Project

General

Profile

Activity

From 09/13/2017 to 10/12/2017

10/12/2017

02:39 PM Bug #21773 (Resolved): restarting active ceph-mgr cause glitches in bps and iops metrics
after doing restart of active ceph-mgr:... Aleksei Gutikov
11:00 AM Bug #20629 (Fix Under Review): Spurious ceph-mgr failovers during mon elections
https://github.com/ceph/ceph/pull/18268 John Spray
05:47 AM Bug #21707: "osd status" command exception if OSD not in pgmap stats
https://github.com/ceph/ceph/pull/18173 Yanhu Cao
12:51 AM Backport #21699 (In Progress): luminous: mgr status module uses base 10 units
Nathan Cutler

10/11/2017

10:22 AM Bug #21752 (Resolved): ceph.in: ceph fs status returns error
Merged, and noted on http://tracker.ceph.com/issues/21699 to pull in the patch rather than creating a separate backpo... John Spray
06:07 AM Bug #21752 (Fix Under Review): ceph.in: ceph fs status returns error
Kefu Chai
01:47 AM Bug #21752: ceph.in: ceph fs status returns error
fixed https://github.com/ceph/ceph/pull/18233 Yanhu Cao
01:38 AM Bug #21752: ceph.in: ceph fs status returns error
see https://github.com/ceph/ceph/pull/17380... Yanhu Cao
10:22 AM Backport #21699: luminous: mgr status module uses base 10 units
When backporting, please also pull in commit from https://github.com/ceph/ceph/pull/18233, which fixes a mistake from... John Spray

10/10/2017

05:21 PM Bug #21752: ceph.in: ceph fs status returns error
John, can you take a look? Patrick Donnelly
05:21 PM Bug #21752 (Resolved): ceph.in: ceph fs status returns error
... Patrick Donnelly

10/09/2017

07:36 AM Feature #17454 (Fix Under Review): Don't force module classes to be called "Module"
https://github.com/ceph/ceph/pull/18526 Kefu Chai

10/08/2017

10:04 PM Feature #17454: Don't force module classes to be called "Module"
I am working on this issue Bhavishya Desai

10/07/2017

09:04 AM Bug #21687: mgr: mark_down of osd without metadata is broken
Maybe we could request OSD's metadata when We kill a OSD connection. how about creating a background thread and reque... Chang Liu

10/06/2017

04:53 PM Bug #21707 (Resolved): "osd status" command exception if OSD not in pgmap stats
... John Spray
10:03 AM Bug #21687: mgr: mark_down of osd without metadata is broken
John Spray wrote:
> Ah -- didn't think about how the OSD conns were used for commands when this change (https://gith...
Chang Liu
09:34 AM Bug #21687: mgr: mark_down of osd without metadata is broken
Ah -- didn't think about how the OSD conns were used for commands when this change (https://github.com/ceph/ceph/pull... John Spray
02:43 AM Bug #21687 (Resolved): mgr: mark_down of osd without metadata is broken
... Sage Weil
03:19 AM Backport #21699 (Resolved): luminous: mgr status module uses base 10 units
https://github.com/ceph/ceph/pull/18257 Nathan Cutler

10/05/2017

06:32 PM Feature #21683 (Resolved): After "ceph mgr module enable", new module's commands should be usable...
Currently, commands for a newly enabled module will appear missing until the mgr daemon has come back up after restar... John Spray
06:27 PM Feature #21682 (Resolved): mgr should raise health alert when a module throws exceptions (post-load)
We have cluster log messages when module startup fails, and we have intention to do the can_run flag for missing deps... John Spray

10/04/2017

10:32 PM Bug #21612: mgr: thread create assert failure when racing with respawn
Sage Weil
10:32 PM Bug #21612: mgr: thread create assert failure when racing with respawn
this looks like a race between thread creates and respawn()... Sage Weil
03:14 PM Bug #21189 (Pending Backport): mgr status module uses base 10 units
Kefu Chai
01:44 PM Bug #21593: segv in PyList_New from PyFormatter
This has obstinately refused to reproduce on my local vstart system, so I've built a branch with the new assertions a... John Spray

10/03/2017

01:30 PM Bug #21593: segv in PyList_New from PyFormatter
Spun off https://github.com/ceph/ceph/pull/18093 while trying to get to bottom of this John Spray
12:24 PM Bug #21399: ceph-mgr module(s) inaccessible after a reboot
Sorry for the delay. I have been playing with this a bit. I am not saying it is related to the executable name being ... Boris Ranto
02:59 AM Backport #21659 (Resolved): luminous: Crash in get_metadata_python during MDS restart
https://github.com/ceph/ceph/pull/18412 Nathan Cutler
02:58 AM Backport #21656 (Resolved): luminous: crash on DaemonPerfCounters::update
https://github.com/ceph/ceph/pull/18675
Nathan Cutler
02:58 AM Backport #21648 (Resolved): luminous: mgr[zabbix] float division by zero
https://github.com/ceph/ceph/pull/18734 Nathan Cutler
02:57 AM Backport #21638 (Resolved): luminous: dashboard OSD list has servers and osds in arbitrary order
Nathan Cutler

10/02/2017

03:34 PM Bug #21593 (In Progress): segv in PyList_New from PyFormatter
John Spray
02:34 AM Bug #21518 (Pending Backport): mgr[zabbix] float division by zero
Sage Weil

10/01/2017

08:56 PM Backport #21452 (Resolved): luminous: prometheus module generates invalid output when counter nam...
Sage Weil
08:56 PM Backport #21443 (Resolved): luminous: Prometheus crash when update
Sage Weil
05:17 PM Bug #21612 (New): mgr: thread create assert failure when racing with respawn
Run: http://pulpito.ceph.com/teuthology-2017-09-26_02:30:02-rados-luminous-distro-basic-smithi/
Job: 1673575
Logs: ...
Yuri Weinstein

09/29/2017

11:06 AM Bug #21599 (Fix Under Review): List of filesystems does not get refreshed after a filesystem dele...
https://github.com/ceph/ceph/pull/18039 John Spray
10:37 AM Bug #21599: List of filesystems does not get refreshed after a filesystem deletion
Never mind, this was easy to reproduce locally. When a filesystem is deleted, ceph-mgr is somehow still holding on t... John Spray
10:29 AM Bug #21599: List of filesystems does not get refreshed after a filesystem deletion
Hmm, the list in the menu is supposed to be updated every 5 seconds.
If you leave the console (the right click->in...
John Spray

09/28/2017

10:56 PM Bug #21599 (Resolved): List of filesystems does not get refreshed after a filesystem deletion
Just tested 12.2.1 and was playing with multiple filesystems (CephFS), creating some and deleting some. After I have ... Jean-Charles Lopez
06:30 PM Bug #21598 (Resolved): Users can do "config-key set" while mgr runs, but it doesn't see settings
This is kind of confusing for users and for module developers.
Originally the config-key stuff was meant to just b...
John Spray
04:22 PM Bug #20629: Spurious ceph-mgr failovers during mon elections
Seen here:
http://pulpito.ceph.com/dzafman-2017-09-27_09:14:10-rados-wip-no-lock-distro-basic-smithi/1679370/
David Zafman
02:55 PM Feature #21594 (Fix Under Review): prometheus meta-series describing OSD<->disk mapping
https://github.com/ceph/ceph/pull/18021 John Spray
02:52 PM Feature #21594 (Resolved): prometheus meta-series describing OSD<->disk mapping
John Spray
02:35 PM Bug #21593: segv in PyList_New from PyFormatter
I'm triggering this with github.com/liewegas/ceph wip-balancer,... Sage Weil
02:34 PM Bug #21593 (Resolved): segv in PyList_New from PyFormatter
... Sage Weil
10:06 AM Bug #20420 (Can't reproduce): Crash in MgrClient::send_report->PerfCountersCollection::with_counters
DaemonState locking has been substantially fixed since this happened -- hopefully whatever the issue was got fixed as... John Spray
06:38 AM Dashboard Bug #21572 (Pending Backport): dashboard OSD list has servers and osds in arbitrary order
Kefu Chai

09/27/2017

02:16 PM Dashboard Bug #21570: dashboard barfs on nulls where it expects numbers
Oops, accidentally put this in fs project, sorry for spam. John Spray
01:37 PM Dashboard Bug #21570 (Fix Under Review): dashboard barfs on nulls where it expects numbers
https://github.com/ceph/ceph/pull/17991 John Spray
01:36 PM Dashboard Bug #21570 (Resolved): dashboard barfs on nulls where it expects numbers

In this particular instance it has nulls for the usage on one of the pools for some unknown reason.
The nulls sh...
John Spray
02:15 PM Dashboard Bug #21572 (Fix Under Review): dashboard OSD list has servers and osds in arbitrary order
https://github.com/ceph/ceph/pull/17993 John Spray
02:10 PM Dashboard Bug #21572 (Resolved): dashboard OSD list has servers and osds in arbitrary order
Doesn't tend to be noticeable on vstart clusters but is very apparent on real clusters.
John Spray

09/26/2017

01:36 PM Bug #21197 (Pending Backport): crash on DaemonPerfCounters::update
Sage Weil
01:36 PM Bug #17737 (Pending Backport): Crash in get_metadata_python during MDS restart
Sage Weil

09/25/2017

09:22 PM Backport #21549 (Resolved): luminous: the dashboard uses absolute links for filesystems and clients
Nathan Cutler
09:21 PM Backport #21547 (Resolved): luminous: ceph-mgr gets process called "exe" after respawn
https://github.com/ceph/ceph/pull/18738 Nathan Cutler

09/23/2017

05:23 PM Bug #17737 (Fix Under Review): Crash in get_metadata_python during MDS restart
https://github.com/ceph/ceph/pull/17933 John Spray
05:11 PM Bug #21197 (Fix Under Review): crash on DaemonPerfCounters::update
https://github.com/ceph/ceph/pull/17932
John Spray
03:53 PM Bug #21197 (In Progress): crash on DaemonPerfCounters::update
John Spray
03:49 PM Documentation #20257 (Closed): ceph-mgr doc updates
Closing this -- I'm very willing to add specific information to the docs, but this request is a bit too general for m... John Spray
03:48 PM Documentation #20257: ceph-mgr doc updates
> a) ceph-mgrs keys and bootstrap process and verify properly bootstraped
We have manual setup instructions, they ...
John Spray
03:29 PM Bug #21367 (Fix Under Review): 'ZabbixSender' object has no attribute 'hostname'
John Spray
03:24 PM Bug #20886 (Can't reproduce): Bad perf counters in ceph-mgr for MDS
John Spray
03:22 PM Bug #20568 (Pending Backport): the dashboard uses absolute links for filesystems and clients
John Spray
02:26 PM Bug #21518 (Fix Under Review): mgr[zabbix] float division by zero
By inspection, I'm assuming this was happening in the avg() function due to lack of data (why there was no data is a ... John Spray

09/22/2017

09:09 PM Bug #21404 (Pending Backport): ceph-mgr gets process called "exe" after respawn
Sage Weil
08:34 PM Backport #21524 (Resolved): luminous: DaemonState members accessed outside of locks
https://github.com/ceph/ceph/pull/18675 Nathan Cutler
08:06 PM Bug #21518: mgr[zabbix] float division by zero
After upgrading all OSDs to bluestore, the issue resolved.
One of OSD had filestore format while others had bluest...
Ilja Slepnev
07:47 PM Bug #21518 (Resolved): mgr[zabbix] float division by zero
Enabled Zabbix module, configured as described on http://docs.ceph.com/docs/master/mgr/zabbix/, but no data is receiv... Ilja Slepnev
02:44 PM Feature #21502: Enable mgr modules to report their "runnability"
Right, although now that I think about it, we could probably also catch ImportErrors from when we try to load the mod... John Spray
02:40 PM Feature #21502: Enable mgr modules to report their "runnability"
This them implies that no import of any "foreign module" may be done in the header of a Mgr Module but should always ... Wido den Hollander
02:25 PM Feature #21502 (Resolved): Enable mgr modules to report their "runnability"
If a module has a dependency which is missing, we should be able to learn that on a mgr (even in standby), so that it... John Spray
02:37 PM Bug #20396 (Can't reproduce): segv in python somewhere from restful test
John Spray
04:30 AM Bug #21158 (Pending Backport): DaemonState members accessed outside of locks
Kefu Chai

09/21/2017

07:43 PM Bug #20222 (Duplicate): v12.0.3 Luminous bluestore 'tp_osd_tp thread tp_osd_tp' had timed out aft...
Highly likely this is #21171; please try latest luminous branch or wait for 12.2.1 (out very very soon now!) Sage Weil
11:57 AM Bug #21122: Kraken to luminous upgrade: Error EINVAL: key for mgr.vpm037 exists but cap mds does ...
I have the same error when I migrate my cluster and create mgr Nicolas Drufin
04:15 AM Backport #21479 (In Progress): luminous: Services reported with blank hostname by mgr
Nathan Cutler
04:13 AM Backport #21452 (In Progress): luminous: prometheus module generates invalid output when counter ...
Nathan Cutler
04:11 AM Backport #21443 (In Progress): luminous: Prometheus crash when update
Nathan Cutler
04:09 AM Backport #21320 (In Progress): luminous: Quieten scary RuntimeError from restful module on startup
Nathan Cutler

09/20/2017

06:33 PM Bug #21399: ceph-mgr module(s) inaccessible after a reboot
The 'exe' thing is really just the process naming - even if there is an issue of sockets not getting torn down, that ... John Spray
01:16 PM Bug #21399: ceph-mgr module(s) inaccessible after a reboot
I think it might be related. If there are any issues with the module not being killed properly before respawn then th... Boris Ranto
01:40 PM Backport #21479 (Resolved): luminous: Services reported with blank hostname by mgr
https://github.com/ceph/ceph/pull/17869 Nathan Cutler
03:24 AM Bug #20887 (Pending Backport): Services reported with blank hostname by mgr
Kefu Chai
03:24 AM Bug #20887 (Resolved): Services reported with blank hostname by mgr
https://github.com/ceph/ceph/pull/17138 Kefu Chai

09/19/2017

11:37 AM Backport #21452 (Resolved): luminous: prometheus module generates invalid output when counter nam...
https://github.com/ceph/ceph/pull/17868 Nathan Cutler
11:37 AM Backport #21443 (Resolved): luminous: Prometheus crash when update
https://github.com/ceph/ceph/pull/17867 Nathan Cutler

09/18/2017

11:31 AM Bug #21356: ceph-mgr admin socket starts failing after many attempts to call nonexistent command
I think it was just the same " do_accept error: '(24) Too many open files'" message many times, right? John Spray
06:14 AM Bug #21356 (Need More Info): ceph-mgr admin socket starts failing after many attempts to call non...
Mark, what log messages were filling your disk? could you pastebin or just paste a sample of it? and what debug level... Kefu Chai
10:36 AM Bug #20899 (Pending Backport): prometheus module generates invalid output when counter names cont...
John Spray

09/15/2017

08:35 PM Bug #21260 (Resolved): ceph mgr versions shows active mgr as "Unknown"
Nathan Cutler
08:35 PM Backport #21342 (Resolved): luminous: ceph mgr versions shows active mgr as "Unknown"
Nathan Cutler
04:28 PM Bug #21404 (Fix Under Review): ceph-mgr gets process called "exe" after respawn
https://github.com/ceph/ceph/pull/17756 John Spray
04:21 PM Bug #21404 (Resolved): ceph-mgr gets process called "exe" after respawn

This is identical to http://tracker.ceph.com/issues/19291, as the respawn code was lifted from ceph-mds for ceph-mg...
John Spray
04:22 PM Bug #21399: ceph-mgr module(s) inaccessible after a reboot
Good to know it's reproducible, if I have the log from the existing instance where this is happening then that would ... John Spray
02:46 PM Bug #21399: ceph-mgr module(s) inaccessible after a reboot
John, this is reproducible, you just need to deploy a cluster with ansible and reboot the machine. For some reason, t... Boris Ranto
11:08 AM Bug #21399: ceph-mgr module(s) inaccessible after a reboot
Hmm, I can't see much from that log -- it looks like it's from a single run of the mgr rather than spanning the reboo... John Spray
10:59 AM Bug #21399 (Closed): ceph-mgr module(s) inaccessible after a reboot
I've enabled and secured the RESTful API and verified that I can access its web UI at https://<active-mgr-node-ip-add... Bara Ancincova
11:50 AM Bug #21253 (Pending Backport): Prometheus crash when update
Kefu Chai
07:10 AM Bug #21367: 'ZabbixSender' object has no attribute 'hostname'
*PR*
https://github.com/ceph/ceph/pull/17751
UEST
jiantao zhu
 

Also available in: Atom