General

Profile

Brad Hubbard's activity

From 11/06/2019 to 12/05/2019

12/05/2019

11:34 PM Messengers Bug #40026 (Duplicate): segfaults in ProtocolV2::write_message()
Brad Hubbard
11:28 PM RADOS Bug #41240 (Can't reproduce): All of the cluster SSDs aborted at around the same time and will no...
Brad Hubbard
11:26 PM mgr Bug #39955 (Need More Info): After upgrade to Nautilus 14.2.1 mon DB is growing too fast when sta...
Brad Hubbard
11:24 PM RADOS Bug #38892 (Closed): /ceph/src/tools/kvstore_tool.cc:266:1: internal compiler error: Segmentation...
Brad Hubbard

12/03/2019

02:53 AM RADOS Bug #38345: mon: segv in MonOpRequest::~MonOpRequest OpHistory::cleanup
I think we can dispense with the session put when we call 'remove_session' since we call it when we replace the sessi... Brad Hubbard

11/28/2019

03:06 AM Ceph Documentation #43049 (New): Allow User Contributed Notes in the documentation
I think it would serve us well to allow these as per https://www.php.net/manual/en/install.unix.php (see bottom of pa... Brad Hubbard

11/27/2019

11:40 PM mgr Bug #38539: tasks.mgr.test_insights.TestInsights#test_health_history fails
This sort of a race really should not be happening in a single threaded module like insights.
I've added the follo...
Brad Hubbard

11/26/2019

05:18 AM RADOS Bug #38345: mon: segv in MonOpRequest::~MonOpRequest OpHistory::cleanup
Theory:
In Monitor::_ms_dispatch() when we detect a feature change we end up with the following sequence.
Monit...
Brad Hubbard
03:45 AM RADOS Bug #38345: mon: segv in MonOpRequest::~MonOpRequest OpHistory::cleanup
... Brad Hubbard
03:20 AM RADOS Bug #38345: mon: segv in MonOpRequest::~MonOpRequest OpHistory::cleanup
I'm wondering if maybe this happens due to the feature change and the session being removed during the upgrade proces... Brad Hubbard

11/25/2019

06:56 AM RADOS Bug #38345: mon: segv in MonOpRequest::~MonOpRequest OpHistory::cleanup
I'm interested to see more cores in case one sheds more light. I've started a few runs in the hope they will fail. Brad Hubbard
05:14 AM RADOS Bug #38345: mon: segv in MonOpRequest::~MonOpRequest OpHistory::cleanup
In both cases so far the Message type appears to be MSG_MON_PAXOS and priority is CEPH_MSG_PRIO_HIGH.... Brad Hubbard
03:40 AM RADOS Bug #38345: mon: segv in MonOpRequest::~MonOpRequest OpHistory::cleanup
Neha sent me another instance of this issue available at http://pulpito.ceph.com/nojha-2019-11-22_18:41:03-rados:upgr... Brad Hubbard
12:58 AM RADOS Bug #38345: mon: segv in MonOpRequest::~MonOpRequest OpHistory::cleanup
... Brad Hubbard

11/24/2019

11:01 PM rgw Bug #42983 (Can't reproduce): double free or corruption in radosgw-admin during rgw testing
http://qa-proxy.ceph.com/teuthology/yuvalif-2019-11-23_19:32:07-rgw:verify-wip-yuval-add-kafka-security2-distro-basic... Brad Hubbard

11/22/2019

09:14 PM mgr Bug #38539: tasks.mgr.test_insights.TestInsights#test_health_history fails
I can reproduce this and am trying to track it down. Brad Hubbard
11:22 AM RADOS Bug #42477: Rados should use the '-o outfile' convention
@Nathan, I think that's the right decision in this case mate. It should be less disruptive hopefully. Brad Hubbard

11/21/2019

02:18 AM mgr Feature #41361: CephInsights - Lacking Sufficient Performance Counters
Actually, I think the data you need is already captured.... Brad Hubbard

11/20/2019

05:42 AM mgr Feature #41361: CephInsights - Lacking Sufficient Performance Counters
Hi Alex,
https://github.com/ceph/ceph/pull/25703 added a warning for bluefs spillover and that could be backported...
Brad Hubbard

11/18/2019

09:13 PM RADOS Bug #42387 (In Progress): ceph_test_admin_socket_output fails in rados qa suite
Brad Hubbard

11/17/2019

10:27 PM RADOS Bug #42477: Rados should use the '-o outfile' convention
Note that the reason I did not set this for backport is that it has the potential to break existing scripts and funct... Brad Hubbard

11/15/2019

05:57 AM mgr Bug #39955: After upgrade to Nautilus 14.2.1 mon DB is growing too fast when state of cluster is ...
Noah Watkins wrote:
> While we don't track much data in the insights module, it looks like the issue here is that th...
Brad Hubbard
05:39 AM RADOS Bug #42824: mimic: rebuild_mondb.cc: FAILED assert(0) in update_osdmap()
... Brad Hubbard

11/14/2019

05:07 AM Ceph Revision 4dd67b9e (ceph): common/admin_socket: Increase socket timeouts
With the move of the 'bench' command to the admin socket the recv
timeout is being exceeded in testing due to the dur...
Brad Hubbard
04:54 AM RADOS Bug #42387: ceph_test_admin_socket_output fails in rados qa suite
As suspected this is due to the length of time the 'bench' command takes to run and return a result to the socket.
...
Brad Hubbard

11/13/2019

10:07 PM RADOS Backport #42242 (Resolved): nautilus: Adding Placement Group id in Large omap log message
Brad Hubbard
07:59 AM RADOS Bug #42387: ceph_test_admin_socket_output fails in rados qa suite
It's only the bench command that causes the issue.... Brad Hubbard

11/12/2019

02:46 AM Ceph Bug #42758 (Resolved): run-tox-ceph-daemon fails with "SyntaxError: (unicode error)"
Brad Hubbard
02:43 AM Ceph Bug #42758 (Resolved): run-tox-ceph-daemon fails with "SyntaxError: (unicode error)"
... Brad Hubbard

11/11/2019

11:20 PM Ceph Revision d2b80607 (ceph): tools/rados: Unmask '-o' to restore original behaviour
0b369e1aff1 masked the original behaviour of '-o' which was to indicate
'outfile' as documented in the man page. Chan...
Brad Hubbard

11/07/2019

11:53 PM mgr Bug #42680: crash in in thread 7f6a445ee700 thread_name:devicehealth
I think this is more likely to be a duplicate of https://tracker.ceph.com/issues/42082 Brad Hubbard
 

Also available in: Atom