Activity
From 04/03/2019 to 05/02/2019
05/02/2019
- 08:35 AM Backport #38877: luminous: mds: high debug logging with many subtrees is slow
- PR - https://github.com/ceph/ceph/pull/27679
- 06:24 AM Backport #39200 (In Progress): mimic: mds: we encountered "No space left on device" when moving h...
- https://github.com/ceph/ceph/pull/27917
- 04:02 AM Backport #39193 (In Progress): mimic: mds: crash during mds restart
- https://github.com/ceph/ceph/pull/27916
05/01/2019
- 08:26 PM Bug #39437 (Resolved): osd: PriorityCache.cc: 265: FAILED ceph_assert(mem_avail >= 0)
- 06:30 PM Bug #39438: workunit fails with EPERM during thrashing
- /ceph/teuthology-archive/pdonnell-2019-04-25_02:44:21-multimds-wip-pdonnell-testing-20190424.232741-distro-basic-smit...
- 09:58 AM Documentation #38729 (Resolved): doc: add LAZYIO
- 09:58 AM Backport #39051 (Resolved): nautilus: doc: add LAZYIO
- 12:58 AM Backport #39051 (In Progress): nautilus: doc: add LAZYIO
- 09:41 AM Documentation #39130 (Resolved): doc: add documentation for `fs set min_compat_client`
- 09:41 AM Backport #39176 (Resolved): nautilus: doc: add documentation for `fs set min_compat_client`
- 01:03 AM Backport #39176 (In Progress): nautilus: doc: add documentation for `fs set min_compat_client`
- 09:28 AM Bug #36384 (Resolved): src/osdc/Journaler.cc: 420: FAILED ceph_assert(!r)
- 09:28 AM Backport #38448 (Resolved): mimic: src/osdc/Journaler.cc: 420: FAILED ceph_assert(!r)
- 09:28 AM Bug #38518 (Resolved): qa: tasks.cephfs.test_misc.TestMisc.test_fs_new hangs because clients cann...
- 09:27 AM Backport #38542 (Resolved): mimic: qa: tasks.cephfs.test_misc.TestMisc.test_fs_new hangs because ...
- 09:27 AM Bug #38487 (Resolved): qa: "Loading libcephfs-jni: Failure!"
- 09:27 AM Backport #38544 (Resolved): mimic: qa: "Loading libcephfs-jni: Failure!"
- 09:26 AM Bug #38723 (Resolved): qa: tolerate longer heartbeat timeouts when using valgrind
- 09:26 AM Backport #38734 (Resolved): mimic: qa: tolerate longer heartbeat timeouts when using valgrind
- 09:26 AM Bug #38491 (Resolved): "log [WRN] : Health check failed: 1 clients failing to respond to capabili...
- 09:24 AM Backport #38670 (Resolved): mimic: "log [WRN] : Health check failed: 1 clients failing to respond...
- 09:24 AM Feature #11172 (Resolved): mds: inode filtering on 'dump cache' asok
- 09:23 AM Backport #38689 (Resolved): mimic: mds: inode filtering on 'dump cache' asok
- 01:21 AM Backport #39471 (In Progress): nautilus: Expose CephFS snapshot creation time to clients
04/30/2019
- 07:28 PM Feature #39098: mds: lock caching for asynchronous unlink
- Zheng Yan wrote:
>
> Yes, this can causes inconsistency. But it's not unique to link count. For example, one clien... - 02:46 PM Bug #39543 (Fix Under Review): cephfs-shell: df command does not always produce correct output
- 01:22 PM Bug #39543 (Resolved): cephfs-shell: df command does not always produce correct output
- Correct output is not produced in the following cases:
1] For non-existing files, there is not error message
2] Whe... - 02:45 PM Backport #39050 (In Progress): nautilus: ceph_volume_client: Too many arguments for "WriteOpCtx"
- 02:35 PM Backport #38876 (In Progress): nautilus: mds: high debug logging with many subtrees is slow
- 04:11 AM Backport #39222 (In Progress): nautilus: mds: behind on trimming and "[dentry] was purgeable but ...
- https://github.com/ceph/ceph/pull/27879
04/29/2019
- 08:41 PM Bug #39406 (Fix Under Review): ceph_volume_client: d_name needs to be converted to string before ...
- 08:40 PM Bug #39405 (Fix Under Review): ceph_volume_client: python program embedded in test_volume_client....
- 06:50 PM Bug #39526 (Fix Under Review): cephfs-shell: teuthology tests
- 06:44 PM Bug #39526 (Resolved): cephfs-shell: teuthology tests
- for:
* mkdir
* get
* put - 04:43 PM Backport #38448: mimic: src/osdc/Journaler.cc: 420: FAILED ceph_assert(!r)
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/26643
merged - 04:42 PM Backport #38542: mimic: qa: tasks.cephfs.test_misc.TestMisc.test_fs_new hangs because clients can...
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/26804
merged - 04:42 PM Backport #38544: mimic: qa: "Loading libcephfs-jni: Failure!"
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/26807
merged - 04:41 PM Backport #38734: mimic: qa: tolerate longer heartbeat timeouts when using valgrind
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/26963
merged - 04:41 PM Backport #38670: mimic: "log [WRN] : Health check failed: 1 clients failing to respond to capabil...
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/27023
merged - 04:40 PM Backport #38689: mimic: mds: inode filtering on 'dump cache' asok
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/27058
merged - 10:29 AM Backport #39211 (In Progress): nautilus: MDSTableServer.cc: 83: FAILED assert(version == tid)
- 10:19 AM Backport #39214 (In Progress): nautilus: mds: there is an assertion when calling Beacon::shutdown()
- 10:11 AM Backport #39232 (In Progress): nautilus: kclient: nofail option not supported
- 09:59 AM Backport #39473 (In Progress): nautilus: mds: fail to resolve snapshot name contains '_'
- 09:55 AM Bug #39511 (Rejected): Cannot remove CephFS snapshot with leading underscore (_)
- Due to a misconfigured cronjob I ended up with CephFS snapshots with leading _
rmdir failed with error "No such file... - 09:54 AM Backport #39468 (In Progress): luminous: There is no punctuation mark or blank between tid and c...
- 09:51 AM Backport #39469 (In Progress): mimic: There is no punctuation mark or blank between tid and clie...
- 09:48 AM Backport #39470 (In Progress): nautilus: There is no punctuation mark or blank between tid and c...
- 08:13 AM Backport #39209 (In Progress): nautilus: mds: mds_cap_revoke_eviction_timeout is not used to init...
- https://github.com/ceph/ceph/pull/27842
- 12:54 AM Backport #39208 (In Progress): luminous: mds: mds_cap_revoke_eviction_timeout is not used to init...
- https://github.com/ceph/ceph/pull/27840
04/28/2019
- 04:12 PM Bug #39510 (Resolved): test_volume_client: test_put_object_versioned is unreliable
- test_put_object_versioned in test_volume_client.py succeeds if it receives a CommandFailedError exception from the em...
04/26/2019
- 06:11 PM Bug #39507 (Fix Under Review): cephfs-shell: mkdir error for relative path
- 05:50 PM Bug #39507 (Resolved): cephfs-shell: mkdir error for relative path
- mkdir does not create directory when relative path is specified.
- 01:23 AM Backport #39198 (In Progress): luminous: mds: we encountered "No space left on device" when movin...
- https://github.com/ceph/ceph/pull/27801
04/25/2019
- 07:47 AM Backport #39430 (In Progress): nautilus: qa: test_sessionmap assumes simple messenger
- 07:45 AM Backport #39473 (Resolved): nautilus: mds: fail to resolve snapshot name contains '_'
- https://github.com/ceph/ceph/pull/27849
- 07:45 AM Backport #39472 (Resolved): mimic: mds: fail to resolve snapshot name contains '_'
- https://github.com/ceph/ceph/pull/28186
- 07:45 AM Backport #39471 (Resolved): nautilus: Expose CephFS snapshot creation time to clients
- https://github.com/ceph/ceph/pull/27901
- 07:44 AM Backport #39470 (Resolved): nautilus: There is no punctuation mark or blank between tid and clie...
- https://github.com/ceph/ceph/pull/27846
- 07:44 AM Backport #39469 (Resolved): mimic: There is no punctuation mark or blank between tid and client_...
- https://github.com/ceph/ceph/pull/27847
- 07:44 AM Backport #39468 (Resolved): luminous: There is no punctuation mark or blank between tid and clie...
- https://github.com/ceph/ceph/pull/27848
04/24/2019
- 07:16 PM Bug #39437 (Fix Under Review): osd: PriorityCache.cc: 265: FAILED ceph_assert(mem_avail >= 0)
- 06:34 PM Bug #39437: osd: PriorityCache.cc: 265: FAILED ceph_assert(mem_avail >= 0)
- https://github.com/ceph/ceph/pull/27763
- 03:42 PM Bug #39437 (In Progress): osd: PriorityCache.cc: 265: FAILED ceph_assert(mem_avail >= 0)
- Mark Nelson wrote:
> Patrick Donnelly wrote:
> > [...]
> >
> > /ceph/teuthology-archive/pdonnell-2019-04-17_06:0... - 03:14 PM Bug #39437: osd: PriorityCache.cc: 265: FAILED ceph_assert(mem_avail >= 0)
- Patrick Donnelly wrote:
> [...]
>
> /ceph/teuthology-archive/pdonnell-2019-04-17_06:07:08-multimds-wip-pdonnell-t... - 04:05 PM Feature #3244 (In Progress): qa: integrate Ganesha into teuthology testing to regularly exercise ...
- 03:00 PM Feature #39098: mds: lock caching for asynchronous unlink
- Zheng Yan wrote:
> Yes, this can causes inconsistency. But it's not unique to link count. For example, one client do... - 07:01 AM Feature #39098: mds: lock caching for asynchronous unlink
- Jeff Layton wrote:
> Zheng Yan wrote:
> > Jeff Layton wrote:
> > > Zheng Yan wrote:
> > > > Sorry, I mean we don'... - 02:33 PM Feature #38829 (Fix Under Review): cephfs-shell: add a "stat" command
- 08:14 AM Backport #39191 (In Progress): luminous: mds: crash during mds restart
- https://github.com/ceph/ceph/pull/27737
- 08:12 AM Backport #39199 (In Progress): nautilus: mds: we encountered "No space left on device" when movin...
- https://github.com/ceph/ceph/pull/27736
04/23/2019
- 08:22 PM Feature #39403 (Fix Under Review): pybind: add the lseek() function to pybind of cephfs
- 06:58 AM Feature #39403 (Resolved): pybind: add the lseek() function to pybind of cephfs
- cephfs pybind: add the lseek() function to pybind of cephfs
- 08:17 PM Bug #39436 (Fix Under Review): qa: upgrade task fails from mimic to master
- 05:07 PM Bug #39436 (Resolved): qa: upgrade task fails from mimic to master
- ...
- 05:53 PM Bug #39266 (Pending Backport): There is no punctuation mark or blank between tid and client_id i...
- 05:52 PM Bug #38832 (Pending Backport): mds: fail to resolve snapshot name contains '_'
- 05:51 PM Feature #38838 (Pending Backport): Expose CephFS snapshot creation time to clients
- 05:41 PM Bug #39438 (New): workunit fails with EPERM during thrashing
- ...
- 05:26 PM Bug #39437 (Resolved): osd: PriorityCache.cc: 265: FAILED ceph_assert(mem_avail >= 0)
- ...
- 05:16 PM Bug #39406: ceph_volume_client: d_name needs to be converted to string before using
- Also seen in QA:...
- 09:35 AM Bug #39406 (Resolved): ceph_volume_client: d_name needs to be converted to string before using
- The "d_name in DirEntry":https://github.com/ceph/ceph/blob/master/src/pybind/cephfs/cephfs.pyx#L723 is obtained by ce...
- 01:06 PM Backport #39430 (Resolved): nautilus: qa: test_sessionmap assumes simple messenger
- https://github.com/ceph/ceph/pull/27772
- 08:51 AM Bug #39405 (Resolved): ceph_volume_client: python program embedded in test_volume_client.py use p...
- These embedded program must use python3 instead.
- 07:35 AM Bug #39404 (Fix Under Review): cephfs-shell: fix string decode for ls command
- 07:33 AM Bug #39404: cephfs-shell: fix string decode for ls command
- Updated with PR ID
- 07:18 AM Bug #39404 (Resolved): cephfs-shell: fix string decode for ls command
- CephFS:~/>>> ls -l
startswith first arg must be bytes or a tuple of bytes, not str
Traceback (most recent call last... - 06:03 AM Backport #39192 (In Progress): nautilus: mds: crash during mds restart
- https://github.com/ceph/ceph/pull/27714
04/22/2019
- 10:03 PM Bug #39329: ceph fs set <xxxxx> min_compat_client <yyyy> DOES NOT warn about connected incompatib...
- I would add the flag to both these commands.
- 09:43 PM Bug #39329: ceph fs set <xxxxx> min_compat_client <yyyy> DOES NOT warn about connected incompatib...
- Марк Коренберг wrote:
> The command ALREADY HAS this flag, but it does destructive actions without requiring it.
... - 09:37 PM Bug #39329: ceph fs set <xxxxx> min_compat_client <yyyy> DOES NOT warn about connected incompatib...
- The command ALREADY HAS this flag, but it does destructive actions without requiring it.
- 09:15 PM Bug #39329: ceph fs set <xxxxx> min_compat_client <yyyy> DOES NOT warn about connected incompatib...
- Sorry, what's the ask here? You want there to be a `--yes-i-really-mean-it` flag?
- 06:50 PM Bug #38803 (Pending Backport): qa: test_sessionmap assumes simple messenger
- Because of the fix to http://tracker.ceph.com/issues/38676 is in nautilus, this must be backported too.
- 06:09 PM Feature #39098: mds: lock caching for asynchronous unlink
- Zheng Yan wrote:
> Jeff Layton wrote:
> > Zheng Yan wrote:
> > > Sorry, I mean we don't need Lx
> >
> > I'm not... - 12:11 PM Bug #39350 (Resolved): df command error
- 12:03 PM Feature #38829 (In Progress): cephfs-shell: add a "stat" command
04/20/2019
- 05:34 PM Bug #39395 (Resolved): ceph: ceph fs auth fails
- Since 12.2.12 documentation-backed example invocation of @ceph fs authorize@ fails all the time:...
04/19/2019
- 03:15 PM Feature #39098: mds: lock caching for asynchronous unlink
- Jeff Layton wrote:
> Zheng Yan wrote:
> > Sorry, I mean we don't need Lx
>
> I'm not sure I understand. What if ... - 11:37 AM Feature #39098: mds: lock caching for asynchronous unlink
- Zheng Yan wrote:
> Sorry, I mean we don't need Lx
I'm not sure I understand. What if other clients have Ls on the... - 01:19 AM Feature #39098: mds: lock caching for asynchronous unlink
- Sorry, I mean we don't need Lx
04/18/2019
- 09:15 PM Feature #39098: mds: lock caching for asynchronous unlink
- Zheng Yan wrote:
> With change in http://tracker.ceph.com/issues/39354, Fx is not needed for async unlink. (Other re... - 07:36 AM Feature #39098: mds: lock caching for asynchronous unlink
- With change in http://tracker.ceph.com/issues/39354, Lx is not needed for async unlink. (Other request will release x...
- 11:06 AM Backport #39379 (In Progress): nautilus: cephfs-shell: python traceback with mkdir inside inexist...
- 09:20 AM Backport #39379 (Resolved): nautilus: cephfs-shell: python traceback with mkdir inside inexistant...
- https://github.com/ceph/ceph/pull/27677
- 11:05 AM Backport #39378 (In Progress): nautilus: cephfs-shell: support mkdir with non-octal mode
- 09:20 AM Backport #39378 (Resolved): nautilus: cephfs-shell: support mkdir with non-octal mode
- https://github.com/ceph/ceph/pull/27677
- 11:05 AM Backport #39377 (In Progress): nautilus: cephfs-shell: python traceback with mkdir when reattempt...
- 09:20 AM Backport #39377 (Resolved): nautilus: cephfs-shell: python traceback with mkdir when reattempt of...
- https://github.com/ceph/ceph/pull/27677
- 11:05 AM Backport #39376 (In Progress): nautilus: cephfs-shell: mkdir creates directory with invalid octal...
- 09:20 AM Backport #39376 (Resolved): nautilus: cephfs-shell: mkdir creates directory with invalid octal mode
- https://github.com/ceph/ceph/pull/27677
- 11:03 AM Backport #39197 (In Progress): nautilus: cephfs-shell: ls command produces error: no "colorize" a...
- 08:24 AM Bug #39349 (Closed): mds: cap revokes leak
- 08:03 AM Bug #39349: mds: cap revokes leak
- Zheng Yan wrote:
> already fixed by https://github.com/ceph/ceph/pull/26713
yes, this is already fixed, please cl... - 07:14 AM Bug #39349: mds: cap revokes leak
- already fixed by https://github.com/ceph/ceph/pull/26713
- 05:35 AM Backport #38877 (In Progress): luminous: mds: high debug logging with many subtrees is slow
- 12:08 AM Feature #38740 (Pending Backport): cephfs-shell: support mkdir with non-octal mode
- 12:08 AM Bug #38739 (Pending Backport): cephfs-shell: python traceback with mkdir inside inexistant directory
- 12:08 AM Bug #38741 (Pending Backport): cephfs-shell: python traceback with mkdir when reattempt of mkdir
- 12:08 AM Bug #38743 (Pending Backport): cephfs-shell: mkdir creates directory with invalid octal mode
04/17/2019
- 08:59 PM Feature #39354: mds: derive wrlock from excl caps
- https://github.com/ceph/ceph/pull/27648
- 01:07 PM Feature #39354 (In Progress): mds: derive wrlock from excl caps
- 12:49 PM Feature #39354 (Closed): mds: derive wrlock from excl caps
- preparation for buffered create/unlink
- 08:42 PM Bug #39349 (Fix Under Review): mds: cap revokes leak
- 07:50 AM Bug #39349: mds: cap revokes leak
- These exists possibilities as follows:
1. some req make mds do simple_xlock on a filelock that's in LOCK_xlock... - 07:25 AM Bug #39349 (Closed): mds: cap revokes leak
- Recently, one of our clusters, after updating to 12.2.11, occasionally reports that "XXX clients failing to respond t...
- 09:31 AM Bug #11314 (In Progress): qa: MDS crashed and the runs hung without ever timing out
- 09:31 AM Feature #5520 (New): osdc: should handle namespaces
- 08:01 AM Bug #39350 (Resolved): df command error
- Commit 417836d, causes the df to produce incorrect output....
- 06:01 AM Bug #39078 (Resolved): fs: we lack a feature bit for nautilus
- 06:01 AM Backport #39187 (Resolved): nautilus: fs: we lack a feature bit for nautilus
04/16/2019
- 11:54 PM Backport #39187: nautilus: fs: we lack a feature bit for nautilus
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/27497
merged - 08:11 PM Bug #39166 (Fix Under Review): mds: error "No space left on device" when create a large number o...
- 06:05 AM Bug #39166: mds: error "No space left on device" when create a large number of dirs
- Patrick Donnelly wrote:
> What version are you running?
Luminous 12.2.10 I have fix this issue by pull request 2... - 08:07 PM Bug #39305 (Fix Under Review): ceph-fuse: client hang because its bad session PipeConnection to mds
- 02:35 AM Bug #39305 (Resolved): ceph-fuse: client hang because its bad session PipeConnection to mds
- There still has a risk client will hang all the time and which happened in my enviroment
a few days ago. Well konw,... - 04:22 PM Bug #39329 (Won't Fix): ceph fs set <xxxxx> min_compat_client <yyyy> DOES NOT warn about connecte...
- ceph fs set <xxxxx> min_compat_client <yyyy> DOES NOT warn about connected incompatible clients
in spite of having... - 10:56 AM Bug #38652 (Resolved): mds|kclient: MDS_CLIENT_LATE_RELEASE warning caused by inline bug on RHEL 7.5
- 10:56 AM Backport #39225 (Resolved): nautilus: mds|kclient: MDS_CLIENT_LATE_RELEASE warning caused by inli...
- 10:55 AM Bug #39060 (Resolved): ls -S command produces AttributeError: 'str' object has no attribute 'decode'
- 10:54 AM Backport #39260 (Resolved): nautilus: ls -S command produces AttributeError: 'str' object has no ...
- 10:54 AM Bug #38804 (Resolved): cephfs-shell: ls always lists hidden files and directories
- 10:54 AM Backport #39217 (Resolved): nautilus: cephfs-shell: ls always lists hidden files and directories
04/15/2019
- 09:10 PM Bug #39166: mds: error "No space left on device" when create a large number of dirs
- What version are you running?
- 08:05 PM Backport #39225: nautilus: mds|kclient: MDS_CLIENT_LATE_RELEASE warning caused by inline bug on R...
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/27500
merged - 08:03 PM Backport #39260: nautilus: ls -S command produces AttributeError: 'str' object has no attribute '...
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/27531
merged - 08:03 PM Backport #39217: nautilus: cephfs-shell: ls always lists hidden files and directories
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/27531
merged
04/12/2019
- 11:54 AM Bug #39266 (Fix Under Review): There is no punctuation mark or blank between tid and client_id i...
- 03:13 AM Bug #39266: There is no punctuation mark or blank between tid and client_id in the output of "ce...
- https://github.com/ceph/ceph/pull/27537
- 02:58 AM Bug #39266 (Resolved): There is no punctuation mark or blank between tid and client_id in the ou...
- There is no punctuation mark or blank between tid and client_id.
The output of "ceph health detail " is listed bel... - 04:49 AM Feature #17309 (Fix Under Review): qa: mon_thrash test for CephFS
04/11/2019
- 05:47 PM Backport #39217 (In Progress): nautilus: cephfs-shell: ls always lists hidden files and directories
- 05:46 PM Backport #39260 (In Progress): nautilus: ls -S command produces AttributeError: 'str' object has ...
- 05:43 PM Backport #39260 (Resolved): nautilus: ls -S command produces AttributeError: 'str' object has no ...
- https://github.com/ceph/ceph/pull/27531
- 04:47 PM Bug #39060 (Pending Backport): ls -S command produces AttributeError: 'str' object has no attribu...
- 08:45 AM Bug #38832 (Fix Under Review): mds: fail to resolve snapshot name contains '_'
- https://github.com/ceph/ceph/pull/27511
- 08:04 AM Bug #38832: mds: fail to resolve snapshot name contains '_'
- There's some confusion here -- we created a snapshot named 'hourly_2019-03-20_172736'. The internal _ chars are causi...
- 07:35 AM Bug #38832 (Need More Info): mds: fail to resolve snapshot name contains '_'
- 07:35 AM Bug #38832: mds: fail to resolve snapshot name contains '_'
- Server::handle_client_mksnap already contains following code:...
04/10/2019
- 09:55 PM Backport #39225 (In Progress): nautilus: mds|kclient: MDS_CLIENT_LATE_RELEASE warning caused by i...
- 09:08 PM Backport #39225 (Resolved): nautilus: mds|kclient: MDS_CLIENT_LATE_RELEASE warning caused by inli...
- https://github.com/ceph/ceph/pull/27500
- 09:14 PM Backport #39187 (In Progress): nautilus: fs: we lack a feature bit for nautilus
- 09:03 PM Backport #39187 (Resolved): nautilus: fs: we lack a feature bit for nautilus
- https://github.com/ceph/ceph/pull/27497
- 09:10 PM Backport #39233 (Resolved): mimic: kclient: nofail option not supported
- https://github.com/ceph/ceph/pull/28090
- 09:10 PM Backport #39232 (Resolved): nautilus: kclient: nofail option not supported
- https://github.com/ceph/ceph/pull/27851
- 09:10 PM Backport #39231 (Resolved): luminous: kclient: nofail option not supported
- https://github.com/ceph/ceph/pull/28436
- 09:07 PM Backport #39223 (Resolved): mimic: mds: behind on trimming and "[dentry] was purgeable but no lon...
- https://github.com/ceph/ceph/pull/29224
- 09:07 PM Backport #39222 (Resolved): nautilus: mds: behind on trimming and "[dentry] was purgeable but no ...
- https://github.com/ceph/ceph/pull/27879
- 09:07 PM Backport #39221 (Resolved): luminous: mds: behind on trimming and "[dentry] was purgeable but no ...
- https://github.com/ceph/ceph/pull/28432
- 09:06 PM Backport #39217 (Resolved): nautilus: cephfs-shell: ls always lists hidden files and directories
- https://github.com/ceph/ceph/pull/27531
- 09:06 PM Backport #39215 (Resolved): mimic: mds: there is an assertion when calling Beacon::shutdown()
- https://github.com/ceph/ceph/pull/29223
- 09:06 PM Backport #39214 (Resolved): nautilus: mds: there is an assertion when calling Beacon::shutdown()
- https://github.com/ceph/ceph/pull/27852
- 09:06 PM Backport #39213 (Resolved): luminous: mds: there is an assertion when calling Beacon::shutdown()
- https://github.com/ceph/ceph/pull/28438
- 09:06 PM Backport #39212 (Resolved): mimic: MDSTableServer.cc: 83: FAILED assert(version == tid)
- https://github.com/ceph/ceph/pull/29222
- 09:06 PM Backport #39211 (Resolved): nautilus: MDSTableServer.cc: 83: FAILED assert(version == tid)
- https://github.com/ceph/ceph/pull/27853
- 09:06 PM Backport #39210 (Resolved): mimic: mds: mds_cap_revoke_eviction_timeout is not used to initialize...
- https://github.com/ceph/ceph/pull/29220
- 09:06 PM Backport #39209 (Resolved): nautilus: mds: mds_cap_revoke_eviction_timeout is not used to initial...
- https://github.com/ceph/ceph/pull/27842
- 09:05 PM Backport #39208 (Resolved): luminous: mds: mds_cap_revoke_eviction_timeout is not used to initial...
- https://github.com/ceph/ceph/pull/27840
- 09:04 PM Backport #39200 (Resolved): mimic: mds: we encountered "No space left on device" when moving huge...
- https://github.com/ceph/ceph/pull/27917
- 09:04 PM Backport #39199 (Resolved): nautilus: mds: we encountered "No space left on device" when moving h...
- https://github.com/ceph/ceph/pull/27736
- 09:04 PM Backport #39198 (Resolved): luminous: mds: we encountered "No space left on device" when moving h...
- https://github.com/ceph/ceph/pull/27801
- 09:04 PM Backport #39197 (Resolved): nautilus: cephfs-shell: ls command produces error: no "colorize" attr...
- https://github.com/ceph/ceph/pull/27677
- 09:03 PM Backport #39193 (Resolved): mimic: mds: crash during mds restart
- https://github.com/ceph/ceph/pull/27916
- 09:03 PM Backport #39192 (Resolved): nautilus: mds: crash during mds restart
- https://github.com/ceph/ceph/pull/27714
- 09:03 PM Backport #39191 (Resolved): luminous: mds: crash during mds restart
- https://github.com/ceph/ceph/pull/27737
- 09:01 PM Backport #39176 (Resolved): nautilus: doc: add documentation for `fs set min_compat_client`
- https://github.com/ceph/ceph/pull/27900
- 04:51 PM Bug #39173: ceph-fuse: Ceph-fuse too slow compared to kernel client
- Sure, It would help to mention the level of performance difference in the documents here http://docs.ceph.com/docs/ma...
- 04:42 PM Bug #39173 (Won't Fix): ceph-fuse: Ceph-fuse too slow compared to kernel client
- Sorry this isn't a bug and it's well documented everywhere. The userspace client libraries (and FUSE) are known to be...
- 04:36 PM Bug #39173 (Won't Fix): ceph-fuse: Ceph-fuse too slow compared to kernel client
- I have a ceph cluster with 10 osd nodes, 50 osds, 5 mons, 5 mds (1 active, 4standby).
Ceph cluster is running latest... - 05:45 AM Bug #39166 (Resolved): mds: error "No space left on device" when create a large number of dirs
- When create dirs under the ceph mount path quickly, it returns "No space left on device" after the number beyond the ...
- 04:32 AM Bug #38681: cephfs-shell: add commands to manipulate snapshots
- filed a separate tracker for quota management
- 04:31 AM Bug #39165 (Resolved): cephfs-shell: add commands to manipulate quotas
- Idea being to make this a trivial one-off command to manipulate quotas on CephFS directories without mounting.
04/09/2019
- 03:10 PM Bug #38681 (Fix Under Review): cephfs-shell: add commands to manipulate snapshots
- need review for snapshot management PR
- 01:18 PM Bug #38679 (Pending Backport): mds: behind on trimming and "[dentry] was purgeable but no longer ...
- 01:18 PM Documentation #39130 (Pending Backport): doc: add documentation for `fs set min_compat_client`
- 12:08 AM Bug #38996 (Pending Backport): cephfs-shell: ls command produces error: no "colorize" attribute f...
04/08/2019
- 11:13 PM Bug #39026 (Pending Backport): mds: crash during mds restart
- 11:12 PM Bug #38994 (Pending Backport): mds: we encountered "No space left on device" when moving huge num...
- 10:42 PM Bug #38803 (Resolved): qa: test_sessionmap assumes simple messenger
- 06:31 PM Bug #22249 (Can't reproduce): Need to restart MDS to release cephfs space
- 06:28 PM Bug #16322 (Can't reproduce): ceph mds getting killed for no reason
- 06:26 PM Bug #38832: mds: fail to resolve snapshot name contains '_'
- Zheng, what was the motivation for the change to append this information to the snap name after the final _? re: 07e7...
- 06:19 PM Cleanup #24745 (Won't Fix): Spurious empty files in CephFS root pool when multiple pools associated
- As John noted, this isn't a bug. It may be we eventually try to relax the requirement for backtraces to be stored on ...
- 06:08 PM Cleanup #4744: mds: pass around LogSegments via std::shared_ptr
- 06:07 PM Feature #38951: client: implement asynchronous unlink/create
- Actually it may be sufficient to just wait on any existing dirops to complete before we do a synchronous one. I alrea...
- 04:31 PM Feature #38951: client: implement asynchronous unlink/create
- I ran fsstress on these patches today and hit a deadlock:...
- 04:46 PM Feature #38849 (Duplicate): mds: proactively merge orphaned inodes into a remaining parent after ...
- 01:49 PM Bug #39061 (Duplicate): Test failure: test_drop_cache_command_timeout_asok (tasks.cephfs.test_mis...
- http://tracker.ceph.com/issues/38348
- 01:46 PM Bug #38996 (Fix Under Review): cephfs-shell: ls command produces error: no "colorize" attribute f...
- 08:06 AM Bug #38996: cephfs-shell: ls command produces error: no "colorize" attribute found error
- Patrick Donnelly wrote:
> I just ran into this issue. Sorry for assuming it's a duplicate.
>
> It looks like this... - 12:14 PM Backport #38877: luminous: mds: high debug logging with many subtrees is slow
- Patrick Donnelly wrote:
> Rishabh, please do this backport. See also this guide: http://tracker.ceph.com/projects/ce... - 11:57 AM Backport #38540 (In Progress): mimic: qa: fsstress with valgrind may timeout
04/06/2019
- 10:07 AM Feature #39098: mds: lock caching for asynchronous unlink
- Zheng Yan wrote:
> If mds revokes Lx or Fx, client should flush the async unlinks (on the file or under the director... - 03:11 AM Feature #39098: mds: lock caching for asynchronous unlink
- Zheng Yan wrote:
> Jeff Layton wrote:
> > Zheng Yan wrote:
> > > mds takes xlock on linklock when unlinking file, ... - 03:08 AM Feature #39098: mds: lock caching for asynchronous unlink
- Jeff Layton wrote:
> Zheng Yan wrote:
> > mds takes xlock on linklock when unlinking file, which always revokes Lsx...
04/05/2019
- 11:10 PM Bug #38803 (Fix Under Review): qa: test_sessionmap assumes simple messenger
- 06:27 PM Bug #38803 (In Progress): qa: test_sessionmap assumes simple messenger
- 09:12 PM Documentation #39130 (Fix Under Review): doc: add documentation for `fs set min_compat_client`
- 07:26 PM Documentation #39130 (Resolved): doc: add documentation for `fs set min_compat_client`
- 06:40 PM Feature #38951: client: implement asynchronous unlink/create
- Ok, I have a prototype implementation that depends on a couple of small MDS patches that are discussed here https://t...
- 05:41 PM Feature #39129 (Resolved): create mechanism to delegate ranges of inode numbers to client
- Create a mechanism by which we can hand out ranges of inode numbers to MDS clients. The clients can then use those to...
- 05:23 PM Feature #39098: mds: lock caching for asynchronous unlink
- Jeff Layton wrote:
> This patch seems to do the right thing for me. I'm not sure whether this is the right approach ... - 05:22 PM Feature #39098: mds: lock caching for asynchronous unlink
- Jeff Layton wrote:
> Zheng Yan wrote:
> > mds takes xlock on linklock when unlinking file, which always revokes Lsx... - 03:31 PM Feature #39098: mds: lock caching for asynchronous unlink
- This patch seems to do the right thing for me. I'm not sure whether this is the right approach though, and we will ne...
- 12:00 PM Feature #39098: mds: lock caching for asynchronous unlink
- Continuing that thought, I see that handle_client_unlink does this unconditionally:...
- 10:51 AM Feature #39098: mds: lock caching for asynchronous unlink
- Zheng Yan wrote:
> mds takes xlock on linklock when unlinking file, which always revokes Lsx. why does client want ... - 09:07 AM Feature #39098: mds: lock caching for asynchronous unlink
- mds takes xlock on linklock when unlinking file, which always revokes Lsx. why does client want to keep Lx after sen...
04/04/2019
- 10:35 PM Bug #38804 (Pending Backport): cephfs-shell: ls always lists hidden files and directories
- 10:32 PM Bug #38996: cephfs-shell: ls command produces error: no "colorize" attribute found error
- I just ran into this issue. Sorry for assuming it's a duplicate.
It looks like this was caused by the cmd2 module ... - 07:47 PM Bug #39020 (Resolved): qa: qa/suites/fs/upgrade testing with upgrades from luminous no longer work
- 07:46 PM Bug #39078 (Pending Backport): fs: we lack a feature bit for nautilus
- Please only backport dcd6e97944f1eefb236c5b680569c0e7c085a692.
- 07:45 PM Bug #39077 (Resolved): fs: add note to release process that new CEPHFS_FEATURE_X bit needs added ...
- Effected in a00151ac as a compiler error.
- 07:40 PM Bug #38835 (Pending Backport): MDSTableServer.cc: 83: FAILED assert(version == tid)
- 07:39 PM Bug #38652 (Pending Backport): mds|kclient: MDS_CLIENT_LATE_RELEASE warning caused by inline bug ...
- 07:33 PM Feature #39098: mds: lock caching for asynchronous unlink
- This does help the MDS to hand out Lx caps on newly created files, but I'm having some trouble getting the cap handli...
- 04:38 PM Feature #39098: mds: lock caching for asynchronous unlink
- Now that I think about it, Lx may not be required for rmdir anyway:
We only need that for files that can be hardli... - 04:06 PM Feature #39098: mds: lock caching for asynchronous unlink
- Thanks Zheng, with both of those patches, I get Lx caps on newly-created files, but not on new directories. This is t...
- 02:22 PM Feature #39098: mds: lock caching for asynchronous unlink
- following path improves the chance that client get Fsx on directory....
- 01:59 PM Feature #39098: mds: lock caching for asynchronous unlink
- please following mds patch...
- 10:57 AM Feature #39098: mds: lock caching for asynchronous unlink
- I added that to the userland client and then used cephfs-shell to create a directory and a file inside it. Neither cr...
- 09:25 AM Feature #39098: mds: lock caching for asynchronous unlink
- please try following patch...
- 01:59 PM Bug #38742 (Fix Under Review): cephfs-shell: entering unrecognized command does not print newline...
- Updated with PR number
- 01:57 PM Feature #38740 (Fix Under Review): cephfs-shell: support mkdir with non-octal mode
- updated with PR number
- 01:56 PM Bug #38739 (Fix Under Review): cephfs-shell: python traceback with mkdir inside inexistant directory
- yes; the shell should print a sensible error in such a case, in my opinion
- 01:54 PM Bug #38741 (Fix Under Review): cephfs-shell: python traceback with mkdir when reattempt of mkdir
- 01:52 PM Bug #38743 (Fix Under Review): cephfs-shell: mkdir creates directory with invalid octal mode
- 01:29 PM Backport #38340 (In Progress): luminous: mds: may leak gather during cache drop
- 12:05 AM Bug #23262: kclient: nofail option not supported
- Kenneth Waegeman wrote:
> Hi,
> Should I do something to backport it?
No that's not necessary.
04/03/2019
- 10:49 PM Bug #39060 (Fix Under Review): ls -S command produces AttributeError: 'str' object has no attribu...
- 07:53 PM Feature #39098 (Resolved): mds: lock caching for asynchronous unlink
- In order to allow the client to asynchronously delete files, we need Fx caps on the parent, and Lx caps on the inode ...
- 02:17 PM Bug #23262: kclient: nofail option not supported
- Hi,
Should I do something to backport it? - 01:15 PM Feature #5520 (In Progress): osdc: should handle namespaces
- 02:32 AM Bug #38832: mds: fail to resolve snapshot name contains '_'
- ...
Also available in: Atom