum >> /var/log/postinstall 2>&1". > > Any other ideas would be appreciated. Thanks for your help! > > /Brian/ > > -- > Brian Long | | | > IT Data Center Systems | .|||. .|||. > Cisco Linux Developer | ..:|||||||:...:|||||||:.. > Phone: (919) 392-7363 | C i s c o S y s t e m s > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Klaus Steden | Senior Systems Administrator | Technicolor Creative Services | TODO: Toronto | 1) Learn to use my new Unix account. klaus.steden at thomson.net | 2) Learn how to change this list. Phone: (416) 585-9995 | Fax: (416) 364-1585 | From skvidal at phy.duke.edu Tue May 31 21:59:02 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Tue, 31 May 2005 17:59:02 -0400 Subject: [Yum] yum in kickstart %post and HP Insight Mgr In-Reply-To: <20050531215550.GA14291@jadzia.bu.edu> References: <1117568770.3743.32.camel@localhost.localdomain> <20050531215550.GA14291@jadzia.bu.edu> Message-ID: <1117576742.18404.60.camel@cutter> On Tue, 2005-05-31 at 17:55 -0400, Matthew Miller wrote: > On Tue, May 31, 2005 at 03:46:10PM -0400, Brian Long wrote: > > written RPMs, this works great. However, I've come across the lovely HP > > Insight Manager RPMS (hpasm, hprsm, etc) which seem to run "service > > hpasm start" during the RPM postinstall (yuck). This wouldn't be too > [...] > > Any ideas would be appreciated. Thanks. > > I know there's the "tsflags=noscripts" option in the config file -- is there > a way to do that from the command line? Then, you could add any other bits > of the %post script to your kickstart file. > Don't encourage them. It does not make sense to add this to the cli in yum in order to work around a broken package. -sv