Red Hat Bugzilla – Bug 994398
qcow2 backing file corrupt?
Last modified: 2016-04-26 11:58:03 EDT
Description of problem:
we have 7 computes running RDO 3.0.
some instances crashed after running several hours. we reboot the instance or rebuild them, can't recovery.
we checked the backing file:
# qemu-img info /home/nova//instances/7dfce16b-d494-4842-abd8-e6bd56567160/disk
file format: qcow2
virtual size: 50G (53687091200 bytes)
disk size: 600M
backing file: /home/nova//instances/_base/74b1c2183d7d7022cc0426781ed35e4b74c83fb4
the md5 checksum is wrong.
# qemu-img check /home/nova//instances/7dfce16b-d494-4842-abd8-e6bd56567160/disk
No errors were found on the image.
yesterday, we find 3 of 7 computes(0.5, 0.7, 0.8) have wrong backing file, and reinstall the backing file, then it worked again.
today, some instances crashed again, and we recheck the md5sum of backing file, 3 server are wrong(0.3， 0.5， 0.7).
please attach SOS report so we can go over all needed logs.
Hi, since this is an issue with using RDO on top of qemu-kvm-rhev, I am going to move this issue to the RHEL qemu-kvm component so the kvm team can take a look.
Peng Yong: we need some extra information to try to debug your problem.
Please attach to this bug report a SOS report and/or at least the version of qemu-kvm, kernel and libvirt packages:
($ rpm -qa | grep qemu)
($ rpm -qa | grep libvirt)
($ rpm -qa | grep kernel)
($ uname -a)
We also need the command line used to invoke qemu. This can be obtained from the output of ps:
($ ps auxww | grep qemu)
Finally, please keep in mind that we're not able to guarantee the timeliness or suitability of a resolution for issues entered here because this is not an official mechanism for requesting support.
If this issue is critical or in any way time sensitive to you or your organization, please raise a ticket through your regular Red Hat support channels to make certain it receives the proper attention and prioritization to assure a timely resolution.
For information on how to contact the Red Hat production support team, please
Thank you for your report.
Peng Yong: given the lack of details, I'm closing this bug for now. If you manage to provide the information requested in comment #5, please reopen it. Thanks.