Activity
From 02/12/2019 to 03/13/2019
03/13/2019
- 10:49 PM Bug #38599 (Duplicate): Ceph Mons crash on Glance image delete
- This looks like a dup of #23915 which was fixed in 12.2.6 (you indicated you were running 12.2.5).
- 09:14 PM Bug #38599: Ceph Mons crash on Glance image delete
- It looks like it might be the same bug, unfortunately. :(
If the rbd team thinks this is not, can you guys transla... - 11:54 AM Documentation #18197: Remove any pre-OpenStack Liberty references from the RBD documentation
- https://releases.openstack.org/
- 11:54 AM Documentation #18197: Remove any pre-OpenStack Liberty references from the RBD documentation
- +1. The documentation talks about versions that are outdated since 2014
- 07:58 AM Bug #38706 (Resolved): flush skips requests in QOS throttler queue
- The flush request must wait for all in-flight operations to complete. But if QOS is enabled currently it does not wai...
03/12/2019
- 11:56 AM Backport #38685 (Resolved): mimic: librbd: avoid aggregate-initializing any static_visitor
- https://github.com/ceph/ceph/pull/27041
- 11:56 AM Backport #38684 (Resolved): mimic: librbd: avoid aggregate-initializing IsWriteOpVisitor
- https://github.com/ceph/ceph/pull/27039
03/11/2019
- 09:08 PM Bug #38660 (Pending Backport): librbd: avoid aggregate-initializing IsWriteOpVisitor
- 12:14 PM Bug #38660 (Resolved): librbd: avoid aggregate-initializing IsWriteOpVisitor
- Just use the regular class ctor with ()'s instead of {}'s
For error info:
https://github.com/ceph/ceph/pull/26838 - 09:08 PM Bug #38659 (Pending Backport): librbd: avoid aggregate-initializing any static_visitor
- 09:08 PM Bug #38659 (Resolved): librbd: avoid aggregate-initializing any static_visitor
- 12:12 PM Bug #38659 (Resolved): librbd: avoid aggregate-initializing any static_visitor
- Clang does not like Boost stat_visitor initialisers with {}.
just using a Ctor works without errors.
See the explan... - 03:49 PM Bug #37877 (Resolved): [libvirt] volume info command might incorrectly attempt to use fast-diff
- A patch has been submitted upstream for this issue [1]
[1] https://www.redhat.com/archives/libvir-list/2019-March/... - 02:54 PM Backport #38674 (Resolved): luminous: Performance improvements for object-map
- https://github.com/ceph/ceph/pull/28137
- 02:54 PM Backport #38673 (Resolved): mimic: Performance improvements for object-map
- https://github.com/ceph/ceph/pull/28136
- 02:45 PM Bug #38538 (Pending Backport): Performance improvements for object-map
- 12:37 PM Bug #38661 (Resolved): deep cp a migration prepared image will results assert
- ...
03/10/2019
03/06/2019
- 03:29 AM Bug #38599 (Duplicate): Ceph Mons crash on Glance image delete
- We had an issue with our cluster where 4/7 mons crashed and failed to come up for longer than 1-2 minutes to establis...
03/05/2019
- 03:20 PM Bug #38585 (Fix Under Review): rbd pool init change breaks standalone tests
- https://github.com/ceph/ceph/pull/26770
- 03:01 PM Bug #38585 (Resolved): rbd pool init change breaks standalone tests
- as of 819290e5cfefd9ffe2720529bae2924c4b91c07d (i think?) rbd pool init foo now needs rados to be up. a bunch of sta...
03/04/2019
- 12:13 PM Backport #38564 (Resolved): luminous: [librbd] race condition possible when validating RBD pool
- https://github.com/ceph/ceph/pull/28140
- 12:13 PM Backport #38563 (Resolved): mimic: [librbd] race condition possible when validating RBD pool
- https://github.com/ceph/ceph/pull/28139
- 08:44 AM Bug #38500 (Pending Backport): [librbd] race condition possible when validating RBD pool
03/02/2019
- 06:38 AM Bug #38553 (Resolved): rbd: race condition in rbd removing
- When we are going to remove a rbd in different process currently, there is a possibility to get a segmentfault. A sim...
03/01/2019
- 04:23 PM Bug #38538 (Fix Under Review): Performance improvements for object-map
- 03:27 PM Bug #38538 (Resolved): Performance improvements for object-map
- The "object_map_update" OSD class method has a some low-hanging improvements that can be made to increase the perform...
- 03:43 PM Bug #38504 (Resolved): librbd: Qos fix bug in were_all_throttled()
- 02:32 AM Backport #38505 (In Progress): mimic: API list_images() Segmentation fault
- https://github.com/ceph/ceph/pull/26707
02/28/2019
- 10:48 PM Bug #38504 (Fix Under Review): librbd: Qos fix bug in were_all_throttled()
- 09:04 AM Bug #38504: librbd: Qos fix bug in were_all_throttled()
- https://github.com/ceph/ceph/pull/26688
- 08:13 AM Bug #38504 (Resolved): librbd: Qos fix bug in were_all_throttled()
- were_all_throttled() function is not correct:
bool were_all_throttled() {
return m_throttled_flag & RB... - 12:36 PM Backport #38509 (Resolved): mimic: [rbd-mirror] LeaderWatcher stuck in loop if pool deleted
- https://github.com/ceph/ceph/pull/28151
- 12:36 PM Backport #38508 (Resolved): luminous: [rbd-mirror] LeaderWatcher stuck in loop if pool deleted
- https://github.com/ceph/ceph/pull/28158
- 12:36 PM Backport #38505 (Resolved): mimic: API list_images() Segmentation fault
- https://github.com/ceph/ceph/pull/26707
02/27/2019
- 07:19 PM Bug #38387 (Pending Backport): [rbd-mirror] LeaderWatcher stuck in loop if pool deleted
- 07:19 PM Bug #38500 (Fix Under Review): [librbd] race condition possible when validating RBD pool
- 02:53 PM Bug #38500 (Resolved): [librbd] race condition possible when validating RBD pool
- The create image state machine attempts to validate the pool for snapshot and overwrite support. However, if multiple...
- 03:31 PM Bug #38468 (Pending Backport): API list_images() Segmentation fault
- 03:15 PM Bug #38453 (Resolved): [rbd-mirror] switch to active/active policy by default
- 03:13 PM Feature #38146 (Resolved): Implement kernel discard min allocation size logic
- 03:13 PM Feature #38146 (Pending Backport): Implement kernel discard min allocation size logic
- 03:01 PM Bug #38463: [rbd-mirror] thrasher can lead to "File exists" error when creating image
- Logs indicate that the leader heartbeat was delayed. It should have timed out in 5 seconds but the notifier didn't re...
02/26/2019
02/25/2019
- 07:14 PM Bug #38453 (Pending Backport): [rbd-mirror] switch to active/active policy by default
- 01:20 PM Bug #38468 (Fix Under Review): API list_images() Segmentation fault
- 06:55 AM Bug #38468: API list_images() Segmentation fault
- *PR:* https://github.com/ceph/ceph/pull/26622
- 06:47 AM Bug #38468 (Resolved): API list_images() Segmentation fault
- bug1:...
02/24/2019
- 05:44 PM Bug #38463 (In Progress): [rbd-mirror] thrasher can lead to "File exists" error when creating image
- ...
- 05:38 PM Bug #38385 (Resolved): [test] rbd-mirror peer monitor address doesn't handle msgr v2
02/22/2019
- 09:13 PM Bug #38453 (Fix Under Review): [rbd-mirror] switch to active/active policy by default
- 07:10 PM Bug #38453 (Resolved): [rbd-mirror] switch to active/active policy by default
- Mimic introduced the active/active policy but it was disabled by default. It has been fairly stable in teuthology tes...
- 05:06 PM Bug #38404 (Resolved): [api] seperate remove verification steps and re-use prior to moving image ...
- 05:06 PM Feature #24226 (Resolved): Image remove state machine should move image to trash as first step
- 03:14 PM Bug #38170 (Resolved): krbd discard no longer guarantees zeroing
- 03:14 PM Backport #38185 (Resolved): mimic: krbd discard no longer guarantees zeroing
- 02:33 PM Backport #38441 (Resolved): mimic: compare-and-write skips compare after copyup without object map
- https://github.com/ceph/ceph/pull/28133
- 02:33 PM Backport #38440 (Resolved): luminous: compare-and-write skips compare after copyup without object...
- https://github.com/ceph/ceph/pull/28134
02/20/2019
- 11:24 PM Bug #38383: compare-and-write skips compare after copyup without object map
- https://github.com/ceph/ceph/pull/26519
- 07:10 PM Bug #38383 (Pending Backport): compare-and-write skips compare after copyup without object map
- 10:14 PM Feature #38146 (Fix Under Review): Implement kernel discard min allocation size logic
- 10:13 PM Bug #38308 (Need More Info): segfault when deleting rbd in python bindings
- @Patrick: how is this reproduced? Frame 0 is showing that it appears that librados isn't even initialized yet since t...
- 10:09 PM Bug #38404 (In Progress): [api] seperate remove verification steps and re-use prior to moving ima...
- 01:22 PM Bug #38404 (Resolved): [api] seperate remove verification steps and re-use prior to moving image ...
- Now that images are moved to the trash as a first step, if they have snapshots (or other issues) it might fail. Re-us...
- 09:30 PM Backport #38185: mimic: krbd discard no longer guarantees zeroing
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/26285
merged
02/19/2019
- 09:11 PM Bug #38387 (Fix Under Review): [rbd-mirror] LeaderWatcher stuck in loop if pool deleted
- 07:47 PM Bug #38387 (Resolved): [rbd-mirror] LeaderWatcher stuck in loop if pool deleted
- ...
- 07:23 PM Bug #38385 (Fix Under Review): [test] rbd-mirror peer monitor address doesn't handle msgr v2
- 07:22 PM Bug #38385 (Resolved): [test] rbd-mirror peer monitor address doesn't handle msgr v2
- An invalid monitor address is stored in the config-key entry for rbd-mirror:
i.e. "[v2:192.168.1.3:6840" should be... - 04:29 PM Bug #38383 (Resolved): compare-and-write skips compare after copyup without object map
- With object map disabled, a copyup turns compare-and-write into an ordinary write. This is because ObjectOperation::...
- 02:30 PM Bug #38346 (Resolved): luminous->nautilus: test_rbd.TestClone.test_unprotect_with_children: Asser...
02/18/2019
- 05:17 PM Bug #36659 (Resolved): [rbd-mirror] forced promotion after killing remote cluster results in stuc...
- 05:16 PM Backport #36692 (Resolved): luminous: [rbd-mirror] forced promotion after killing remote cluster ...
- 05:06 PM Backport #36692: luminous: [rbd-mirror] forced promotion after killing remote cluster results in ...
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/25720
merged - 05:15 PM Bug #37541 (Resolved): [journal] max journal order is incorrectly set at 64
- 05:15 PM Backport #37901 (Resolved): luminous: [journal] max journal order is incorrectly set at 64
- 05:04 PM Backport #37901: luminous: [journal] max journal order is incorrectly set at 64
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/25955
merged - 05:14 PM Bug #36327 (Resolved): Broken parameter parsing in /etc/ceph/rbdmap
- 05:14 PM Backport #37926 (Resolved): luminous: Broken parameter parsing in /etc/ceph/rbdmap
- 05:04 PM Backport #37926: luminous: Broken parameter parsing in /etc/ceph/rbdmap
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/26000
merged - 05:13 PM Backport #38186 (Resolved): luminous: krbd discard no longer guarantees zeroing
- 05:03 PM Backport #38186: luminous: krbd discard no longer guarantees zeroing
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/26286
merged - 04:26 PM Bug #38346 (Fix Under Review): luminous->nautilus: test_rbd.TestClone.test_unprotect_with_childre...
- 02:27 PM Bug #38364: [sparsify] verify that image isn't using an EC data pool
- ... alternatively, when the "sparsify" cls method is unavailable, librbd could fallback to using "CMPEXT" + "REMOVE"/...
- 02:19 PM Bug #38364 (Resolved): [sparsify] verify that image isn't using an EC data pool
- The cls operation needed for sparsify isn't supported on EC pools, so provide a nice failure message when attempting ...
- 02:26 PM Feature #13706 (Resolved): Online sparsify RBD images from the CLI
- 09:23 AM Bug #38315 (Resolved): journal: stalled io due to race when closing and advancing object set
02/16/2019
- 09:06 AM Bug #38346 (In Progress): luminous->nautilus: test_rbd.TestClone.test_unprotect_with_children: As...
- 09:06 AM Bug #38346: luminous->nautilus: test_rbd.TestClone.test_unprotect_with_children: AssertionError: ...
- The problem is that after upgrade to the master the librbd client starts to support clone v2 format, and with the def...
- 07:15 AM Bug #38346: luminous->nautilus: test_rbd.TestClone.test_unprotect_with_children: AssertionError: ...
- Or actually the errors above were also after real failure in the tests tear down, and the failure is:...
- 06:54 AM Bug #38346: luminous->nautilus: test_rbd.TestClone.test_unprotect_with_children: AssertionError: ...
- test_rbd.test_rename error and the segfault are just consequences of an earlier test failure, due to imperfect test c...
02/15/2019
- 06:45 PM Bug #38346: luminous->nautilus: test_rbd.TestClone.test_unprotect_with_children: AssertionError: ...
- This is on role client.0, post-upgrade to master. The workunit (i.e., the bash) is luminous, fwiw...
- 06:42 PM Bug #38346 (Resolved): luminous->nautilus: test_rbd.TestClone.test_unprotect_with_children: Asser...
- ...
02/14/2019
- 11:01 AM Bug #38315 (Fix Under Review): journal: stalled io due to race when closing and advancing object set
- 10:45 AM Bug #38315 (Resolved): journal: stalled io due to race when closing and advancing object set
- I am observing this running DiffIterateTest/1.DiffIterateStress unittest in loop with journaling enabled:...
- 12:38 AM Bug #38308 (Can't reproduce): segfault when deleting rbd in python bindings
- This happens with 12.2.10 and 12.2.11, this backtrace is from 12.2.10. The cluster is running 12.2.10....
02/12/2019
Also available in: Atom