Activity
From 10/11/2015 to 11/09/2015
11/09/2015
- 09:27 PM Bug #13727 (Resolved): TestClsRbd.snapshot failure
- 09:26 PM Bug #13727: TestClsRbd.snapshot failure
- Second issue covered by #13731
- 06:38 PM Bug #13727: TestClsRbd.snapshot failure
- For 'TestClsRbd.add_remove_child', OSD log shows that a child was added to the image -- yet the get child call failed...
- 02:07 PM Bug #13727 (Fix Under Review): TestClsRbd.snapshot failure
- *master PR*: https://github.com/ceph/ceph/pull/6504
- 02:06 PM Bug #13727 (In Progress): TestClsRbd.snapshot failure
- 01:59 PM Bug #13727: TestClsRbd.snapshot failure
- ...
- 01:58 PM Bug #13727 (Resolved): TestClsRbd.snapshot failure
- ...
- 07:07 PM Bug #13731 (Fix Under Review): TestClsRbd.add_remove_child failure
- https://github.com/ceph/ceph/pull/6511
- 04:44 PM Bug #13731 (Resolved): TestClsRbd.add_remove_child failure
- ...
- 05:58 PM Backport #13733 (Resolved): misdirected op in rbd balance-reads test
- https://github.com/ceph/ceph/pull/6629
- 04:26 PM Bug #13726 (Fix Under Review): QEMU hangs after creating snapshot and stopping VM
- *jewel PR*: https://github.com/ceph/ceph/pull/6508
- 03:01 PM Bug #13726 (In Progress): QEMU hangs after creating snapshot and stopping VM
- 02:59 PM Bug #13726: QEMU hangs after creating snapshot and stopping VM
- In-flight AIO read request forces an image refresh due to out-of-band snapshot creation. Detecting the newly created...
- 02:02 PM Bug #13726: QEMU hangs after creating snapshot and stopping VM
- I have attached the client log,
the snapshot create command was :
rbd -p pooltest --image vm-162-disk-1 snap cr... - 01:42 PM Bug #13726 (Resolved): QEMU hangs after creating snapshot and stopping VM
- When RBD cache is disabled, taking a snapshot and stopping the VM results in a hung QEMU process. Setting "rbd_non_b...
- 02:03 PM Bug #13491 (Pending Backport): misdirected op in rbd balance-reads test
- 01:40 PM Feature #13500 (Resolved): Optimize clone write path if object-map is enabled
- 01:40 PM Bug #13720 (Resolved): librbd: perf counter section name hase empty id
11/07/2015
- 09:27 PM Bug #13720 (Fix Under Review): librbd: perf counter section name hase empty id
- 08:00 PM Bug #13720: librbd: perf counter section name hase empty id
- https://github.com/ceph/ceph/pull/6494/
- 07:48 PM Bug #13720 (Resolved): librbd: perf counter section name hase empty id
- The expected format for librbd perf counter section name is: librbd-%{id}-%{pool}/name[@%{snap}]. But in generated na...
- 07:22 PM Bug #13719: perf dump for librbd socket generates invalid xml due to slash in attribute name
- The easiest fix (workaround) could be not to use "/" as a separator between pool and image name.
- 07:19 PM Bug #13719 (Resolved): perf dump for librbd socket generates invalid xml due to slash in attribut...
- % ceph --format xml --admin-daemon out/client.admin.19285.asok perf dump > /tmp/perf.xml
% xmllint /tmp/perf.xml
/t...
11/06/2015
- 03:29 PM Bug #8000: SLAB: Unable to allocate memory on node 0
- I created issue #13712 since this report is closed.
- 03:18 PM Bug #8000: SLAB: Unable to allocate memory on node 0
- I seem to have the exact same problem on kernel 4.2.3 so I backtracked to a working kernel which was 3.16.
This bu... - 02:41 PM Bug #13711 (Resolved): RBD: formatter session may not be rightfully closed.
- RBD: in do_list and do_disk_usage, formatter sessions may not be rightfully closed on error exit.
11/05/2015
- 09:10 PM Feature #12678 (Resolved): Support the ability to rename snapshots
- 05:00 PM Feature #13706 (Resolved): Online sparsify RBD images from the CLI
- With the future introduction of the CMPEXT (compare extent) OSD op, it will become possible to safely detect zeroed-b...
- 03:59 PM Backport #13705 (In Progress): rbd : enable feature objectmap
- 03:58 PM Backport #13705 (Resolved): rbd : enable feature objectmap
- https://github.com/ceph/ceph/pull/6477
- 03:35 PM Bug #13558 (Pending Backport): rbd : enable feature objectmap
- 01:37 PM Backport #13619: rbd clone issue
- Assigning to Jason because he merged the master PR, not because any action is expected of him.
- 01:36 PM Backport #13619 (In Progress): rbd clone issue
11/04/2015
- 08:22 PM Backport #13695 (Resolved): init-rbdmap uses distro-specific functions
- https://github.com/ceph/ceph/pull/6528
- 08:22 PM Backport #13694 (Rejected): init-rbdmap uses distro-specific functions
- 07:40 PM Bug #13553 (Pending Backport): rbd clone issue
11/03/2015
- 11:43 AM Feature #2468 (Fix Under Review): librbd: provide a way for a user to flush and invalidate the cache
- 09:20 AM Feature #2468: librbd: provide a way for a user to flush and invalidate the cache
- https://github.com/ceph/ceph/pull/6453
11/02/2015
- 09:55 PM Bug #13674 (Fix Under Review): Avoid re-writing old-format image header on resize
- *master PR*: https://github.com/ceph/ceph/pull/6447
- 09:35 PM Bug #13674 (Resolved): Avoid re-writing old-format image header on resize
- Race conditions are possible between snapshot creation and resize operations (especially if there are issues with wat...
- 07:43 PM Bug #13304 (Fix Under Review): "TestLibRBD.Metadata , TestLibRBD.UpdateFeatures" tests failed in ...
- *PR*" https://github.com/ceph/ceph-qa-suite/pull/669
- 07:03 PM Bug #13432: "Mutex.cc: 105: FAILED assert(r == 0)" in rbd run
- the fix is pending upstream for fio.
not a blocker for ceph. once fio bug is fixed, the teuthology test has to be m...
10/31/2015
- 03:45 PM Bug #13432: "Mutex.cc: 105: FAILED assert(r == 0)" in rbd run
- Also in run: http://pulpito.ceph.com/teuthology-2015-10-27_23:00:01-rbd-infernalis-distro-basic-multi/
Jobs: ['11295...
10/29/2015
- 07:27 PM Bug #13636: rbd: pure virtual method called
- infernalis is effectively closed pending its eminent release. After its release, a jewel branch will be created from...
- 03:05 PM Bug #13636: rbd: pure virtual method called
- Hi - if you target infernalis with the fix, we would only have to backport to hammer and firefly. AFAIK master is sti...
- 02:07 PM Bug #13636: rbd: pure virtual method called
- ThreadPool::WorkQueueVal @ 0x7f0cb09d6360 is the ImageCtx::op_work_queue from the active image. Thread 4 shows that ...
- 01:44 PM Bug #13636: rbd: pure virtual method called
- ...
- 01:26 PM Bug #13636 (In Progress): rbd: pure virtual method called
- 05:13 AM Bug #13636 (Resolved): rbd: pure virtual method called
- From http://qa-proxy.ceph.com/teuthology/teuthology-2015-10-23_23:00:08-rbd-master---basic-multi/1122358/teuthology.l...
- 03:16 PM Bug #12415 (Pending Backport): init-rbdmap uses distro-specific functions
- Merged upstream (infernalis).
10/28/2015
- 06:02 PM Feature #13633: Verify self-managed snapshot functionality on image create
- User will see something like the following error if the pool is not in self-managed snapshot mode:...
- 05:56 PM Feature #13633 (Resolved): Verify self-managed snapshot functionality on image create
- If a user accidentally creates a pool snapshot on an RBD pool, images can no longer be snapshotted. There is no way ...
- 09:28 AM Backport #13541: LTTng-UST should be optionally enabled
- @Nathan : Here is the PR that I am working on -https://github.com/ceph/ceph/pull/6357-. It looks non-trivial backport...
10/27/2015
- 08:34 PM Feature #13504 (Fix Under Review): rbd: API: options on image create
- 07:52 PM Backport #13619: rbd clone issue
- Master PR -not- merged -yet-.
- 07:51 PM Backport #13619 (Resolved): rbd clone issue
- https://github.com/ceph/ceph/pull/6474
- 03:01 PM Backport #13388: librbd: reads larger than cache size hang
- Test case fix is non-trivial to backport.
- 02:45 PM Bug #13558 (Fix Under Review): rbd : enable feature objectmap
- 02:44 PM Bug #13558: rbd : enable feature objectmap
- master PR : https://github.com/ceph/ceph/pull/6339
- 02:22 PM Bug #13560 (Fix Under Review): unknown argument --quiet in udevadm settle
- *master PR*: https://github.com/ceph/ceph/pull/6394
- 01:22 PM Bug #13608 (Duplicate): can not clone a image with deep-flatten image-feature.
- Closing as duplicate of #13553
- 02:55 AM Bug #13608 (Duplicate): can not clone a image with deep-flatten image-feature.
- ceph version I used: ceph -v output:
ceph version 9.1.0-299-g89b2b9b
tried to clone a image with deep-flatten ima... - 01:22 PM Bug #13553 (Fix Under Review): rbd clone issue
- *master PR*: https://github.com/ceph/ceph/pull/6334
- 12:42 PM Backport #13567 (Resolved): FAILED assert(m_ictx->owner_lock.is_locked())
- 12:41 PM Backport #13541: LTTng-UST should be optionally enabled
- @Abhishek: You are working on this, but have not opened a PR yet - correct?
10/26/2015
- 03:32 PM Bug #13560 (In Progress): unknown argument --quiet in udevadm settle
- 02:15 PM Bug #13559: FAILED assert(m_ictx->owner_lock.is_locked())
- To confirm: this issue only affected v0.94.4, and v0.94.3 is unaffected?
What are the exact commit(s) that introdu... - 02:06 PM Bug #12415 (Fix Under Review): init-rbdmap uses distro-specific functions
- This should be fixed by
https://github.com/ceph/ceph/pull/6363
although the backports might be a bit tricky.
10/24/2015
- 01:13 PM Feature #13504: rbd: API: options on image create
- https://github.com/ceph/ceph/pull/6369
10/23/2015
- 06:21 PM Bug #13491 (Fix Under Review): misdirected op in rbd balance-reads test
- 06:21 PM Bug #13491: misdirected op in rbd balance-reads test
- *master PR*: https://github.com/ceph/ceph/pull/6364
- 03:53 PM Bug #13491 (In Progress): misdirected op in rbd balance-reads test
- 03:44 PM Feature #13465 (Fix Under Review): librbd: internal way to write multiple extents to an object at...
- *master PR*: https://github.com/ceph/ceph/pull/6359
- 01:07 AM Bug #13559: FAILED assert(m_ictx->owner_lock.is_locked())
- *ceph-qa-suite PR*: https://github.com/ceph/ceph-qa-suite/pull/652
10/22/2015
- 07:01 PM Backport #13541 (In Progress): LTTng-UST should be optionally enabled
- 06:23 PM Bug #13559: FAILED assert(m_ictx->owner_lock.is_locked())
- Updated interim fix for hammer will be available under wip-13567-hammer branch
- 09:28 AM Bug #13559: FAILED assert(m_ictx->owner_lock.is_locked())
- Jason Dillaman wrote:
> The interim fix packages are available via Ceph development repo (see http://docs.ceph.com/d... - 09:07 AM Bug #13559: FAILED assert(m_ictx->owner_lock.is_locked())
- https://github.com/ceph/ceph/pull/6344
- 03:00 PM Backport #13567: FAILED assert(m_ictx->owner_lock.is_locked())
- see also matching ceph-qa-suite pull request at https://github.com/ceph/ceph-qa-suite/pull/653
- 02:52 PM Backport #13567 (In Progress): FAILED assert(m_ictx->owner_lock.is_locked())
- 09:08 AM Backport #13567 (Resolved): FAILED assert(m_ictx->owner_lock.is_locked())
- https://github.com/ceph/ceph/pull/6348
- 02:51 PM Backport #13387 (In Progress): librbd: reads larger than cache size hang
10/21/2015
- 09:15 PM Bug #13559: FAILED assert(m_ictx->owner_lock.is_locked())
- -The interim fix packages are available via Ceph development repo (see http://docs.ceph.com/docs/master/install/get-p...
- 08:44 PM Bug #13559: FAILED assert(m_ictx->owner_lock.is_locked())
- Mihai, I've downloaded the required ceph packages, libraries, dependencies from ceph repo and installed it manually. ...
- 06:40 PM Bug #13559: FAILED assert(m_ictx->owner_lock.is_locked())
- The suggestion of disabling writeback caching with ...
- 05:50 PM Bug #13559: FAILED assert(m_ictx->owner_lock.is_locked())
- Andrei Mikhailovsky wrote:
> Tried downgrading ceph on host servers to version 0.94.3 as per jdillaman's suggestion ... - 03:30 PM Bug #13559: FAILED assert(m_ictx->owner_lock.is_locked())
- Tried downgrading ceph on host servers to version 0.94.3 as per jdillaman's suggestion on #ceph and the problem is go...
- 03:16 PM Bug #13559: FAILED assert(m_ictx->owner_lock.is_locked())
- See also #13545
- 02:56 PM Bug #13559 (In Progress): FAILED assert(m_ictx->owner_lock.is_locked())
- 01:15 PM Bug #13559 (Resolved): FAILED assert(m_ictx->owner_lock.is_locked())
- After upgrading from 0.94.1 to 0.94.4 I am having issues starting _SOME_ virtual guests with KVM. The following error...
- 08:45 PM Bug #13549: Locking assertion failure in librbd/LibrbdWriteback.cc
- Duplicate of the later ticket: http://tracker.ceph.com/issues/13559
- 12:50 AM Bug #13549 (Duplicate): Locking assertion failure in librbd/LibrbdWriteback.cc
- When attempting to boot a new QEMU VM, after an upgrade from 0.94.3 to 0.94.4:
librbd/LibrbdWriteback.cc: In funct... - 02:56 PM Bug #13545 (Duplicate): QEMU processes using librbd crashing with FAILED assert(m_ictx->owner_loc...
- Marking duplicate of #13559
- 01:22 PM Bug #13560 (Resolved): unknown argument --quiet in udevadm settle
- Seems 'rbd unmap' calls udevadm settle in krbd.c line 524:
} else {
/*
* libudev does not ... - 11:19 AM Bug #13558 (Resolved): rbd : enable feature objectmap
- how to produce this bug:
rbd -p rbd create --new-format --size 1024 test1
rbd feature enable test1 exclusive-lock... - 07:19 AM Bug #13553 (Resolved): rbd clone issue
- rbd CLI : rbd clone --image-feature layering rbd/test@snap rbd/clone
output : rbd: image format can only be set wh...
10/20/2015
- 06:39 PM Bug #13545 (Duplicate): QEMU processes using librbd crashing with FAILED assert(m_ictx->owner_loc...
- I performed an upgrade of Ceph from 0.94.2 to 0.94.4 in a lab scenario (we had been waiting for 0.94.4 to resolve the...
- 06:18 PM Bug #13115: librbd python bindings eat exceptions in callbacks
- I had a couple weeks full of IRQs and am just getting back to Ceph work now; I have another Ceph cluster to deploy th...
- 03:49 PM Bug #13115: librbd python bindings eat exceptions in callbacks
- Any updates on this? It'd be good to convert to cython sooner than later since there's bound to be more changes to th...
- 03:27 PM Bug #13221: rbd-replay: cryptic error when 'rbd' pool does not exist
- For reference, the fix for this issue is being backported to Hammer as a part of the patch series in #13460. See http...
- 02:13 PM Feature #13465 (In Progress): librbd: internal way to write multiple extents to an object at once
- 01:15 PM Feature #13494 (Fix Under Review): Improve rbd CLI bash completion
- *PR*: https://github.com/ceph/ceph/pull/6316
- 10:51 AM Backport #13541 (Resolved): LTTng-UST should be optionally enabled
- https://github.com/ceph/ceph/pull/6415
10/19/2015
- 12:35 PM Bug #13457 (Fix Under Review): Test and document "--cluster" option for RBD
- https://github.com/ceph/ceph/pull/6309
10/16/2015
- 07:09 PM Feature #13274 (Pending Backport): LTTng-UST should be optionally enabled
- 07:09 PM Feature #13274 (Fix Under Review): LTTng-UST should be optionally enabled
- *PR*: https://github.com/ceph/ceph/pull/6249
- 03:41 PM Feature #13494 (In Progress): Improve rbd CLI bash completion
- 06:20 AM Feature #13504: rbd: API: options on image create
- Discussed on ceph-dev (Subject: CEPH_RBD_API: options on image create).
http://www.spinics.net/lists/ceph-devel/ms... - 06:18 AM Feature #13504 (Resolved): rbd: API: options on image create
- It would be nice to pass options as key/value configuration pairs on image create instead of expanding rbd_create/rbd...
10/15/2015
- 09:34 PM Backport #13461 (Fix Under Review): librbd: object map invalid but not flagged as such
- 06:50 PM Backport #13461 (In Progress): librbd: object map invalid but not flagged as such
- 06:49 PM Bug #13496 (Resolved): librbd: rebuild_object_map() crashes when no object map exists
- 01:51 AM Bug #13496: librbd: rebuild_object_map() crashes when no object map exists
- https://github.com/ceph/ceph/pull/6270
- 01:45 AM Bug #13496 (Resolved): librbd: rebuild_object_map() crashes when no object map exists
- Doing something like:...
- 06:49 PM Bug #13483 (Resolved): 'Command failed ... rbd/qemu-iotests.sh'
- 04:47 PM Bug #13483 (Fix Under Review): 'Command failed ... rbd/qemu-iotests.sh'
- *PR*: https://github.com/ceph/ceph/pull/6284
- 03:43 PM Bug #13483: 'Command failed ... rbd/qemu-iotests.sh'
- http://qa-proxy.ceph.com/teuthology/teuthology-2015-10-13_23:00:07-rbd-infernalis-distro-basic-multi/1106716/teutholo...
- 12:41 AM Bug #13483 (In Progress): 'Command failed ... rbd/qemu-iotests.sh'
- 06:48 PM Backport #13460 (Fix Under Review): rbd-replay-prep: crashes on incorrect trace file
- 06:25 PM Backport #13460 (In Progress): rbd-replay-prep: crashes on incorrect trace file
- 03:28 PM Feature #13500 (Resolved): Optimize clone write path if object-map is enabled
- If we know that the clone does not have an associated object, we can skip the initial write attempt (which has an ass...
- 12:48 AM Feature #13494 (Resolved): Improve rbd CLI bash completion
- PR 5987 introduces a refactored rbd CLI that should make it trivial to dynamically generate the bash completion direc...
10/14/2015
- 11:09 PM Bug #13491 (Resolved): misdirected op in rbd balance-reads test
- http://pulpito.ceph.com/teuthology-2015-10-09_23:00:08-rbd-master---basic-multi/1098381/
and
http://pulpito.ce...
10/13/2015
- 08:37 PM Bug #13483 (Resolved): 'Command failed ... rbd/qemu-iotests.sh'
- http://qa-proxy.ceph.com/teuthology/teuthology-2015-10-10_23:00:02-rbd-infernalis-distro-basic-multi/1100362/teutholo...
- 07:52 PM Bug #13432: "Mutex.cc: 105: FAILED assert(r == 0)" in rbd run
- *fio PR*: https://github.com/axboe/fio/pull/121
- 07:34 PM Bug #13432: "Mutex.cc: 105: FAILED assert(r == 0)" in rbd run
- Race condition within fio rbd engine where a 'fio_rbd_iou' object is reused prior to retrieving the return value from...
- 03:30 PM Bug #13432 (In Progress): "Mutex.cc: 105: FAILED assert(r == 0)" in rbd run
- 12:06 PM Bug #13371 (Closed): "[ FAILED ] LibRBD.*" tests in upgrade:firefly-hammer-x:stress-split-infer...
- Appears to be resource starvation while running the test on a VPS node. The error code (-4) is from the monitor timi...
10/12/2015
- 11:00 PM Feature #13465 (Resolved): librbd: internal way to write multiple extents to an object at once
- To take advantage of scattered writes from the ObjectCacher (https://github.com/ceph/ceph/pull/5990), librbd needs to...
- 10:54 PM Documentation #13464 (Closed): rbd metadata docs
- New in Infernalis, these are for user-defined purposes, but setting conf_ followed by an rbd option (e.g. conf_rbd_ca...
- 08:44 PM Bug #13304: "TestLibRBD.Metadata , TestLibRBD.UpdateFeatures" tests failed in upgrade:client-upgr...
- Awaiting retest after PR 6215 (https://github.com/ceph/ceph/pull/6215) merges.
- 08:33 PM Bug #13371 (In Progress): "[ FAILED ] LibRBD.*" tests in upgrade:firefly-hammer-x:stress-split-...
- 07:59 PM Backport #13461 (Resolved): librbd: object map invalid but not flagged as such
- https://github.com/ceph/ceph/pull/6289
- 07:58 PM Backport #13460 (Resolved): rbd-replay-prep: crashes on incorrect trace file
- https://github.com/ceph/ceph/pull/6286
- 06:54 PM Bug #13372 (Pending Backport): librbd: object map invalid but not flagged as such
- 06:50 PM Bug #13405 (Resolved): rbd-replay-prep trace output is not version controlled
- 06:50 PM Bug #13384 (Resolved): rbd-replay-prep stack overflow while converting 'rbd bench-write' trace
- 06:50 PM Bug #13378 (Resolved): rbd-replay-prep requires >4 hrs to convert 'rbd bench-write' trace
- 06:50 PM Bug #13325 (Resolved): rbd-replay-prep doesn't handle new IO APIs
- 06:50 PM Bug #13221 (Resolved): rbd-replay: cryptic error when 'rbd' pool does not exist
- 06:49 PM Bug #13220 (Pending Backport): rbd-replay-prep: crashes on incorrect trace file
- 06:02 PM Bug #13457 (Resolved): Test and document "--cluster" option for RBD
- Test and document "--cluster" option for RBD
For using multiple ceph cluster from same RBD clients.
Also available in: Atom