Bug 1208527 - Previewing live snapshot with ram volume fails silently to restore VM's ram parameters
Summary: Previewing live snapshot with ram volume fails silently to restore VM's ram p...
Keywords:
Status: CLOSED DUPLICATE of bug 1145658
Alias: None
Product: Red Hat Enterprise Virtualization Manager
Classification: Red Hat
Component: ovirt-engine
Version: 3.5.1
Hardware: Unspecified
OS: Unspecified
unspecified
high
Target Milestone: ovirt-3.6.3
: 3.6.0
Assignee: Idan Shaby
QA Contact:
URL:
Whiteboard: storage
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2015-04-02 12:52 UTC by Ori Gofen
Modified: 2016-05-26 01:50 UTC (History)
9 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2015-04-04 15:15:18 UTC
oVirt Team: Storage
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)
logs (2.08 MB, application/x-gzip)
2015-04-02 12:52 UTC, Ori Gofen
no flags Details

Description Ori Gofen 2015-04-02 12:52:07 UTC
Created attachment 1010149 [details]
logs

Description of problem:

When creating a live snapshot with ram volume enabled, it's ram volume is created randomly on a storage domain on dc, not necessarily the same domain which contain's VM's storage see BZ #1186230 for more input on the matter.
oVirt is not monitoring those ram volumes thus, detaching a domain which contains ram volumes is allowed without a proper warning, furthermore, when attempting to preview this snapshot, VM's ram is not reconstructed obviously.
vdsm does send a traceback to log:
"Traceback (most recent call last):
  File "/usr/share/vdsm/API.py", line 185, in create
    fname = self._cif.prepareVolumePath(paramFilespec)
  File "/usr/share/vdsm/clientIF.py", line 277, in prepareVolumePath
    raise vm.VolumeError(drive)
VolumeError: Bad volume specification {'device': 'disk', 'imageID': u'1d53dc1e-b867-4590-a0f2-50f597043bd9', 'domainID': u'fb0d07bd-8953-41dd-8600-b6e367231ebb', 'volumeID': u'd5385fa5-0c25-4e60-9a1d-4d5551f6306e', 'poolID': u'00000002-0002-0002-0002-000000000018'}"

but ui wise this failed operation is not mentioned at all, and our client can never know from looking at the web-admin it's VM is running with "fresh" cache.

Version-Release number of selected component (if applicable):
vt14.1


How reproducible:
100%

Steps to Reproduce:
-- Have a DC with 2 domains or more
1.create VM+os (one disk not on master domain)
2.run an application which update cache frequently, or hot plug virtio disk and mount it to some dir
3.create live snapshot with ram enabled
4.locate the domain which contains the ram volume
5.detach this domain from dc
6. preview the live snapshot
7.the application have failed | the mount point is gone

Actual results:
reviewing live snapshot with ram volume fails silently to restore VM's ram parameters

Expected results:
* when detaching a domain that contains ram snapshot a pop up warning window   should be displayed | or operation should be CDA

* when vdsm fails to reconstruct VM's ram upon live snapshot preview, an ERROR msg should be displayed on web-admin status bar.

Additional info:

Comment 1 Allon Mureinik 2015-04-04 15:15:18 UTC

*** This bug has been marked as a duplicate of bug 1145658 ***


Note You need to log in before you can comment on or make changes to this bug.