Activity
From 05/26/2020 to 06/24/2020
06/24/2020
- 04:38 AM Bug #45986: importing rbd diff does not apply zero sequences correctly
- you're right, I was setting this using ceph config set osd ..., but this is apparently not the right way to do it.. w...
06/23/2020
- 08:58 AM Backport #46150 (Resolved): octopus: [object-map] possible race condition when disabling object m...
- https://github.com/ceph/ceph/pull/36656
- 08:58 AM Backport #46149 (Resolved): nautilus: [object-map] possible race condition when disabling object ...
- https://github.com/ceph/ceph/pull/41787
06/22/2020
- 11:54 AM Bug #46139: profile rbd does not allow the use of RBD_INFO
- PR: https://github.com/ceph/ceph/pull/35699
- 10:52 AM Bug #46139 (Resolved): profile rbd does not allow the use of RBD_INFO
- pretty much like https://tracker.ceph.com/issues/39973 but this time a client restricted to a namespace cannot create...
- 06:06 AM Bug #45437: Segmentation fault in ImageCtx and ObjectCacher::FlusherThread of removed image to Ce...
- Jason Dillaman wrote:
> Sven Anderson wrote:
> > Of course, as I wrote in the description "the image object still c...
06/21/2020
- 03:47 PM Bug #46083 (Pending Backport): [object-map] possible race condition when disabling object map wit...
06/18/2020
- 08:06 PM Backport #46088 (Resolved): nautilus: [prometheus] auto-configure RBD metric exports for all RBD ...
- https://github.com/ceph/ceph/pull/36412
- 08:06 PM Backport #46087 (Resolved): octopus: [prometheus] auto-configure RBD metric exports for all RBD p...
- https://github.com/ceph/ceph/pull/36411
- 07:23 PM Bug #46083 (Fix Under Review): [object-map] possible race condition when disabling object map wit...
- 05:46 PM Bug #46083 (Resolved): [object-map] possible race condition when disabling object map with active IO
- If the object-map feature is disabled while there is in-flight IO, there is a possibility that the ObjectMap contains...
- 02:52 PM Bug #45986: importing rbd diff does not apply zero sequences correctly
- Nikola Ciprich wrote:
> one more update, after disabling rbd_skip_partial_discard (and restarting OSDs) I'm still ab... - 08:24 AM Bug #45986: importing rbd diff does not apply zero sequences correctly
- one more update, after disabling rbd_skip_partial_discard (and restarting OSDs) I'm still able to reproduce the probl...
- 05:53 AM Bug #45986: importing rbd diff does not apply zero sequences correctly
- sure thing, attaching the log. n.
- 01:52 PM Bug #46076 (Fix Under Review): TestMockImageReplayerSnapshotReplayer.UnlinkRemoteSnapshot race on...
- 01:46 PM Bug #46076 (Resolved): TestMockImageReplayerSnapshotReplayer.UnlinkRemoteSnapshot race on shut down
- ...
- 12:49 PM Bug #46034 (Pending Backport): [prometheus] auto-configure RBD metric exports for all RBD pools
06/17/2020
- 10:33 PM Bug #45986 (Need More Info): importing rbd diff does not apply zero sequences correctly
- If it's only affecting Nautilus and later releases, I would suspect it's related to the "rbd_skip_partial_discard" se...
- 09:25 PM Bug #45986: importing rbd diff does not apply zero sequences correctly
- Jason, can you please take a look to see if this bug is in the RBD layer.
- 12:16 PM Bug #46034: [prometheus] auto-configure RBD metric exports for all RBD pools
- Re-used the existing "rbd_stats_pools" config option. It now accepts a '*' wildcard for the pool or namespace.
06/16/2020
- 02:36 PM Bug #46034 (Fix Under Review): [prometheus] auto-configure RBD metric exports for all RBD pools
- 12:10 PM Bug #46034 (Resolved): [prometheus] auto-configure RBD metric exports for all RBD pools
- Add a new "rbd_stats_all_pools" config setting to the prometheus exporter. When enabled, all pools assigned to RBD wi...
06/15/2020
- 07:22 PM Bug #45504 (Resolved): [python] Image create(...) method defaults to "old_format = True"
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 07:22 PM Bug #45589 (Resolved): mgr/rbd_support: rename "rbd_trash_trash_purge_schedule" oid to "rbd_trash...
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 07:21 PM Bug #45714 (Resolved): [rbd-mirror] ensure ops are canceled during replayer shut down before waiting
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 07:21 PM Bug #45798 (Resolved): librbd: make rbd_read_from_replica_policy actually work
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 07:21 PM Bug #45803 (Resolved): [rbd-mirror] loss of lock might result in asserting failure when starting ...
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
06/12/2020
- 08:56 AM Bug #45987 (Fix Under Review): [iscsi-target-cli page]: add systemctl commands for enabling and s...
- 08:02 AM Bug #45987 (Resolved): [iscsi-target-cli page]: add systemctl commands for enabling and starting ...
- 20
https://docs.ceph.com/docs/master/rbd/iscsi-target-cli/
IS:
# systemctl daemon-reload
# systemctl enable r... - 07:45 AM Bug #45986 (Resolved): importing rbd diff does not apply zero sequences correctly
- while investigating problems with incorrect backup checkums, we've discovered there's probably bug in rbd import comm...
06/11/2020
- 05:33 PM Backport #45885 (Resolved): octopus: [rbd-mirror] ensure ops are canceled during replayer shut do...
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/35440
m... - 05:32 PM Backport #45801 (Resolved): octopus: Blacklist leads to potential rewatch live-lock loop
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/35439
m... - 05:32 PM Backport #45799 (Resolved): octopus: librbd: make rbd_read_from_replica_policy actually work
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/35438
m... - 05:32 PM Backport #45763 (Resolved): octopus: [rbd-mirror] image replayer stop might race with remove and ...
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/35437
m... - 05:32 PM Backport #45597 (Resolved): octopus: mgr/rbd_support: rename "rbd_trash_trash_purge_schedule" oid...
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/35436
m... - 05:32 PM Backport #45580 (Resolved): octopus: [python] Image create(...) method defaults to "old_format = ...
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/35435
m... - 05:31 PM Backport #45881 (Resolved): octopus: [rbd-mirror] loss of lock might result in asserting failure ...
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/35430
m...
06/10/2020
- 04:21 PM Backport #45885: octopus: [rbd-mirror] ensure ops are canceled during replayer shut down before w...
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/35440
merged - 04:21 PM Backport #45801: octopus: Blacklist leads to potential rewatch live-lock loop
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/35439
merged - 04:20 PM Backport #45799: octopus: librbd: make rbd_read_from_replica_policy actually work
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/35438
merged - 04:20 PM Backport #45763: octopus: [rbd-mirror] image replayer stop might race with remove and instace rep...
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/35437
merged - 04:19 PM Backport #45597: octopus: mgr/rbd_support: rename "rbd_trash_trash_purge_schedule" oid to "rbd_tr...
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/35436
merged - 04:18 PM Backport #45580: octopus: [python] Image create(...) method defaults to "old_format = True"
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/35435
merged - 04:18 PM Backport #45881: octopus: [rbd-mirror] loss of lock might result in asserting failure when starti...
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/35430
merged
06/08/2020
- 04:27 PM Bug #45437: Segmentation fault in ImageCtx and ObjectCacher::FlusherThread of removed image to Ce...
- Sven Anderson wrote:
> Of course, as I wrote in the description "the image object still contains a pointer to a Ceph... - 04:16 PM Bug #45437: Segmentation fault in ImageCtx and ObjectCacher::FlusherThread of removed image to Ce...
- Jason Dillaman wrote:
> an open RBD image was left open and then the RADOS cluster connection was closed. This leave...
06/06/2020
- 08:56 AM Bug #44567 (Resolved): [test] NBD workunit does not wait for unmap disconnect delay
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 08:32 AM Backport #44647 (Resolved): mimic: [test] NBD workunit does not wait for unmap disconnect delay
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/34737
m... - 08:26 AM Backport #45885 (In Progress): octopus: [rbd-mirror] ensure ops are canceled during replayer shut...
- 08:25 AM Backport #45801 (In Progress): octopus: Blacklist leads to potential rewatch live-lock loop
- 08:25 AM Backport #45799 (In Progress): octopus: librbd: make rbd_read_from_replica_policy actually work
- 08:24 AM Backport #45763 (In Progress): octopus: [rbd-mirror] image replayer stop might race with remove a...
- 08:23 AM Backport #45597 (In Progress): octopus: mgr/rbd_support: rename "rbd_trash_trash_purge_schedule" ...
- 08:20 AM Backport #45580 (In Progress): octopus: [python] Image create(...) method defaults to "old_format...
- 07:50 AM Backport #45881 (In Progress): octopus: [rbd-mirror] loss of lock might result in asserting failu...
06/05/2020
- 05:25 PM Bug #45437 (Won't Fix): Segmentation fault in ImageCtx and ObjectCacher::FlusherThread of removed...
- This is a bug in go-ceph [1] -- an open RBD image was left open and then the RADOS cluster connection was closed. Thi...
- 03:38 PM Backport #44647: mimic: [test] NBD workunit does not wait for unmap disconnect delay
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/34737
merged
06/04/2020
- 12:42 PM Backport #45800 (In Progress): nautilus: Blacklist leads to potential rewatch live-lock loop
- 11:56 AM Backport #45885 (Resolved): octopus: [rbd-mirror] ensure ops are canceled during replayer shut do...
- https://github.com/ceph/ceph/pull/35440
- 11:55 AM Backport #45881 (Resolved): octopus: [rbd-mirror] loss of lock might result in asserting failure ...
- https://github.com/ceph/ceph/pull/35430
- 10:57 AM Bug #45714 (Pending Backport): [rbd-mirror] ensure ops are canceled during replayer shut down bef...
- 10:56 AM Bug #45803 (Pending Backport): [rbd-mirror] loss of lock might result in asserting failure when s...
06/03/2020
- 12:18 PM Bug #45518 (Resolved): [librbd] The 'copy' method defaults to the source image format
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 11:50 AM Bug #45714: [rbd-mirror] ensure ops are canceled during replayer shut down before waiting
- Note: backport to octopus should skip the first commit (the revert) but instead cherry-pick 4a23b4be68d00ac45216f7208...
- 11:48 AM Bug #45409 (Resolved): rbd-mirror: src/librbd/journal/Replay.cc: 264: FAILED ceph_assert(!m_shut_...
- Dropping octopus backport. It will be fixed via the backport for #45714.
- 11:47 AM Backport #45581 (Rejected): octopus: rbd-mirror: src/librbd/journal/Replay.cc: 264: FAILED ceph_a...
- Closing this backport since the original fix introduced a potential hang in replay shutdown.
06/02/2020
- 04:53 PM Bug #45714 (Fix Under Review): [rbd-mirror] ensure ops are canceled during replayer shut down bef...
- 12:26 PM Bug #45798: librbd: make rbd_read_from_replica_policy actually work
- It doesn't, disregard that.
The kernel client ends up marking them because read-from-replica is implemented in lib... - 12:11 PM Bug #45798: librbd: make rbd_read_from_replica_policy actually work
- Ilya Dryomov wrote:
> Note that now that it actually works, we will hit #45795.
Is librbd marking notify ops w/ t... - 11:40 AM Bug #45798: librbd: make rbd_read_from_replica_policy actually work
- Note that now that it actually works, we will hit #45795.
06/01/2020
- 08:53 PM Bug #45714 (In Progress): [rbd-mirror] ensure ops are canceled during replayer shut down before w...
- 08:52 PM Bug #45803 (Fix Under Review): [rbd-mirror] loss of lock might result in asserting failure when s...
- 05:31 PM Bug #45803 (Resolved): [rbd-mirror] loss of lock might result in asserting failure when starting ...
- If the exclusive lock is lost (e.g. by being blacklisted), the pre-release state machine will close the journal. Howe...
- 08:19 PM Backport #45578 (Resolved): octopus: [librbd] The 'copy' method defaults to the source image format
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/35255
m... - 07:20 PM Backport #45578: octopus: [librbd] The 'copy' method defaults to the source image format
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/35255
merged - 03:44 PM Bug #45694: "TestMigration.StressLive" fails
- spotted again https://github.com/ceph/ceph/pull/35337
- 12:40 PM Backport #45801 (Resolved): octopus: Blacklist leads to potential rewatch live-lock loop
- https://github.com/ceph/ceph/pull/35439
- 12:40 PM Backport #45800 (Resolved): nautilus: Blacklist leads to potential rewatch live-lock loop
- https://github.com/ceph/ceph/pull/35385
- 12:40 PM Backport #45799 (Resolved): octopus: librbd: make rbd_read_from_replica_policy actually work
- https://github.com/ceph/ceph/pull/35438
- 12:15 PM Bug #45798 (Pending Backport): librbd: make rbd_read_from_replica_policy actually work
- 07:46 AM Bug #45798 (Fix Under Review): librbd: make rbd_read_from_replica_policy actually work
- 07:43 AM Bug #45798 (Resolved): librbd: make rbd_read_from_replica_policy actually work
- Neither rbd_read_from_replica_policy=balance nor rbd_read_from_replica_policy=localize actually work because wrong fl...
05/30/2020
05/29/2020
- 08:32 PM Feature #45313 (Resolved): rbd: add sustained burst capability with controlled burst duration to RBD
- 04:28 PM Bug #45437 (Need More Info): Segmentation fault in ImageCtx and ObjectCacher::FlusherThread of re...
- How does this crash manifest itself using the "rbd" CLI or other RBD tooling? Your problem description is pretty vagu...
- 04:00 PM Backport #45579 (Resolved): nautilus: [python] Image create(...) method defaults to "old_format =...
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/35183
m... - 04:00 PM Backport #45577 (Resolved): nautilus: [librbd] The 'copy' method defaults to the source image format
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/35182
m... - 11:31 AM Backport #45276 (Need More Info): mimic: [rbd-mirror] image replayer stop might race with remove ...
- 11:30 AM Backport #45764 (Need More Info): nautilus: [rbd-mirror] image replayer stop might race with remo...
- Note: to be done together with https://tracker.ceph.com/issues/45275
- 11:30 AM Backport #45764 (Resolved): nautilus: [rbd-mirror] image replayer stop might race with remove and...
- https://github.com/ceph/ceph/pull/41792
- 11:29 AM Backport #45763 (Resolved): octopus: [rbd-mirror] image replayer stop might race with remove and ...
- https://github.com/ceph/ceph/pull/35437
- 11:29 AM Backport #45275: nautilus: [rbd-mirror] image replayer stop might race with remove and instace re...
- Note: could (should?) be done together with https://tracker.ceph.com/issues/45716
05/28/2020
- 08:40 PM Bug #45715 (Fix Under Review): Blacklist leads to potential rewatch live-lock loop
- 06:09 PM Bug #45715 (In Progress): Blacklist leads to potential rewatch live-lock loop
- 12:30 AM Bug #45668 (Resolved): librbd: fix compile error with CephContext declaration
05/27/2020
- 02:45 PM Bug #45731 (Duplicate): TestDeepCopy.Snaps_LargerDstObjSize fails in jenkins "make check" run
- ...
05/26/2020
- 09:20 PM Backport #45578 (In Progress): octopus: [librbd] The 'copy' method defaults to the source image f...
- 06:24 PM Backport #45579: nautilus: [python] Image create(...) method defaults to "old_format = True"
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/35183
merged - 06:23 PM Backport #45577: nautilus: [librbd] The 'copy' method defaults to the source image format
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/35182
merged - 04:49 PM Bug #45072: [rbd-mirror] image replayer stop might race with remove and instace replayer shut down
- Additional fixes here: https://tracker.ceph.com/issues/45716
- 04:48 PM Bug #45716 (Pending Backport): [rbd-mirror] image replayer stop might race with remove and instac...
- This tracker ticket covers additional fixes related to #45072
- 04:48 PM Bug #45716 (Resolved): [rbd-mirror] image replayer stop might race with remove and instace replay...
- http://qa-proxy.ceph.com/teuthology/jdillaman-2020-04-09_09:42:22-rbd-wip-jd-testing-distro-basic-smithi/4938679/teut...
- 04:44 PM Bug #45715 (Resolved): Blacklist leads to potential rewatch live-lock loop
- ...
- 04:41 PM Bug #45714 (Resolved): [rbd-mirror] ensure ops are canceled during replayer shut down before waiting
- If rbd-mirror attempts to shut down its journal replayer, it might shut down after a op has started but before the Op...
Also available in: Atom