Activity
From 11/09/2021 to 12/08/2021
12/08/2021
- 07:12 PM Bug #52804 (Resolved): pacific: Hybrid Allocator exhibits high tail latency for writes in Octopus
- 07:12 PM Bug #53087 (Resolved): os/bluestore/AvlAllocator: introduce bluestore_avl_alloc_ff_max_* options
- 07:11 PM Backport #53101 (Resolved): pacific: os/bluestore/AvlAllocator: introduce bluestore_avl_alloc_ff_...
- 04:19 PM Backport #53101: pacific: os/bluestore/AvlAllocator: introduce bluestore_avl_alloc_ff_max_* options
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43745
merged - 07:11 PM Bug #53086 (Resolved): os/bluestore/AvlAllocator: specialize _block_picker() and cleanups
- 07:11 PM Backport #53103 (Resolved): pacific: os/bluestore/AvlAllocator: specialize _block_picker() and c...
- 04:19 PM Backport #53103: pacific: os/bluestore/AvlAllocator: specialize _block_picker() and cleanups
- Mauricio Oliveira wrote:
> https://github.com/ceph/ceph/pull/43745
merged - 07:11 PM Bug #53085 (Resolved): os/bluestore: Improve _block_picker function
- 07:10 PM Backport #53105 (Resolved): pacific: os/bluestore: Improve _block_picker function
- 04:18 PM Backport #53105: pacific: os/bluestore: Improve _block_picker function
- Mauricio Oliveira wrote:
> https://github.com/ceph/ceph/pull/43745
merged - 05:41 PM Bug #47453: checksum failures lead to assert on OSD shutdown in lab tests
- Sridhar Seshasayee wrote:
> /a/yuriw-2021-12-07_16:02:55-rados-wip-yuri11-testing-2021-12-06-1619-distro-default-smi... - 05:28 PM Bug #47453: checksum failures lead to assert on OSD shutdown in lab tests
- /a/yuriw-2021-12-07_16:02:55-rados-wip-yuri11-testing-2021-12-06-1619-distro-default-smithi/6550960...
12/07/2021
- 10:15 PM Bug #53184: failed to start new osd due to SIGSEGV in BlueStore::read()
- Thank you! I'll run the reproducer with this fix (probably v16.2.7-rc?).
- 06:08 PM Bug #53184: failed to start new osd due to SIGSEGV in BlueStore::read()
- Curious if this might be relevant: https://github.com/ceph/ceph/pull/41889 (Pacific backport - to be published in 16....
12/06/2021
- 04:32 PM Bug #53502 (Rejected): ceph_test_objectstore: ObjectStore/StoreTestOmapUpgrade.WithOmapHeader/2 f...
- Closing this as the issue was only seen during teuthology testing of yet to be merged master PRs.
- 04:25 PM Bug #53502: ceph_test_objectstore: ObjectStore/StoreTestOmapUpgrade.WithOmapHeader/2 failed with ...
- Sridhar Seshasayee wrote:
> Igor Fedotov wrote:
> > IIUC this isn't a valid bug for the current Ceph code base - it... - 04:17 PM Bug #53502: ceph_test_objectstore: ObjectStore/StoreTestOmapUpgrade.WithOmapHeader/2 failed with ...
- Igor Fedotov wrote:
> IIUC this isn't a valid bug for the current Ceph code base - it's an issue with the mentioned ... - 10:41 AM Bug #53502: ceph_test_objectstore: ObjectStore/StoreTestOmapUpgrade.WithOmapHeader/2 failed with ...
- IIUC this isn't a valid bug for the current Ceph code base - it's an issue with the mentioned not yet merged PR.
- 09:07 AM Bug #53502 (Rejected): ceph_test_objectstore: ObjectStore/StoreTestOmapUpgrade.WithOmapHeader/2 f...
- /a/yuriw-2021-12-03_15:27:18-rados-wip-yuri11-testing-2021-12-02-1451-distro-default-smithi/6542898
*Appears to be...
12/03/2021
- 06:02 PM Bug #53184: failed to start new osd due to SIGSEGV in BlueStore::read()
- I understood. I'll try to modify Ceph (and Rook?) to collect requested information.
- 12:24 PM Bug #53184: failed to start new osd due to SIGSEGV in BlueStore::read()
- Satoru Takeuchi wrote:
> > Curious if run Ceph commands following mkfs in the same container/pod? Could you please r... - 10:31 AM Bug #53483 (Resolved): Bluestore: Function not available for other platforms
- Which results in a linker error on FreeBSD and Apple
/usr/local/bin/x86_64-unknown-freebsd14.0-ld: /usr/local/bin/...
12/02/2021
- 04:39 PM Bug #52676 (Resolved): qa/standalone/osd/osd-bluefs-volume-ops.sh: TEST_bluestore2: return 1
- 03:30 PM Bug #52804 (Fix Under Review): pacific: Hybrid Allocator exhibits high tail latency for writes in...
- 03:12 PM Bug #53381 (Resolved): NCB-FSCK functionality was broken by SMR updates/cleanups #42762
- 03:11 PM Bug #52502 (Need More Info): src/os/bluestore/BlueStore.cc: FAILED ceph_assert(collection_ref)
- Dimitri, are you able to reproduce this issue? We have merged several fixes in the area recently.
- 03:10 PM Bug #53466 (Resolved): OSD is unable to allocate free space for BlueFS
- We have a ceph cluster using rook on kubernetes with some OSDs that are in CrashLoopBackOff state.
We have the sus... - 03:09 PM Bug #52939 (Rejected): lockdep cycle under BlueFS::_compact_log_async_LD_NF_D()
- The fix will be merged in https://github.com/ceph/ceph/pull/43794, not a bug at this time.
- 03:08 PM Bug #53359: bluestore: missing block.db symlinks leads to confusing crash
- Adam will take a look.
12/01/2021
- 11:00 AM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- Here is a summary for another similar case shared by the affected cluster's admin. I presume it provides good enough ...
- 07:26 AM Bug #53184: failed to start new osd due to SIGSEGV in BlueStore::read()
- > Curious if run Ceph commands following mkfs in the same container/pod? Could you please run the ceph-kvstore-tool's...
- 01:56 AM Bug #52464 (Can't reproduce): FAILED ceph_assert(current_shard->second->valid())
11/30/2021
- 03:06 PM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- I upgraded the cluster in the end of october, and by crawling a chat that i had with a coworkers the correct record g...
- 02:48 PM Bug #53426 (In Progress): rbd_directory (and some others) corrupted after update from 15.2.10 to ...
- 02:48 PM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- Florian Florensa wrote:
> Here is the output, and the keys prefix in there looks familiar with what i was seeing in ... - 01:45 PM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- Here is the output, and the keys prefix in there looks familiar with what i was seeing in the omapvals
- 12:21 PM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- Florian, this 17-byte prefix in the keys makes me think you're bitten by https://tracker.ceph.com/issues/53062
I s... - 12:07 PM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- I found another image that exhibit the same issue as the one above.
It seems like there is some kind of prefix that ... - 11:57 AM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- After inspecting the rbd_header of the broken images, all of its key exhibit the same issue, they are all prefixed by...
- 10:12 AM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- All the one that encountered this issue so far have all been created before the upgrade to ceph 16
- 09:51 AM Bug #53184: failed to start new osd due to SIGSEGV in BlueStore::read()
- Satoru Takeuchi wrote:
> > Sorry, I found that my test script, which I originally hit this problem, was wrong. I'll ... - 02:15 AM Bug #53184: failed to start new osd due to SIGSEGV in BlueStore::read()
- > Sorry, I found that my test script, which I originally hit this problem, was wrong. I'll confirm whether this probl...
11/29/2021
- 05:41 PM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- Is the problem observed for images created before upgrade or after?
- 05:34 PM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- Ah, sorry, when reading the first time I was too fast and somehow missed your description about the non-binary chars ...
- 05:22 PM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- Florian Florensa wrote:
> After checking, there are some images in the trash, but the one we have issues with in the... - 05:17 PM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- After checking, there are some images in the trash, but the one we have issues with in these cluster were not deleted...
- 05:12 PM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- Actually this behavior looks expected for a partially removed image.
Some time ago the behavior of rbd remove comm... - 02:22 PM Bug #53426: rbd_directory (and some others) corrupted after update from 15.2.10 to 16.2.6
- for the missing namespace, its key was looking like that :...
- 01:59 PM Bug #53426 (Duplicate): rbd_directory (and some others) corrupted after update from 15.2.10 to 16...
- After an upgrade from 15.2.10 to 16.2.6, i noticed that some images were doing strange things ie, an image is apperin...
- 11:16 AM Backport #53307: pacific: OMAP upgrade to PER-PG format result in skipped first key.
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43922
m... - 11:16 AM Backport #53292: pacific: OSD might wrongly attempt to use "slow" device when single device is ba...
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43970
m... - 11:13 AM Backport #52432: octopus: BlueFS might end-up with huge WAL files when upgrading OMAPs
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/42958
m...
11/26/2021
- 11:53 AM Support #53410: bluestore read rocksdb high latency
- osd scrub object will scan db, to find target key, this is a huge expense.
670 ObjectMap::ObjectMapIterator iter ... - 08:45 AM Support #53410 (Closed): bluestore read rocksdb high latency
- problem detail log: https://paste.openstack.org/show/811291/
get a k/v from rocksdb spend 48s, so I gone to check my...
11/25/2021
- 10:24 AM Bug #49170 (Resolved): BlueFS might end-up with huge WAL files when upgrading OMAPs
- 10:23 AM Backport #52432 (Resolved): octopus: BlueFS might end-up with huge WAL files when upgrading OMAPs
- 05:37 AM Bug #49256: src/os/bluestore/BlueStore.cc: FAILED ceph_assert(!c)
- We encountered this issue on 14.2.11 version. Any idea about how to quickly workaround it?
11/24/2021
- 05:20 PM Backport #53392 (Resolved): octopus: BlueFS truncate() and poweroff can create corrupted files
- https://github.com/ceph/ceph/pull/44860
- 05:20 PM Backport #53391 (Resolved): pacific: BlueFS truncate() and poweroff can create corrupted files
- 05:20 PM Backport #53390 (Rejected): nautilus: BlueFS truncate() and poweroff can create corrupted files
- 05:18 PM Bug #53129 (Pending Backport): BlueFS truncate() and poweroff can create corrupted files
- 09:37 AM Bug #53381 (Resolved): NCB-FSCK functionality was broken by SMR updates/cleanups #42762
- SMR updates/cleanups #42762 did a major query-replace on shared_alloc.a replacing a proper call to allocator with aha...
- 03:04 AM Bug #53184: failed to start new osd due to SIGSEGV in BlueStore::read()
- > > Or may be you can reproduce the issue for a different new OSD? Then it would be great to have 'mkfs' log with de...
11/22/2021
- 04:50 PM Bug #53359: bluestore: missing block.db symlinks leads to confusing crash
- fyi here is a better stack trace...
- 04:22 PM Bug #53359: bluestore: missing block.db symlinks leads to confusing crash
- >I thought we had a flag somewhere indicating whether a db and/or wal was expected so that we could provide a meaning...
- 04:09 PM Bug #53359 (New): bluestore: missing block.db symlinks leads to confusing crash
- A regression in ceph-volume (master branch) led to the block.db symlink not getting created. This leads to OSDs that ...
11/19/2021
- 11:40 PM Bug #53260 (Resolved): OMAP upgrade to PER-PG format result in skipped first key.
- 11:39 PM Backport #53307 (Resolved): pacific: OMAP upgrade to PER-PG format result in skipped first key.
- 11:36 PM Backport #53307: pacific: OMAP upgrade to PER-PG format result in skipped first key.
- Backport Bot wrote:
> https://github.com/ceph/ceph/pull/43922
merged - 08:51 PM Bug #53139 (Resolved): OSD might wrongly attempt to use "slow" device when single device is backi...
- 08:50 PM Backport #53292 (Resolved): pacific: OSD might wrongly attempt to use "slow" device when single d...
- 06:57 PM Backport #53292: pacific: OSD might wrongly attempt to use "slow" device when single device is ba...
- Igor Fedotov wrote:
> https://github.com/ceph/ceph/pull/43970
merged
11/18/2021
- 12:30 PM Bug #53261 (Duplicate): pacific: OMAP upgrade to PER-PG format result in skipped first key.
- Duplicates #53307
11/17/2021
- 10:04 PM Backport #53307 (In Progress): pacific: OMAP upgrade to PER-PG format result in skipped first key.
- 05:40 PM Backport #53307 (Resolved): pacific: OMAP upgrade to PER-PG format result in skipped first key.
- https://github.com/ceph/ceph/pull/43922
- 05:35 PM Bug #53260 (Pending Backport): OMAP upgrade to PER-PG format result in skipped first key.
11/16/2021
- 05:16 PM Backport #53292 (In Progress): pacific: OSD might wrongly attempt to use "slow" device when singl...
- https://github.com/ceph/ceph/pull/43970
- 05:10 PM Backport #53292 (Resolved): pacific: OSD might wrongly attempt to use "slow" device when single d...
- https://github.com/ceph/ceph/pull/43970
- 05:06 PM Bug #53139 (Pending Backport): OSD might wrongly attempt to use "slow" device when single device ...
- 05:03 PM Bug #53139: OSD might wrongly attempt to use "slow" device when single device is backing the store
- https://github.com/ceph/ceph/pull/43818 merged
- 07:24 AM Backport #52768: pacific: bluestore repair might cause invalid write
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43731
m... - 07:24 AM Backport #51763: pacific: Missed shared block repair doesn't fix related issues
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43731
m... - 07:23 AM Backport #52934: pacific: os/bluestore: _do_write_small fix head_pad
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43756
m... - 07:21 AM Backport #53124: pacific: OMAP upgrade to PER-PG format result in ill-formatted OMAP keys.
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43793
m... - 07:19 AM Backport #53100: octopus: os/bluestore/AvlAllocator: introduce bluestore_avl_alloc_ff_max_* options
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43747
m... - 07:19 AM Backport #53102: octopus: os/bluestore/AvlAllocator: specialize _block_picker() and cleanups
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43747
m... - 07:19 AM Backport #53104: octopus: os/bluestore: Improve _block_picker function
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43747
m... - 07:19 AM Backport #48477: octopus: osd: fix bluestore avl allocator
- This update was made using the script "backport-resolve-issue".
backport PR https://github.com/ceph/ceph/pull/43747
m...
11/15/2021
- 09:20 PM Backport #52768 (Resolved): pacific: bluestore repair might cause invalid write
- 09:05 PM Backport #52768: pacific: bluestore repair might cause invalid write
- Igor Fedotov wrote:
> https://github.com/ceph/ceph/pull/43731
merged - 09:19 PM Backport #51763 (Resolved): pacific: Missed shared block repair doesn't fix related issues
- 09:05 PM Backport #51763: pacific: Missed shared block repair doesn't fix related issues
- Igor Fedotov wrote:
> https://github.com/ceph/ceph/pull/43731
merged - 09:19 PM Bug #51762 (Pending Backport): Missed shared block repair doesn't fix related issues
- 09:19 PM Bug #51762 (Resolved): Missed shared block repair doesn't fix related issues
- 04:53 PM Bug #53260 (Fix Under Review): OMAP upgrade to PER-PG format result in skipped first key.
- 03:36 PM Backport #52934 (Resolved): pacific: os/bluestore: _do_write_small fix head_pad
- 03:35 PM Backport #52934: pacific: os/bluestore: _do_write_small fix head_pad
- Igor Fedotov wrote:
> https://github.com/ceph/ceph/pull/43756
merged - 10:12 AM Bug #53266 (Resolved): default osd_fast_shutdown=true would cause NCB to recover allocation map o...
- NCB doesn't save allocation map on regular OSD shutdown since this setting results in non-graceful one.
Hence each s...
11/13/2021
- 10:39 AM Bug #53261 (Duplicate): pacific: OMAP upgrade to PER-PG format result in skipped first key.
- This is a regression introduced by fix to omap upgrade: https://github.com/ceph/ceph/pull/43687
The problem is that ... - 10:36 AM Bug #53260 (Resolved): OMAP upgrade to PER-PG format result in skipped first key.
- This is a regression introduced by fix to omap upgrade: https://github.com/ceph/ceph/pull/43687
The problem is that ...
11/11/2021
- 11:55 AM Backport #51764 (In Progress): octopus: Missed shared block repair doesn't fix related issues
- https://github.com/ceph/ceph/pull/43887
- 10:32 AM Backport #52767 (In Progress): octopus: bluestore repair might cause invalid write
- https://github.com/ceph/ceph/pull/43885
- 09:37 AM Backport #53195 (In Progress): pacific: fsck/repair uses invalid prefix when removing undecodable...
- https://github.com/ceph/ceph/pull/43882
- 09:36 AM Backport #53196 (In Progress): octopus: fsck/repair uses invalid prefix when removing undecodable...
- https://github.com/ceph/ceph/pull/43883
- 08:45 AM Backport #53100 (Resolved): octopus: os/bluestore/AvlAllocator: introduce bluestore_avl_alloc_ff_...
- 08:44 AM Backport #53102 (Resolved): octopus: os/bluestore/AvlAllocator: specialize _block_picker() and c...
- 08:43 AM Backport #53104 (Resolved): octopus: os/bluestore: Improve _block_picker function
- 08:43 AM Fix #48272 (Resolved): osd: fix bluestore avl allocator
- 08:43 AM Backport #48477 (Resolved): octopus: osd: fix bluestore avl allocator
11/10/2021
- 11:35 PM Backport #53100: octopus: os/bluestore/AvlAllocator: introduce bluestore_avl_alloc_ff_max_* options
- Mauricio Oliveira wrote:
> https://github.com/ceph/ceph/pull/43747
merged - 11:35 PM Backport #53102: octopus: os/bluestore/AvlAllocator: specialize _block_picker() and cleanups
- Mauricio Oliveira wrote:
> https://github.com/ceph/ceph/pull/43747
merged - 11:35 PM Backport #53104: octopus: os/bluestore: Improve _block_picker function
- Mauricio Oliveira wrote:
> https://github.com/ceph/ceph/pull/43747
merged - 11:35 PM Backport #48477: octopus: osd: fix bluestore avl allocator
- Mauricio Oliveira wrote:
> https://github.com/ceph/ceph/pull/43747
merged - 11:48 AM Bug #53185 (Fix Under Review): FSCK removes allocation file when called in DEEP mode causing next...
11/09/2021
- 01:07 PM Bug #53184: failed to start new osd due to SIGSEGV in BlueStore::read()
- > To check the relevant content in DB could you please run ceph-kvstore-tool bluestore-kv <path-to-osd> list C >out a...
- 11:03 AM Bug #51034 (Closed): osd: failed to initialize OSD in Rook
- Presumably fixed by: https://github.com/ceph/ceph/pull/42424
- 02:59 AM Bug #51034: osd: failed to initialize OSD in Rook
- Please close this ticket, thank you very much for handling this problem!
- 02:58 AM Bug #51034: osd: failed to initialize OSD in Rook
- I confirmed this problem disappeared in ceph v16.2.6 and rook v1.7.5 environment. However, another problem, which was...
Also available in: Atom