-
Bug
-
Resolution: Unresolved
-
Major
-
rhel-9.5
-
qemu-kvm-9.1.0-15.el9
-
No
-
Moderate
-
rhel-sst-virtualization-storage
-
ssg_virtualization
-
26
-
None
-
False
-
-
None
-
None
-
Pass
-
None
-
None
What were you trying to do that didn't work?
Attempting to migrate a VM paused for a second time results into an assertion failure:
(qemu) qemu-kvm: ../block.c:6979: int bdrv_inactivate_recurse(BlockDriverState *): Assertion `!(bs->open_flags & BDRV_O_INACTIVE)' failed.
and subsequent qemu crash.
The issue is addressed by the series ending at upstream commit:
commit 8597af76153a87068b675d8099063c3ad8695773
Author: Peter Xu <peterx@redhat.com>
Date: Fri Dec 6 18:08:38 2024 -0500
migration/block: Rewrite disk activation
This patch proposes a flag to maintain disk activation status globally. It
mostly rewrites disk activation mgmt for QEMU, including COLO and QMP
command xen_save_devices_state.
The crash was originally observed on OpenStack. See the original issue filed against libvirt for further information: https://issues.redhat.com/browse/RHEL-48801
The libvirt issue will be marked as test-only to facilitate testing on the libvirt level.
- blocks
-
RHEL-48801 Live migration of a paused domain twice causing qemu crash
-
- Integration
-
- links to
-
RHBA-2024:139949 qemu-kvm bug fix and enhancement update