Activity
From 11/23/2021 to 12/22/2021
12/22/2021
- 04:19 PM Bug #50787: rbd diff between two snapshots lists entire image content with 'whole-object' switch
- It would be really appreciated if it can be pushed to Nautilus.
- 03:20 PM Bug #48850: "FAILED ceph_assert(m_pending_ops == 0)" in TestImageReplayer/3.SnapshotUnprotect
- https://jenkins.ceph.com/job/ceph-pull-requests/87500/consoleFull#-19741017167725d52c-2930-43cb-b77a-ae0a919c2170
- 12:10 PM Bug #53613 (Rejected): [pwl] Failed to start IOs when SSD mode persistent write back cache is ena...
- This is an issue with a particular downstream build, closing.
- 11:02 AM Bug #53613: [pwl] Failed to start IOs when SSD mode persistent write back cache is enabled in cep...
- Preethi Nataraj wrote:
> After setting the rbd_PWL to 20. Captured the logs http://pastebin.test.redhat.com/1017506.... - 10:47 AM Bug #53613: [pwl] Failed to start IOs when SSD mode persistent write back cache is enabled in cep...
- After setting the rbd_PWL to 20. Captured the logs http://pastebin.test.redhat.com/1017506. Issue still seen in downs...
- 09:13 AM Bug #53613: [pwl] Failed to start IOs when SSD mode persistent write back cache is enabled in cep...
- jianpeng ma wrote:
> @Deepika Upadhyay, https://github.com/ceph/ceph/pull/44199 looks good. Does 16.2.7-3.el8cp cont... - 03:09 AM Bug #53613: [pwl] Failed to start IOs when SSD mode persistent write back cache is enabled in cep...
- Deepika Upadhyay wrote:
> @CONGMIN YIN are you seeing this issue as well? I will verify this today otherwise
No, ... - 02:01 AM Bug #53613: [pwl] Failed to start IOs when SSD mode persistent write back cache is enabled in cep...
- @Deepika Upadhyay, https://github.com/ceph/ceph/pull/44199 looks good. Does 16.2.7-3.el8cp contain this?
12/21/2021
- 03:12 PM Bug #53674 (Fix Under Review): Reading from an image being live-migrated from an encrypted export...
12/20/2021
- 02:34 PM Bug #53674 (Fix Under Review): Reading from an image being live-migrated from an encrypted export...
- Consider the following use-case, with the following sequence of user operations:
1. user creates an RBD image
2. ... - 05:41 AM Bug #53613: [pwl] Failed to start IOs when SSD mode persistent write back cache is enabled in cep...
- @CONGMIN YIN are you seeing this issue as well? I will verify this today otherwise
12/15/2021
- 02:40 PM Bug #49558 (Can't reproduce): make check: run-rbd-unit-tests-61.sh (Failed)
- This ticket was about an actual ceph_assert in the mutex code. AFAIK we haven't seen it since. The comments highlig...
- 02:30 PM Bug #49558: make check: run-rbd-unit-tests-61.sh (Failed)
- https://jenkins.ceph.com/job/ceph-pull-requests/87196/consoleFull#-1353280663f609609-74b6-4c11-b0c1-dab0624390b5
<... - 07:09 AM Bug #53613 (Rejected): [pwl] Failed to start IOs when SSD mode persistent write back cache is ena...
We upgraded the cluster to the latest and saw Io's failed to start. (Triggered Ios from RBD bench and FIO both)
[...
12/13/2021
- 08:54 AM Cleanup #53592 (New): qa/tasks/qemu - use librbd encryption directly via qemu
- qa/tasks/qemu currently sets-up an NBD device when testing with encrypted rbd images.
This is a temporary workaround... - 08:14 AM Feature #53508: [pwl] Can't find the existed pool file when the cache file is lost
- use rbd command to solve....
- 07:38 AM Bug #53587 (Fix Under Review): Failure in qemu_xfstests_luks{1,2} on centos 8.stream
12/12/2021
- 02:40 PM Bug #53587 (Resolved): Failure in qemu_xfstests_luks{1,2} on centos 8.stream
- These tests currently pass on all distros, except for centos 8.stream.
See here:
https://pulpito.ceph.com/trociny-2...
12/09/2021
- 11:59 AM Bug #52258 (Closed): [pwl] The write back time of cache is too long
- Author comments:
Tracker: https://tracker.ceph.com/issues/52258 This is not a bug, write back should talk some time... - 11:48 AM Bug #50832 (Resolved): [pwl ssd] first_free_entry corruption on media (segfault in bufferlist::sp...
- 11:44 AM Bug #49819 (Closed): [pwl ssd] assert in retire_entries() during QEMU xfstest workload
- Not observed as of now, feel free to open if otherwise
- 10:05 AM Bug #53537: [rbd-mirror] ceph_assert on rollback_snapshot info during a forcepromote while snapsh...
- Deepika Upadhyay wrote:
> more logs present at: /home/ideepika/tracker-53537 on vossi03 sepia lab
What was happen... - 08:21 AM Bug #53537: [rbd-mirror] ceph_assert on rollback_snapshot info during a forcepromote while snapsh...
- Deepika Upadhyay wrote:
> more logs present at: /home/ideepika/tracker-53537 on vossi03 sepia lab... - 09:24 AM Bug #49876 (Resolved): [luks] sporadic failure in TestLibRBD.TestEncryptionLUKS2
- see related issue: https://github.com/ceph/ceph/pull/44103
- 09:16 AM Bug #50734 (Fix Under Review): [pwl][test] make recovery.yaml actually trigger recovery
- 01:54 AM Bug #47560 (Closed): rbd-nbd: add --legacy and make --try-netlink as deprecated
- Close it for now, will reopen it when needed.
12/08/2021
- 05:18 PM Bug #53537: [rbd-mirror] ceph_assert on rollback_snapshot info during a forcepromote while snapsh...
- more logs present at: /home/ideepika/tracker-53537 on vossi03 sepia lab
- 01:01 PM Bug #53537 (In Progress): [rbd-mirror] ceph_assert on rollback_snapshot info during a forcepromot...
- ...
- 04:30 AM Bug #53517 (Duplicate): [pwl ssd] pacific: DeepCopy failure with pwl ssd mode
- ...
12/07/2021
- 08:40 AM Feature #53508: [pwl] Can't find the existed pool file when the cache file is lost
- Reproducing:
* run: rbd bench --io-pattern rand --io-type write test/image2
* kill -9: force kill rbd bench before ... - 08:03 AM Feature #53508 (Closed): [pwl] Can't find the existed pool file when the cache file is lost
- If the cache file is lost, the error is always reported and cannot be solved....
12/06/2021
- 03:59 AM Bug #50734: [pwl][test] make recovery.yaml actually trigger recovery
- I tried, will submit fix soon, just set SIGKILL is OK....
- 03:17 AM Bug #49819: [pwl ssd] assert in retire_entries() during QEMU xfstest workload
- Maybe the bug has been solved? The segfault in bufferlist::splice() maybe https://tracker.ceph.com/issues/51419, whic...
12/03/2021
- 05:20 PM Bug #48850: "FAILED ceph_assert(m_pending_ops == 0)" in TestImageReplayer/3.SnapshotUnprotect
- ...
- 10:50 AM Backport #53170 (Resolved): pacific: [pwl] deadlock on AbstractWriteLog::m_lock during shutdown
- https://github.com/ceph/ceph/pull/43772
- 10:05 AM Bug #52235 (Resolved): [pwl] deadlock on AbstractWriteLog::m_lock during shutdown
- 09:50 AM Backport #51669 (Resolved): pacific: [pwl ssd] segfault in AbstractWriteLog::get_context() during...
- https://github.com/ceph/ceph/pull/43772
- 09:50 AM Bug #50951 (Resolved): [pwl ssd] segfault in AbstractWriteLog::get_context() during fio workload
- backport https://github.com/ceph/ceph/pull/43772
- 09:20 AM Bug #53477: [rbd-mirror] mirror demon crash
- The messages reported in the description is not an issue actually. This is generated on cleanup by "status" command, ...
- 07:14 AM Bug #53477: [rbd-mirror] mirror demon crash
- https://pulpito.ceph.com/ideepika-2021-12-02_12:23:25-rbd-wip-deepika-testing-2021-12-02-1334-distro-default-smithi/6...
12/02/2021
- 09:00 PM Bug #53477 (New): [rbd-mirror] mirror demon crash
- ...
- 08:32 AM Bug #53460 (New): rbd-mirror: split-brain after failover if rbd-mirror is started only after promote
- The following scenario leads to the split-brain failure:
1) start rbd-mirror on site1 (site1 is local and site2 is...
12/01/2021
- 05:29 AM Bug #53434 (Fix Under Review): DiffIterateTest/0.DiffIterate failed w/ librbd pwl cache.
- 05:27 AM Bug #53108 (Fix Under Review): [pwl] TestMigration.Stress* failure with pwl cache
11/30/2021
- 08:18 PM Bug #53440 (New): TestLibRBD.ConcurentOperations hang
- https://jenkins.ceph.com/job/ceph-pull-requests/86489/consoleFull#20912794656733401c-e9d0-4737-9832-6594c5da0afa
<pr... - 07:30 AM Bug #53434: DiffIterateTest/0.DiffIterate failed w/ librbd pwl cache.
- pr https://github.com/ceph/ceph/pull/44144 fix this bug
- 07:18 AM Bug #53434 (Resolved): DiffIterateTest/0.DiffIterate failed w/ librbd pwl cache.
- [ RUN ] DiffIterateTest/0.DiffIterate
using new format!
wrote [4104167~28361,7507937~35127,8211521~20835,1064... - 07:27 AM Bug #51531: rbd_fsx_nbd "Size error" after resize with a timed out notification
- ...
11/29/2021
- 11:21 AM Bug #52675 (Resolved): [rbd-mirror] unbreak one-way snapshot-based mirroring
- While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are ...
- 11:16 AM Backport #52733 (Resolved): pacific: [rbd-mirror] unbreak one-way snapshot-based mirroring
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43315
m... - 11:15 AM Backport #53028 (Resolved): pacific: rbd diff between two snapshots lists entire image content wi...
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43805
m... - 07:56 AM Feature #53373: [pwl] old data may lose if host IP change after restarting
- The current design uses host IP to judge whether the existing cache file is valid, but in the distributed environment...
- 07:45 AM Backport #53421 (Resolved): pacific: librbd/crypto: fix various memory leaks
- https://github.com/ceph/ceph/pull/44998
- 07:36 AM Bug #53419 (Pending Backport): librbd/crypto: fix various memory leaks
- 07:34 AM Bug #53419 (Resolved): librbd/crypto: fix various memory leaks
- Fix all current (i.e. contained in master branch) errors that are reported by running:...
11/28/2021
- 01:38 PM Bug #53417 (Fix Under Review): librbd/crypto: Uninitialized image data may be gibberish
- 11:50 AM Bug #53417: librbd/crypto: Uninitialized image data may be gibberish
- To reproduce, you may need to read from an area where the relevant rados object exists, so first write a small amount...
- 11:21 AM Bug #53417 (Fix Under Review): librbd/crypto: Uninitialized image data may be gibberish
- By convention, librbd returns zeros when reading uninitialized image data.
When using encryption, this convention is...
11/25/2021
- 09:34 AM Bug #49876: [luks] sporadic failure in TestLibRBD.TestEncryptionLUKS2
- I reproduced this issue with debug enabled on pwl as well.
Looks like that pwl is flushing both writes concurrently,...
11/24/2021
- 04:26 PM Backport #53387 (In Progress): octopus: rbd-mirror: TestMockMirrorStatusUpdater.RemoveImmediateUp...
- https://github.com/ceph/ceph/pull/43663
- 04:00 PM Backport #53387 (Resolved): octopus: rbd-mirror: TestMockMirrorStatusUpdater.RemoveImmediateUpdat...
- https://github.com/ceph/ceph/pull/43663
- 04:15 PM Backport #53386 (In Progress): pacific: rbd-mirror: TestMockMirrorStatusUpdater.RemoveImmediateUp...
- 04:00 PM Backport #53386 (Resolved): pacific: rbd-mirror: TestMockMirrorStatusUpdater.RemoveImmediateUpdat...
- https://github.com/ceph/ceph/pull/44094
- 03:55 PM Bug #53375 (Pending Backport): rbd-mirror: TestMockMirrorStatusUpdater.RemoveImmediateUpdate may ...
- 09:52 AM Bug #43274: unittest_rbd_mirror: Exception: SegFault
- ...
11/23/2021
- 02:32 PM Bug #53375 (Fix Under Review): rbd-mirror: TestMockMirrorStatusUpdater.RemoveImmediateUpdate may ...
- 12:10 PM Bug #53375 (Resolved): rbd-mirror: TestMockMirrorStatusUpdater.RemoveImmediateUpdate may get stuck
- https://jenkins.ceph.com/job/ceph-pull-requests/86004/console...
- 07:12 AM Feature #53373: [pwl] old data may lose if host IP change after restarting
- https://github.com/ceph/ceph/pull/43839 is a feasible solution. Adding id in root structure, we'd better using string...
- 07:08 AM Feature #53373 (New): [pwl] old data may lose if host IP change after restarting
- If multiple clients successively write to the same image and go down halfway, and then the client that went down earl...
- 06:05 AM Bug #52277 (Closed): [pwl] IO hang when the single IO size * io_depth > cache size
- 06:04 AM Bug #52277: [pwl] IO hang when the single IO size * io_depth > cache size
- As https://tracker.ceph.com/issues/52599 this solution applied, current issue won't appear. Internal flush request(f...
- 02:18 AM Bug #53368 (Closed): rbd_cache configuration is meaningless
- @ yunqing wang closing, as I suppose you resolved by taking it out of QEMU environment, if not please reopen with mor...
- 01:31 AM Bug #53368: rbd_cache configuration is meaningless
- sorry I used it in qemu, which this was covered by it.
- 01:17 AM Bug #53368 (Closed): rbd_cache configuration is meaningless
- When set rbd_cache = false, the cache is still enabled.
I tested in v15.2.15, maybe alse in master.
Also available in: Atom