Description of problem: During valgrind test run following leaks were seen, John mentioned that there are known issues with upstream jewel as well and tracking this for downstream once its fixed, I will link it to upstream tracker if there is one. 2017-02-07T03:08:06.594 INFO:teuthology.orchestra.run.clara004:Running: "sudo zgrep '<kind>' /var/log/ceph/valgrind/* /dev/null | sort | uniq" 2017-02-07T03:08:06.600 INFO:teuthology.orchestra.run.clara003:Running: "sudo zgrep '<kind>' /var/log/ceph/valgrind/* /dev/null | sort | uniq" 2017-02-07T03:08:06.689 INFO:teuthology.orchestra.run.clara004.stdout:/var/log/ceph/valgrind/client.0.log: <kind>Leak_StillReachable</kind> 2017-02-07T03:08:06.690 INFO:teuthology.orchestra.run.clara004.stdout:/var/log/ceph/valgrind/mds.a.log: <kind>Leak_DefinitelyLost</kind> 2017-02-07T03:08:06.690 INFO:teuthology.orchestra.run.clara004.stdout:/var/log/ceph/valgrind/mon.a.log: <kind>Leak_StillReachable</kind> 2017-02-07T03:08:06.690 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/client.0.log kind <kind>Leak_StillReachable</kind> 2017-02-07T03:08:06.691 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/client.0.log 2017-02-07T03:08:06.691 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mds.a.log kind <kind>Leak_DefinitelyLost</kind> 2017-02-07T03:08:06.691 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mon.a.log kind <kind>Leak_StillReachable</kind> 2017-02-07T03:08:06.691 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/mon.a.log 2017-02-07T03:08:06.697 INFO:teuthology.orchestra.run.clara003.stdout:/var/log/ceph/valgrind/mds.a-s.log: <kind>Leak_DefinitelyLost</kind> 2017-02-07T03:08:06.697 INFO:teuthology.orchestra.run.clara003.stdout:/var/log/ceph/valgrind/mon.b.log: <kind>Leak_StillReachable</kind> 2017-02-07T03:08:06.698 INFO:teuthology.orchestra.run.clara003.stdout:/var/log/ceph/valgrind/mon.c.log: <kind>Leak_StillReachable</kind> 2017-02-07T03:08:06.698 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mds.a-s.log kind <kind>Leak_DefinitelyLost</kind> 2017-02-07T03:08:06.698 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mon.b.log kind <kind>Leak_StillReachable</kind> 2017-02-07T03:08:06.699 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/mon.b.log 2017-02-07T03:08:06.699 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mon.c.log kind <kind>Leak_StillReachable</kind> 2017-02-07T03:08:06.699 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/mon.c.log Version-Release number of selected component (if applicable): 10.2.5-22.el7cp How reproducible: 1/1 Logs: http://magna002.ceph.redhat.com/vasu-2017-02-06_22:49:33-fs-jewel---basic-multi/264408/teuthology.log
Kefu, Do you want to look at the mon leak issue seen during fs or rgw runs, I know there are few known issues but trying to distinguish new from old is becoming difficult with failed cases, Let me know if you need a different bz to track mon issue. http://magna002.ceph.redhat.com/vasu-2017-02-08_16:40:39-rgw-jewel---basic-multi/264717/teuthology.log 2017-02-08T18:19:10.864 INFO:teuthology.orchestra.run.pluto010:Running: "sudo zgrep '<kind>' /var/log/ceph/valgrind/* /dev/null | sort | uniq" 2017-02-08T18:19:10.869 INFO:teuthology.orchestra.run.pluto008:Running: "sudo zgrep '<kind>' /var/log/ceph/valgrind/* /dev/null | sort | uniq" 2017-02-08T18:19:10.940 INFO:teuthology.orchestra.run.pluto008.stdout:/var/log/ceph/valgrind/mon.b.log: <kind>Leak_StillReachable</kind> 2017-02-08T18:19:10.942 INFO:teuthology.orchestra.run.pluto010.stdout:/var/log/ceph/valgrind/mon.a.log: <kind>Leak_StillReachable</kind> 2017-02-08T18:19:10.942 INFO:teuthology.orchestra.run.pluto010.stdout:/var/log/ceph/valgrind/mon.c.log: <kind>Leak_StillReachable</kind> 2017-02-08T18:19:10.942 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mon.a.log kind <kind>Leak_StillReachable</kind> 2017-02-08T18:19:10.943 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/mon.a.log 2017-02-08T18:19:10.943 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mon.c.log kind <kind>Leak_StillReachable</kind> 2017-02-08T18:19:10.943 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/mon.c.log 2017-02-08T18:19:10.943 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mon.b.log kind <kind>Leak_StillReachable</kind> 2017-02-08T18:19:10.944 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/mon.b.log
Vasu, the leak reports of mon are the same issue. no need to create a different bz for them. i will take a look at it later on.
but by inspecting the back trace, the mds leak is a different one.
I have separated monitor leak to a cloned bz, still seeing mds leaks 2017-05-04T00:41:33.643 INFO:teuthology.orchestra.run.clara002:Running: "sudo zgrep '<kind>' /var/log/ceph/valgrind/* /dev/null | sort | uniq" 2017-05-04T00:41:33.648 INFO:teuthology.orchestra.run.pluto003:Running: "sudo zgrep '<kind>' /var/log/ceph/valgrind/* /dev/null | sort | uniq" 2017-05-04T00:41:33.725 INFO:teuthology.orchestra.run.pluto003.stdout:/var/log/ceph/valgrind/mds.a-s.log: <kind>Leak_DefinitelyLost</kind> 2017-05-04T00:41:33.725 INFO:teuthology.orchestra.run.pluto003.stdout:/var/log/ceph/valgrind/mon.b.log: <kind>Leak_StillReachable</kind> 2017-05-04T00:41:33.725 INFO:teuthology.orchestra.run.pluto003.stdout:/var/log/ceph/valgrind/mon.c.log: <kind>Leak_StillReachable</kind> 2017-05-04T00:41:33.738 INFO:teuthology.orchestra.run.clara002.stdout:/var/log/ceph/valgrind/client.0.log: <kind>Leak_StillReachable</kind> 2017-05-04T00:41:33.739 INFO:teuthology.orchestra.run.clara002.stdout:/var/log/ceph/valgrind/mds.a.log: <kind>Leak_DefinitelyLost</kind> 2017-05-04T00:41:33.739 INFO:teuthology.orchestra.run.clara002.stdout:/var/log/ceph/valgrind/mon.a.log: <kind>Leak_StillReachable</kind> 2017-05-04T00:41:33.739 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/client.0.log kind <kind>Leak_StillReachable</kind> 2017-05-04T00:41:33.740 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/client.0.log 2017-05-04T00:41:33.740 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mds.a.log kind <kind>Leak_DefinitelyLost</kind> 2017-05-04T00:41:33.740 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mon.a.log kind <kind>Leak_StillReachable</kind> 2017-05-04T00:41:33.740 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/mon.a.log 2017-05-04T00:41:33.740 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mds.a-s.log kind <kind>Leak_DefinitelyLost</kind> 2017-05-04T00:41:33.740 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mon.b.log kind <kind>Leak_StillReachable</kind> 2017-05-04T00:41:33.741 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/mon.b.log 2017-05-04T00:41:33.741 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mon.c.log kind <kind>Leak_StillReachable</kind> 2017-05-04T00:41:33.741 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/mon.c.log http://magna002.ceph.redhat.com/vasu-2017-05-03_20:52:32-fs-jewel---basic-multi/268296/teuthology.log
Hello Vasu, I'm the new PTL for CephFS. Can you tell me where I can find the valgrind logs for these leaks?
This needs to be re-tested. The logs have been lost.
Hi Patrick, Sorry I missed your comment, I will rerun this on new build and update the logs link here. I could recreate this multiple times.
Moving this to 3.1. We can't proceed without a log.
Seeing this in new regression runs at http://magna002.ceph.redhat.com/vasu-2017-10-18_22:32:22-fs-luminous---basic-multi/278194/teuthology.log 2017-10-19T23:21:37.300 INFO:teuthology.orchestra.run.clara013.stdout:/var/log/ceph/valgrind/mds.a-s.log: <kind>Leak_DefinitelyLost</kind> 2017-10-19T23:21:37.300 INFO:teuthology.orchestra.run.clara013.stdout:/var/log/ceph/valgrind/mds.a-s.log: <kind>Leak_PossiblyLost</kind> 2017-10-19T23:21:37.301 INFO:teuthology.orchestra.run.clara013.stdout:/var/log/ceph/valgrind/mon.b.log: <kind>Leak_StillReachable</kind> 2017-10-19T23:21:37.301 INFO:teuthology.orchestra.run.clara013.stdout:/var/log/ceph/valgrind/mon.c.log: <kind>Leak_StillReachable</kind> 2017-10-19T23:21:37.301 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mds.a-s.log kind <kind>Leak_DefinitelyLost</kind> 2017-10-19T23:21:37.301 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mds.a-s.log kind <kind>Leak_PossiblyLost</kind> 2017-10-19T23:21:37.301 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mon.b.log kind <kind>Leak_StillReachable</kind> 2017-10-19T23:21:37.302 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/mon.b.log 2017-10-19T23:21:37.302 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mon.c.log kind <kind>Leak_StillReachable</kind> 2017-10-19T23:21:37.302 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/mon.c.log 2017-10-19T23:21:37.304 INFO:teuthology.orchestra.run.clara012.stdout:/var/log/ceph/valgrind/client.0.log: <kind>Leak_StillReachable</kind> 2017-10-19T23:21:37.305 INFO:teuthology.orchestra.run.clara012.stdout:/var/log/ceph/valgrind/mds.a.log: <kind>Leak_DefinitelyLost</kind> 2017-10-19T23:21:37.305 INFO:teuthology.orchestra.run.clara012.stdout:/var/log/ceph/valgrind/mds.a.log: <kind>Leak_PossiblyLost</kind> 2017-10-19T23:21:37.305 INFO:teuthology.orchestra.run.clara012.stdout:/var/log/ceph/valgrind/mon.a.log: <kind>Leak_StillReachable</kind> 2017-10-19T23:21:37.305 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/client.0.log kind <kind>Leak_StillReachable</kind> 2017-10-19T23:21:37.305 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/client.0.log 2017-10-19T23:21:37.305 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mds.a.log kind <kind>Leak_DefinitelyLost</kind> 2017-10-19T23:21:37.306 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mds.a.log kind <kind>Leak_PossiblyLost</kind> 2017-10-19T23:21:37.306 DEBUG:tasks.ceph:file /var/log/ceph/valgrind/mon.a.log kind <kind>Leak_StillReachable</kind> 2017-10-19T23:21:37.306 ERROR:tasks.ceph:saw valgrind issue <kind>Leak_StillReachable</kind> in /var/log/ceph/valgrind/mon.a.log
Patrick, New logs with 12.2.4-4 build http://magna002.ceph.redhat.com/vasu-2018-03-23_18:08:23-fs-luminous-distro-basic-multi/292943/teuthology.log