Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1196678

Summary: Failing to migrate VMs between FC19 and FC20 hosts
Product: [Retired] oVirt Reporter: Brian Sipos <BSipos>
Component: vdsmAssignee: Dan Kenigsberg <danken>
Status: CLOSED DUPLICATE QA Contact: Gil Klein <gklein>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 3.5CC: amureini, bazulay, BSipos, bugs, ecohen, gklein, iheim, lsurette, mgoldboi, michal.skrivanek, ofrenkel, rbalakri, yeylon
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard: virt
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2015-03-23 12:56:11 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:
Embargoed:
Attachments:
Description Flags
engine log from a few seconds before the failed migration
none
migration-source VDSM log from up to two minutes before the failed migration none

Description Brian Sipos 2015-02-26 14:18:06 UTC
Description of problem:
I currently have ovirt-engine running on a CentOS VM and two ovirt hosts (one on Fedora 19 on on Fedora 20).

Version-Release number of selected component (if applicable):
ovirt-engine (on EL6) is 3.5.1.1

How reproducible:
Consistently, for migration in either direction.

Steps to Reproduce:
1. Start up a VM, it is allocated to a particular host.
2. Select the running VM in the VM list and select Migrate.
3. Observe the ovirt-engine log contains (with timestamps less than a minute apart):
Migration started (VM: vm, Source: host-fc20, Destination: host-fc19, User: ...).
Migration failed due to Error: Fatal error during migration. Trying to migrate to another Host (VM: vm, Source: host-fc20, Destination: host-fc19).

Actual results:
The VM fails to migrate.

Expected results:
The VM should end up running on the alternate host.

Additional info:
I can provide individual "vdsm.log" from the hosts or "engine.log" from the engine as needed.
If this matters, the cluster is running in compatibility version 3.4 since it was created when 3.4 was latest version. I should be able to bump this compat. version although I have never attempted to do so.

Comment 1 Omer Frenkel 2015-03-01 08:06:00 UTC
please attach engine.log and vdsm.log from source host for the time of the migration.
thanks!

Comment 2 Brian Sipos 2015-03-02 14:09:11 UTC
Created attachment 997101 [details]
engine log from a few seconds before the failed migration

Comment 3 Brian Sipos 2015-03-02 14:10:04 UTC
Created attachment 997102 [details]
migration-source VDSM log from up to two minutes before the failed migration

Comment 4 Brian Sipos 2015-03-02 19:08:09 UTC
Upon further inspection, it appears that I have blocked the libvirt port (TCP 16514) on one host's firewalls. Now that the port is open, I am running across what appears to be a known FC19 issue in libvirtd (https://bugzilla.redhat.com/show_bug.cgi?id=1018178).

Comment 5 Michal Skrivanek 2015-03-23 12:56:11 UTC
ok. thanks for finding out. no need to track it again here...

*** This bug has been marked as a duplicate of bug 1018178 ***