live migration (block migration) fails at post_live_migration_at_destination function, but the status of this instance is still "migrating".

Bug #1379581 reported by Charlotte Han
18
This bug affects 2 people
Affects Status Importance Assigned to Milestone
OpenStack Compute (nova)
Fix Released
High
Pawel Koniszewski
Nominated for Mitaka by Pawel Koniszewski

Bug Description

I have two compute node, host opencos114-93 and host opencos179-24.
A migration failed at function of " post_live_migration_at_destination", but the status of this instance is still migrating.
Log is as follows:
http://paste.openstack.org/show/121143/

Revision history for this message
Charlotte Han (hanrong) wrote :

A Exception of "NovaException" is thrown and it would be not captured when the function "post_live_migration_at_destination" is in process.
I hope the the Exception of "NovaException" will be captured, and roll back.
We should either put the Instance status into ERROR, if we can't rollback.

Charlotte Han (hanrong)
summary: live migration (block migration) fails at
post_live_migration_at_destination function, but the status of this
- instance is still "migrating",
+ instance is still "migrating".
Revision history for this message
Joe Gordon (jogo) wrote :

can you share the logs on paste.openstack.org so they don't get wrapped and are easier to read. Also what version of nova are you using?

Changed in nova:
status: New → Incomplete
Revision history for this message
Joe Gordon (jogo) wrote :

Marking as incomplete since we don't know what version of nova this is

Revision history for this message
Charlotte Han (hanrong) wrote :
Revision history for this message
Charlotte Han (hanrong) wrote :

the version of nova is icehouse 2014.1.2.

Charlotte Han (hanrong)
Changed in nova:
status: Incomplete → New
Charlotte Han (hanrong)
description: updated
Revision history for this message
Joe Gordon (jogo) wrote :

Are you using nova network or neutron? "NovaException: Unexpected vif_type=binding_failed"
 makes me thing this is a networking issue.

tags: added: network
Revision history for this message
Charlotte Han (hanrong) wrote :

I am using neutron.
if the network is error, I hope the migrating Instance change its status.

Revision history for this message
John Garbutt (johngarbutt) wrote :

I am seeing similar issues, when neutron operations fail during post_live_migrate, it leaves things in a very messy state.

I am working on a patch to fix this now.

Changed in nova:
importance: Undecided → High
status: New → In Progress
assignee: nobody → John Garbutt (johngarbutt)
tags: added: compute live-migrate
Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix proposed to nova (master)

Fix proposed to branch: master
Review: https://review.openstack.org/129881

Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Change abandoned on nova (master)

Change abandoned by Joe Gordon (<email address hidden>) on branch: master
Review: https://review.openstack.org/129881
Reason: This review is > 4 weeks without comment, and failed Jenkins the last time it was checked. We are abandoning this for now. Feel free to reactivate the review by pressing the restore button and leaving a 'recheck' comment to get fresh test results.

Charlotte Han (hanrong)
Changed in nova:
assignee: John Garbutt (johngarbutt) → Rong Han ZTE (hanrong)
Revision history for this message
Charlotte Han (hanrong) wrote :

Hi John Garbutt, since it was over 4 weeks after your last change, I hope you do not mind if I take over this bug and give it try.

Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix proposed to nova (master)

Fix proposed to branch: master
Review: https://review.openstack.org/186634

Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Change abandoned on nova (master)

Change abandoned by Michael Still (<email address hidden>) on branch: master
Review: https://review.openstack.org/186634
Reason: This patch has been stalled for quite a while, so I am going to abandon it to keep the code review queue sane. Please restore the change when it is ready for review.

Changed in nova:
assignee: Rong Han ZTE (hanrong) → Pawel Koniszewski (pawel-koniszewski)
Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix proposed to nova (master)

Fix proposed to branch: master
Review: https://review.openstack.org/235994

Paul Murray (pmurray)
tags: added: live-migration
removed: live-migrate
Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix merged to nova (master)

Reviewed: https://review.openstack.org/235994
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=89b1fecce116bc44f558e76cbb5dc43497ea67cc
Submitter: Jenkins
Branch: master

commit 89b1fecce116bc44f558e76cbb5dc43497ea67cc
Author: Pawel Koniszewski <email address hidden>
Date: Mon Feb 1 11:56:59 2016 +0100

    Update instance host in post live migration even when exception occurs

    Currently when, e.g., port binding fails on destination host nova
    loses track of running VM. Operator needs to change record in DB
    manually in order to recover VM in nova and then perform operations
    on destination host to repair such VM. Because VM is on destination
    host already it should be updated regardless of post live migration
    at destination result.

    Change-Id: Ibb5158f453abd9717e6d2ab501295351ca9d0dcf
    Closes-Bug: #1379581

Changed in nova:
status: In Progress → Fix Released
Revision history for this message
Doug Hellmann (doug-hellmann) wrote : Fix included in openstack/nova 13.0.0.0b3

This issue was fixed in the openstack/nova 13.0.0.0b3 development milestone.

Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix proposed to nova (stable/liberty)

Fix proposed to branch: stable/liberty
Review: https://review.openstack.org/358439

Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix merged to nova (stable/liberty)

Reviewed: https://review.openstack.org/358439
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=53fffe02bbabdd71d3cd829d73fba0c19c9b9175
Submitter: Jenkins
Branch: stable/liberty

commit 53fffe02bbabdd71d3cd829d73fba0c19c9b9175
Author: Pawel Koniszewski <email address hidden>
Date: Mon Feb 1 11:56:59 2016 +0100

    Update instance host in post live migration even when exception occurs

    Currently when, e.g., port binding fails on destination host nova
    loses track of running VM. Operator needs to change record in DB
    manually in order to recover VM in nova and then perform operations
    on destination host to repair such VM. Because VM is on destination
    host already it should be updated regardless of post live migration
    at destination result.

    (cherry picked from commit 89b1fecce116bc44f558e76cbb5dc43497ea67cc)

    Change-Id: Ibb5158f453abd9717e6d2ab501295351ca9d0dcf
    Closes-Bug: #1379581
    Conflicts:
     nova/tests/unit/compute/test_compute_mgr.py

tags: added: in-stable-liberty
Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix included in openstack/nova 12.0.5

This issue was fixed in the openstack/nova 12.0.5 release.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.