I've experienced a similar problem with HP boxes, but only when mixing some fibre channel devices (labeled as sd[a-z]) and some cciss devices. (/dev/cciss/c0d0). It seems that anaconda favors putting grub on the /dev/sda fiber device per default, which is a bad thing since it's an external storage cabinet and we'd rather boot off the local drives on the cciss controller. I had to go into "Advanced Boot Loader Options" and specify the drive ordering. There are kickstart flags for this as well. James S. Martin, RHCE Contractor Administrative Office of the United States Courts Washington, DC (202) 502-2394 anaconda-devel-list-bounces@xxxxxxxxxx wrote on 09/05/2006 05:36:17 AM: > Hi, > > I have a particular problem, only verified with RHEL4AS U4, where Anaconda > happily installs and reboots but there does not seem a bootloader > installed (or it is installed at the wrong location). > > This happens on a HP Proliant DL580 G2 with the cciss driver. If I install > the bootloader with grub-install myself (using a rescue cd and > mount/chroot) then the grub device.map lists /dev/cciss/c0d0 as hd4 and > sda as hd0. > > Not sure if that is the problem, but manually overwriting the device.map > and making /dev/cciss/c0d0 hd0 works as expected. > > Now, instead of asking if this is indeed the problem, I was wondering how > someone would go and debug this. Is there some way to know exactly what > anaconda has been doing at the end of the installation, without modifying > anaconda, without even rebooting and kickstarting again ? > > I would expect something like a complete 'set -x' and command output of > *everything* it did. Or an strace like output. Something that can match > behaviour with expectation. Or something you can grep for things, during > or after the installation without loosing time or iterating over things. > > The ultimate goal is to provide sufficient information (and maybe even a > patch) to put in a bugzilla report, which I do not have right now. > > Thanks in avdance for any help pursuiing this :) > -- dag wieers, dag@xxxxxxxxxx, http://dag.wieers.com/ -- > [all I want is a warm bed and a kind word and unlimited power] > > _______________________________________________ > Anaconda-devel-list mailing list > Anaconda-devel-list@xxxxxxxxxx > https://www.redhat.com/mailman/listinfo/anaconda-devel-list
