Created attachment 1341223 [details] engine log Description of problem: Failed to delete snapshot, error in engine log is execution failed: VDSGenericException: VDSErrorException: Failed to GetVolumeInfoVDS, error = Volume does not exist: (u'bbf536f2-5bb6-44ed-9de2-5c77e67c3af2',), code = 201 Version-Release number of selected component (if applicable): ovirt-engine-4.2.0-0.0.master.20171019181349.gite12eda4.el7.centos.noarch How reproducible: always Steps to Reproduce: 1. have VM with OS installed, not running 2. go to VM detail -> Snapshots, create snapshot 3. run the VM 4. when VM is up, create other snapshot 5. when snapshot is successfully created, delete the first one snapshot Actual results: delete fails Expected results: delete is successful Additional info: Second delete is successful. I tried it on two different engines with the same result.
Created attachment 1341224 [details] vdsm log
Ala, can you take a look at this, please? It looks eerily familiar.
Lucie, can you specify the VDSM version you used too please?
Hi Lucie, The error seen in Vdsm log is expected and it corresponds to this in seen in engine log: 2017-10-20 13:47:23,499+02 ERROR [org.ovirt.engine.core.vdsbroker.vdsbroker.GetVolumeInfoVDSCommand] (EE-ManagedThreadFactory-engineScheduled-Thread-100) [cbe5137e-44ef-4fd9-998f-734b04b9cb0c] Command 'GetVolumeInfoVDSCommand(HostName = vmhost04, GetVolumeInfoVDSCommandParameters:{expectedEngineErrors='[VolumeDoesNotExist]', hostId='f80f228c-135a-4376-ab6c-dd506db45abc', storagePoolId='59db601a-03c8-00b2-00ec-000000000216', storageDomainId='351d0071-0b19-4ca1-b2aa-773adc504ccc', imageGroupId='e206235c-3363-45f0-9917-9b702ac7c1db', imageId='b8333b7b-19c9-479a-9765-72527b154f07'})' execution failed: VDSGenericException: VDSErrorException: Failed to GetVolumeInfoVDS, error = Volume does not exist: (u'b8333b7b-19c9-479a-9765-72527b154f07',), code = 201 This is actually a (good) indication that merge completed, because we perform this volume check after merge completed and base volume deleted. In addition, in the engine log I see the steps of cold merge (PrepareMerge and FinalizeMerge), can you please double check whether this is cold or live merge? Lastly, based on the engine logs, the merge did succeed (referring to merging snapshot test deletion of vm2-01): 2017-10-20 13:47:27,059+02 INFO [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (EE-ManagedThreadFactory-engineScheduled-Thread-62) [cbe5137e-44ef-4fd9-998f-734b04b9cb0c] EVENT_ID: USER_REMOVE_SNAPSHOT_FINISHED_SUCCESS(356), Snapshot 'test' deletion for VM 'vm2-01' has been completed.
The logs seem partial. If possible, please reproduce the issue and upload the logs.
Tested on two hosts with vdsm-4.20.3-160.git5446b34.el7.centos.x86_64 and vdsm-4.20.3-209.git65452bc.el7.centos.x86_64 It happened on two different data storages, two VMs running on different hosts. I delete the snapshot while VM is running (in that run I created other snapshot, see steps to reproduce).
Can you please confirm that you are using file based storage?
Yes, it is data nfs storage.
This is the same as BZ 1505150. A fix is already on the way. closing as duplicate. *** This bug has been marked as a duplicate of bug 1505150 ***