Bug #42788
closedunittest_rgw_dmclock_scheduler Failed
0%
Description
on aarch64 centos 7,unittest_rgw_dmclock_scheduler test failed
detail message:
171 - unittest_rgw_dmclock_scheduler (Failed)
[root@ceph2 build]# bin/unittest_rgw_dmclock_scheduler
[==========] Running 8 tests from 1 test suite.
[----------] Global test environment set-up.
[----------] 8 tests from Queue
[ RUN ] Queue.SyncRequest
[ OK ] Queue.SyncRequest (0 ms)
[ RUN ] Queue.RateLimit
[ OK ] Queue.RateLimit (1 ms)
[ RUN ] Queue.AsyncRequest
[ OK ] Queue.AsyncRequest (0 ms)
[ RUN ] Queue.Cancel
[ OK ] Queue.Cancel (0 ms)
[ RUN ] Queue.CancelClient
[ OK ] Queue.CancelClient (0 ms)
[ RUN ] Queue.CancelOnDestructor
[ OK ] Queue.CancelOnDestructor (0 ms)
[ RUN ] Queue.CrossExecutorRequest
[ OK ] Queue.CrossExecutorRequest (0 ms)
[ RUN ] Queue.SpawnAsyncRequest
/root/projects/src_ceph/ceph-master/src/test/rgw/test_rgw_dmclock_scheduler.cc:436: Failure
Value of: context.stopped()
Actual: false
Expected: true
[ FAILED ] Queue.SpawnAsyncRequest (1 ms)
[----------] 8 tests from Queue (2 ms total)
[----------] Global test environment tear-down
[==========] 8 tests from 1 test suite ran. (2 ms total)
[ PASSED ] 7 tests.
[ FAILED ] 1 test, listed below:
[ FAILED ] Queue.SpawnAsyncRequest
1 FAILED TESTUpdated by Brad Hubbard over 4 years ago
- Project changed from Ceph to rgw-testing
- Category deleted (
common)
Updated by Sebastian Wagner over 3 years ago
- Related to Bug #40069: unittest_rgw_dmclock_scheduler added
Updated by Sebastian Wagner over 3 years ago
- Related to Bug #46734: unittest_rgw_dmclock_scheduler: Queue.SyncRequest: ***Timeout 3600.01 sec: -1 Errors while parsing config file! added
Updated by Tomohiro Misono almost 3 years ago
Hi,
I got the same error in my development machine.
AFAIU, io_context::poll() does not guarantee that context is stopped:
https://github.com/ceph/ceph/blob/master/src/test/rgw/test_rgw_dmclock_scheduler.cc#L424
Changing poll() to run() solves the problem in my environment. Is it the correct way to fix the problem?
Updated by Kefu Chai almost 3 years ago
Tomohiro, i think that should be right fix. probably some requests posted to the io_context is not ready yet when poll() gets called.
Casey, sorry for hijacking this issue from you. it's happening more often recently.
i created https://github.com/ceph/ceph/pull/42237
Updated by Casey Bodley almost 3 years ago
- Project changed from rgw-testing to rgw
Updated by Casey Bodley almost 3 years ago
- Status changed from New to Fix Under Review
- Pull request ID set to 42425
Updated by Kefu Chai almost 3 years ago
- Status changed from Fix Under Review to Pending Backport
- Backport set to pacific
Updated by Backport Bot almost 3 years ago
- Copied to Backport #51771: pacific: unittest_rgw_dmclock_scheduler Failed added
Updated by Loïc Dachary over 2 years ago
- Status changed from Pending Backport to Resolved
While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are in status "Resolved" or "Rejected".