Project

General

Profile

Actions

Bug #41028

closed

Segmentation fault in rbd/test_librbd.sh

Added by Yuri Weinstein over 4 years ago. Updated over 4 years ago.

Status:
Won't Fix
Priority:
Normal
Assignee:
-
Target version:
-
% Done:

0%

Source:
Q/A
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
rados
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

Run: http://pulpito.ceph.com/yuriw-2019-07-30_15:39:36-rados-wip-yuri3-testing-2019-07-25-2227-luminous-distro-basic-smithi/
Jobs: 4165353
Logs: yuriw-2019-07-30_15:39:36-rados-wip-yuri3-testing-2019-07-25-2227-luminous-distro-basic-smithi/4165353/teuthology.log

019-07-30T17:32:49.353 INFO:teuthology.orchestra.run.smithi121:Running:
2019-07-30T17:32:49.354 INFO:teuthology.orchestra.run.smithi121:> sudo adjust-ulimits ceph-coverage /home/ubuntu/cephtest/archive/coverage timeout 120 ceph --cluster ceph osd unset noscrub
2019-07-30T17:32:49.358 INFO:tasks.ceph.osd.1.smithi121.stderr:2019-07-30 17:32:49.353299 7fc64eb6a700 -1 received  signal: Hangup from  PID: 175675 task name: /usr/bin/python /usr/bin/daemon-helper kill ceph-osd -f --cluster ceph -i 1  UID: 0
2019-07-30T17:32:49.367 INFO:tasks.workunit.client.0.smithi168.stderr:Segmentation fault (core dumped)
2019-07-30T17:32:49.367 DEBUG:teuthology.orchestra.run:got remote process result: 139
2019-07-30T17:32:49.367 INFO:tasks.workunit:Stopping ['rbd/test_librbd.sh'] on client.0...

Actions #1

Updated by David Zafman over 4 years ago

This is the core file for this run. It is NOT ceph-osd as might be assumed by the log output:

remote/smithi168/coredump/1564507968.10856.core: ELF 64-bit LSB core file x86-64, version 1 (SYSV), SVR4-style, from 'ceph_test_librbd'

Actions #2

Updated by Neha Ojha over 4 years ago

  • Project changed from RADOS to rbd
Actions #3

Updated by Neha Ojha over 4 years ago

  • Subject changed from Segmentation fault in rados to Segmentation fault in rbd/test_librbd.sh
Actions #4

Updated by Jason Dillaman over 4 years ago

  • Status changed from New to Won't Fix

Looks like a sporadic duplicate watch/notify callback bug in the unit test (due to thrashing no doubt), but Jewel is EOL.

Actions

Also available in: Atom PDF