Red Hat Bugzilla – Bug 1344479
Live Merge completes but the 'job' table entry in the database is not marked as finished
Last modified: 2017-03-21 17:19:07 EDT
Description of problem: Since upgrading to 3.6.6, we are now seeing instances of Live Merges completing, but the tasks in the Tasks pane in the RHEV Admin Portal are not being marked as having finished (an hourglass is displayed for the Finalizing step). The 'RemoveDiskSnapshots' entry in the job table in the database is not getting changed to 'FINISHED'. Thus it appears to the end-user that the snapshot deletion has not completed. Version-Release number of selected component (if applicable): RHEV 3.6.6 How reproducible: Not, but appears to be a repeatable problem. Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
Tentatively targeting to 3.6.8 until we have an RCA.
Hi Gordon, the operation that was performed was live merge of snapshot? or snapshot's disk live deletion?
It was "snapshot's disk live deletion", i.e. a single disk live snapshot deletion as opposed to an entire live snapshot deletion. Please refer to comment #6, which provides the steps on how to reproduce this. When I opened the bug I didn't know exactly what caused it, hence the reproduction steps were added later. Unfortunately, I'm going to be out of the office for an extended period of time, so I hope this helps to clarify the issue. Regards, GFW.
*** Bug 1347243 has been marked as a duplicate of this bug. ***
Detailed logs can be found as attachment to bug 1347243
Hi, I think the solution does not cover the second bug as described in BZ1347243. The single disk live merge command starts its action with logging the wrong message: "Disks from snapshot deletion has been completed" See engine log of BZ1347243: ... 2016-06-16 13:21:13,630 INFO Running command: RemoveDiskSnapshotsCommand 2016-06-16 13:21:13,738 INFO deletion has been completed <- WRONG! 2016-06-16 13:21:16,444 INFO Waiting on Live Merge child commands to complete ... 2016-06-16 13:22:42,169 INGO Merge command has completed ...
That issue reported in BZ 1344516 and already fixed.
Good to see. Thanks for that info.
Ala - although the fix is obvious, many customers have faced this issue. Can you please provide some doctext highlighting it?
This bug does not reproduce on the 4.0 branch (the fix is for 3.6 only), but it was cloned to follow the z-stream process. Setting to ON_QA so QA can keep me honest here.
Tested with the following code: --------------------------------------- vdsm-4.18.4-2.el7ev.x86_64 rhevm-4.0.2-0.2.rc1.el7ev.noarch Tested using the following scenario: --------------------------------------- 1. Create a VM with one disk. 2. Create a snapshot. 3. Start VM. 4. Got to main 'Storage' tab. 5. Select SD in which disk resides. 6. Go to 'Disk Snapshots' tab. 7. Select snapshot. 8. Delete snapshot. The engine log reports that the Live Merge completed successfully The events log in the Web Admin reports that the Live Merge completed successfully The tasks window reports all stages of the Live Merge as having completed successfully Moving to VERIFIED!
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHEA-2016-1743.html