OpenStack Compute (nova)

Allocations records aren't removed after rescheduling

Series pike
Bug #1712718

Bug #1712718 reported by Alex Xu on 2017-08-24

This bug affects 1 person

Affects		Status	Importance	Assigned to	Milestone
	OpenStack Compute (nova)	Fix Released	High	Matt Riedemann
	Pike	Fix Committed	High	Matt Riedemann

Bug Description

When all the nodes are Pike release, the allocations records aren't removed from the destination host which failed to boot up the instance, after the instance are rescheduled in another host, the allocation records for the failed host still exists.

See original description

Tags:

Alex Xu (xuhj) on 2017-08-24

description:	updated
tags:	added: placement
tags:	added: scheduler

Revision history for this message

Alex Xu (xuhj) wrote on 2017-08-24:

There ia functional test for this bug https://review.openstack.org/#/c/496995/

OpenStack Infra (hudson-openstack) on 2017-08-24

Changed in nova:
assignee:	nobody → Alex Xu (xuhj)
status:	New → In Progress

Alex Xu (xuhj) on 2017-08-24

description:	updated
tags:	added: pike-rc-potential

Alex Xu (xuhj) on 2017-08-24

Changed in nova:
importance:	Undecided → High

OpenStack Infra (hudson-openstack) on 2017-08-24

Changed in nova:
assignee:	Alex Xu (xuhj) → Matt Riedemann (mriedem)

Revision history for this message

OpenStack Infra (hudson-openstack) wrote on 2017-08-24: Fix proposed to nova (stable/pike)

Fix proposed to branch: stable/pike
Review: https://review.openstack.org/497542

Revision history for this message

OpenStack Infra (hudson-openstack) wrote on 2017-08-24: Fix merged to nova (master)

Reviewed: https://review.openstack.org/496995
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=c19ebcbd58bc673e336bc5fa3b815c7b0a20d628
Submitter: Jenkins
Branch: master

commit c19ebcbd58bc673e336bc5fa3b815c7b0a20d628
Author: He Jie Xu <email address hidden>
Date: Thu Aug 24 11:38:31 2017 +0800

Remove allocation when booting instance rescheduled or aborted

    In Pike release, there is no auto-heal in the resource tracker.
    So when instance booting failed, the allocation records aren't
    cleaned up.

This patch clean up the allocation records when the instance
rescheduled and aborted.

Change-Id: I95d2f4c9392883052188fb7901451530068502db
Closes-Bug: #1712718

Changed in nova:
status:	In Progress → Fix Released

Revision history for this message

OpenStack Infra (hudson-openstack) wrote on 2017-08-24: Fix merged to nova (stable/pike)

Reviewed: https://review.openstack.org/497542
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=27fa0f03d266975e37dda6f4ccae1a58f7d5a029
Submitter: Jenkins
Branch: stable/pike

commit 27fa0f03d266975e37dda6f4ccae1a58f7d5a029
Author: He Jie Xu <email address hidden>
Date: Thu Aug 24 11:38:31 2017 +0800

Remove allocation when booting instance rescheduled or aborted

    In Pike release, there is no auto-heal in the resource tracker.
    So when instance booting failed, the allocation records aren't
    cleaned up.

This patch clean up the allocation records when the instance
rescheduled and aborted.

    Change-Id: I95d2f4c9392883052188fb7901451530068502db
    Closes-Bug: #1712718
    (cherry picked from commit c19ebcbd58bc673e336bc5fa3b815c7b0a20d628)