Bug 728953 - grub will not boot from non-primary virtio disk
Summary: grub will not boot from non-primary virtio disk
Alias: None
Product: Red Hat Enterprise Linux 4
Classification: Red Hat
Component: grub
Version: 4.8
Hardware: Unspecified
OS: Unspecified
Target Milestone: rc
: ---
Assignee: Peter Jones
QA Contact: Release Test Team
Depends On: 728947
Blocks: 728951
TreeView+ depends on / blocked
Reported: 2011-08-08 13:19 UTC by Matthew Booth
Modified: 2012-06-20 16:10 UTC (History)
0 users

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of: 728947
Last Closed: 2012-06-20 16:10:04 UTC
Target Upstream Version:

Attachments (Terms of Use)

Description Matthew Booth 2011-08-08 13:19:19 UTC
+++ This bug was initially created as a clone of Bug #728947 +++

Description of problem:
A guest has 2 virtio disks: vda and vdb. /boot resides on /dev/vdb1. Grub is installed on /dev/vda, which is also where the guest is configured to boot from.

Firstly, a fresh installation in the above configuration will cause the boot to fail with 'Hard Disk Error' (screenshot rhel6-grub-hd_error)[1].

If you attempt to fix the problem by manually running 'grub-install /dev/vda' from the rescue environment you  get Error 21 trying to load stage 1.5 (screenshot rhel6-error_21).

The above configuration works fine if the disks are IDE. It also works fine if the drives are swapped so that /dev/vdb1 becomes /dev/vda1 and grub is manually installed on the new /dev/vda. The problem seems to be specific to attempting to read /boot from a drive other than the first one.

I have replicated this problem on RHEL 4.8, RHEL 5.4 and RHEL 6.0, all of which support VirtIO. I have also replicated the problem on a RHEL 5 KVM host, which uses Bochs BIOS, and an F14 host, which uses SeaBIOS. The behaviour is the same in all cases, although the error messages from RHEL 4 and RHEL 5 grub are slightly more verbose.

[1] I'd expect the behaviour to be identical immediately after installation to behaviour after running 'grub-install /dev/vda', so could this indicate a secondary problem in anaconda?

Comment 1 Jiri Pallich 2012-06-20 16:10:04 UTC
Thank you for submitting this issue for consideration in Red Hat Enterprise Linux. The release for which you requested us to review is now End of Life. 
Please See https://access.redhat.com/support/policy/updates/errata/

If you would like Red Hat to re-consider your feature request for an active release, please re-open the request via appropriate support channels and provide additional supporting details about the importance of this issue.

Note You need to log in before you can comment on or make changes to this bug.