Project

General

Profile

Activity

From 04/21/2019 to 05/20/2019

05/20/2019

04:10 PM Feature #38816: Deferred writes do not work for random writes
Sage Weil wrote:
> Vitaliy Filippov wrote:
> > 1) Bluestore doesn't honor the max_deferred_txc parameter. It starts...
Vitaliy Filippov
03:20 PM Feature #38816 (In Progress): Deferred writes do not work for random writes
Nathan Cutler

05/16/2019

09:56 AM Bug #36482: High amount of Read I/O on BlueFS/DB when listing omap keys
Just to mention - find two more onodes at customer's cluster with slow omap listing - both spend all the time while s... Igor Fedotov

05/15/2019

07:03 PM Bug #38272 (In Progress): "no available blob id" assertion might occur
Igor Fedotov
04:07 PM Backport #39446 (Resolved): nautilus: OSD crashed in BitmapAllocator::init_add_free()
Igor Fedotov
04:05 PM Backport #39446: nautilus: OSD crashed in BitmapAllocator::init_add_free()
Igor Fedotov wrote:
> https://github.com/ceph/ceph/pull/27740
merged
Yuri Weinstein
09:25 AM Bug #36268: Unable to recover from ENOSPC in BlueFS
Sage Weil wrote:
> Alternative fix for mimic and luminous: https://github.com/ceph/ceph/pull/26735
hello,sage wei...
鹏 张

05/10/2019

05:01 PM Bug #39618: Runaway memory usage on Bluestore OSD
Attaching an image showing memory consumption between Nautilus and Luminous on identical hardware and indentical conf... Richard Hesse
10:55 AM Backport #39672 (Resolved): nautilus: os/bluestore: fix missing discard in BlueStore::_kv_sync_th...
https://github.com/ceph/ceph/pull/28258 Nathan Cutler

05/09/2019

05:29 PM Bug #39618: Runaway memory usage on Bluestore OSD
Igor Fedotov wrote:
> Richard, you have the following line in your config:
> osd.* dev bluestore_cache_autotune fal...
Richard Hesse
01:20 PM Bug #39618: Runaway memory usage on Bluestore OSD
Richard, you have the following line in your config:
osd.* dev bluestore_cache_autotune false
Which disables cac...
Igor Fedotov

05/08/2019

09:12 PM Backport #39638 (In Progress): luminous: fsck on mkfs breaks ObjectStore/StoreTestSpecificAUSize....
https://github.com/ceph/ceph/pull/27056 Igor Fedotov
09:10 PM Backport #39638 (Resolved): luminous: fsck on mkfs breaks ObjectStore/StoreTestSpecificAUSize.Blo...
https://github.com/ceph/ceph/pull/27056 Igor Fedotov
08:02 PM Bug #38795 (Pending Backport): fsck on mkfs breaks ObjectStore/StoreTestSpecificAUSize.BlobReuseO...
Igor Fedotov
07:20 PM Bug #38795: fsck on mkfs breaks ObjectStore/StoreTestSpecificAUSize.BlobReuseOnOverwrite
Igor Fedotov wrote:
> https://github.com/ceph/ceph/pull/27055
merged
Yuri Weinstein
05:09 PM Bug #39621 (Pending Backport): os/bluestore: fix missing discard in BlueStore::_kv_sync_thread
Sage Weil
12:57 AM Bug #39621 (Resolved): os/bluestore: fix missing discard in BlueStore::_kv_sync_thread
Discard comand should be issued before space released. tang junhui
12:19 PM Bug #39103 (Resolved): bitmap allocator osd make %use of osd over 100%
Igor Fedotov
08:51 AM Bug #39318: w_await high when rockdb compacting
i found an issue same my issue
https://tracker.ceph.com/issues/23819
hoan nv
03:29 AM Bug #39318: w_await high when rockdb compacting
use rocksdb tuning options in tuning guide : https://ceph.com/community/bluestore-default-vs-tuned-performance-compar... hoan nv

05/07/2019

05:40 PM Bug #39618: Runaway memory usage on Bluestore OSD
All of my clients are CephFS FWIW Richard Hesse
05:06 PM Bug #39618 (Can't reproduce): Runaway memory usage on Bluestore OSD
It seems that memory usage on Bluestore OSD's in Nautilus is runaway or has a leak. Auto-tuning has no effect, and ca... Richard Hesse

05/06/2019

03:49 AM Bug #39318: w_await high when rockdb compacting
@Mark
Do you need more information?
hoan nv

05/03/2019

12:49 PM Bug #36482: High amount of Read I/O on BlueFS/DB when listing omap keys
Some notes on the similar issue I'm investigating.
I've got Luminous OSD image that contains at least one object wit...
Igor Fedotov
09:32 AM Bug #38329 (Resolved): OSD crashes in get_str_map while creating with ceph-volume
Nathan Cutler
09:32 AM Backport #38587 (Resolved): mimic: OSD crashes in get_str_map while creating with ceph-volume
Nathan Cutler

05/02/2019

10:10 PM Backport #38587: mimic: OSD crashes in get_str_map while creating with ceph-volume
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/26810
merged
Yuri Weinstein
03:48 PM Bug #39569 (Duplicate): os/bluestore: fix for FreeBSD iocb structure #27458
Willem Jan Withagen

05/01/2019

10:48 PM Backport #39565: luminous: ceph-bluestore-tool: bluefs-bdev-expand silently bypasses main device ...
https://github.com/ceph/ceph/pull/27912 Igor Fedotov
10:42 PM Backport #39565 (In Progress): luminous: ceph-bluestore-tool: bluefs-bdev-expand silently bypasse...
Igor Fedotov
09:03 PM Backport #39565 (Resolved): luminous: ceph-bluestore-tool: bluefs-bdev-expand silently bypasses m...
https://github.com/ceph/ceph/pull/27912 Nathan Cutler
03:59 PM Bug #39144 (Pending Backport): ceph-bluestore-tool: bluefs-bdev-expand silently bypasses main dev...
Igor Fedotov
03:58 PM Bug #39144: ceph-bluestore-tool: bluefs-bdev-expand silently bypasses main device (slot 2)
https://github.com/ceph/ceph/pull/27447 merged Yuri Weinstein
03:57 PM Backport #39445 (Resolved): mimic: OSD crashed in BitmapAllocator::init_add_free()
Igor Fedotov
03:46 PM Backport #39445: mimic: OSD crashed in BitmapAllocator::init_add_free()
Igor Fedotov wrote:
> https://github.com/ceph/ceph/pull/27738
merged
Yuri Weinstein
03:56 PM Backport #39255 (Resolved): mimic: occaionsal ObjectStore/StoreTestSpecificAUSize.Many4KWritesTes...
Igor Fedotov
03:50 PM Backport #39255: mimic: occaionsal ObjectStore/StoreTestSpecificAUSize.Many4KWritesTest/2 failure
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/27570
merged
Yuri Weinstein
03:56 PM Backport #39246 (Resolved): mimic: os/bluestore: fix length overflow
Igor Fedotov
03:52 PM Backport #39246: mimic: os/bluestore: fix length overflow
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/27366
mergedReviewed-by: Sage Weil <sage@redhat.com>
Yuri Weinstein

04/30/2019

10:07 AM Bug #38363: Failure in assert when calling: ceph-volume lvm prepare --bluestore --data /dev/sdg
SCSI defines data protection in SBC-3, "4.22 Protection Information Model".
( http://www.t10.org/drafts.htm#SBC_Fami...
Adam Kupczyk

04/29/2019

10:32 PM Bug #39529: bluestore transaction apparently lost on osd restart
The full run logs can be found at rex004:/slow/sjust/savedlogs/sjust-2019-04-26_14:00:33-rados-wip-sjust-peering-refa... Samuel Just
10:18 PM Bug #39529: bluestore transaction apparently lost on osd restart
Sage: here's the instance I was talking about. Samuel Just
10:16 PM Bug #39529: bluestore transaction apparently lost on osd restart
The symptom was that the omap entries for the pg info after restart didn't reflect the pg log and last_update changes... Samuel Just
10:10 PM Bug #39529 (Need More Info): bluestore transaction apparently lost on osd restart
... Samuel Just
04:01 PM Bug #37282: rocksdb: submit_transaction_sync error: Corruption: block checksum mismatch code = 2
Sage Weil wrote:
> Dan van der Ster wrote:
> > > would it be possible to try that OSD one more time
> >
> > do y...
Dan van der Ster
01:23 PM Bug #37282: rocksdb: submit_transaction_sync error: Corruption: block checksum mismatch code = 2
Dan van der Ster wrote:
> > would it be possible to try that OSD one more time
>
> do you mean to zap/recreate it...
Sage Weil
01:52 PM Documentation #39522 (Resolved): fix and improve doc regarding manual bluestore cache settings.
Seems like bluestore_cache_kv_max no longer exists and defaults have changed. Jan Fajerski

04/26/2019

02:35 AM Bug #39318: w_await high when rockdb compacting
My config rocksdb.... hoan nv
01:14 AM Feature #38816: Deferred writes do not work for random writes
master: https://github.com/ceph/ceph/pull/27789
nautilus: https://github.com/ceph/ceph/pull/27819 merged
Konstantin Shalygin

04/25/2019

06:40 PM Bug #39318: w_await high when rockdb compacting
w_await is literally how long it's taking for the device to service requests. If the latency is increasing that mean... Mark Nelson
02:22 PM Bug #39318: w_await high when rockdb compacting
This sounds like the SSD is just busy from compaction and/or client work. The w_await latency does sound a bit high ... Sage Weil
05:29 PM Bug #37282: rocksdb: submit_transaction_sync error: Corruption: block checksum mismatch code = 2
> would it be possible to try that OSD one more time
do you mean to zap/recreate it and try again or just start it...
Dan van der Ster
05:26 PM Bug #37282: rocksdb: submit_transaction_sync error: Corruption: block checksum mismatch code = 2
Dan, that dump appears to have multiple errors:... Sage Weil
02:23 PM Bug #37282: rocksdb: submit_transaction_sync error: Corruption: block checksum mismatch code = 2
Same devices (HDD for data and ssd partition for block.db) for both failures.
We have left the osd down since the ...
Dan van der Ster
02:18 PM Bug #37282: rocksdb: submit_transaction_sync error: Corruption: block checksum mismatch code = 2
Taking a look at this. It's interesting that it happened twice on the same device(s)... did it occur again after tha... Sage Weil
03:52 PM Feature #38816: Deferred writes do not work for random writes
Vitaliy Filippov wrote:
> 1) Bluestore doesn't honor the max_deferred_txc parameter. It starts to flush operations a...
Sage Weil
03:49 PM Feature #38816: Deferred writes do not work for random writes
Igor Fedotov wrote:
> @Mark, thanks for the update.
> From you perf counter dumps (after?.json) one can see the fol...
Sage Weil
02:44 PM Bug #36482 (In Progress): High amount of Read I/O on BlueFS/DB when listing omap keys
Igor Fedotov
01:30 PM Bug #20236: bluestore: ObjectStore/StoreTestSpecificAUSize.Many4KWritesNoCSumTest/2 failure
/a/sage-2019-04-24_22:14:24-rados-wip-sage2-testing-2019-04-24-0829-distro-basic-smithi/3890129... Sage Weil

04/24/2019

11:57 AM Bug #39144: ceph-bluestore-tool: bluefs-bdev-expand silently bypasses main device (slot 2)
Adding "[needs luminous backport]" prefix to subject line so that, just in case the bug gets accidentally marked "Res... Nathan Cutler
11:21 AM Backport #39444: luminous: OSD crashed in BitmapAllocator::init_add_free()
https://github.com/ceph/ceph/pull/27739 Igor Fedotov
11:06 AM Backport #39444 (In Progress): luminous: OSD crashed in BitmapAllocator::init_add_free()
Igor Fedotov
10:41 AM Backport #39444 (Resolved): luminous: OSD crashed in BitmapAllocator::init_add_free()
https://github.com/ceph/ceph/pull/27739 Igor Fedotov
11:21 AM Backport #39446: nautilus: OSD crashed in BitmapAllocator::init_add_free()
https://github.com/ceph/ceph/pull/27740 Igor Fedotov
11:05 AM Backport #39446 (In Progress): nautilus: OSD crashed in BitmapAllocator::init_add_free()
Igor Fedotov
10:41 AM Backport #39446 (Resolved): nautilus: OSD crashed in BitmapAllocator::init_add_free()
https://github.com/ceph/ceph/pull/27740 Igor Fedotov
11:07 AM Backport #39445: mimic: OSD crashed in BitmapAllocator::init_add_free()
https://github.com/ceph/ceph/pull/27738 Igor Fedotov
11:05 AM Backport #39445 (In Progress): mimic: OSD crashed in BitmapAllocator::init_add_free()
Igor Fedotov
10:41 AM Backport #39445 (Resolved): mimic: OSD crashed in BitmapAllocator::init_add_free()
https://github.com/ceph/ceph/pull/27738 Igor Fedotov

04/23/2019

06:06 PM Bug #39334 (Pending Backport): OSD crashed in BitmapAllocator::init_add_free()
Sage Weil

04/22/2019

02:43 AM Bug #39318: w_await high when rockdb compacting
update.
my log in attach file.
rocksdb of osd compact after 30 minutes. i thinks it too fast.
when osd has t...
hoan nv
 

Also available in: Atom