I am trying to re-install a RHEL 5.3 server that previously had SAN disks assigned to it (and still has) so that I can leave the SAN disks presented and re-install RHEL 5.x without it crashing. My understanding from the manual (
http://www.redhat.com/docs/manuals/e...2-options.html) was that you could use kick the following to kick off an install:
linux ks=blahblah
ignoredisk --drives=sdb,sdc
With the above, I was expecting the SAN disks wouldn't cause the installer to crash. I know I can have the SAN disks unmapped and RHEL 5.x installs properly (tested it previously) but would like to avoid this as I have to rely on other people for that to happen (not to mention have them represent the disks after the installation is complete).
However, an old bugzilla note (
https://bugzilla.redhat.com/show_bug.cgi?id=366341) has someone suggesting that “ignoredisk really only refers to that certain disks should not be used when creating partitions and clearing devices."
So which one is it?
I see that newer releases of Fedora include a useful option that would be ideal where you specify the local disk you want used for the installation and the SAN disks are ignored, but I don’t have that luxury as I have to install with RHEL 5 (from
http://fedoraproject.org/wiki/Anacon...art#ignoredisk) :
ignoredisk --only-use=[disk1,disk2,...]