Bug #13350
closed"Error cancelling global ust listener thread" errors
0%
Description
See lots of those, seem benign
Run: http://pulpito.ceph.com/teuthology-2015-10-02_10:48:43-upgrade:hammer-x:point-to-point-x-infernalis-distro-basic-vps/
Logs: http://qa-proxy.ceph.com/teuthology/teuthology-2015-10-02_10:48:43-upgrade:hammer-x:point-to-point-x-infernalis-distro-basic-vps/1084366/teuthology.log
2015-10-02T11:58:48.861 INFO:tasks.workunit.client.0.vpm088.stderr:libust[20605/20605]: Error: Error cancelling global ust listener thread: No such process (in lttng_ust_exit() at lttng-ust-comm.c:1517) 2015-10-02T11:58:48.861 INFO:tasks.workunit.client.0.vpm088.stderr:libust[20605/20605]: Error: Error cancelling local ust listener thread: No such process (in lttng_ust_exit() at lttng-ust-comm.c:1526) 2015-10-02T11:58:51.073 INFO:tasks.workunit.client.0.vpm088.stdout:process_2_[20606]: listing objects. 2015-10-02T11:58:51.074 INFO:tasks.workunit.client.0.vpm088.stdout:process_2_[20606]: saw 0 objects 2015-10-02T11:58:51.074 INFO:tasks.workunit.client.0.vpm088.stdout:process_2_[20606]: shutting down. 2015-10-02T11:58:51.074 INFO:tasks.workunit.client.0.vpm088.stderr:libust[20606/20606]: Error: Error cancelling global ust listener thread: No such process (in lttng_ust_exit() at lttng-ust-comm.c:1517) 2015-10-02T11:58:51.074 INFO:tasks.workunit.client.0.vpm088.stderr:libust[20606/20606]: Error: Error cancelling local ust listener thread: No such process (in lttng_ust_exit() at lttng-ust-comm.c:1526)
From #ceph-devel chat:
sounds like a process died while someone was trying to collect data from it dmick @ 4:15 lttng is the tracing framework, for gathering execution data 4:16 gmeno yuriw: Gregorys-MacBook-Pro:ceph gmeno$ git blame ceph.spec.in | grep lttng gmeno @ 4:16 gmeno e267128c ceph.spec.in (Nathan Cutler 2015-07-15 12:54:20 +0200 169) # lttng and babeltrace for rbd-replay-prep gmeno @ 4:16
Updated by Yuri Weinstein over 8 years ago
- Release set to infernalis
- ceph-qa-suite upgrade/client-upgrade added
Updated by Yuri Weinstein over 8 years ago
- Project changed from teuthology to Ceph
Updated by Yuri Weinstein over 8 years ago
- Assignee set to Nathan Cutler
Nathan, can you take a look at this, pls?
Updated by Nathan Cutler over 8 years ago
Hi Yuri. This test failure is happening in Ubuntu 14.04, right? My commit cited in the description affects RPM distributions only.
Updated by Nathan Cutler over 8 years ago
(Not that I don't want to look at it, but I will have to do some digging/learning and that will take some time.)
Updated by Nathan Cutler over 8 years ago
A quick jab at google indicates that the error might be related to a missing LD_PRELOAD=liblttng-ust-fork.so.0
- maybe this is a regression?
Upgrading from v0.87.x Giant ---------------------------- * librbd and librados include lttng tracepoints on distros with liblttng 2.4 or later (only Ubuntu Trusty for the ceph.com packages). When running a daemon that uses these libraries, i.e. an application that calls fork(2) or clone(2) without exec(3), you must set LD_PRELOAD=liblttng-ust-fork.so.0 to prevent a crash in the lttng atexit handler when the process exits. The only ceph tool that requires this is rbd-fuse.
Updated by Nathan Cutler over 8 years ago
It would seem that this commit has stopped working for some reason:
commit c4939f9aeae73089ab82c1783d7afd9baafd784f Author: Adam Crume <adamcrume@gmail.com> Date: Mon Jul 28 16:32:15 2014 -0700 lttng: Preload liblttng-ust-fork.so in TESTS_ENVIRONMENT This adds LD_PRELOAD=liblttng-ust-fork.so to TESTS_ENVIRONMENT. This prevents lttng from complaining when processes are forked. The complaints otherwise taint the output and cause tests to fail. Signed-off-by: Adam Crume <adamcrume@gmail.com>
Updated by Loïc Dachary over 8 years ago
http://tracker.ceph.com/issues/13274 seems related
Updated by Loïc Dachary over 8 years ago
Something happened, somewhere, after http://pulpito.ceph.com/teuthology-2015-10-05_16:05:09-upgrade:hammer-hammer-distro-basic-vps/ : noone of the teuthology.log show the error above.
It does not come from the ceph repository:
git log --merges --format='%H' 79385a85beea9bccd82c99b6bda653f0224c4fcd..ceph/hammer | while read sha1 ; do echo ; git log --format='%aD "* %s":https://github.com/ceph/ceph/commit/%H' ${sha1}^1..${sha1} ; done | cat Thu, 8 Oct 2015 00:09:24 +0200 "* Merge pull request #6192 from dachary/wip-13401-hammer":https://github.com/ceph/ceph/commit/70e612a395f7f5daf2fa289d8ae522824ef7a38f Fri, 21 Aug 2015 16:40:34 -0400 "* crush/CrushTester: test fewer inputs when running crushtool":https://github.com/ceph/ceph/commit/2a28114053fbf4ceee09558790ad3cf755270d5f Sun, 20 Sep 2015 23:42:45 +0200 "* tests: update to match crushmap validation message":https://github.com/ceph/ceph/commit/abc5b5f139b05da0be0fbaa99282be16386980db Fri, 18 Sep 2015 09:42:47 -0400 "* mon/OSDMonitor: fix crush injection error message":https://github.com/ceph/ceph/commit/25bd2778f8b7f5eb5245efebca56a7348dd064eb Fri, 18 Sep 2015 09:41:25 -0400 "* mon/OSDMonitor: only test crush ruleset for the newly created pool":https://github.com/ceph/ceph/commit/6635530aa94b1f21c3052e64d4dfe3952f98fe83 Fri, 18 Sep 2015 09:40:13 -0400 "* crush/CrushTester: allow testing by ruleset":https://github.com/ceph/ceph/commit/cc1fedda02fbf3c95d7a00f0b395db5439e96f90 Tue, 6 Oct 2015 11:11:14 -0400 "* Merge pull request #5887 from Abhishekvrshny/wip-13044-hammer":https://github.com/ceph/ceph/commit/065086246290cfa02f67c37c95a2ccdb05872f21 Wed, 5 Aug 2015 15:19:13 +0800 "* crypto: fix unbalanced ceph::crypto::init/ceph::crypto:shutdown":https://github.com/ceph/ceph/commit/3a50b904a2ffd613b695ead1b26c93278044d7df
nor ceph-cm-ansible
ceph/ceph-cm-ansible$ git log --since 2015-10-01 --merges --format='%H' master | while read sha1 ; do echo ; git log --format='%aD "* %s":https://github.com/ceph/ceph/commit/%H' ${sha1}^1..${sha1} ; done | cat Tue, 6 Oct 2015 11:23:35 -0500 "* Merge pull request #139 from ceph/wip-dban":https://github.com/ceph/ceph/commit/107f3c46e31f31132622b1f691b0e741e5a61b23 Tue, 22 Sep 2015 13:25:25 -0600 "* Add dummy dban-2.3.0 entry":https://github.com/ceph/ceph/commit/f28b827b3d1412bf70691eb54c7a6e12dd040d86 Tue, 6 Oct 2015 11:23:06 -0500 "* Merge pull request #148 from ceph/wip-teuthology":https://github.com/ceph/ceph/commit/ea87b162c49d45ff16f044369efda7210cb1b295 Fri, 2 Oct 2015 16:35:20 -0600 "* teuthology: Serve logs over http":https://github.com/ceph/ceph/commit/f3d1d93a22e627d1a08dedc1d33aebe0fe8fd1e2 Fri, 2 Oct 2015 11:50:44 -0500 "* Merge pull request #147 from ceph/wip-teuthology":https://github.com/ceph/ceph/commit/d5eae5d035b9a24f5786193b6a9e9e5055e0c59d Tue, 29 Sep 2015 13:45:03 -0600 "* teuthology: manage teuthology system users":https://github.com/ceph/ceph/commit/6c63105da1d9a3c99e8fd75504ef7464dcbb6fb8 Fri, 2 Oct 2015 09:13:50 -0500 "* Merge pull request #138 from ceph/wip-puddle-updates":https://github.com/ceph/ceph/commit/ec9f2c4ffac6aa4070cc67b7f69020689249193b Mon, 21 Sep 2015 12:17:05 -0600 "* puddle: new RH-COMMON advisory undergoing testing":https://github.com/ceph/ceph/commit/8a71cf9362d3506c51389f03c80b0699595becab Fri, 18 Sep 2015 16:43:47 -0600 "* puddle: output RHCEPH 1.3 to separate locations":https://github.com/ceph/ceph/commit/bfc268d7d11ce4757f44a8729830dff9ecb93df9 Thu, 1 Oct 2015 17:46:24 -0500 "* Merge pull request #146 from ceph/wip-paddles":https://github.com/ceph/ceph/commit/5482c8a1d7313f07ac3942378a4984e43a7acfeb Thu, 1 Oct 2015 10:53:40 -0600 "* pulpito: Restart if the config changes":https://github.com/ceph/ceph/commit/e9732daf99495f6f950ab7d1b081f4db5bdbafa2 Wed, 30 Sep 2015 16:38:49 -0600 "* paddles: Add new paddles role":https://github.com/ceph/ceph/commit/06cf08ddb74dd276790153a0bf7c10e2c5767a2d Thu, 1 Oct 2015 10:36:58 -0600 "* Merge pull request #145 from ceph/wip-hostname":https://github.com/ceph/ceph/commit/989b9d54dd285ff09a6f64f8f26b03241231fbb5 Wed, 30 Sep 2015 16:27:58 -0400 "* Update documentation for lab_domain var":https://github.com/ceph/ceph/commit/91273540ddc760b8d3e24026e716edd7ffd4c204 Wed, 30 Sep 2015 15:11:26 -0400 "* testnode: set_hostname should be a global task":https://github.com/ceph/ceph/commit/ae09ced8721f635138dad6c27be7b9a53deda5f0
Updated by Loïc Dachary over 8 years ago
teuthology-suite --verbose --machine-type vps --suite upgrade/hammer --suite-branch hammer --filter="upgrade:hammer/older/{0-cluster/start.yaml 1-install/v0.94.yaml 2-workload/testrados.yaml 3-upgrade-sequence/upgrade-osd-mon-mds.yaml 4-final/{monthrash.yaml osdthrash.yaml testrados.yaml} distros/ubuntu_14.04.yaml}" --ceph hammer --priority 101
That demonstrate this failure does not exist when running tests on the hammer branch.
Updated by Nathan Cutler over 8 years ago
This warning from #13430 is interesting: 2015-10-08T11:24:58.470 INFO:tasks.workunit.client.0.plana30.stderr:ERROR: ld.so: object 'liblttng-ust-fork.so.0' from LD_PRELOAD cannot be preloaded: ignored.
Updated by Nathan Cutler over 8 years ago
Updated by Samuel Just over 8 years ago
- Status changed from New to Can't reproduce
Updated by Yuri Weinstein over 8 years ago
- Status changed from Can't reproduce to New