Description of problem: rados test fails with random ms fails following test test/librados/watch_notify.cc:309: Failure LibRadosWatchNotify.Watch2Timeout Version-Release number of selected component (if applicable): ceph version 10.1.1-1.el7cp (61adb020219fbad4508050b5f0a792246ba74dae) How reproducible: 1/1 Steps to Reproduce: rados/test.sh with random ms Actual results: 2016-04-09T23:35:23.140 INFO:tasks.workunit.client.0.clara004.stdout: api_watch_notify: [ RUN ] LibRadosWatchNotify.AioWatchDelete 2016-04-09T23:35:23.141 INFO:tasks.workunit.client.0.clara004.stdout: api_watch_notify: waiting up to 300 for disconnect notification ... 2016-04-09T23:35:23.142 INFO:tasks.workunit.client.0.clara004.stdout: api_watch_notify: watch_notify2_test_errcb cookie 140434457569840 err -107 2016-04-09T23:35:23.143 INFO:tasks.workunit.client.0.clara004.stdout: api_watch_notify: [ OK ] LibRadosWatchNotify.AioWatchDelete (1021 ms) 2016-04-09T23:35:23.144 INFO:tasks.workunit.client.0.clara004.stdout: api_watch_notify: [ RUN ] LibRadosWatchNotify.Watch2Timeout 2016-04-09T23:35:23.145 INFO:tasks.workunit.client.0.clara004.stdout: api_watch_notify: waiting up to 900 for osd to time us out ... 2016-04-09T23:35:23.146 INFO:tasks.workunit.client.0.clara004.stdout: api_watch_notify: test/librados/watch_notify.cc:309: Failure 2016-04-09T23:35:23.148 INFO:tasks.workunit.client.0.clara004.stdout: api_watch_notify: Value of: left > 0 2016-04-09T23:35:23.149 INFO:tasks.workunit.client.0.clara004.stdout: api_watch_notify: Actual: false 2016-04-09T23:35:23.150 INFO:tasks.workunit.client.0.clara004.stdout: api_watch_notify: Expected: true 2016-04-09T23:35:23.151 INFO:tasks.workunit.client.0.clara004.stdout: api_watch_notify: [ FAILED ] LibRadosWatchNotify.Watch2Timeout (899114 ms) Expected results: Additional info: http://magna002.ceph.redhat.com/vasu-2016-04-09_11:26:14-rados:monthrash-jewel---basic-clara/217843/teuthology.log
Sage has a PR
Looks like this was fixed upstream in v10.2.0.
Pending backport to hammer.
Oops, I changed the wrong one. This is in 10.2.0, so I marked it POST?
It's in 10.2.0, so it's MODIFIED I guess
Verified in regression run.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHBA-2016-1755.html