Activity
From 11/02/2021 to 12/01/2021
12/01/2021
- 07:35 PM Bug #53214: qa: "dd: error reading '/sys/kernel/debug/ceph/2a934501-6731-4052-a836-f42229a869be.c...
- This patch doesn't appear to be applicable to Pacific or Octopus since the get_op_read_count method doesn't exist in ...
- 04:09 AM Bug #53214 (Pending Backport): qa: "dd: error reading '/sys/kernel/debug/ceph/2a934501-6731-4052-...
- 10:54 AM Bug #53216 (Resolved): qa: "RuntimeError: value of attributes should be either str or None. clien...
- 09:50 AM Bug #53360: pacific: client: "handle_auth_bad_method server allowed_methods [2] but i only suppor...
- `volume_client` script threw a traceback with "ModuleNotFoundError: No module named 'ceph_volume_client'"::...
- 09:26 AM Bug #53360: pacific: client: "handle_auth_bad_method server allowed_methods [2] but i only suppor...
- From monitor logs::...
- 07:27 AM Bug #50223: qa: "client.4737 isn't responding to mclientcaps(revoke)"
- Recent instances:
http://pulpito.front.sepia.ceph.com/yuriw-2021-11-08_15:19:37-fs-wip-yuri2-testing-2021-11-06-1322... - 06:31 AM Bug #40002 (In Progress): mds: not trim log under heavy load
- 06:29 AM Feature #10764 (In Progress): optimize memory usage of MDSCacheObject
- 06:21 AM Bug #52397 (Resolved): pacific: qa: test_acls (tasks.cephfs.test_acls.TestACLs) failed
- This has been fixed by the following commit in git://git.ceph.com/xfstests-dev.git:...
- 04:18 AM Bug #53436 (Duplicate): mds, mon: mds beacon messages get dropped? (mds never reaches up:active s...
- This is a known bug long time ago.
- 02:26 AM Bug #53436: mds, mon: mds beacon messages get dropped? (mds never reaches up:active state)
- From remote/smithi154/log/ceph-mds.d.log.gz, we can see that the mon.1 connection was broken:...
- 04:10 AM Backport #53446 (Rejected): octopus: mds: opening connection to up:replay/up:creating daemon caus...
- 04:10 AM Backport #53445 (Resolved): pacific: mds: opening connection to up:replay/up:creating daemon caus...
- https://github.com/ceph/ceph/pull/44296
- 04:10 AM Backport #53444 (Resolved): octopus: qa: "dd: error reading '/sys/kernel/debug/ceph/2a934501-6731...
- https://github.com/ceph/ceph/pull/44270
- 04:10 AM Backport #53443 (New): pacific: qa: "dd: error reading '/sys/kernel/debug/ceph/2a934501-6731-4052...
- 04:09 AM Bug #53082 (Resolved): ceph-fuse: segmenetation fault in Client::handle_mds_map
- 04:07 AM Bug #53194 (Pending Backport): mds: opening connection to up:replay/up:creating daemon causes mes...
- 04:06 AM Feature #52725 (Resolved): qa: mds_dir_max_entries workunit test case
- 04:05 AM Feature #47277 (Resolved): implement new mount "device" syntax for kcephfs
11/30/2021
- 01:48 PM Bug #53436: mds, mon: mds beacon messages get dropped? (mds never reaches up:active state)
- Seems the same issue with https://tracker.ceph.com/issues/51705.
- 01:39 PM Bug #53436 (Triaged): mds, mon: mds beacon messages get dropped? (mds never reaches up:active state)
- 12:46 PM Bug #53436 (Duplicate): mds, mon: mds beacon messages get dropped? (mds never reaches up:active s...
- Seen in this run - https://pulpito.ceph.com/vshankar-2021-11-24_07:14:27-fs-wip-vshankar-testing-20211124-094330-test...
- 11:37 AM Feature #40633 (In Progress): mds: dump recent log events for extraordinary events
- 09:20 AM Bug #48711 (Closed): mds: standby-replay mds abort when replay metablob
- No updates from haitao yet, closing this.
- 01:31 AM Bug #16739 (Fix Under Review): Client::setxattr always sends setxattr request to MDS
- 01:17 AM Feature #18514 (Resolved): qa: don't use a node for each kclient
- I had pushed several patches by adding unsharing network namespace support to fix this. More detail please see https:...
11/29/2021
- 11:21 AM Bug #52625 (Resolved): qa: test_kill_mdstable (tasks.cephfs.test_snapshots.TestSnapshots)
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 11:20 AM Bug #52949 (Resolved): RuntimeError: The following counters failed to be set on mds daemons: {'md...
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 11:20 AM Bug #52975 (Resolved): MDSMonitor: no active MDS after cluster deployment
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 11:20 AM Bug #52994 (Resolved): client: do not defer releasing caps when revoking
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 11:19 AM Bug #53155 (Resolved): MDSMonitor: assertion during upgrade to v16.2.5+
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 11:17 AM Backport #53121: pacific: mds: collect I/O sizes from client for cephfs-top
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43784
m... - 11:15 AM Backport #53217 (Resolved): pacific: test: Implement cephfs-mirror trasher test for HA active/active
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43924
m... - 11:15 AM Backport #53164: pacific: mds: tcmalloc::allocate_full_cpp_throw_oom(unsigned long)+0xf3)
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43815
m... - 11:15 AM Backport #52678 (Resolved): pacific: qa: test_kill_mdstable (tasks.cephfs.test_snapshots.TestSnap...
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43702
m... - 11:14 AM Backport #53231 (Resolved): pacific: MDSMonitor: assertion during upgrade to v16.2.5+
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43890
m... - 11:14 AM Backport #53006: pacific: RuntimeError: The following counters failed to be set on mds daemons: {...
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43828
m... - 05:40 AM Bug #39634 (Fix Under Review): qa: test_full_same_file timeout
- 05:34 AM Bug #39634: qa: test_full_same_file timeout
- In /ceph/teuthology-archive/yuriw-2021-11-08_20:21:11-fs-wip-yuri5-testing-2021-11-08-1003-pacific-distro-basic-smith...
11/26/2021
- 09:50 AM Bug #48673: High memory usage on standby replay MDS
- Patrick Donnelly wrote:
> I've been able to reproduce this. Will try to track down the cause...
The same situatio...
11/25/2021
- 05:41 AM Bug #48812 (New): qa: test_scrub_pause_and_resume_with_abort failure
- This has started to show up again: https://pulpito.ceph.com/vshankar-2021-11-24_07:14:27-fs-wip-vshankar-testing-2021...
- 05:34 AM Bug #39634: qa: test_full_same_file timeout
- Checked all the other OSDs, they all didn't reach the "mon osd full ratio: 0.7". Only the osd.4 did.
That means the ... - 05:30 AM Bug #39634: qa: test_full_same_file timeout
- When the test_full test case was deleting the "large_file_b" and "large_file_a", from /ceph/teuthology-archive/yuriw-...
11/24/2021
- 11:40 AM Feature #53310: Add admin socket command to trim caps
- Patrick mentioned about these config options:
- mds_session_cache_liveness_decay_rate
- mds_session_cache_livenes... - 11:22 AM Bug #53360: pacific: client: "handle_auth_bad_method server allowed_methods [2] but i only suppor...
- ceph-fuse fails way before `install.upgrade` in run. Looks like the failure is when everything is nautilus::...
11/23/2021
- 06:18 PM Fix #52591 (Fix Under Review): mds: mds_oft_prefetch_dirfrags = false is not qa tested
- 04:08 PM Bug #52094 (Duplicate): Tried out Quincy: All MDS Standby
- 01:54 PM Feature #53310: Add admin socket command to trim caps
- Brief background - this request came up from some community members. They run a file system scanning job every day (?...
- 01:40 PM Bug #53360 (Triaged): pacific: client: "handle_auth_bad_method server allowed_methods [2] but i o...
- 01:22 PM Bug #52487 (Fix Under Review): qa: Test failure: test_deep_split (tasks.cephfs.test_fragment.Test...
- 01:20 PM Bug #53300 (Duplicate): qa: cluster [WRN] Scrub error on inode
- Duplicate of https://tracker.ceph.com/issues/50250
- 09:24 AM Bug #50946 (Duplicate): mgr/stats: exception ValueError in perf stats
- 09:19 AM Bug #50946: mgr/stats: exception ValueError in perf stats
- This issue seems duplicate of https://tracker.ceph.com/issues/48473
It will automatically get resolved once https://... - 04:17 AM Feature #49811 (Resolved): mds: collect I/O sizes from client for cephfs-top
- 04:15 AM Backport #53121 (Resolved): pacific: mds: collect I/O sizes from client for cephfs-top
11/22/2021
- 05:37 PM Documentation #53236 (Resolved): doc: ephemeral pinning with subvolumegroups
- 05:36 PM Backport #53245 (Resolved): pacific: doc: ephemeral pinning with subvolumegroups
- 05:33 PM Bug #53360 (Duplicate): pacific: client: "handle_auth_bad_method server allowed_methods [2] but i...
- Nautilus ceph-fuse client fails to start for Pacific upgrade tests:...
- 02:58 PM Backport #53121: pacific: mds: collect I/O sizes from client for cephfs-top
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43784
merged - 02:27 PM Bug #53293 (Resolved): qa: v16.2.4 mds crash caused by centos stream kernel
- 02:41 AM Bug #53082 (Fix Under Review): ceph-fuse: segmenetation fault in Client::handle_mds_map
- 01:24 AM Backport #53164 (Resolved): pacific: mds: tcmalloc::allocate_full_cpp_throw_oom(unsigned long)+0xf3)
11/20/2021
- 05:03 PM Backport #53347 (Resolved): pacific: qa: v16.2.4 mds crash caused by centos stream kernel
- 12:47 AM Backport #53347 (In Progress): pacific: qa: v16.2.4 mds crash caused by centos stream kernel
11/19/2021
- 11:45 PM Backport #53347 (Resolved): pacific: qa: v16.2.4 mds crash caused by centos stream kernel
- https://github.com/ceph/ceph/pull/44034
- 11:44 PM Bug #53293 (Pending Backport): qa: v16.2.4 mds crash caused by centos stream kernel
- 02:17 PM Backport #53217: pacific: test: Implement cephfs-mirror trasher test for HA active/active
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43924
merged - 02:14 PM Backport #53164: pacific: mds: tcmalloc::allocate_full_cpp_throw_oom(unsigned long)+0xf3)
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43815
merged - 04:45 AM Backport #53332 (Resolved): pacific: ceph-fuse seems to need root permissions to mount (ceph-fuse...
- https://github.com/ceph/ceph/pull/44272
- 04:45 AM Backport #53331 (Resolved): octopus: ceph-fuse seems to need root permissions to mount (ceph-fuse...
- https://github.com/ceph/ceph/pull/44271
- 04:41 AM Documentation #53054 (Pending Backport): ceph-fuse seems to need root permissions to mount (ceph-...
11/18/2021
- 03:23 PM Backport #53120: pacific: client: do not defer releasing caps when revoking
- https://github.com/ceph/ceph/pull/43782 merged
- 03:21 PM Backport #52678: pacific: qa: test_kill_mdstable (tasks.cephfs.test_snapshots.TestSnapshots)
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43702
merged - 02:51 PM Bug #48473 (Fix Under Review): fs perf stats command crashes
- 02:31 PM Bug #53314 (Duplicate): qa: fs/upgrade/mds_upgrade_sequence test timeout
- 09:12 AM Bug #53314: qa: fs/upgrade/mds_upgrade_sequence test timeout
- @Xiubo, I think the PR https://github.com/ceph/ceph/pull/43784 is causing this.
- 09:09 AM Bug #53314 (Duplicate): qa: fs/upgrade/mds_upgrade_sequence test timeout
- The qa suite mds_upgrade_sequence becomes dead with job timeout because of mds crash.
-------------
ceph versi... - 12:31 PM Bug #48773: qa: scrub does not complete
- Another Instance
http://qa-proxy.ceph.com/teuthology/yuriw-2021-11-17_19:02:43-fs-wip-yuri10-testing-2021-11-17-08... - 08:50 AM Bug #39634: qa: test_full_same_file timeout
- Will work on it.
- 08:47 AM Bug #52396 (Duplicate): pacific: qa: ERROR: test_perf_counters (tasks.cephfs.test_openfiletable.O...
- This is duplicated to https://tracker.ceph.com/issues/53218.
- 03:55 AM Cleanup #51406 (Fix Under Review): mgr/volumes/fs/operations/versions/op_sm.py: fix various flake...
11/17/2021
- 09:29 PM Feature #53310 (New): Add admin socket command to trim caps
- Add an admin socket command to cause the MDS to reclaim state from a client. This would simply involve calling reclai...
- 04:30 PM Backport #53232 (Resolved): pacific: MDSMonitor: no active MDS after cluster deployment
- 03:59 PM Backport #53232: pacific: MDSMonitor: no active MDS after cluster deployment
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43891
merged - 04:14 PM Backport #53231: pacific: MDSMonitor: assertion during upgrade to v16.2.5+
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43890
merged - 03:49 PM Backport #53006 (Resolved): pacific: RuntimeError: The following counters failed to be set on mds...
- 03:16 PM Backport #53006: pacific: RuntimeError: The following counters failed to be set on mds daemons: {...
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43828
merged - 03:48 PM Backport #53120 (Resolved): pacific: client: do not defer releasing caps when revoking
- 01:41 PM Bug #50946: mgr/stats: exception ValueError in perf stats
- Nikhil, please take this one.
- 11:00 AM Backport #53304 (New): octopus: Improve API documentation for struct ceph_client_callback_args
- 11:00 AM Backport #53303 (New): pacific: Improve API documentation for struct ceph_client_callback_args
- 10:57 AM Documentation #53004 (Pending Backport): Improve API documentation for struct ceph_client_callbac...
- 10:56 AM Bug #51964: qa: test_cephfs_mirror_restart_sync_on_blocklist failure
- I'll take a look at the failure soon.
- 10:35 AM Bug #51964: qa: test_cephfs_mirror_restart_sync_on_blocklist failure
- Also seen in this run
http://pulpito.front.sepia.ceph.com/yuriw-2021-11-08_15:19:37-fs-wip-yuri2-testing-2021-11-0... - 08:46 AM Bug #53082 (In Progress): ceph-fuse: segmenetation fault in Client::handle_mds_map
- 07:35 AM Bug #53300 (Duplicate): qa: cluster [WRN] Scrub error on inode
- "2021-11-09T18:59:42.703093+0000 mds.l (mds.0) 19 : cluster [WRN] Scrub error on inode 0x100000012cc (/client.0/tmp/b...
- 07:18 AM Bug #39634: qa: test_full_same_file timeout
- Seen in the below pacific run
http://pulpito.front.sepia.ceph.com/yuriw-2021-11-08_20:21:11-fs-wip-yuri5-testing-2... - 06:42 AM Bug #51705: qa: tasks.cephfs.fuse_mount:mount command failed
- New instance
http://pulpito.front.sepia.ceph.com/yuriw-2021-11-12_00:33:28-fs-wip-yuri7-testing-2021-11-11-1339-pa... - 06:34 AM Backport #52875: pacific: qa: test_dirfrag_limit
- Seen in this pacific test run as well. Should go away once the fix is backported
http://pulpito.front.sepia.ceph.c... - 06:25 AM Bug #52396: pacific: qa: ERROR: test_perf_counters (tasks.cephfs.test_openfiletable.OpenFileTable)
- Seen in this pacific run as well
http://pulpito.front.sepia.ceph.com/yuriw-2021-11-12_00:33:28-fs-wip-yuri7-testin... - 06:00 AM Bug #53216 (Fix Under Review): qa: "RuntimeError: value of attributes should be either str or Non...
- The parameters' order is incorrect and missing the 'client_config'.
- 04:38 AM Backport #53218 (In Progress): pacific: qa: Test failure: test_perf_counters (tasks.cephfs.test_o...
- 01:04 AM Bug #53192: High cephfs MDS latency and CPU load with snapshots and unlink operations
- Andras Pataki wrote:
> Thanks Patrick! How safe do you feel this patch is? Does it need a lot of testing or is it ...
11/16/2021
- 08:13 PM Bug #53293 (Fix Under Review): qa: v16.2.4 mds crash caused by centos stream kernel
- 07:59 PM Bug #53293 (Resolved): qa: v16.2.4 mds crash caused by centos stream kernel
- breaks fs:upgrade:mds_upgrade_sequence tests:
http://pulpito.front.sepia.ceph.com/yuriw-2021-11-13_15:31:06-rados-... - 07:23 AM Backport #52633 (Resolved): pacific: mds,client: add flag to MClientSession for reject reason
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43251
m... - 07:23 AM Backport #52628 (Resolved): pacific: pybind/mgr/volumes: first subvolume permissions set perms on...
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43223
m... - 04:37 AM Bug #48473: fs perf stats command crashes
- Nikhil, please take this one.
- 04:34 AM Documentation #53054 (Fix Under Review): ceph-fuse seems to need root permissions to mount (ceph-...
11/15/2021
- 05:42 PM Bug #53192: High cephfs MDS latency and CPU load with snapshots and unlink operations
- Thanks Patrick! How safe do you feel this patch is? Does it need a lot of testing or is it safe to deploy?
- 01:52 PM Bug #53179 (Triaged): Crash when unlink in corrupted cephfs
- Most likely the same as: https://tracker.ceph.com/issues/41147.
I guess this should have been fixed a while back. ... - 01:50 PM Bug #53216 (Triaged): qa: "RuntimeError: value of attributes should be either str or None. client...
- 01:46 PM Backport #53245 (In Progress): pacific: doc: ephemeral pinning with subvolumegroups
- 01:38 PM Bug #53246 (Triaged): rhel 8.4 and centos stream unable to install cephfs-java
- 01:06 PM Backport #53217 (In Progress): pacific: test: Implement cephfs-mirror trasher test for HA active/...
11/12/2021
- 06:25 PM Bug #53192: High cephfs MDS latency and CPU load with snapshots and unlink operations
- Andras, I think there are other inefficiencies I've not yet identified but this fix addresses your specific problem.
- 06:09 PM Bug #53192 (Fix Under Review): High cephfs MDS latency and CPU load with snapshots and unlink ope...
- 10:28 AM Bug #53192: High cephfs MDS latency and CPU load with snapshots and unlink operations
- That's great news! I'm also trying to set up a test environment to reproduce the issue outside our main cluster (eye...
- 12:19 AM Bug #53192: High cephfs MDS latency and CPU load with snapshots and unlink operations
- Just want to share an update that I was able to reproduce the problem. I'll test out a fix soon...
- 03:51 PM Bug #53246 (Triaged): rhel 8.4 and centos stream unable to install cephfs-java
- ...
- 03:25 PM Backport #53245 (Resolved): pacific: doc: ephemeral pinning with subvolumegroups
- https://github.com/ceph/ceph/pull/43925
- 03:20 PM Documentation #53236 (Pending Backport): doc: ephemeral pinning with subvolumegroups
- 06:09 AM Fix #48683: mds/MDSMap: print each flag value in MDSMap::dump
- Sven Kieske wrote:
> any chance to backport this, at least to pacific or octopus?
This is a feature (classified w...
11/11/2021
- 07:25 PM Bug #53192: High cephfs MDS latency and CPU load with snapshots and unlink operations
- Thanks for opening a ticket and doing some digging, Andras,
Andras Pataki wrote:
> I've done some further debuggi... - 05:31 PM Documentation #53236 (Fix Under Review): doc: ephemeral pinning with subvolumegroups
- 05:04 PM Documentation #53236 (Resolved): doc: ephemeral pinning with subvolumegroups
- 05:00 PM Fix #48683: mds/MDSMap: print each flag value in MDSMap::dump
- any chance to backport this, at least to pacific or octopus?
- 03:39 PM Backport #53232 (In Progress): pacific: MDSMonitor: no active MDS after cluster deployment
- 03:20 PM Backport #53232 (Resolved): pacific: MDSMonitor: no active MDS after cluster deployment
- https://github.com/ceph/ceph/pull/43891
- 03:38 PM Backport #53231 (In Progress): pacific: MDSMonitor: assertion during upgrade to v16.2.5+
- 03:20 PM Backport #53231 (Resolved): pacific: MDSMonitor: assertion during upgrade to v16.2.5+
- https://github.com/ceph/ceph/pull/43890
- 03:18 PM Bug #52975 (Pending Backport): MDSMonitor: no active MDS after cluster deployment
- 03:17 PM Bug #53155 (Pending Backport): MDSMonitor: assertion during upgrade to v16.2.5+
- 03:17 PM Bug #53150 (Resolved): pybind/mgr/cephadm/upgrade: tolerate MDS failures during upgrade straddlin...
- backport will be tracked by #53155
- 11:07 AM Feature #46166 (Fix Under Review): mds: store symlink target as xattr in data pool inode for disa...
- 07:01 AM Feature #53228 (New): cephfs/quota: Set a limit on minimum quota setting
- As of now there is no minimum limit on quota setting. This can be set to lower than the
CEPH_BLOCK size (i.e. 4M) a...
11/10/2021
- 10:41 PM Bug #48805: mds: "cluster [WRN] Scrub error on inode 0x1000000039d (/client.0/tmp/blogbench-1.0/s...
- Came across a failure that looks related to this one in a recent Pacific run: http://pulpito.front.sepia.ceph.com/yur...
- 10:38 PM Backport #50253: pacific: mds: "cluster [WRN] Scrub error on inode 0x1000000039d (/client.0/tmp/b...
- Saw another failure related to this issue in a recent Pacific run: http://pulpito.front.sepia.ceph.com/yuriw-2021-11-...
- 10:18 PM Bug #23797: qa: cluster [WRN] Health check failed: 1 osds down (OSD_DOWN)
- Related failure found in a Pacific run: http://pulpito.front.sepia.ceph.com/yuriw-2021-11-02_19:49:55-rados-wip-yuri8...
- 07:05 PM Backport #53218 (Resolved): pacific: qa: Test failure: test_perf_counters (tasks.cephfs.test_open...
- https://github.com/ceph/ceph/pull/43979
- 07:01 PM Backport #53217 (Resolved): pacific: test: Implement cephfs-mirror trasher test for HA active/active
- https://github.com/ceph/ceph/pull/43924
- 07:00 PM Bug #52887 (Pending Backport): qa: Test failure: test_perf_counters (tasks.cephfs.test_openfileta...
- 06:59 PM Feature #50372 (Pending Backport): test: Implement cephfs-mirror trasher test for HA active/active
- 06:55 PM Bug #53216 (Resolved): qa: "RuntimeError: value of attributes should be either str or None. clien...
- ...
- 06:18 PM Backport #52633: pacific: mds,client: add flag to MClientSession for reject reason
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43251
merged - 06:18 PM Backport #52628: pacific: pybind/mgr/volumes: first subvolume permissions set perms on /volumes a...
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43223
merged - 06:17 PM Bug #53214 (Fix Under Review): qa: "dd: error reading '/sys/kernel/debug/ceph/2a934501-6731-4052-...
- 05:59 PM Bug #53214: qa: "dd: error reading '/sys/kernel/debug/ceph/2a934501-6731-4052-a836-f42229a869be.c...
- Ahh, that's because it _is_ a directory now, since this patch:
https://lore.kernel.org/ceph-devel/19b30242-15ed-87... - 05:32 PM Bug #53214 (Resolved): qa: "dd: error reading '/sys/kernel/debug/ceph/2a934501-6731-4052-a836-f42...
- ...
- 01:23 PM Bug #53045: stat->fsid is not unique among filesystems exported by the ceph server
- I have patches for the userland client that mimic what the kernel does, but building ceph under f35 is currently brok...
11/09/2021
- 04:35 PM Bug #53192: High cephfs MDS latency and CPU load with snapshots and unlink operations
- I've done some further debugging to understand the MDS performance problem that has been impacting us more. The fini...
- 01:14 AM Bug #52975 (Fix Under Review): MDSMonitor: no active MDS after cluster deployment
11/08/2021
- 10:07 PM Bug #53194 (Fix Under Review): mds: opening connection to up:replay/up:creating daemon causes mes...
- 05:49 PM Bug #53194 (Resolved): mds: opening connection to up:replay/up:creating daemon causes message drop
- Found a QA run where MDS was stuck in up:resolve:
https://pulpito.ceph.com/pdonnell-2021-11-05_19:13:39-fs:upgrade... - 04:52 PM Bug #53192 (Fix Under Review): High cephfs MDS latency and CPU load with snapshots and unlink ope...
- We have recently enabled snapshots on our large Nautilus cluster (running 14.2.20) and our fairly smooth running ceph...
- 02:35 PM Backport #52953 (In Progress): octopus: mds: crash when journaling during replay
- 02:26 PM Backport #52952 (In Progress): pacific: mds: crash when journaling during replay
- 02:23 PM Bug #52975 (In Progress): MDSMonitor: no active MDS after cluster deployment
- 04:21 AM Bug #52975: MDSMonitor: no active MDS after cluster deployment
- Thanks for the reproducer Igor.
commit cbd9a7b354abb06cd395753f93564bdc687cdb04 ("mon,mds: use per-MDS compat to i... - 01:07 AM Bug #49132: mds crashed "assert_condition": "state == LOCK_XLOCK || state == LOCK_XLOCKDONE",
- Andras Pataki wrote:
> I haven't had luck keeping the MDS running well with higher log levels unfortunately. Howeve...
11/06/2021
- 02:54 AM Bug #49922: MDS slow request lookupino #0x100 on rank 1 block forever on dispatched
- We are triggering the new warning: https://tracker.ceph.com/issues/53180
- 12:54 AM Bug #53179 (Duplicate): Crash when unlink in corrupted cephfs
- We have a corrupted cephfs that breaks every time after the repair when files are removed....
11/05/2021
- 08:14 PM Backport #53006 (In Progress): pacific: RuntimeError: The following counters failed to be set on ...
- 08:02 PM Backport #53006 (Need More Info): pacific: RuntimeError: The following counters failed to be set ...
- I will need to work on this because it pulls in some commits that don't have a tracker assigned.
- 03:17 AM Backport #53163 (In Progress): octopus: mds: tcmalloc::allocate_full_cpp_throw_oom(unsigned long)...
- 02:56 AM Backport #53164 (In Progress): pacific: mds: tcmalloc::allocate_full_cpp_throw_oom(unsigned long)...
11/04/2021
- 09:00 PM Backport #53165 (New): pacific: qa/vstart_runner: tests crashes due incompatiblity
- 08:56 PM Backport #53164 (Resolved): pacific: mds: tcmalloc::allocate_full_cpp_throw_oom(unsigned long)+0xf3)
- https://github.com/ceph/ceph/pull/43815
- 08:56 PM Backport #53163 (Resolved): octopus: mds: tcmalloc::allocate_full_cpp_throw_oom(unsigned long)+0xf3)
- https://github.com/ceph/ceph/pull/43816
- 08:55 PM Backport #53162 (In Progress): pacific: qa: test_standby_count_wanted failure
- https://github.com/ceph/ceph/pull/50760
- 08:55 PM Bug #53043 (Pending Backport): qa/vstart_runner: tests crashes due incompatiblity
- 08:52 PM Bug #52995 (Pending Backport): qa: test_standby_count_wanted failure
- 08:51 PM Bug #51023 (Pending Backport): mds: tcmalloc::allocate_full_cpp_throw_oom(unsigned long)+0xf3)
- 08:29 PM Bug #51964: qa: test_cephfs_mirror_restart_sync_on_blocklist failure
- /ceph/teuthology-archive/pdonnell-2021-11-04_15:43:53-fs-wip-pdonnell-testing-20211103.023355-distro-basic-smithi/648...
- 02:27 PM Bug #53155 (Fix Under Review): MDSMonitor: assertion during upgrade to v16.2.5+
- 02:21 PM Bug #53155 (Resolved): MDSMonitor: assertion during upgrade to v16.2.5+
- ...
11/03/2021
- 11:44 PM Bug #53150 (Fix Under Review): pybind/mgr/cephadm/upgrade: tolerate MDS failures during upgrade s...
- 08:38 PM Bug #53150 (Resolved): pybind/mgr/cephadm/upgrade: tolerate MDS failures during upgrade straddlin...
- If a v16.2.4 or older MDS fails and rejoins, the compat set assigned to it is the empty set (because it sends no comp...
- 11:49 AM Backport #52823 (In Progress): pacific: mgr/nfs: add more log messages
- 11:43 AM Bug #53074 (Resolved): pybind/mgr/cephadm: upgrade sequence does not continue if no MDS are active
- 10:42 AM Bug #51824: pacific scrub ~mds_dir causes stray related ceph_assert, abort and OOM
- Main Name wrote:
> Same issue with roughly 1.6M folders.
>
> * Generated a Folder tree with 1611111 Folders
> * ... - 09:52 AM Bug #51824: pacific scrub ~mds_dir causes stray related ceph_assert, abort and OOM
- Same issue with roughly 1.6M folders.
* Generated a Folder tree with 1611111 Folders
* Make snapshot
* Delete Fo... - 09:19 AM Bug #49132: mds crashed "assert_condition": "state == LOCK_XLOCK || state == LOCK_XLOCKDONE",
- I haven't had luck keeping the MDS running well with higher log levels unfortunately. However, I do have one more da...
- 06:57 AM Bug #53126 (Triaged): In the 5.4.0 kernel, the mount of ceph-fuse fails
- 06:54 AM Bug #52487 (In Progress): qa: Test failure: test_deep_split (tasks.cephfs.test_fragment.TestFragm...
- The check here[0] results in `num_strays` being zero _right after_ the journal was flushed::...
- 06:36 AM Feature #50372 (Fix Under Review): test: Implement cephfs-mirror trasher test for HA active/active
- 06:35 AM Backport #51415 (In Progress): octopus: mds: "FAILED ceph_assert(r == 0 || r == -2)"
- 02:55 AM Backport #53121 (In Progress): pacific: mds: collect I/O sizes from client for cephfs-top
- 02:53 AM Backport #53120 (In Progress): pacific: client: do not defer releasing caps when revoking
11/02/2021
- 10:15 PM Bug #50622 (Resolved): msg: active_connections regression
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 10:09 PM Bug #52572 (Resolved): "cluster [WRN] 1 slow requests" in smoke pacific
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 10:08 PM Bug #52820 (Resolved): Ceph monitor crash after upgrade from ceph 15.2.14 to 16.2.6
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 10:07 PM Bug #52874 (Resolved): Monitor might crash after upgrade from ceph to 16.2.6
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 10:05 PM Backport #52999 (Resolved): pacific: Ceph monitor crash after upgrade from ceph 15.2.14 to 16.2.6
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43615
m... - 01:34 PM Backport #52999: pacific: Ceph monitor crash after upgrade from ceph 15.2.14 to 16.2.6
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43615
merged - 10:04 PM Backport #52998 (Resolved): pacific: Monitor might crash after upgrade from ceph to 16.2.6
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43614
m... - 01:33 PM Backport #52998: pacific: Monitor might crash after upgrade from ceph to 16.2.6
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43614
merged - 10:03 PM Backport #52679: pacific: "cluster [WRN] 1 slow requests" in smoke pacific
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43562
m... - 09:54 PM Backport #51199 (Resolved): octopus: msg: active_connections regression
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43310
m... - 03:25 PM Bug #53126: In the 5.4.0 kernel, the mount of ceph-fuse fails
- Might be related to #53082
- 06:41 AM Bug #53126 (Closed): In the 5.4.0 kernel, the mount of ceph-fuse fails
- Hello everyone,
I use ubuntu18.04.5 server and the ceph version is 14.2.22.
After upgrading the kernel to 5.4.0, th... - 01:41 PM Bug #53082: ceph-fuse: segmenetation fault in Client::handle_mds_map
- Venky, I will take it.
- 03:19 AM Bug #52887 (Fix Under Review): qa: Test failure: test_perf_counters (tasks.cephfs.test_openfileta...
- 03:02 AM Bug #52887: qa: Test failure: test_perf_counters (tasks.cephfs.test_openfiletable.OpenFileTable)
- The `self.wait_until_true(lambda: self._check_oft_counter('omap_total_removes', 1), timeout=30)` last check was at `2...
Also available in: Atom