Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 874120

Summary: While installing VM following error is displyed File "/usr/share/vdsm/storage/task.py", line 861, in _run
Product: [Retired] oVirt Reporter: Valentin Bajrami <valentin.bajrami>
Component: vdsmAssignee: Dan Kenigsberg <danken>
Status: CLOSED INSUFFICIENT_DATA QA Contact: Haim <hateya>
Severity: medium Docs Contact:
Priority: unspecified    
Version: 3.1 RCCC: abaron, acathrow, amureini, bazulay, dneary, dyasny, iheim, mgoldboi, tnisan, yeylon, ykaul
Target Milestone: ---   
Target Release: 3.3.4   
Hardware: x86_64   
OS: Linux   
Whiteboard: storage
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2013-06-05 08:22:34 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: Storage RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
vdsm.log
none
Full vdsm.log
none
Full engine.log none

Description Valentin Bajrami 2012-11-07 14:54:51 UTC
Description of problem:

When provisioning a Virtual Machine the following error from vdsm.log is shown


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


How reproducible:

Provision a new virtual machine

Steps to Reproduce:
1. Using web interface
2. [new server]
3. Half the provision process the NFS storage is unexpectedly detached/unplugged
  
Actual results:
See attachment.


Expected results:


Additional info:

Comment 1 Valentin Bajrami 2012-11-07 14:56:13 UTC
Created attachment 640114 [details]
vdsm.log

Comment 2 Ayal Baron 2012-11-07 21:59:46 UTC
(In reply to comment #0)
> Description of problem:
> 
> When provisioning a Virtual Machine the following error from vdsm.log is
> shown
> 
> 
> Version-Release number of selected component (if applicable):
> 
> 
> How reproducible:
> 
> Provision a new virtual machine
> 
> Steps to Reproduce:
> 1. Using web interface
> 2. [new server]
> 3. Half the provision process the NFS storage is unexpectedly
> detached/unplugged

What do you mean? did you disconnect the storage?
If so, what is the problem?

Also, please provide full vdsm.log and engine log.
>   
> Actual results:
> See attachment.
> 
> 
> Expected results:
> 
> 
> Additional info:

Comment 3 Valentin Bajrami 2012-11-09 13:11:09 UTC
Created attachment 641526 [details]
Full vdsm.log

Hi Ayal,

See the full vdsm.log file attached to this bug report.

If you need anything else that makes debugging easier, just let me know.

Valentin

Comment 4 Valentin Bajrami 2012-11-09 13:15:40 UTC
Created attachment 641527 [details]
Full engine.log

I did not disconnect the nfs mounts myself. During the provisionin of the VM, it just got disconnected / unplugged as result changing the virtualization host state to Non-Responsive. Minor details below

Running engine and vdsm on the same node.

Comment 5 Ayal Baron 2012-11-10 23:14:26 UTC
(In reply to comment #4)
> Created attachment 641527 [details]
> Full engine.log
> 
> I did not disconnect the nfs mounts myself. During the provisionin of the
> VM, it just got disconnected / unplugged as result changing the
> virtualization host state to Non-Responsive. Minor details below
> 
> Running engine and vdsm on the same node.

So I'm not sure what you expected exactly, your storage got disconnected while you were trying to create a VM on it.
The host can no longer run any VM because the storage is unavailable...

Comment 6 Valentin Bajrami 2012-11-11 14:52:15 UTC
That's the part I don't understand. Why would the storage be disconnected when a new VM is being installed?  This seems to me like a bug or maybe because the nfs server runs on the same machine? 

Can you suggest any workaround / permanent solution to this problem? I see a lot of similair problems on the ovirt mailing list.

Comment 7 Dave Neary 2013-03-18 09:47:46 UTC
Hi Ayal,

What action is required here? The behaviour seems to be surprising the user - would this be classed as "not a bug", "rfe", or is there something here we need to fix?

Thanks,
Dave.

Comment 8 Ayal Baron 2013-03-18 21:04:08 UTC
(In reply to comment #6)
> That's the part I don't understand. Why would the storage be disconnected
> when a new VM is being installed?  This seems to me like a bug or maybe
> because the nfs server runs on the same machine? 
> 
> Can you suggest any workaround / permanent solution to this problem? I see a
> lot of similair problems on the ovirt mailing list.

I totally missed the reply, sorry.

In general (just from the description) this is an NFS issue, it has nothing to do with oVirt.  If the mount is not operational then clearly we cannot go on provisioning the VM (I/O will fail).
In general communication shouldn't be leaving the host so I'm not sure why NFS would fail in this scenario though.
Looking at the vdsm.log I only see that the master domain is not found leaving me thinking that the problem is hiding in a previous log.
Are you able to reproduce this? (if so please reattach log immediately after it happens so I could review it.  better yet to use sosreport so I have all relevant logs).

When you say it just got disconnected / unplugged, what do you mean? did the network cable get disconnected or something else?

Comment 9 Ayal Baron 2013-03-18 21:04:38 UTC
Please set needinfo on me once additional info is available.