rgwAdminAPIFailed in master ci

[ZhiQiang Fan]

2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager Traceback (most recent call last):
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager File "/opt/stack/new/ceilometer/ceilometer/agent/manager.py", line 200, in poll_and_notify
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager for sample in samples:
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager File "/opt/stack/new/ceilometer/ceilometer/objectstore/rgw.py", line 206, in get_samples
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager cache, resources):
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager File "/opt/stack/new/ceilometer/ceilometer/objectstore/rgw.py", line 86, in _iter_accounts
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager ksclient, tenants))
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager File "/opt/stack/new/ceilometer/ceilometer/objectstore/rgw.py", line 102, in _get_account_info
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager yield t.id, getattr(rgw_client, api_method)(t.id)
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager File "/opt/stack/new/ceilometer/ceilometer/objectstore/rgw_client.py", line 70, in get_usage
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager json_data = self._make_request(path, req_params)
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager File "/opt/stack/new/ceilometer/ceilometer/objectstore/rgw_client.py", line 49, in _make_request
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager {'status': r.status_code, 'reason': r.reason})
2016-04-21 09:40:43.968 23094 ERROR ceilometer.agent.manager RGWAdminAPIFailed: RGW AdminOps API returned 400 Bad Request

it seems our rgw_client doesn't up to date with new ceph admin api

[ZhiQiang Fan (aji-zqfan)]

http://logs.openstack.org/53/308853/1/check/gate-tempest-dsvm-ceilometer-mysql-full/19ebd08/logs/screen-ceilometer-acentral.txt.gz?level=ERROR#_2016-04-21_09_40_43_968

[Sumant Murke (sumant-murke)]

How to reproduce this ? Are you running on devstack ?

[gordon chung (chungg)]

yeah, its visible in gate. we should look into this.

[gordon chung (chungg)]

we arguably should not post backtrace for warning https://github.com/openstack/ceilometer/blob/dbbc05f45570c94cf3caec241d821b303a797483/ceilometer/agent/manager.py#L223

[OpenStack Infra (hudson-openstack)]

Related fix proposed to branch: master
Review: https://review.openstack.org/392878

[gordon chung (chungg)]

we should stop polling rgw if not available.

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/392878
Committed: https://git.openstack.org/cgit/openstack/ceilometer/commit/?id=03298e645d934aa307ab24cd8103e8edadcc6ebc
Submitter: Jenkins
Branch: master

commit 03298e645d934aa307ab24cd8103e8edadcc6ebc
Author: gord chung <email address hidden>
Date: Wed Nov 2 16:39:40 2016 +0000

    stop assuming ceph/swift share same endpoint

    showing backtrace on a handled and warning logged exception seems
    aggressive, just return as error.

    also, stop setting ceph/swift to same endpoint

    Change-Id: I33be2d433c86cd67add5fac92fed1b66219d48b6
    Closes-Bug: #1573241

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/ceilometer 8.0.0 release.