Bug #3048: rados bench: use after free? - Ceph - Ceph

Actions

Copy link

Bug #3048

closed

rados bench: use after free?

Added by Sage Weil over 11 years ago. Updated over 11 years ago.

Status:

Resolved

Priority:

Urgent

Assignee:

Category:

Target version:

% Done:

Source:

Development

Tags:

Backport:

Regression:

Severity:

Reviewed:

Affected Versions:

ceph-qa-suite:

Pull request ID:

Crash signature (v1):

Crash signature (v2):

Description


2012-08-25T21:38:54.250 INFO:teuthology.orchestra.run.err:2012-08-25 21:38:54.220370 7f08d7fff700 -1 ./common/Mutex.h: In function 'void Mutex::Lock(bool)' thread 7f08d7fff700 time 2012-08-25 21:38:54.219572
2012-08-25T21:38:54.250 INFO:teuthology.orchestra.run.err:./common/Mutex.h: 113: FAILED assert(r == 0)
2012-08-25T21:38:54.250 INFO:teuthology.orchestra.run.err:
2012-08-25T21:38:54.250 INFO:teuthology.orchestra.run.err: ceph version 0.50-383-gaa91cf8 (commit:aa91cf81af548463ae583872e319316871771e94)
2012-08-25T21:38:54.250 INFO:teuthology.orchestra.run.err: 1: (Mutex::Lock(bool)+0x118) [0x463bf8]
2012-08-25T21:38:54.251 INFO:teuthology.orchestra.run.err: 2: (_aio_cb(void*, void*)+0x16) [0x45a5b6]
2012-08-25T21:38:54.251 INFO:teuthology.orchestra.run.err: 3: (librados::C_AioSafe::finish(int)+0x20) [0x7f08e9f91070]
2012-08-25T21:38:54.251 INFO:teuthology.orchestra.run.err: 4: (Finisher::finisher_thread_entry()+0x340) [0x7f08ea00f1c0]
2012-08-25T21:38:54.251 INFO:teuthology.orchestra.run.err: 5: (Finisher::FinisherThread::entry()+0x15) [0x7f08e9f7a335]
2012-08-25T21:38:54.251 INFO:teuthology.orchestra.run.err: 6: (Thread::_entry_func(void*)+0x12) [0x47b192]
2012-08-25T21:38:54.251 INFO:teuthology.orchestra.run.err: 7: (()+0x7e9a) [0x7f08e9c36e9a]
2012-08-25T21:38:54.251 INFO:teuthology.orchestra.run.err: 8: (clone()+0x6d) [0x7f08e865c4bd]
2012-08-25T21:38:54.251 INFO:teuthology.orchestra.run.err: NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this.

ubuntu@teuthology:/a/teuthology-2012-08-25_19:00:04-regression-master-testing-gcov/8815 for core etc.

i'm guessing the rados bench cleanup changes are responsible?