http://bugzilla.suse.com/show_bug.cgi?id=906248 Charles Arnold <carnold@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |carnold@suse.com --- Comment #5 from Charles Arnold <carnold@suse.com> --- (In reply to Juergen Gross from comment #4)

Sorry, I don't think the component of this bug should be Xen. It's just a matter of the installation medium to contain a kernel and initrd bootable as Xen pv-domain. Putting those on the DVD like for the x86_64 one should be okay, just remove the xen_server pattern, as this is available for 64 bits only.

You are correct. However, they will not remaster the media once it ships so the 32bit media will just not have a kernel/initrd for Xen in the boot/i386 directory. As a work around you may be able to create a network install source and place the Xen kernel/initrd in the boot/i386 directory (vmlinuz-xen and initrd-xen). You would have to generate the initrd. Vm-install was designed to try and extract the kernel/initrd from the RPMs found on the media if the kernel and initrd files are not in the boot directory. This code path hasn't been needed in a long time and has some problems with changes in os13.x. I fixed some of these problems and pushed the changes out to the Virtualization/vm-install repo. The 32bit os13.2 VM now boots for me but hangs coming up. I think something is wrong or missing from the initrd that is generated by vm-install. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.suse.com/show_bug.cgi?id=906248 Charles Arnold <carnold@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Assignee|jdouglas@suse.com |snwint@suse.com QA Contact|qa-bugs@suse.de |jdouglas@suse.com --- Comment #7 from Charles Arnold <carnold@suse.com> --- (In reply to Juergen Gross from comment #6)

Charles, thanks for the information how to manage to install the system. Meanwhile I've found another solution for my problem, as my main requirement is a rather new 32-bit pv-domain capable of running an upstream kernel: I'm using a debian vm now.

I hope the required changes for openSUSE will make it for the factory build.

If 32bit versions of openSUSE releases are planned for the future then this will need to be fixed in Factory. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.suse.com/show_bug.cgi?id=906248 Stephan Kulow <coolo@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Flags|needinfo?(coolo@suse.com) | --- Comment #9 from Stephan Kulow <coolo@suse.com> --- this is between you and Charles. If the agreements is that xen is fine on 32bit, I have no objections :) -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.suse.com/show_bug.cgi?id=906248 --- Comment #11 from Jürgen Groß <jgross@suse.com> --- (In reply to Stephan Kulow from comment #9)

this is between you and Charles. If the agreements is that xen is fine on 32bit, I have no objections :)

The question is not whether it's fine to run xen on 32bit, but to run 32bit systems on xen. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.suse.com/show_bug.cgi?id=906248 --- Comment #13 from Steffen Winterfeldt <snwint@suse.com> --- That may be true but there's only one define in the spec file for xen. I'll change it a bit. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.suse.com/show_bug.cgi?id=906248 Jason Douglas <jdouglas@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- QA Contact|jdouglas@suse.com |qa-bugs@suse.de -- You are receiving this mail because: You are on the CC list for the bug.