Activity
From 12/04/2019 to 01/02/2020
01/02/2020
- 09:19 PM Bug #43407: mds crash after update to v14.2.5
- Yes I had 3 filesystems (namespaces), one for every mds daemon, and the setup was working up to the update to v14.2.5...
- 07:00 PM Bug #43407: mds crash after update to v14.2.5
- Were you using multiple MDS before?
Can you increase MDS debugging:
ceph config set mds debug_mds 10
and res... - 08:10 PM Feature #24461: cephfs: improve file create performance buffering file unlink/create operations
- Looking at my home-grown testcase, the results look pretty good, but an untarring a random kernel tarball is consider...
01/01/2020
- 09:38 AM Documentation #43154: doc: migrate best practice recommendations to relevant docs
- https://docs.ceph.com/docs/master/cephfs/fuse/ - This is the location of the FUSE docs.
12/31/2019
- 12:46 PM Bug #43440 (Resolved): client: chdir does not raise error if a file is passed
- ...
- 06:14 AM Feature #41566 (In Progress): mds: support rolling upgrades
- 04:13 AM Feature #43435: kclient:send client provided metric flags in client metadata
- Patch is ready and the test output is:...
- 04:11 AM Bug #43438 (Fix Under Review): cephfs-journal-tool: will crash without any extra argument
- 04:10 AM Bug #43438: cephfs-journal-tool: will crash without any extra argument
- The fixing PR: https://github.com/ceph/ceph/pull/32452
- 04:01 AM Bug #43438 (In Progress): cephfs-journal-tool: will crash without any extra argument
- 04:00 AM Bug #43438 (Resolved): cephfs-journal-tool: will crash without any extra argument
- ...
12/30/2019
- 04:16 PM Bug #41565 (Fix Under Review): mds: detect MDS<->MDS messages that are not versioned
- 05:42 AM Feature #43435 (In Progress): kclient:send client provided metric flags in client metadata
- 05:42 AM Feature #43435 (Resolved): kclient:send client provided metric flags in client metadata
- This will send the kclient provided metric flags to the MDS server.
12/27/2019
12/26/2019
- 05:26 PM Cleanup #43425 (Fix Under Review): mds: reorg snap header
- 03:01 PM Cleanup #43425 (Resolved): mds: reorg snap header
- 03:03 PM Cleanup #43426 (Resolved): mds: reorg mdstypes header
- 03:02 PM Cleanup #43424 (Fix Under Review): mds: reorg inode_backtrace header
- 01:58 PM Cleanup #43424 (Resolved): mds: reorg inode_backtrace header
- 06:15 AM Feature #43423: mds: collect and show the dentry lease metric
- https://tracker.ceph.com/issues/24285
- 06:12 AM Feature #43423: mds: collect and show the dentry lease metric
- Locally the patch is ready, but depend on https://github.com/ceph/ceph/pull/26004, which hasn't been merged yet.
<... - 06:10 AM Feature #43423 (Resolved): mds: collect and show the dentry lease metric
- Kclient will collect the dentry lease metric and send it to the MDS, currently this hasn't been shown in the perf stats.
12/25/2019
- 11:20 AM Bug #43410 (Won't Fix): mds:When the directory level is above 3000, the following assertions will...
12/24/2019
- 09:02 AM Cleanup #43418 (Fix Under Review): mds: reorg flock header
- 07:38 AM Cleanup #43418 (Resolved): mds: reorg flock header
- 06:19 AM Bug #43410: mds:When the directory level is above 3000, the following assertions will appear
- Zheng Yan wrote:
> mds call FOO::adjust_nested_auth_pins functions for each directory level, which caused stack over... - 06:14 AM Bug #43410: mds:When the directory level is above 3000, the following assertions will appear
- Zheng Yan wrote:
> full calltrace ?
I use gdb for mounting, this information is complete - 02:31 AM Bug #43410: mds:When the directory level is above 3000, the following assertions will appear
- mds call FOO::adjust_nested_auth_pins functions for each directory level, which caused stack overflow. mimic and late...
- 01:53 AM Bug #43410: mds:When the directory level is above 3000, the following assertions will appear
- full calltrace ?
12/23/2019
- 11:14 AM Cleanup #43408 (Fix Under Review): mds: reorg StrayManager header
- 10:51 AM Cleanup #43408 (Resolved): mds: reorg StrayManager header
- 11:14 AM Bug #43410: mds:When the directory level is above 3000, the following assertions will appear
- I don't see any exceptions in the log print
@Patrick Donnelly
@Zheng Yan - 11:10 AM Bug #43410 (Won't Fix): mds:When the directory level is above 3000, the following assertions will...
- When I use the script to continuously create directories,
but the directory level is above 3000, the following asse... - 11:10 AM Bug #43409 (Closed): mds:When the directory level is above 3000, the following assertions will ap...
- 11:09 AM Bug #43409 (Closed): mds:When the directory level is above 3000, the following assertions will ap...
- When I use the script to continuously create directories,
but the directory level is above 3000, the following asse...
12/22/2019
- 11:36 PM Bug #43407 (Triaged): mds crash after update to v14.2.5
- All MDS crashed and not able to restart after update from v14.2.4 to v14.2.5
*systemctl status:*...
12/21/2019
- 03:57 AM Feature #24461: cephfs: improve file create performance buffering file unlink/create operations
- Zheng Yan wrote:
> Patrick Donnelly wrote:
> > The baseline performance is surprising I think. That's with the same...
12/20/2019
12/19/2019
- 08:31 PM Bug #43393 (Resolved): qa: add support/qa for cephfs-shell on CentOS 9 / RHEL9
- 08:00 PM Documentation #41688 (Resolved): doc: client config reference improvements
- 07:57 PM Bug #43250 (Resolved): qa/test_cephfs_shell: TestDu.test_du_works_for_hardlinks fails
- 07:46 PM Bug #43392 (Resolved): MDSMonitor: support automatic failover to standbys with stronger affinity
- Initial work by Sage: https://github.com/ceph/ceph/pull/32015
The next step is to failover to a standby with stron... - 02:38 PM Bug #43329 (Fix Under Review): cephfs-shell: AttributeError when undefined an conf opt is attempt...
- 02:30 PM Cleanup #43387 (Fix Under Review): mds: reorg SnapServer header
- 02:23 PM Cleanup #43387 (Resolved): mds: reorg SnapServer header
- 02:16 PM Cleanup #43386 (Fix Under Review): mds: reorg SnapRealm header
- 02:08 PM Cleanup #43386 (Resolved): mds: reorg SnapRealm header
- 01:48 PM Feature #39129: create mechanism to delegate ranges of inode numbers to client
- You're right. I just pushed a patch to be squashed in on top of the existing series. I'm testing it now with the clie...
- 01:40 PM Feature #39129: create mechanism to delegate ranges of inode numbers to client
- Jeff Layton wrote:
> That's not a bad idea. We'd have to keep track of a separate set of newly-added ino_t's to send... - 12:09 PM Feature #39129: create mechanism to delegate ranges of inode numbers to client
- That's not a bad idea. We'd have to keep track of a separate set of newly-added ino_t's to send in the reply, but tha...
- 09:01 AM Feature #39129: create mechanism to delegate ranges of inode numbers to client
- Jeff Layton wrote:
> I have patches for this for the MDS, and the kernel, but I keep hitting a race where the client...
12/18/2019
- 09:12 PM Feature #39129: create mechanism to delegate ranges of inode numbers to client
- Jeff Layton wrote:
> > > The big problem is that all of the creates are not necessarily processed in a strict order ... - 09:07 PM Feature #39129: create mechanism to delegate ranges of inode numbers to client
- Patrick Donnelly wrote:
> Jeff Layton wrote:
> > The problem there is that the second set would grow without bound.... - 08:43 PM Feature #39129: create mechanism to delegate ranges of inode numbers to client
- Jeff Layton wrote:
> The problem there is that the second set would grow without bound. It's not a lot of info per i... - 08:26 PM Feature #39129: create mechanism to delegate ranges of inode numbers to client
- The problem there is that the second set would grow without bound. It's not a lot of info per inode, but it's enough ...
- 08:04 PM Feature #39129: create mechanism to delegate ranges of inode numbers to client
- Jeff Layton wrote:
> I have patches for this for the MDS, and the kernel, but I keep hitting a race where the client... - 07:33 PM Feature #39129: create mechanism to delegate ranges of inode numbers to client
- I have patches for this for the MDS, and the kernel, but I keep hitting a race where the client adds an already-used ...
- 08:53 PM Bug #41329: mds: reject sessionless messages
- Follow-up: https://github.com/ceph/ceph/pull/32318
I've asked Zheng to make another tracker ticket. - 03:39 PM Cleanup #43369 (Fix Under Review): mds: reorg SnapClient header
- 02:28 PM Cleanup #43369 (Resolved): mds: reorg SnapClient header
- 02:39 PM Bug #43362 (Fix Under Review): client: disallow changing fuse_default_permissions option at runtime
- 04:46 AM Bug #43362 (Resolved): client: disallow changing fuse_default_permissions option at runtime
- If fuse_default_permissions is false when initializing fuse, then ceph-fuse will use its own permission check. If cha...
- 01:33 PM Cleanup #43367 (Fix Under Review): mds: reorg SimpleLock header
- 01:00 PM Cleanup #43367 (Resolved): mds: reorg SimpleLock header
- 12:40 PM Cleanup #43366 (Fix Under Review): mds: reorg SessionMap header
- 12:34 PM Cleanup #43366 (Resolved): mds: reorg SessionMap header
- 09:05 AM Bug #43336 (Fix Under Review): qa: test_unmount_for_evicted_client hangs
- 06:56 AM Bug #43336: qa: test_unmount_for_evicted_client hangs
- I think it was caused by
[ 150.326253] ceph: mdsc_handle_session corrupt message mds0 len 75^M
- 05:15 AM Cleanup #42465 (Resolved): mds: reorg MDSRank header
- 04:30 AM Feature #24461: cephfs: improve file create performance buffering file unlink/create operations
- Patrick Donnelly wrote:
> The baseline performance is surprising I think. That's with the same MDS patches? The usua... - 12:23 AM Documentation #43162 (Resolved): doc: "adding an MDS" in deployment is out-of-date
12/17/2019
- 06:58 PM Bug #42088 (Fix Under Review): 'ceph -s' does not show standbys if there are no filesystems
- 06:58 PM Bug #42088 (Pending Backport): 'ceph -s' does not show standbys if there are no filesystems
- 05:18 PM Bug #43039 (Need More Info): client: shutdown race fails with status 141
- Jeff Layton wrote:
> (Handing back to Patrick for now)
>
> Is this problem still occurring in teuthology?
Havn... - 03:36 PM Feature #24461: cephfs: improve file create performance buffering file unlink/create operations
- Ok, I think I've figured out what's going on with the inode number reuse.
I changed the code to not remove ino_t e... - 12:08 PM Backport #42713: nautilus: mgr: daemon state for mds not available
- I'll send the backport tomorrow.
- 10:36 AM Documentation #43162 (In Progress): doc: "adding an MDS" in deployment is out-of-date
- 09:27 AM Feature #43349 (Fix Under Review): mgr/volumes: provision subvolumes with config metadata storage...
- 09:25 AM Feature #43349 (Resolved): mgr/volumes: provision subvolumes with config metadata storage in cephfs
- Patrick had this idea a while back, but this never got implemented. Currently, there is no storage area when a subvol...
- 09:20 AM Backport #43348 (Resolved): nautilus: mds: crash(FAILED assert(omap_num_objs <= MAX_OBJECTS))
- https://github.com/ceph/ceph/pull/32756
- 09:20 AM Backport #43347 (Resolved): mimic: mds: crash(FAILED assert(omap_num_objs <= MAX_OBJECTS))
- https://github.com/ceph/ceph/pull/32757
- 09:18 AM Backport #43345 (Resolved): nautilus: mds: metadata changes may be lost when MDS is restarted
- https://github.com/ceph/ceph/pull/30843
- 09:18 AM Backport #43344 (Rejected): mimic: mds: metadata changes may be lost when MDS is restarted
- 09:16 AM Backport #43343 (Resolved): nautilus: mds: client does not response to cap revoke After session s...
- https://github.com/ceph/ceph/pull/32909
- 09:16 AM Backport #43342 (Rejected): mimic: mds: client does not response to cap revoke After session stal...
- 05:06 AM Backport #43338 (In Progress): nautilus: qa/tasks: add remaining tests for fs volume
- 04:28 AM Backport #43338 (Resolved): nautilus: qa/tasks: add remaining tests for fs volume
- https://github.com/ceph/ceph/pull/33122/
- 03:30 AM Bug #43326: mds: batch getattr/lookup bug
- One symptom is client lookup request hang
- 01:27 AM Feature #43337 (New): fs: support relatime correctly for CephFS
- As of now, CephFS does not seem to handle atime.
The relatime mount option (since Linux 2.6.30) is meant to not o... - 12:31 AM Bug #42872 (Pending Backport): qa/tasks: add remaining tests for fs volume
- 12:08 AM Bug #40784 (Pending Backport): mds: metadata changes may be lost when MDS is restarted
- 12:06 AM Bug #42826 (Pending Backport): mds: client does not response to cap revoke After session stale->r...
- 12:04 AM Cleanup #42464 (Resolved): mds: reorg MDSMap header
- 12:00 AM Bug #36094 (Pending Backport): mds: crash(FAILED assert(omap_num_objs <= MAX_OBJECTS))
12/16/2019
- 11:58 PM Feature #43182 (Resolved): mds: increase default cache size to 4GB
- 11:34 PM Bug #43336 (Resolved): qa: test_unmount_for_evicted_client hangs
- ...
- 04:59 PM Bug #42365: client: FAILED assert(dir->readdir_cache[dirp->cache_index] == dn)
- ceph-post-file: 123801df-99cc-4c0a-a76c-9b6c8a614394
- 03:51 PM Bug #42365: client: FAILED assert(dir->readdir_cache[dirp->cache_index] == dn)
- hit 3 more times in 13.2.5, I had catch a coredump
- 02:35 PM Bug #43329 (Resolved): cephfs-shell: AttributeError when undefined an conf opt is attemptted to read
- conf_get() from pybind/cephfs/cephfs.pyx returns None when passed argument is not present as config file option which...
- 01:27 PM Bug #43248 (In Progress): cephfs-shell: do not drop into shell after running command-line command
- 08:20 AM Bug #43326 (Fix Under Review): mds: batch getattr/lookup bug
- 08:13 AM Bug #43326 (Resolved): mds: batch getattr/lookup bug
12/15/2019
- 08:09 AM Backport #42462 (Resolved): nautilus: doc: MDS and metadata pool hardware requirements/recommenda...
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/31116
m...
12/14/2019
12/13/2019
- 07:51 PM Feature #24461: cephfs: improve file create performance buffering file unlink/create operations
- Doing some testing today with xfstests, during generic/531 test, I saw some of these pop up in the kernel ring buffer...
- 07:29 PM Documentation #43222 (Resolved): doc: mention multimds in dev guide's list of integration test su...
- 07:29 PM Documentation #43220 (Resolved): doc: clarify difference fs and kcephfs suite in dev guide
- 07:07 PM Bug #43133 (Resolved): vstop.sh: Mounts are not cleaned up
- 06:21 AM Bug #43251 (Fix Under Review): mds: track client provided metric flags in session
- 01:19 AM Bug #43149: kclient: umount will stuck for around 1 minutes sometimes
- The fixing commit: https://github.com/ceph/ceph-client/commit/992dd028db77657b5eb164d0825a991d5c14ec78
- 01:17 AM Bug #43295 (Fix Under Review): kclient: keep the session state until it is released
- The fixing commit: https://github.com/ceph/ceph-client/commit/38d173ab657c9b77ad3ab0f8c9b83245959cdb63
- 01:16 AM Bug #43295 (In Progress): kclient: keep the session state until it is released
Let's keep the session state until its memories is released.- 01:16 AM Bug #43295 (Resolved): kclient: keep the session state until it is released
- When reconnecting the session but if it is denied by the MDS due
to client was in blacklist or something else, kclie... - 01:11 AM Feature #43294 (Fix Under Review): mount.ceph: give a hint message when no mds is up or cluster i...
- 01:11 AM Feature #43294 (In Progress): mount.ceph: give a hint message when no mds is up or cluster is lag...
- 01:11 AM Feature #43294: mount.ceph: give a hint message when no mds is up or cluster is laggy
- The relating PR: https://github.com/ceph/ceph/pull/32164
- 01:10 AM Feature #43294 (Resolved): mount.ceph: give a hint message when no mds is up or cluster is laggy
- The kclient will return EHOSTUNREACH when no MDS is up or the cluster is laggy.
Check it and give a hint. - 01:00 AM Feature #4386 (Fix Under Review): kclient: Mount error message when no MDS present
- 01:00 AM Feature #4386: kclient: Mount error message when no MDS present
- Return -EHOSTUNREACH instead if no MDS is up or the cluster is laggy.
- 12:55 AM Bug #43293 (Resolved): kclient: trigger the reclaim work once there has enough pending caps
- This will fix it: https://github.com/ceph/ceph-client/commit/bba1560bd4a46aa0d16bb7d81abd9d0eb47dea36.
- 12:54 AM Bug #43293 (Resolved): kclient: trigger the reclaim work once there has enough pending caps
- For corner case the reclaim work won't be fired even we have a large number of pending caps in time as expected.
12/12/2019
- 09:39 PM Bug #42923 (Pending Backport): pybind / cephfs: remove static typing in LibCephFS.chown
- 02:28 PM Bug #43249 (Fix Under Review): cephfs-shell: exit failure when non-interactive command fails
- 06:29 AM Documentation #37746: doc: how to mount a subdir with ceph-fuse/kclient
- Марк Коренберг wrote:
> Well,
> 1. I don't consider using options in the device part. I just found a working examp... - 05:40 AM Backport #43271 (In Progress): nautilus: qa/tasks: Fix raises that doesn't re-raise in test_volum...
- 05:30 AM Backport #43271 (Resolved): nautilus: qa/tasks: Fix raises that doesn't re-raise in test_volumes.py
- https://github.com/ceph/ceph/pull/33122
- 12:09 AM Bug #43270 (Fix Under Review): kclient: retry the same mds later after the new session is opened
- This is the fixing patch: https://github.com/ceph/ceph-client/commit/5be1d0c54652ae3ba0a452bb3b12950e20597d0e
- 12:08 AM Bug #43270 (Resolved): kclient: retry the same mds later after the new session is opened
- With max_mds > 1 and for a request which are choosing a random
mds rank and if the relating session is not opened ye...
12/11/2019
- 09:15 PM Feature #24461: cephfs: improve file create performance buffering file unlink/create operations
- The baseline performance is surprising I think. That's with the same MDS patches? The usual full tilt create/second r...
- 06:19 PM Feature #24461: cephfs: improve file create performance buffering file unlink/create operations
- Changed my test rig around a bit so I could give bluestore a LV backed by an SSD, and rebuilt the kernel w/o KASAN.
... - 04:57 PM Feature #24461: cephfs: improve file create performance buffering file unlink/create operations
- I've pushed the current patch stack to https://github.com/ceph/ceph-client/tree/wip-async-dirops .
It's still very... - 04:45 PM Feature #24461: cephfs: improve file create performance buffering file unlink/create operations
- The patchset currently has a module option to enable this that defaults of "off". So I can do some apples to apples t...
- 04:36 PM Feature #24461: cephfs: improve file create performance buffering file unlink/create operations
- I finally have a working patchset for this, and so far, the results are somewhat lackluster. I'm seeing about the sam...
- 05:51 PM Cleanup #41951 (Resolved): mds: obsolete mds_cache_size
- 05:43 PM Cleanup #42866 (Resolved): mds: reorg ScrubStack header
- 05:41 PM Cleanup #42865 (Resolved): mds: reorg ScrubHeader header
- 05:40 PM Cleanup #42864 (Resolved): mds: reorg ScatterLock header
- 05:39 PM Cleanup #42813 (Resolved): mds: reorg RecoveryQueue header
- 05:39 PM Cleanup #42792 (Resolved): mds: reorg OpenFileTable header
- 05:37 PM Bug #41694 (Pending Backport): qa/tasks: Fix raises that doesn't re-raise in test_volumes.py
- 03:30 PM Bug #43247 (Resolved): qa: test_cephfs_shell.TestSnapshots.test_snap FAIL
- 01:50 PM Documentation #37746: doc: how to mount a subdir with ceph-fuse/kclient
- Well,
1. I don't consider using options in the device part. I just found a working example and wondered that there ... - 05:28 AM Documentation #37746: doc: how to mount a subdir with ceph-fuse/kclient
- Марк Коренберг wrote:
> In order not to loose linked info from https://forum.proxmox.com/threads/mount-cephfs-using-... - 11:54 AM Bug #43249 (In Progress): cephfs-shell: exit failure when non-interactive command fails
- 09:59 AM Bug #43251 (Resolved): mds: track client provided metric flags in session
- With PR https://github.com/ceph/ceph/pull/26004, MDS will start tracking client provided metrics. However, the set of...
- 07:10 AM Bug #43250 (Fix Under Review): qa/test_cephfs_shell: TestDu.test_du_works_for_hardlinks fails
- 06:57 AM Bug #43250 (In Progress): qa/test_cephfs_shell: TestDu.test_du_works_for_hardlinks fails
- 06:42 AM Bug #43250 (Resolved): qa/test_cephfs_shell: TestDu.test_du_works_for_hardlinks fails
- Got this yesterday locally -...
12/10/2019
- 11:37 PM Bug #43247 (Fix Under Review): qa: test_cephfs_shell.TestSnapshots.test_snap FAIL
- 10:58 PM Bug #43247: qa: test_cephfs_shell.TestSnapshots.test_snap FAIL
- master: http://pulpito.ceph.com/pdonnell-2019-12-10_20:51:09-fs-master-distro-basic-smithi/
- 08:52 PM Bug #43247 (Resolved): qa: test_cephfs_shell.TestSnapshots.test_snap FAIL
- ...
- 11:22 PM Bug #43249 (Resolved): cephfs-shell: exit failure when non-interactive command fails
- If a one-shot command fails, the cephfs-shell should exit with a non-zero status:...
- 11:20 PM Bug #43248 (Resolved): cephfs-shell: do not drop into shell after running command-line command
- e.g....
- 09:34 PM Cleanup #42468 (Resolved): mds: reorg MDSTable header
- 09:33 PM Cleanup #42564 (Resolved): mds: reorg Migrator header
- 09:31 PM Cleanup #42793 (Resolved): mds: reorg PurgeQueue header
- 09:03 AM Documentation #43222 (Resolved): doc: mention multimds in dev guide's list of integration test su...
- 07:16 AM Documentation #43220 (In Progress): doc: clarify difference fs and kcephfs suite in dev guide
- 07:12 AM Documentation #43220 (Resolved): doc: clarify difference fs and kcephfs suite in dev guide
- 05:57 AM Backport #43219 (In Progress): nautilus: mgr/volumes: ERROR: test_subvolume_create_with_desired_u...
- 05:49 AM Backport #43219 (Resolved): nautilus: mgr/volumes: ERROR: test_subvolume_create_with_desired_uid_...
- https://github.com/ceph/ceph/pull/31741
- 05:44 AM Bug #43038 (Pending Backport): mgr/volumes: ERROR: test_subvolume_create_with_desired_uid_gid (ta...
- 05:42 AM Bug #43218 (Rejected): kclient: when looking up the snap dirs sometime will hit WARN_ON
- Hit this twice in 30 minutes, the following are the warning:
76 <7>[ 3254.346712] ceph: readdir fetching 100... - 05:22 AM Feature #4386: kclient: Mount error message when no MDS present
- And maybe we could return the -ESTALE or some other specified errornos to the userland to mount.ceph and then the mou...
- 05:20 AM Feature #4386: kclient: Mount error message when no MDS present
- Checked the new mount API, we still need the fix when the mount request timedout due to there is no any MDS is up or ...
- 03:52 AM Documentation #22204 (Resolved): doc: scrub_path is missing in the docs
- 12:37 AM Documentation #42016 (Resolved): doc: layout rest of intro page
12/09/2019
- 11:59 PM Bug #43216 (Resolved): MDSMonitor: removes MDS coming out of quorum election
- Event sequence:
- 2019-12-07T12:26:26.854 mon_thrash kills mon.a(leader)
- 2019-12-07T12:27:07.843 mon_thrash rev... - 10:12 PM Bug #43133 (Fix Under Review): vstop.sh: Mounts are not cleaned up
- 06:55 PM Feature #26996 (Fix Under Review): cephfs: get capability cache hits by clients to provide intros...
- 06:29 PM Bug #43191 (Fix Under Review): test_cephfs_shell: set `colors` to Never for cephfs-shell
- 06:17 AM Bug #43191 (Resolved): test_cephfs_shell: set `colors` to Never for cephfs-shell
- Originally, the plan was to use setUpClass and tearDownClass for tests[1] but I missed pushing that modification befo...
- 03:13 PM Documentation #43210 (In Progress): doc: MDS config reference improvements
- https://docs.ceph.com/docs/master/cephfs/mds-config-ref/
Add details on how to apply a configuration option, fetch... - 03:06 PM Backport #43085 (In Progress): nautilus: pybind / cephfs: remove static typing in LibCephFS.chown
- 02:56 PM Backport #43085 (New): nautilus: pybind / cephfs: remove static typing in LibCephFS.chown
- Reopening. I will update this PR with this fix: https://github.com/ceph/ceph/pull/31741
- 02:57 PM Feature #40929 (In Progress): pybind/mgr/mds_autoscaler: create mgr plugin to deploy and configur...
- 02:56 PM Fix #41782 (Fix Under Review): mds: allow stray directories to fragment and switch from 10 stray ...
- Update:
Stray dirs are not being dropped from 10 to 1. Zheng recommended having more stray dirs.
Only fragmentation... - 02:44 PM Bug #43208 (Fix Under Review): mds: unsafe req may result in data remaining in the datapool
- 12:56 PM Bug #43208 (Resolved): mds: unsafe req may result in data remaining in the datapool
- when client create file, if early_reply is set true, the metadata has not write to journal and the file data is succe...
- 02:42 PM Bug #43039: client: shutdown race fails with status 141
- (Handing back to Patrick for now)
Is this problem still occurring in teuthology? - 11:55 AM Feature #36253 (Fix Under Review): cephfs: clients should send usage metadata to MDSs for adminis...
- 11:54 AM Feature #24285 (Fix Under Review): mgr: add module which displays current usage of file system (`...
12/08/2019
12/07/2019
- 12:14 AM Feature #43182 (Resolved): mds: increase default cache size to 4GB
- 1GB is too low as a default and usually results in cache size warnings at that size; the MDS will struggle to maintai...
12/06/2019
- 10:50 PM Backport #42440: mimic: mds: create a configurable snapshot limit
- Nathan Cutler wrote:
> feature backport - does it need a release note?
Yes. - 12:58 PM Backport #42440 (Need More Info): mimic: mds: create a configurable snapshot limit
- feature backport - does it need a release note?
- 10:50 PM Backport #42441: nautilus: mds: create a configurable snapshot limit
- Nathan Cutler wrote:
> feature backport - does it need a release note?
Yes. - 12:58 PM Backport #42441 (Need More Info): nautilus: mds: create a configurable snapshot limit
- feature backport - does it need a release note?
- 01:32 PM Backport #43143 (In Progress): nautilus: mds: tolerate no snaprealm encoded in on-disk root inode
- 01:32 PM Backport #43141 (In Progress): nautilus: tools/cephfs: linkages injected by cephfs-data-scan have...
- 01:31 PM Backport #43138 (In Progress): nautilus: mds: reports unrecognized message for mgrclient messages
- 01:29 PM Backport #43137 (In Progress): nautilus: pybind/mgr/volumes: idle connection drop is not working
- 01:27 PM Bug #42923 (Resolved): pybind / cephfs: remove static typing in LibCephFS.chown
- 01:26 PM Backport #43085 (Rejected): nautilus: pybind / cephfs: remove static typing in LibCephFS.chown
- The code being fixed does not exist in nautilus.
- 01:25 PM Backport #43001 (In Progress): nautilus: qa: ignore "ceph.dir.pin: No such attribute" for (old) k...
- 01:21 PM Backport #42951 (In Progress): nautilus: Test failure: test_subvolume_snapshot_ls (tasks.cephfs.t...
- 01:21 PM Backport #42949 (In Progress): nautilus: mds: inode lock stuck at unstable state after evicting c...
- 01:20 PM Backport #43170 (In Progress): nautilus: nautilus: qa: ignore RECENT_CRASH for multimds snapshot ...
- 01:20 PM Backport #43170 (Resolved): nautilus: nautilus: qa: ignore RECENT_CRASH for multimds snapshot tes...
- https://github.com/ceph/ceph/pull/32072
- 01:19 PM Bug #42922 (Pending Backport): nautilus: qa: ignore RECENT_CRASH for multimds snapshot testing
- 01:18 PM Backport #42738 (Need More Info): nautilus: mgr/volumes: cleanup libcephfs handles on mgr shutdown
- 01:18 PM Backport #42713 (Need More Info): nautilus: mgr: daemon state for mds not available
- 01:17 PM Backport #42650 (In Progress): nautilus: mds: no assert on frozen dir when scrub path
- 12:58 PM Backport #42631 (In Progress): nautilus: client: FAILED assert(cap == in->auth_cap)
- 10:11 AM Bug #39947 (Resolved): cephfs-shell: add CI testing with flake8
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 10:11 AM Bug #40202 (Resolved): cephfs-shell: Error messages are printed to stdout
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 10:10 AM Bug #40430 (Resolved): cephfs-shell: No error message is printed on ls of invalid directories
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 10:10 AM Bug #40476 (Resolved): cephfs-shell: cd with no args has no effect
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 10:10 AM Bug #40836 (Resolved): cephfs-shell: flake8 blank line and indentation error
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 10:10 AM Cleanup #40992 (Resolved): cephfs-shell: Multiple flake8 errors
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 10:09 AM Bug #41164 (Resolved): cephfs-shell: onecmd throws TypeError
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
12/05/2019
- 11:58 PM Bug #42643 (Resolved): vstart.sh: highlight presence of stray conf file
- 10:01 PM Backport #41283 (Rejected): nautilus: cephfs-shell: No error message is printed on ls of invalid ...
- We're not backporting cephfs-shell fixes to Nautilus anymore.
- 10:01 PM Backport #41268 (Rejected): nautilus: cephfs-shell: onecmd throws TypeError
- We're not backporting cephfs-shell fixes to Nautilus anymore.
- 10:01 PM Backport #41118 (Rejected): nautilus: cephfs-shell: add CI testing with flake8
- We're not backporting cephfs-shell fixes to Nautilus anymore.
- 10:01 PM Backport #41112 (Rejected): nautilus: cephfs-shell: cd with no args has no effect
- We're not backporting cephfs-shell fixes to Nautilus anymore.
- 10:01 PM Backport #41105 (Rejected): nautilus: cephfs-shell: flake8 blank line and indentation error
- We're not backporting cephfs-shell fixes to Nautilus anymore.
- 10:01 PM Backport #41089 (Rejected): nautilus: cephfs-shell: Multiple flake8 errors
- We're not backporting cephfs-shell fixes to Nautilus anymore.
- 10:00 PM Backport #40898 (Rejected): nautilus: cephfs-shell: Error messages are printed to stdout
- 09:45 PM Bug #42894 (Fix Under Review): kclient: if there has at least one MDS still not laggy the mount w...
- 09:45 PM Bug #42760 (Fix Under Review): kclient: get random mds not work as expected
- 09:45 PM Bug #42515 (Fix Under Review): fs: OpenFileTable object shards have too many k/v pairs
- 09:37 PM Bug #42088 (New): 'ceph -s' does not show standbys if there are no filesystems
- 09:36 PM Bug #26901 (New): mds: no throttlers set on incoming messages
- 09:36 PM Bug #21507 (New): mds: debug logs near respawn are not flushed
- 09:36 PM Bug #21058 (New): mds: remove UNIX file permissions binary dependency
- 09:35 PM Bug #20938 (New): CephFS: concurrent access to file from multiple nodes blocks for seconds
- 09:35 PM Bug #19812 (New): client: not swapping directory caps efficiently leads to very slow create chains
- 09:35 PM Bug #18883 (New): qa: failures in samba suite
- 09:35 PM Bug #17847 (New): "Fuse mount failed to populate /sys/ after 31 seconds" in jewel 10.2.4
- 09:35 PM Bug #17594 (New): cephfs: permission checking not working (MDS should enforce POSIX permissions)
- 09:35 PM Bug #16881 (New): RuntimeError: Files in flight high water is unexpectedly low (0 / 6)
- 09:35 PM Bug #16920 (New): mds.inodes* perf counters sound like the number of inodes but they aren't
- 09:35 PM Bug #16556 (New): LibCephFS.InterProcessLocking failing on master and jewel
- 09:35 PM Bug #9105 (New): ~ObjectCacher behaves poorly on EBLACKLISTED
- 09:35 PM Bug #9101 (New): multimds: unlinked file is not pruned from replica mds caches
- 09:34 PM Bug #4023 (New): kclient: d_revalidate is abusing d_parent
- 09:34 PM Bug #2277 (New): qa: flock test broken
- 09:23 PM Bug #42252 (Rejected): mimic: test_21501 (tasks.cephfs.test_volume_client.TestVolumeClient) failure
- Fixed by https://github.com/ceph/ceph/pull/31017
- 06:56 PM Documentation #43162 (Resolved): doc: "adding an MDS" in deployment is out-of-date
- https://docs.ceph.com/docs/master/cephfs/add-remove-mds/#adding-an-mds
See: https://github.com/ceph/ceph/pull/32... - 06:13 PM Documentation #42016 (Fix Under Review): doc: layout rest of intro page
- 06:11 PM Documentation #43155 (Closed): CephFS Documentation Sprint 4
- 05:20 PM Documentation #43154 (Resolved): doc: migrate best practice recommendations to relevant docs
- Best practices doc:
https://docs.ceph.com/docs/master/cephfs/best-practices/
Should just put these recommendati... - 03:01 PM Bug #43149 (In Progress): kclient: umount will stuck for around 1 minutes sometimes
- During umount, if there the last request reply is only a safe one without unsafe, the umount won't have any chance to...
- 02:55 PM Bug #43149 (Resolved): kclient: umount will stuck for around 1 minutes sometimes
- While running some test, in one terminal run a script to do creating/deleting/listing large amount of directories wit...
- 02:54 PM Feature #38851 (Rejected): mount.ceph.fuse: support secretfile option
- 02:53 PM Bug #43061: ceph fs add_data_pool doesn't set pool metadata properly
- Ramana Raja wrote:
> [...]
> `add_data_pool` sets the pool's meta data properly if the pool's application metadata ... - 11:10 AM Backport #43144 (Rejected): mimic: mds: tolerate no snaprealm encoded in on-disk root inode
- 11:10 AM Backport #43143 (Resolved): nautilus: mds: tolerate no snaprealm encoded in on-disk root inode
- https://github.com/ceph/ceph/pull/32079
- 11:10 AM Backport #43142 (Rejected): mimic: tools/cephfs: linkages injected by cephfs-data-scan have first...
- 11:07 AM Backport #43141 (Resolved): nautilus: tools/cephfs: linkages injected by cephfs-data-scan have fi...
- https://github.com/ceph/ceph/pull/32078
- 11:07 AM Backport #43138 (Resolved): nautilus: mds: reports unrecognized message for mgrclient messages
- https://github.com/ceph/ceph/pull/32077
- 11:07 AM Backport #43137 (Resolved): nautilus: pybind/mgr/volumes: idle connection drop is not working
- https://github.com/ceph/ceph/pull/33116
- 08:58 AM Bug #36094: mds: crash(FAILED assert(omap_num_objs <= MAX_OBJECTS))
- For the sake of completeness, here the crash logging with the extra debug output:...
- 02:20 AM Bug #36094 (Fix Under Review): mds: crash(FAILED assert(omap_num_objs <= MAX_OBJECTS))
- 06:45 AM Bug #43133 (Resolved): vstop.sh: Mounts are not cleaned up
- When vstop is run while the cephFS is mounted, mount processes are retained and can't be killed.
And also the mount ...
12/04/2019
- 10:59 PM Bug #43129 (New): qa: `fs dump` fails during snaptests
- ...
- 10:54 PM Bug #42636 (Resolved): qa: AttributeError: can't set attribute
- 10:53 PM Bug #42636 (Pending Backport): qa: AttributeError: can't set attribute
- 10:50 PM Bug #43036 (Pending Backport): mds: reports unrecognized message for mgrclient messages
- 10:18 PM Bug #42675 (Pending Backport): mds: tolerate no snaprealm encoded in on-disk root inode
- 10:03 PM Bug #43125: qa: ceph_volume_client not available "ModuleNotFoundError: No module named 'ceph_volu...
- Can't seem to reproduce on master:
http://pulpito.ceph.com/pdonnell-2019-12-04_20:54:30-fs-master-distro-basic-smi... - 08:46 PM Bug #43125 (Can't reproduce): qa: ceph_volume_client not available "ModuleNotFoundError: No modul...
- ...
- 09:58 PM Bug #42829 (Pending Backport): tools/cephfs: linkages injected by cephfs-data-scan have first == ...
- 09:57 PM Bug #38452 (Resolved): mds: assert crash loop while unlinking file
- 09:45 PM Bug #43113 (Pending Backport): pybind/mgr/volumes: idle connection drop is not working
- 03:57 PM Documentation #16300 (Resolved): doc: fuse_disable_pagecache
Also available in: Atom