From hawk at tbi.univie.ac.at Thu Mar 1 13:59:33 2007 From: hawk at tbi.univie.ac.at (Richard Neuboeck) Date: Thu, 01 Mar 2007 14:59:33 +0100 Subject: large repo installation delay Message-ID: <45E6DC45.9020205@tbi.univie.ac.at> Hi! I'm using kickstart to install our workstations and cluster. Since I'm installing packages from several repositories I thought I would try a direct install from our FC6 mirror via http and add several repo entries (extras, updates, livna and our custom repo). [snip] ... install url --url=http://... repo --name=extras --baseurl=http://... ... [/snap] The installation works fine this way. However it takes about 45minutes to complete. Right after 'moving (1) to step basepkgsel' (log message) anaconda uses the workstations CPU with 100%. This state takes a little bit more than 30min. During this phase it only shows a blue screen. After that phase the typical 'Checking dependencies in packages selected for installation...' bar appears. I have not figured out yet why anaconda takes so long. Is this a common phenomenon or am I doing something wrong? To speed things up I'm using a script that takes the package listing from the ks.cfg file and creates the installation tree containing only packages needed for installation. This kickstart run takes about 15min. The gap between 'moving (1) to step basepkgsel' and the 'Checking dependencies...' bar is hardly noticeable. Thanks for feedback. Richard -- /dev/null From email at jasonkohles.com Thu Mar 1 15:55:59 2007 From: email at jasonkohles.com (Jason Kohles) Date: Thu, 1 Mar 2007 10:55:59 -0500 Subject: large repo installation delay In-Reply-To: <45E6DC45.9020205@tbi.univie.ac.at> References: <45E6DC45.9020205@tbi.univie.ac.at> Message-ID: On Mar 1, 2007, at 8:59 AM, Richard Neuboeck wrote: > Hi! > > I'm using kickstart to install our workstations and cluster. > > Since I'm installing packages from several repositories I thought I > would try a direct install from our FC6 mirror via http and add > several > repo entries (extras, updates, livna and our custom repo). > > The installation works fine this way. However it takes about 45minutes > to complete. Right after 'moving (1) to step basepkgsel' (log message) > anaconda uses the workstations CPU with 100%. This state takes a > little > bit more than 30min. During this phase it only shows a blue screen. > After that phase the typical 'Checking dependencies in packages > selected > for installation...' bar appears. I have not figured out yet why > anaconda takes so long. > > Is this a common phenomenon or am I doing something wrong? > > To speed things up I'm using a script that takes the package listing > from the ks.cfg file and creates the installation tree containing only > packages needed for installation. This kickstart run takes about > 15min. > The gap between 'moving (1) to step basepkgsel' and the 'Checking > dependencies...' bar is hardly noticeable. > If you have only one repo, yum can just download and use the repodata from that one location. I suspect that what you are waiting for is the time it takes for yum to retrieve and merge the repodata from the various repositories, in order to do to things like check dependencies against each other. -- Jason Kohles email at jasonkohles.com http://www.jasonkohles.com/ "A witty saying proves nothing." -- Voltaire From William.Ramthun at Level3.com Thu Mar 1 16:49:58 2007 From: William.Ramthun at Level3.com (Ramthun, William) Date: Thu, 1 Mar 2007 09:49:58 -0700 Subject: "pump" times out -- extend a few seconds more? In-Reply-To: Message-ID: Chip - thanks again. (I confused nicedelay and nicdelay, but have is sorted out now.) I tried each option individually: nicdelay=5/50/500/50000 linksleep=5/50/500/50000 both seem to have no effect in the time to bring up the host - at least for my boot loader. I did notice that (1) Anaconda did not hit the web server where my ks.cfg is located when ks=http//config is provided on the boot line and (2) as soon as Anaconda 'times out' and the 'choose a language' menu is presented, the interface is immediately ping-able. So while waiting for things to boot, I ended up with a couple of questions: 1) So is the objective of the 'nicdelay' and 'linksleep' options to pause Anaconda long enough so that physical network negotiation can happen to prevent Anaconda from timing out? Or is it just to delay the physical network from coming up? 2) With the boot line option 'ks='. What is supposed to look like if the ks.cfg is incorporated into the initrd for a rhel3u5 install? Any and all help is appreciated; including links to documentation. -Bill > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Shabazian, Chip > Sent: Tuesday, February 27, 2007 12:32 PM > To: Discussion list about Kickstart > Subject: RE: "pump" times out -- extend a few seconds more? > > Like I said, I've never used it, but it is SUPPOSED to exist: > > https://rhn.redhat.com/errata/RHEA-2006-0443.html > > nicdelay= > > If you want to download the srpm and take a look at loader, I think > that's where you will find the code. > > Chip > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, William > Sent: Tuesday, February 27, 2007 7:34 AM > To: Discussion list about Kickstart > Subject: RE: "pump" times out -- extend a few seconds more? > > Chip - thanks for the help. > > What I meant to say is that I am not sure how to use "nicedelay". Is > this a boot: line option? I couldn't find any reference on how to use > it. > > -Bill > > > -----Original Message----- > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > > bounces at redhat.com] On Behalf Of Shabazian, Chip > > Sent: Friday, February 23, 2007 1:38 PM > > To: Discussion list about Kickstart > > Subject: RE: "pump" times out -- extend a few seconds more? > > > > It's applicable because you said your problem was: > > > > > the network needs to "come up" before Anaconda times out. > > > > That's exactly what nicdelay is supposed to help you with. The > problem > > is, it takes ~ 30 seconds for the nic to negotiate with the switch. > By > > delaying bringing the nic up until enough time has passed that the > > negotiation has taken place, you should then be able to configure the > > nic, get your kickstart file, and be off and running. > > > > Now, I say "supposed" to, because I've never had to use it, and heard > > mixed results from those who have. > > > > Remember, the nic gets completely cycled and needs to renegotiate 3 > > times during a build: > > > > at power on > > when getting the kickstart file > > when starting the build > > > > Chip > > > > > > > > -----Original Message----- > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, > William > > Sent: Friday, February 23, 2007 12:26 PM > > To: Discussion list about Kickstart > > Subject: RE: "pump" times out -- extend a few seconds more? > > > > I did a cursory search of nicedelay and I don't see how it is > applicable > > for my situation... Maybe I just don't know how to use it. > > > > Portfast is enabled. > > > > I tried the eth0_ethtool option, but that didn't appear to help. > > > > I thought about placing the ks.cfg files in the initrd, but didn't > > really want to go there. > > > > Thanks > > > > > -----Original Message----- > > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > > > bounces at redhat.com] On Behalf Of Shabazian, Chip > > > Sent: Friday, February 23, 2007 10:56 AM > > > To: Discussion list about Kickstart > > > Subject: RE: "pump" times out -- extend a few seconds more? > > > > > > I think there was a "nicdelay" or something similar introduced > around > > > RHEL3U7, so you might want to try that. Also, make sure that > portfast > > > > > is enabled on the switches. The other thing we do is set the > > > speed/duplex via ethtool to avoid autonegotion delays: > > > > > > eth0_ethtool="autoneg=on speed=1000 duplex=full" > > > > > > When we have to build via iLO, we embed the kickstart file in the > > initrd > > > and then use ks=file: > > > > > > > > > > > > -----Original Message----- > > > From: kickstart-list-bounces at redhat.com > > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, > > William > > > Sent: Friday, February 23, 2007 9:50 AM > > > To: Discussion list about Kickstart > > > Subject: RE: "pump" times out -- extend a few seconds more? > > > > > > In our normal data centers, we build HP DL385/585 daily as well. > The > > > data center with issues is from an acquired company with older Cisco > > > switches (5 years +). > > > > > > Installing RHEL 3 Update 5 > > > > > > The install client accesses the RHEL 3 media server via eth0. When > > > Anaconda times-out, I can run thru the install setup manually and > > > RHEL3U5 will install over the network. The problem is the Kickstart > > > config is highly configured and doing a manual install is not really > > an > > > option that I want to consider. > > > > > > What I want is to access the ks.cfg via the network. Which implies > > that > > > the network needs to "come up" before Anaconda times out. > > > > > > OR > > > > > > Read the ks.cfg from the cdrom. But the cdrom is virually mounted > > over > > > the ILO port and is not accessible like ks=cdrom:/ks.cfg from the > boot > > > > > prompt. I don't know what the device is. > > > -Bill > > > > > > > -----Original Message----- > > > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > > > > bounces at redhat.com] On Behalf Of Shabazian, Chip > > > > Sent: Friday, February 23, 2007 10:34 AM > > > > To: Discussion list about Kickstart > > > > Subject: RE: "pump" times out -- extend a few seconds more? > > > > > > > > We build HP DL 380/385/580/585 servers every day while connected > to > > > > Cisco switches that we don't have any access to. Are you using > RHEL > > 3 > > > > > > > or 4? If 4, do you have spare NIC's in the box? RHEL 4 enumerates > > the > > > > > > > bus backwards from RHEL 3, so if you have a NIC, it's going to be > > eth0 > > > > > > > instead of the onboard... > > > > > > > > -----Original Message----- > > > > From: kickstart-list-bounces at redhat.com > > > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, > > > William > > > > Sent: Friday, February 23, 2007 8:01 AM > > > > To: Discussion list about Kickstart > > > > Subject: RE: "pump" times out -- extend a few seconds more? > > > > > > > > > -----Original Message----- > > > > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > > > > > bounces at redhat.com] On Behalf Of nate > > > > > Sent: Friday, January 12, 2007 10:33 AM > > > > > To: kickstart-list at redhat.com > > > > > Subject: Re: "pump" times out -- extend a few seconds more? > > > > > > > > > > Bill-Schoolcraft said: > > > > > > Hello Family, > > > > > > > > > > > > I have a series of switches, in racks, and one of them seems > to > > be > > > a > > > > bit > > > > > > slow so when kickstart goes for it's dhcp request, pump times > > out. > > > > The > > > > > > initial PXE takes much more time to get an "ack" at the > initial > > > > contact > > > > > > than the same boxes in other racks with other switches but > then > > > > finally > > > > > > starts chugging away -- it's the next dhcp request that fails, > > > times > > > > out. > > > > > > > > > > > > Today I took the same box, draped a cat6 cable to another, > less > > > > > > populated switch, and it kicked fine. > > > > > > > > > > > > (question) > > > > > > So my question is, how or where can I place an argument to > tell > > > > > > kickstart's "pump" request to try longer before timing out. > > > > > > > > > > just curious, what type of switch? many older cisco switches > have > > > STP > > > > > on by default, if you set the port(s) to 'portfast' the links > come > > > up > > > > > immediately rather than waiting ~45 seconds. > > > > > > > > > > I'm not a cisco guy but if your switches are cisco the command > is > > > > > something like > > > > > > > > > > int (interface) > > > > > spanning-tree portfast > > > > > ^Z > > > > > wr mem > > > > > > > > > > > > > > > e.g. on my old 3500s (IOS 12.0 is the latest they'll run) int > > fa0/5 > > > > > spanning-tree portfast > > > > > > > > > > I haven't encountered any other switch vendors that have this > > > > > 'feature' enabled by default. You can tell if it's on without > even > > > > > > > logging in(to the switch) by checking to see if the link light > for > > > the > > > > > > > > > port spends a long time in the orange color before turning > green. > > > I've > > > > > > > > > been told more modern cisco switches ship with this feature off > by > > > > > > > default. > > > > > > > > > > as for extending the pump times, I'm not sure, so I can't answer > > > your > > > > > specific question :) maybe the above will help a bit.. > > > > > > > > > > I did a quick search on pump timeout, and what I did see was > that > > it > > > > > > > > may be possible to extend the timeout by adding a > > > > > /etc/pump_device.conf and using the option 'timeout > number>'. > > > > > > > > > > though you'll need to add that to the initrd image..not the > > easiest > > > > > thing in the world. > > > > > > > > > > nate > > > > > > > > > > > > > > > > > I am experiencing a similar problem; however, my situation is > > slightly > > > > > > > different. > > > > > > > > I have HP DL385's connected to a 48 port 10/100 module in a Cisco > > 6509 > > > > > > > (with Sup 2 routing module). I'm using the virtual CDROM via the > HP > > > ILO > > > > (Integrated Lights Out) interface to perform the initial boot. I > > > mount > > > > a customized version of RHEL3U5 CD1. My kickstart file is > specified > > > as > > > > an url. The system boots and Anaconda starts, but before the link > > > > negotiation can complete Anaconda times-out and I am dumped into > an > > > > interactive install. > > > > > > > > A couple of data points: > > > > > > > > 1) The HP DL385 on-board NICs are manufactured by Broadcom. I've > > read > > > > > > > about a known bug with link negotiation between the Broadcom NICs > > and > > > > Cisco 48-port line cards. The link negotiation works, it just > takes > > a > > > > > > > "longer" time than with any other kind of NIC. > > > > > > > > 2) Setting the switch port NIC to enable "spanning-tree portfast" > > does > > > > > > > shorten the link negotiation time, but not enough to get around my > > > > problem. > > > > > > > > Work Arounds: > > > > > > > > 1) I will prebuild the kickstart configs and place them on my > > > customized > > > > RHEL3U5 CD. I've been unsuccessful at reading the kickstart > config > > > from > > > > the CDROM. The driver supporting the ILO virtual cdrom does not > > name > > > > the cdrom device as "cdrom". Still reading up on this. > > > > > > > > That's it so far, > > > > -Bill > > > > > > > > _______________________________________________ > > > > 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 > > > > _______________________________________________ > > 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 dlutter at redhat.com Thu Mar 1 18:42:55 2007 From: dlutter at redhat.com (David Lutterkort) Date: Thu, 01 Mar 2007 10:42:55 -0800 Subject: large repo installation delay In-Reply-To: <45E6DC45.9020205@tbi.univie.ac.at> References: <45E6DC45.9020205@tbi.univie.ac.at> Message-ID: <1172774575.27848.73.camel@localhost.localdomain> On Thu, 2007-03-01 at 14:59 +0100, Richard Neuboeck wrote: > To speed things up I'm using a script that takes the package listing > from the ks.cfg file and creates the installation tree containing only > packages needed for installation. This kickstart run takes about 15min. > The gap between 'moving (1) to step basepkgsel' and the 'Checking > dependencies...' bar is hardly noticeable. Not that I have a fix for this, but what are the differences in total number of packages between the case where you have all repos and where you use the custom-tailored repo ? David From Chip.Shabazian at bankofamerica.com Thu Mar 1 19:20:36 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Thu, 01 Mar 2007 11:20:36 -0800 Subject: "pump" times out -- extend a few seconds more? In-Reply-To: Message-ID: 1) The way *I* understand it is that it is supposed to delay the driver load while the physical negotiation takes place. But, I've never used or tested it. 2) ks=file:filename Here is a small, easy script I use to insert kickstart files into the initrd. I just put this in the isolinux folder, and all of my ks.cfg files in a subfolder called configs: #!/bin/sh mv initrd.img initrd.img.gz gunzip initrd.img.gz mount -o loop initrd.img /mnt/init/ cp --reply=yes configs/* /mnt/init/ umount /mnt/init/ gzip initrd.img mv initrd.img.gz initrd.img Chip -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, William Sent: Thursday, March 01, 2007 8:50 AM To: Discussion list about Kickstart Subject: RE: "pump" times out -- extend a few seconds more? Chip - thanks again. (I confused nicedelay and nicdelay, but have is sorted out now.) I tried each option individually: nicdelay=5/50/500/50000 linksleep=5/50/500/50000 both seem to have no effect in the time to bring up the host - at least for my boot loader. I did notice that (1) Anaconda did not hit the web server where my ks.cfg is located when ks=http//config is provided on the boot line and (2) as soon as Anaconda 'times out' and the 'choose a language' menu is presented, the interface is immediately ping-able. So while waiting for things to boot, I ended up with a couple of questions: 1) So is the objective of the 'nicdelay' and 'linksleep' options to pause Anaconda long enough so that physical network negotiation can happen to prevent Anaconda from timing out? Or is it just to delay the physical network from coming up? 2) With the boot line option 'ks='. What is supposed to look like if the ks.cfg is incorporated into the initrd for a rhel3u5 install? Any and all help is appreciated; including links to documentation. -Bill > -----Original Message----- > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > bounces at redhat.com] On Behalf Of Shabazian, Chip > Sent: Tuesday, February 27, 2007 12:32 PM > To: Discussion list about Kickstart > Subject: RE: "pump" times out -- extend a few seconds more? > > Like I said, I've never used it, but it is SUPPOSED to exist: > > https://rhn.redhat.com/errata/RHEA-2006-0443.html > > nicdelay= > > If you want to download the srpm and take a look at loader, I think > that's where you will find the code. > > Chip > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, William > Sent: Tuesday, February 27, 2007 7:34 AM > To: Discussion list about Kickstart > Subject: RE: "pump" times out -- extend a few seconds more? > > Chip - thanks for the help. > > What I meant to say is that I am not sure how to use "nicedelay". Is > this a boot: line option? I couldn't find any reference on how to use > it. > > -Bill > > > -----Original Message----- > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > > bounces at redhat.com] On Behalf Of Shabazian, Chip > > Sent: Friday, February 23, 2007 1:38 PM > > To: Discussion list about Kickstart > > Subject: RE: "pump" times out -- extend a few seconds more? > > > > It's applicable because you said your problem was: > > > > > the network needs to "come up" before Anaconda times out. > > > > That's exactly what nicdelay is supposed to help you with. The > problem > > is, it takes ~ 30 seconds for the nic to negotiate with the switch. > By > > delaying bringing the nic up until enough time has passed that the > > negotiation has taken place, you should then be able to configure the > > nic, get your kickstart file, and be off and running. > > > > Now, I say "supposed" to, because I've never had to use it, and heard > > mixed results from those who have. > > > > Remember, the nic gets completely cycled and needs to renegotiate 3 > > times during a build: > > > > at power on > > when getting the kickstart file > > when starting the build > > > > Chip > > > > > > > > -----Original Message----- > > From: kickstart-list-bounces at redhat.com > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, > William > > Sent: Friday, February 23, 2007 12:26 PM > > To: Discussion list about Kickstart > > Subject: RE: "pump" times out -- extend a few seconds more? > > > > I did a cursory search of nicedelay and I don't see how it is > applicable > > for my situation... Maybe I just don't know how to use it. > > > > Portfast is enabled. > > > > I tried the eth0_ethtool option, but that didn't appear to help. > > > > I thought about placing the ks.cfg files in the initrd, but didn't > > really want to go there. > > > > Thanks > > > > > -----Original Message----- > > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > > > bounces at redhat.com] On Behalf Of Shabazian, Chip > > > Sent: Friday, February 23, 2007 10:56 AM > > > To: Discussion list about Kickstart > > > Subject: RE: "pump" times out -- extend a few seconds more? > > > > > > I think there was a "nicdelay" or something similar introduced > around > > > RHEL3U7, so you might want to try that. Also, make sure that > portfast > > > > > is enabled on the switches. The other thing we do is set the > > > speed/duplex via ethtool to avoid autonegotion delays: > > > > > > eth0_ethtool="autoneg=on speed=1000 duplex=full" > > > > > > When we have to build via iLO, we embed the kickstart file in the > > initrd > > > and then use ks=file: > > > > > > > > > > > > -----Original Message----- > > > From: kickstart-list-bounces at redhat.com > > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, > > William > > > Sent: Friday, February 23, 2007 9:50 AM > > > To: Discussion list about Kickstart > > > Subject: RE: "pump" times out -- extend a few seconds more? > > > > > > In our normal data centers, we build HP DL385/585 daily as well. > The > > > data center with issues is from an acquired company with older Cisco > > > switches (5 years +). > > > > > > Installing RHEL 3 Update 5 > > > > > > The install client accesses the RHEL 3 media server via eth0. When > > > Anaconda times-out, I can run thru the install setup manually and > > > RHEL3U5 will install over the network. The problem is the Kickstart > > > config is highly configured and doing a manual install is not really > > an > > > option that I want to consider. > > > > > > What I want is to access the ks.cfg via the network. Which implies > > that > > > the network needs to "come up" before Anaconda times out. > > > > > > OR > > > > > > Read the ks.cfg from the cdrom. But the cdrom is virually mounted > > over > > > the ILO port and is not accessible like ks=cdrom:/ks.cfg from the > boot > > > > > prompt. I don't know what the device is. > > > -Bill > > > > > > > -----Original Message----- > > > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > > > > bounces at redhat.com] On Behalf Of Shabazian, Chip > > > > Sent: Friday, February 23, 2007 10:34 AM > > > > To: Discussion list about Kickstart > > > > Subject: RE: "pump" times out -- extend a few seconds more? > > > > > > > > We build HP DL 380/385/580/585 servers every day while connected > to > > > > Cisco switches that we don't have any access to. Are you using > RHEL > > 3 > > > > > > > or 4? If 4, do you have spare NIC's in the box? RHEL 4 enumerates > > the > > > > > > > bus backwards from RHEL 3, so if you have a NIC, it's going to be > > eth0 > > > > > > > instead of the onboard... > > > > > > > > -----Original Message----- > > > > From: kickstart-list-bounces at redhat.com > > > > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, > > > William > > > > Sent: Friday, February 23, 2007 8:01 AM > > > > To: Discussion list about Kickstart > > > > Subject: RE: "pump" times out -- extend a few seconds more? > > > > > > > > > -----Original Message----- > > > > > From: kickstart-list-bounces at redhat.com [mailto:kickstart-list- > > > > > bounces at redhat.com] On Behalf Of nate > > > > > Sent: Friday, January 12, 2007 10:33 AM > > > > > To: kickstart-list at redhat.com > > > > > Subject: Re: "pump" times out -- extend a few seconds more? > > > > > > > > > > Bill-Schoolcraft said: > > > > > > Hello Family, > > > > > > > > > > > > I have a series of switches, in racks, and one of them seems > to > > be > > > a > > > > bit > > > > > > slow so when kickstart goes for it's dhcp request, pump times > > out. > > > > The > > > > > > initial PXE takes much more time to get an "ack" at the > initial > > > > contact > > > > > > than the same boxes in other racks with other switches but > then > > > > finally > > > > > > starts chugging away -- it's the next dhcp request that fails, > > > times > > > > out. > > > > > > > > > > > > Today I took the same box, draped a cat6 cable to another, > less > > > > > > populated switch, and it kicked fine. > > > > > > > > > > > > (question) > > > > > > So my question is, how or where can I place an argument to > tell > > > > > > kickstart's "pump" request to try longer before timing out. > > > > > > > > > > just curious, what type of switch? many older cisco switches > have > > > STP > > > > > on by default, if you set the port(s) to 'portfast' the links > come > > > up > > > > > immediately rather than waiting ~45 seconds. > > > > > > > > > > I'm not a cisco guy but if your switches are cisco the command > is > > > > > something like > > > > > > > > > > int (interface) > > > > > spanning-tree portfast > > > > > ^Z > > > > > wr mem > > > > > > > > > > > > > > > e.g. on my old 3500s (IOS 12.0 is the latest they'll run) int > > fa0/5 > > > > > spanning-tree portfast > > > > > > > > > > I haven't encountered any other switch vendors that have this > > > > > 'feature' enabled by default. You can tell if it's on without > even > > > > > > > logging in(to the switch) by checking to see if the link light > for > > > the > > > > > > > > > port spends a long time in the orange color before turning > green. > > > I've > > > > > > > > > been told more modern cisco switches ship with this feature off > by > > > > > > > default. > > > > > > > > > > as for extending the pump times, I'm not sure, so I can't answer > > > your > > > > > specific question :) maybe the above will help a bit.. > > > > > > > > > > I did a quick search on pump timeout, and what I did see was > that > > it > > > > > > > > may be possible to extend the timeout by adding a > > > > > /etc/pump_device.conf and using the option 'timeout > number>'. > > > > > > > > > > though you'll need to add that to the initrd image..not the > > easiest > > > > > thing in the world. > > > > > > > > > > nate > > > > > > > > > > > > > > > > > I am experiencing a similar problem; however, my situation is > > slightly > > > > > > > different. > > > > > > > > I have HP DL385's connected to a 48 port 10/100 module in a Cisco > > 6509 > > > > > > > (with Sup 2 routing module). I'm using the virtual CDROM via the > HP > > > ILO > > > > (Integrated Lights Out) interface to perform the initial boot. I > > > mount > > > > a customized version of RHEL3U5 CD1. My kickstart file is > specified > > > as > > > > an url. The system boots and Anaconda starts, but before the link > > > > negotiation can complete Anaconda times-out and I am dumped into > an > > > > interactive install. > > > > > > > > A couple of data points: > > > > > > > > 1) The HP DL385 on-board NICs are manufactured by Broadcom. I've > > read > > > > > > > about a known bug with link negotiation between the Broadcom NICs > > and > > > > Cisco 48-port line cards. The link negotiation works, it just > takes > > a > > > > > > > "longer" time than with any other kind of NIC. > > > > > > > > 2) Setting the switch port NIC to enable "spanning-tree portfast" > > does > > > > > > > shorten the link negotiation time, but not enough to get around my > > > > problem. > > > > > > > > Work Arounds: > > > > > > > > 1) I will prebuild the kickstart configs and place them on my > > > customized > > > > RHEL3U5 CD. I've been unsuccessful at reading the kickstart > config > > > from > > > > the CDROM. The driver supporting the ILO virtual cdrom does not > > name > > > > the cdrom device as "cdrom". Still reading up on this. > > > > > > > > That's it so far, > > > > -Bill > > > > > > > > _______________________________________________ > > > > 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 > > > > _______________________________________________ > > 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 hawk at tbi.univie.ac.at Fri Mar 2 09:33:33 2007 From: hawk at tbi.univie.ac.at (Richard Neuboeck) Date: Fri, 02 Mar 2007 10:33:33 +0100 Subject: large repo installation delay In-Reply-To: <1172774575.27848.73.camel@localhost.localdomain> References: <45E6DC45.9020205@tbi.univie.ac.at> <1172774575.27848.73.camel@localhost.localdomain> Message-ID: <45E7EF6D.6040008@tbi.univie.ac.at> David Lutterkort wrote: > Not that I have a fix for this, but what are the differences in total > number of packages between the case where you have all repos and where > you use the custom-tailored repo ? Our custom kickstart tree consists of about 700 rpm packages (those 700 packages listed in the ks.cfg file) wheres the total of all repos is about 10000 packages (core, extras, updates, livna and our own repo). However if one is installing directly from fedora mirrors this amount of packages should be pretty common since livna and our own repo only add about 500 packages. So as a result a 'normal' kickstart installation should always take a lot of time to check packages for installation. I've traced the kickstart steps to the 'install' method in yum/__init__.py. As described in the comment of this method it 'tries to mark for install the item specified'. Specified items are of course all the packages listed in the ks.cfg. Greetings Richard -- /dev/null From Vivek.Kalia at Euroclear.com Fri Mar 2 17:48:54 2007 From: Vivek.Kalia at Euroclear.com (Vivek.Kalia at Euroclear.com) Date: Fri, 2 Mar 2007 17:48:54 +0000 Subject: "pump" times out -- extend a few seconds more? In-Reply-To: <20070302170012.47A05732D1@hormel.redhat.com> Message-ID: Regarding experience of parameters nicdelay=50 linksleep=50 I used them today to circumvent the Anaconda "time-out" and can confirm that they worked in the required manner. My previous http builds were failing to contact the apache server due to the time-out. With these values in place, I could see Anaconda waiting about 50 seconds while the NIC initialised; the build then completed normally. I am still investigating the root cause of the time-out, but this seems a good circumvention. Vivek Kalia This e-mail message, including any attachments transmitted with it, is CONFIDENTIAL and may contain legally privileged information.This message is intended solely for the use of the individual or entity to whom it is addressed. If you are not the intended recipient you should not read, copy, distribute, disclose or otherwise use this information. If you have received this message in error, please notify us immediately and delete it from your system. The integrity and security of this message cannot be guaranteed and it may be subject to data corruption and unauthorised amendment, for which we accept no liability. Euroclear reserves the right to retain email messages on its systems and to the extent and under circumstances permitted by applicable law, to monitor and intercept email messages to and from its systems. Euroclear is the marketing name for the Euroclear System, Euroclear plc, Euroclear SA/NV and their affiliates. http://www.euroclear.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From William.Ramthun at Level3.com Mon Mar 5 16:50:46 2007 From: William.Ramthun at Level3.com (Ramthun, William) Date: Mon, 5 Mar 2007 09:50:46 -0700 Subject: "pump" times out -- extend a few seconds more? In-Reply-To: Message-ID: Summarizing my situation: Client HW: HP DL385 Client OS: RHEL3U5 The boot cd is mounted via the virtual cdrom thru HP's ILO port. My isolinux.cfg contains the entry: label ap10903-02-net kernel vmlinuz append ks=http://10.1.181.252/cgi-bin/avamar/dl385_data_node initrd=initrd.img text And on the boot: line I use: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 I've used both nicdelay and linksleep with values of 50, 500, 5000, and 50000 with no discernable boot delays. The result is: I see vmliniz and initrd loading. I am presented the "welcome to red hat linux" blue screen (I'm in text mode) Then I see the usb and raid controller drivers load Then I see a "determine hostname and domain" ( I think this is where the timeout occurs ) Then I am prompted to "choose a language" In a different terminal window I am "pinging" the 10.196.154.122 IP during the whole boot process. Once the "choose a language" prompt appears, then the interface begins responding to my pings. My next step is to include the Kickstart config in the initrd. If anyone sees any issues with my method, please drop me a note. Thanks -Bill ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Vivek.Kalia at Euroclear.com Sent: Friday, March 02, 2007 10:49 AM To: kickstart-list at redhat.com Subject: RE: "pump" times out -- extend a few seconds more? Regarding experience of parameters nicdelay=50 linksleep=50 I used them today to circumvent the Anaconda "time-out" and can confirm that they worked in the required manner. My previous http builds were failing to contact the apache server due to the time-out. With these values in place, I could see Anaconda waiting about 50 seconds while the NIC initialised; the build then completed normally. I am still investigating the root cause of the time-out, but this seems a good circumvention. Vivek Kalia This e-mail message, including any attachments transmitted with it, is CONFIDENTIAL and may contain legally privileged information.This message is intended solely for the use of the individual or entity to whom it is addressed. If you are not the intended recipient you should not read, copy, distribute, disclose or otherwise use this information. If you have received this message in error, please notify us immediately and delete it from your system. The integrity and security of this message cannot be guaranteed and it may be subject to data corruption and unauthorised amendment, for which we accept no liability. Euroclear reserves the right to retain email messages on its systems and to the extent and under circumstances permitted by applicable law, to monitor and intercept email messages to and from its systems. Euroclear is the marketing name for the Euroclear System, Euroclear plc, Euroclear SA/NV and their affiliates. http://www.euroclear.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From Luca.Mariani at anritsu.com Mon Mar 5 18:03:37 2007 From: Luca.Mariani at anritsu.com (Mariani, Luca) Date: Mon, 5 Mar 2007 19:03:37 +0100 Subject: kernel parameters Message-ID: <2066A5EDF16FF848AAE6F1DEDAD977C6280176@rom-exc-002.main.intgin.net> Hi all, I'm upgrading a bootable CD which uses kickstart to automatically install REHL4 Upgrade3 and a subset of programs. The upgrade is needed by a new hardware that requests a special kernel parameter to boot the system (in particular the kernel parameter is 'pci=nommconf' and it should be needed to use the Smart Array Controller) I've been able to add this parameter in kickstart config file and now the boot is working fine (I'm using GRUB as bootloader), but I must type the kernel parameter by hand at installation prompt... Is there a way to configure kickstart (or something else) in order to autamatically pass this kernel parameter to installation process too? Thanks Regards /Luca -------------- next part -------------- An HTML attachment was scrubbed... URL: From CallahanT at tessco.com Mon Mar 5 18:33:44 2007 From: CallahanT at tessco.com (Callahan, Tom) Date: Mon, 5 Mar 2007 13:33:44 -0500 Subject: kernel parameters Message-ID: <80F8765339230441923652E935615D3B0756BBC5@cliff.tessco.com> You have to modify the syslinux files on the boot CD and rebuild the CD. Change the append= lines and add the option you need. Thanks, Tom Callahan TESSCO Technologies Desk: (410)-229-1361 Cell: (410)-588-7605 Email: callahant at tessco.com _____ size=2 width="100%" align=center tabindex=-1> From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Monday, March 05, 2007 1:04 PM To: kickstart-list at redhat.com Subject: kernel parameters Hi all, I'm upgrading a bootable CD which uses kickstart to automatically install REHL4 Upgrade3 and a subset of programs. The upgrade is needed by a new hardware that requests a special kernel parameter to boot the system (in particular the kernel parameter is 'pci=nommconf' and it should be needed to use the Smart Array Controller) I've been able to add this parameter in kickstart config file and now the boot is working fine (I'm using GRUB as bootloader), but I must type the kernel parameter by hand at installation prompt... Is there a way to configure kickstart (or something else) in order to autamatically pass this kernel parameter to installation process too? Thanks Regards /Luca -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Mon Mar 5 19:59:29 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Mon, 05 Mar 2007 11:59:29 -0800 Subject: "pump" times out -- extend a few seconds more? In-Reply-To: Message-ID: Try the following: If you are connected to a gigabit switch: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 nicdelay=50 linksleep=50 ETHTOOL_OPTS="autoneg on speed 1000 duplex full" If you are connected to a 100Mb switch: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 nicdelay=50 linksleep=50 ETHTOOL_OPTS="autoneg off speed 100 duplex full" ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, William Sent: Monday, March 05, 2007 8:51 AM To: Discussion list about Kickstart Subject: RE: "pump" times out -- extend a few seconds more? Summarizing my situation: Client HW: HP DL385 Client OS: RHEL3U5 The boot cd is mounted via the virtual cdrom thru HP's ILO port. My isolinux.cfg contains the entry: label ap10903-02-net kernel vmlinuz append ks=http://10.1.181.252/cgi-bin/avamar/dl385_data_node initrd=initrd.img text And on the boot: line I use: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 I've used both nicdelay and linksleep with values of 50, 500, 5000, and 50000 with no discernable boot delays. The result is: I see vmliniz and initrd loading. I am presented the "welcome to red hat linux" blue screen (I'm in text mode) Then I see the usb and raid controller drivers load Then I see a "determine hostname and domain" ( I think this is where the timeout occurs ) Then I am prompted to "choose a language" In a different terminal window I am "pinging" the 10.196.154.122 IP during the whole boot process. Once the "choose a language" prompt appears, then the interface begins responding to my pings. My next step is to include the Kickstart config in the initrd. If anyone sees any issues with my method, please drop me a note. Thanks -Bill ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Vivek.Kalia at Euroclear.com Sent: Friday, March 02, 2007 10:49 AM To: kickstart-list at redhat.com Subject: RE: "pump" times out -- extend a few seconds more? Regarding experience of parameters nicdelay=50 linksleep=50 I used them today to circumvent the Anaconda "time-out" and can confirm that they worked in the required manner. My previous http builds were failing to contact the apache server due to the time-out. With these values in place, I could see Anaconda waiting about 50 seconds while the NIC initialised; the build then completed normally. I am still investigating the root cause of the time-out, but this seems a good circumvention. Vivek Kalia This e-mail message, including any attachments transmitted with it, is CONFIDENTIAL and may contain legally privileged information.This message is intended solely for the use of the individual or entity to whom it is addressed. If you are not the intended recipient you should not read, copy, distribute, disclose or otherwise use this information. If you have received this message in error, please notify us immediately and delete it from your system. The integrity and security of this message cannot be guaranteed and it may be subject to data corruption and unauthorised amendment, for which we accept no liability. Euroclear reserves the right to retain email messages on its systems and to the extent and under circumstances permitted by applicable law, to monitor and intercept email messages to and from its systems. Euroclear is the marketing name for the Euroclear System, Euroclear plc, Euroclear SA/NV and their affiliates. http://www.euroclear.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From Luca.Mariani at anritsu.com Tue Mar 6 10:09:00 2007 From: Luca.Mariani at anritsu.com (Mariani, Luca) Date: Tue, 6 Mar 2007 11:09:00 +0100 Subject: kernel parameters In-Reply-To: <80F8765339230441923652E935615D3B0756BBC5@cliff.tessco.com> Message-ID: <2066A5EDF16FF848AAE6F1DEDAD977C62801BF@rom-exc-002.main.intgin.net> Hi Tom, thanks a lot for your quickly response... I'll try this ASAP Bye Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Callahan, Tom Sent: 05 March 2007 19:34 To: Discussion list about Kickstart Subject: RE: kernel parameters You have to modify the syslinux files on the boot CD and rebuild the CD. Change the append= lines and add the option you need. Thanks, Tom Callahan TESSCO Technologies Desk: (410)-229-1361 Cell: (410)-588-7605 Email: callahant at tessco.com size=2 width="100%" align=center tabindex=-1> From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Monday, March 05, 2007 1:04 PM To: kickstart-list at redhat.com Subject: kernel parameters Hi all, I'm upgrading a bootable CD which uses kickstart to automatically install REHL4 Upgrade3 and a subset of programs. The upgrade is needed by a new hardware that requests a special kernel parameter to boot the system (in particular the kernel parameter is 'pci=nommconf' and it should be needed to use the Smart Array Controller) I've been able to add this parameter in kickstart config file and now the boot is working fine (I'm using GRUB as bootloader), but I must type the kernel parameter by hand at installation prompt... Is there a way to configure kickstart (or something else) in order to autamatically pass this kernel parameter to installation process too? Thanks Regards /Luca -------------- next part -------------- An HTML attachment was scrubbed... URL: From Luca.Mariani at anritsu.com Tue Mar 6 15:51:49 2007 From: Luca.Mariani at anritsu.com (Mariani, Luca) Date: Tue, 6 Mar 2007 16:51:49 +0100 Subject: kernel parameters In-Reply-To: <2066A5EDF16FF848AAE6F1DEDAD977C62801BF@rom-exc-002.main.intgin.net> Message-ID: <2066A5EDF16FF848AAE6F1DEDAD977C6280255@rom-exc-002.main.intgin.net> Hi Tom, now it can start but soon anaconda breaks reporting the following Exception: "An error occurrder unmounting the CD. Please make sure you're not accessing /mnt/source from the shell on tty2 and then clik OK to retry" Shell on tty2 is on / from /tmp/anaconda.log file I see this: ... ... * moving (1) to step readcomps * exception in unmountCD: (16,'Device or resouce busy') I did the following steps: - modified isolinux.cfg file to add the kernel parameter I need - modified some .msg files to customize this new CD - modified the kickstart files to add the kernel parameter to boot loader (as I did yesterday) - recreate the ISO Any ideas? I really don't known how to investigate on it... thanks /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: 06 March 2007 11:09 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, thanks a lot for your quickly response... I'll try this ASAP Bye Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Callahan, Tom Sent: 05 March 2007 19:34 To: Discussion list about Kickstart Subject: RE: kernel parameters You have to modify the syslinux files on the boot CD and rebuild the CD. Change the append= lines and add the option you need. Thanks, Tom Callahan TESSCO Technologies Desk: (410)-229-1361 Cell: (410)-588-7605 Email: callahant at tessco.com size=2 width="100%" align=center tabindex=-1> From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Monday, March 05, 2007 1:04 PM To: kickstart-list at redhat.com Subject: kernel parameters Hi all, I'm upgrading a bootable CD which uses kickstart to automatically install REHL4 Upgrade3 and a subset of programs. The upgrade is needed by a new hardware that requests a special kernel parameter to boot the system (in particular the kernel parameter is 'pci=nommconf' and it should be needed to use the Smart Array Controller) I've been able to add this parameter in kickstart config file and now the boot is working fine (I'm using GRUB as bootloader), but I must type the kernel parameter by hand at installation prompt... Is there a way to configure kickstart (or something else) in order to autamatically pass this kernel parameter to installation process too? Thanks Regards /Luca -------------- next part -------------- An HTML attachment was scrubbed... URL: From Colin.Coe at woodside.com.au Tue Mar 6 21:45:43 2007 From: Colin.Coe at woodside.com.au (Coe, Colin C. (Unix Engineer)) Date: Wed, 7 Mar 2007 06:45:43 +0900 Subject: kernel parameters In-Reply-To: <2066A5EDF16FF848AAE6F1DEDAD977C6280255@rom-exc-002.main.intgin.net> Message-ID: Hi Luca Is there a reason you can't call grubby from the %post section to add the parameters you need? CC -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Wednesday, 7 March 2007 12:52 AM To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, now it can start but soon anaconda breaks reporting the following Exception: "An error occurrder unmounting the CD. Please make sure you're not accessing /mnt/source from the shell on tty2 and then clik OK to retry" Shell on tty2 is on / from /tmp/anaconda.log file I see this: ... ... * moving (1) to step readcomps * exception in unmountCD: (16,'Device or resouce busy') I did the following steps: - modified isolinux.cfg file to add the kernel parameter I need - modified some .msg files to customize this new CD - modified the kickstart files to add the kernel parameter to boot loader (as I did yesterday) - recreate the ISO Any ideas? I really don't known how to investigate on it... thanks /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: 06 March 2007 11:09 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, thanks a lot for your quickly response... I'll try this ASAP Bye Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Callahan, Tom Sent: 05 March 2007 19:34 To: Discussion list about Kickstart Subject: RE: kernel parameters You have to modify the syslinux files on the boot CD and rebuild the CD. Change the append= lines and add the option you need. Thanks, Tom Callahan TESSCO Technologies Desk: (410)-229-1361 Cell: (410)-588-7605 Email: callahant at tessco.com size=2 width="100%" align=center tabindex=-1> From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Monday, March 05, 2007 1:04 PM To: kickstart-list at redhat.com Subject: kernel parameters Hi all, I'm upgrading a bootable CD which uses kickstart to automatically install REHL4 Upgrade3 and a subset of programs. The upgrade is needed by a new hardware that requests a special kernel parameter to boot the system (in particular the kernel parameter is 'pci=nommconf' and it should be needed to use the Smart Array Controller) I've been able to add this parameter in kickstart config file and now the boot is working fine (I'm using GRUB as bootloader), but I must type the kernel parameter by hand at installation prompt... Is there a way to configure kickstart (or something else) in order to autamatically pass this kernel parameter to installation process too? Thanks Regards /Luca NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- An HTML attachment was scrubbed... URL: From William.Ramthun at Level3.com Wed Mar 7 00:01:29 2007 From: William.Ramthun at Level3.com (Ramthun, William) Date: Tue, 6 Mar 2007 17:01:29 -0700 Subject: "pump" times out -- extend a few seconds more? In-Reply-To: Message-ID: Chip, Thanks for the help! I did try setting ethtool_opts="autoneg off speed 100 duplex full" along with the nicdelay=50 earlier and received no love for my efforts. I did build the 20 configs and wrapped them in the initrd.img - and that worked. I did have to edit all 20 configs for a cut/paste line-wrap error. But now the 20 systems are built. I really appreciate the help in working to a solution. Since this problem is solved for now and for all my other installation needs the ks=http:// works, I am not sure that I will ever learn what my network problem was. (So typically when an issue is solved, does the person with the issue and resolution write a summary for the list?) Regards, -Bill ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip Sent: Monday, March 05, 2007 12:59 PM To: Discussion list about Kickstart Subject: RE: "pump" times out -- extend a few seconds more? Try the following: If you are connected to a gigabit switch: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 nicdelay=50 linksleep=50 ETHTOOL_OPTS="autoneg on speed 1000 duplex full" If you are connected to a 100Mb switch: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 nicdelay=50 linksleep=50 ETHTOOL_OPTS="autoneg off speed 100 duplex full" ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, William Sent: Monday, March 05, 2007 8:51 AM To: Discussion list about Kickstart Subject: RE: "pump" times out -- extend a few seconds more? Summarizing my situation: Client HW: HP DL385 Client OS: RHEL3U5 The boot cd is mounted via the virtual cdrom thru HP's ILO port. My isolinux.cfg contains the entry: label ap10903-02-net kernel vmlinuz append ks=http://10.1.181.252/cgi-bin/avamar/dl385_data_node initrd=initrd.img text And on the boot: line I use: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 I've used both nicdelay and linksleep with values of 50, 500, 5000, and 50000 with no discernable boot delays. The result is: I see vmliniz and initrd loading. I am presented the "welcome to red hat linux" blue screen (I'm in text mode) Then I see the usb and raid controller drivers load Then I see a "determine hostname and domain" ( I think this is where the timeout occurs ) Then I am prompted to "choose a language" In a different terminal window I am "pinging" the 10.196.154.122 IP during the whole boot process. Once the "choose a language" prompt appears, then the interface begins responding to my pings. My next step is to include the Kickstart config in the initrd. If anyone sees any issues with my method, please drop me a note. Thanks -Bill ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Vivek.Kalia at Euroclear.com Sent: Friday, March 02, 2007 10:49 AM To: kickstart-list at redhat.com Subject: RE: "pump" times out -- extend a few seconds more? Regarding experience of parameters nicdelay=50 linksleep=50 I used them today to circumvent the Anaconda "time-out" and can confirm that they worked in the required manner. My previous http builds were failing to contact the apache server due to the time-out. With these values in place, I could see Anaconda waiting about 50 seconds while the NIC initialised; the build then completed normally. I am still investigating the root cause of the time-out, but this seems a good circumvention. Vivek Kalia This e-mail message, including any attachments transmitted with it, is CONFIDENTIAL and may contain legally privileged information.This message is intended solely for the use of the individual or entity to whom it is addressed. If you are not the intended recipient you should not read, copy, distribute, disclose or otherwise use this information. If you have received this message in error, please notify us immediately and delete it from your system. The integrity and security of this message cannot be guaranteed and it may be subject to data corruption and unauthorised amendment, for which we accept no liability. Euroclear reserves the right to retain email messages on its systems and to the extent and under circumstances permitted by applicable law, to monitor and intercept email messages to and from its systems. Euroclear is the marketing name for the Euroclear System, Euroclear plc, Euroclear SA/NV and their affiliates. http://www.euroclear.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Wed Mar 7 02:03:40 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Tue, 06 Mar 2007 18:03:40 -0800 Subject: "pump" times out -- extend a few seconds more? In-Reply-To: Message-ID: > (So typically when an issue is solved, does the person with the issue and resolution write a summary for the list?) Usually all I see is either a "it worked" or those with problems stop asking questions :( If you ever want to make a "generic" boot cd, you can assign the IP address on the boot: line, then have just what you need to get the machine up in the embedded ks.cfg, and use wget, or nfs (not sure about ftp/tftp) to pull in the system specific config, then use %include to add it into the ks.cfg. This is the way we are moving to. Glad you got things working, Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, William Sent: Tuesday, March 06, 2007 4:01 PM To: Discussion list about Kickstart Subject: RE: "pump" times out -- extend a few seconds more? Chip, Thanks for the help! I did try setting ethtool_opts="autoneg off speed 100 duplex full" along with the nicdelay=50 earlier and received no love for my efforts. I did build the 20 configs and wrapped them in the initrd.img - and that worked. I did have to edit all 20 configs for a cut/paste line-wrap error. But now the 20 systems are built. I really appreciate the help in working to a solution. Since this problem is solved for now and for all my other installation needs the ks=http:// works, I am not sure that I will ever learn what my network problem was. (So typically when an issue is solved, does the person with the issue and resolution write a summary for the list?) Regards, -Bill ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip Sent: Monday, March 05, 2007 12:59 PM To: Discussion list about Kickstart Subject: RE: "pump" times out -- extend a few seconds more? Try the following: If you are connected to a gigabit switch: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 nicdelay=50 linksleep=50 ETHTOOL_OPTS="autoneg on speed 1000 duplex full" If you are connected to a 100Mb switch: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 nicdelay=50 linksleep=50 ETHTOOL_OPTS="autoneg off speed 100 duplex full" ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, William Sent: Monday, March 05, 2007 8:51 AM To: Discussion list about Kickstart Subject: RE: "pump" times out -- extend a few seconds more? Summarizing my situation: Client HW: HP DL385 Client OS: RHEL3U5 The boot cd is mounted via the virtual cdrom thru HP's ILO port. My isolinux.cfg contains the entry: label ap10903-02-net kernel vmlinuz append ks=http://10.1.181.252/cgi-bin/avamar/dl385_data_node initrd=initrd.img text And on the boot: line I use: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 I've used both nicdelay and linksleep with values of 50, 500, 5000, and 50000 with no discernable boot delays. The result is: I see vmliniz and initrd loading. I am presented the "welcome to red hat linux" blue screen (I'm in text mode) Then I see the usb and raid controller drivers load Then I see a "determine hostname and domain" ( I think this is where the timeout occurs ) Then I am prompted to "choose a language" In a different terminal window I am "pinging" the 10.196.154.122 IP during the whole boot process. Once the "choose a language" prompt appears, then the interface begins responding to my pings. My next step is to include the Kickstart config in the initrd. If anyone sees any issues with my method, please drop me a note. Thanks -Bill ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Vivek.Kalia at Euroclear.com Sent: Friday, March 02, 2007 10:49 AM To: kickstart-list at redhat.com Subject: RE: "pump" times out -- extend a few seconds more? Regarding experience of parameters nicdelay=50 linksleep=50 I used them today to circumvent the Anaconda "time-out" and can confirm that they worked in the required manner. My previous http builds were failing to contact the apache server due to the time-out. With these values in place, I could see Anaconda waiting about 50 seconds while the NIC initialised; the build then completed normally. I am still investigating the root cause of the time-out, but this seems a good circumvention. Vivek Kalia This e-mail message, including any attachments transmitted with it, is CONFIDENTIAL and may contain legally privileged information.This message is intended solely for the use of the individual or entity to whom it is addressed. If you are not the intended recipient you should not read, copy, distribute, disclose or otherwise use this information. If you have received this message in error, please notify us immediately and delete it from your system. The integrity and security of this message cannot be guaranteed and it may be subject to data corruption and unauthorised amendment, for which we accept no liability. Euroclear reserves the right to retain email messages on its systems and to the extent and under circumstances permitted by applicable law, to monitor and intercept email messages to and from its systems. Euroclear is the marketing name for the Euroclear System, Euroclear plc, Euroclear SA/NV and their affiliates. http://www.euroclear.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From Luca.Mariani at anritsu.com Wed Mar 7 09:55:00 2007 From: Luca.Mariani at anritsu.com (Mariani, Luca) Date: Wed, 7 Mar 2007 10:55:00 +0100 Subject: kernel parameters In-Reply-To: Message-ID: <2066A5EDF16FF848AAE6F1DEDAD977C62802E5@rom-exc-002.main.intgin.net> Hi CC, the only reason that let me modify kickstart files in order to add kernel parameters is that it was the only way I knew to do it do you think it should be better using grubby? I'll try this way too. thanks /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: 06 March 2007 22:46 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Luca Is there a reason you can't call grubby from the %post section to add the parameters you need? CC -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Wednesday, 7 March 2007 12:52 AM To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, now it can start but soon anaconda breaks reporting the following Exception: "An error occurrder unmounting the CD. Please make sure you're not accessing /mnt/source from the shell on tty2 and then clik OK to retry" Shell on tty2 is on / from /tmp/anaconda.log file I see this: ... ... * moving (1) to step readcomps * exception in unmountCD: (16,'Device or resouce busy') I did the following steps: - modified isolinux.cfg file to add the kernel parameter I need - modified some .msg files to customize this new CD - modified the kickstart files to add the kernel parameter to boot loader (as I did yesterday) - recreate the ISO Any ideas? I really don't known how to investigate on it... thanks /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: 06 March 2007 11:09 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, thanks a lot for your quickly response... I'll try this ASAP Bye Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Callahan, Tom Sent: 05 March 2007 19:34 To: Discussion list about Kickstart Subject: RE: kernel parameters You have to modify the syslinux files on the boot CD and rebuild the CD. Change the append= lines and add the option you need. Thanks, Tom Callahan TESSCO Technologies Desk: (410)-229-1361 Cell: (410)-588-7605 Email: callahant at tessco.com size=2 width="100%" align=center tabindex=-1> From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Monday, March 05, 2007 1:04 PM To: kickstart-list at redhat.com Subject: kernel parameters Hi all, I'm upgrading a bootable CD which uses kickstart to automatically install REHL4 Upgrade3 and a subset of programs. The upgrade is needed by a new hardware that requests a special kernel parameter to boot the system (in particular the kernel parameter is 'pci=nommconf' and it should be needed to use the Smart Array Controller) I've been able to add this parameter in kickstart config file and now the boot is working fine (I'm using GRUB as bootloader), but I must type the kernel parameter by hand at installation prompt... Is there a way to configure kickstart (or something else) in order to autamatically pass this kernel parameter to installation process too? Thanks Regards /Luca NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Colin.Coe at woodside.com.au Wed Mar 7 10:49:43 2007 From: Colin.Coe at woodside.com.au (Coe, Colin C. (Unix Engineer)) Date: Wed, 7 Mar 2007 19:49:43 +0900 Subject: kernel parameters References: <2066A5EDF16FF848AAE6F1DEDAD977C62802E5@rom-exc-002.main.intgin.net> Message-ID: Hi Luca 'grubby' is the tool to automate modifications to /etc/grub.conf (and the config files of other boot loaders). Amongst other things, in my %post section I have grubby --grub --update-kernel=`grubby --default-kernel` --args "${KERNEL_PARAMS} root=LABEL=/" Where $KERNEL_PARAMS are programatically determined previous in the %post section. Hope this helps. CC ________________________________ From: kickstart-list-bounces at redhat.com on behalf of Mariani, Luca Sent: Wed 3/7/2007 6:55 PM To: Discussion list about Kickstart Subject: RE: kernel parameters Hi CC, the only reason that let me modify kickstart files in order to add kernel parameters is that it was the only way I knew to do it do you think it should be better using grubby? I'll try this way too. thanks /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: 06 March 2007 22:46 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Luca Is there a reason you can't call grubby from the %post section to add the parameters you need? CC -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Wednesday, 7 March 2007 12:52 AM To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, now it can start but soon anaconda breaks reporting the following Exception: "An error occurrder unmounting the CD. Please make sure you're not accessing /mnt/source from the shell on tty2 and then clik OK to retry" Shell on tty2 is on / from /tmp/anaconda.log file I see this: ... ... * moving (1) to step readcomps * exception in unmountCD: (16,'Device or resouce busy') I did the following steps: - modified isolinux.cfg file to add the kernel parameter I need - modified some .msg files to customize this new CD - modified the kickstart files to add the kernel parameter to boot loader (as I did yesterday) - recreate the ISO Any ideas? I really don't known how to investigate on it... thanks /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: 06 March 2007 11:09 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, thanks a lot for your quickly response... I'll try this ASAP Bye Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Callahan, Tom Sent: 05 March 2007 19:34 To: Discussion list about Kickstart Subject: RE: kernel parameters You have to modify the syslinux files on the boot CD and rebuild the CD. Change the append= lines and add the option you need. Thanks, Tom Callahan TESSCO Technologies Desk: (410)-229-1361 Cell: (410)-588-7605 Email: callahant at tessco.com size=2 width="100%" align=center tabindex=-1> From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Monday, March 05, 2007 1:04 PM To: kickstart-list at redhat.com Subject: kernel parameters Hi all, I'm upgrading a bootable CD which uses kickstart to automatically install REHL4 Upgrade3 and a subset of programs. The upgrade is needed by a new hardware that requests a special kernel parameter to boot the system (in particular the kernel parameter is 'pci=nommconf' and it should be needed to use the Smart Array Controller) I've been able to add this parameter in kickstart config file and now the boot is working fine (I'm using GRUB as bootloader), but I must type the kernel parameter by hand at installation prompt... Is there a way to configure kickstart (or something else) in order to autamatically pass this kernel parameter to installation process too? Thanks Regards /Luca NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- A non-text attachment was scrubbed... Name: winmail.dat Type: application/ms-tnef Size: 13838 bytes Desc: not available URL: From Luca.Mariani at anritsu.com Wed Mar 7 14:26:02 2007 From: Luca.Mariani at anritsu.com (Mariani, Luca) Date: Wed, 7 Mar 2007 15:26:02 +0100 Subject: kernel parameters In-Reply-To: Message-ID: <2066A5EDF16FF848AAE6F1DEDAD977C6280385@rom-exc-002.main.intgin.net> Hi CC, thanks a lot for you help... however by now I'm not able to perform the installation owing to anaconda failures on mounting or umounting the CD :( /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: 07 March 2007 11:50 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Luca 'grubby' is the tool to automate modifications to /etc/grub.conf (and the config files of other boot loaders). Amongst other things, in my %post section I have grubby --grub --update-kernel=`grubby --default-kernel` --args "${KERNEL_PARAMS} root=LABEL=/" Where $KERNEL_PARAMS are programatically determined previous in the %post section. Hope this helps. CC ________________________________ From: kickstart-list-bounces at redhat.com on behalf of Mariani, Luca Sent: Wed 3/7/2007 6:55 PM To: Discussion list about Kickstart Subject: RE: kernel parameters Hi CC, the only reason that let me modify kickstart files in order to add kernel parameters is that it was the only way I knew to do it do you think it should be better using grubby? I'll try this way too. thanks /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: 06 March 2007 22:46 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Luca Is there a reason you can't call grubby from the %post section to add the parameters you need? CC -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Wednesday, 7 March 2007 12:52 AM To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, now it can start but soon anaconda breaks reporting the following Exception: "An error occurrder unmounting the CD. Please make sure you're not accessing /mnt/source from the shell on tty2 and then clik OK to retry" Shell on tty2 is on / from /tmp/anaconda.log file I see this: ... ... * moving (1) to step readcomps * exception in unmountCD: (16,'Device or resouce busy') I did the following steps: - modified isolinux.cfg file to add the kernel parameter I need - modified some .msg files to customize this new CD - modified the kickstart files to add the kernel parameter to boot loader (as I did yesterday) - recreate the ISO Any ideas? I really don't known how to investigate on it... thanks /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: 06 March 2007 11:09 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, thanks a lot for your quickly response... I'll try this ASAP Bye Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Callahan, Tom Sent: 05 March 2007 19:34 To: Discussion list about Kickstart Subject: RE: kernel parameters You have to modify the syslinux files on the boot CD and rebuild the CD. Change the append= lines and add the option you need. Thanks, Tom Callahan TESSCO Technologies Desk: (410)-229-1361 Cell: (410)-588-7605 Email: callahant at tessco.com size=2 width="100%" align=center tabindex=-1> From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Monday, March 05, 2007 1:04 PM To: kickstart-list at redhat.com Subject: kernel parameters Hi all, I'm upgrading a bootable CD which uses kickstart to automatically install REHL4 Upgrade3 and a subset of programs. The upgrade is needed by a new hardware that requests a special kernel parameter to boot the system (in particular the kernel parameter is 'pci=nommconf' and it should be needed to use the Smart Array Controller) I've been able to add this parameter in kickstart config file and now the boot is working fine (I'm using GRUB as bootloader), but I must type the kernel parameter by hand at installation prompt... Is there a way to configure kickstart (or something else) in order to autamatically pass this kernel parameter to installation process too? Thanks Regards /Luca NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Luca.Mariani at anritsu.com Wed Mar 7 14:49:03 2007 From: Luca.Mariani at anritsu.com (Mariani, Luca) Date: Wed, 7 Mar 2007 15:49:03 +0100 Subject: kernel parameters In-Reply-To: <2066A5EDF16FF848AAE6F1DEDAD977C6280385@rom-exc-002.main.intgin.net> Message-ID: <2066A5EDF16FF848AAE6F1DEDAD977C6280395@rom-exc-002.main.intgin.net> Hi, I'm sorry... the problem while trying to mount CD is not an anaconda's failure cause it happens befor running anaconda Bye ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: 07 March 2007 15:26 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi CC, thanks a lot for you help... however by now I'm not able to perform the installation owing to anaconda failures on mounting or umounting the CD :( /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: 07 March 2007 11:50 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Luca 'grubby' is the tool to automate modifications to /etc/grub.conf (and the config files of other boot loaders). Amongst other things, in my %post section I have grubby --grub --update-kernel=`grubby --default-kernel` --args "${KERNEL_PARAMS} root=LABEL=/" Where $KERNEL_PARAMS are programatically determined previous in the %post section. Hope this helps. CC ________________________________ From: kickstart-list-bounces at redhat.com on behalf of Mariani, Luca Sent: Wed 3/7/2007 6:55 PM To: Discussion list about Kickstart Subject: RE: kernel parameters Hi CC, the only reason that let me modify kickstart files in order to add kernel parameters is that it was the only way I knew to do it do you think it should be better using grubby? I'll try this way too. thanks /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: 06 March 2007 22:46 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Luca Is there a reason you can't call grubby from the %post section to add the parameters you need? CC -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Wednesday, 7 March 2007 12:52 AM To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, now it can start but soon anaconda breaks reporting the following Exception: "An error occurrder unmounting the CD. Please make sure you're not accessing /mnt/source from the shell on tty2 and then clik OK to retry" Shell on tty2 is on / from /tmp/anaconda.log file I see this: ... ... * moving (1) to step readcomps * exception in unmountCD: (16,'Device or resouce busy') I did the following steps: - modified isolinux.cfg file to add the kernel parameter I need - modified some .msg files to customize this new CD - modified the kickstart files to add the kernel parameter to boot loader (as I did yesterday) - recreate the ISO Any ideas? I really don't known how to investigate on it... thanks /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: 06 March 2007 11:09 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, thanks a lot for your quickly response... I'll try this ASAP Bye Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Callahan, Tom Sent: 05 March 2007 19:34 To: Discussion list about Kickstart Subject: RE: kernel parameters You have to modify the syslinux files on the boot CD and rebuild the CD. Change the append= lines and add the option you need. Thanks, Tom Callahan TESSCO Technologies Desk: (410)-229-1361 Cell: (410)-588-7605 Email: callahant at tessco.com size=2 width="100%" align=center tabindex=-1> From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Monday, March 05, 2007 1:04 PM To: kickstart-list at redhat.com Subject: kernel parameters Hi all, I'm upgrading a bootable CD which uses kickstart to automatically install REHL4 Upgrade3 and a subset of programs. The upgrade is needed by a new hardware that requests a special kernel parameter to boot the system (in particular the kernel parameter is 'pci=nommconf' and it should be needed to use the Smart Array Controller) I've been able to add this parameter in kickstart config file and now the boot is working fine (I'm using GRUB as bootloader), but I must type the kernel parameter by hand at installation prompt... Is there a way to configure kickstart (or something else) in order to autamatically pass this kernel parameter to installation process too? Thanks Regards /Luca NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Luca.Mariani at anritsu.com Wed Mar 7 18:28:44 2007 From: Luca.Mariani at anritsu.com (Mariani, Luca) Date: Wed, 7 Mar 2007 19:28:44 +0100 Subject: kernel parameters In-Reply-To: <2066A5EDF16FF848AAE6F1DEDAD977C6280395@rom-exc-002.main.intgin.net> Message-ID: <2066A5EDF16FF848AAE6F1DEDAD977C62803DF@rom-exc-002.main.intgin.net> Hi all, eventually it seems to work... I had to rebuild all step by step trying to find out where the problem was... :( I was not able to understand hat I was doing wrong I whish to thanks all the people that helps me BR /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: 07 March 2007 15:49 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi, I'm sorry... the problem while trying to mount CD is not an anaconda's failure cause it happens befor running anaconda Bye ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: 07 March 2007 15:26 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi CC, thanks a lot for you help... however by now I'm not able to perform the installation owing to anaconda failures on mounting or umounting the CD :( /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: 07 March 2007 11:50 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Luca 'grubby' is the tool to automate modifications to /etc/grub.conf (and the config files of other boot loaders). Amongst other things, in my %post section I have grubby --grub --update-kernel=`grubby --default-kernel` --args "${KERNEL_PARAMS} root=LABEL=/" Where $KERNEL_PARAMS are programatically determined previous in the %post section. Hope this helps. CC ________________________________ From: kickstart-list-bounces at redhat.com on behalf of Mariani, Luca Sent: Wed 3/7/2007 6:55 PM To: Discussion list about Kickstart Subject: RE: kernel parameters Hi CC, the only reason that let me modify kickstart files in order to add kernel parameters is that it was the only way I knew to do it do you think it should be better using grubby? I'll try this way too. thanks /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Coe, Colin C. (Unix Engineer) Sent: 06 March 2007 22:46 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Luca Is there a reason you can't call grubby from the %post section to add the parameters you need? CC -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Wednesday, 7 March 2007 12:52 AM To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, now it can start but soon anaconda breaks reporting the following Exception: "An error occurrder unmounting the CD. Please make sure you're not accessing /mnt/source from the shell on tty2 and then clik OK to retry" Shell on tty2 is on / from /tmp/anaconda.log file I see this: ... ... * moving (1) to step readcomps * exception in unmountCD: (16,'Device or resouce busy') I did the following steps: - modified isolinux.cfg file to add the kernel parameter I need - modified some .msg files to customize this new CD - modified the kickstart files to add the kernel parameter to boot loader (as I did yesterday) - recreate the ISO Any ideas? I really don't known how to investigate on it... thanks /Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: 06 March 2007 11:09 To: Discussion list about Kickstart Subject: RE: kernel parameters Hi Tom, thanks a lot for your quickly response... I'll try this ASAP Bye Luca ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Callahan, Tom Sent: 05 March 2007 19:34 To: Discussion list about Kickstart Subject: RE: kernel parameters You have to modify the syslinux files on the boot CD and rebuild the CD. Change the append= lines and add the option you need. Thanks, Tom Callahan TESSCO Technologies Desk: (410)-229-1361 Cell: (410)-588-7605 Email: callahant at tessco.com size=2 width="100%" align=center tabindex=-1> From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Mariani, Luca Sent: Monday, March 05, 2007 1:04 PM To: kickstart-list at redhat.com Subject: kernel parameters Hi all, I'm upgrading a bootable CD which uses kickstart to automatically install REHL4 Upgrade3 and a subset of programs. The upgrade is needed by a new hardware that requests a special kernel parameter to boot the system (in particular the kernel parameter is 'pci=nommconf' and it should be needed to use the Smart Array Controller) I've been able to add this parameter in kickstart config file and now the boot is working fine (I'm using GRUB as bootloader), but I must type the kernel parameter by hand at installation prompt... Is there a way to configure kickstart (or something else) in order to autamatically pass this kernel parameter to installation process too? Thanks Regards /Luca NOTICE: This email and any attachments are confidential. They may contain legally privileged information or copyright material. You must not read, copy, use or disclose them without authorisation. If you are not an intended recipient, please contact us at once by return email and then delete both messages and all attachments. -------------- next part -------------- An HTML attachment was scrubbed... URL: From William.Ramthun at Level3.com Wed Mar 7 18:56:54 2007 From: William.Ramthun at Level3.com (Ramthun, William) Date: Wed, 7 Mar 2007 11:56:54 -0700 Subject: "pump" times out -- extend a few seconds more? In-Reply-To: Message-ID: Actually the kickstart infrastructure that I have setup today uses a generic boot cd. I have the SA's specify the IP on the boot: line along with the ks=http://, the ks url is actually a cgi script that reads a static config file. The config file has markers in it that are replaced with actual values based on certain things. For instance, the cgi looks at the IP making the connection, does a reverse lookup to get the fqdn, the hostname is plucked from the fqdn, and the actual hostname replaces the hostname_marker in the static config file just before it's sent to the install client. This allows me to have one config for a type of install, but allows it to be customizable on a per client basis. I just ran into the stupid server NIC to switchport timeout problem. I guess I could get to the same place if I had a generic ks.cfg imbedded in the initrd, and used wget and %include. This actually might work better than what I have today as it should work with the timeout problem where as my stuff did not. Hmmm. -Bill ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip Sent: Tuesday, March 06, 2007 7:04 PM To: Discussion list about Kickstart Subject: RE: "pump" times out -- extend a few seconds more? > (So typically when an issue is solved, does the person with the issue and resolution write a summary for the list?) Usually all I see is either a "it worked" or those with problems stop asking questions :( If you ever want to make a "generic" boot cd, you can assign the IP address on the boot: line, then have just what you need to get the machine up in the embedded ks.cfg, and use wget, or nfs (not sure about ftp/tftp) to pull in the system specific config, then use %include to add it into the ks.cfg. This is the way we are moving to. Glad you got things working, Chip ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, William Sent: Tuesday, March 06, 2007 4:01 PM To: Discussion list about Kickstart Subject: RE: "pump" times out -- extend a few seconds more? Chip, Thanks for the help! I did try setting ethtool_opts="autoneg off speed 100 duplex full" along with the nicdelay=50 earlier and received no love for my efforts. I did build the 20 configs and wrapped them in the initrd.img - and that worked. I did have to edit all 20 configs for a cut/paste line-wrap error. But now the 20 systems are built. I really appreciate the help in working to a solution. Since this problem is solved for now and for all my other installation needs the ks=http:// works, I am not sure that I will ever learn what my network problem was. (So typically when an issue is solved, does the person with the issue and resolution write a summary for the list?) Regards, -Bill ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Shabazian, Chip Sent: Monday, March 05, 2007 12:59 PM To: Discussion list about Kickstart Subject: RE: "pump" times out -- extend a few seconds more? Try the following: If you are connected to a gigabit switch: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 nicdelay=50 linksleep=50 ETHTOOL_OPTS="autoneg on speed 1000 duplex full" If you are connected to a 100Mb switch: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 nicdelay=50 linksleep=50 ETHTOOL_OPTS="autoneg off speed 100 duplex full" ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Ramthun, William Sent: Monday, March 05, 2007 8:51 AM To: Discussion list about Kickstart Subject: RE: "pump" times out -- extend a few seconds more? Summarizing my situation: Client HW: HP DL385 Client OS: RHEL3U5 The boot cd is mounted via the virtual cdrom thru HP's ILO port. My isolinux.cfg contains the entry: label ap10903-02-net kernel vmlinuz append ks=http://10.1.181.252/cgi-bin/avamar/dl385_data_node initrd=initrd.img text And on the boot: line I use: Boot: ap10903-02-net ksdevice=eth0 ip=10.196.254.122 netmask=255.255.255.0 gateway=10.196.254.1 I've used both nicdelay and linksleep with values of 50, 500, 5000, and 50000 with no discernable boot delays. The result is: I see vmliniz and initrd loading. I am presented the "welcome to red hat linux" blue screen (I'm in text mode) Then I see the usb and raid controller drivers load Then I see a "determine hostname and domain" ( I think this is where the timeout occurs ) Then I am prompted to "choose a language" In a different terminal window I am "pinging" the 10.196.154.122 IP during the whole boot process. Once the "choose a language" prompt appears, then the interface begins responding to my pings. My next step is to include the Kickstart config in the initrd. If anyone sees any issues with my method, please drop me a note. Thanks -Bill ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Vivek.Kalia at Euroclear.com Sent: Friday, March 02, 2007 10:49 AM To: kickstart-list at redhat.com Subject: RE: "pump" times out -- extend a few seconds more? Regarding experience of parameters nicdelay=50 linksleep=50 I used them today to circumvent the Anaconda "time-out" and can confirm that they worked in the required manner. My previous http builds were failing to contact the apache server due to the time-out. With these values in place, I could see Anaconda waiting about 50 seconds while the NIC initialised; the build then completed normally. I am still investigating the root cause of the time-out, but this seems a good circumvention. Vivek Kalia This e-mail message, including any attachments transmitted with it, is CONFIDENTIAL and may contain legally privileged information.This message is intended solely for the use of the individual or entity to whom it is addressed. If you are not the intended recipient you should not read, copy, distribute, disclose or otherwise use this information. If you have received this message in error, please notify us immediately and delete it from your system. The integrity and security of this message cannot be guaranteed and it may be subject to data corruption and unauthorised amendment, for which we accept no liability. Euroclear reserves the right to retain email messages on its systems and to the extent and under circumstances permitted by applicable law, to monitor and intercept email messages to and from its systems. Euroclear is the marketing name for the Euroclear System, Euroclear plc, Euroclear SA/NV and their affiliates. http://www.euroclear.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From debian at herakles.homelinux.org Wed Mar 7 14:23:35 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Wed, 07 Mar 2007 23:23:35 +0900 Subject: kernel parameters In-Reply-To: <2066A5EDF16FF848AAE6F1DEDAD977C6280176@rom-exc-002.main.intgin.net> References: <2066A5EDF16FF848AAE6F1DEDAD977C6280176@rom-exc-002.main.intgin.net> Message-ID: <45EECAE7.8030601@herakles.homelinux.org> Mariani, Luca wrote: > Hi all, > > I'm upgrading a bootable CD which uses kickstart to automatically > install REHL4 Upgrade3 and a subset of programs. > > The upgrade is needed by a new hardware that requests a special kernel > parameter to boot the system (in particular the kernel parameter is > 'pci=nommconf' and it should be needed to use the Smart Array > Controller) > > I've been able to add this parameter in kickstart config file and now > the boot is working fine (I'm using GRUB as bootloader), but I must type > the kernel parameter by hand at installation prompt... > > > Is there a way to configure kickstart (or something else) in order to > autamatically pass this kernel parameter to installation process too? I'd do a simple edit in %pre - if you wish, you can simply replace the menu, though I'm generally a little more conservative. This has proven effective: [ -f /boot/grub/grub.conf.orig ] \ || cp -p /boot/grub/grub.conf /boot/grub/grub.conf.orig cat <<. >/boot/grub/menu.extra title Reinstall Interprise Linux root (hd0,0) kernel /boot/vmlinuz vga=6 initrd /boot/initrd title Kickstart Reinstall Interprise Linux ES (2.6.9-1.648_EL) root (hd0,0) kernel /boot/vmlinuz vga=6 ks=http://RHEL.demo.lan/ks/ks initrd /boot/initrd # If this is yor last line, your email system's broken. . cat /boot/grub/grub.conf.orig \ /boot/grub/menu.extra >/boot/grub/grub.conf mv -vf /boot/grub/grub.conf.orig /boot/grub/menu.extra ~root -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From tbergema at eso.org Fri Mar 9 13:20:36 2007 From: tbergema at eso.org (Timo Bergemann) Date: Fri, 09 Mar 2007 14:20:36 +0100 Subject: packages not to update during the KS installation Message-ID: <45F15F24.9060008@eso.org> Hello all, I have the following situation with the following problem: four repositorys: 1.) base 2.) updates 3.) extras 4.) eso ( our own repository ) i am installing using the options: repo --name=updates ..... repo --name=extras .... repo --name=eso .... %packages --resolvedeps after the installation everything is already updated, which also very fine :) But, the kernel I do not want to update. usually I would like to have the kernel and its family (-devel ...)i want to have in repository ESO. As soon there is a newer kernel in the repository "updates" it gets of course installed . Is there a way of saying something like: do never update kernel from repository "update"? I do not want to exclude the kernel get not mirrored into our updates repo, is there another way? I am searching for a way like "exclude=kernel,kernel-smp,kernel-devel,kernel-smp-devel" which I could give in /etc/yum.repos.d/*.repo. Unfortunaly during the installation such files not exists, or am i wrong here? Thank you for your help Timo Bergemann -- \|/ (o o) _______________________oOO__(.)__OOo_________________________ ______ ______ _______ / ____/ / ____/ / ___ / Timo Bergemann / /___ / /___ / / / / European Southern Observatory / ____/ /___ / / / / / Karl-Schwarzschildstr. 2 / /___ ___ / / / /__/ / D-85748 Garching bei Muenchen /_____/ /_____/ /______/ Germany fax : +49 89 3200 6380 e-mail : tbergema at eso.org voice : +49 89 3200 6760 www : http://www.eso.org _____________________________________________________________ From bill at wiliweld.com Wed Mar 14 05:06:17 2007 From: bill at wiliweld.com (Bill-S) Date: Tue, 13 Mar 2007 22:06:17 -0700 (PDT) Subject: Adding Broadcom BCM5708 (nic) module to kickstart? Message-ID: <20070313215917.V53874@corten11-bsd.billschoolcraft.com> Hello Family, I was curious as to how to add a newer nic driver to a pre-exisiting kickstart setup. Broadcom Corporation NetXtreme II BCM5708 Gigabit Ethernet (rev 11) With RHEL4 and the new DELL-2950's, it appears that there is no support for the following network card. I ran CentOS-4.4 "live" and got the following info for the nic came up then: ############################# /etc/modprobe.conf ############################# alias eth1 bnx2 alias eth0 bnx2 alias scsi_hostadapter megaraid_sas alias usb-controller ehci-hcd alias usb-controller1 uhci-hcd ############################# [root at localhost ~]# grep bnx2 /lib/modules/2.6.9-42.livecd.c4/modules.dep /lib/modules/2.6.9-42.livecd.c4/kernel/drivers/net/bnx2.ko: ############################ TIA From cjk at techma.com Wed Mar 14 12:43:37 2007 From: cjk at techma.com (Kovacs, Corey J.) Date: Wed, 14 Mar 2007 08:43:37 -0400 Subject: Adding Broadcom BCM5708 (nic) module to kickstart? In-Reply-To: <20070313215917.V53874@corten11-bsd.billschoolcraft.com> References: <20070313215917.V53874@corten11-bsd.billschoolcraft.com> Message-ID: <7DCE72B3C36E2A45B7580F887EE4948C18FDCC@TMAEMAIL.techma.com> What version of RHEL did you use? I run RHEL4.4 on HP DL series machines which have the same nic and it works just fine. CentOS 4.4 corrosponds to RHEL4.4 so if you tried an earlier release of RHEL then that might explain it. Be aware that the bnx2 driver shipping with RHEL4.4 has a nasty (but rarely seen) bug which took our cluster nodes down every few hours. Something the dlm was doing triggered it. Corey Kovacs Senior Systems Engineer Technology Management Associates 703.279.6168 (B) 855-6168 (R) -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Bill-S Sent: Wednesday, March 14, 2007 1:06 AM To: Kickstart Subject: Adding Broadcom BCM5708 (nic) module to kickstart? Hello Family, I was curious as to how to add a newer nic driver to a pre-exisiting kickstart setup. Broadcom Corporation NetXtreme II BCM5708 Gigabit Ethernet (rev 11) With RHEL4 and the new DELL-2950's, it appears that there is no support for the following network card. I ran CentOS-4.4 "live" and got the following info for the nic came up then: ############################# /etc/modprobe.conf ############################# alias eth1 bnx2 alias eth0 bnx2 alias scsi_hostadapter megaraid_sas alias usb-controller ehci-hcd alias usb-controller1 uhci-hcd ############################# [root at localhost ~]# grep bnx2 /lib/modules/2.6.9-42.livecd.c4/modules.dep /lib/modules/2.6.9-42.livecd.c4/kernel/drivers/net/bnx2.ko: ############################ TIA _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From bill at wiliweld.com Wed Mar 14 14:38:04 2007 From: bill at wiliweld.com (Bill-S) Date: Wed, 14 Mar 2007 07:38:04 -0700 (PDT) Subject: Adding Broadcom BCM5708 (nic) module to kickstart? In-Reply-To: <7DCE72B3C36E2A45B7580F887EE4948C18FDCC@TMAEMAIL.techma.com> References: <20070313215917.V53874@corten11-bsd.billschoolcraft.com> <7DCE72B3C36E2A45B7580F887EE4948C18FDCC@TMAEMAIL.techma.com> Message-ID: <20070314073629.D53874@corten11-bsd.billschoolcraft.com> At Wed, 14 Mar 2007 it looks like Kovacs, Corey J. composed: > What version of RHEL did you use? I run RHEL4.4 on HP DL series machines > which have the same nic and it works just fine. CentOS 4.4 corrosponds to > RHEL4.4 so if you tried an earlier release of RHEL then that might explain > it. > > Be aware that the bnx2 driver shipping with RHEL4.4 has a nasty (but rarely > seen) bug which took our cluster nodes down every few hours. Something the > dlm was doing triggered it. > > Corey Kovacs > Senior Systems Engineer > Technology Management Associates > 703.279.6168 (B) > 855-6168 (R) > > Thanks Corey, We use the first release of RHEL4 shipped from RedHat. I was wondering if I could add the CentOS-4.4 driver to the pre-existing kickstart engine we are using? Sound plausible? Thanks Corey > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Bill-S > Sent: Wednesday, March 14, 2007 1:06 AM > To: Kickstart > Subject: Adding Broadcom BCM5708 (nic) module to kickstart? > > Hello Family, > > I was curious as to how to add a newer nic driver to a pre-exisiting > kickstart setup. > > Broadcom Corporation NetXtreme II BCM5708 Gigabit Ethernet (rev 11) > > With RHEL4 and the new DELL-2950's, it appears that there is no support for > the following network card. I ran CentOS-4.4 "live" and got the following > info for the nic came up then: > > ############################# > /etc/modprobe.conf > ############################# > alias eth1 bnx2 > alias eth0 bnx2 > alias scsi_hostadapter megaraid_sas > alias usb-controller ehci-hcd > alias usb-controller1 uhci-hcd > ############################# > > [root at localhost ~]# > > grep bnx2 /lib/modules/2.6.9-42.livecd.c4/modules.dep > > /lib/modules/2.6.9-42.livecd.c4/kernel/drivers/net/bnx2.ko: > > ############################ > > TIA > > > > _______________________________________________ > 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 > -- Bill Schoolcraft <*> http://wiliweld.com ~ "Unix is very simple, but it takes a genius to understand the simplicity." (Dennis Ritchie) From adas at Yodlee.com Tue Mar 20 07:22:50 2007 From: adas at Yodlee.com (Abhijit Das) Date: Tue, 20 Mar 2007 00:22:50 -0700 Subject: KickStart Issue Message-ID: <947003D94A6BCB4EB71FDC1DCB47E024028CDA59@rwsmail.corp.yodlee.com> Dear Gurus I need a help here. I have create a RHEL4ASU4 32bit source server and configured dhcp/nfs/kickstart on it. I am also using the initrd.img and vmlinuz from the first CD to pxe boot the destination machines. I use the pxelinux concept of kickstarting. Here is my default file LABEL pxe MENU LABEL pxe kernel vmlinuz-pxe append ksdevice=eth0 console=tty0 load_ramdisk=1 initrd=initrd-pxe.img network ks=nfs:192.168.203.180:/kickstart/rhelas4u4/i386/kickstart.cfg As you see, I am passing the kickstart parameters via additional boot loader parameters. /kickstart/rhelas4u4/i386 contains the entire distribution of RHEL4ASU4 Now am using a kickstart configuration file (client's file) wherein he calls many post install and pre install scripts. I get the following error :- IOError: [Errno 2] No such file or directory : '/tmp/xxxxxx' This directory is mentioned in the %pre script to be created. But looks like it is unable to. Before that I get bunch of File "/usr/lib/anaconda/kickstary.py" , line numbers messages but no errors on that. Would appreciate your help and direction regards this Thanks - Abhijit Das -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chip.Shabazian at bankofamerica.com Tue Mar 20 15:35:30 2007 From: Chip.Shabazian at bankofamerica.com (Shabazian, Chip) Date: Tue, 20 Mar 2007 08:35:30 -0700 Subject: KickStart Issue In-Reply-To: <947003D94A6BCB4EB71FDC1DCB47E024028CDA59@rwsmail.corp.yodlee.com> Message-ID: The way kickstart works is that it copies your kickstart file from the remote location to the local filesystem as /tmp/ks.cfg, so it sounds like it's not getting your kickstart file. The easiest way to troubleshoot is to put a sleep statement in your %pre and/or %post, then go to the Alt-F2 terminal and poke around. You will see your ks.cfg file in /tmp if it's been copied across, etc. ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Abhijit Das Sent: Tuesday, March 20, 2007 12:23 AM To: kickstart-list at redhat.com Cc: Abhijit Das Subject: KickStart Issue Dear Gurus I need a help here. I have create a RHEL4ASU4 32bit source server and configured dhcp/nfs/kickstart on it. I am also using the initrd.img and vmlinuz from the first CD to pxe boot the destination machines. I use the pxelinux concept of kickstarting. Here is my default file LABEL pxe MENU LABEL pxe kernel vmlinuz-pxe append ksdevice=eth0 console=tty0 load_ramdisk=1 initrd=initrd-pxe.img network ks=nfs:192.168.203.180:/kickstart/rhelas4u4/i386/kickstart.cfg As you see, I am passing the kickstart parameters via additional boot loader parameters. /kickstart/rhelas4u4/i386 contains the entire distribution of RHEL4ASU4 Now am using a kickstart configuration file (client's file) wherein he calls many post install and pre install scripts. I get the following error :- IOError: [Errno 2] No such file or directory : '/tmp/xxxxxx' This directory is mentioned in the %pre script to be created. But looks like it is unable to. Before that I get bunch of File "/usr/lib/anaconda/kickstary.py" , line numbers messages but no errors on that. Would appreciate your help and direction regards this Thanks - Abhijit Das -------------- next part -------------- An HTML attachment was scrubbed... URL: From adas at Yodlee.com Wed Mar 21 00:03:28 2007 From: adas at Yodlee.com (Abhijit Das) Date: Tue, 20 Mar 2007 17:03:28 -0700 Subject: SUMMARY: RE: KickStart Issue In-Reply-To: <947003D94A6BCB4EB71FDC1DCB47E024028CDA59@rwsmail.corp.yodlee.com> Message-ID: <947003D94A6BCB4EB71FDC1DCB47E024028CDA64@rwsmail.corp.yodlee.com> Thanks very much to Chip Shabazian (BofA) who advised me to put debug statements like sleep and echo and run the commands manually and check the stages. The actual problem was that I copied the file from a windows machine and probably linux messed up with the file format. I did a dos2unix on the file and was able to get past the errors below. Thanks - Abhijit Das > _____________________________________________ > From: Abhijit Das > Sent: Tuesday, March 20, 2007 12:23 AM > To: 'kickstart-list at redhat.com' > Cc: Abhijit Das > Subject: KickStart Issue > > Dear Gurus > > I need a help here. I have create a RHEL4ASU4 32bit source server and > configured dhcp/nfs/kickstart on it. I am also using the initrd.img > and vmlinuz from the first CD to pxe boot the destination machines. I > use the pxelinux concept of kickstarting. Here is my default file > > LABEL pxe > MENU LABEL pxe > kernel vmlinuz-pxe > append ksdevice=eth0 console=tty0 load_ramdisk=1 > initrd=initrd-pxe.img network > ks=nfs:192.168.203.180:/kickstart/rhelas4u4/i386/kickstart.cfg > > As you see, I am passing the kickstart parameters via additional boot > loader parameters. > /kickstart/rhelas4u4/i386 contains the entire distribution of > RHEL4ASU4 > > Now am using a kickstart configuration file (client's file) wherein he > calls many post install and pre install scripts. I get the following > error :- > > IOError: [Errno 2] No such file or directory : '/tmp/xxxxxx' > This directory is mentioned in the %pre script to be created. But > looks like it is unable to. > > Before that I get bunch of File "/usr/lib/anaconda/kickstary.py" , > line numbers messages but no errors on that. > > Would appreciate your help and direction regards this > > Thanks > - Abhijit Das > -------------- next part -------------- An HTML attachment was scrubbed... URL: From cjk at techma.com Wed Mar 21 12:09:13 2007 From: cjk at techma.com (Kovacs, Corey J.) Date: Wed, 21 Mar 2007 08:09:13 -0400 Subject: SUMMARY: RE: KickStart Issue In-Reply-To: <947003D94A6BCB4EB71FDC1DCB47E024028CDA64@rwsmail.corp.yodlee.com> References: <947003D94A6BCB4EB71FDC1DCB47E024028CDA59@rwsmail.corp.yodlee.com> <947003D94A6BCB4EB71FDC1DCB47E024028CDA64@rwsmail.corp.yodlee.com> Message-ID: <7DCE72B3C36E2A45B7580F887EE4948C18FDDB@TMAEMAIL.techma.com> Just a note, there is no "probably linux messed up with the file format" issue here. Any *NIX variant treats linefeeds differently than DOS based utilites. What happens is DOS based utilites put different control characters at the end of text lines than *NIX utilities do. For as long as it has been this way, to expect a text file edited on a windows machine to appear as expected on a *NIX machine is not even in the realm of reasonable. Furthermore, since *NIX systems have been around much longer than DOS based systems, I'd say that "probably Windows messed up with the file format"... But thats just me :) Aside from all that, I'm glad you were able to figure it out and get things going. Have fun :) Regards. Corey Kovacs Senior Systems Engineer Technology Management Associates 703.279.6168 (B) 855-6168 (R) ________________________________ From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Abhijit Das Sent: Tuesday, March 20, 2007 8:03 PM To: kickstart-list at redhat.com Subject: SUMMARY: RE: KickStart Issue Thanks very much to Chip Shabazian (BofA) who advised me to put debug statements like sleep and echo and run the commands manually and check the stages. The actual problem was that I copied the file from a windows machine and probably linux messed up with the file format. I did a dos2unix on the file and was able to get past the errors below. Thanks - Abhijit Das _____________________________________________ From: Abhijit Das Sent: Tuesday, March 20, 2007 12:23 AM To: 'kickstart-list at redhat.com' Cc: Abhijit Das Subject: KickStart Issue Dear Gurus I need a help here. I have create a RHEL4ASU4 32bit source server and configured dhcp/nfs/kickstart on it. I am also using the initrd.img and vmlinuz from the first CD to pxe boot the destination machines. I use the pxelinux concept of kickstarting. Here is my default file LABEL pxe MENU LABEL pxe kernel vmlinuz-pxe append ksdevice=eth0 console=tty0 load_ramdisk=1 initrd=initrd-pxe.img network ks=nfs:192.168.203.180:/kickstart/rhelas4u4/i386/kickstart.cfg As you see, I am passing the kickstart parameters via additional boot loader parameters. /kickstart/rhelas4u4/i386 contains the entire distribution of RHEL4ASU4 Now am using a kickstart configuration file (client's file) wherein he calls many post install and pre install scripts. I get the following error :- IOError: [Errno 2] No such file or directory : '/tmp/xxxxxx' This directory is mentioned in the %pre script to be created. But looks like it is unable to. Before that I get bunch of File "/usr/lib/anaconda/kickstary.py" , line numbers messages but no errors on that. Would appreciate your help and direction regards this Thanks - Abhijit Das -------------- next part -------------- An HTML attachment was scrubbed... URL: From Eric.Doutreleau at int-evry.fr Wed Mar 21 14:19:49 2007 From: Eric.Doutreleau at int-evry.fr (Eric Doutreleau) Date: Wed, 21 Mar 2007 15:19:49 +0100 Subject: issue with the parameter xconfig in kickstart file Message-ID: <46013F05.4020409@int-evry.fr> Hi i begin to switch to FC6 for my linux machine and i have the following problem with the kickstart file. I wold like to have a default resolution for the machine to 1024x768 i put that on my kickstart file xconfig --defaultdesktop=GNOME --depth=24 --resolution=1024x768 --startxonboot monitor --monitor="Monitor 1280x1024" but the xorg.conf file is not configured. on the machine here what i have # Xorg configuration created by pyxf86config Section "ServerLayout" Identifier "Default Layout" Screen 0 "Screen0" 0 0 InputDevice "Keyboard0" "CoreKeyboard" EndSection Section "InputDevice" Identifier "Keyboard0" Driver "kbd" Option "XkbModel" "pc105" Option "XkbLayout" "fr" Option "XkbVariant" "latin9" EndSection Section "Device" Identifier "Videocard0" Driver "i810" EndSection Section "Screen" Identifier "Screen0" Device "Videocard0" DefaultDepth 24 SubSection "Display" Viewport 0 0 Depth 24 EndSubSection EndSection there s no default mode defined. how can i do that with kickstart? thanks in advance for any help From clumens at redhat.com Wed Mar 21 15:17:56 2007 From: clumens at redhat.com (Chris Lumens) Date: Wed, 21 Mar 2007 11:17:56 -0400 Subject: issue with the parameter xconfig in kickstart file In-Reply-To: <46013F05.4020409@int-evry.fr> References: <46013F05.4020409@int-evry.fr> Message-ID: <20070321151755.GK16460@exeter.boston.redhat.com> > i begin to switch to FC6 for my linux machine and i have the following > problem with the kickstart file. > > I wold like to have a default resolution for the machine to 1024x768 > > i put that on my kickstart file > > xconfig --defaultdesktop=GNOME --depth=24 --resolution=1024x768 > --startxonboot > monitor --monitor="Monitor 1280x1024" > > but the xorg.conf file is not configured. > on the machine here what i have Yeah, this was broken in FC6. Can you try a Fedora 7 test release and see if it works for you there? I made some commits that should fix this. - Chris From Eric.Doutreleau at int-evry.fr Wed Mar 21 15:21:06 2007 From: Eric.Doutreleau at int-evry.fr (Eric Doutreleau) Date: Wed, 21 Mar 2007 16:21:06 +0100 Subject: issue with the parameter xconfig in kickstart file In-Reply-To: <20070321151755.GK16460@exeter.boston.redhat.com> References: <46013F05.4020409@int-evry.fr> <20070321151755.GK16460@exeter.boston.redhat.com> Message-ID: <46014D62.8040906@int-evry.fr> Chris Lumens a ?crit : >> i begin to switch to FC6 for my linux machine and i have the following >> problem with the kickstart file. >> >> I wold like to have a default resolution for the machine to 1024x768 >> >> i put that on my kickstart file >> >> xconfig --defaultdesktop=GNOME --depth=24 --resolution=1024x768 >> --startxonboot >> monitor --monitor="Monitor 1280x1024" >> >> but the xorg.conf file is not configured. >> on the machine here what i have >> > > Yeah, this was broken in FC6. Can you try a Fedora 7 test release and > see if it works for you there? I made some commits that should fix > this. > > - Chris > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > well i could try a FC7 test release for testing but i can't use it on my production machine Is there any way to have the patch include in an update of FC6? From steven.hajducko at digitalinsight.com Wed Mar 21 19:09:04 2007 From: steven.hajducko at digitalinsight.com (Hajducko, Steven) Date: Wed, 21 Mar 2007 12:09:04 -0700 Subject: Digitally signed RPMs in Kickstart Message-ID: <21F8C5D33731F44BB9DCC396A12F9F210184AAC2@WLVEXM01.corp.ad.diginsite.com> Is it possible to have Kickstart verify digital signatures on RPMs during the installation process? We want to have a kickstart server that serves several different network tiers, however, our security team is concerned that if an attacker were to get onto the system, they could replace some of the RPM's with his own and each time a new system was kickstarted, would have the bogus RPM. -- sh -------------- next part -------------- An HTML attachment was scrubbed... URL: From clumens at redhat.com Wed Mar 21 19:43:34 2007 From: clumens at redhat.com (Chris Lumens) Date: Wed, 21 Mar 2007 15:43:34 -0400 Subject: Digitally signed RPMs in Kickstart In-Reply-To: <21F8C5D33731F44BB9DCC396A12F9F210184AAC2@WLVEXM01.corp.ad.diginsite.com> References: <21F8C5D33731F44BB9DCC396A12F9F210184AAC2@WLVEXM01.corp.ad.diginsite.com> Message-ID: <20070321194334.GO16460@exeter.boston.redhat.com> > Is it possible to have Kickstart verify digital signatures on RPMs > during the installation process? > > We want to have a kickstart server that serves several different network > tiers, however, our security team is concerned that if an attacker were > to get onto the system, they could replace some of the RPM's with his > own and each time a new system was kickstarted, would have the bogus > RPM. Nope. See the very old bug: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=998 - Chris From clumens at redhat.com Wed Mar 21 19:44:51 2007 From: clumens at redhat.com (Chris Lumens) Date: Wed, 21 Mar 2007 15:44:51 -0400 Subject: issue with the parameter xconfig in kickstart file In-Reply-To: <46014D62.8040906@int-evry.fr> References: <46013F05.4020409@int-evry.fr> <20070321151755.GK16460@exeter.boston.redhat.com> <46014D62.8040906@int-evry.fr> Message-ID: <20070321194451.GP16460@exeter.boston.redhat.com> > well i could try a FC7 test release for testing but i can't use it on my > production machine > Is there any way to have the patch include in an update of FC6? No, unfortunately we don't release updates of anaconda for previous versions. - Chris From psvdavey at sealand-systems.co.uk Thu Mar 22 04:12:04 2007 From: psvdavey at sealand-systems.co.uk (Peter Davey) Date: Thu, 22 Mar 2007 05:12:04 +0100 Subject: SUMMARY: RE: KickStart Issue Message-ID: <31627707.838711174536724593.JavaMail.servlet@kundenserver> An HTML attachment was scrubbed... URL: From John.Beamon at turner.com Fri Mar 23 16:05:09 2007 From: John.Beamon at turner.com (Beamon, John) Date: Fri, 23 Mar 2007 12:05:09 -0400 Subject: Install to local drive instead of a fibre attached SAN Message-ID: <10D60FF054C7BC40870DDB85A09B28F599ABBC@ATLMSG12.turner.com> We've had a few isolated incidents of kickstart installs assigning "/dev/sda" to a fibre attached SAN volume instead of the internal scsi disks. My environment is RHEL 4U4 on mixed 32- and 64-bit Intel hardware. Is there a way to script in %pre the exclusion of drives attached by an HBA? I've rewritten /tmp/modprobe.conf with "scsi_hostadapterN qla2300" as both first and last entry in the scsi hosts, but the lettering of the drive doesn't change. It appears to be set by the kernel image during boot. Can that be re-ordered? The closest thing I've found is that I can 'rmmod qla2300' in %pre, but the better long-term plan would be to configure internal or external by choice so we have the option of installing to the SAN later. -j -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.413 / Virus Database: 268.18.17/730 - Release Date: 3/22/2007 From smah at vmware.com Fri Mar 23 16:14:30 2007 From: smah at vmware.com (Stephen Mah) Date: Fri, 23 Mar 2007 09:14:30 -0700 Subject: Install to local drive instead of a fibre attached SAN In-Reply-To: <10D60FF054C7BC40870DDB85A09B28F599ABBC@ATLMSG12.turner.com> References: <10D60FF054C7BC40870DDB85A09B28F599ABBC@ATLMSG12.turner.com> Message-ID: <4603FCE6.90306@vmware.com> I did a quick google. Try passing "nostorage" http://archives.devshed.com/forums/linux-97/ignoring-san-devices-during-kickstart-1499210.html Beamon, John wrote: > We've had a few isolated incidents of kickstart installs assigning "/dev/sda" to a fibre attached SAN volume instead of the internal scsi disks. My environment is RHEL 4U4 on mixed 32- and 64-bit Intel hardware. Is there a way to script in %pre the exclusion of drives attached by an HBA? I've rewritten /tmp/modprobe.conf with "scsi_hostadapterN qla2300" as both first and last entry in the scsi hosts, but the lettering of the drive doesn't change. It appears to be set by the kernel image during boot. Can that be re-ordered? The closest thing I've found is that I can 'rmmod qla2300' in %pre, but the better long-term plan would be to configure internal or external by choice so we have the option of installing to the SAN later. > > -j > > From smah at vmware.com Fri Mar 23 16:21:12 2007 From: smah at vmware.com (Stephen Mah) Date: Fri, 23 Mar 2007 09:21:12 -0700 Subject: Install to local drive instead of a fibre attached SAN In-Reply-To: <4603FCE6.90306@vmware.com> References: <10D60FF054C7BC40870DDB85A09B28F599ABBC@ATLMSG12.turner.com> <4603FCE6.90306@vmware.com> Message-ID: <4603FE78.80604@vmware.com> Also, you can change the kickstart file's disk order I believe. we use cciss so even if a san is attached, it goes to sda (local storage) per the kickstart file. but I had to change the bootloader order via: bootloader --location=mbr --driveorder=cciss/c0d0,sda might work on the actual partitioning directive. Haven't tried it. Stephen Mah wrote: > I did a quick google. Try passing "nostorage" > > http://archives.devshed.com/forums/linux-97/ignoring-san-devices-during-kickstart-1499210.html > > > Beamon, John wrote: >> We've had a few isolated incidents of kickstart installs assigning >> "/dev/sda" to a fibre attached SAN volume instead of the internal >> scsi disks. My environment is RHEL 4U4 on mixed 32- and 64-bit Intel >> hardware. Is there a way to script in %pre the exclusion of drives >> attached by an HBA? I've rewritten /tmp/modprobe.conf with >> "scsi_hostadapterN qla2300" as both first and last entry in the scsi >> hosts, but the lettering of the drive doesn't change. It appears to >> be set by the kernel image during boot. Can that be re-ordered? The >> closest thing I've found is that I can 'rmmod qla2300' in %pre, but >> the better long-term plan would be to configure internal or external >> by choice so we have the option of installing to the SAN later. >> >> -j >> >> > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From John.Beamon at turner.com Fri Mar 23 20:08:16 2007 From: John.Beamon at turner.com (Beamon, John) Date: Fri, 23 Mar 2007 16:08:16 -0400 Subject: Install to local drive instead of a fibre attached SAN In-Reply-To: <4603FCE6.90306@vmware.com> References: <10D60FF054C7BC40870DDB85A09B28F599ABBC@ATLMSG12.turner.com> <4603FCE6.90306@vmware.com> Message-ID: <10D60FF054C7BC40870DDB85A09B28F599ABDC@ATLMSG12.turner.com> That helped. I didn't have the terms to search for. There's a boot option "nostorage" that will skip disk drivers. You then put "device scsi qla2300" to load only Qlogic as "/dev/sda" or "device scsi qla2300:mptscsih" to load both Qlogic and mptbase cards. I do NOT find that this method allows you to specify Qlogic as "/dev/sdb" or "/dev/sda" in a two-controller list. I also stumbled across a native kickstart boot option "latefcload", which loads fibre channel drivers last among all scsi modules. This is a good basic way to install the OS on local drives without having to worry about nuking an attached SAN volume. Thanks, Stephen. -j -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Stephen Mah Sent: Friday, March 23, 2007 12:15 PM To: Discussion list about Kickstart Subject: Re: Install to local drive instead of a fibre attached SAN I did a quick google. Try passing "nostorage" http://archives.devshed.com/forums/linux-97/ignoring-san-devices-during-kickstart-1499210.html Beamon, John wrote: > We've had a few isolated incidents of kickstart installs assigning "/dev/sda" to a fibre attached SAN volume instead of the internal scsi disks. My environment is RHEL 4U4 on mixed 32- and 64-bit Intel hardware. Is there a way to script in %pre the exclusion of drives attached by an HBA? I've rewritten /tmp/modprobe.conf with "scsi_hostadapterN qla2300" as both first and last entry in the scsi hosts, but the lettering of the drive doesn't change. It appears to be set by the kernel image during boot. Can that be re-ordered? The closest thing I've found is that I can 'rmmod qla2300' in %pre, but the better long-term plan would be to configure internal or external by choice so we have the option of installing to the SAN later. > > -j > > _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.413 / Virus Database: 268.18.17/730 - Release Date: 3/22/2007 -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.413 / Virus Database: 268.18.17/730 - Release Date: 3/22/2007 From jce at zot.com Fri Mar 23 23:43:31 2007 From: jce at zot.com (Chris Edillon) Date: Fri, 23 Mar 2007 18:43:31 -0500 Subject: Install to local drive instead of a fibre attached SAN In-Reply-To: <10D60FF054C7BC40870DDB85A09B28F599ABDC@ATLMSG12.turner.com> References: <10D60FF054C7BC40870DDB85A09B28F599ABBC@ATLMSG12.turner.com> <4603FCE6.90306@vmware.com> <10D60FF054C7BC40870DDB85A09B28F599ABDC@ATLMSG12.turner.com> Message-ID: <1174693412.3188.4.camel@ruthless> On Fri, 2007-03-23 at 16:08 -0400, Beamon, John wrote: > That helped. I didn't have the terms to search for. There's a > boot option "nostorage" that will skip disk drivers. You then > put "device scsi qla2300" to load only Qlogic as "/dev/sda" or > "device scsi qla2300:mptscsih" to load both Qlogic and mptbase > cards. I do NOT find that this method allows you to specify > Qlogic as "/dev/sdb" or "/dev/sda" in a two-controller list. > > I also stumbled across a native kickstart boot option "latefcload", > which loads fibre channel drivers last among all scsi modules. > This is a good basic way to install the OS on local drives without > having to worry about nuking an attached SAN volume. > i wish i had found that information about a year and a half ago. on that project i ended up hacking the initrd.img and stage2.img so the HBA drivers weren't loaded during kickstart. i take it you had success with "latefcload"? chris From drkludge at cox.net Mon Mar 26 17:55:37 2007 From: drkludge at cox.net (Greg Morgan) Date: Mon, 26 Mar 2007 10:55:37 -0700 Subject: "pump" times out -- extend a few seconds more? In-Reply-To: References: Message-ID: <46080919.3070005@cox.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ramthun, William wrote: > > (So typically when an issue is solved, does the person with the issue > and resolution write a summary for the list?) > I took the first stab at trying to organizing some of the issues relating to Anaconda and Network issues that I have seen on the list. You can find the information in the wiki at http://fedoraproject.org/wiki/AnacondaNetworkIssues I thought at this point I'd do a "release early and release often" page. There are some areas of the discussion where I do not have a network environment to make sure the information is correct. I was afraid of revising the email exchange into non-sense. :-( I hope this helps. Please advise, Greg -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFGCAkZxyxe5L6mr7IRArrHAJwP4dhvJReA23Nmc66Ay12Sfi4xJACfRjgD kUu1HbYaXXW1yNV+Qj455p8= =d6S7 -----END PGP SIGNATURE----- From ronald645 at gmail.com Wed Mar 28 13:26:51 2007 From: ronald645 at gmail.com (Ronald) Date: Wed, 28 Mar 2007 15:26:51 +0200 Subject: Repo error Message-ID: <460A6D1B.7020709@gmail.com> This is my kickstart file: auth --enablemd5 --enableshadow bootloader --location=mbr --password=blablabla firewall --disabled firstboot --reconfig install --partition=sdb2 --dir=/back-up/Os/fc/ interactive keyboard us-acentos lang nl_NL langsupport nl_NL mouse --device=input/mouse0 logitech part /mnt/shared --usepart=sdb1 --noformat --fstype=vfat part /mnt/data --usepart=sdb2 --noformat --fstype=ext3 part / --usepart=sda5 --fstype=ext3 rootpw blablabla selinux --disabled xconfig --startxonboot repo --name=extra --baseurl=http://download.fedora.redhat.com/pub/fedora/linux/extras/6/i386/ I'm using Fedora Core 6 and I'm trying to create a kickstart file for installation on multiple machines. My LAN-card works fine when I activate the Fedora Extra in a 'normal' installation. But I want to add external repository's with this kickstart (it seems that kickstart doesn't include Fedora Extra's by default so I try to add it), but it seems that: repo --name=extra --baseurl=http://download.fedora.redhat.com/pub/fedora/linux/extras/6/i386/ is wrong. I don't know why, or how. The error Anaconda gives is: "Unable to read package metadata. This may be due to a missing repodata directory. Please ensure that your install tree has been correctly generated. Cannot open / read repomd.xml for repository: extra." Can someone help me? From timm at fnal.gov Wed Mar 28 13:43:06 2007 From: timm at fnal.gov (Steven Timm) Date: Wed, 28 Mar 2007 08:43:06 -0500 (CDT) Subject: how to force inclusion of i386 packages in x86_64 distro Message-ID: Hi-- I have an x86_64 install that I am doing, which needs to be sure that about 20 i386-flavored rpms of various shared libraries are included in the install as well. Is there any way to force this in a regular kickstart file, if so, how? Thanks Steve Timm -- ------------------------------------------------------------------ Steven C. Timm, Ph.D (630) 840-8525 timm at fnal.gov http://home.fnal.gov/~timm/ Fermilab Computing Division, Scientific Computing Facilities, Grid Facilities Department, FermiGrid Services Group, Assistant Group Leader. From slarty.tj at gmail.com Wed Mar 28 15:13:36 2007 From: slarty.tj at gmail.com (Matt P) Date: Wed, 28 Mar 2007 10:13:36 -0500 Subject: Kickstarting an RX4640 Itanium Message-ID: <859a78260703280813g25e0dc42wc7b0b087d959bb5c@mail.gmail.com> I've been trying to build a bootable Kickstart DVD for RHEL 4 (AS4u3). I've got the booting part working no problem, all the packages are on the disc. We can even use the disc for a manual/interactive install. The problem I'm having with the process is getting Kickstart to partition the HDDs. I've got this whole process working like a champ on x86 hardware, but this /boot/efi has become a thorn in my side. The specifics are as follows: HP RX4640 4 Internal HDD, mapped into 2 RAID1 discs. The controller is recognized as CCISS The error I get is something like "can't find device cciss/c0d0 to use for /boot/efi" (sorry I don't have it word for word). At that point the only option is "OK" and reboot the machine. As I said above, if I use the same DVD and do a manual/interactive install everything works. Even if I let the installer build the partitions automatically it finds the cciss/c0d0 and labels it correctly /boot/efi. To add further confusion after the manual/interactive install is complete and I compare the "part" section of my ks.cfg with the one anaconda builds (/root/anaconda-ks.cfg) they are exactly the same.... Below is the section of my ks.cfg in question: part /boot/efi --fstype vfat --noformat --onpart cciss/c0d0p1 I've tried several permutations of --onpart, --ondisk, on that line. Also clearpart --all, --linux, --initlabel just before that line. Does anyone have any magic bullets, or examples of known working itanium ks.cfg's? Any help would be much appreciated. From ebrown at lanl.gov Wed Mar 28 15:01:16 2007 From: ebrown at lanl.gov (Ed Brown) Date: Wed, 28 Mar 2007 09:01:16 -0600 Subject: Repo error In-Reply-To: <460A6D1B.7020709@gmail.com> References: <460A6D1B.7020709@gmail.com> Message-ID: <460A833C.2080301@lanl.gov> Ronald wrote: > it seems that: > > repo --name=extra > --baseurl=http://download.fedora.redhat.com/pub/fedora/linux/extras/6/i386/ > > is wrong. I don't know why, or how. The error Anaconda gives is: > > "Unable to read package metadata. This may be due to a missing repodata > directory. Please ensure that your install tree has been correctly > generated. Cannot open / read repomd.xml for repository: extra." Looks like attempts to read from the repodata directory are being redirected to the repoview directory. ?? -Ed From Greg.Bulman at pdi.dreamworks.com Wed Mar 28 17:06:52 2007 From: Greg.Bulman at pdi.dreamworks.com (Greg Bulman) Date: Wed, 28 Mar 2007 10:06:52 -0700 Subject: how to force inclusion of i386 packages in x86_64 distro In-Reply-To: References: Message-ID: <460AA0AC.80104@pdi.com> In your list of RPMS you can add the dot notated architecture extensions. To include the i386 cyrus-sasl-gssapi on an x86_64 OS install list this cyrus-sasl-gssapi.i386 To exclude openssl096b (both of them since they both by default install) do this -openssl096b.i386 -openssl096b.x86_64 This assumes the i386 rpms are in the channels or repositories already and you just want the installer to pick them up. -Greg Bulman Steven Timm wrote: > > Hi-- > I have an x86_64 install that I am doing, which needs to be sure > that about 20 i386-flavored rpms of various shared libraries > are included in the install as well. Is there > any way to force this in a regular kickstart file, if so, how? > > Thanks > > Steve Timm > > From bryan.stillwell at hp.com Wed Mar 28 17:51:46 2007 From: bryan.stillwell at hp.com (Bryan Stillwell) Date: Wed, 28 Mar 2007 11:51:46 -0600 Subject: Kickstarting an RX4640 Itanium Message-ID: <460AAB32.4030500@hp.com> Matt, I work in HP's Open Source and Linux Organization and I've done multiple kickstart installs on Itanium hardware and everything looks right to me. It sounds like the cciss driver isn't being loaded during your kickstart install for some reason. Can you switch to the shell (Alt-F2) and run lsmod (or 'cat /proc/modules') and see if the cciss driver is loaded when you receive that message? You can also try adding the following line to the top of your kickstart script to make sure the cciss driver is loaded: device scsi cciss BTW, how did you manage to fit four internal drives in an rx4640? There's only room for two. Bryan From slarty.tj at gmail.com Wed Mar 28 18:12:49 2007 From: slarty.tj at gmail.com (Matt P) Date: Wed, 28 Mar 2007 13:12:49 -0500 Subject: Kickstarting an RX4640 Itanium In-Reply-To: <460AAB32.4030500@hp.com> References: <460AAB32.4030500@hp.com> Message-ID: <859a78260703281112j40f669ecp847d087d0c3af325@mail.gmail.com> Yes, you're correct, I've got Itatium on the brain, there are only 2 internal drives. I'm in the process of building out 70+ machine which are a mix of Itanium (RX4640) and x86 (DL class). The x86 machines have 4 HDDs, my mistake. I'm pretty sure the Module gets loaded. I didn't use lsmod but fdisk does see the disk and the partitions that HP SmartStart builds which includes efi.... I'll give that "device scsi cciss" a shot. the worst part about testing this is I've got to burn a new DVD each time I make a change to the config. Then sneaker net it over to the customer site, etc, etc... It's a royal pain. On 3/28/07, Bryan Stillwell wrote: > Matt, > > I work in HP's Open Source and Linux Organization and I've done multiple > kickstart installs on Itanium hardware and everything looks right to me. > It sounds like the cciss driver isn't being loaded during your > kickstart install for some reason. Can you switch to the shell (Alt-F2) > and run lsmod (or 'cat /proc/modules') and see if the cciss driver is > loaded when you receive that message? > > You can also try adding the following line to the top of your kickstart > script to make sure the cciss driver is loaded: > > device scsi cciss > > BTW, how did you manage to fit four internal drives in an rx4640? > There's only room for two. > > Bryan > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From bryan.stillwell at hp.com Wed Mar 28 19:00:58 2007 From: bryan.stillwell at hp.com (Bryan Stillwell) Date: Wed, 28 Mar 2007 13:00:58 -0600 Subject: Kickstarting an RX4640 Itanium In-Reply-To: <859a78260703281112j40f669ecp847d087d0c3af325@mail.gmail.com> References: <460AAB32.4030500@hp.com> <859a78260703281112j40f669ecp847d087d0c3af325@mail.gmail.com> Message-ID: <460ABB6A.5020107@hp.com> Matt, Here's a kickstart script that I just tested with a network kickstart and it worked fine on an rx2660. Bryan Matt P wrote: > Yes, you're correct, I've got Itatium on the brain, there are only 2 > internal drives. I'm in the process of building out 70+ machine which > are a mix of Itanium (RX4640) and x86 (DL class). The x86 machines > have 4 HDDs, my mistake. > > I'm pretty sure the Module gets loaded. I didn't use lsmod but fdisk > does see the disk and the partitions that HP SmartStart builds which > includes efi.... I'll give that "device scsi cciss" a shot. the worst > part about testing this is I've got to burn a new DVD each time I make > a change to the config. Then sneaker net it over to the customer site, > etc, etc... It's a royal pain. > > > On 3/28/07, Bryan Stillwell wrote: >> Matt, >> >> I work in HP's Open Source and Linux Organization and I've done multiple >> kickstart installs on Itanium hardware and everything looks right to me. >> It sounds like the cciss driver isn't being loaded during your >> kickstart install for some reason. Can you switch to the shell (Alt-F2) >> and run lsmod (or 'cat /proc/modules') and see if the cciss driver is >> loaded when you receive that message? >> >> You can also try adding the following line to the top of your kickstart >> script to make sure the cciss driver is loaded: >> >> device scsi cciss >> >> BTW, how did you manage to fit four internal drives in an rx4640? >> There's only room for two. >> >> Bryan -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: rhel4u3-cciss2.ks URL: From dan at half-asleep.com Wed Mar 28 20:36:18 2007 From: dan at half-asleep.com (Daniel Segall) Date: Wed, 28 Mar 2007 16:36:18 -0400 Subject: Identifying SCSI or IDE drives in %pre Message-ID: <515ba2a65fb25dfe1e2d204a85e6ab7a@mail.half-asleep.com> I deploy kickstarts to a number of different configurations. The servers have IDE, SAS, or SCSI drives. My partitioning is always the same, but if it's an IDE server, I do a software raid1. The SCSI & SAS servers have hardware raid1. Currently I manage multiple configs to make this happen. I'd like to start using list-harddrives in %pre, and generate the partition tables from that. It's really simple, if sd = yes, then partition normally. hd = yes then partition with raid config. I know this has been covered before, and I have searched and found some examples, but not exactly what I'm looking for. I am not a programmer, so if someone could help me with the logic here, that would be sweet. Thanks! -- -Dan From smah at vmware.com Wed Mar 28 20:56:29 2007 From: smah at vmware.com (Stephen Mah) Date: Wed, 28 Mar 2007 13:56:29 -0700 Subject: Identifying SCSI or IDE drives in %pre In-Reply-To: <515ba2a65fb25dfe1e2d204a85e6ab7a@mail.half-asleep.com> References: <515ba2a65fb25dfe1e2d204a85e6ab7a@mail.half-asleep.com> Message-ID: <460AD67D.4030706@vmware.com> http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/pdf/rhel-sag-en.pdf should be on page 20. There's an example script that does that. You will have to use the %include directive. Daniel Segall wrote: > I deploy kickstarts to a number of different configurations. The servers have IDE, SAS, or SCSI drives. My partitioning is always the same, but if it's an IDE server, I do a software raid1. The SCSI & SAS servers have hardware raid1. Currently I manage multiple configs to make this happen. I'd like to start using list-harddrives in %pre, and generate the partition tables from that. It's really simple, if sd = yes, then partition normally. hd = yes then partition with raid config. > > I know this has been covered before, and I have searched and found some examples, but not exactly what I'm looking for. I am not a programmer, so if someone could help me with the logic here, that would be sweet. > > Thanks! > > -- > -Dan > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From slarty.tj at gmail.com Wed Mar 28 21:02:36 2007 From: slarty.tj at gmail.com (Matt P) Date: Wed, 28 Mar 2007 16:02:36 -0500 Subject: Kickstarting an RX4640 Itanium In-Reply-To: <460ABB6A.5020107@hp.com> References: <460AAB32.4030500@hp.com> <859a78260703281112j40f669ecp847d087d0c3af325@mail.gmail.com> <460ABB6A.5020107@hp.com> Message-ID: <859a78260703281402qeab525fue1f234224c2bf925@mail.gmail.com> Worked like a Champ! Thanks Bryan. Looks like my problem was not having the "--size=" included in the /boot/efi line. I knew it had to be something fairly simply, since I wasn't able to find anyone on the net that was having much trouble at all Kickstarting Itanium hardware. Thanks again! On 3/28/07, Bryan Stillwell wrote: > Matt, > > Here's a kickstart script that I just tested with a network kickstart > and it worked fine on an rx2660. > > Bryan > > Matt P wrote: > > Yes, you're correct, I've got Itatium on the brain, there are only 2 > > internal drives. I'm in the process of building out 70+ machine which > > are a mix of Itanium (RX4640) and x86 (DL class). The x86 machines > > have 4 HDDs, my mistake. > > > > I'm pretty sure the Module gets loaded. I didn't use lsmod but fdisk > > does see the disk and the partitions that HP SmartStart builds which > > includes efi.... I'll give that "device scsi cciss" a shot. the worst > > part about testing this is I've got to burn a new DVD each time I make > > a change to the config. Then sneaker net it over to the customer site, > > etc, etc... It's a royal pain. > > > > > > On 3/28/07, Bryan Stillwell wrote: > >> Matt, > >> > >> I work in HP's Open Source and Linux Organization and I've done multiple > >> kickstart installs on Itanium hardware and everything looks right to me. > >> It sounds like the cciss driver isn't being loaded during your > >> kickstart install for some reason. Can you switch to the shell (Alt-F2) > >> and run lsmod (or 'cat /proc/modules') and see if the cciss driver is > >> loaded when you receive that message? > >> > >> You can also try adding the following line to the top of your kickstart > >> script to make sure the cciss driver is loaded: > >> > >> device scsi cciss > >> > >> BTW, how did you manage to fit four internal drives in an rx4640? > >> There's only room for two. > >> > >> Bryan > > install > nfs --server=test.test --dir=/dist/redhat/rhel4/as/U3/ia64 > lang en_US.UTF-8 > langsupport --default=en_US.UTF-8 en_US.UTF-8 > keyboard us > network --device eth0 --bootproto dhcp > rootpw --iscrypted XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX > firewall --disabled > selinux --disabled > authconfig --enableshadow --enablemd5 > timezone --utc America/Denver > bootloader --location=mbr --append="console=ttyS0" > clearpart --all --initlabel=gpt > part /boot/efi --fstype vfat --size=100 --ondisk=cciss/c1d0 > part swap --fstype swap --size=2048 --ondisk=cciss/c1d0 > part / --fstype ext3 --size=256 --grow --ondisk=cciss/c1d0 > > %packages > e2fsprogs > kernel > dosfstools > elilo > > %post > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From dan at half-asleep.com Wed Mar 28 21:12:53 2007 From: dan at half-asleep.com (Daniel Segall) Date: Wed, 28 Mar 2007 17:12:53 -0400 Subject: Identifying SCSI or IDE drives in %pre In-Reply-To: <460AD67D.4030706@vmware.com> References: <460AD67D.4030706@vmware.com> Message-ID: Right. This would be one of the examples I've seen many times. I just don't know how to change that script to look for a type of disk, rather than a number of disks. -Dan On Wed, 28 Mar 2007 13:56:29 -0700, Stephen Mah wrote: > http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/pdf/rhel-sag-en.pdf > should be on page 20. There's an example script that does that. > You will have to use the %include directive. > > Daniel Segall wrote: >> I deploy kickstarts to a number of different configurations. The servers > have IDE, SAS, or SCSI drives. My partitioning is always the same, but if > it's an IDE server, I do a software raid1. The SCSI & SAS servers have > hardware raid1. Currently I manage multiple configs to make this happen. > I'd like to start using list-harddrives in %pre, and generate the > partition tables from that. It's really simple, if sd = yes, then > partition normally. hd = yes then partition with raid config. >> >> I know this has been covered before, and I have searched and found some > examples, but not exactly what I'm looking for. I am not a programmer, so > if someone could help me with the logic here, that would be sweet. >> >> Thanks! >> >> -- >> -Dan >> From rrutherf at tripos.com Wed Mar 28 22:11:46 2007 From: rrutherf at tripos.com (Rodney Rutherford) Date: Wed, 28 Mar 2007 17:11:46 -0500 Subject: Identifying SCSI or IDE drives in %pre In-Reply-To: <515ba2a65fb25dfe1e2d204a85e6ab7a@mail.half-asleep.com> References: <515ba2a65fb25dfe1e2d204a85e6ab7a@mail.half-asleep.com> Message-ID: <460AE822.10904@tripos.com> Here is the simple test I do: # # Determine primary drive type (IDE or SATA/SCSI) # DRIVE='' parted -s /dev/hda > /dev/null 2>&1 if [ "$?" = 0 ] then DRIVE=hda else parted -s /dev/sda > /dev/null 2>&1 if [ "$?" = 0 ] then DRIVE=sda fi fi # # Then you can just configure your drives based on matching the type # if [ "$DRIVE" = "hda" ] then # echo your desired IDE partitioning > /tmp/part-include else # echo your desired SATA/SCSI partitioning > /tmp/part-include fi # # Then in the main body of the kickstart config before the # %pre, you would specify an include where the partitioning # normally would be: # %include /tmp/part-include Rodney Daniel Segall wrote: > I deploy kickstarts to a number of different configurations. The servers > have IDE, SAS, or SCSI drives. My partitioning is always the same, but > if it's an IDE server, I do a software raid1. The SCSI & SAS servers > have hardware raid1. Currently I manage multiple configs to make this > happen. I'd like to start using list-harddrives in %pre, and generate > the partition tables from that. It's really simple, if sd = yes, then > partition normally. hd = yes then partition with raid config. > > I know this has been covered before, and I have searched and found some > examples, but not exactly what I'm looking for. I am not a programmer, > so if someone could help me with the logic here, that would be sweet. > > Thanks! > > -- > -Dan > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From debian at herakles.homelinux.org Thu Mar 29 02:07:38 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Thu, 29 Mar 2007 10:07:38 +0800 Subject: Identifying SCSI or IDE drives in %pre In-Reply-To: References: <460AD67D.4030706@vmware.com> Message-ID: <460B1F6A.8020506@herakles.homelinux.org> Daniel Segall wrote: > Right. This would be one of the examples I've seen many times. I just don't know how to change that script to look for a type of disk, rather than a number of disks. Does /proc/partitions have what you want? (beware, I think SATA looks much like SCSI). Can you tell from the PCI bus? -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From Bob.Huisman at filterworks.com Thu Mar 29 09:12:18 2007 From: Bob.Huisman at filterworks.com (Bob Huisman) Date: Thu, 29 Mar 2007 11:12:18 +0200 Subject: Setting environment variable Message-ID: <9313697AF97C424595D8040FE226731E16962D@TINY.socs.lan> I am currently writing a kickstart script that must install a base system and some custom packages. Each of those packages has 2 'versions', a fully automated one and one that just deploys the software. Which version to deploy is determined by an environment variable read in the .spec file. I have been thinking about making a package which sets some of the environment variables, and have that installed too, but I understood that the order of installation of packages is not always the same... As the system has just been installed, there are no pre-set environment variables which can be used. I was wondering whether and how it is possible to set an environment variable before installation of any custom packages? I am using RHEL 5 with RHN Satellite 5 beta. Anyone who has a suggestion on how to deal with this problem? Thanks in advance! Bob From binand at gmail.com Thu Mar 29 13:02:53 2007 From: binand at gmail.com (Binand Sethumadhavan) Date: Thu, 29 Mar 2007 15:02:53 +0200 Subject: Setting environment variable In-Reply-To: <9313697AF97C424595D8040FE226731E16962D@TINY.socs.lan> References: <9313697AF97C424595D8040FE226731E16962D@TINY.socs.lan> Message-ID: On 29/03/07, Bob Huisman wrote: > I have been thinking about making a package which sets some of the > environment variables, and have that installed too, but I understood > that the order of installation of packages is not always the same... You can control the order of installation by making your environment-setting package as a dependency for the other packages. > As the system has just been installed, there are no pre-set environment > variables which can be used. I was wondering whether and how it is > possible to set an environment variable before installation of any > custom packages? I'd create a file in /etc/sysconfig for my custom environment (via a separate package or by just echo whatever > /etc/sysconfig/myenviron), and source that file in the %pre section of my specfile. Binand From jason at rampaginggeek.com Thu Mar 29 13:12:59 2007 From: jason at rampaginggeek.com (Jason Edgecombe) Date: Thu, 29 Mar 2007 09:12:59 -0400 Subject: Identifying SCSI or IDE drives in %pre In-Reply-To: <515ba2a65fb25dfe1e2d204a85e6ab7a@mail.half-asleep.com> References: <515ba2a65fb25dfe1e2d204a85e6ab7a@mail.half-asleep.com> Message-ID: <460BBB5B.5030305@rampaginggeek.com> Daniel Segall wrote: > I deploy kickstarts to a number of different configurations. The servers have IDE, SAS, or SCSI drives. My partitioning is always the same, but if it's an IDE server, I do a software raid1. The SCSI & SAS servers have hardware raid1. Currently I manage multiple configs to make this happen. I'd like to start using list-harddrives in %pre, and generate the partition tables from that. It's really simple, if sd = yes, then partition normally. hd = yes then partition with raid config. > > I know this has been covered before, and I have searched and found some examples, but not exactly what I'm looking for. I am not a programmer, so if someone could help me with the logic here, that would be sweet. > Put this in the main part of you kickstart config: %include /tmp/part-include My partitioning logic ignores removable drives like usb devices. Add more devices to the "for" line if you have a large number of drives. Here is my pre section: %pre #----- partitioning logic below-------------- # pick the first drive that is not removable and is over MINSIZE DIR="/sys/block" # minimum size of hard drive needed specified in GIGABYTES MINSIZE=6 ROOTDRIVE="" # /sys/block/*/size is in 512 byte chunks for DEV in sda sdb hda hdb; do if [ -d $DIR/$DEV ]; then REMOVABLE=`cat $DIR/$DEV/removable` if (( $REMOVABLE == 0 )); then echo $DEV SIZE=`cat $DIR/$DEV/size` GB=$(($SIZE/2**21)) if [ $GB -gt $MINSIZE ]; then echo "$(($SIZE/2**21))" if [ -z $ROOTDRIVE ]; then ROOTDRIVE=$DEV fi fi fi fi done echo "ROOTDRIVE=$ROOTDRIVE" cat << EOF >> /tmp/part-include bootloader --location=mbr --driveorder=$ROOTDRIVE clearpart --all --drives=$ROOTDRIVE --initlabel part /boot --fstype ext3 --size=300 --ondisk=$ROOTDRIVE part pv.4 --size=0 --grow --ondisk=$ROOTDRIVE volgroup VolGroup00 --pesize=32768 pv.4 logvol swap --fstype swap --name=LogVol01 --vgname=VolGroup00 --size=1024 --grow --maxsize=2048 logvol / --fstype ext3 --name=root --vgname=VolGroup00 --size=4096 --grow --maxsize=20480 logvol /var/log --fstype ext3 --name=var-log --vgname=VolGroup00 --size=512 logvol /tmp --fstype ext3 --name=tmp --vgname=VolGroup00 --size=1024 --maxsize=4096 logvol /usr/vice/cache --fstype ext3 --name=usr-vice-cache --vgname=VolGroup00 --size=1024 logvol /home --fstype ext3 --name=home --vgname=VolGroup00 --size=512 --grow EOF Sincerely, Jason Edgecombe From debian at herakles.homelinux.org Thu Mar 29 22:20:06 2007 From: debian at herakles.homelinux.org (John Summerfield) Date: Fri, 30 Mar 2007 06:20:06 +0800 Subject: Setting environment variable In-Reply-To: <9313697AF97C424595D8040FE226731E16962D@TINY.socs.lan> References: <9313697AF97C424595D8040FE226731E16962D@TINY.socs.lan> Message-ID: <460C3B96.6020108@herakles.homelinux.org> Bob Huisman wrote: > I am currently writing a kickstart script that must install a base > system and some custom packages. Each of those packages has 2 > 'versions', a fully automated one and one that just deploys the > software. Which version to deploy is determined by an environment > variable read in the .spec file. > > I have been thinking about making a package which sets some of the > environment variables, and have that installed too, but I understood > that the order of installation of packages is not always the same... > > As the system has just been installed, there are no pre-set environment > variables which can be used. I was wondering whether and how it is > possible to set an environment variable before installation of any > custom packages? > > I am using RHEL 5 with RHN Satellite 5 beta. Anyone who has a suggestion > on how to deal with this problem? > > Thanks in advance! One can set environment variables from the commandline when booting the kernel. Any "key=val" pairs appear in the environment. -- Cheers John -- spambait 1aaaaaaa at coco.merseine.nu Z1aaaaaaa at coco.merseine.nu Please do not reply off-list From Bob.Huisman at filterworks.com Fri Mar 30 07:39:33 2007 From: Bob.Huisman at filterworks.com (Bob Huisman) Date: Fri, 30 Mar 2007 09:39:33 +0200 Subject: Setting environment variable In-Reply-To: Message-ID: <9313697AF97C424595D8040FE226731E169675@TINY.socs.lan> -----Original Message----- From: Binand Sethumadhavan [mailto:binand at gmail.com] Sent: donderdag 29 maart 2007 15:03 To: Discussion list about Kickstart Subject: Re: Setting environment variable On 29/03/07, Bob Huisman wrote: >> I have been thinking about making a package which sets some of the >> environment variables, and have that installed too, but I understood >> that the order of installation of packages is not always the same... > You can control the order of installation by making your > environment-setting package as a dependency for the other packages. The problem with this is that the package can also be installed separately, where the extra commands depending on that environment variable should not be displayed. >> As the system has just been installed, there are no pre-set environment >> variables which can be used. I was wondering whether and how it is >> possible to set an environment variable before installation of any >> custom packages? > I'd create a file in /etc/sysconfig for my custom environment (via a > separate package or by just echo whatever > /etc/sysconfig/myenviron), > and source that file in the %pre section of my specfile. That would be an option, but would require me to build separate packages for the kickstart installation, which is not the idea of installing all 'versions' of the software with one package. I'm starting to think that would be the only option though...thanks for the input :) Bob From amm at lanl.gov Fri Mar 30 19:29:40 2007 From: amm at lanl.gov (Aaron M Morrison) Date: Fri, 30 Mar 2007 13:29:40 -0600 Subject: RHEL5 Client package list question Message-ID: <1175282980.23468.8.camel@itchy.lanl.gov> Howdy, has anyone noticed any features with RHEL5 + anaconda and packages selected vs packages actually installed. I installed a system via CD with manual package selection, and attempted to reuse the %package section from the resulting anaconda-ks.cfg file on a system that uses kickstart/nfs/pxeboot and I get drastically different sets of RPMS installed. I know that since RHEL5 is yum based, perhaps there more going on behind the scenes...is there an option that Kickstart requires that I could be missing? Package line from original anaconda-ks.cfg and kickstart file for 2nd system %packages @admin-tools @base @core @development-libs @development-tools @editors @gnome-software-development @system-tools @text-internet @x-software-development kernel-devel Here's a diff of sorted rpm -qa > /tmp/foo output from both systems (first 25 lines) bash-3.00$ diff -u cd_install.rpms pxe_ks_install.rpms|head -n 25 --- cd_install.rpms 2007-03-30 13:05:28.000000000 -0600 +++ pxe_ks_install.rpms 2007-03-30 13:05:59.000000000 -0600 @@ -1,24 +1,14 @@ -a2ps-4.13b-57.1.el5 acl-2.2.39-1.1 acpid-1.0.4-5 -adjtimex-1.20-2.1 -agg-2.4-2.1 -aide-0.12-7 -alacarte-0.10.0-1.fc6 alsa-lib-1.0.12-3.el5 alsa-lib-devel-1.0.12-3.el5 alsa-utils-1.0.12-3.fc6 -amanda-2.5.0p2-4 -amanda-client-2.5.0p2-4 amtu-1.0.4-4 am-utils-6.1.5-4.1.el5 anacron-2.3-45.el5 -antlr-2.7.6-4jpp.2 apmd-3.2.2-5 apr-1.2.7-11 apr-util-1.2.7-6 -arptables_jf-0.0.8-8 -arpwatch-2.1a13-15.8.1 aspell-0.60.3-7.1 Thanks in advance amm