ea2ea492a3
If the source compute service is down when a resize or cold migrate is initiated the prep_resize cast from the selected destination compute service to the source will fail/hang. The API can validate the source compute service is up or fail the operation with a 409 response if the source service is down. Note that a host status of "MAINTENANCE" means the service is up but disabled by an administrator which is OK for resize/cold migrate. The solution here works the validation into the check_instance_host decorator which surprisingly isn't used in more places where the source host is involved like reboot, rebuild, snapshot, etc. This change just handles the resize method but is done in such a way that the check_instance_host decorator could be applied to those other methods and perform the is-up check as well. The decorator is made backward compatible by default. Note that Instance._save_services is added because during resize the Instance is updated and the services field is set but not actually changed, but Instance.save() handles object fields differently so we need to implement the no-op _save_services method to avoid a failure. Change-Id: I85423c7bcacff3bc465c22686d0675529d211b59 Closes-Bug: #1856925
9 lines
288 B
YAML
9 lines
288 B
YAML
---
|
|
fixes:
|
|
- |
|
|
This release contains a fix for `bug 1856925`_ such that ``resize`` and
|
|
``migrate`` server actions will be rejected with a 409 ``HTTPConflict``
|
|
response if the source compute service is down.
|
|
|
|
.. _bug 1856925: https://bugs.launchpad.net/nova/+bug/1856925
|