Red Hat Bugzilla – Bug 967602
vdsm: we fail to destroy vm that was paused on EIO with "Drive is not a vdsm image"
Last modified: 2016-02-10 13:41:30 EST
Created attachment 753644 [details]
Description of problem:
I live storage migrated a vm disk and blocked connectivity to the storage from the hsm only (host that the vm was running on).
when the vm paused I tried to destroy it and failed in vdsm with Drive is not a vdsm image followed by:
m volExtensionChunk:1024 watermarkLimit:536870912
Traceback (most recent call last):
File "/usr/share/vdsm/clientIF.py", line 334, in teardownVolumePath
res = self.irs.teardownImage(drive['domainID'],
File "/usr/share/vdsm/libvirtvm.py", line 1071, in __getitem__
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. in iscsi storage with two hosts and two domains create a vm from template and run it on the hsm
2. LSM the vm disk
3. block connectivity to the storage in hsm only when engine logs "SyncImageGroupDataVDSCommand"
4. when the vm pauses try to destroy it (power off)
we fail to destroy the vm with no alert in UI.
we should be able to destroy the vm
Additional info: logs
Fede, is this solved already?
*** This bug has been marked as a duplicate of bug 980054 ***