Project

General

Profile

Activity

From 06/19/2016 to 07/18/2016

07/18/2016

06:07 PM Bug #16689 (Fix Under Review): mkfs.xfs slow performance with discards and object map
*PR*: https://github.com/ceph/ceph/pull/10332 Jason Dillaman
05:37 PM Bug #16689 (In Progress): mkfs.xfs slow performance with discards and object map
Jason Dillaman
02:38 PM Bug #16689: mkfs.xfs slow performance with discards and object map
Attaching the raw lttng files for librbd runs. The new image is 10G instead of the 100G in the previous case, but all... Vaibhav Bhembre
04:51 PM Bug #16717 (Resolved): "[ FAILED ] TestLibRBD.TestCreateLsDeletePP" in upgrade:client-upgrade-j...
Run: http://pulpito.ceph.com/teuthology-2016-07-13_02:45:02-upgrade:client-upgrade-jewel-distro-basic-smithi/
Job: 3...
Yuri Weinstein
12:26 PM Bug #16708 (Resolved): Sporadic failure in TestImageReplayer.StartReplayAndWrite
From Mykola:... Jason Dillaman
12:22 PM Bug #16223 (In Progress): journal should prefetch small chunks of the object during replay
Jason Dillaman
11:53 AM Bug #16707 (Resolved): rbd-replay-prep doesn't record discard IO events
Jason Dillaman
07:07 AM Bug #16529: "[ FAILED ] TestClsRbd.mirror_image" in upgrade:jewel-x-master-distro-basic-vps
http://pulpito.ceph.com/teuthology-2016-07-17_04:20:03-upgrade:jewel-x-master-distro-basic-vps/319486/
http://pulpit...
Loïc Dachary

07/17/2016

12:29 AM Bug #16689: mkfs.xfs slow performance with discards and object map
I am attaching 2 files for lttng trace run captured while running: _mkfs.xfs -s size=4096 -f /dev/sda_.
* replay.1...
Vaibhav Bhembre

07/16/2016

02:28 PM Feature #15468: Feature : cephx user management for RBD images
Few more things: I really wants to hide from such user as much information about cluster, as possible. Right now anyo... George Shuklin

07/15/2016

06:10 PM Documentation #16704 (Duplicate): RBD replay needs to document new configuration option
The "rbd_tracing" configuration option needs to be set to true to enable tracing. Jason Dillaman
09:36 AM Backport #16701 (Resolved): jewel: rbd-mirror: image sync throttle needs to use pool id + image i...
https://github.com/ceph/ceph/pull/10678 Nathan Cutler

07/14/2016

09:38 PM Bug #16689 (Resolved): mkfs.xfs slow performance with discards and object map
Examples:
object-map enabled:
time mkfs.xfs -s size=4096 -f /dev/sda
real 9m10.882s
user 0m0.000s
sys 0m0.012s...
Jason Dillaman
02:41 AM Bug #16623: segfault in unittest_rbd_mirror
Always some problems when on the bleeding edge. Jason Dillaman
01:42 AM Bug #16623 (Resolved): segfault in unittest_rbd_mirror
Damn, I thought about a problem in the gmock/gtest code but dismissed it as "unlikely"
Sure enough, it works fine ...
Brad Hubbard

07/13/2016

04:43 PM Bug #16536 (Pending Backport): rbd-mirror: image sync throttle needs to use pool id + image id to...
Mykola Golub
03:16 PM Bug #16623: segfault in unittest_rbd_mirror
@Brad: Josh just merged the upgraded googletest/googlemock changes. It works for me now, but can you pull the latest ... Jason Dillaman
03:05 PM Bug #16623: segfault in unittest_rbd_mirror
Work for switching to the newer googletest framework was already in-progress:
*PR*: https://github.com/ceph/ceph/p...
Jason Dillaman
02:58 PM Bug #16623: segfault in unittest_rbd_mirror
Using the latest gmock/gtest environment appears to fix the issue. Jason Dillaman
02:45 PM Bug #16623: segfault in unittest_rbd_mirror
Yup -- able to instantly reproduce under F24. Jason Dillaman
01:02 PM Bug #16623: segfault in unittest_rbd_mirror
Turns out that I am on F23 -- I thought I had already upgraded. I found a similar bug report where GCC 6 + F24 resul... Jason Dillaman
10:07 AM Bug #16623: segfault in unittest_rbd_mirror
I spent a lot of time looking at this today but I couldn't pin it down but I do have some findings.
The following ...
Brad Hubbard
02:45 AM Bug #16623: segfault in unittest_rbd_mirror
Hi @Jason,
Following our discussion on IRC I did the following.
# git clone --recursive https://github.com/ceph...
Brad Hubbard

07/12/2016

11:01 PM Bug #16623: segfault in unittest_rbd_mirror
@jason yes, I can Brad Hubbard
01:13 PM Bug #16623 (Need More Info): segfault in unittest_rbd_mirror
@Brad: are you able to repeat this issue? Jason Dillaman
03:01 PM Bug #16538 (In Progress): rbd-mirror: image deleter should use pool id + global image uuid for key
Jason Dillaman
02:33 PM Backport #16658 (Resolved): jewel: rbd-mirror: gracefully handle being blacklisted
https://github.com/ceph/ceph/pull/10684 Loïc Dachary
04:12 AM Bug #16536 (Fix Under Review): rbd-mirror: image sync throttle needs to use pool id + image id to...
*PR*: https://github.com/ceph/ceph/pull/10254 Jason Dillaman
01:16 AM Bug #16536 (In Progress): rbd-mirror: image sync throttle needs to use pool id + image id to form...
Jason Dillaman
03:10 AM Bug #16654: the option 'rbd_cache_writethrough_until_flush=true' dosn't work
Sorry, the correct test data is :
1. Test in the VM with 'rbd_cache_writethrough_until_flush=false' , the randwrit...
Qi Yu
03:04 AM Bug #16654 (Resolved): the option 'rbd_cache_writethrough_until_flush=true' dosn't work
Env: my ceph cluster with 252 SATA osds. the test VM kernerl version is ' 3.13.0-86-generic'
test cmd: sudo fio -...
Qi Yu

07/11/2016

08:48 PM Bug #16622 (Fix Under Review): rbd-mirror: snap rename does not correctly replicate
*PR*: https://github.com/ceph/ceph/pull/10249 Jason Dillaman

07/10/2016

05:36 AM Bug #16349: rbd-mirror: gracefully handle being blacklisted
-Note: same master PR as #16622- nevermind Nathan Cutler

07/09/2016

11:43 PM Bug #15034: Possible inconsistent IO when using configured for multiple op threads
*PR*: https://github.com/ceph/ceph/pull/9023 -- still requires re-enabling the librbd option. Jason Dillaman
11:40 PM Bug #16349 (Pending Backport): rbd-mirror: gracefully handle being blacklisted
*PR*: PR: https://github.com/ceph/ceph/pull/9970 Jason Dillaman
11:40 PM Bug #16622 (In Progress): rbd-mirror: snap rename does not correctly replicate
Jason Dillaman
11:39 PM Bug #16622 (Pending Backport): rbd-mirror: snap rename does not correctly replicate
Jason Dillaman
11:37 PM Bug #4446 (Closed): librbd: crash from opensolaris vm
2 years since last update -- closing. Jason Dillaman

07/08/2016

04:19 AM Bug #16623: segfault in unittest_rbd_mirror
jason, mind taking a look? Kefu Chai
04:12 AM Bug #16623 (Resolved): segfault in unittest_rbd_mirror
@Core was generated by `/home/brad/working/src/ceph/build/bin/unittest_rbd_mirror'.
Program terminated with signal S...
Brad Hubbard

07/07/2016

10:25 PM Bug #16622 (Resolved): rbd-mirror: snap rename does not correctly replicate
Jason Dillaman
03:16 AM Bug #16608 (Resolved): bashism in src/rbdmap
the shebang line reads @#!/bin/sh@@
while we are using array in this script, ...
Kefu Chai

07/06/2016

03:15 PM Bug #15259 (In Progress): performance: journaling results in 4X slowdown when writes are not bloc...
Ricardo Dias
09:43 AM Backport #16601 (In Progress): jewel: librbd: fix missing return statement if failed to get mirro...
Loïc Dachary
09:39 AM Backport #16601 (Resolved): jewel: librbd: fix missing return statement if failed to get mirror i...
https://github.com/ceph/ceph/pull/10144 Loïc Dachary
09:39 AM Bug #16600 (Resolved): librbd: fix missing return statement if failed to get mirror image state
https://github.com/ceph/ceph/pull/10136 Loïc Dachary

07/05/2016

06:20 PM Backport #16593 (Resolved): jewel: FAILED assert(object_no < m_object_map.size())
https://github.com/ceph/ceph/pull/10647 Nathan Cutler
08:45 AM Bug #16561 (Pending Backport): FAILED assert(object_no < m_object_map.size())
Mykola Golub

07/04/2016

07:16 PM Bug #16449 (Fix Under Review): Prevent the creation of a clone from a non-primary mirrored image
PR: https://github.com/ceph/ceph/pull/10123 Mykola Golub
10:28 AM Bug #16449 (In Progress): Prevent the creation of a clone from a non-primary mirrored image
Mykola Golub
05:20 PM Bug #16386 (Fix Under Review): Discard hangs when 'rbd_skip_partial_discard' is enabled
PR: https://github.com/ceph/ceph/pull/10060 Mykola Golub

07/03/2016

11:34 AM Backport #16576 (Resolved): jewel: rbd-mirror: FAILED assert(m_local_image_ctx->object_map != nul...
https://github.com/ceph/ceph/pull/10646 Loïc Dachary

07/01/2016

12:29 PM Cleanup #16539 (Resolved): rbd-mirror: remove ceph_test_rbd_mirror_image_replay test case
Jason Dillaman
06:10 AM Cleanup #16539 (Fix Under Review): rbd-mirror: remove ceph_test_rbd_mirror_image_replay test case
PR: https://github.com/ceph/ceph/pull/10083
QA PR: https://github.com/ceph/ceph-qa-suite/pull/1072
Mykola Golub
06:08 AM Cleanup #16539: rbd-mirror: remove ceph_test_rbd_mirror_image_replay test case
Nathan, it is not related to #16529 Mykola Golub
12:12 PM Feature #16571 (Resolved): Discard journal event should record state of rbd_skip_partial_discard ...
When RBD mirroring verification is added, if two clusters have different settings for this option, the images will di... Jason Dillaman
11:19 AM Bug #16558 (Pending Backport): rbd-mirror: FAILED assert(m_local_image_ctx->object_map != nullptr)"
Mykola Golub

06/30/2016

10:17 PM Bug #16563 (Duplicate): rbd-mirror: missing op finish event should be treated as an error
Issue fixed under #16362 Jason Dillaman
07:42 PM Bug #16563 (New): rbd-mirror: missing op finish event should be treated as an error
Jason Dillaman
07:29 PM Bug #16563 (Duplicate): rbd-mirror: missing op finish event should be treated as an error
If an op doesn't successfully complete and record its success in the journal via an OpFinishEvent, treat the non-prim... Jason Dillaman
07:26 PM Bug #16561 (Fix Under Review): FAILED assert(object_no < m_object_map.size())
*PR*: https://github.com/ceph/ceph/pull/10072 Jason Dillaman
05:52 PM Bug #16561 (Resolved): FAILED assert(object_no < m_object_map.size())
Steps to reproduce:
1. clone a large image
2. create a snapshot on the clone
3. shrink the clone twice
The semi...
Jason Dillaman
02:45 PM Bug #16558 (Fix Under Review): rbd-mirror: FAILED assert(m_local_image_ctx->object_map != nullptr)"
*PR*: https://github.com/ceph/ceph/pull/10065 Jason Dillaman
02:20 PM Bug #16558 (Resolved): rbd-mirror: FAILED assert(m_local_image_ctx->object_map != nullptr)"
The exclusive lock for the local image was lost due to cluster starvation during image sync. The daemon should gracef... Jason Dillaman
02:19 PM Feature #16557 (Resolved): Update on-disk exclusive lock tag when image watcher is lost
Under heavy-cluster loads, it's quite possible that the watcher on the RBD image header will time-out. If the exclusi... Jason Dillaman
12:52 PM Bug #16555 (Resolved): librbd should permit removal of image being bootstrapped by rbd-mirror
If an image is being bootstrapped, it implies that the rbd-mirror daemon currently has the image open. The removal AP... Jason Dillaman
08:56 AM Bug #16404 (Resolved): Sporadic test failures when image update notification missed
Loïc Dachary
08:56 AM Feature #16197 (Resolved): Improve rbd-mirror test case coverage
Loïc Dachary
08:55 AM Bug #16289 (Resolved): "rbd: create error: (38) Function not implemented" in upgrade:client
Loïc Dachary
08:00 AM Backport #16512: jewel: rbd-mirror: potential race condition when restarting image replayer
merge conflict at c97f7245a015dbdda25584134840d023fd65cdd1 from https://github.com/ceph/ceph/pull/9958/commits Loïc Dachary
07:58 AM Backport #16511: jewel: rbd-mirror: 'wait_for_scheduled_deletion' callback might deadlock
merge conflict at c136f443366fe804057a9c1f8c01c3f0a33a4d63 from https://github.com/ceph/ceph/pull/9964/commits Loïc Dachary
07:56 AM Backport #16485 (In Progress): jewel: Whitelist EBUSY error from "snap unprotect" for journal replay
Loïc Dachary
07:55 AM Backport #16484 (In Progress): jewel: ExclusiveLock object leaked when switching to snapshot
Loïc Dachary
07:54 AM Backport #16483 (In Progress): jewel: Close journal and object map before flagging exclusive lock...
Loïc Dachary
07:53 AM Backport #16482 (In Progress): jewel: Timeout sending mirroring notification shouldn't result in ...
Loïc Dachary
07:52 AM Backport #16460 (In Progress): jewel: Crash when utilizing advisory locking API functions
Loïc Dachary
07:50 AM Backport #16459 (In Progress): jewel: rbd-mirror should disable proxied maintenance ops for non-p...
Loïc Dachary
07:49 AM Backport #16232 (Resolved): jewel: Improve rbd-mirror test case coverage
Loïc Dachary
07:49 AM Backport #16458: jewel: Potential crash during journal::Replay shut down
merge conflict at 3112a93b49c24f9ae101a7f252c8e708e0d3e260 Loïc Dachary
07:32 AM Backport #16438 (In Progress): jewel: Delete image when a resync is requested
Loïc Dachary
07:25 AM Backport #16426 (In Progress): jewel: Possible race condition during journal transition from repl...
Loïc Dachary
07:23 AM Backport #16425 (In Progress): jewel: rbd-mirror: potential race condition accessing local image ...
Loïc Dachary
07:22 AM Backport #16424 (In Progress): jewel: Journal needs to handle duplicate maintenance op tids
Loïc Dachary
07:20 AM Backport #16423 (In Progress): jewel: Journal duplicate op detection can cause lockdep error
Loïc Dachary
07:19 AM Backport #16373: jewel: rbd-mirror: gracefully handle missing sync point snapshots
https://github.com/ceph/ceph/pull/9730/commits/d16698f96c65b9627cc1f64ef80b7f1e39d69d45 conflicts Loïc Dachary
07:16 AM Backport #16371 (In Progress): jewel: rbd-mirror: ensure replay status formatter has completed be...
Loïc Dachary
07:16 AM Cleanup #16539: rbd-mirror: remove ceph_test_rbd_mirror_image_replay test case
Is this related to #16529 ? Nathan Cutler
07:15 AM Backport #16372 (In Progress): jewel: Unable to disable journaling feature if in unexpected mirro...
Loïc Dachary
07:13 AM Backport #16315 (In Progress): jewel: When journaling is enabled, a flush request shouldn't flush...
Loïc Dachary
06:50 AM Bug #16386 (In Progress): Discard hangs when 'rbd_skip_partial_discard' is enabled
Mykola Golub
12:53 AM Subtask #16545 (New): rbd-mirror multi-peer: register local image as client of remote peer image ...
If a new peer image is discovered after the ImageReplayer state machine as started the bootstrap stage, the local ima... Jason Dillaman

06/29/2016

11:44 PM Subtask #16544 (New): rbd-mirror multi-peer: image deleter should only schedule deletion if last ...
The image deleter currently takes a shortcut since it only has to handle a single peer. With multiple peers, the imag... Jason Dillaman
11:42 PM Subtask #16543 (New): rbd-mirror multi-peer: ImageReplayer should restart when a new mirror peer ...
This condition would indicate a failover to a new peer. By restarting the state machine, it would re-invoke the boots... Jason Dillaman
11:39 PM Subtask #16542 (New): rbd-mirror multi-peer: bootstrap state machine should support multiple remo...
In the case where the local image doesn't exist or is not primary and its last journal tag isn't linked to a specific... Jason Dillaman
11:37 PM Subtask #16541 (New): rbd-mirror multi-peer: ImageReplayer should support multiple registered rem...
No longer pass the remote librados connection, remote mirror uuid, remote pool id, and remote image id via the constr... Jason Dillaman
11:32 PM Feature #16540 (New): rbd-mirror: support multiple peers per pool in mesh topology
Update rbd-mirror to support transitioning the primary status between N peer clusters. Jason Dillaman
10:03 PM Cleanup #16539 (Resolved): rbd-mirror: remove ceph_test_rbd_mirror_image_replay test case
It is essentially duplicated by ceph_test_rbd_mirror + the rbd-mirror workunits. Its special hooks into ImageReplayer... Jason Dillaman
09:57 PM Bug #16538 (Resolved): rbd-mirror: image deleter should use pool id + global image uuid for key
Waiting for a deletion or canceling pending waiters should use the pool id + global image uuid as the key for referen... Jason Dillaman
09:50 PM Bug #16536 (Resolved): rbd-mirror: image sync throttle needs to use pool id + image id to form un...
There is a possibility that the image id will collide between pools, so it should also use the local pool id to ensur... Jason Dillaman
04:52 PM Bug #16529 (Resolved): "[ FAILED ] TestClsRbd.mirror_image" in upgrade:jewel-x-master-distro-ba...
Run: http://pulpito.ceph.com/teuthology-2016-06-29_04:20:02-upgrade:jewel-x-master-distro-basic-vps/
Jobs: 283220, 2...
Yuri Weinstein
09:05 AM Backport #16486 (In Progress): jewel: Object map/fast-diff invalidated if journal replays the sam...
Loïc Dachary
08:59 AM Backport #16513: jewel: rbd-mirror: image-replayer: segfault when removing resync listener
It does not apply cleanly, probably needs to be clustered with other replay related backports. Loïc Dachary
06:37 AM Backport #16513 (Resolved): jewel: rbd-mirror: image-replayer: segfault when removing resync list...
https://github.com/ceph/ceph/pull/10678 Loïc Dachary
08:55 AM Backport #16514 (In Progress): jewel: Image removal doesn't necessarily clean up all rbd_mirrorin...
Loïc Dachary
06:37 AM Backport #16514 (Resolved): jewel: Image removal doesn't necessarily clean up all rbd_mirroring e...
https://github.com/ceph/ceph/pull/10009 Loïc Dachary
08:44 AM Backport #16324 (Resolved): jewel: rbd: create error: (38) Function not implemented in upgrade:cl...
Loïc Dachary
07:38 AM Backport #16518 (In Progress): jewel: TaskFinisher: cancel all tasks wait until finisher done
Loïc Dachary
07:21 AM Backport #16518 (Resolved): jewel: TaskFinisher: cancel all tasks wait until finisher done
https://github.com/ceph/ceph/pull/9752 Loïc Dachary
07:37 AM Backport #16520 (In Progress): jewel: librbd: potential use after free on refresh error
Loïc Dachary
07:35 AM Backport #16520 (Resolved): jewel: librbd: potential use after free on refresh error
https://github.com/ceph/ceph/pull/9952 Loïc Dachary
07:34 AM Bug #16519 (Resolved): librbd: potential use after free on refresh error
https://github.com/ceph/ceph/pull/9949 Loïc Dachary
07:20 AM Bug #16517 (Resolved): TaskFinisher: cancel all tasks wait until finisher done
https://github.com/ceph/ceph/pull/9682 Loïc Dachary
06:37 AM Backport #16512 (Resolved): jewel: rbd-mirror: potential race condition when restarting image rep...
https://github.com/ceph/ceph/pull/10684 Loïc Dachary
06:37 AM Backport #16511 (Resolved): jewel: rbd-mirror: 'wait_for_scheduled_deletion' callback might deadlock
https://github.com/ceph/ceph/pull/10678 Loïc Dachary

06/28/2016

07:54 PM Bug #16269 (Resolved): "AssertionError" in upgrade:client-upgrade-jewel-distro-basic-smithi
Fix will be included in v10.2.3 (or anything against the current jewel branch HEAD) Jason Dillaman
07:46 PM Bug #16269: "AssertionError" in upgrade:client-upgrade-jewel-distro-basic-smithi
Run: http://pulpito.ceph.com/teuthology-2016-06-26_02:45:02-upgrade:client-upgrade-jewel-distro-basic-smithi/
Jobs: ...
Yuri Weinstein
07:53 PM Backport #16422 (Resolved): jewel: Sporadic test failures when image update notification missed
Jason Dillaman
06:14 PM Bug #16489 (Pending Backport): rbd-mirror: potential race condition when restarting image replayer
Mykola Golub
06:12 PM Bug #16471 (Pending Backport): Image removal doesn't necessarily clean up all rbd_mirroring entries
Mykola Golub
06:11 PM Bug #16491 (Pending Backport): rbd-mirror: 'wait_for_scheduled_deletion' callback might deadlock
Mykola Golub
06:57 AM Bug #16502 (Closed): fio ended up with soft lockup - CPU#1 stuck for 22s!
*Ceph Client*
_//DISTRO_
Ubuntu 14.04.4 LTS - trusty
_//KERNEL_
4.2.0-36-generic #42~14.04.1-Ubuntu
_//PACKA...
Shinobu Kinjo

06/27/2016

04:24 PM Bug #16491 (Fix Under Review): rbd-mirror: 'wait_for_scheduled_deletion' callback might deadlock
*PR*: https://github.com/ceph/ceph/pull/9964 Jason Dillaman
01:12 PM Bug #16491 (Resolved): rbd-mirror: 'wait_for_scheduled_deletion' callback might deadlock
If there is a race and there is no scheduled deletion, it's possible for the callback to complete within the caller's... Jason Dillaman
04:14 PM Bug #16488 (Pending Backport): rbd-mirror: image-replayer: segfault when removing resync listener
Mykola Golub
10:22 AM Bug #16488 (Fix Under Review): rbd-mirror: image-replayer: segfault when removing resync listener
PR: https://github.com/ceph/ceph/pull/9945 Mykola Golub
09:28 AM Bug #16488 (Resolved): rbd-mirror: image-replayer: segfault when removing resync listener
Transcript from Mykola Golub:... Ricardo Dias
03:51 PM Backport #15360 (Resolved): hammer: Linking to -lrbd causes process startup times to balloon
Nathan Cutler
01:09 PM Bug #16489 (Fix Under Review): rbd-mirror: potential race condition when restarting image replayer
*PR*: https://github.com/ceph/ceph/pull/9958 Jason Dillaman
12:31 PM Bug #16489 (Resolved): rbd-mirror: potential race condition when restarting image replayer
"ImageReplayer::handle_shut_down" advances the state machine before closing the local and remote IoCtxs. If "Replayer... Jason Dillaman
07:56 AM Backport #16486 (Resolved): jewel: Object map/fast-diff invalidated if journal replays the same s...
https://github.com/ceph/ceph/pull/10010 Loïc Dachary
07:55 AM Backport #16485 (Resolved): jewel: Whitelist EBUSY error from "snap unprotect" for journal replay
https://github.com/ceph/ceph/pull/10055 Loïc Dachary
07:55 AM Backport #16484 (Resolved): jewel: ExclusiveLock object leaked when switching to snapshot
https://github.com/ceph/ceph/pull/10054 Loïc Dachary
07:55 AM Backport #16483 (Resolved): jewel: Close journal and object map before flagging exclusive lock as...
https://github.com/ceph/ceph/pull/10053 Loïc Dachary
07:55 AM Backport #16482 (Resolved): jewel: Timeout sending mirroring notification shouldn't result in fai...
https://github.com/ceph/ceph/pull/10052 Loïc Dachary

06/25/2016

05:01 PM Bug #16470 (Pending Backport): Timeout sending mirroring notification shouldn't result in failure
Mykola Golub
12:31 PM Bug #16478 (Resolved): Non-primary image is recording journal events during image sync
When the snapshot state is being synchronized from the primary to the non-primary, these events are being incorrectly... Jason Dillaman
12:26 PM Bug #16473 (In Progress): Demotion/promotion results in the image syncing again from the beginning
Jason Dillaman

06/24/2016

07:25 PM Bug #16450 (Pending Backport): Close journal and object map before flagging exclusive lock as rel...
Mykola Golub
07:24 PM Bug #16350 (Pending Backport): Object map/fast-diff invalidated if journal replays the same snap ...
Mykola Golub
07:23 PM Bug #16446 (Pending Backport): ExclusiveLock object leaked when switching to snapshot
Mykola Golub
07:23 PM Bug #16445 (Pending Backport): Whitelist EBUSY error from "snap unprotect" for journal replay
Mykola Golub
06:13 PM Bug #16473 (Resolved): Demotion/promotion results in the image syncing again from the beginning
Demote the primary image and promote the original non-primary image. The rbd-mirror for the newly demoted non-primar... Jason Dillaman
02:55 PM Bug #16471 (Fix Under Review): Image removal doesn't necessarily clean up all rbd_mirroring entries
*PR*: https://github.com/ceph/ceph/pull/9923 Jason Dillaman
02:09 PM Bug #16471 (Resolved): Image removal doesn't necessarily clean up all rbd_mirroring entries
If an image was partially created (due to a failure), attempting to remove the image will not clean up its rbd_mirror... Jason Dillaman
02:12 PM Bug #16470 (Fix Under Review): Timeout sending mirroring notification shouldn't result in failure
*PR*: https://github.com/ceph/ceph/pull/9922 Jason Dillaman
02:00 PM Bug #16470 (Resolved): Timeout sending mirroring notification shouldn't result in failure
Log and ignore any notification failure when sending mirroring status updates:... Jason Dillaman
03:14 AM Cleanup #16465 (Resolved): rbd discard ret value truncated
rbd discard use 'int' as return type, but 'len' user passed is 'unit64',
in some case, the ret value will be truncat...
Huan Zhang Zhang

06/23/2016

07:09 PM Backport #16460 (Resolved): jewel: Crash when utilizing advisory locking API functions
https://github.com/ceph/ceph/pull/10051 Nathan Cutler
07:09 PM Backport #16459 (Resolved): jewel: rbd-mirror should disable proxied maintenance ops for non-prim...
https://github.com/ceph/ceph/pull/10050 Nathan Cutler
07:09 PM Backport #16458 (Resolved): jewel: Potential crash during journal::Replay shut down
https://github.com/ceph/ceph/pull/10645 Nathan Cutler
05:40 PM Bug #16450 (Fix Under Review): Close journal and object map before flagging exclusive lock as rel...
*PR*: https://github.com/ceph/ceph/pull/9905 Jason Dillaman
02:16 PM Bug #16450 (In Progress): Close journal and object map before flagging exclusive lock as released
Jason Dillaman
02:16 PM Bug #16450 (Resolved): Close journal and object map before flagging exclusive lock as released
The journal might have in-flight IO which will check if the image is the exclusive lock owner. Therefore, ensure that... Jason Dillaman
05:28 PM Bug #16364 (Pending Backport): Crash when utilizing advisory locking API functions
Mykola Golub
05:27 PM Bug #16433 (Pending Backport): Potential crash during journal::Replay shut down
Mykola Golub
05:26 PM Bug #16411 (Pending Backport): rbd-mirror should disable proxied maintenance ops for non-primary ...
Mykola Golub
02:28 PM Bug #16350 (Fix Under Review): Object map/fast-diff invalidated if journal replays the same snap ...
*PR*: https://github.com/ceph/ceph/pull/9899 Jason Dillaman
02:25 PM Bug #16350 (In Progress): Object map/fast-diff invalidated if journal replays the same snap remov...
Jason Dillaman
02:14 PM Bug #14059 (Resolved): Can't build Ceph with --with-rbd (Cython)
Jewel doesn't officially support stock CentOS 6 and no known issues with CentOS 7. Jason Dillaman
01:57 PM Bug #16449 (Resolved): Prevent the creation of a clone from a non-primary mirrored image
Jason Dillaman
01:30 PM Bug #16446 (Fix Under Review): ExclusiveLock object leaked when switching to snapshot
*PR*: https://github.com/ceph/ceph/pull/9896 Jason Dillaman
01:21 PM Bug #16446 (Resolved): ExclusiveLock object leaked when switching to snapshot
The exclusive lock is properly shut down but its object is not deleted when switching from image HEAD to a snapshot. Jason Dillaman
12:49 PM Bug #16445 (Fix Under Review): Whitelist EBUSY error from "snap unprotect" for journal replay
*PR*: https://github.com/ceph/ceph/pull/9895 Jason Dillaman
12:26 PM Bug #16445 (Resolved): Whitelist EBUSY error from "snap unprotect" for journal replay
If the OpFinishEvent's return code is "-EBUSY" for a snap unprotect event, do not attempt to replay the snap unprotec... Jason Dillaman
07:43 AM Backport #16438 (Resolved): jewel: Delete image when a resync is requested
https://github.com/ceph/ceph/pull/10678 Loïc Dachary
12:42 AM Bug #16269 (Fix Under Review): "AssertionError" in upgrade:client-upgrade-jewel-distro-basic-smithi
Jason Dillaman
12:40 AM Bug #16269 (In Progress): "AssertionError" in upgrade:client-upgrade-jewel-distro-basic-smithi
Inadvertently fixed by tracker ticket #16422 Jason Dillaman
12:39 AM Backport #16422 (In Progress): jewel: Sporadic test failures when image update notification missed
Jason Dillaman

06/22/2016

11:02 PM Bug #15561: RBD image can be listed but not opened
@Ke Ke: any update or can this be closed? Jason Dillaman
10:31 PM Bug #15220 (Can't reproduce): Deadlock in Objecter::_throttle_op
Haven't seen this reproduce in several months -- closing for now. Jason Dillaman
10:30 PM Bug #14059 (Need More Info): Can't build Ceph with --with-rbd (Cython)
@Piotr: is this still an issue for you and the Jewel release? Jason Dillaman
10:27 PM Bug #8385 (Closed): RBD / QEMU Crash: Invalid fastbin entry (free)
Closing ticket as there have been no similar issues noted against Hammer or later releases. Jason Dillaman
10:22 PM Bug #16364 (Fix Under Review): Crash when utilizing advisory locking API functions
*PR*: https://github.com/ceph/ceph/pull/9882 Jason Dillaman
10:13 PM Bug #16364 (In Progress): Crash when utilizing advisory locking API functions
Jason Dillaman
10:12 PM Bug #16433 (Fix Under Review): Potential crash during journal::Replay shut down
*PR*: https://github.com/ceph/ceph/pull/9881 Jason Dillaman
10:02 PM Bug #16433 (Resolved): Potential crash during journal::Replay shut down
Replay<I>::handle_op_complete shouldn't be holding the mutex when invoking the callback:... Jason Dillaman
10:03 PM Subtask #15670 (Pending Backport): Delete image when a resync is requested
Jason Dillaman
09:30 PM Bug #16394 (Need More Info): Ceph RBD
@Junming: unfortunately, this isn't enough information to be acted upon. What were you doing when this occurred? Can... Jason Dillaman
09:08 PM Backport #16426 (Resolved): jewel: Possible race condition during journal transition from replay ...
https://github.com/ceph/ceph/pull/10047 Nathan Cutler
09:08 PM Backport #16425 (Resolved): jewel: rbd-mirror: potential race condition accessing local image jou...
https://github.com/ceph/ceph/pull/10046 Nathan Cutler
09:08 PM Backport #16424 (Resolved): jewel: Journal needs to handle duplicate maintenance op tids
https://github.com/ceph/ceph/pull/10045 Nathan Cutler
09:08 PM Backport #16423 (Resolved): jewel: Journal duplicate op detection can cause lockdep error
https://github.com/ceph/ceph/pull/10044 Nathan Cutler
09:08 PM Backport #16422 (Resolved): jewel: Sporadic test failures when image update notification missed
https://github.com/ceph/ceph/pull/9883 Nathan Cutler
05:13 PM Bug #16404 (Pending Backport): Sporadic test failures when image update notification missed
Mykola Golub
02:20 PM Bug #16411 (Fix Under Review): rbd-mirror should disable proxied maintenance ops for non-primary ...
*PR*: https://github.com/ceph/ceph/pull/9867 Jason Dillaman
12:43 PM Bug #16411 (In Progress): rbd-mirror should disable proxied maintenance ops for non-primary image
Jason Dillaman
12:29 PM Bug #16411 (Resolved): rbd-mirror should disable proxied maintenance ops for non-primary image
When rbd-mirror is replaying updates into a local image, it owns the exclusive lock and currently is accepting mainte... Jason Dillaman
06:51 AM Bug #16362 (Pending Backport): Journal needs to handle duplicate maintenance op tids
Mykola Golub

06/21/2016

07:19 PM Bug #15150 (Duplicate): librbd: AioImageRequestWQ::aio_write triggered helgrind assert
Tracking helgrind issues under tracker #14846 Jason Dillaman
07:18 PM Bug #15007 (Resolved): Cannot rebuild object map in in-flight IO
This issue is resolved in Jewel and we have stress-test IO cases to verify it properly functions. Infernalis is effec... Jason Dillaman
06:59 PM Bug #16404 (Fix Under Review): Sporadic test failures when image update notification missed
*PR*: https://github.com/ceph/ceph/pull/9850 Jason Dillaman
06:58 PM Bug #16404 (In Progress): Sporadic test failures when image update notification missed
Jason Dillaman
06:34 PM Bug #16404 (Resolved): Sporadic test failures when image update notification missed
Hammer and Infernalis automatically mark the image as updated when proxying requests to the exclusive lock holder to ... Jason Dillaman
04:52 PM Bug #16363 (Pending Backport): Journal duplicate op detection can cause lockdep error
Mykola Golub
04:51 PM Bug #16198 (Pending Backport): Possible race condition during journal transition from replay to r...
Mykola Golub
04:51 PM Bug #16230 (Pending Backport): rbd-mirror: potential race condition accessing local image journal
Mykola Golub
02:07 AM Bug #16394: Ceph RBD
./log/SubsystemMap.h: In function 'bool ceph::log::SubsystemMap::should_gather(unsigned int, int)' thread 2abc8630270... junming rao
02:03 AM Bug #16394 (Rejected): Ceph RBD
junming rao

06/20/2016

04:04 PM Bug #16349 (In Progress): rbd-mirror: gracefully handle being blacklisted
Jason Dillaman
04:03 PM Bug #16362 (Fix Under Review): Journal needs to handle duplicate maintenance op tids
*PR*: https://github.com/ceph/ceph/pull/9823 Jason Dillaman
02:36 PM Bug #16362 (In Progress): Journal needs to handle duplicate maintenance op tids
Jason Dillaman
02:34 PM Bug #16363 (Fix Under Review): Journal duplicate op detection can cause lockdep error
*PR*: https://github.com/ceph/ceph/pull/9819 Jason Dillaman
02:26 PM Bug #16363 (In Progress): Journal duplicate op detection can cause lockdep error
Jason Dillaman
01:42 PM Bug #16198 (Fix Under Review): Possible race condition during journal transition from replay to r...
*PR*: https://github.com/ceph/ceph/pull/9816 Jason Dillaman
 

Also available in: Atom