Activity
From 04/23/2015 to 05/22/2015
05/22/2015
- 07:51 PM Bug #11562 (Resolved): docs: rbd-diff developer document lists image size as u64 instead of le64
- 03:19 PM Bug #11562 (Fix Under Review): docs: rbd-diff developer document lists image size as u64 instead ...
- *master PR*: https://github.com/ceph/ceph/pull/4748
- 03:18 PM Bug #11562 (In Progress): docs: rbd-diff developer document lists image size as u64 instead of le64
- 06:40 PM Bug #11743 (Resolved): Possible crash while concurrently writing and shrinking an image
- An image shrink will result in a flush and cache invalidation. If there is a pending write in the cache, the flush w...
- 03:01 PM Bug #11593 (Fix Under Review): diff_iterate API call should use callback's return code
- *master PR*: https://github.com/ceph/ceph/pull/4744
- 03:01 PM Feature #11625 (Fix Under Review): diff_iterate should utilize concurrent RADOS operations
- *master PR*: https://github.com/ceph/ceph/pull/4744
- 08:36 AM Backport #11731 (In Progress): Objectcacher setting max object counts too low
- 12:16 AM Backport #11731 (Resolved): Objectcacher setting max object counts too low
- https://github.com/ceph/ceph/pull/4639
- 12:15 AM Backport #11730 (Resolved): Objectcacher setting max object counts too low
- 12:15 AM Backport #11730 (Resolved): Objectcacher setting max object counts too low
- https://github.com/ceph/ceph/pull/4497
05/21/2015
05/20/2015
- 07:42 PM Bug #11579 (Fix Under Review): Copyup operation for deep flatten cannot use current parent extents
- *master PR*: https://github.com/ceph/ceph/pull/4730
05/17/2015
- 01:09 AM Bug #10116: Ceph vm guest disk lockup when using fio
- I haven't tested it yet, but its possible this kernel bug is what was biting me:
https://groups.google.com/forum/#!t...
05/15/2015
- 03:24 PM Feature #11625 (In Progress): diff_iterate should utilize concurrent RADOS operations
- 03:23 PM Bug #11537 (Fix Under Review): librbd: crash when two clients try to write to an exclusive locked...
- *master PR*: https://github.com/ceph/ceph/pull/4695
05/14/2015
- 02:00 PM Feature #11625 (Resolved): diff_iterate should utilize concurrent RADOS operations
- The diff_iterate API (which is used for many rbd CLI tools) operates on each block serially. Performance could be im...
05/13/2015
- 04:26 PM Bug #11551 (Resolved): fsx: krbd mode fails to clone
- 02:24 PM Bug #11551: fsx: krbd mode fails to clone
- *master PR*: https://github.com/ceph/ceph/pull/4669
- 12:12 PM Bug #11551: fsx: krbd mode fails to clone
- *next PR*: https://github.com/ceph/ceph/pull/4665
- 12:11 PM Bug #11551 (Fix Under Review): fsx: krbd mode fails to clone
- 05:47 AM Feature #7746 (Fix Under Review): Capacity Management: rbd df
- 05:45 AM Feature #7746 (Duplicate): Capacity Management: rbd df
- 05:46 AM Feature #10154 (Resolved): librbd: use early snapshot context for copyup operations so snapshots ...
- 05:45 AM Feature #2167 (Duplicate): rbd: report the allocated size/space of a RBD with "rbd info <rbd name>"
05/12/2015
- 04:44 PM Bug #11593 (Resolved): diff_iterate API call should use callback's return code
- If the callback from diff_iterate returns an error, this return code is ignored by diff_iterate and is not bubbled up...
- 04:17 PM Feature #11405: qa: test rbd notify-based proxying across versions
- *ceph-qa-suite PR*: https://github.com/ceph/ceph-qa-suite/pull/438
- 04:04 PM Feature #11405: qa: test rbd notify-based proxying across versions
- *ceph master PR*: https://github.com/ceph/ceph/pull/4644
- 04:03 PM Feature #11405 (Fix Under Review): qa: test rbd notify-based proxying across versions
- 04:15 PM Feature #4605 (Duplicate): rbd xfstests: define qa user, group, etc.
- 04:14 PM Feature #4136 (Duplicate): rbd xfstests: update list of tests run
- 04:12 PM Feature #4089 (Closed): rbd: improve small I/O performance
- 04:11 PM Feature #2586 (Rejected): rbd: check/take locks on --lock
- 04:11 PM Feature #5187 (Resolved): rbd: allow unmap using mapped image name
- 04:03 PM Feature #11372 (Resolved): qa: run librbd tests under valgrind
- 10:25 AM Bug #11478 (Resolved): Queued AIO reference counters not properly updated
- 10:24 AM Bug #11370 (Resolved): A retransmit of proxied flatten request can result in -EINVAL
- 10:24 AM Bug #11369 (Resolved): Periodic failure of TestLibRBD.DiffIterateStress
05/11/2015
- 06:08 PM Bug #11587 (Resolved): inconsistency with rbd-replay utilities and --with-debug option
- In v0.94.1, @make install@ installs @/usr/bin/rbd-replay@ and @/usr/bin/rbd-replay-many@ even when the user does not ...
05/10/2015
- 03:54 PM Bug #7385: Objectcacher setting max object counts too low
- * firefly backport https://github.com/ceph/ceph/pull/4639
05/08/2015
- 07:41 PM Bug #11577 (Resolved): fsx deadlocked in several of the most recent runs
- 05:34 PM Bug #11577 (Fix Under Review): fsx deadlocked in several of the most recent runs
- *master PR*: https://github.com/ceph/ceph/pull/4623
- 11:58 AM Bug #11577 (In Progress): fsx deadlocked in several of the most recent runs
- 11:57 AM Bug #11577 (Resolved): fsx deadlocked in several of the most recent runs
- http://qa-proxy.ceph.com/teuthology/jdillaman-2015-05-06_21:47:33-rbd-master---basic-multi/878909/teuthology.log
htt... - 05:26 PM Bug #11579 (Resolved): Copyup operation for deep flatten cannot use current parent extents
- If an image has shrunk in size after an existing snapshot, it's possible that the copyup operation will only read the...
- 11:55 AM Bug #11536 (Resolved): object map invalid during fsx
- No object map issues detected on new FSX run.
05/07/2015
- 03:40 PM Bug #11553 (Closed): rbd: get_flags calls cause 'rbd info' to fail on pre-hammer osds
- Not an issue -- librbd's get_flags method does not invoke the OSD get_flags class method.
- 01:29 PM Bug #11553 (In Progress): rbd: get_flags calls cause 'rbd info' to fail on pre-hammer osds
- 01:28 PM Bug #11553 (Fix Under Review): rbd: get_flags calls cause 'rbd info' to fail on pre-hammer osds
- 01:21 PM Bug #11553 (In Progress): rbd: get_flags calls cause 'rbd info' to fail on pre-hammer osds
- 03:39 PM Bug #11557 (Resolved): test/librbd/test_librbd.cc:2295: Failure DiffIterateTest/1.DiffIterateStress
- 01:29 PM Bug #11557 (Fix Under Review): test/librbd/test_librbd.cc:2295: Failure DiffIterateTest/1.DiffIte...
- *master PR*: https://github.com/ceph/ceph/pull/4603
- 09:12 AM Bug #11557: test/librbd/test_librbd.cc:2295: Failure DiffIterateTest/1.DiffIterateStress
- probably because of https://github.com/ceph/ceph/pull/4593
- 09:09 AM Bug #11557 (Resolved): test/librbd/test_librbd.cc:2295: Failure DiffIterateTest/1.DiffIterateStress
- at commit:d20a387788d0a2cb0f97c9f2fb34720ce2861f89...
- 03:06 PM Bug #11562 (Resolved): docs: rbd-diff developer document lists image size as u64 instead of le64
05/06/2015
- 11:43 PM Bug #11553 (Closed): rbd: get_flags calls cause 'rbd info' to fail on pre-hammer osds
- cls_rbd::get_flags() behaves sensibly when the flags aren't present, but librbd will return an error if the
method d... - 11:11 PM Bug #11549 (Resolved): Infernalis librbd cannot open images on pre-Infernalis OSDs
- 06:56 PM Bug #11549 (Fix Under Review): Infernalis librbd cannot open images on pre-Infernalis OSDs
- *master PR*: https://github.com/ceph/ceph/pull/4589
- 06:40 PM Bug #11549 (Resolved): Infernalis librbd cannot open images on pre-Infernalis OSDs
- The addition of metadata to RBD images currently prevents opening images from older OSDs. Since metadata isn't a fea...
- 09:41 PM Bug #11551 (Resolved): fsx: krbd mode fails to clone
- http://typica002.front.sepia.ceph.com/teuthology-2015-05-05_23:06:01-krbd-master-testing-basic-typica/11129/teutholog...
- 09:30 PM Feature #7746 (Fix Under Review): Capacity Management: rbd df
- 09:30 PM Feature #7746: Capacity Management: rbd df
- *master PR*: https://github.com/ceph/ceph/pull/4592
- 07:06 PM Bug #11537 (In Progress): librbd: crash when two clients try to write to an exclusive locked image
- 07:06 PM Bug #11536 (In Progress): object map invalid during fsx
- Should be fixed by PR4530 which merged yesterday. I will reschedule the FSX tests to verify before marking as resolved.
- 06:25 PM Feature #11405: qa: test rbd notify-based proxying across versions
- FYI skeleton for a new suite is here wip_11413_2
- 05:34 PM Bug #11370: A retransmit of proxied flatten request can result in -EINVAL
- *hammer PR*: https://github.com/ceph/ceph/pull/4585
- 09:47 AM Bug #11370: A retransmit of proxied flatten request can result in -EINVAL
- * incorrect and closed hammer backport https://github.com/ceph/ceph/pull/4565
- 09:49 AM Bug #11478: Queued AIO reference counters not properly updated
- * hammer backport https://github.com/ceph/ceph/pull/4566
- 09:30 AM Bug #11369: Periodic failure of TestLibRBD.DiffIterateStress
- * hammer backport https://github.com/ceph/ceph/pull/4564
- 09:28 AM Bug #11363 (Resolved): ImageWatcher should cancel in-flight ops on watch error
- 09:27 AM Bug #11342 (Resolved): librbd notification race condition on snap_create
- 09:26 AM Bug #11113 (Resolved): RBD snapshot, created while flattening image, is unremovable
- 09:25 AM Bug #5488 (Resolved): librbd: deadlock in image refresh
05/05/2015
- 11:15 PM Bug #11537 (Resolved): librbd: crash when two clients try to write to an exclusive locked image
- Using two instances of rbd bench-write on the same image at the same time:...
- 09:10 PM Bug #11536 (Resolved): object map invalid during fsx
- A bunch of jobs in this run hit this in similar circumstances, noticing it after cloning and trying to resize an imag...
- 04:09 PM Bug #10051 (Won't Fix): kernel-mounted RBD image may block shutdown
05/04/2015
- 03:36 PM Feature #11372 (Fix Under Review): qa: run librbd tests under valgrind
- *ceph-qa-suite*: https://github.com/ceph/ceph-qa-suite/pull/426
*ceph master*: https://github.com/ceph/ceph/pull/4461 - 03:35 PM Feature #11372 (Resolved): qa: run librbd tests under valgrind
- 03:31 PM Feature #11405 (In Progress): qa: test rbd notify-based proxying across versions
04/29/2015
- 11:42 PM Fix #7787 (Resolved): rbd diff takes longer as images grow larger
- 06:03 PM Bug #11363: ImageWatcher should cancel in-flight ops on watch error
- * hammer backport https://github.com/ceph/ceph/pull/4499
- 06:00 PM Bug #11342: librbd notification race condition on snap_create
- * hammer backport https://github.com/ceph/ceph/pull/4498
- 05:57 PM Bug #7385: Objectcacher setting max object counts too low
- * hammer backport https://github.com/ceph/ceph/pull/4497
- 05:35 PM Bug #5488: librbd: deadlock in image refresh
- * hammer backport https://github.com/ceph/ceph/pull/4496
- 04:37 PM Bug #11487 (Duplicate): Remove big size image so slowly.
- 03:20 PM Bug #11502 (Resolved): "[ FAILED ] TestLibRBD.LockingPP" in rbd-next-distro-basic-mult run
- Run: http://pulpito.ceph.com/teuthology-2015-04-27_23:00:01-rbd-next-distro-basic-multi/
Jobs: ['866113', '866154']
...
04/28/2015
- 11:43 PM Tasks #11472 (Closed): Remove big size image so slowly.
- This is because rbd does not track which objects exists without the objectmap feature (available in hammer).
With ... - 04:08 PM Feature #11372 (Fix Under Review): qa: run librbd tests under valgrind
- 01:54 AM Bug #11487 (Duplicate): Remove big size image so slowly.
- Ceph server 0.94.1
Ceph client 0.94.1.
I try to create new image 8 PB size and use rbd command and librados library...
04/27/2015
- 09:08 PM Bug #11478 (Pending Backport): Queued AIO reference counters not properly updated
- 08:20 PM Bug #11478 (Fix Under Review): Queued AIO reference counters not properly updated
- *next PR*: https://github.com/ceph/ceph/pull/4472
- 08:05 PM Bug #11478 (Resolved): Queued AIO reference counters not properly updated
- When queuing an aio_write or aio_discard while waiting for the exclusive lock, the reference counters for the associa...
- 08:16 PM Bug #11477 (Resolved): test: rbd cram test failures
- 06:37 PM Bug #11477 (Fix Under Review): test: rbd cram test failures
- *master PR*: https://github.com/ceph/ceph/pull/4471
- 06:33 PM Bug #11477 (Resolved): test: rbd cram test failures
- http://qa-proxy.ceph.com/teuthology/teuthology-2015-04-22_23:00:02-rbd-master-testing-basic-multi/858971/teuthology.log
- 08:13 PM Bug #11475 (Resolved): test: multiple failures of Python rbd test cases
- 05:47 PM Bug #11475 (Fix Under Review): test: multiple failures of Python rbd test cases
- *master PR*: https://github.com/ceph/ceph/pull/4470
- 05:20 PM Bug #11475 (Resolved): test: multiple failures of Python rbd test cases
- http://qa-proxy.ceph.com/teuthology/teuthology-2015-04-22_23:00:02-rbd-master-testing-basic-multi/858858/teuthology.log
- 08:12 PM Bug #11380 (Resolved): "[ FAILED ] cls_rbd.get_features" in upgrade:giant-x-next-distro-basic-m...
- 04:11 PM Bug #11380: "[ FAILED ] cls_rbd.get_features" in upgrade:giant-x-next-distro-basic-multi
- *next PR*: https://github.com/ceph/ceph/pull/4469
- 03:44 PM Bug #11380 (Pending Backport): "[ FAILED ] cls_rbd.get_features" in upgrade:giant-x-next-distro...
- 01:43 PM Bug #11370 (Pending Backport): A retransmit of proxied flatten request can result in -EINVAL
- 07:37 AM Tasks #11472 (Closed): Remove big size image so slowly.
- Ceph server 0.94.1
Ceph client 0.94.1.
I try to create new image 8 PB size and use rbd rm 8PB image that is so slow...
04/24/2015
- 06:46 PM Bug #7385 (Pending Backport): Objectcacher setting max object counts too low
- commit:0b378942c4f1b79cb65967f2d3466728ca1c8d5b
- 04:42 PM Bug #11380: "[ FAILED ] cls_rbd.get_features" in upgrade:giant-x-next-distro-basic-multi
- It does not seem to be on next branch and therefor jobs failing.
Needs backport?
04/23/2015
Also available in: Atom