yahoo-eng-team team mailing list archive
-
yahoo-eng-team team
-
Mailing list archive
-
Message #60016
[Bug 1652155] Re: We don't need to log a warning when we can't delete an allocation for an instance that doesn't exist
Reviewed: https://review.openstack.org/414349
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=8ea3c838d8637b602f5d1d8d3f3279579c3994c8
Submitter: Jenkins
Branch: master
commit 8ea3c838d8637b602f5d1d8d3f3279579c3994c8
Author: Matt Riedemann <mriedem@xxxxxxxxxx>
Date: Thu Dec 22 18:03:32 2016 -0500
Ignore 404s when deleting allocation records
We don't need to emit a warning when trying to delete
an allocation record that doesn't exist. We see these
warnings all over the CI runs because of Tempest tests
that delete a server while it's building, in which case
the allocation record may have never been created.
Change-Id: I472fe1a091dc2524c50873d4fd916bcc535f51c7
Closes-Bug: #1652155
** Changed in: nova
Status: In Progress => Fix Released
--
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Compute (nova).
https://bugs.launchpad.net/bugs/1652155
Title:
We don't need to log a warning when we can't delete an allocation for
an instance that doesn't exist
Status in OpenStack Compute (nova):
Fix Released
Bug description:
These are all over the n-cpu logs in master branch CI runs now:
http://logs.openstack.org/75/405575/6/check/gate-tempest-dsvm-neutron-
full-ubuntu-
xenial/e1c5575/logs/screen-n-cpu.txt.gz?level=TRACE#_2016-12-22_16_52_19_551
2016-12-22 16:52:19.551 7348 WARNING nova.scheduler.client.report
[req-8eeeb512-bfe6-4f23-a9b1-92eff8b55fb3 tempest-
DeleteServersTestJSON-565294557 tempest-
DeleteServersTestJSON-565294557] Unable to delete allocation for
instance 19d35113-6c72-42ca-8b73-b76f630bb211: (404 404 Not Found
The resource could not be found.
No allocations for consumer '19d35113-6c72-42ca-8b73-b76f630bb211' )
--
If we're trying to delete something that doesn't exist, we don't need
to treat it as a warning or something that the operator needs to
investigate.
Based on the name of the tenant, these are probably from tests where
the instance is deleted while it's building.
This also shows up a ton in gate runs:
http://logstash.openstack.org/#dashboard/file/logstash.json?query=message%3A%5C%22Unable%20to%20delete%20allocation%20for%20instance%5C%22%20AND%20message%3A%5C%22404%20404%20Not%20Found%5C%22%20AND%20tags%3A%5C%22screen-n-cpu.txt%5C%22&from=7d
To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1652155/+subscriptions
References