Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1601498

Summary: Migration of VM
Product: [oVirt] ovirt-engine Reporter: Vladimir <vvlopuk>
Component: GeneralAssignee: bugs <bugs>
Status: CLOSED NOTABUG QA Contact: meital avital <mavital>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 4.2.5.1CC: bugs, michal.skrivanek, vvlopuk
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-07-18 20:20:38 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: Virt RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
VM
none
Log_zip_1
none
Log_zip_2
none
Log_zip_3
none
Log_zip_4
none
Log_zip_5 none

Description Vladimir 2018-07-16 13:53:07 UTC
Created attachment 1459186 [details]
VM

Description of problem: Migration of VM.


Version-Release number of selected component (if applicable):4.2.5.1-1.el7


How reproducible:


Steps to Reproduce:

Hello.
I have a problem with VM migration.
If you disable 1 node, the VM does not automatically migrate.

Actual results:


Expected results:


Additional info:

Comment 1 Michal Skrivanek 2018-07-17 14:59:02 UTC
sorry, that doesn't really give us enough information. Can you please provide more details about what you did and how, how does the setup look like, ovirt-engine logs, etc.

The picture just says it's an Unknown VM meaning it used to run on a host which "disappeared". That's expected. It will either appear again when it starts responding on its own(e.g. temporary network outage), or you have to manually clear it out if you happen to reinstall/fix the host (e.g. hw failure)

Comment 2 Vladimir 2018-07-18 05:15:13 UTC
Hey.
At some point, HW crashed on the server node1, after which it moved to the "Non responsive" state, while VMs running on it did not migrate to working nodes.

In the attachment, all log files.

Comment 3 Vladimir 2018-07-18 05:16:57 UTC
Created attachment 1459615 [details]
Log_zip_1

Comment 4 Vladimir 2018-07-18 05:17:38 UTC
Created attachment 1459616 [details]
Log_zip_2

Comment 5 Vladimir 2018-07-18 05:18:16 UTC
Created attachment 1459617 [details]
Log_zip_3

Comment 6 Vladimir 2018-07-18 05:19:19 UTC
Created attachment 1459618 [details]
Log_zip_4

Comment 7 Vladimir 2018-07-18 05:19:59 UTC
Created attachment 1459619 [details]
Log_zip_5

Comment 8 Michal Skrivanek 2018-07-18 20:20:38 UTC
(In reply to Vladimir from comment #2)
> Hey.
> At some point, HW crashed on the server node1, after which it moved to the
> "Non responsive" state, while VMs running on it did not migrate to working
> nodes.
> 
> In the attachment, all log files.

Ah, then it’s clear. You expected a restart elsewhere, not a live migration. Since you apparently didn’t configure power management for that host oVirt has no way how to determine that the host is really not running the VM anymore and therefore cannot start it elsewhere
Use power management for your hosts and/or High Available VMs with a storage lease where we can determine the actual state using storage rather than network reachability