Red Hat Bugzilla – Bug 960988
engine: we report failure and then success for creation of the same live snapshot
Last modified: 2016-02-10 15:04:56 EST
Created attachment 745241 [details]
Description of problem:
2013-May-08, 15:27 Snapshot 'Auto-generated for Live Storage Migration' creation for VM 'grrr-03' has been completed.
2013-May-08, 15:27 Failed to create live snapshot 'Auto-generated for Live Storage Migration' for VM 'grrr-03'. VM restart is recommended.
root@gold-vdsd ~]# lvs |grep 1a3e182b-cf6d-4206-aec1-847b348a60e1
1a3e182b-cf6d-4206-aec1-847b348a60e1 d326916a-89b6-41c8-9c14-4b9cf8a1c979 -wi------ 1.00g
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. run a vm on the hsm host live migrate a disk
2. stop vdsm service on hsm only
we report the snapshot as failed and as completed.
we should not report a failure unless it actually failed.
Additional info: logs
same issue happens when we block src domain during DeleteImageGroup is logged in the engine log.
the mapping has changed (meaning that the LSM actually succeeded) and engine reports the task first as succeeded and than as failed:
User admin@internal have failed to move disk NFS-RHEL6_iSCSI_Disk1 to domain tiger-01.
User admin@internal finished moving disk NFS-RHEL6_iSCSI_Disk1 to domain tiger-01.
still planned for 3.4?
The issue should already be resolved by commit I7c69f663836e74691c968f4a15c5f3012479a8b0
which addressed bug 1018876.
verified using is29
Closing - RHEV 3.3 Released