test_hotplug_nic intermittently fails teardown due to "One or more ports have an IP allocation from this subnet."

[Matt Riedemann]

Seen here:

http://logs.openstack.org/92/627892/9/check/nova-next/d402065/job-output.txt.gz#_2019-01-30_03_05_59_790327

2019-01-30 03:05:59.790327 | primary | Captured traceback-1:
2019-01-30 03:05:59.790348 | primary | ~~~~~~~~~~~~~~~~~~~~~
2019-01-30 03:05:59.790375 | primary | Traceback (most recent call last):
2019-01-30 03:05:59.790440 | primary | File "tempest/lib/common/utils/test_utils.py", line 84, in call_and_ignore_notfound_exc
2019-01-30 03:05:59.790467 | primary | return func(*args, **kwargs)
2019-01-30 03:05:59.790518 | primary | File "tempest/lib/services/network/subnets_client.py", line 52, in delete_subnet
2019-01-30 03:05:59.790549 | primary | return self.delete_resource(uri)
2019-01-30 03:05:59.790607 | primary | File "tempest/lib/services/network/base.py", line 41, in delete_resource
2019-01-30 03:05:59.790636 | primary | resp, body = self.delete(req_uri)
2019-01-30 03:05:59.790688 | primary | File "tempest/lib/common/rest_client.py", line 311, in delete
2019-01-30 03:05:59.790733 | primary | return self.request('DELETE', url, extra_headers, headers, body)
2019-01-30 03:05:59.790774 | primary | File "tempest/lib/common/rest_client.py", line 676, in request
2019-01-30 03:05:59.790805 | primary | self._error_checker(resp, resp_body)
2019-01-30 03:05:59.790850 | primary | File "tempest/lib/common/rest_client.py", line 797, in _error_checker
2019-01-30 03:05:59.790885 | primary | raise exceptions.Conflict(resp_body, resp=resp)
2019-01-30 03:05:59.790930 | primary | tempest.lib.exceptions.Conflict: Conflict with state of target resource
2019-01-30 03:05:59.791051 | primary | Details: {u'detail': u'', u'type': u'SubnetInUse', u'message': u'Unable to complete operation on subnet e6e7e892-bd6b-453b-9e6f-08d889c83190: One or more ports have an IP allocation from this subnet.'}

From the server logs:

http://logs.openstack.org/92/627892/9/check/nova-next/d402065/logs/screen-q-svc.txt.gz#_Jan_30_03_05_44_715787

Jan 30 03:05:44.715787 ubuntu-xenial-ovh-bhs1-0002242443 neutron-server[25564]: INFO neutron.db.db_base_plugin_v2 [None req-62a95088-0965-4d58-a43d-6450fee48399 tempest-TestNetworkBasicOps-1823641274 tempest-TestNetworkBasicOps-1823641274] Found port (3080b6d8-f622-496b-a749-0b7def78fb6c, 10.1.0.26) having IP allocation on subnet e6e7e892-bd6b-453b-9e6f-08d889c83190, cannot delete

My guess is it's a timing issue during the delete and we're maybe not waiting long enough in tempest? Unless something failed when trying to delete port 3080b6d8-f622-496b-a749-0b7def78fb6c?

http://logstash.openstack.org/#dashboard/file/logstash.json?query=message%3A%5C%22Body%3A%5C%22%20AND%20message%3A%5C%22%3A%20One%20or%20more%20ports%20have%20an%20IP%20allocation%20from%20this%20subnet%5C%22%20AND%20tags%3A%5C%22console%5C%22&from=7d

15 hits in 7 days, check and gate, all failures.

I want to say I've seen this before in the gate but couldn't find an existing bug for it, unless it was already fixed once and this is a regression.

[Slawek Kaplonski (slaweq)]

I didn't saw issues like that which causes directly job failure for long time.
I know that things like that may happen but it's more like red herring and the root cause is other issue which happened earlier.
So lets close this bug for now, we can always reopen it if needed.

[Launchpad Janitor (janitor)]

[Expired for neutron because there has been no activity for 60 days.]