Bug #42478: qa: AttributeError: can't set attribute - CephFS - Ceph

Actions

Copy link

Bug #42478

closed

qa: AttributeError: can't set attribute

Added by Patrick Donnelly over 4 years ago. Updated over 4 years ago.

Status:

Resolved

Priority:

Urgent

Assignee:

Jos Collin

Category:

Target version:

Ceph - v15.0.0

% Done:

Source:

Q/A

Tags:

Backport:

Regression:

Severity:

3 - minor

Reviewed:

Affected Versions:

ceph-qa-suite:

Component(FS):

qa-suite

Labels (FS):

Pull request ID:

31207

Crash signature (v1):

Crash signature (v2):

Description

2019-10-24T10:21:01.042 INFO:tasks.mds_thrash.fs.[cephfs]:waiting till mds map indicates mds.b is laggy/crashed, in failed state, or mds.b is removed from mdsmap
2019-10-24T10:21:03.044 INFO:teuthology.orchestra.run.smithi102:Running:
2019-10-24T10:21:03.044 INFO:teuthology.orchestra.run.smithi102:> sudo adjust-ulimits ceph-coverage /home/ubuntu/cephtest/archive/coverage timeout 120 ceph --cluster ceph fs dump --format=json
2019-10-24T10:21:12.795 INFO:teuthology.orchestra.run.smithi019:Running:
2019-10-24T10:21:12.795 INFO:teuthology.orchestra.run.smithi019:> sudo logrotate /etc/logrotate.d/ceph-test.conf
2019-10-24T10:21:12.798 INFO:teuthology.orchestra.run.smithi102:Running:
2019-10-24T10:21:12.799 INFO:teuthology.orchestra.run.smithi102:> sudo logrotate /etc/logrotate.d/ceph-test.conf
2019-10-24T10:21:42.839 INFO:teuthology.orchestra.run.smithi019:Running:
2019-10-24T10:21:42.839 INFO:teuthology.orchestra.run.smithi019:> sudo logrotate /etc/logrotate.d/ceph-test.conf
2019-10-24T10:21:42.843 INFO:teuthology.orchestra.run.smithi102:Running:
2019-10-24T10:21:42.843 INFO:teuthology.orchestra.run.smithi102:> sudo logrotate /etc/logrotate.d/ceph-test.conf
2019-10-24T10:22:12.887 INFO:teuthology.orchestra.run.smithi019:Running:
2019-10-24T10:22:12.888 INFO:teuthology.orchestra.run.smithi019:> sudo logrotate /etc/logrotate.d/ceph-test.conf
2019-10-24T10:22:12.894 INFO:teuthology.orchestra.run.smithi102:Running:
2019-10-24T10:22:12.894 INFO:teuthology.orchestra.run.smithi102:> sudo logrotate /etc/logrotate.d/ceph-test.conf
2019-10-24T10:22:42.934 INFO:teuthology.orchestra.run.smithi019:Running:
2019-10-24T10:22:42.934 INFO:teuthology.orchestra.run.smithi019:> sudo logrotate /etc/logrotate.d/ceph-test.conf
2019-10-24T10:22:42.938 INFO:teuthology.orchestra.run.smithi102:Running:
2019-10-24T10:22:42.938 INFO:teuthology.orchestra.run.smithi102:> sudo logrotate /etc/logrotate.d/ceph-test.conf
2019-10-24T10:23:03.064 DEBUG:teuthology.orchestra.run:got remote process result: 124
2019-10-24T10:23:03.068 ERROR:teuthology:Uncaught exception (Hub)
Traceback (most recent call last):
  File "/home/teuthworker/src/git.ceph.com_git_teuthology_master/virtualenv/local/lib/python2.7/site-packages/gevent/greenlet.py", line 536, in run
    result = self._run(*self.args, **self.kwargs)
  File "/home/teuthworker/src/git.ceph.com_ceph-c_wip-pdonnell-testing-20191024.021348/qa/tasks/mds_thrash.py", line 139, in _run
    self.exception = e
AttributeError: can't set attribute
2019-10-24T10:23:03.758 INFO:tasks.daemonwatchdog.daemon_watchdog:thrasher on fs.cephfs failed
2019-10-24T10:23:03.759 INFO:tasks.daemonwatchdog.daemon_watchdog:BARK! unmounting mounts and killing all daemons

From: /ceph/teuthology-archive/pdonnell-2019-10-24_06:19:20-fs-wip-pdonnell-testing-20191024.021348-distro-basic-smithi/4439995/teuthology.log

Related issues 1 (0 open — 1 closed)

Actions

Copy link

Updated by Patrick Donnelly over 4 years ago

Description updated (diff)

Actions

Copy link

Updated by Jos Collin over 4 years ago

Status changed from New to Need More Info

This happens when there is no setter in thrasher.py. Can you show me your thrasher.py and mds_thrash.py, so that I get more clarity? I hope your branch is latest.

Actions

Copy link

Updated by Patrick Donnelly over 4 years ago

Jos Collin wrote:

This happens when there is no setter in thrasher.py. Can you show me your thrasher.py and mds_thrash.py, so that I get more clarity? I hope your branch is latest.

https://github.com/ceph/ceph-ci/tree/wip-pdonnell-testing-20191024.021348

Actions

Copy link

Updated by Jos Collin over 4 years ago

Status changed from Need More Info to In Progress

https://github.com/ceph/ceph-ci/blob/0772e8a667e86de7945704f53c601d09a49232f1/qa/tasks/mds_thrash.py#L21
https://github.com/ceph/ceph-ci/blob/0772e8a667e86de7945704f53c601d09a49232f1/qa/tasks/mds_thrash.py#L101

We should avoid using MRO and better call Greenlet.__init__() as specified here: http://www.gevent.org/api/gevent.greenlet.html#subclassing-greenlet and call Thrasher.__init__ separately. This avoids the doubt whether Thrasher is instantiated via Greenlet or not.

Actions

Copy link