Bug 1220115 - Traceback on VM destroy in mom log
Summary: Traceback on VM destroy in mom log
Keywords:
Status: CLOSED WONTFIX
Alias: None
Product: Red Hat Enterprise Virtualization Manager
Classification: Red Hat
Component: mom
Version: 3.5.0
Hardware: All
OS: All
medium
medium
Target Milestone: ---
: 3.5.4
Assignee: Martin Sivák
QA Contact: Shira Maximov
URL:
Whiteboard: sla
Depends On: 1170512
Blocks:
TreeView+ depends on / blocked
 
Reported: 2015-05-10 09:12 UTC by rhev-integ
Modified: 2022-05-16 06:38 UTC (History)
11 users (show)

Fixed In Version: mom-0.4.1-5.el6ev
Doc Type: Bug Fix
Doc Text:
Clone Of: 1170512
Environment:
Last Closed: 2015-07-05 07:44:02 UTC
oVirt Team: SLA
Target Upstream Version:
Embargoed:

Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Issue Tracker RHV-46026 0 None None None 2022-05-16 06:38:23 UTC
oVirt gerrit 35881 0 master ABANDONED don't print stacktrace when the VM is powered down while running collectors Never
oVirt gerrit 35883 0 master MERGED print more user friendly error when VM is powered down while running collectors Never

Comment 2 Shira Maximov 2015-05-28 08:19:43 UTC 
The bug wasn't fixed, 
steps to repurduce the bug : 
1. create 20 vm's from pool 
2. select all the vms and start them and the shutdown the vm very quickly.
3. look at mom log.

you can still see the error :
2015-05-28 11:13:02,600 - mom.vdsmInterface - ERROR - Virtual machine does not exist
2015-05-28 11:13:02,600 - mom.vdsmInterface - ERROR - Traceback (most recent call last):
  File "/usr/lib/python2.6/site-packages/mom/HypervisorInterfaces/vdsmInterface.py", line 171, in getVmCpuTuneInfo
    self._check_status(response)
  File "/usr/lib/python2.6/site-packages/mom/HypervisorInterfaces/vdsmInterface.py", line 46, in _check_status
    raise vdsmException(response, self.logger)
vdsmException


tried to verify on :
MOM version : mom-0.4.1-5.el6ev.noarch
RHEVM Version: 3.5.3-0.2.el6ev
Comment 3 Sandro Bonazzola 2015-05-28 12:45:54 UTC 
> tried to verify on :
> MOM version : mom-0.4.1-5.el6ev.noarch
> RHEVM Version: 3.5.3-0.2.el6ev

Martin, shouldn't be mom at 0.4.4[1]?

[1] http://koji.fedoraproject.org/koji/buildinfo?buildID=637140
Comment 4 Martin Sivák 2015-05-28 12:47:51 UTC 
Nope, RHEV uses 0.4.1 as the base version and fixes (only handful of them) are added as patches.
Comment 5 Martin Sivák 2015-05-28 12:52:20 UTC 
But the log is still supposed to contain the traceback. The patch just adds the explanation message.

Commit message:

The stacktrace will still get logged, but there will be human readable
reason why it happened. Please note that this situation is rare so we shouldn't
have problems with log floods
Comment 6 Shira Maximov 2015-05-28 15:13:42 UTC 
i think that the fix should be change because it was not implemented correctly
as jmoskovc wrote: 
The proper fix here is to handle the exception in nicer way and just print a warning that the VM dissapeared while mom was processing it.

in the fix, the message is not more clear, it's still writen that "the vm does not exists" instead of "the VM dissapeared while mom was processing it"
also, it should be a warning and not an error.
Comment 7 Roy Golan 2015-06-04 08:20:53 UTC 
Martin can we refine the message be an info so users won't see errors. after all this is a valid use case - A VM can go down during a mom cycle.
Comment 8 Roy Golan 2015-07-05 07:44:02 UTC 
we're leaving the stack trace as written in comment 5. we want to know what's the underlying error.
Note You need to log in before you can comment on or make changes to this bug.