Project

General

Profile

Activity

From 07/14/2022 to 08/12/2022

08/12/2022

05:22 PM Documentation #57062: Document access patterns that have good/pathological performance on CephFS
I think that a good place for this info to be added would be https://docs.ceph.com/en/quincy/cephfs/app-best-practice... Niklas Hambuechen
12:03 PM Documentation #57115 (New): Explanation for cache pressure
Following up on the "thread":https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/DOUQNI5YQ74YB3FS5ZOQI2MS... Eugen Block
09:54 AM Bug #56996: Transient data read corruption from other machine
Confirmed with Venky, when the **CInode::filelock** is in **LOCK_MIX** state we won't guarantee the data consistency ... Xiubo Li
09:31 AM Feature #40633 (Resolved): mds: dump recent log events for extraordinary events
Venky Shankar
09:23 AM Backport #57113 (Resolved): pacific: Intermittent ParsingError failure in mgr/volumes module dur...
https://github.com/ceph/ceph/pull/47112 Backport Bot
09:23 AM Backport #57112 (In Progress): quincy: Intermittent ParsingError failure in mgr/volumes module d...
https://github.com/ceph/ceph/pull/47747 Backport Bot
09:12 AM Bug #55583 (Pending Backport): Intermittent ParsingError failure in mgr/volumes module during "c...
Venky Shankar
09:11 AM Backport #57111 (Resolved): quincy: mds: handle deferred client request core when mds reboot
https://github.com/ceph/ceph/pull/53363 Backport Bot
09:11 AM Backport #57110 (Resolved): pacific: mds: handle deferred client request core when mds reboot
https://github.com/ceph/ceph/pull/53362 Backport Bot
09:10 AM Bug #56116 (Pending Backport): mds: handle deferred client request core when mds reboot
Venky Shankar
05:54 AM Bug #54460: snaptest-multiple-capsnaps.sh test failure
Milind Changire wrote:
> client-type: fuse
>
> * Iteratively running shell scripts under *qa/workunits/fs/snaps/*...
Venky Shankar

08/11/2022

04:08 PM Bug #57048: osdc/Journaler: better handle ENOENT during replay as up:standby-replay
Greg Farnum wrote:
> Patrick Donnelly wrote:
> > Venky Shankar wrote:
> > > Patrick,
> > >
> > > Do you mean a ...
Patrick Donnelly
03:12 PM Bug #57048: osdc/Journaler: better handle ENOENT during replay as up:standby-replay
Patrick Donnelly wrote:
> Venky Shankar wrote:
> > Patrick,
> >
> > Do you mean a standby-replay MDS should tole...
Greg Farnum
01:37 PM Bug #57048: osdc/Journaler: better handle ENOENT during replay as up:standby-replay
Venky Shankar wrote:
> Patrick,
>
> Do you mean a standby-replay MDS should tolerate missing journal objects?
...
Patrick Donnelly
01:40 PM Backport #51337 (Rejected): nautilus: mds: avoid journaling overhead for setxattr("ceph.dir.subvo...
Patrick Donnelly
06:56 AM Bug #57087: qa: test_fragmented_injection (tasks.cephfs.test_data_scan.TestDataScan) failure
Note that the test successfully passed on the re-run
https://pulpito.ceph.com/yuriw-2022-08-10_20:34:29-fs-wip-yuri6...
Kotresh Hiremath Ravishankar
03:44 AM Bug #54253: Avoid OOM exceeding 10x MDS cache limit on restart after many files were opened
Unfortunately I must report that I'm still hitting this issue even with Ceph 16.2.7 and... Niklas Hambuechen

08/10/2022

05:30 PM Feature #56140 (Fix Under Review): cephfs: tooling to identify inode (metadata) corruption
Patrick Donnelly
05:20 PM Feature #57091 (Resolved): mds: modify scrub to catch dentry corruption
Such as "first" snapshot being an invalid value. Patrick Donnelly
05:01 PM Feature #57090 (Resolved): MDSMonitor,mds: add MDSMap flag to prevent clients from connecting
During some recovery situations, it would be useful to have MDS up but prevent clients from establishing sessions. Us... Patrick Donnelly
03:18 PM Backport #56979: quincy: mgr/volumes: Subvolume creation failed on FIPs enabled system
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/47368
mergedReviewed-by: Ramana Raja <rraja@redhat.com>
Yuri Weinstein
02:34 PM Bug #55216 (Resolved): cephfs-shell: creates directories in local file system even if file not found
PR along with backport PRs merged. Marking as resolved. Dhairya Parmar
02:30 PM Backport #55627 (Resolved): pacific: cephfs-shell: creates directories in local file system even ...
merged Dhairya Parmar
01:59 PM Feature #55715 (Resolved): pybind/mgr/cephadm/upgrade: allow upgrades without reducing max_mds
Adam King
11:13 AM Bug #54271: mds/OpenFileTable.cc: 777: FAILED ceph_assert(omap_num_objs == num_objs)
We will wait for this to happen in recent versions. Kotresh Hiremath Ravishankar
11:11 AM Bug #54271: mds/OpenFileTable.cc: 777: FAILED ceph_assert(omap_num_objs == num_objs)
Lowering the priority as this is seen only in nautilus and not seen in supported versions. Kotresh Hiremath Ravishankar
10:48 AM Bug #56644: qa: test_rapid_creation fails with "No space left on device"
https://pulpito.ceph.com/yuriw-2022-08-04_20:54:08-fs-wip-yuri6-testing-2022-08-04-0617-pacific-distro-default-smithi... Kotresh Hiremath Ravishankar
10:35 AM Bug #57087 (Pending Backport): qa: test_fragmented_injection (tasks.cephfs.test_data_scan.TestDat...
Seen in https://pulpito.ceph.com/yuriw-2022-08-04_20:54:08-fs-wip-yuri6-testing-2022-08-04-0617-pacific-distro-defaul... Kotresh Hiremath Ravishankar
10:01 AM Bug #51276 (Resolved): mds: avoid journaling overhead for setxattr("ceph.dir.subvolume") for no-o...
Konstantin Shalygin
10:00 AM Backport #51337 (Resolved): nautilus: mds: avoid journaling overhead for setxattr("ceph.dir.subvo...
Nautilus is EOL Konstantin Shalygin
09:49 AM Bug #51267: CommandFailedError: Command failed (workunit test fs/snaps/snaptest-multiple-capsnaps...
Seen in https://pulpito.ceph.com/yuriw-2022-08-04_20:54:08-fs-wip-yuri6-testing-2022-08-04-0617-pacific-distro-defaul... Kotresh Hiremath Ravishankar
08:18 AM Bug #57083 (Fix Under Review): ceph-fuse: monclient(hunting): handle_auth_bad_method server allow...
Xiubo Li
07:56 AM Bug #57083: ceph-fuse: monclient(hunting): handle_auth_bad_method server allowed_methods [2] but ...
The **nautilus** is using **python2**, while the **pacific** qa suite is using **python3** and the qa test suite seem... Xiubo Li
07:37 AM Bug #57083: ceph-fuse: monclient(hunting): handle_auth_bad_method server allowed_methods [2] but ...
From **remote/smithi029/log/ceph-mon.a.log.gz**: ... Xiubo Li
07:26 AM Bug #57083: ceph-fuse: monclient(hunting): handle_auth_bad_method server allowed_methods [2] but ...
The root cause is that in **nautilus** the **qa/workunits/fs/upgrade/volume_client** script is using **python2** to r... Xiubo Li
07:21 AM Bug #57083: ceph-fuse: monclient(hunting): handle_auth_bad_method server allowed_methods [2] but ...
From **remote/smithi029/log/ceph-mon.a.log.gz**: ... Xiubo Li
07:10 AM Bug #57083 (Resolved): ceph-fuse: monclient(hunting): handle_auth_bad_method server allowed_metho...
From https://pulpito.ceph.com/yuriw-2022-07-24_15:34:38-fs-wip-yuri2-testing-2022-07-15-0755-pacific-distro-default-s... Xiubo Li
07:54 AM Bug #53360 (Duplicate): pacific: client: "handle_auth_bad_method server allowed_methods [2] but i...
Missed this existing tracker. Will track this in https://tracker.ceph.com/issues/57083 tracker. Have found root cause... Xiubo Li
07:37 AM Bug #57084 (Resolved): Permissions of the .snap directory do not inherit ACLs
when using CephFS with POSIX ACLs I noticed that the .snap directory does not inherit the ACLs from its parent but on... Robert Sander
07:26 AM Backport #53861: pacific: qa: tasks.cephfs.fuse_mount:mount command failed
Created a new tracker to fix it https://tracker.ceph.com/issues/57083. Xiubo Li
06:50 AM Backport #53861: pacific: qa: tasks.cephfs.fuse_mount:mount command failed
Xiubo Li wrote:
> Kotresh Hiremath Ravishankar wrote:
> > Xiubo,
> >
> > Looks like this is seen again in this p...
Xiubo Li
07:23 AM Bug #55572: qa/cephfs: omit_sudo doesn't have effect when passed to run_shell()
I think this needs to be backported. Nikhil mentioned that the PR https://github.com/ceph/ceph/pull/47112 in pacific ... Kotresh Hiremath Ravishankar
07:08 AM Bug #57071 (Fix Under Review): mds: consider mds_cap_revoke_eviction_timeout for get_late_revokin...
Venky Shankar

08/09/2022

04:13 PM Backport #56527: pacific: mds: src/mds/MDLog.cc: 283: FAILED ceph_assert(!mds->is_ any_replay())
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/47111
merged
Yuri Weinstein
04:12 PM Backport #56152: pacific: mgr/snap_schedule: schedule updates are not persisted across mgr restart
https://github.com/ceph/ceph/pull/46797 merged Yuri Weinstein
12:55 PM Bug #56529: ceph-fs crashes on getfattr
Frank Schilder wrote:
> Thanks for the quick answer. Then, I guess, the patch to the ceph-fs clients will handle thi...
Venky Shankar
12:54 PM Bug #56529: ceph-fs crashes on getfattr
Frank Schilder wrote:
> Thanks for the quick answer. Then, I guess, the patch to the ceph-fs clients will handle thi...
Xiubo Li
12:47 PM Bug #56529: ceph-fs crashes on getfattr
Thanks for the quick answer. Then, I guess, the patch to the ceph-fs clients will handle this once it is approved? I ... Frank Schilder
12:40 PM Bug #56529: ceph-fs crashes on getfattr
Frank Schilder wrote:
> Hi all,
>
> this story continues, this time with a _valid_ vxattr name. I just observed e...
Xiubo Li
12:33 PM Bug #56529: ceph-fs crashes on getfattr
Hi all,
this story continues, this time with a _valid_ vxattr name. I just observed exactly the same problem now w...
Frank Schilder
11:40 AM Bug #57072 (Pending Backport): Quincy 17.2.3 pybind/mgr/status: assert metadata failed
`ceph fs status` return AssertionError
Error EINVAL: Traceback (most recent call last):
File "/usr/share/ceph/m...
cld slnt
10:24 AM Backport #53861: pacific: qa: tasks.cephfs.fuse_mount:mount command failed
Kotresh Hiremath Ravishankar wrote:
> Xiubo,
>
> Looks like this is seen again in this pacific run ?
>
> https...
Xiubo Li
10:13 AM Backport #53861: pacific: qa: tasks.cephfs.fuse_mount:mount command failed
Xiubo,
Looks like this is seen again in this pacific run ?
https://pulpito.ceph.com/yuriw-2022-07-24_15:34:38-f...
Kotresh Hiremath Ravishankar
10:24 AM Bug #56644: qa: test_rapid_creation fails with "No space left on device"
Seen in recent pacific run https://pulpito.ceph.com/yuriw-2022-07-24_15:34:38-fs-wip-yuri2-testing-2022-07-15-0755-pa... Kotresh Hiremath Ravishankar
09:27 AM Bug #57071 (Fix Under Review): mds: consider mds_cap_revoke_eviction_timeout for get_late_revokin...
Even though mds_cap_revoke_eviction_timeout is set to zero, ceph-mon reports some clients failing to respond to capab... Yongseok Oh
09:01 AM Bug #57048: osdc/Journaler: better handle ENOENT during replay as up:standby-replay
Patrick,
Do you mean a standby-replay MDS should tolerate missing journal objects? How can it end up in such a sit...
Venky Shankar
08:58 AM Bug #56808: crash: LogSegment* MDLog::get_current_segment(): assert(!segments.empty())
Looks similar to https://tracker.ceph.com/issues/51589 which was fixed a while ago.
Kotresh, please RCA this.
Venky Shankar
08:16 AM Backport #57058 (In Progress): pacific: mgr/volumes: Handle internal metadata directories under '...
Nikhilkumar Shelke
08:06 AM Backport #57057 (In Progress): quincy: mgr/volumes: Handle internal metadata directories under '/...
Nikhilkumar Shelke
07:07 AM Bug #54462: Command failed (workunit test fs/snaps/snaptest-git-ceph.sh) on smithi055 with status...
This maybe duplicated to https://tracker.ceph.com/issues/55332. Xiubo Li
06:55 AM Bug #54462: Command failed (workunit test fs/snaps/snaptest-git-ceph.sh) on smithi055 with status...
Seen in this run too.
https://pulpito.ceph.com/yuriw-2022-08-02_21:20:37-fs-wip-yuri7-testing-2022-07-27-0808-quin...
Kotresh Hiremath Ravishankar
06:51 AM Bug #56592: mds: crash when mounting a client during the scrub repair is going on
More info:
I was just simulating the cu case we hit by just removing one object of the directory from the metadata...
Xiubo Li
06:47 AM Bug #56592: mds: crash when mounting a client during the scrub repair is going on
Venky Shankar wrote:
> Xiubo,
>
> Were you trying to mount /mydir when it was getting repaired?
No, I was just...
Xiubo Li
06:23 AM Bug #56592: mds: crash when mounting a client during the scrub repair is going on
Xiubo,
Were you trying to mount /mydir when it was getting repaired?
Venky Shankar
06:47 AM Bug #51964: qa: test_cephfs_mirror_restart_sync_on_blocklist failure
Seen in this quincy run https://pulpito.ceph.com/yuriw-2022-08-02_21:20:37-fs-wip-yuri7-testing-2022-07-27-0808-quinc... Kotresh Hiremath Ravishankar
06:30 AM Bug #56830: crash: cephfs::mirror::PeerReplayer::pick_directory()
Dhairya,
Please take a look at this. I think there is some sort of race that is causing this crash while iterating...
Venky Shankar
06:25 AM Bug #57014: cephfs-top: add an option to dump the computed values to stdout
Jos, please take this one. Venky Shankar
05:54 AM Bug #56996 (In Progress): Transient data read corruption from other machine
Xiubo Li
04:48 AM Bug #56996: Transient data read corruption from other machine
Witold Baryluk wrote:
> Ok. I still do not understand why this can happen:
>
> writer: write("a"); write("b"); wr...
Xiubo Li
04:59 AM Bug #57065 (Closed): qa: test_query_client_ip_filter fails with latest 'perf stats' structure cha...
test_query_client_ip_filter fails with the below error in tests [1] and [2]. This happens when PR [3] is tested.
<...
Jos Collin
04:41 AM Bug #57064 (Need More Info): qa: test_add_ancestor_and_child_directory failure
Seen in recent quincy run https://pulpito.ceph.com/yuriw-2022-08-04_11:54:20-fs-wip-yuri8-testing-2022-08-03-1028-qui... Kotresh Hiremath Ravishankar
02:47 AM Bug #56067: Cephfs data loss with root_squash enabled
Patrick Donnelly wrote:
> Please open a PR for discussion.
https://github.com/ceph/ceph/pull/47506 . Please take ...
Ramana Raja
02:45 AM Bug #56067 (Fix Under Review): Cephfs data loss with root_squash enabled
Ramana Raja

08/08/2022

07:03 PM Documentation #57062 (New): Document access patterns that have good/pathological performance on C...
I have a CephFS 16.2.7 with 200 M small files (between 1 KB and 100 KB; ther are a few larger ones up to 200 MB) and ... Niklas Hambuechen
03:28 PM Bug #56048: ceph.mirror.info is not removed from target FS when mirroring is disabled
Hi Venky,
I tried it again, now with 17.2.1, and I could reproduce the issue. The mgr debug log is below.
As fa...
Andreas Teuchert
01:08 PM Bug #56048: ceph.mirror.info is not removed from target FS when mirroring is disabled
Andreas Teuchert wrote:
> When disabling mirroring on a FS with "ceph fs snapshot mirror disable <source-fs>" the "c...
Venky Shankar
02:32 PM Bug #56996: Transient data read corruption from other machine
Ok. I still do not understand why this can happen:
writer: write("a"); write("b"); write("c");
reader (other cl...
Witold Baryluk
06:33 AM Bug #56996: Transient data read corruption from other machine
Witold Baryluk wrote:
> What about when there is one writer and one reader?
This will depend on whether they are ...
Xiubo Li
01:18 PM Feature #56643: scrub: add one subcommand or option to add the missing objects back
Venky Shankar wrote:
> Xiubo Li wrote:
> > When we are scrub repairing the metadatas and some objects may get lost ...
Xiubo Li
01:02 PM Feature #56643: scrub: add one subcommand or option to add the missing objects back
Xiubo Li wrote:
> When we are scrub repairing the metadatas and some objects may get lost due to some reasons. After...
Venky Shankar
01:01 PM Bug #56249: crash: int Client::_do_remount(bool): abort
Xiubo Li wrote:
> Should be fixed by https://tracker.ceph.com/issues/54049.
Looks the same. However, I'm not sure...
Venky Shankar
09:41 AM Bug #56506: pacific: Test failure: test_rebuild_backtraceless (tasks.cephfs.test_data_scan.TestDa...
Milind Changire wrote:
> Adding any more condition to the assertion expression and passing the assertion is not goin...
Venky Shankar
08:07 AM Bug #56506: pacific: Test failure: test_rebuild_backtraceless (tasks.cephfs.test_data_scan.TestDa...
Adding any more condition to the assertion expression and passing the assertion is not going to do any good.
Since M...
Milind Changire
05:37 AM Bug #56506: pacific: Test failure: test_rebuild_backtraceless (tasks.cephfs.test_data_scan.TestDa...
Never mind - I see the err coming from JournalPointer. If the MDS is respawning/shutting down could that condition ad... Venky Shankar
05:29 AM Bug #56506: pacific: Test failure: test_rebuild_backtraceless (tasks.cephfs.test_data_scan.TestDa...
Milind Changire wrote:
> This seems to be a race between an mds respawn and the MDLog::_recovery_thread()
> In Paci...
Venky Shankar
08:55 AM Backport #57058 (Resolved): pacific: mgr/volumes: Handle internal metadata directories under '/vo...
https://github.com/ceph/ceph/pull/47512 Backport Bot
08:55 AM Backport #57057 (Resolved): quincy: mgr/volumes: Handle internal metadata directories under '/vol...
https://github.com/ceph/ceph/pull/47511 Backport Bot
08:54 AM Bug #55762 (Pending Backport): mgr/volumes: Handle internal metadata directories under '/volumes'...
Venky Shankar

08/05/2022

09:26 PM Bug #56067: Cephfs data loss with root_squash enabled
Greg Farnum wrote:
>
>
> But now I have another question -- does this mean that a kclient which only has access ...
Ramana Raja
04:21 PM Bug #56506: pacific: Test failure: test_rebuild_backtraceless (tasks.cephfs.test_data_scan.TestDa...
This seems to be a race between an mds respawn and the MDLog::_recovery_thread()
In Pacific, the MDLog::_recovery_th...
Milind Changire
01:15 PM Bug #57048 (Pending Backport): osdc/Journaler: better handle ENOENT during replay as up:standby-r...
... Patrick Donnelly
06:37 AM Backport #57042 (In Progress): quincy: pybind/mgr/volumes: interface to check the presence of sub...
Neeraj Pratap Singh
04:42 AM Bug #48673: High memory usage on standby replay MDS
We seem to be running into this pretty frequently and easily with standby-replay configuration. Venky Shankar

08/04/2022

11:43 PM Bug #57044 (Fix Under Review): mds: add some debug logs for "crash during construction of interna...
Xiubo Li
11:42 PM Bug #57044 (Resolved): mds: add some debug logs for "crash during construction of internal request"
... Xiubo Li
07:26 PM Bug #56802 (Duplicate): crash: void MDLog::_submit_entry(LogEvent*, MDSLogContextBase*): assert(!...
Patrick Donnelly
03:33 PM Bug #55897: test_nfs: update of export's access type should not trigger NFS service restart
/a/yuriw-2022-08-03_20:33:43-rados-wip-yuri8-testing-2022-08-03-1028-quincy-distro-default-smithi/6957515 Laura Flores
02:37 PM Backport #57041 (In Progress): pacific: pybind/mgr/volumes: interface to check the presence of su...
Neeraj Pratap Singh
01:15 PM Backport #57041 (Resolved): pacific: pybind/mgr/volumes: interface to check the presence of subvo...
https://github.com/ceph/ceph/pull/47460 Backport Bot
01:15 PM Backport #57042 (Resolved): quincy: pybind/mgr/volumes: interface to check the presence of subvol...
https://github.com/ceph/ceph/pull/47474 Backport Bot
01:10 PM Feature #55821 (Pending Backport): pybind/mgr/volumes: interface to check the presence of subvolu...
Venky Shankar
12:19 PM Bug #56996: Transient data read corruption from other machine
What about when there is one writer and one reader? Witold Baryluk
12:36 AM Bug #56996: Transient data read corruption from other machine
I am not very sure this is a bug.
If there are multiple clients and they are in any of:...
Xiubo Li
10:59 AM Fix #51177: pybind/mgr/volumes: investigate moving calls which may block on libcephfs into anothe...
Kotresh, please take a look at this. Venky Shankar

08/03/2022

02:46 PM Bug #56644: qa: test_rapid_creation fails with "No space left on device"
Rishabh,
Do we know why the space issue started to show up recently?
Venky Shankar
02:19 PM Bug #56517 (Resolved): fuse_ll.cc: error: expected identifier before ‘{’ token 1379 | {
Venky Shankar
10:36 AM Bug #57014 (Resolved): cephfs-top: add an option to dump the computed values to stdout
It would be nice if cephfs-top dumps it's computed values to stdout in json format. The json should contain all the f... Jos Collin
08:16 AM Backport #56462 (In Progress): pacific: mds: crash due to seemingly unrecoverable metadata error
Xiubo Li
08:15 AM Backport #56462 (Need More Info): pacific: mds: crash due to seemingly unrecoverable metadata error
Xiubo Li
08:12 AM Backport #56461 (In Progress): quincy: mds: crash due to seemingly unrecoverable metadata error
Xiubo Li
06:13 AM Bug #56506: pacific: Test failure: test_rebuild_backtraceless (tasks.cephfs.test_data_scan.TestDa...
Milind, please RCA this. Venky Shankar
12:04 AM Fix #51177: pybind/mgr/volumes: investigate moving calls which may block on libcephfs into anothe...
Downstream BZ - https://bugzilla.redhat.com/show_bug.cgi?id=2114615 Vikhyat Umrao

08/02/2022

02:09 PM Bug #56626 (Closed): "ceph fs volume create" fails with error ERANGE
Closing the bug. Changes in devstack-plugin-ceph, https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/85152... Ramana Raja
02:03 PM Bug #55858: Pacific 16.2.7 MDS constantly crashing
I've noticed a commonality when this is being triggered, Singularity is being used https://en.wikipedia.org/wiki/Sing... Mike Lowe
08:15 AM Bug #56802: crash: void MDLog::_submit_entry(LogEvent*, MDSLogContextBase*): assert(!mds->is_any_...
Maybe this is relevant information to reproduce the crash:
I have NFS Ganesha running to export CephFS and when I ...
Andreas Teuchert
06:47 AM Bug #56988: mds: memory leak suspected
Here is a graph of the memory summary without and with the automated restart. Ramin Torabi
06:34 AM Bug #56988: mds: memory leak suspected
I have automated restarting a single MDS-Server when MDS memory consumption is 80GB (roughly twice the configured mds... Ramin Torabi
06:28 AM Bug #56695 (Fix Under Review): [RHEL stock] pjd test failures(a bug that need to wait the unlink ...
Xiubo Li
05:42 AM Bug #56695: [RHEL stock] pjd test failures(a bug that need to wait the unlink to finish)
Patrick Donnelly wrote:
> [...]
>
> /ceph/teuthology-archive/pdonnell-2022-07-22_19:42:58-fs-wip-pdonnell-testing...
Xiubo Li
02:50 AM Bug #56695: [RHEL stock] pjd test failures(a bug that need to wait the unlink to finish)
Xiubo Li wrote:
> Tried **4.18.0-348.20.1.el8_5.x86_64** and couldn't reproduce it.
>
> Will try the exact same ...
Xiubo Li
02:37 AM Bug #56695: [RHEL stock] pjd test failures(a bug that need to wait the unlink to finish)
Tried **4.18.0-348.20.1.el8_5.x86_64** and couldn't reproduce it.
Will try the exact same version of **kernel-4.1...
Xiubo Li

08/01/2022

04:34 PM Bug #56996 (In Progress): Transient data read corruption from other machine
Kernel cephfs on both sides.
* ceph version 15.2.15 (2dfb18841cfecc2f7eb7eb2afd65986ca4d95985) octopus (stable)
*...
Witold Baryluk
09:47 AM Bug #56695: [RHEL stock] pjd test failures(a bug that need to wait the unlink to finish)
Test this with the latest **testing** kclient branch, I couldn't reproduce it.
Will switch to use the distro kerne...
Xiubo Li
09:46 AM Bug #56695: [RHEL stock] pjd test failures(a bug that need to wait the unlink to finish)
Xiubo Li wrote:
> Currently the kclient's **testing** branch has merged the fscryption name related patches, which w...
Xiubo Li
09:10 AM Bug #56695 (In Progress): [RHEL stock] pjd test failures(a bug that need to wait the unlink to fi...
Currently the kclient's **testing** branch has merged the fscryption name related patches, which will limit the **MAX... Xiubo Li
09:08 AM Bug #56633 (Need More Info): mds: crash during construction of internal request
Locally I couldn't reproduce it. And by reading the code I couldn't figure out in which case will the internal reques... Xiubo Li
08:59 AM Bug #53573: qa: test new clients against older Ceph clusters
Xiubo Li wrote:
> The tracker [1] has done the test for new clients with nautilus ceph simultaneously.
>
> [1] ht...
Xiubo Li
08:51 AM Bug #53573: qa: test new clients against older Ceph clusters
The tracker [1] has done the test for new clients with nautilus ceph simultaneously.
[1] https://tracker.ceph.com/...
Xiubo Li
07:01 AM Bug #56988 (Need More Info): mds: memory leak suspected
We are runnung a cephfs pacific cluster in production:
MDS version: ceph version 16.2.9 (4c3647a322c0ff5a1dd2344...
Ramin Torabi

07/30/2022

12:39 PM Backport #56978 (In Progress): pacific: mgr/volumes: Subvolume creation failed on FIPs enabled sy...
Kotresh Hiremath Ravishankar
11:45 AM Backport #56978 (Resolved): pacific: mgr/volumes: Subvolume creation failed on FIPs enabled system
https://github.com/ceph/ceph/pull/47369 Backport Bot
12:31 PM Backport #56979 (In Progress): quincy: mgr/volumes: Subvolume creation failed on FIPs enabled system
Kotresh Hiremath Ravishankar
11:45 AM Backport #56979 (Resolved): quincy: mgr/volumes: Subvolume creation failed on FIPs enabled system
https://github.com/ceph/ceph/pull/47368 Backport Bot
11:45 AM Backport #56980 (Rejected): octopus: mgr/volumes: Subvolume creation failed on FIPs enabled system
Backport Bot
11:41 AM Bug #56727 (Pending Backport): mgr/volumes: Subvolume creation failed on FIPs enabled system
Ramana Raja

07/29/2022

02:55 PM Bug #56626: "ceph fs volume create" fails with error ERANGE
Once we confirm that removing osd_pool_default_pgp_num and osd_pool_default_pg_num in devstack-plugin-ceph works, we ... Ramana Raja
07:17 AM Bug #56626: "ceph fs volume create" fails with error ERANGE
Tested the deployment with "osd_pool_default_pg_autoscale_mode = off" in bootstrap_conf and seems to fix the issue. H... Victoria Martinez de la Cruz
01:01 PM Backport #50126 (Rejected): octopus: pybind/mgr/volumes: deadlock on async job hangs finisher thread
Patrick Donnelly

07/28/2022

09:00 PM Bug #56626 (Need More Info): "ceph fs volume create" fails with error ERANGE
This does not seem like a bug in the mgr/volumes code. The mgr/volumes module creates FS pools using `osd pool create... Ramana Raja
01:29 PM Backport #53714 (Resolved): pacific: mds: fails to reintegrate strays if destdn's directory is fu...
Patrick Donnelly
01:26 PM Bug #56633: mds: crash during construction of internal request
Xiubo volunteered yesterday and said he's started work on this in standup today. Greg Farnum
07:53 AM Bug #46140 (Resolved): mds: couldn't see the logs in log file before the daemon get aborted
Checked the code, all the **assert()/abort()** have been fixed. Closing it. Xiubo Li
07:37 AM Bug #46140 (New): mds: couldn't see the logs in log file before the daemon get aborted
Xiubo Li
07:37 AM Bug #46140: mds: couldn't see the logs in log file before the daemon get aborted
I recalled it, we need to switch `assert()` to `ceph_assert()`. And the `ceph_assert()` will help dump the recent log... Xiubo Li
02:20 AM Bug #56830 (Can't reproduce): crash: cephfs::mirror::PeerReplayer::pick_directory()

http://telemetry.front.sepia.ceph.com:4000/d/jByk5HaMz/crash-spec-x-ray?orgId=1&var-sig_v2=d6f26d40363a53f0bed9a466...
Telemetry Bot
02:19 AM Bug #56808 (In Progress): crash: LogSegment* MDLog::get_current_segment(): assert(!segments.empty())

http://telemetry.front.sepia.ceph.com:4000/d/jByk5HaMz/crash-spec-x-ray?orgId=1&var-sig_v2=70db1b6eecab75317a1e77bd...
Telemetry Bot
02:19 AM Bug #56802 (Duplicate): crash: void MDLog::_submit_entry(LogEvent*, MDSLogContextBase*): assert(!...

http://telemetry.front.sepia.ceph.com:4000/d/jByk5HaMz/crash-spec-x-ray?orgId=1&var-sig_v2=902003e195a320e2927d5e39...
Telemetry Bot
02:16 AM Bug #56774 (Duplicate): crash: Client::_get_vino(Inode*)

http://telemetry.front.sepia.ceph.com:4000/d/jByk5HaMz/crash-spec-x-ray?orgId=1&var-sig_v2=90a2f49686f20a5d71a3cdc3...
Telemetry Bot
01:41 AM Bug #56067: Cephfs data loss with root_squash enabled
Ramana Raja wrote:
> I made the following change to the Locker code, and then checked how kclient and fuse client be...
Ramana Raja
01:15 AM Bug #56605 (In Progress): Snapshot and xattr scanning in cephfs-data-scan
Greg Farnum wrote:
> Xiubo Li wrote:
> > Our purpose here is to recover the snaprealms and snaptable from the data ...
Xiubo Li

07/27/2022

04:59 PM Bug #56727 (Fix Under Review): mgr/volumes: Subvolume creation failed on FIPs enabled system
Ramana Raja
11:06 AM Bug #56727 (Resolved): mgr/volumes: Subvolume creation failed on FIPs enabled system

The subvolume creation hits the following traceback on fips enabled system....
Kotresh Hiremath Ravishankar
02:02 PM Bug #56067: Cephfs data loss with root_squash enabled
Please open a PR for discussion. Patrick Donnelly
12:29 PM Bug #56067: Cephfs data loss with root_squash enabled
I made the following change to the Locker code, and then checked how kclient and fuse client behaved with root_squash... Ramana Raja
03:59 AM Bug #56067: Cephfs data loss with root_squash enabled
Patrick Donnelly wrote:
> Good work tracking that down Ramana! I don't think it's reasonable to try to require the c...
Greg Farnum
01:44 PM Bug #56605: Snapshot and xattr scanning in cephfs-data-scan
Xiubo Li wrote:
> Our purpose here is to recover the snaprealms and snaptable from the data pool. It's hard to do th...
Greg Farnum
08:17 AM Bug #56605: Snapshot and xattr scanning in cephfs-data-scan
The **listsnaps** could list the snapids of the objects:... Xiubo Li
07:32 AM Bug #56605: Snapshot and xattr scanning in cephfs-data-scan
> We should be able to see that we're missing snapshots by listing snaps on objects?
Yeah. If a file was snapshote...
Xiubo Li
07:10 AM Bug #56605: Snapshot and xattr scanning in cephfs-data-scan
Greg Farnum wrote:
> Xiubo Li wrote:
> > Here is my test case locally https://github.com/lxbsz/ceph/tree/wip-56605-...
Xiubo Li
05:42 AM Bug #56605: Snapshot and xattr scanning in cephfs-data-scan
Xiubo Li wrote:
> Here is my test case locally https://github.com/lxbsz/ceph/tree/wip-56605-draft.
>
> By using:
...
Greg Farnum
01:30 PM Feature #55121: cephfs-top: new options to limit and order-by
Jos Collin wrote:
> Greg Farnum wrote:
> > Can't fs top already change the sort order? I thought that was done in N...
Jos Collin
01:11 PM Documentation #56730: doc: update snap-schedule notes regarding 'start' time
Adding chat discussion from #cephfs IRC channel :
<gauravsitlani> Hi team i have a quick question regarding : http...
Gaurav Sitlani
01:06 PM Documentation #56730 (Resolved): doc: update snap-schedule notes regarding 'start' time
Add notes to snap-schedule mgr plugin documentation about the handling of time zone for the 'start' time.
Primary ...
Milind Changire
12:55 PM Bug #46140 (Closed): mds: couldn't see the logs in log file before the daemon get aborted
After a brief discussion with @Xiubo Li, we decided to close this tracker as this issue was encountered while debuggi... Dhairya Parmar
11:50 AM Bug #55112 (Resolved): cephfs-shell: saving files doesn't work as expected
Dhairya Parmar
11:49 AM Backport #55629 (Resolved): pacific: cephfs-shell: saving files doesn't work as expected
Dhairya Parmar
11:49 AM Bug #55242 (Resolved): cephfs-shell: put command should accept both path mandatorily and validate...
Dhairya Parmar
11:49 AM Backport #55625 (Resolved): pacific: cephfs-shell: put command should accept both path mandatoril...
Dhairya Parmar
11:36 AM Bug #40860 (Resolved): cephfs-shell: raises incorrect error when regfiles are passed to be deleted
Dhairya Parmar
11:36 AM Documentation #54551 (Resolved): docs.ceph.com/en/pacific/cephfs/add-remove-mds/#adding-an-mds ca...
Dhairya Parmar
11:35 AM Backport #55238 (Resolved): pacific: docs.ceph.com/en/pacific/cephfs/add-remove-mds/#adding-an-md...
Dhairya Parmar
10:04 AM Bug #56659: mgr: crash after upgrade pacific to main
Patrick,
Your patch to fix the libsqlite3-mod-ceph dependency and the eventual crash has worked to resolve the crash...
Milind Changire

07/26/2022

08:44 PM Bug #56659 (Duplicate): mgr: crash after upgrade pacific to main
Laura Flores
02:31 PM Backport #56712 (In Progress): pacific: mds: standby-replay daemon always removed in MDSMonitor::...
Patrick Donnelly
01:05 PM Backport #56712 (Resolved): pacific: mds: standby-replay daemon always removed in MDSMonitor::pre...
https://github.com/ceph/ceph/pull/47282 Backport Bot
02:30 PM Backport #56713 (In Progress): quincy: mds: standby-replay daemon always removed in MDSMonitor::p...
Patrick Donnelly
01:05 PM Backport #56713 (Resolved): quincy: mds: standby-replay daemon always removed in MDSMonitor::prep...
https://github.com/ceph/ceph/pull/47281 Backport Bot
01:03 PM Bug #56666 (Pending Backport): mds: standby-replay daemon always removed in MDSMonitor::prepare_b...
Patrick Donnelly
12:14 PM Bug #56605: Snapshot and xattr scanning in cephfs-data-scan
Here is my test case locally https://github.com/lxbsz/ceph/tree/wip-56605-draft.
By using:...
Xiubo Li

07/25/2022

03:20 PM Bug #56698 (Resolved): client: FAILED ceph_assert(_size == 0)
... Patrick Donnelly
03:17 PM Bug #56697 (New): qa: fs/snaps fails for fuse
... Patrick Donnelly
02:46 PM Bug #56695 (Resolved): [RHEL stock] pjd test failures(a bug that need to wait the unlink to finish)
... Patrick Donnelly
02:38 PM Bug #56694 (Fix Under Review): qa: avoid blocking forever on hung umount
Patrick Donnelly
02:34 PM Bug #56694 (Rejected): qa: avoid blocking forever on hung umount
/ceph/teuthology-archive/pdonnell-2022-07-22_19:42:58-fs-wip-pdonnell-testing-20220721.235756-distro-default-smithi/6... Patrick Donnelly
11:18 AM Bug #56626 (In Progress): "ceph fs volume create" fails with error ERANGE
Kotresh Hiremath Ravishankar
11:16 AM Bug #56626: "ceph fs volume create" fails with error ERANGE
Hi Victoria,
I am not very familiar with the osd configs but as per code if 'osd_pool_default_pg_autoscale_mode' i...
Kotresh Hiremath Ravishankar
06:39 AM Bug #55858 (Need More Info): Pacific 16.2.7 MDS constantly crashing
Kotresh Hiremath Ravishankar
04:59 AM Backport #56469 (In Progress): quincy: mgr/volumes: display in-progress clones for a snapshot
Nikhilkumar Shelke

07/24/2022

06:20 PM Bug #56067: Cephfs data loss with root_squash enabled
Patrick Donnelly wrote:
> I don't think it's reasonable to try to require the client mount to keep track of which ap...
Ramana Raja

07/23/2022

05:27 PM Bug #55759 (Resolved): mgr/volumes: subvolume ls with groupname as '_nogroup' crashes
Nikhilkumar Shelke
05:27 PM Bug #55822 (Resolved): mgr/volumes: Remove incorrect 'size' in the output of 'snapshot info' command
Nikhilkumar Shelke
05:25 PM Backport #56103 (Resolved): quincy: mgr/volumes: subvolume ls with groupname as '_nogroup' crashes
Nikhilkumar Shelke

07/22/2022

06:18 PM Feature #50470 (Resolved): cephfs-top: multiple file system support
Neeraj Pratap Singh
06:17 PM Bug #52982 (Resolved): client: Inode::hold_caps_until should be a time from a monotonic clock
Neeraj Pratap Singh
06:17 PM Backport #55937 (Resolved): pacific: client: Inode::hold_caps_until should be a time from a monot...
Neeraj Pratap Singh
05:31 PM Bug #55971 (Resolved): LibRadosMiscConnectFailure.ConnectFailure test failure
Laura Flores
05:30 PM Backport #56005 (Resolved): pacific: LibRadosMiscConnectFailure.ConnectFailure test failure
Laura Flores
05:30 PM Backport #56004 (Resolved): quincy: LibRadosMiscConnectFailure.ConnectFailure test failure
Laura Flores
04:37 PM Backport #55936 (Resolved): quincy: client: Inode::hold_caps_until should be a time from a monoto...
Patrick Donnelly
12:07 PM Backport #55936: quincy: client: Inode::hold_caps_until should be a time from a monotonic clock
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/46563
merged
Yuri Weinstein
04:37 PM Backport #56013 (Resolved): quincy: quota support for subvolumegroup
Patrick Donnelly
12:10 PM Backport #56013: quincy: quota support for subvolumegroup
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/46667
merged
Yuri Weinstein
04:37 PM Backport #56108 (Resolved): quincy: mgr/volumes: Remove incorrect 'size' in the output of 'snapsh...
Patrick Donnelly
12:12 PM Backport #56108: quincy: mgr/volumes: Remove incorrect 'size' in the output of 'snapshot info' co...
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/46804
merged
Yuri Weinstein
04:36 PM Bug #56067: Cephfs data loss with root_squash enabled
Good work tracking that down Ramana! I don't think it's reasonable to try to require the client mount to keep track o... Patrick Donnelly
12:13 PM Backport #56103: quincy: mgr/volumes: subvolume ls with groupname as '_nogroup' crashes
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/46805
merged
Yuri Weinstein
12:09 PM Backport #54578: quincy: pybind/cephfs: Add mapping for Ernno 13:Permission Denied and adding pat...
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/46647
merged
Yuri Weinstein
02:58 AM Bug #56605: Snapshot and xattr scanning in cephfs-data-scan
Greg Farnum wrote:
> Matan Breizman wrote:
> > Meaning,
> > > We can see the 1000098a1a5.00000000 object is still...
Xiubo Li
02:52 AM Bug #56605 (Need More Info): Snapshot and xattr scanning in cephfs-data-scan
Matan Breizman wrote:
> Meaning,
> > We can see the 1000098a1a5.00000000 object is still in the data pool: ...
> ...
Greg Farnum
12:33 AM Bug #56638: Restore the AT_NO_ATTR_SYNC define in libcephfs
John Mulligan wrote:
> I'm setting the backport field now for pacific & quincy. I hope I am setting it properly. Ple...
Xiubo Li
12:21 AM Bug #56666 (Fix Under Review): mds: standby-replay daemon always removed in MDSMonitor::prepare_b...
Patrick Donnelly

07/21/2022

10:25 PM Bug #56067: Cephfs data loss with root_squash enabled
Greg Farnum wrote:
> Hmm. Is the kernel client just losing track of root_squash when flushing caps? That is a differ...
Ramana Raja
12:49 PM Bug #56067: Cephfs data loss with root_squash enabled
Hmm. Is the kernel client just losing track of root_squash when flushing caps? That is a different path than the more... Greg Farnum
12:36 PM Bug #56067 (In Progress): Cephfs data loss with root_squash enabled
Ramana Raja
02:14 AM Bug #56067: Cephfs data loss with root_squash enabled
With vstart cluster (ceph main branch), I was able to reproduce the issue with a kernel client (5.17.11-200.fc35.x86_... Ramana Raja
08:19 PM Bug #56666 (Resolved): mds: standby-replay daemon always removed in MDSMonitor::prepare_beacon
If a standby-replay daemon's beacon makes it to MDSMonitor::prepare_beacon (rarely), it's automatically removed by th... Patrick Donnelly
02:54 PM Bug #56638: Restore the AT_NO_ATTR_SYNC define in libcephfs
I'm setting the backport field now for pacific & quincy. I hope I am setting it properly. Please correct it if I've f... John Mulligan
12:09 PM Bug #56605: Snapshot and xattr scanning in cephfs-data-scan
Hi Xiubo, Thank you for the detailed information!
From a RADOS standpoint everything is working as expected.
We a...
Matan Breizman
10:22 AM Bug #54283: qa/cephfs: is_mounted() depends on a mutable variable
Rishabh Dave wrote:
> The PR for this ticket needed fix for "ticket 56476":https://tracker.ceph.com/issues/56476 in ...
Rishabh Dave
08:48 AM Bug #56659 (Duplicate): mgr: crash after upgrade pacific to main
... Milind Changire

07/20/2022

05:50 PM Bug #56605 (In Progress): Snapshot and xattr scanning in cephfs-data-scan
Radoslaw Zarzynski
01:10 AM Bug #56605: Snapshot and xattr scanning in cephfs-data-scan
Let me describe how the cephfs act for this:
**1**, For the directory and it's contents, which are all metadata in...
Xiubo Li
01:25 PM Bug #55858: Pacific 16.2.7 MDS constantly crashing
Hi Mike,
We would need more information on this to proceed further.
1. Output of 'ceph fs dump' ?
2. Was multi...
Kotresh Hiremath Ravishankar
09:03 AM Bug #56063: Snapshot retention config lost after mgr restart
After updating to 17.2.1 I'm not observing the issue anymore. Now, after failing over the mgr, the retention policy i... Andreas Teuchert
08:04 AM Bug #56507 (Duplicate): pacific: Test failure: test_rapid_creation (tasks.cephfs.test_fragment.Te...
Rishabh Dave
08:04 AM Bug #56644: qa: test_rapid_creation fails with "No space left on device"
h3. From https://tracker.ceph.com/issues/56507 -
https://pulpito.ceph.com/yuriw-2022-07-06_13:57:53-fs-wip-yuri4-t...
Rishabh Dave
07:01 AM Bug #56644 (Triaged): qa: test_rapid_creation fails with "No space left on device"
http://pulpito.front.sepia.ceph.com/rishabh-2022-07-08_23:53:34-fs-wip-rishabh-testing-2022Jul08-1820-testing-default... Rishabh Dave
07:49 AM Bug #55716 (Resolved): cephfs-shell: Cmd2ArgparseError is imported without version check
The PR was merged by Venky a couple months ago - https://github.com/ceph/ceph/pull/46337#event-6657873439 Rishabh Dave
07:32 AM Bug #56416 (Resolved): qa/cephfs: delete path from cmd args after use
Rishabh Dave
06:01 AM Feature #56643 (New): scrub: add one subcommand or option to add the missing objects back
When we are scrub repairing the metadatas and some objects may get lost due to some reasons. After the repair finishe... Xiubo Li
01:45 AM Bug #56638 (Fix Under Review): Restore the AT_NO_ATTR_SYNC define in libcephfs
Xiubo Li
01:37 AM Bug #56638 (In Progress): Restore the AT_NO_ATTR_SYNC define in libcephfs
Xiubo Li

07/19/2022

11:43 PM Backport #55928 (In Progress): quincy: mds: FAILED ceph_assert(dir->get_projected_version() == di...
Hit this in downstream too. Xiubo Li
11:40 PM Backport #55929 (In Progress): pacific: mds: FAILED ceph_assert(dir->get_projected_version() == d...
Patrick Donnelly wrote:
> Xiubo, please do this backport.
Done.
Xiubo Li
04:12 PM Backport #55929 (Need More Info): pacific: mds: FAILED ceph_assert(dir->get_projected_version() =...
Xiubo, please do this backport. Patrick Donnelly
06:14 PM Bug #56632: qa: test_subvolume_snapshot_clone_quota_exceeded fails CommandFailedError
This test passed on main branch - http://pulpito.front.sepia.ceph.com/rishabh-2022-07-19_12:12:03-fs:volumes-main-dis... Rishabh Dave
04:03 PM Bug #56632 (Resolved): qa: test_subvolume_snapshot_clone_quota_exceeded fails CommandFailedError
100% reproducible so far.
http://pulpito.front.sepia.ceph.com/rishabh-2022-07-08_23:53:34-fs-wip-rishabh-testing-2...
Rishabh Dave
05:46 PM Bug #56638 (Resolved): Restore the AT_NO_ATTR_SYNC define in libcephfs
While working on an unrelated topic but building against the current 'quincy' branch - but not a released quincy - we... John Mulligan
04:34 PM Bug #56634 (New): qa: workunit snaptest-intodir.sh fails with MDS crash
http://pulpito.front.sepia.ceph.com/rishabh-2022-07-08_23:53:34-fs-wip-rishabh-testing-2022Jul08-1820-testing-default... Rishabh Dave
04:16 PM Bug #56633 (Need More Info): mds: crash during construction of internal request
... Patrick Donnelly
02:20 PM Bug #56626 (Triaged): "ceph fs volume create" fails with error ERANGE
Patrick Donnelly
02:20 PM Bug #56626: "ceph fs volume create" fails with error ERANGE
Kotresh, PTAL. Patrick Donnelly
01:43 PM Bug #56626 (Closed): "ceph fs volume create" fails with error ERANGE
Trying to create a CephFS filesystem within a cluster deployed with cephadm fails
Steps followed
1. sudo cephad...
Victoria Martinez de la Cruz
05:02 AM Bug #56605: Snapshot and xattr scanning in cephfs-data-scan
I think about this more, even the *xattrs* are not lost, we still couldn't recovery the snapshot from the data pool. ... Xiubo Li
01:41 AM Bug #56605: Snapshot and xattr scanning in cephfs-data-scan
Greg Farnum wrote:
> Do you have logs/shell output or can you reproduce this, demonstrating the presence of the xatt...
Xiubo Li
12:53 AM Bug #43216 (Resolved): MDSMonitor: removes MDS coming out of quorum election
Patrick Donnelly
12:53 AM Backport #52636 (Resolved): pacific: MDSMonitor: removes MDS coming out of quorum election
Patrick Donnelly

07/18/2022

08:38 PM Documentation #49406: Exceeding osd nearfull ratio causes write throttle.
After wondering for a long time why my clusters get slow at some point, I finally found this as well.
It would be ...
Niklas Hambuechen
03:24 PM Cleanup #4744: mds: pass around LogSegments via std::shared_ptr
For those following along, most MDS operations involve something like "mut->ls = get_current_segment()", and the poss... Greg Farnum
10:40 AM Cleanup #4744: mds: pass around LogSegments via std::shared_ptr
>
> Yeah, IMO it should be a good habit to use the shared_ptr to avoid potential use-after-free bugs as we hit in c...
Tamar Shacked
09:42 AM Cleanup #4744: mds: pass around LogSegments via std::shared_ptr
Tamar Shacked wrote:
> The issue suggests spreading LogSegment* as shared_ptr while class MDLog manages those ptrs l...
Xiubo Li
09:26 AM Cleanup #4744: mds: pass around LogSegments via std::shared_ptr
The issue suggests spreading LogSegment* as shared_ptr while class MDLog manages those ptrs lifetime (creates/stores/... Tamar Shacked
03:23 PM Bug #56605: Snapshot and xattr scanning in cephfs-data-scan
Do you have logs/shell output or can you reproduce this, demonstrating the presence of the xattr before taking the sn... Greg Farnum
02:33 PM Bug #56605 (In Progress): Snapshot and xattr scanning in cephfs-data-scan
We are doing the recovery by steps with a *alternate metadata pool*, more detail please see https://docs.ceph.com/en/... Xiubo Li
06:36 AM Bug #56592 (Triaged): mds: crash when mounting a client during the scrub repair is going on
... Xiubo Li
06:30 AM Feature #55715 (Fix Under Review): pybind/mgr/cephadm/upgrade: allow upgrades without reducing ma...
Dhairya Parmar
03:46 AM Fix #55567 (Resolved): cephfs-shell: rm returns just the error code and not proper error msg
Rishabh Dave
03:46 AM Backport #56591 (Rejected): pacific: qa: iogen workunit: "The following counters failed to be set...
Backport Bot
03:45 AM Backport #56590 (New): quincy: qa: iogen workunit: "The following counters failed to be set on md...
Backport Bot
03:45 AM Feature #48911 (Resolved): cephfs-shell needs "ln" command equivalent
Rishabh Dave
03:43 AM Bug #54108 (Pending Backport): qa: iogen workunit: "The following counters failed to be set on md...
Rishabh Dave
01:37 AM Bug #55778 (Resolved): client: choose auth MDS for getxattr with the Xs caps
Xiubo Li
01:37 AM Backport #56109 (Resolved): quincy: client: choose auth MDS for getxattr with the Xs caps
Xiubo Li
01:37 AM Bug #55824 (Resolved): ceph-fuse[88614]: ceph mount failed with (65536) Unknown error 65536
Xiubo Li
01:36 AM Backport #56106 (Resolved): quincy: ceph-fuse[88614]: ceph mount failed with (65536) Unknown erro...
Xiubo Li
01:35 AM Bug #53504 (Resolved): client: infinite loop "got ESTALE" after mds recovery
Xiubo Li
01:35 AM Backport #55934 (Resolved): quincy: client: infinite loop "got ESTALE" after mds recovery
Xiubo Li
01:35 AM Bug #55253 (Resolved): client: switch to glibc's STATX macros
Xiubo Li
01:35 AM Backport #55994 (Resolved): quincy: client: switch to glibc's STATX macros
Xiubo Li
01:34 AM Bug #53741 (Resolved): crash just after MDS become active
Xiubo Li
01:34 AM Backport #56015 (Resolved): quincy: crash just after MDS become active
Xiubo Li

07/15/2022

08:30 PM Bug #56577 (Pending Backport): mds: client request may complete without queueing next replay request
We received a report of a situation of a cluster with a single active MDS stuck in up:clientreplay. The status was:
...
Patrick Donnelly
03:29 PM Bug #52430: mds: fast async create client mount breaks racy test
Copying tracebacks for convenience (recently saw same test fail for different reason) -... Rishabh Dave
02:43 PM Backport #56106: quincy: ceph-fuse[88614]: ceph mount failed with (65536) Unknown error 65536
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/46801
merged
Yuri Weinstein
02:42 PM Backport #56109: quincy: client: choose auth MDS for getxattr with the Xs caps
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/46800
merged
Yuri Weinstein
02:41 PM Backport #56015: quincy: crash just after MDS become active
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/46681
merged
Yuri Weinstein
02:40 PM Backport #55994: quincy: client: switch to glibc's STATX macros
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/46680
merged
Yuri Weinstein
02:39 PM Backport #55926: quincy: Unexpected file access behavior using ceph-fuse
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/46595
merged
Yuri Weinstein
02:39 PM Backport #55933: quincy: crash: void Server::set_trace_dist(ceph::ref_t<MClientReply>&, CInode*, ...
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/46566
merged
Yuri Weinstein
02:38 PM Backport #55934: quincy: client: infinite loop "got ESTALE" after mds recovery
Backport Bot wrote:
> https://github.com/ceph/ceph/pull/46558
merged
Yuri Weinstein
10:05 AM Bug #56532 (Fix Under Review): client stalls during vstart_runner test
Xiubo Li
01:02 AM Bug #56532: client stalls during vstart_runner test
From Milind's reproducing logs, there has two different error code, which are *1* and *32*:... Xiubo Li
05:49 AM Backport #56468 (In Progress): pacific: mgr/volumes: display in-progress clones for a snapshot
Nikhilkumar Shelke
02:46 AM Backport #56527 (In Progress): pacific: mds: src/mds/MDLog.cc: 283: FAILED ceph_assert(!mds->is_ ...
Kotresh Hiremath Ravishankar
02:44 AM Backport #56526 (In Progress): quincy: mds: src/mds/MDLog.cc: 283: FAILED ceph_assert(!mds->is_ a...
Kotresh Hiremath Ravishankar

07/14/2022

01:00 PM Bug #56537 (Fix Under Review): cephfs-top: wrong/infinitely changing wsp values
Jos Collin
11:18 AM Bug #48773: qa: scrub does not complete
Saw this in my Quincy backport reviews as well -
https://pulpito.ceph.com/yuriw-2022-07-08_17:05:01-fs-wip-yuri2-tes...
Rishabh Dave
10:46 AM Backport #56152 (In Progress): pacific: mgr/snap_schedule: schedule updates are not persisted acr...
Venky Shankar
10:40 AM Bug #56446: Test failure: test_client_cache_size (tasks.cephfs.test_client_limits.TestClientLimits)
Rishabh, did you get to RCA this? Venky Shankar
06:09 AM Bug #56522: Do not abort MDS on unknown messages
Milind Changire wrote:
> Xiubo Li wrote:
> > Milind Changire wrote:
> > > I had started the GETVXATTR RPC implemen...
Venky Shankar
05:31 AM Bug #56522: Do not abort MDS on unknown messages
Milind Changire wrote:
> Xiubo Li wrote:
> > Milind Changire wrote:
> > > I had started the GETVXATTR RPC implemen...
Xiubo Li
05:14 AM Bug #56522: Do not abort MDS on unknown messages
Xiubo Li wrote:
> Milind Changire wrote:
> > I had started the GETVXATTR RPC implementation with the introduction o...
Milind Changire
04:20 AM Bug #56522: Do not abort MDS on unknown messages
Milind Changire wrote:
> I had started the GETVXATTR RPC implementation with the introduction of a feature bit for t...
Xiubo Li
01:29 AM Bug #56553 (Fix Under Review): client: do not uninline data for read
Xiubo Li
01:20 AM Bug #56553 (Resolved): client: do not uninline data for read
We don't even ask for and to be sure that we have been granted the Fw caps when reading, we shouldn't write contents ... Xiubo Li
 

Also available in: Atom