Bug 1003516

Summary: failed to start VM after adding payload over an existing payload
Product: Red Hat Enterprise Virtualization Manager Reporter: sefi litmanovich <slitmano>
Component: ovirt-engineAssignee: Shahar Havivi <shavivi>
Status: CLOSED NOTABUG QA Contact: sefi litmanovich <slitmano>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 3.3.0CC: acathrow, iheim, lpeer, lsvaty, michal.skrivanek, pstehlik, Rhev-m-bugs, slitmano, talayan, yeylon
Target Milestone: ---Keywords: Triaged
Target Release: 3.3.0   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard: virt
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2013-09-04 07:38:19 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
logs - engine and vdsm none

Description sefi litmanovich 2013-09-02 08:55:40 UTC 
Description of problem:
after updating a vm twice with different payloads through RESTapi, vm fails to start.

Version-Release number of selected component (if applicable):
is12

Steps to Reproduce:

1. Create Vm and install RHEL 6.4
2. Stop vm and create template
3. create 5 vms pool from template.
4. with the RESTapi update one vm from the pool with payload:

PUT .../api/vms/da37a39c-807b-4150-b369-a8698a696acb/

<vm>
<payloads>
<payload type="cdrom">
<file name="file1.update.cdrom">
<content>'some content cdrom 1!'</content>
</file>
</payload>
</payloads>
</vm>

5. start vm.
6. ssh vm and run command: mkdir /tmp/cdrom_payload1; mount /dev/cdrom1 /tmp/cdrom_payload1;
7. file.update.cdrom should be vissibl with its content in folder /tmp/cdrom_payload1.
8. umount /dev/cdrom1
9. check that file doesn't exit anymore.
10. stop vm
11. with the RESTapi update the same vm with a new payload:

PUT .../api/vms/da37a39c-807b-4150-b369-a8698a696acb/

<vm>
<payloads>
<payload type="cdrom">
<file name="file2.update.cdrom">
<content>'some content cdrom 2!'</content>
</file>
</payload>
</payloads>
</vm>

12. start vm in order to reproduce the same procedure.


Actual results:

vm fails to start, both from GUI and RESTapi.

Expected results:

vm should start and new payload should be available for mounting as prior payload.

Additional info:

logs attached
Comment 1 sefi litmanovich 2013-09-02 08:56:48 UTC 
Created attachment 792767 [details]
logs - engine and vdsm
Comment 2 Shahar Havivi 2013-09-03 11:22:31 UTC 
Hi,
I cannot reproduce the error.
Looking at the log its look like you have recurring error:
---------------------------
Unknown libvirterror: ecode: 42 edom: 10 level: 2 message: Domain not found: no domain with matching uuid 'da37a39c-807b-4150-b369-a8698a696acb
---------------------------
It may be related to storage?

Please try to reproduce the problem with less steps (I don't think you need to create 5 VMs, only update the payload)
If you do get the error again please upload the logs and post the exact time that the error occur in the VDSM and Engine (for easier pin point the problem).
Comment 3 sefi litmanovich 2013-09-03 12:54:33 UTC 
Hi,

You are right, after created a new vm unrelated to the pool I created earlir it worked fine.
Comment 4 Shahar Havivi 2013-09-03 15:28:04 UTC 
(In reply to sefi litmanovich from comment #3)
> Hi,
> 
> You are right, after created a new vm unrelated to the pool I created earlir
> it worked fine.

So can we close the bug?