Activity
From 07/30/2015 to 08/28/2015
08/28/2015
- 11:31 PM rgw Bug #12830 (Closed): S3 PUT Object returns 411 Length Required after upgrading to 0.94.3
- rgw 0.94.3, nginx 1.8.0, RHEL 6.6.
After upgrading from 0.94.2, S3 PUT Object operations return 411 Length Require... - 10:37 PM Bug #12437 (Pending Backport): Mutex Assert from PipeConnection::try_get_pipe
- 10:11 PM Revision cfce56fa (ceph): rados/basic/msgr: enable all experimental features if random/async
- Signed-off-by: Kefu Chai <kchai@redhat.com>
- 09:50 PM Revision d155519b (ceph): Revert "rados/basic/msgr: enable all experimental features if random/as...
- This reverts commit 40db8376ab42756e7c2f45469153099183439744.
Getting errors trying to schedule with this commit. - 09:49 PM Revision 788c5584 (ceph): Merge pull request #5333 from ceph/wip-12437
- osd: Keep a reference count on Connection while calling send_mes…
Reviewed-by: Samuel Just <sjust@redhat.com>
Review... - 09:20 PM Bug #12803 (Duplicate): osd: op_commit on cancelled op
- 04:07 PM Bug #12803: osd: op_commit on cancelled op
- /a/sage-2015-08-26_09:07:57-rados-wip-sage-testing---basic-multi/1033667
- 09:16 PM Bug #12828: ReplicatedBackend: populate recovery_info.size for clone (bug symptom is size mismatc...
- /a/samuelj-2015-08-26_12:18:34-rados-wip-sam-testing-distro-basic-multi/1033873/remote
- 09:15 PM Bug #12828: ReplicatedBackend: populate recovery_info.size for clone (bug symptom is size mismatc...
- https://github.com/ceph/ceph/pull/5706
I'm out of town until 9/14 - 07:44 PM Bug #12828 (Resolved): ReplicatedBackend: populate recovery_info.size for clone (bug symptom is s...
- This apparently never got filled in. It wasn't actually a problem since we do populate the the copy and clone subset...
- 09:15 PM Bug #12829: buffer::list::write_fd on an empty buffer after seeking past the end of the file does...
- https://github.com/ceph/ceph/pull/5706
I'm out of town until 9/14 - 08:21 PM Bug #12829 (Resolved): buffer::list::write_fd on an empty buffer after seeking past the end of th...
- 09:15 PM Bug #12809: hit set clear repops fired in same epoch as map change -- segfault since they fall in...
- https://github.com/ceph/ceph/pull/5707
I'm out of town until 9/14 - 07:41 PM Revision 33ad04c7 (ceph): Merge remote-tracking branch 'gh/next'
- 07:20 PM Revision bbe9f577 (ceph): Merge remote-tracking branch 'gh/infernalis'
- 06:44 PM Backport #12751 (Resolved): kvm die with assert(m_seed < old_pg_num)
- 07:06 AM Backport #12751 (In Progress): kvm die with assert(m_seed < old_pg_num)
- 06:44 PM Backport #12813 (Resolved): is_new_interval: size change should cause new interval
- 06:43 PM Revision ed162d49 (ceph): Merge pull request #5691 from jdurgin/wip-10399-hammer
- is_new_interval() fixes
Reviewed-by: Samuel Just <sjust@redhat.com> - 05:33 PM Revision cc21514b (ceph): ceph-disk: {CentOS,RHEL} >= 7 && Fedora >= 22 are systemd
- http://tracker.ceph.com/issues/12786 Fixes: #12786
Signed-off-by: Loic Dachary <ldachary@redhat.com> - 05:19 PM Revision cbaac86f (ceph): Merge pull request #5690 from dzafman/wip-pgls-zafman
- cls: Fix successful return found by compiler warning
Reviewed-by: Sage Weil <sage@redhat.com> - 04:34 PM Revision d5be625c (ceph): Merge pull request #548 from tchaikov/wip-enable-exp-features
- rados/basic/msgr: enable all experimental features if random/async
Reviewed-by: Samuel Just <sjust@redhat.com> - 04:23 PM Revision 40db8376 (ceph): rados/basic/msgr: enable all experimental features if random/async
- Signed-off-by: Kefu Chai <kchai@redhat.com>
- 04:21 PM Bug #12775 (Pending Backport): common: do not insert emtpy ptr when rebuild emtpy bufferlist
- 04:21 PM Bug #12785 (Resolved): shard 4 missing -2/00000014/temp_recovering_0.14_25'135_118_head/head
- 04:20 PM Bug #9221 (Resolved): LibRadosTwoPoolsPP.PromoteOn2ndRead
- 04:20 PM Revision 8534ae6d (ceph): Merge pull request #5704 from tchaikov/wip-adventure-mode
- common: 'enable experimental data corrupting features' now understand '*'
Reviewed-by: Sage Weil <sage@redhat.com> - 04:18 PM Revision 6d5d0726 (ceph): Merge pull request #5667 from liewegas/wip-9221
- ceph_test_rados_api_tier: make PromoteOn2ndRead tolerate thrashing
Reviewed-by: Samuel Just <sjust@redhat.com> - 04:17 PM Revision d75824d9 (ceph): Merge pull request #5666 from liewegas/wip-12785
- osd: fix off-by-one in hobject_t::is_temp() (bug 12785)
Reviewed-by: Samuel Just <sjust@redhat.com> - 04:15 PM Revision a8959820 (ceph): common: 'enable experimental data corrupting features' now understands '*'
- '*' allows any feature.
Signed-off-by: Kefu Chai <kchai@redhat.com> - 04:14 PM Revision c2a9641c (ceph): Merge pull request #5609 from rohanmars/wip-timegm
- added boost timegm impl for cross platform support
In the future it would be nice to avoid duplication of the code i... - 04:14 PM rados-java Support #12826 (Resolved): Rados-java ioctx.read(objectName, 20, 0, buf) returns negative value
- -2 is ENOENT. Probably the object doesn't exist in the pool you're trying to find it from. (The code you've pasted ce...
- 04:08 PM rados-java Support #12826 (Resolved): Rados-java ioctx.read(objectName, 20, 0, buf) returns negative value
- Please have a look at my program:
ioctx = cluster.ioCtxCreate(poolName);
byte[] bu... - 04:13 PM Revision 4c001839 (ceph): Merge pull request #5627 from wonzhq/agent-work
- osd: do evict before flush in agent_work
Reviewed-by: Sage Weil <sage@redhat.com> - 04:13 PM Revision 620c4177 (ceph): Merge pull request #5640 from liewegas/wip-mutex-lock
- mutex: avoid trylock unless instrumentation is enabled
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Piot... - 04:12 PM Revision 7942db55 (ceph): Merge pull request #5656 from XinzeChi/wip-buffer-emtpyptr
- common: fix insert empty ptr when bufferlist rebuild
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Sage Wei... - 03:58 PM RADOS Bug #12687: osd thrashing + pg import/export can cause maybe_went_rw intervals to be missed
- maybe/probably also this run on 1/d6c871c1/plana798404-412/head in 1.1, run /a/sage-2015-08-26_09:07:57-rados-wip-sag...
- 03:45 PM Revision 9c86b8e6 (ceph): Merge pull request #5453 from wonzhq/recover-primary
- osd: recovery fixes
Reviewed-by: Sage Weil <sage@redhat.com> - 03:40 PM Revision 2b17da39 (ceph): Merge pull request #5415 from wonzhq/do-recovery
- osd: do_recovery cleanup
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com> - 03:14 PM Revision 4a345435 (ceph): Merge pull request #5703 from ceph/wip-boost-random-cmake
- CMake: update for boost_random
- 02:34 PM Revision a3fc6e83 (ceph): CMake: update for boost_random
- This became a dependency in dbcaa544
Signed-off-by: John Spray <john.spray@redhat.com> - 02:29 PM Bug #12825: "test/librados/tier.cc" EINVAL on hammer -> infernalis
- oops, this should stay open
- 02:27 PM Bug #12825 (Rejected): "test/librados/tier.cc" EINVAL on hammer -> infernalis
- already backported.. see commit:58e62662f6ef04ac76470090d1d958467e34194a
- 02:22 PM Bug #12825 (Resolved): "test/librados/tier.cc" EINVAL on hammer -> infernalis
- In new run http://pulpito.ceph.com/teuthology-2015-08-27_13:22:31-upgrade:firefly-hammer-x:parallel-infernalis-distro...
- 02:14 PM rgw Bug #12690: rgw: test_multipart_upload_incorrect_etag fails
- Run: http://pulpito.ceph.com/teuthology-2015-08-28_05:00:06-smoke-master-distro-basic-multi/
Job: 1035629 - 01:27 PM Bug #12100: OSD crash, uexpected aio error in FileJournal.cc
- Just when we thought it wouldn't come back, last night we saw a crash with this same trace again on one of the 48 OSD...
- 01:25 PM Bug #12824 (Can't reproduce): osd/ReplicatedPG.cc: 10604: FAILED assert(obc) in hit_set_remove_all
- ...
- 01:21 PM Bug #12746: failed assert ctx.manager.pg_inconsistent(pool, 0) on repair_test.yaml
- /a/sage-2015-08-26_09:07:57-rados-wip-sage-testing---basic-multi/1033691
- 01:19 PM Bug #12681: tcmalloc doesn't release memory back to system
- Do you have any special config settings, or is everything on the default?
- 01:13 PM CephFS Feature #10369: qa-suite: detect unexpected MDS failovers and daemon crashes
- We just keep re-creating this feature: #12821
- 01:13 PM CephFS Bug #12821 (Duplicate): mds_thrasher: handle MDSes failing on startup
- #10369
- 01:11 PM CephFS Bug #12821: mds_thrasher: handle MDSes failing on startup
- This is kind of a special case of http://tracker.ceph.com/issues/10369 -- there is a more general need for something ...
- 11:31 AM CephFS Bug #12821 (Duplicate): mds_thrasher: handle MDSes failing on startup
- http://pulpito.ceph.com/teuthology-2015-08-21_23:04:01-fs-master---basic-multi/1026045/...
- 01:10 PM devops Bug #12812 (Rejected): Something fails to collect rocksdb sources for dist
- see Makefile-rocksdb.am for hte hacky way that this is include din 'make dist' with autotools. NOte that i am pullin...
- 01:09 PM Revision afa92e57 (ceph): common/SubProcess: silence compiler warnings
- * the STD{IN,OUT,ERR}_FILENO are closed when being dup2'ed so we
can not write to cout or cerr in child processes, ... - 01:02 PM CephFS Bug #12822 (Resolved): ceph-fuse crash in test_client_recovery
- If we actually see that it's crashing with the timeout we can reopen this.
- 12:58 PM CephFS Bug #12822: ceph-fuse crash in test_client_recovery
- I would expect us to see a backtrace from ceph-fuse stderr in the case of an actual crash. Seems more like the clien...
- 11:53 AM CephFS Bug #12822 (Resolved): ceph-fuse crash in test_client_recovery
- http://pulpito.ceph.com/teuthology-2015-08-17_23:04:01-fs-master---basic-multi/1020395/
Sadly there are absolutely... - 01:01 PM Revision 1c250b53 (ceph): Merge pull request #547 from ceph/wip-12822
- tasks/cephfs: timeout on wait in test_network_death
Reviewed-by: Greg Farnum <gfarnum@redhat.com> - 12:57 PM Revision ffe79e9f (ceph): tasks/cephfs: timeout on wait in test_network_death
- So that we can get a failure instead of a hang in
Signed-off-by: John Spray <john.spray@redhat.com> - 12:12 PM CephFS Feature #12823 (Rejected): cephfs_test_runner: print test names when executing them
- They're already logged, like this:...
- 11:57 AM CephFS Feature #12823 (Rejected): cephfs_test_runner: print test names when executing them
- It looks like we don't print out test names. When running through a whole suite that makes telling where we are in a ...
- 12:10 PM CephFS Bug #12776: qa: standby MDS not shutting down, "reached maximum tries (50) after waiting for 300 ...
- Actually, I just tried sending SIGTERM to a standby mds here, and it's getting stuck too.
- 12:08 PM CephFS Bug #12776: qa: standby MDS not shutting down, "reached maximum tries (50) after waiting for 300 ...
- It's getting the signal, but not making it through shutdown:...
- 11:48 AM CephFS Bug #12820: stuck looping on 'ls /sys/fs/fuse/connections'
- http://pulpito.ceph.com/teuthology-2015-08-17_23:04:01-fs-master---basic-multi/1020354/
- 11:24 AM CephFS Bug #12820 (Resolved): stuck looping on 'ls /sys/fs/fuse/connections'
- http://pulpito.ceph.com/teuthology-2015-08-21_23:04:01-fs-master---basic-multi/1025967/...
- 11:47 AM CephFS Bug #12612 (Can't reproduce): fuse jobs fail to start on centos7
- Okay, haven't seen this particular one again, just the new one #12820.
- 11:32 AM Revision cfcacb89 (ceph): mon: LogMonitor: handle boolean options consistently
- 'mon_cluster_log_to_syslog' gets a string of key/value pairs, values
in the pair being booleans, and keys being optio... - 10:46 AM Revision fbbe5b05 (ceph): mailmap: make h3c mailmap more robust
- Signed-off-by: Loic Dachary <ldachary@redhat.com>
- 10:44 AM Revision 51e6b710 (ceph): mailmap: sort {organization,mail}map
- Signed-off-by: Loic Dachary <ldachary@redhat.com>
- 10:23 AM Revision 34411efa (ceph): Merge pull request #5701 from guce/h3c_mail_organization_map
- h3c mail organization map
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 10:19 AM Revision c901e855 (ceph): doc:radosgw: correct typos of the command removing a subuser
- Fix typos in the example command removing a subuser, and delete the 'The Ceph Object Gateway' tag as it should not ap...
- 10:15 AM rgw Bug #12819 (Resolved): rgw: objects starting with underscore in bucket with versioning enabled ar...
- This issue still occurs after upgrading to 0.94.3, but only with objects created with versioning enabled.
Steps to... - 10:11 AM CephFS Bug #12808: smbtorture failure on scan-pipe
- It doesn't look like that in the error logs to me; I think it just failed to allocate — but perhaps I'm misreading th...
- 02:19 AM CephFS Bug #12808: smbtorture failure on scan-pipe
- this test case makes smbd allocate tens of GBs memory. maybe smbd got killed during the test
- 10:04 AM CephFS Bug #12806: nfs restart failures
- Zheng, do we have any idea how the machines are getting into that duplicated export state? It looks pretty clear that...
- 01:35 AM CephFS Bug #12806 (Resolved): nfs restart failures
- 01:35 AM CephFS Bug #12806: nfs restart failures
- ...
- 10:03 AM Linux kernel client Bug #12811 (Fix Under Review): ENXIO from kernel client (cephfs)
- https://github.com/ceph/ceph-client/commit/4a18ede97ba5029587b362446360cfd42379f9d0...
- 08:57 AM Linux kernel client Bug #12811: ENXIO from kernel client (cephfs)
- I compared osdmaps of good and bad mounts, found that the 'exist' flag is missing in bad mount's osdmap....
- 09:56 AM Revision e92d2f38 (ceph): h3c mail organization map
- update .mailmap and .organizationmap for h3c organization
Signed-off-by: Ce Gu <guce@h3c.com> - 09:38 AM CephFS Bug #12657 (Can't reproduce): Failure in TestStrays.test_ops_throttle
- Hmm, now this looks like a teuthology burp. The stats polling is meant to happen every second, but in this instance ...
- 08:37 AM Stable releases Tasks #11990 (Resolved): hammer v0.94.3
- 08:31 AM Bug #12818 (Resolved): ceph-disk prepare fails because of --setuser
- This can be reproduced with ceph-disk prepare on CentOS 7. I'm running into this while working on https://github.com/...
- 08:18 AM CephFS Bug #12777: qa: leftover files in cephtest directory
- merged to next and master.
- 01:56 AM CephFS Bug #12777: qa: leftover files in cephtest directory
- 08:17 AM Revision a3f8a44d (ceph): Merge remote-tracking branch 'origin/next'
- 06:42 AM Backport #12817 (Resolved): build_incremental() could take 40678 ms to finish
- https://github.com/ceph/ceph/pull/5697
- 06:40 AM Backport #12816 (Resolved): build_incremental() could take 40678 ms to finish
- https://github.com/ceph/ceph/pull/5698
- 06:11 AM Bug #12696: ceph-osd starts before network, fails
- Full startup log:
Please see attached file - 06:09 AM Bug #12696: ceph-osd starts before network, fails
- Sage Weil wrote:
> - what OS/distro is this?
>
> - is the ceph service enabled? the final start should have done... - 06:00 AM Revision 0bc909e5 (ceph): mon: add a cache layer over MonitorDBStore
- the cache of of leveldb does not perform well under some condition,
so we need a cache in our own stack.
* add an op... - 05:58 AM Revision 411769c1 (ceph): mon: add a cache layer over MonitorDBStore
- the cache of of leveldb does not perform well under some condition,
so we need a cache in our own stack.
* add an op... - 04:57 AM Bug #12012 (Fix Under Review): IO error on kvm/rbd with an erasure coded pool tier (after an upgr...
- https://github.com/ceph/ceph/pull/5695
- 03:05 AM Linux kernel client Bug #12815 (Closed): Stop the rbd map command due to the linux kernal crush.
- My ceph version is 0.87
[root@song ~]# uname -a
Linux song 3.10.0-123.el7.x86_64 #1 SMP Thu Jun 4 17:17:49 CST 20... - 02:08 AM Revision 52bc4886 (ceph): Merge pull request #5114 from branch-predictor/bp-reduced-memzero
- ObjectStore/FileJournal: reduce impact of memset()
Reviewed-by: Sage Weil <sage@redhat.com> - 01:50 AM Bug #12730 (Fix Under Review): common/Subprocess: build warnings
- https://github.com/ceph/ceph/pull/5693
- 01:16 AM Revision fe0729d3 (ceph): Merge pull request #545 from ceph/wip-12777-next
- tasks/cephfs: fix test_pool_perm teardown
08/27/2015
- 11:54 PM Bug #12814 (Can't reproduce): flipping the overlay from forward to seems to reorder writes
- ubuntu@teuthology:/a/samuelj-2015-08-26_12:18:34-rados-wip-sam-testing-distro-basic-multi/1033948
- 11:43 PM Backport #12813 (Resolved): is_new_interval: size change should cause new interval
- https://github.com/ceph/ceph/pull/5691
- 11:21 PM Revision a9c1601e (ceph): cls: Fix successful return found by compiler warning
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 10:31 PM Revision bee86660 (ceph): Objecter: pg_interval_t::is_new_interval needs pgid from previous pool
- When increasing the pg_num of a pool, an assert would fail since the
calculated pgid seed would be for the pool's new... - 10:25 PM Revision b5418b9a (ceph): osd_types::is_new_interval: size change triggers new interval
- Fixes: 11771
Signed-off-by: Samuel Just <sjust@redhat.com>
(cherry picked from commit ff79959c037a7145f7104b06d9e6a64... - 10:24 PM Bug #11771 (Pending Backport): is_new_interval: size change should cause new interval
- commit:ff79959c037a7145f7104b06d9e6a64492fdb95f
- 09:54 PM devops Bug #12812: Something fails to collect rocksdb sources for dist
- "trunk" tarball... not sure what that is
- 09:41 PM devops Bug #12812: Something fails to collect rocksdb sources for dist
- BTW, it appears that there's no way for me to dispose of an invalid issue.
- 09:40 PM devops Bug #12812: Something fails to collect rocksdb sources for dist
- I'm talking about the trunk tarball here, on 0.94. But strangely, I just tried it on 9.0.3 and "make dist" works as e...
- 09:09 PM devops Bug #12812: Something fails to collect rocksdb sources for dist
- The tarballs for v0.94 and v0.94.3 both include @src/rocksdb@:...
- 08:52 PM devops Bug #12812 (Rejected): Something fails to collect rocksdb sources for dist
- Originally, this started because the current RPM spec cannot be used to build 0.94 (either in-tree ceph.spec.in or th...
- 09:46 PM devops Feature #11881: ceph-disk support for multipath
- ...
- 08:51 PM devops Feature #11881: ceph-disk support for multipath
- Integration tests for multipath now pass on CentOS 7 non regression tests pass on Ubuntu 14.04. They should all pass ...
- 09:42 PM Bug #12809: hit set clear repops fired in same epoch as map change -- segfault since they fall in...
- 06:08 PM Bug #12809 (Resolved): hit set clear repops fired in same epoch as map change -- segfault since t...
- ubuntu@teuthology:/ceph/teuthology-archive/samuelj-2015-08-26_12:18:34-rados-wip-sam-testing-distro-basic-multi/10338...
- 09:32 PM Revision cfd6a6f0 (ceph): implement the ceph-disk suite
- Run the ceph-disk workunit with a mon setup with ceph-deploy. It is
prefered to the install done by the ceph tasks be... - 09:23 PM Revision d6c72c3a (ceph): ceph-deploy: implement only_mon
- If only_mon is true, do not attempt to deploy osds.
Signed-off-by: Loic Dachary <loic@dachary.org> - 09:20 PM Revision b1817831 (ceph): ceph-deploy: implement keep_running
- When keep_running is true, do not shutdown the cluster, leave it as it
is for other workunits or tasks to use. This e... - 09:19 PM Revision e5db657e (ceph): ceph-deploy: ceph report on failure
- When ceph-deploy fails, run ceph report to get more information about
the state of the cluster at the time of the fai... - 09:03 PM Revision c09c1192 (ceph): osd: Send reads to other shards if erasure coded chunk reads fail
- Handle errors in a common way whether redundant reads or not
Signed-off-by: David Zafman <dzafman@redhat.com> - 09:03 PM Revision d3b06edf (ceph): test: Fix comment in test-erasure-eio.sh
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 09:03 PM Revision 70e000a9 (ceph): test: Fix to expect no errors on 1 bad shard and errors with 2 bad shards
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 09:03 PM Revision 5eb2a77d (ceph): mon: add a new pool setting to configure fast read for EC pool
- Signed-off-by: Guang Yang <yguang@yahoo-inc.com>
- 09:03 PM Revision 54090f18 (ceph): mon: add a configuration for default fast read setting
- Signed-off-by: Guang Yang <yguang@yahoo-inc.com>
- 09:03 PM Revision ae1df24d (ceph): osd: Fix admin socket help output
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 09:03 PM Revision 1febe89d (ceph): osd: Avoid confusion by changing EC decode total_chunk_size to total_da...
- The length of the buffers being decoded may only be part of the chunk, so
less than the HashInfo total_chunk_size.
S... - 09:03 PM Revision 45014921 (ceph): common, osd: Remove osd_read_eio_on_bad_digest config variable
- In order to handle erasure coded reads where additional chunks would be able to
substitute for some bad chunks, we re... - 09:03 PM Revision da2987d7 (ceph): osd: Fix ECBackend to handle mismatch of total chunk size
- Fixes: #12200
Signed-off-by: David Zafman <dzafman@redhat.com> - 09:03 PM Revision 08f81a9b (ceph): osd: Check for EC decode errors, though none are possible at this time
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 09:03 PM Revision 21e9f69d (ceph): osd: Check CRC when able to on async read
- Fixes: #12000
Signed-off-by: David Zafman <dzafman@redhat.com> - 09:03 PM Revision f3eea4aa (ceph): test: Fix incorrect syntax in check for subread all feature
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 09:03 PM Revision bbdae537 (ceph): test: Enable EIO test code but expect error instead of osd crash
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 09:03 PM Revision a7c6b6a9 (ceph): test: Adding testing of shard with incorrect size
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 09:03 PM Revision 5bfa75c2 (ceph): osd: Drop errors if enough copies are good redundant reads come in
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 09:01 PM Revision 131214dc (ceph): ec: add support for fast read on PGBackend/ECBackend async read
- Extend the PGBackend::objects_read_async interface to support *fast read*,
add the implemenation for ECBackend, in wh... - 08:35 PM Revision 9db81226 (ceph): rgw: lock obj expirer shards when processing
- to prevent multiple rgws processing the same shard concurrently
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 08:10 PM Revision 62f44236 (ceph): Update rados_api.yaml(cherry picked from commit 5b9d41e3ea0f07171ce59f0...
- Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
- 08:09 PM Revision da5d3c28 (ceph): Initial check in to add firefly-hammer-x suite #12641
- Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
(cherry picked from commit 998ba8b58c5c70e8a110c3e1cf29cd24ae9692b9) - 08:05 PM Revision c4a9a4b7 (ceph): rgw: objexp related fixes
- minor issues following code review
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 08:03 PM Revision 850148c2 (ceph): Merge pull request #528 from ceph/wip-12641
- Initial check in to add firefly-hammer-x suite #12641
- 07:58 PM Revision 0a7c4e39 (ceph): Merge pull request #5688 from dzafman/wip-decode-zafman
- osd: Decode use_gmt_hitset with a unique version
Reviewed-by: Sage Weil <sage@redhat.com> - 07:34 PM Revision e734b0a4 (ceph): radosgw-admin: a new command to run objects expirer
- $ radosgw-admin objects expire
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 07:05 PM Revision 3dbea3c3 (ceph): rgw: rename obj expiration hint oids
- Give a more clear name
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 07:03 PM Revision eee424c2 (ceph): rgw: init object expirer thread
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 06:59 PM Revision 46e7f367 (ceph): Merge pull request #5679 from wonzhq/store-test
- test: disable newstore test until it's merged
Reviewed-by: Samuel Just <sjust@redhat.com> - 06:56 PM Linux kernel client Bug #12811: ENXIO from kernel client (cephfs)
- during this time the apama osds were flapping and we were rebalancing lots of data... either of those may be the trig...
- 06:47 PM Linux kernel client Bug #12811 (Resolved): ENXIO from kernel client (cephfs)
- this is on the lab cluster. It has happened twice in 2 days.
In each case, the requests are sent to the second ac... - 06:52 PM Revision 478b14ee (ceph): rgw: verify Swift object lifetime at POST.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 06:51 PM Revision f2f23c2d (ceph): rgw: implement object_is_expired function.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
Conflict... - 06:45 PM Revision 65949bd1 (ceph): rgw: add support for X-Delete-After HTTP header of Swift API.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 06:44 PM Revision aa5f1b8c (ceph): rgw: a few fixes, guard bufferlist decodes
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 06:36 PM Revision 4f9a8431 (ceph): rgw: add basic support for X-Delete-At header of Swift API.
- Fixes: #4099
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 06:32 PM Revision 16a6b5df (ceph): Merge pull request #534 from ceph/wip-12625-master
- Added singleton hammer-x test to address #525
Reviewed-by: Samuel Just <sjust@redhat.com> - 06:25 PM Revision 2bc5a48f (ceph): osd: Decode use_gmt_hitset with a unique version
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 06:17 PM Bug #12810 (Won't Fix): Init script status command ignore cluster value
- The init script status commands don't work properly when the cluster name is not ceph. For example:...
- 06:11 PM Revision 470f9704 (ceph): include/ceph_features: define HAMMER_0_94_4 feature
- This is to constrain upgrades past hammer to version that include
the appropriate compatibility fixes (e.g., hobject_... - 06:11 PM Revision f668c6cc (ceph): mon: use HAMMER_0_94_4 feature to require sufficiently new hammer
- - Do not allow post-hammer OSDs to start if the running hammer is
not a new enough hammer.
- Do not allow pre-hammer ... - 06:11 PM Revision 38465f0d (ceph): osd: refuse to boot if any pre-hammer or old hammer (<v0.94.4) are running
- Old OSDs don't understand us.
This is useful if the osd is upgraded before the mons.
Signed-off-by: Sage Weil <sage... - 06:09 PM Revision f0283890 (ceph): include/ceph_features: define HAMMER_0_94_4 feature
- This is to constrain upgrades past hammer to version that include
the appropriate compatibility fixes (e.g., hobject_... - 06:08 PM Revision 14e02bc9 (ceph): PG::handle_advance_map: on_pool_change after handling the map change
- Otherwise, the is_active() checks in the hitset code can erroneously
return true firing off repops stamped with the n... - 06:05 PM Revision faac0b1f (ceph): rgw: create a worker thread for object expiration
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 05:40 PM Revision cdce7a21 (ceph): rgw: integrate Swift object expiration-related things with CMake.
- Fixes: #4099
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 05:40 PM Revision 05c90e68 (ceph): rgw: split rgw-object-expirer.
- Fixes: #4099
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 05:40 PM Revision 1fa376c2 (ceph): rgw: make object removal atomic in rgw-object-expirer.
- Fixes: #4099
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 05:40 PM Revision 3b0636e9 (ceph): rgw: make the rgw-object-expirer's options more human readable.
- Fixes: #4096
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 05:39 PM Revision f572430b (ceph): rgw: move objexp pool creation into separate function.
- Fixes: #4099
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 05:39 PM Revision db27ea92 (ceph): rgw: add garbage collector daemon for expired objects.
- Fixes: #4099
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 05:39 PM Revision 0d792c9d (ceph): rgw: add support for object expiration in rgw_rados.cc.
- Fixes: #4099
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 05:39 PM Revision 7675aca5 (ceph): cls: add timeindex class for radosgw's objects expiration.
- Fixes: #4099
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 05:39 PM Revision 9f476136 (ceph): rgw: define attribute for storing object expiration info.
- Fixes: #4099
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 05:29 PM Revision 20c76523 (ceph): ceph-object-corpus: remove hammer foo and bar coll_t's
- Signed-off-by: Sage Weil <sage@redhat.com>
- 05:27 PM Revision 4a1cb825 (ceph): test/encoding/readable: handle nondeterministic items in corpus too
- In paritcular, 0.94.2-207-g88e7ee7/objects/ExplicitHashHitSet
Signed-off-by: Sage Weil <sage@redhat.com> - 05:14 PM Revision 60e58b01 (ceph): Merge pull request #5545 from ceph/wip-objclass-filters
- Allow object classes to define PGLS filters
Reviewed-by: David Zafman <dzafman@redhat.com> - 05:14 PM Revision f078a675 (ceph): ceph-object-corpus: add 0.94.2-207-g88e7ee7 hammer objects
- Signed-off-by: Sage Weil <sage@redhat.com>
- 05:12 PM Bug #12536 (Resolved): "FAILED assert(!log.null() || olog.tail == eversion_t())"
- 05:11 PM Bug #12410 (Resolved): OSDMonitor::preprocess_get_osdmap: must send the last map as well
- 05:11 PM Backport #12530 (Resolved): OSDMonitor::preprocess_get_osdmap: must send the last map as well
- merged wip-12536-hammer
- 05:10 PM Backport #12571 (Resolved): "FAILED assert(!log.null() || olog.tail == eversion_t())"
- merged wip-12536-hammer
- 05:07 PM Revision 41a245af (ceph): Merge remote-tracking branch 'gh/wip-12536-hammer' into hammer
- This includes wip-osd-compat-hammer.
- 05:05 PM Revision cbe101e7 (ceph): Merge remote-tracking branch 'gh/wip-osd-compat'
- Reviewed-by: Samuel Just <sjust@redhat.com>
- 04:55 PM CephFS Bug #12777 (Fix Under Review): qa: leftover files in cephtest directory
Oops, this is CephFSTestCase.tearDown not getting called (so there's still a client mount, so the dir can't be remo...- 01:34 PM CephFS Bug #12777: qa: leftover files in cephtest directory
- This shows up in the logs as 'rmdir -- /home/ubuntu/cephtest'
- 04:53 PM Revision df8cecf4 (ceph): tasks/cephfs: fix test_pool_perm teardown
- Fixes: #12777
Signed-off-by: John Spray <john.spray@redhat.com> - 04:41 PM Revision 68d1d8ee (ceph): Merge pull request #5683 from tchaikov/wip-hitset-gmt-yet-another-featu...
- osd: do not let OSD_HITSET_GMT reuse the feature bit
Reviewed-by: Sage Weil <sage@redhat.com> - 04:39 PM Revision 5a4f6a86 (ceph): osd: do not let OSD_HITSET_GMT reuse the feature bit
- * to ease the backport to hammer
Signed-off-by: Kefu Chai <kchai@redhat.com> - 04:32 PM Revision d43c10e0 (ceph): memstore: use thread_local page vector
- avoid allocating page_vector storage for every read/write request. this
improves the sequential write bandwidth in os... - 04:31 PM Revision 4e6548ca (ceph): memstore: PageSetObject for MemStore integration
- adds config variable memstore_page_set=true to switch between the
PageSet and bufferlist implementations for object d... - 04:29 PM Bug #12681: tcmalloc doesn't release memory back to system
- Hi, Greg
I'm using RHEL-6 with kernel 2.6.32, the tcmalloc is 4.1.0 with gperftools-2.0. I use the same OS and tcm... - 04:20 PM CephFS Bug #12806: nfs restart failures
- iirc starting the nfs service on el7 depends on rpcbind, nfs-lock already being up, so that might be what's missing here
- 01:19 PM CephFS Bug #12806 (Resolved): nfs restart failures
- ...
- 03:55 PM Bug #12638 (Pending Backport): build_incremental() could take 40678 ms to finish
- 03:38 PM Bug #12755 (New): selinux: ceph-mon denial on wip-user
- Sage, while the post-script got improved, the bahaviour might still not be fixed, here. It is likely that the file is...
- 12:39 AM Bug #12755 (Resolved): selinux: ceph-mon denial on wip-user
- 03:21 PM Revision f1e1ada7 (ceph): Merge pull request #5524 from tchaikov/wip-monstore-cache
- mon: add a cache layer over MonitorDBStore
Reviewed-by: Joao Eduardo Luis <joao@suse.de>
Reviewed-by: Sage Weil <sag... - 03:21 PM Revision f585d0e2 (ceph): Merge pull request #544 from ceph/wip-ms-type-async
- rados: test both simple, async, and a mix
Reviewed-by: Samuel Just <sjust@redhat.com> - 02:38 PM Revision 671a9dfa (ceph): rados: test both simple, async, and a mix
- - in general, test simple vs async vs random
- not for msgr-less workloads
- not for thrash-erasure-*.. the regular t... - 02:11 PM Messengers Bug #12745 (Resolved): ms/simple: Messenger/MessengerTest.SyntheticInjectTest2/1 hang
- 02:10 PM Revision cba062e4 (ceph): Merge pull request #5669 from yuyuyu101/wip-12745
- ms/simple: Messenger/MessengerTest.SyntheticInjectTest2/1 hang
Reviewed-by: Sage Weil <sage@redhat.com> - 02:06 PM Revision 3eb36fc7 (ceph): doc/release-notes: v0.94.3
- Fix up the release timeline link for v9.0.3 too.
Signed-off-by: Sage Weil <sage@redhat.com> - 01:46 PM CephFS Bug #12808 (New): smbtorture failure on scan-pipe
- Log summary line: "Command failed on burnupi59 with status 137: 'TESTDIR=/home/ubuntu/cephtest bash -s'"...
- 01:35 PM CephFS Bug #12807 (Duplicate): rmdir cephtest failing
- #12777
- 01:30 PM CephFS Bug #12807 (Duplicate): rmdir cephtest failing
- http://pulpito.ceph.com/teuthology-2015-08-22_23:04:02-fs-next---basic-multi/1027445/
http://pulpito.ceph.com/teutho... - 01:13 PM Revision e749b214 (ceph): objclass: enable unregistering filter factory
- Signed-off-by: John Spray <john.spray@redhat.com>
- 01:13 PM Revision 60d51fc4 (ceph): cls: add a filter to the hello class for testing
- Signed-off-by: John Spray <john.spray@redhat.com>
- 01:13 PM Revision 2777438b (ceph): test: add a test for filter in cls hello
- Signed-off-by: John Spray <john.spray@redhat.com>
- 01:13 PM Revision f5df1e48 (ceph): osd: separate filter init from construction
- ...so that implementations can readily handle
decode errors in client args and return an
error code.
Signed-off-by: ... - 01:05 PM Messengers Bug #12805 (Can't reproduce): msg/simple/SimpleMessenger.cc: 238: FAILED assert(!cleared)
- ceph version 9.0.2-1556-g5a20f28 (5a20f2804f92e82ec9ee19e08135287322f5e961)
http://pulpito.ceph.com/teuthology-201... - 12:15 PM Bug #12803 (Fix Under Review): osd: op_commit on cancelled op
- https://github.com/ceph/ceph/pull/5680
- 08:43 AM Bug #12803 (Duplicate): osd: op_commit on cancelled op
- http://qa-proxy.ceph.com/teuthology/yuan-2015-08-20_00:19:52-rados-wip-intel-testing---basic-multi/1024108/...
- 09:32 AM Revision e2c3c863 (ceph): rgw: add key parameter conflict check for radosgw-admin command line.
- 1.--access-key & --gen-access-key
radosgw-admin key create --uid=aaa --access-key='111' --gen-access-key --gen-secret... - 08:53 AM Revision 328d30cb (ceph): AsyncConnection: Fix uninitialized variable compile warning
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 08:53 AM Revision 24f4d22d (ceph): TestMsgr: Fix forever hang under lossless policy and one is WAIT anothe...
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 08:40 AM Revision e05e35b7 (ceph): Merge pull request #5678 from ceph/wip-12208
- rgw: delete finisher only after finalizing watches
- 08:39 AM Revision edbb22f0 (ceph): Merge pull request #5677 from ceph/wip-11455
- rgw: init some manifest fields when handling explicit objs
- 08:38 AM Revision 7515c047 (ceph): Merge pull request #5651 from theanalyst/rgw/fixiso8601
- rgw: be more flexible with iso8601 timestamps
- 08:32 AM Bug #12802 (Resolved): support of non-block connect in async messenger
- Async messenger should avoid all potential blocking point, but we use a block connect now.
So we must add this feature. - 08:18 AM Feature #12801 (Resolved): Make "ceph osd metadata" command support dumping all osds info
- Now "ceph osd metadata" only support single osd metadata dump, it would be silly if devops want to get the all osds m...
- 07:18 AM Bug #12100: OSD crash, uexpected aio error in FileJournal.cc
- Not close to this date no.
In June we deployed the cluster using ceph-deploy deploying dumpling. during june we di... - 04:27 AM Revision 10fd1a56 (ceph): test: don't unmount when no store is created
- When store is not created in SetUp for some reasons, unmounting it would
lead to segment fault.
Signed-off-by: Zhiqi... - 03:17 AM Revision 83ba5978 (ceph): test: handle the case when ObjectStore::create returns NULL
- When the objectstore type doesn't exist, NULL is returned in SetUp.
Handle the NULL return code to avoid a segment fa... - 03:08 AM Revision 6a735d62 (ceph): test: disable newstore test until it's merged
- Newstore hasn't been merged. It leads to a segment fault in one of the
teuthology testing job since ObjectStore::crea... - 01:17 AM devops Bug #11675 (Can't reproduce): "error: Can't download https://pypi.python.org..." in ceph-deploy-h...
- 01:16 AM devops Bug #11662 (Can't reproduce): "error: Can't download https://pypi.python.org..." in ceph-deploy-h...
- 01:15 AM devops Bug #11384 (Rejected): "Failed to fetch package version" in smoke-master-distro-basic-multi run
- 01:15 AM devops Bug #10532 (Won't Fix): Failed "joining thrashosds" - suspected low memory config on vps
- 01:13 AM devops Tasks #6932 (Rejected): Create a tar.gz file with packages per-distro
- 01:12 AM devops Bug #9499 (Can't reproduce): osds do not start after reboot (centos7, dm-crypt)
- 01:12 AM devops Bug #11581 (Closed): "port 22: No route to host" errors in the labs
- closing this... i don't see anything useful here!
- 01:10 AM devops Bug #12300 (Rejected): build ...dumpling/noarch/ceph-release-1-0.el7.noarch.rpm
- it was trying to pull from gitbuilder.ceph.redhat.com .. but you're running it on vps, which has no access to the red...
- 01:06 AM rgw Bug #12407 (Need More Info): With root as default user, unable to have multiple RGW instances run...
- 01:06 AM rgw Bug #12407: With root as default user, unable to have multiple RGW instances running
- Pavan, can you also let me know what version of debian you are using? Thanks!
- 01:05 AM rgw Bug #12407: With root as default user, unable to have multiple RGW instances running
- ...
- 12:41 AM Revision b224eb45 (ceph): Merge pull request #4456 from ceph/wip-user
- run ceph daemons as user and group ceph
- 12:38 AM Revision c825fd76 (ceph): Merge pull request #5674 from ceph/wip-selinux-post-script
- ceph.spec.in: Restart services only if they are running
Reviewed-by: Milan Broz <mbroz@redhat.com> - 12:35 AM Revision b89d7521 (ceph): global_init: ignore --set{user,group} if not root
- Assume these options can only drop privileges; normal users can't setuid
root (and even if they could we wouldn't wan... - 12:35 AM Revision 7cd07490 (ceph): PendingReleaseNotes: more notes about the 'ceph' user
- Signed-off-by: Sage Weil <sage@redhat.com>
- 12:35 AM Revision 18e0c777 (ceph): debian: rename ceph-common.postinst
- There was a trailing space!
Signed-off-by: Sage Weil <sage@redhat.com> - 12:35 AM Revision 7c960168 (ceph): debian/ceph-common.postinst: fix /var/log/ceph permissions
- Signed-off-by: Sage Weil <sage@redhat.com>
- 12:35 AM Revision 94da8c1e (ceph): debian/ceph-common.postinst: fix adduser, addgroup
- Drop the extra groups thing--it's broken anyway.
Signed-off-by: Sage Weil <sage@redhat.com> - 12:35 AM Revision c8605539 (ceph): debian/ceph-common.dirs: install /var/lib/ceph
- It is the ceph user's home dir and we need to set the ownership.
Signed-off-by: Sage Weil <sage@redhat.com> - 12:35 AM Revision 48f98e1d (ceph): upstart: setuser ceph
- Signed-off-by: Sage Weil <sage@redhat.com>
- 12:35 AM Revision 743b15b4 (ceph): debian: /var/run/ceph should be owned by ceph:ceph
- Fix upstart and sysvinit scripts to create /var/run/ceph properly.
Chown existing dir on upgrade.
Signed-off-by: Sa... - 12:35 AM Revision 960139ed (ceph): PendingReleaseNotes: some notes about upgrade and ceph user
- Signed-off-by: Sage Weil <sage@redhat.com>
- 12:35 AM Revision aef00eb8 (ceph): ceph-disk: fix get_ceph_user
- getpwnam throws an exception
Signed-off-by: Sage Weil <sage@redhat.com> - 12:34 AM Revision 4dfe0a8a (ceph): global: add --setuser and --setgroup options
- These are done after reading config files/environment and before log files
are opened. Allow a name or id to be spec... - 12:34 AM Revision 7c9fdf44 (ceph): systemd: make ceph-osd setuid/gid to ceph:ceph
- Signed-off-by: Sage Weil <sage@redhat.com>
- 12:34 AM Revision c7ee798a (ceph): set nofile ulimit in /etc/security/limits.d/ceph only
- Specify the nofile ulimit in one standard place, where everyone expects it
to be. Drop it from the ceph-osd unit fil... - 12:34 AM Revision ed0cd42d (ceph): ceph.spec.in: add ceph user/group
- Add support of create ceph user/group.
TODO: Add correct number for SUSE.
Signed-off-by: Danny Al-Gaaf <danny.al-ga... - 12:34 AM Revision 596c9b68 (ceph): ceph.spec.in: install ceph.limits.d
- limits.d expect the file to end with *.conf. Add the
installed file to package list of the ceph package.
Signed-off-... - 12:34 AM Revision 2d4f3a91 (ceph): ceph.spec.in: Fixup uid/gid setting
- We need to set the variables in the %pre phase, otherwise they are not properly
evaluated. Also use /sbin/nolongin in... - 12:34 AM Revision d9df52b4 (ceph): ceph.spec.in: Fix ceph.limits.d path
- Signed-off-by: Boris Ranto <branto@redhat.com>
- 12:34 AM Revision e95904f7 (ceph): ceph.spec.in: /var/lib/ceph is owned by ceph package
- Signed-off-by: Boris Ranto <branto@redhat.com>
- 12:34 AM Revision ceb93e8e (ceph): ceph.spec.in: User and group must be created in ceph-common pre-install...
- The package ceph-common uses ceph user and group but they are both created in
ceph package %pre phase. We need to mov... - 12:34 AM Revision 25f68aee (ceph): init-ceph.in: Set ceph user and group when running the daemons
- Signed-off-by: Boris Ranto <branto@redhat.com>
- 12:34 AM Revision bbedc8e5 (ceph): ceph-osd-prestart.sh: ensure data dir is root or ceph before start
- Signed-off-by: Sage Weil <sage@redhat.com>
- 12:34 AM Revision 8bd35bd6 (ceph): Set Ceph device partitions owner to ceph user in udev.
- Signed-off-by: Milan Broz <mbroz@redhat.com>
- 12:34 AM Revision 52e978e4 (ceph): Set keys owner to ceph user if exists.
- Also fix directory access rigths.
Signed-off-by: Milan Broz <mbroz@redhat.com> - 12:34 AM Revision 09db67fd (ceph): ceph-disk: set owner of created files to ceph
- Signed-off-by: Milan Broz <mbroz@redhat.com>
- 12:34 AM Revision 28fdac32 (ceph): global: implement setuser_match_path
- Allow the --setuser and --setgroup to be conditional on the
specified user/group matching the ownership of a given pa... - 12:34 AM Revision 8f3185ba (ceph): systemd: use --setuser and --setgroup for all daemons
- Allow all daemons drop privilege themselves, instead of letting
systemd do it.
Among other things, this means that a... - 12:34 AM Revision 71a0a029 (ceph): debian: create ceph user and group
- Use Debian base-passwd allocated UID/GID pair.
Signed-off-by: Sage Weil <sage@redhat.com> - 12:34 AM Revision 2ba3d61d (ceph): debian: chown ceph:ceph /var/llib/ceph
- Do not do it recursively--there may already be huge amounts of data
here.
Signed-off-by: Sage Weil <sage@redhat.com> - 12:34 AM Revision 75226509 (ceph): debian: chown -R ceph:ceph /var/log/ceph
- The number of log files is generally bounded; safe to chown these.
Allow ceph group members to write to this dir.
S... - 12:34 AM Revision 3c569382 (ceph): ceph.spec: chown and chmod /var/lib/ceph and /var/log/ceph
- Signed-off-by: Sage Weil <sage@redhat.com>
- 12:34 AM Revision ec1ee5e9 (ceph): systemd: run mon and mds as ceph:ceph
- Signed-off-by: Sage Weil <sage@redhat.com>
- 12:34 AM Revision b8893f6b (ceph): systemd: chown ceph:ceph /var/run/ceph
- Signed-off-by: Sage Weil <sage@redhat.com>
- 12:34 AM Revision 6532e1c4 (ceph): debian: fix /var/lib/ceph/* directory ownership
- These dirs are owned by the package; make sure they are owend by the ceph
user.
Signed-off-by: Sage Weil <sage@redha... - 12:01 AM rgw Bug #12434: "Error in `radosgw': free()" in upgrade:firefly-firefly-distro-basic-vps
- Depending on how often are see this issue. Base versions will always have bugs in them, otherwise we wouldn't have ne...
08/26/2015
- 11:43 PM rgw Bug #12434: "Error in `radosgw': free()" in upgrade:firefly-firefly-distro-basic-vps
- The purpose of this test do from v0.80.4 -> firefly upgrade.
Do we need to remove v0.80.4 from the configurations?
... - 10:42 PM rgw Bug #12434 (Duplicate): "Error in `radosgw': free()" in upgrade:firefly-firefly-distro-basic-vps
- Duplicate (8846). The issue has long been resolved, but this upgrade test runs the broken version (0.80.4).
- 11:16 PM Bug #12100: OSD crash, uexpected aio error in FileJournal.cc
- Was there any change or update not only for ceph?
Pontus Lindgren wrote:
> I am also experiencing a similar issue... - 11:00 AM Bug #12100: OSD crash, uexpected aio error in FileJournal.cc
I am also experiencing a similar issue:
2015-08-16 08:11:54.227567 7f13d68de700 0 log_channel(cluster) log [WR...- 10:08 PM Revision 97aed59f (ceph): rgw: delete finisher only after finalizing watches
- Fixes: #12208
The watch error path might try to schedule a finisher work, delete finisher
only after watch destructi... - 09:49 PM Revision 01a9a792 (ceph): osbench: add multithreaded objectstore benchmark
- usage: ceph_objectstore_bench [flags]
--size
total size in bytes
--block-size
... - 09:49 PM Revision 26f716e9 (ceph): memstore: use intrusive_ptr instead of shared_ptr
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 09:49 PM Revision 5d8307a1 (ceph): memstore: add Object interface to hide bufferlist
- this prepares MemStore for a new object data implementation that
replaces bufferlist
Signed-off-by: Casey Bodley <cb... - 09:49 PM Revision 54739a59 (ceph): memstore: protect object omap with a mutex
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 09:49 PM Revision 61cd2da2 (ceph): memstore: protect object xattrs with a mutex
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 09:49 PM Revision 46f92f0d (ceph): memstore: BufferlistObject uses spinlock for data
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 09:49 PM Revision 79454822 (ceph): memstore: move collection lock into get_object
- only hold the collection lock while accessing the hash/object maps,
don't use it to serialize entire operations
Sign... - 09:49 PM Revision b0882fb6 (ceph): memstore: replace apply_lock with sequencer
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 09:49 PM Revision 51d2553f (ceph): memstore: add PageSet for MemStore object data
- introduce class PageSet as an alternative to bufferlist for storing
object data
PageSet uses an avl set to manage it... - 09:49 PM Revision dd7fe61e (ceph): memstore: add unit test for PageSet
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 09:38 PM Revision d7bf8cb5 (ceph): rgw: init some manifest fields when handling explicit objs
- Fixes: #11455
When dealing with old manifest that has explicit objs, we also
need to set the head size and head objec... - 09:36 PM rgw Bug #11455 (Fix Under Review): COPYing an old object onto itself produces a truncated object
- 08:37 PM Bug #12738 (In Progress): scrub bogus results when missing a clone
- 07:20 PM rbd Feature #12112 (Fix Under Review): rbd: add --object-size option
- https://github.com/ceph/ceph/pull/6830
- 07:16 PM Bug #12787: ceph-disk does not activate with systemd on some CentOS 7
- It worked once, I'm not sure why but that suggests a race condition somewhere ?
- 12:41 AM Bug #12787 (Resolved): ceph-disk does not activate with systemd on some CentOS 7
- Steps to reproduce on CentOS 7
* ceph-disk prepare /dev/vdb
The ceph-osd is not running and mounting /dev/vdb1 ... - 07:01 PM Bug #12798 (Resolved): ceph_test_objectstore: keyvaluestore fails tests
- disabled for now; reenable in test/objectstore/store_test.cc and then fix any issues!
- 07:01 PM Bug #12788 (Resolved): os/newstore/NewStore.cc: 2988: FAILED assert(!c)
- freaking rmkey bug in RocksDBStore.cc
- 01:41 AM Bug #12788 (Resolved): os/newstore/NewStore.cc: 2988: FAILED assert(!c)
- ...
- 06:58 PM rgw Bug #12724 (Resolved): rgw: wrong error handling in RGWGetObj_ObjStore_SWIFT
- 06:56 PM rgw Bug #12750 (Pending Backport): rgw: wrong removal during iteration over Swift metadata in filter_...
- 06:44 PM rgw Bug #12767 (Duplicate): common/Mutex.cc: 100: FAILED assert(r == 0) from RGWWatcher::handle_error...
- Duplicate of 12208
- 06:40 PM rgw Feature #12666: rgw: expose the number of *stuck threads* via admin socket
- https://github.com/ceph/ceph/pull/5557
- 06:21 PM rgw Bug #12370: test_s3.test_object_copy_canned_acl ... FAIL
- https://github.com/ceph/ceph/pull/5575
- 06:18 PM rgw Bug #12370 (Pending Backport): test_s3.test_object_copy_canned_acl ... FAIL
- 06:18 PM rgw Bug #12759 (Duplicate): s3tests.functional.test_s3.test_object_copy_canned_acl failure
- Duplicate of issue #12370
- 06:11 PM Revision ef053614 (ceph): Merge pull request #5625 from liewegas/wip-memcpy
- buffer: put inlined memcpy into a header; x86_64 only
Reviewed-by: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
Reviewed... - 05:39 PM Revision 95cefea9 (ceph): 0.94.3
- 04:40 PM Backport #12395: register_new_pgs() should check ruleno instead of its index
- since the assignee of https://github.com/ceph/ceph/pull/5404 is smithfarm.
- 04:22 PM Bug #9732 (Pending Backport): ReplicatedPG::hit_set_trim osd/ReplicatedPG.cc: 11006: FAILED asser...
- 04:21 PM Tasks #12797: create the upgrade test suite for gmt and sortbitmap change
- per sage
> you need to use suite_branch wip-12536-hammer-x-test, which installs wip-12536-hammer instead of hammer... - 04:19 PM Tasks #12797 (In Progress): create the upgrade test suite for gmt and sortbitmap change
- for more context, see http://tracker.ceph.com/issues/9732#note-16
- 04:15 PM Revision 31143c72 (ceph): Merge pull request #5585 from tchaikov/wip-4941
- mon,osd: use GMT time for the object name of hitsets
Reviewed-by: Sage Weil <sage@redhat.com> - 03:17 PM devops Bug #12796 (Rejected): missing libcephfs-jni1-devel package?...
- http://pulpito.ceph.redhat.com/teuthology-2015-08-22_18:12:02-hadoop-hammer---basic-magna/179587/...
- 02:55 PM devops Bug #12791 (Resolved): ceph.spec.in does SUSE-specific sysvinit stuff on unsupported SUSE systems
- 01:47 PM devops Bug #12791 (In Progress): ceph.spec.in does SUSE-specific sysvinit stuff on unsupported SUSE systems
- oops, missed a bit: https://github.com/ceph/ceph/pull/5675
- 01:33 PM devops Bug #12791 (Resolved): ceph.spec.in does SUSE-specific sysvinit stuff on unsupported SUSE systems
- Oops, no backport to hammer, because no systemd in hammer.
- 01:21 PM devops Bug #12791 (Pending Backport): ceph.spec.in does SUSE-specific sysvinit stuff on unsupported SUSE...
- 01:21 PM devops Bug #12791: ceph.spec.in does SUSE-specific sysvinit stuff on unsupported SUSE systems
- Master PR: https://github.com/ceph/ceph/pull/5661 (merged)
- 01:19 PM devops Bug #12791 (Resolved): ceph.spec.in does SUSE-specific sysvinit stuff on unsupported SUSE systems
- As far as the Ceph server daemons (osd, mon, rgw) are concerned, all sysvinit-based versions of SLES and openSUSE are...
- 02:52 PM Revision dda1016d (ceph): Merge pull request #5671 from ceph/wip-msgr-fix-random
- Messenger: Fix rand() generate the same sequence numbers
Reviewed-by: Kefu Chai <kchai@redhat.com> - 02:50 PM Revision 3c2d3c93 (ceph): Merge pull request #5675 from ceph/wip-12791
- ceph.spec.in: remove obsolete SUSE-specific code
- 02:34 PM Revision b610588b (ceph): ceph.spec.in: remove obsolete SUSE-specific code
- http://tracker.ceph.com/issues/12791 Fixes: #12791
Signed-off-by: Nathan Cutler <ncutler@suse.com> - 01:35 PM devops Backport #12792 (Rejected): ceph.spec.in does SUSE-specific sysvinit stuff on unsupported SUSE sy...
- for a second I thought we could backport https://github.com/ceph/ceph/pull/5661 to hammer, but there is no systemd in...
- 01:20 PM devops Backport #12792 (Rejected): ceph.spec.in does SUSE-specific sysvinit stuff on unsupported SUSE sy...
- 11:58 AM Bug #12755: selinux: ceph-mon denial on wip-user
- The PR for the safer process restart procedure:
https://github.com/ceph/ceph/pull/5674 - 10:25 AM Bug #12755: selinux: ceph-mon denial on wip-user
- We need to stop/start the service in order to get its SELinux context fixed when installing the policy.
btw: There... - 11:56 AM Revision df21a6e2 (ceph): osd: expose PGLSFilter in objclass interface
- Signed-off-by: John Spray <john.spray@redhat.com>
- 11:54 AM Revision c318129b (ceph): ceph.spec.in: Restart services only if they are running
- Signed-off-by: Boris Ranto <branto@redhat.com>
- 11:01 AM rgw Feature #12790: rgw: multipart uploads copy part support
- PR available here:
-https://github.com/ceph/ceph/pull/5673- (closed) - 10:36 AM rgw Feature #12790 (Resolved): rgw: multipart uploads copy part support
- Implement multipart uploads copy part support
- 09:56 AM Revision 2280ca52 (ceph): fix rebug rados
- Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
- 08:37 AM Revision 83b93eb7 (ceph): Merge pull request #5639 from rzarzynski/wip-12750
- rgw: fix removal during iteration over Swift metadata in filter_out_temp_url()
- 04:21 AM Revision 55cec07b (ceph): Messenger: Fix rand() generate the same sequence numbers
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 03:41 AM Bug #12789: Long tail latency due to journal aio io_submit takes long time to return?
- https://github.com/ceph/ceph/pull/5670
- 03:40 AM Bug #12789 (Resolved): Long tail latency due to journal aio io_submit takes long time to return?
- There are some (not much) long tail latency due to journal aio io_submit takes long time to return. We add some logs ...
- 03:34 AM Revision ef1434a6 (ceph): Narrow journal aio_lock locking scope in write_aio_bl
- Aio io_submit in write_aio_bl with aio_lock might take long time to return, so write_finish_thread_entry can't proces...
- 02:33 AM Revision 2eac0ff9 (ceph): Merge pull request #540 from ceph/wip-hobj-sort
- suites/rados/thrash: randomize hobject sort order
Reviewed-by: Samuel Just <sjust@redhat.com> - 01:54 AM Revision 2d2f0eb3 (ceph): test: add test case for insert empty ptr when buffer rebuild
- Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
- 01:54 AM Revision 15e5ebe1 (ceph): common: fix code format
- Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
- 01:54 AM Revision fb1b6ddd (ceph): common: fix insert empty ptr when bufferlist rebuild
- Fixes: #12775
Signed-off-by: Xinze Chi <xmdxcxz@gmail.com> - 01:48 AM Revision 23ebeb36 (ceph): Merge pull request #5629 from kylinstorage/wip-nocache_skip_promote-v3
- Avoid cache pollution by rbd export etc v2
Reviewed-by: Sage Weil <sage@redhat.com> - 01:47 AM Bug #12747 (Resolved): osd/PGBackend.cc: 294: FAILED assert(ec_impl) - wrong path on rpm distros
- 01:47 AM Revision 8c08b6b0 (ceph): Merge pull request #5636 from liewegas/wip-12747
- make EC plugin path static
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 01:47 AM Bug #12766 (Resolved): osd: segv in osd bench
- 01:47 AM Revision 4bd73150 (ceph): Merge pull request #5646 from liewegas/wip-12766
- osd: wait for cleanup from bench
- 01:46 AM Bug #12736 (Resolved): boost uuid makes valgrind complain
- 01:46 AM Revision f889151e (ceph): Merge pull request #5650 from liewegas/wip-12736
- uuid: use boost::random:random_device
Reviewed-by: Kefu Chai <kchai@redhat.com> - 01:33 AM Revision ccc47d32 (ceph): suites/rados/thrash: randomize hobject sort order
- Signed-off-by: Sage Weil <sage@redhat.com>
08/25/2015
- 11:13 PM devops Feature #11881: ceph-disk support for multipath
- Partitioning via multipath on Ubuntu 14.04 fails in a strange way, reported bug at https://bugs.launchpad.net/ubuntu/...
- 10:29 AM devops Feature #11881: ceph-disk support for multipath
- > We ended up adding a rule to call a script on mpath add which would read the GPT partition and create links for the...
- 09:58 AM devops Feature #11881: ceph-disk support for multipath
- Actually we were unable to create the parttype uuid links using the udev rules.
The part-uuid and type-uuid do not ... - 10:22 PM Bug #9221: LibRadosTwoPoolsPP.PromoteOn2ndRead
- https://github.com/ceph/ceph/pull/5667
- 09:55 PM Bug #9221 (In Progress): LibRadosTwoPoolsPP.PromoteOn2ndRead
- 10:20 PM Revision 347ac0f8 (ceph): ceph_test_rados_api_tier: make PromoteOn2ndRead tolerate thrashing
- Repeate the test up to 20 times until we get a read that doesn't
trigger promote.
Fixes: #9221 (again)
Signed-off-by... - 09:53 PM Bug #12775: common: do not insert emtpy ptr when rebuild emtpy bufferlist
- 02:22 PM Bug #12775 (Fix Under Review): common: do not insert emtpy ptr when rebuild emtpy bufferlist
- 06:43 AM Bug #12775: common: do not insert emtpy ptr when rebuild emtpy bufferlist
- https://github.com/ceph/ceph/pull/5656 is waitting for review.
- 06:01 AM Bug #12775 (Resolved): common: do not insert emtpy ptr when rebuild emtpy bufferlist
- when bufferlist len is 0, rebuild would new a prt with len 0 and insert into bufferlist.
The follow test case woul... - 09:53 PM Bug #12755 (In Progress): selinux: ceph-mon denial on wip-user
- 06:39 PM Bug #12755: selinux: ceph-mon denial on wip-user
- oh, it's in ceph.spec:...
- 06:36 PM Bug #12755: selinux: ceph-mon denial on wip-user
- ...
- 12:03 PM Bug #12755: selinux: ceph-mon denial on wip-user
- SELinux treats root as a regular user by default. This means that SELinux will deny access to a file owned by a rando...
- 06:09 AM Bug #12755: selinux: ceph-mon denial on wip-user
- Anyway, short info how to interpret it:
- install policycore-utils
- we can list "ceph_t" AVC rom autidt usin... - 06:00 AM Bug #12755: selinux: ceph-mon denial on wip-user
- Probably it just need some new rule in selinux policy. How it can be reproduced, is is part of a regular tests run?
- 09:44 PM Bug #12785: shard 4 missing -2/00000014/temp_recovering_0.14_25'135_118_head/head
- off by one error in hobject_t::is_temp() breaks the temp cleanup:...
- 09:18 PM Bug #12785 (Resolved): shard 4 missing -2/00000014/temp_recovering_0.14_25'135_118_head/head
- http://pulpito.ceph.com/teuthology-2015-08-22_23:06:01-krbd-master-testing-basic-multi/1027470/...
- 09:43 PM Revision 8a08acce (ceph): common/hobject_t: fix is_temp() off-by-one
- pool 0 -> -2 for its temp objects.
Fixes: #12785
Signed-off-by: Sage Weil <sage@redhat.com> - 09:42 PM Revision 7cc8d86c (ceph): ceph_test_msgr: parse CEPH_ARGS
- Signed-off-by: Sage Weil <sage@redhat.com>
- 08:40 PM Revision dfd142f2 (ceph): include/inline_memcpy: use __builtin_memcpy instead of explicit ptr copies
- The gcc version of this is smart enough to handle any alignment
issues that exist and *should* generate the fastest c... - 08:40 PM Revision 98c0606c (ceph): include/inline_memcpy: make prototype resemble memcpy's
- Suggested-by: Steve Capper <steve.capper@linaro.org>
Signed-off-by: Sage Weil <sage@redhat.com> - 08:00 PM Documentation #12784 (Resolved): Update the Latest Release Information in "os-recommendations.rst"
- Update http://docs.ceph.com/docs/master/start/os-recommendations page with the "Giant" and "Hammer" information under...
- 07:19 PM Messengers Bug #12745: ms/simple: Messenger/MessengerTest.SyntheticInjectTest2/1 hang
- ...
- 07:18 PM Messengers Bug #12745: ms/simple: Messenger/MessengerTest.SyntheticInjectTest2/1 hang
- ...
- 07:16 PM Messengers Bug #12745: ms/simple: Messenger/MessengerTest.SyntheticInjectTest2/1 hang
- I think the test is buggy. It randomly does drop_connection(), which will mark_down() one end of a stateful lossless...
- 02:29 AM Messengers Bug #12745: ms/simple: Messenger/MessengerTest.SyntheticInjectTest2/1 hang
- Hmm, it seemed like a rare case, I can't reproduce this in local env.
- 06:59 PM rbd Bug #12765 (Pending Backport): Failure in test_librbd_python.sh
- 06:40 PM Bug #12614 (Pending Backport): recursive lock of md_config_t (0)
- 06:39 PM Revision 5d58dd2e (ceph): Merge pull request #5654 from ceph/wip-12614
- config: skip lockdep for intentionally recursive md_config_t lock
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-... - 06:31 PM Bug #12782: Straw2 bucket created even when "tunable straw_calc_version 1" is created
- Apparently "allowed_bucket_algs" is the one which dictates that buckets should be STRAW2 and not "straw_calc_version"...
- 06:09 PM Bug #12782 (Closed): Straw2 bucket created even when "tunable straw_calc_version 1" is created
- - Hammer cluster running on kernel 3.19
- Tunables set to "hammer" via "ceph osd crush tunables hammer"
- Tunables ... - 06:16 PM Revision cca0d587 (ceph): Merge pull request #5647 from ceph/wip-12765
- librbd: error closing image while set to invalid snapshot
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 05:56 PM Documentation #12771 (Closed): Update the Latest Release Information in "get-packages.rst"
- I'll open a new tracker issue as I find more documents where the release information needs updated.
- 03:18 PM Documentation #12771 (In Progress): Update the Latest Release Information in "get-packages.rst"
- 03:16 PM Documentation #12771: Update the Latest Release Information in "get-packages.rst"
- I updated the "get-packages.rst" file with the "Hammer" release. Still going through the rest of the documentation c...
- 05:45 PM devops Bug #12768: Error: Package: 1:ceph-0.87.2-8.g6c33a35.el7.x86_64 (Ceph)
- Tamil, as we discussed we need to coordinate what to do with rhel specific tests, like in this example.
Questions:
... - 05:27 PM Revision 5b9d41e3 (ceph): Update rados_api.yaml
- 05:21 PM Bug #12781 (Resolved): ceph-disk list fails on dmcrypt devices
- ceph-disk list fails on dmcrypt devices...
- 05:11 PM Revision 998ba8b5 (ceph): Initial check in to add firefly-hammer-x suite #12641
- Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
- 05:01 PM Revision fc02a8ad (ceph): added boost timegm impl for cross platform support
- Signed-off-by: Rohan Mars <code@rohanmars.com>
- 04:08 PM Revision 6b4b85a5 (ceph): Merge pull request #4677 from tchaikov/wip-fix-type-error-in-crush-rule
- pybind: Avoid type error in crush rule (str, not an int)
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 03:54 PM Revision 4628c517 (ceph): Merge pull request #5659 from badone/doc_writeback_typo
- doc: Fix typo in writeback throttle section
- 03:54 PM Revision dfb9fea3 (ceph): Merge pull request #5660 from badone/doc_snaps_typo
- doc: Fix typo and redundant word in snaps section
- 03:43 PM rbd Bug #12017 (Can't reproduce): ERROR: test_rbd.TestImage.test_update_features
- If you have more info on how to produce this issue please reopen.
- 03:39 PM Linux kernel client Bug #12691: ffsb osd thrash test - osd/ReplicatedPG.cc: 2348: FAILED assert(0 == "out of order op")
- Can you post the yaml config for this test? We haven't seen it in the usual nightlies, and it'd be good to add it the...
- 03:27 PM Bug #10399 (Pending Backport): kvm die with assert(m_seed < old_pg_num)
- 03:23 PM Revision 6886d453 (ceph): Merge pull request #5648 from ceph/wip-10399
- Objecter: pg_interval_t::is_new_interval needs pgid from previous pool
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 03:14 PM Revision a5aa4150 (ceph): Merge pull request #5663 from ritz303/wip-doc-12771
- doc: Added "Hammer" in the list of major releases.
- 03:12 PM Revision da6d5cfc (ceph): osd: bug fix hit_set_map size for tier pool
- Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
- 03:03 PM Bug #11511 (Resolved): FAILED assert(cursor.data_offset <= oi.size)
- 79961bc3386e4a926e628552586020ff17d9d086
- 03:02 PM Revision 582f0f64 (ceph): doc: Added "Hammer" in the list of major releases.
- Fixes: #12771
Signed-off-by: ritz303 <ritz_303@yahoo.com> - 03:00 PM Bug #10487 (Can't reproduce): logclient: setting clog_to_monitor=false on live osd crashes
- i wasn't able to reproduce this either. :(
- 02:59 PM Bug #12223 (Pending Backport): WBThrottle::clear_object: signal on cond when we reduce throttle v...
- 02:59 PM Revision 855ae1fd (ceph): bug fix: osd: avoid multi set osd_op.outdata in tier pool
- There are two read op on the same object for ec pool. First op read
miss happend, calling do_proxy_read and promote_o... - 02:58 PM Bug #12263 (Resolved): ReplicatedPG::finish_promote: do not pre-fill clones vector with 0s
- 02:57 PM Bug #8193 (Resolved): HitSetTrim test in test/librados/tier.cc needs to be skipped if thrasher ru...
- 02:57 PM Bug #11994 (Resolved): "unknown op append_excl" in upgrade:giant-x-next-distro-basic-vps run
- 02:55 PM Bug #10543 (Duplicate): "FAILED assert(m_seed < old_pg_num)" in upgrade:giant-x-hammer-distro-bas...
- 02:47 PM Bug #10930 (Pending Backport): mon: map_cache can become inaccurate if osd does not receive the o...
- 02:45 PM Bug #12681: tcmalloc doesn't release memory back to system
- We usually see this happening with specific releases of tcmalloc and host operating systems, although it's not someth...
- 02:33 PM Bug #12681: tcmalloc doesn't release memory back to system
- We're hping to switch to jemalloc... tcmalloc is annoying in many other ways as well.
- 02:40 PM Support #12760: FileSystem become read-only after over 6 hour massive read and write
- The MDS got back an error when it tried to write out the directory. What's the output of "ceph -s"? Can you upload th...
- 02:38 PM Bug #12688 (Duplicate): pg import can cause bogus interval (first > last)
- 02:34 PM Bug #12720: probably import/export related: osd/PG.cc: 6893: FAILED assert(pg->peer_info.count(so))
- 02:34 PM Bug #12720: probably import/export related: osd/PG.cc: 6893: FAILED assert(pg->peer_info.count(so))
- Sage says same_primary_since was probably wrong.
- 02:30 PM RADOS Bug #12687: osd thrashing + pg import/export can cause maybe_went_rw intervals to be missed
- 02:29 PM Bug #11332 (Can't reproduce): mon failed to read inc osdmap
- 02:26 PM Bug #12738: scrub bogus results when missing a clone
- 02:25 PM Bug #12746: failed assert ctx.manager.pg_inconsistent(pool, 0) on repair_test.yaml
- maybe add a flush pgs?
- 02:23 PM Bug #12748: bug with cache/tiering and snapshot reads [merged, needs a test in ceph-qa-suite]
- 02:23 PM Bug #12696 (Can't reproduce): ceph-osd starts before network, fails
- 02:07 PM Revision 6be5606b (ceph): Merge pull request #5661 from SUSE/wip-suse-sysvinit-cleanup
- ceph.spec.in: drop sysvinit-specific macros that run only on SUSE
Reviewed-by: Ken Dreyer <kdreyer@redhat.com> - 02:00 PM CephFS Bug #11789: knfs mount fails with "getfh failed: Function not implemented"
- We think this is the NFS kernel module not being loaded. If this is still happening we need to figure out why.
- 01:58 PM CephFS Bug #12653 (Fix Under Review): fuse mounted file systems fails SAMBA CTDB ping_pong rw test with ...
- 01:56 PM CephFS Bug #11783 (Fix Under Review): protocol: flushing caps on MDS restart can go bad
- 01:52 PM CephFS Bug #11784 (Can't reproduce): ceph-fuse hang on unmount (stuck dentry refs)
- 01:51 PM CephFS Bug #9994: ceph-qa-suite: nfs mount timeouts
- 01:51 PM CephFS Bug #12365 (Resolved): kcephfs: hang on umount
- Haven't seen this since then.
- 01:26 PM Bug #12779: valgrind failures in the monitor
- we switched to libboost to generated uuid recently.
- 01:07 PM Bug #12779 (Duplicate): valgrind failures in the monitor
- #12736
- 01:02 PM Bug #12779 (Duplicate): valgrind failures in the monitor
- ...
- 01:25 PM rbd Feature #12678: Support the ability to rename snapshots
- great! Would it be possible to mark this to be backported to hammer?
- 01:14 PM rbd Feature #12678: Support the ability to rename snapshots
- *master PR*: https://github.com/ceph/ceph/pull/5645
- 01:14 PM rbd Feature #12678 (In Progress): Support the ability to rename snapshots
- 01:19 PM rgw Bug #12780 (Resolved): rgw: segments are read during HEAD on Swift DLO
- $ curl -i "$publicURL/Disc Images2/HP Pavilion dv7-6c95dx Recovery Disc (1 of 3).ISO" -X HEAD -H "X-Auth-Token: $tok...
- 12:59 PM CephFS Bug #12777: qa: leftover files in cephtest directory
- http://pulpito.ceph.com/teuthology-2015-08-21_23:04:01-fs-master---basic-multi/1026047/
- 12:13 PM CephFS Bug #12777 (Resolved): qa: leftover files in cephtest directory
- http://pulpito.ceph.com/teuthology-2015-08-17_23:04:01-fs-master---basic-multi/1020426/
http://pulpito.ceph.com/teut... - 12:33 PM CephFS Bug #11482: kclient: intermittent log warnings "client.XXXX isn't responding to mclientcaps(revoke)"
- http://pulpito.ceph.com/teuthology-2015-08-17_23:08:05-kcephfs-master-testing-basic-multi/1020518/
Doesn't have th... - 12:10 PM CephFS Bug #12776 (Resolved): qa: standby MDS not shutting down, "reached maximum tries (50) after waiti...
- http://pulpito.ceph.com/teuthology-2015-08-17_23:04:01-fs-master---basic-multi/1020415/
The standby MDS doesn't lo... - 11:49 AM Revision d0386d23 (ceph): ceph.spec.in: drop sysvinit-specific macros that run only on openSUSE/SLE
- All sysvinit-based versions of openSUSE/SLE are EOL as far as upstream Ceph is
concerned.
Signed-off-by: Nathan Cutl... - 11:35 AM Revision 665de0a9 (ceph): Merge pull request #5657 from badone/master
- doc: Fix typo in recovery reservation section
Reviewed-by: Kefu Chai <kchai@redhat.com> - 11:31 AM Revision 699fca80 (ceph): doc: Fix typo and redundant word in snaps section
- Fix typo and remove identifying from "filtering identifying" in
doc/dev/osd_internals/snaps.rst
Signed-off-by: Brad ... - 11:14 AM Revision 4dffc167 (ceph): doc: Fix typo in writeback throttle section
- Fix typo in doc/dev/osd_internals/wbthrottle.rst
Signed-off-by: Brad Hubbard <bhubbard@redhat.com> - 11:04 AM Revision 6ee7068c (ceph): doc: Fix typo in writeback throttle section
- Fix typo in doc/dev/osd_internals/wbthrottle.rst
Signed-off-by: Brad Hubbard <bhubbard@redhat.com> - 10:51 AM Revision 07a28d6c (ceph): doc: Fix typo in recovery reservation section
- Fix typo in doc/dev/osd_internals/recovery_reservation.rst
Signed-off-by: Brad Hubbard <bhubbard@redhat.com> - 10:11 AM Revision b2028b79 (ceph): Merge pull request #5607 from rzarzynski/wip-12724
- rgw: fix error handling during GET on object through Swift API.
- 08:10 AM Revision 8c2f8bf5 (ceph): osd: no bother to create RecoveryCtx if no recovery op is started
- If there is no recovery op started, there is no need to create a RecoveryCtx
and later delete all of the things in it... - 08:10 AM Revision 178d4d5c (ceph): osd: remove unused parameter of start_recovery_ops
- The RecoveryCtx parameter is not needed.
Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com> - 03:26 AM Bug #12774: ceph add new osd on existing host, status keep down
- 2015-08-25 02:20:16.316941 7f3c8e273800 0 ceph version 0.80.9 (b5a67f0e1d15385bc0d60a6da6e7fc810bde6047), process ce...
- 03:21 AM Bug #12774 (Can't reproduce): ceph add new osd on existing host, status keep down
- Description of problem:
health cluster have 3 hosts, one mon and one osd per host, cluster status is 248 active + cl... - 03:22 AM Revision f77949fe (ceph): bug fix: osd: requeue_scrub when kick_object_context_blocked
- when read miss in writeback cache pool, osd do_proxy_read first
and maybe promote it. but in this case, the op is not... - 02:15 AM Revision 71eb7be2 (ceph): Merge pull request #5563 from varadakari/wip-kvstore-keyformat
- Change key format to preserve order of keys in backend db.
Reviewed-by: Sage Weil <sage@redhat.com>
08/24/2015
- 11:51 PM rbd Bug #12764 (Pending Backport): Failure in 'cram -v -- /home/ubuntu/cephtest/archive/cram.client.0...
- 03:45 PM rbd Bug #12764 (Fix Under Review): Failure in 'cram -v -- /home/ubuntu/cephtest/archive/cram.client.0...
- 03:45 PM rbd Bug #12764 (Resolved): Failure in 'cram -v -- /home/ubuntu/cephtest/archive/cram.client.0/*.t'
- *master PR*: https://github.com/ceph/ceph/pull/5649
- 03:14 PM rbd Bug #12764 (In Progress): Failure in 'cram -v -- /home/ubuntu/cephtest/archive/cram.client.0/*.t'
- 01:14 PM rbd Bug #12764 (Resolved): Failure in 'cram -v -- /home/ubuntu/cephtest/archive/cram.client.0/*.t'
- http://qa-proxy.ceph.com/teuthology/teuthology-2015-08-22_23:00:06-rbd-next-distro-basic-multi/1026973/teuthology.log
- 11:24 PM Bug #12614 (Fix Under Review): recursive lock of md_config_t (0)
- This is a false positive, disabled in:
https://github.com/ceph/ceph/pull/5654 - 10:40 PM Revision 41522699 (ceph): config: skip lockdep for intentionally recursive md_config_t lock
- lockdep can't handle recursive locks, resulting in false positive
reports for certain set_val_or_die() calls, like vi... - 10:15 PM Revision 9f2aad87 (ceph): Merge pull request #5649 from ceph/wip-12764
- tests: fixed rbd cli cram integration tests
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 10:04 PM Bug #12766: osd: segv in osd bench
- 03:03 PM Bug #12766 (Fix Under Review): osd: segv in osd bench
- 02:53 PM Bug #12766 (Resolved): osd: segv in osd bench
- ...
- 10:04 PM Bug #12736: boost uuid makes valgrind complain
- 04:38 PM Bug #12736 (Fix Under Review): boost uuid makes valgrind complain
- https://github.com/ceph/ceph/pull/5650
- 09:32 PM Documentation #12771 (Closed): Update the Latest Release Information in "get-packages.rst"
- Update the http://ceph.com/docs/master/install/get-packages page with the latest Ceph release (Hammer) in the "Add Ce...
- 09:27 PM Revision 473f927f (ceph): Add error codes.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 57154560 (ceph): FIXUP: Shuffle auth order back down.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision ca9db615 (ceph): Improve RGWHandler for S3Website.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 07192746 (ceph): Remove config rgw_s3website_mode, only hostname mode remains.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision ab90f5ab (ceph): Clean up headers.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision f7e85c36 (ceph): Mark where to add some functionality in future.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 4a6f968b (ceph): rgw: add 301 http text.
- HTTP/1.1 code 301 is missing a message.
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org> - 09:27 PM Revision 3bdbe59e (ceph): rgw: prepare for handlers to have custom error handler routines
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 1dc9ac66 (ceph): RGWWebsite: first pass at error redirect and error page handling.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 61b0ce65 (ceph): S3Website: Ensure redirect_all has a 302 response, just like AmazonS3
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision f7d04dd6 (ceph): S3Website: "good" error status is actually not zero, use the function i...
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 731c73b3 (ceph): s3website: Prepwork for x-amz-website-redirect-location header.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 8613b692 (ceph): Use more specific error codes for retarget of website requests
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision a32f768b (ceph): WIP: static-site errordoc work.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision a8db2984 (ceph): Fixup.
- Signed-off-by: Robin H. Johnson <robin.johnson@dreamhost.com>
- 09:27 PM Revision 47169921 (ceph): rgw-website: append the key, not overwrite.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision e06b46aa (ceph): rgw-website: handle redirect_all via RGWBWRoutingRule.
- Rather than duplicate code to construct the redirect if redirect_all is
used, construct a catch-all RoutingRule with ... - 09:27 PM Revision 5e729002 (ceph): rgw: Add note about future rgw performance optimization for regions wit...
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 1fb56621 (ceph): RGWRegion: AmazonS3 detects website endpoint storage
- AmazonS3 detects S3website mode via DNS:
${bucket}.s3-website-${region}.amazonaws.com.
VS:
${bucket}.s3-${region}.ama... - 09:27 PM Revision 2608983a (ceph): rgw: Document rgw_dns_name
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision eb27d6f2 (ceph): WIP-FIXME-TODO: Planning for hostname/endpoints specific to APIs.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision f378b5b6 (ceph): WIP-FIXME: Static fetch works without breaking anything else, uses buck...
- WIP-FIXME rgw-website: refactor to have RGW*S3Website classes for website-specific stuff
Signed-off-by: Robin H. Joh... - 09:27 PM Revision 2e4b1ba0 (ceph): FIXUP: Fix comment.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision fa02d4a0 (ceph): common: teach formatters about headers & footers.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 12e0063c (ceph): rgw: use new formatter header/footer ability.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision b3682b55 (ceph): rgw: More fields in error output to match main S3.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 96f9a018 (ceph): rgw: pretty-print s3website error output.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 4651b21f (ceph): More s3website json region data.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision ec168282 (ceph): rgw: tweak dns-based s3website detection for other changes FIXME-REBASE
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 8f0d0bdc (ceph): Do not automatically output header; make the dev do it
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 76f9f8f9 (ceph): Avoid dupe reset.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 5dd7bdc1 (ceph): WIP-FIXUP: Muck with bucket detection again.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision a82267e5 (ceph): DEBUG: Ensure we always go past the debug output for now.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 54aff3e5 (ceph): Optimize website detection differently.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision f3c6ac7d (ceph): common/*Formatters: Split Formatters
- No functionality changes in this commit, just shuffle lines to new files
and adding #include headers to ensure it sti... - 09:27 PM Revision eb14de2e (ceph): common/HTMLFormatter: add new formatter
- StaticSites/S3Website has error output in HTML, for which we need a new
formatter. It's implemented by extending the ... - 09:27 PM Revision 2f8075c2 (ceph): common/*Formatters: handling of headers & footers.
- Formatters now gain first-class headers and footers, along with
consistent handling of having header/footer only outp... - 09:27 PM Revision e45469ab (ceph): rgw: define data structures for static website config
- move website rules to separate file, check rules,
and implement key modification functionality
Signed-off-by: Yehuda... - 09:27 PM Revision 6d791d6d (ceph): rgw: retarget requests
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 09:27 PM Revision d2b48f9f (ceph): rgw: initial implementation of website REST api
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 09:27 PM Revision b31e6ef8 (ceph): rgw: enable website as subresource
- also enable GetBucketWebsite op.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 09:27 PM Revision 34ba4372 (ceph): rgw: xml encoder / decoder
- similar to the json decoder, for easier xml parsing.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 09:27 PM Revision 0fb16bdb (ceph): rgw: add xml encoders for website conf
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 09:27 PM Revision 33c57102 (ceph): rgw: set bucket website operation
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 09:27 PM Revision dc1a457d (ceph): rgw: no need to explicitly allocate XMLObj
- let the parser do the allocation if needed. Also, keep unallocated
objects in a list.
Signed-off-by: Yehuda Sadeh <y... - 09:27 PM Revision 17ce9de3 (ceph): rgw: fix xml encoding for certain website operations
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 09:27 PM Revision 2d50cbb7 (ceph): rgw: add remove website api
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 09:27 PM Revision dcb66f1c (ceph): rgw: apply redirect rules in static website
- only when key matches, not checking for return code yet.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 09:27 PM Revision 7bb0ddab (ceph): rgw-website: do not truncate nuke key if valid.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision dbadaa96 (ceph): rgw-website: ListBucket not valid for websites
- ListBucket/StatBucket are not a valid action for website mode, redirect
it to GetObj instead, where it will modified ... - 09:27 PM Revision a55cb566 (ceph): rgw-website: cleanup unused variable: rules
- With the XML dumping moved, this variable is not required.
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org> - 09:27 PM Revision 84dfb4ef (ceph): rgw-website: Do not apply retarget for an authenticated request.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:27 PM Revision 7482acf4 (ceph): rgw-website: Use default hostname/protocol if redirect does not specify.
- Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
- 09:02 PM Bug #9987: mon: min_last_epoch_complete tracking broken
- For the record, the fix at https://github.com/ceph/ceph/pull/3220 shipped in v0.80.8.
- 08:52 PM Revision 8ef2c965 (ceph): buffer: modify inline memory ops to use packed structs
- packed structs can be used to ensure alignment is not an issue.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 08:15 PM devops Bug #12768: Error: Package: 1:ceph-0.87.2-8.g6c33a35.el7.x86_64 (Ceph)
- That does not ring a bell. Since I'm on vacation it's probably wise to assign this to someone else.
- 05:10 PM devops Bug #12768: Error: Package: 1:ceph-0.87.2-8.g6c33a35.el7.x86_64 (Ceph)
- Loic, can you take a look at this?
This is a jobs from /ceph-qa-suite/suites/upgrade/giant-x/stress-split-erasure-co... - 04:13 PM devops Bug #12768: Error: Package: 1:ceph-0.87.2-8.g6c33a35.el7.x86_64 (Ceph)
- Sepia rhel vps nodes are not entitled, would python-flask be something coming from rhel entitlements?
- 04:11 PM devops Bug #12768: Error: Package: 1:ceph-0.87.2-8.g6c33a35.el7.x86_64 (Ceph)
- Looks like it's missing python-flask?...
- 04:05 PM devops Bug #12768 (Rejected): Error: Package: 1:ceph-0.87.2-8.g6c33a35.el7.x86_64 (Ceph)
- Run: http://pulpito.ceph.com/teuthology-2015-08-23_15:08:07-upgrade:giant-x-hammer-distro-basic-vps/
Job: 1028058
L... - 06:51 PM Revision dbcaa544 (ceph): uuid: use boost::random:random_device
- The boost mt code uses uninitialized memory for extra randomness,
which is a bad idea in general but more importantly... - 06:45 PM CephFS Bug #12715 (Resolved): "[ERR] bad backtrace on dir ino 600" in cluster log"
- 06:42 PM CephFS Bug #12715: "[ERR] bad backtrace on dir ino 600" in cluster log"
- https://github.com/ceph/ceph-qa-suite/pull/539
- 06:36 PM CephFS Bug #12715: "[ERR] bad backtrace on dir ino 600" in cluster log"
- whitelist "bad backtrace on dir ino" warning message (per irc chat with Greg)
- 05:30 PM CephFS Bug #12715: "[ERR] bad backtrace on dir ino 600" in cluster log"
- Also in run:
http://pulpito.ceph.com/teuthology-2015-08-21_08:42:54-upgrade:firefly-x-hammer-distro-basic-vps/
Jobs... - 03:37 AM CephFS Bug #12715: "[ERR] bad backtrace on dir ino 600" in cluster log"
- the test uses 0.80.4 (7c241cfaa6c8c068bc9da8578ca00b9f4fc7567f). the newest firefly include the fix (commit a5970963)
- 06:42 PM Revision 4fdd1b8d (ceph): Merge pull request #539 from ceph/wip-12715
- Whitelisted "bad backtrace on dir ino" to fis #12715
- 06:39 PM Revision 79ea4b6b (ceph): Whitelisted "bad backtrace on dir ino" to fis #12715
- Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
- 06:29 PM rgw Bug #12761 (Fix Under Review): Keystone Fernet tokens break auth
- 06:19 PM rgw Bug #12761 (In Progress): Keystone Fernet tokens break auth
- master pr: https://github.com/ceph/ceph/pull/5651
- 01:49 PM rgw Bug #12761: Keystone Fernet tokens break auth
- Looks like rgw's parser expects milliseconds precision only and fails when seconds tells microseconds as well
- 03:57 AM rgw Bug #12761 (Resolved): Keystone Fernet tokens break auth
- When using Fernet tokens in Keystone (as opposed to UUID or PKI), RGW does not handle them correctly due to the times...
- 05:41 PM Revision 136242b5 (ceph): rgw: be more flexible with iso8601 timestamps
- make parsing 8601 more flexible by not restricting the length of seconds
to 5, this allows timestamp to be specified ... - 04:59 PM Revision 03471229 (ceph): Merge pull request #5617 from rzarzynski/wip-12728
- rgw: url_decode bucket name and prefix from X-Object-Manifest during GET on Swift DLO.
Reviewed-by: Yehuda Sadeh <ye... - 03:42 PM Revision fd725771 (ceph): tests: fixed rbd cli cram integration tests
- Fixes: #12764
Backport: infernalis
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 03:33 PM rgw Bug #12767 (Duplicate): common/Mutex.cc: 100: FAILED assert(r == 0) from RGWWatcher::handle_error...
- ...
- 03:32 PM devops Tasks #10587: split ceph-mon, ceph-osd, and ceph-mds in the Debian and Red Hat packages
- To clarify the work that remains to be done here:
# The "wip-10587-split-servers" branch needs to be rebased onto ... - 03:13 PM Revision 9063eaa6 (ceph): Merge pull request #538 from ceph/wip-12758
- ceph_deploy: if a branch is given, use it when installing rgw
- 03:12 PM Revision f20f7a23 (ceph): Objecter: pg_interval_t::is_new_interval needs pgid from previous pool
- When increasing the pg_num of a pool, an assert would fail since the
calculated pgid seed would be for the pool's new... - 03:09 PM Revision 766da54d (ceph): ceph_deploy: if a branch is given, use it when installing rgw
- Fixes: http://tracker.ceph.com/issues/12758
Signed-off-by: Andrew Schoen <aschoen@redhat.com> - 03:02 PM rbd Bug #12765 (Fix Under Review): Failure in test_librbd_python.sh
- *master PR*: https://github.com/ceph/ceph/pull/5647
- 01:16 PM rbd Bug #12765 (Resolved): Failure in test_librbd_python.sh
- http://qa-proxy.ceph.com/teuthology/teuthology-2015-08-22_23:00:06-rbd-next-distro-basic-multi/1026985/teuthology.log
- 03:02 PM Revision cd6ac726 (ceph): librbd: error closing image while set to invalid snapshot
- With cache disabled, closing the image will result in a flush
which might result in the image being refreshed. If th... - 03:00 PM Revision 622d22e2 (ceph): osd: wait for cleanup from bench
- We need to wait for cleanup to flush before we destroy the
sequencer or else we get a use-after-free.
Introduced e7b... - 02:31 PM Revision c5895d3f (ceph): bug fix: osd: do not cache unused buffer in attrs
- attrs only reference the origin bufferlist (decode from MOSDPGPush or
ECSubReadReply message) whose size is much grea... - 01:36 PM Revision 888a6337 (ceph): doc/release-notes: v9.0.3
- Signed-off-by: Sage Weil <sage@redhat.com>
- 01:22 PM CephFS Bug #12710 (Resolved): fsstress.sh fails
- 01:22 PM CephFS Bug #12709 (Resolved): hammer chmod.sh fails
- fixed by commit 81a311a744987564b70852fdacfd915523c73b5d
- 01:21 PM CephFS Bug #12711 (Resolved): mds get damaged
- 12:35 PM Revision b2daae6a (ceph): Merge remote-tracking branch 'gh/next'
- 12:35 PM Revision 66980795 (ceph): Merge remote-tracking branch 'gh/next'
- 12:07 PM Revision fba67606 (ceph): Merge pull request #5623 from yuyuyu101/wip-12729
- Fix compile warning unused-result
Reviewed-by: Kefu Chai <kchai@redhat.com> - 11:09 AM rgw Bug #11239: civetweb defaults to libssl.so and libcrypto.so when versions not passed
- It should work if you create a couple of symlinks:...
- 10:39 AM rgw Bug #11239: civetweb defaults to libssl.so and libcrypto.so when versions not passed
- Also it seems the config file parser whines about the trailing 's' on the port number:...
- 10:52 AM Bug #12673: cache agent is idle although one object is left in the cache
- 08:36 AM CephFS Bug #12676 (Resolved): MDSMap assertion in MDCache::trim (multimds)
- ...
- 08:33 AM CephFS Bug #12321 (Can't reproduce): MDS crash when try to connect clients
- 03:52 AM CephFS Bug #12321: MDS crash when try to connect clients
- John Spray wrote:
> Hi, do you have any updates for us on this? If the system is unavailable for any more debug the... - 06:30 AM CephFS Bug #12753 (Resolved): cls_cephfs_client encodes time_t directly
- fixed by 1213dde3d207d0d91ccecfca4dd6af1bdee0ed65
- 06:15 AM Linux kernel client Bug #12697: [RBD] rbd_dev_id_put causes system crash
- Ilya Dryomov wrote:
> Any chance you could share the relevant Go pieces, pre and post adjustment? I suspect your co... - 05:54 AM Linux kernel client Bug #12763 (Closed): rbd: unmap failed: (16) Device or resource busy
- My ceph version is 0.87.
The linux system info
[root@client27 ~]# uname -a
Linux client27 3.10.0-123.el7.x86_64 ... - 03:49 AM Support #12760 (New): FileSystem become read-only after over 6 hour massive read and write
- The following is the full mds log, nothing else:
1 2015-08-21 21:51:39.224967 7f0f4117b700 1 mds.0.cache.dir(10000... - 03:38 AM Revision d63508f7 (ceph): make-check: support MAKEOPTS overrides.
- It is useful in some cases to exercise more control over the parameters
passed to make for make check. I am using it ... - 01:18 AM Revision d2e4fe36 (ceph): set skip promote flag if NOCACHE or DONTNEED flag set
- Signed-off-by: Yunchuan Wen <yunchuanwen@ubuntukylin.com>
Reviewed-by: Li Wang <liwang@ubuntukylin.com>
Reviewed-by: ...
08/23/2015
- 06:02 PM Revision 151c0511 (ceph): common/Mutex: avoid trylock on lock if instrumentation is not enabled
- Benchmarks have shown that the trylock in the lock path has a high
latency cost. Only pay the penalty if instrumenta... - 12:25 PM Bug #12614: recursive lock of md_config_t (0)
- /var/lib/teuthworker/archive/sage-2015-08-22_06:35:33-rados-wip-sage-testing-distro-basic-multi/1026420
- 12:21 PM rgw Bug #12759 (Duplicate): s3tests.functional.test_s3.test_object_copy_canned_acl failure
- On smoke suite and master,...
- 09:32 AM Revision 42f8c5da (ceph): osd: use GMT time for the object name of hitsets
- * bump the encoding version of pg_hit_set_info_t to 2, so we can
tell if the corresponding hit_set is named using l... - 09:32 AM Revision 03a1a3cf (ceph): mon: add "ceph osd pool set $pool use_gmt_hitset true" cmd
- allow "ceph osd pool set $pool use_gmt_hitset <true|1>" as long as
the cluster supports gmt hitset.
Fixes: #9732
Sig... - 09:32 AM Revision cc2bcf76 (ceph): mon: print use_gmt_hitset in "ceph osd pool get"
- Signed-off-by: Kefu Chai <kchai@redhat.com>
- 09:32 AM Revision ff087f9e (ceph): osd: remove useless hitset code
- * we do not persist current hit set using get_hit_set_current_object()
anymore, instead we always append current hi... - 09:32 AM Revision 2c7fe2bc (ceph): osd: remove unused ctor
- Signed-off-by: Kefu Chai <kchai@redhat.com>
08/22/2015
- 09:24 PM Revision df03d84f (ceph): Merge pull request #523 from dachary/wip-10887-shec
- erasure-code: CEPH_FEATURE_ERASURE_CODE_PLUGINS_V3 integration tests
Reviewed-by: Takeshi Miyamae <miyamae.takeshi@j... - 09:22 PM Revision 2d92f4cb (ceph): Merge pull request #5493 from dachary/wip-10887-shec
- erasure-code: shec plugin feature
Reviewed-by: Takeshi Miyamae <miyamae.takeshi@jp.fujitsu.com>
Reviewed-by: Sage We... - 01:47 PM Bug #12696: ceph-osd starts before network, fails
- - what OS/distro is this?
- is the ceph service enabled? the final start should have done an activate-all which w... - 01:46 PM Bug #12696 (Need More Info): ceph-osd starts before network, fails
- 11:22 AM Bug #12755 (Closed): selinux: ceph-mon denial on wip-user
- type=AVC msg=audit(1440160179.917:2120): avc: denied { dac_override } for pid=8523 comm="ceph-mon" capability=1 s...
- 01:22 AM Revision 5a20f280 (ceph): Merge pull request #5631 from wonzhq/finish-ctx
- osd: consolidate encoding snaps into pg log code in finish_ctx
Reviewed-by: Kefu Chai <kchai@redhat.com>
08/21/2015
- 11:20 PM rbd Bug #12604 (Resolved): rbd map failure in tasks/rbd_workunit_suites_iozone.yaml
- The originating cause for the issue has been reverted.
- 11:14 PM rbd Bug #12618 (Need More Info): incomplete rbd delete(command:rbd rm) damaged the rbd block
- What's the issue here that you are trying to address? The error message indicates what you need to do -- what for th...
- 11:13 PM rbd Bug #12743 (Duplicate): incomplete rbd delete(command:rbd rm) damaged the rbd block
- 12:04 AM rbd Bug #12743 (Duplicate): incomplete rbd delete(command:rbd rm) damaged the rbd block
- #When I try to delete a rbd block(the size is 20G),I use the command
rbd -p rbd rm myblock
#When this command is ... - 11:12 PM rbd Bug #12700 (Duplicate): incomplete rbd delete(command:rbd rm) damaged the rbd block
- 10:58 PM Feature #12754 (Resolved): pg auto repair for EC pool
- For EC pool, since we know which replica is corrupted, it makes sense to do auto repair with scrubbing.
- 09:53 PM Revision c4aa935f (ceph): Merge pull request #5611 from ceph/wip-newstore-misc
- osd: final prereq patches for newstore
Reviewed-by: Samuel Just <sjust@redhat.com> - 09:52 PM Revision 0045b8dd (ceph): PendingReleaseNotes: make a note about KeyValueStore on-disk format change
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:45 PM Revision ea8609b2 (ceph): mon/OSDMonitor: debug why pool creation fails
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:45 PM Revision 0f82f461 (ceph): crush/CrushTester: test fewer inputs when running crushtool
- If there are a lot of crush rules (say, 100) then the test can
take a long time. 100 values per rule should be enoug... - 09:40 PM CephFS Bug #12753 (Resolved): cls_cephfs_client encodes time_t directly
- fail to build on i386.
we should never encode time_t directly.. cast to uint32_t or uint64_t so it is sized explic... - 08:54 PM CephFS Bug #12715: "[ERR] bad backtrace on dir ino 600" in cluster log"
- Zheng, Sage's mentioned that this may have been fixed by you, can you take a look?
- 08:19 PM CephFS Bug #12715: "[ERR] bad backtrace on dir ino 600" in cluster log"
- This is an old bug, right? We should just whitelist this?
- 07:06 PM CephFS Bug #12715: "[ERR] bad backtrace on dir ino 600" in cluster log"
- Run http://pulpito.ceph.com/teuthology-2015-08-21_08:42:54-upgrade:firefly-x-hammer-distro-basic-vps/
Jobs: 1024928,... - 08:43 PM Backport #12752 (Resolved): kvm die with assert(m_seed < old_pg_num)
- https://github.com/ceph/ceph/pull/5726
- 08:42 PM Backport #12751 (Resolved): kvm die with assert(m_seed < old_pg_num)
- https://github.com/ceph/ceph/pull/5691
- 08:41 PM Bug #10399: kvm die with assert(m_seed < old_pg_num)
- giant is retired
- 03:45 PM Bug #10399 (Fix Under Review): kvm die with assert(m_seed < old_pg_num)
- *master PR*: https://github.com/ceph/ceph/pull/5648
- 03:12 PM Bug #10399 (In Progress): kvm die with assert(m_seed < old_pg_num)
- Easily repeatable when "rbd bench-write" is running in the background while you double the number of PGs.
- 08:33 PM rgw Bug #12750 (Fix Under Review): rgw: wrong removal during iteration over Swift metadata in filter_...
- Pull request: https://github.com/ceph/ceph/pull/5639.
- 08:32 PM rgw Bug #12750 (In Progress): rgw: wrong removal during iteration over Swift metadata in filter_out_t...
- 08:25 PM rgw Bug #12750 (Resolved): rgw: wrong removal during iteration over Swift metadata in filter_out_temp...
- During iteration over STL map in _RGWPutMetadataAccount::filter_out_temp_url_ erase operation on the same map might t...
- 08:29 PM Revision 7c08c54f (ceph): rgw: fix dangerous removal from STL map in filter_out_temp_url().
- Fixes: #12750
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 08:03 PM Revision 660ae5bc (ceph): osd: always load erasure plugins from the configured directory
- Ignore the profile 'directory' field.
This ensures that we can always find plugins even when teh cluster
is installe... - 08:03 PM Revision 5df12716 (ceph): do not include directory in ec profiles
- Signed-off-by: Sage Weil <sage@redhat.com>
- 08:03 PM Revision 88bfd796 (ceph): test/erasure-code: drop directory from profile
- Signed-off-by: Sage Weil <sage@redhat.com>
- 07:46 PM Revision 7295612d (ceph): 9.0.3
- 07:27 PM Revision 271513f8 (ceph): erasure-code: shec plugin feature
- There is one new plugin (shec). When upgrading a cluster, there
must be a protection against the following scenario:
... - 07:04 PM Bug #11771: is_new_interval: size change should cause new interval
- ff79959c037a7145f7104b06d9e6a64492fdb95f
- 06:54 PM Bug #11771: is_new_interval: size change should cause new interval
- Yes, yes it should.
- 06:35 PM Feature #12749 (New): add an option for ceph_test_rados to have a very skewed distribution for ac...
- We have a hole in our cache/tiering testing where all head objects are pretty much hot, and we tend not to evict them...
- 06:28 PM Bug #12748: bug with cache/tiering and snapshot reads [merged, needs a test in ceph-qa-suite]
- Specifically, ssc->exists should have been checked.
- 06:07 PM Bug #12748 (Resolved): bug with cache/tiering and snapshot reads [merged, needs a test in ceph-qa...
- From mailing list (http://article.gmane.org/gmane.comp.file-systems.ceph.user/22901), courtesy of Ilya:
I think I ... - 04:49 PM rbd Bug #12708 (Resolved): Even if RBD export-diff failure, will still create a local file
- 04:26 PM Revision 5e99a578 (ceph): mon: add a cache layer over MonitorDBStore
- the cache of of leveldb does not perform well under some condition,
so we need a cache in our own stack.
* add an op... - 03:59 PM Bug #12747 (Fix Under Review): osd/PGBackend.cc: 294: FAILED assert(ec_impl) - wrong path on rpm ...
- https://github.com/ceph/ceph/pull/5636
- 03:47 PM Bug #12747: osd/PGBackend.cc: 294: FAILED assert(ec_impl) - wrong path on rpm distros
- When a profile is created, the directory is filled in by the monitor. But... the OSDs may be different distros, whic...
- 01:46 AM Bug #12747 (Resolved): osd/PGBackend.cc: 294: FAILED assert(ec_impl) - wrong path on rpm distros
- ...
- 01:49 PM Revision 929cf29f (ceph): Merge pull request #5634 from ceph/wip-fix-FTBFS-boost
- rbd: fix the FTBFS on old boost introduced by 2050d08
- 01:41 PM Revision 2d13a472 (ceph): rbd: fix the FTBFS on old boost introduced by 2050d08
- in boost 1.49, BOOST_SCOPE_EXIT() does not accept capture_tuple,
only `(capture) (capture) ...` is supported.
Signed... - 12:11 PM Revision bfc259ef (ceph): Merge pull request #5583 from ceph/wip-12676
- mds: check mds up before calling MDSMap::get_mds_info
Reviewed-by: Greg Farnum <gfarnum@redhat.com> - 11:33 AM Revision 64251118 (ceph): Merge pull request #5587 from caibo2014/master
- rbd: remove the output file if rbd export-diff fails
Reviewed-by: Kefu Chai <kchai@redhat.com> - 11:22 AM Revision 2daccff5 (ceph): Merge pull request #5632 from ceph/wip-fix-ftbfs-4998
- osd: fix the FTBFS introduced by be28319
Reviewed-by: Greg Farnum <gfarnum@redhat.com> - 10:29 AM Linux kernel client Bug #12697: [RBD] rbd_dev_id_put causes system crash
- Any chance you could share the relevant Go pieces, pre and post adjustment? I suspect your code isn't doing any udev...
- 10:23 AM Linux kernel client Bug #12697: [RBD] rbd_dev_id_put causes system crash
- Hi Ilya,
Looks like this is caused by rbd unmap, but it is due to our codes to map rbd firstly, then unmap rbd via... - 09:24 AM Revision d1c40868 (ceph): osd: fix the FTBFS introduced by be28319
- the signature of Scrubber::write_blocked_by_scrub() was changed in #5470.
Signed-off-by: Kefu Chai <kchai@redhat.com> - 08:51 AM Revision e06207a3 (ceph): Merge pull request #4998 from wonzhq/hitset-remove-all
- osd: implement hit_set_remove_all
Reviewed-by: Kefu Chai <kchai@redhat.com> - 08:08 AM Revision 5fedc841 (ceph): osd: consolidate encoding snaps into pg log code in finish_ctx
- 'MODIFY', 'PROMOTE' and 'CLEAN' are sharing the same encoding snaps
code.
Signed-off-by: Zhiqiang Wang <zhiqiang.wan... - 07:48 AM Revision 5812adb6 (ceph): osd: no need to check current hitset object is degraded when persisting
- hit set
The current hit set object doesn't exist yet. We are about to create it.
Signed-off-by: Zhiqiang Wang <zhiq... - 07:48 AM Revision fd38902d (ceph): osd: only remove all the hit set objects on primary osd
- Only the primary PG is allowed to remove all the hit set objects. And
the PG should be in the active or peered states... - 07:45 AM Revision be28319b (ceph): osd: implement hit_set_remove_all
- When hit set is not configured on startup or on a change, remove all
previous hitsets.
Signed-off-by: Zhiqiang Wang ... - 06:38 AM Revision 9c325d2b (ceph): skip promote if flag set when doing proxy read
- Signed-off-by: Yunchuan Wen <yunchuanwen@ubuntukylin.com>
Reviewed-by: Li Wang <liwang@ubuntukylin.com> - 06:16 AM Revision faa964d3 (ceph): Fix compile warning unused-result
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 05:44 AM CephFS Bug #12732: very slow read when a file has holes.
- '(rc < 0 && rc != -ENOENT)' should work. please send a patch to ceph-devel@vger.kernel.org
- 05:16 AM rgw Bug #12537: [CVE-2015-5245] RGW returns requested bucket name raw in "Bucket" response header
- Sage,
When can the Hammer backport be expected? Is there a time-line to expect this?
Vimal - 03:18 AM Revision b33209b4 (ceph): Merge pull request #5315 from tchaikov/wip-10930
- mon: remove osd_epoch to avoid out-dated osdmap_cache
Reviewed-by: Sage Weil <sage@redhat.com> - 02:56 AM Revision 1546d57d (ceph): osd: do either flush or evict but not both in agent_work
- An cache pool object is either dirty or not. It's unlikely the agent
will do both flush and evict at the same time fo... - 02:51 AM Revision 5eb80b81 (ceph): osd: do evict before flush in agent_work
- This is to avoid the extreme case that the agent continuously does
flush, but not evict. This may lead to the cache p... - 01:41 AM Bug #12746 (Resolved): failed assert ctx.manager.pg_inconsistent(pool, 0) on repair_test.yaml
- "/a/sage-2015-08-19_19:18:53-rados-wip-newstore-misc-distro-basic-multi/1023449":http://pulpito.ceph.com/sage-2015-08...
- 01:36 AM Messengers Bug #12745: ms/simple: Messenger/MessengerTest.SyntheticInjectTest2/1 hang
- ...
- 01:35 AM Messengers Bug #12745 (Resolved): ms/simple: Messenger/MessengerTest.SyntheticInjectTest2/1 hang
- /a/sage-2015-08-19_19:18:53-rados-wip-newstore-misc-distro-basic-multi/1023384
ceph_test_msgr... - 12:09 AM Bug #12744 (Resolved): ceph -s output shows "N pgs backfill", should be "waiting for backfill"
- at least I think so. Excerpt:
23 pgs backfill
10 pgs backfilling
...
21 active+remapped+wait_backfill
9 ac...
08/20/2015
- 11:10 PM Feature #12742 (Resolved): modify cache/tiering rados suite tests to cycle between write-back mod...
- I think we aren't getting the coverage on forward mode we think we are.
- 10:46 PM rgw Bug #12735 (Pending Backport): rgw: missing handling of encoding-type=url when listing keys in bu...
- 05:12 PM rgw Bug #12735 (Resolved): rgw: missing handling of encoding-type=url when listing keys in bucket
- Opening this bug so that we could track this fix:
https://github.com/ceph/ceph/pull/5579
This is not a full encod... - 10:45 PM Revision 8314384a (ceph): Merge pull request #5579 from jbweber/s3_encoding_type
- rgw: implement s3 encoding-type for get bucket
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 10:33 PM Revision 180ca7b5 (ceph): rgw: implement s3 encoding-type for get bucket
- This change introduces handling for the encoding-type request
parameter on the get bucket operation. An object key ma... - 09:50 PM Feature #12740 (Resolved): ceph-objetstore-tool: add operations to remove clone ids from a head o...
- Sometimes, various kinds of on disk corruption can result in a head/snapdir believing that a clone exists which actua...
- 09:43 PM Bug #12738 (Resolved): scrub bogus results when missing a clone
- 2015-08-20 20:15:47.150457 mon.0 10.12.2.2:6789/0 1313 : cluster [INF] pgmap v5060458: 2048 pgs: 1 active+clean+scrub...
- 09:18 PM Revision d164ec15 (ceph): perf_serialize: fix i386 build
- test/perf_local.cc: In function ‘double perf_serialize()’:
error: test/perf_local.cc:683:20: inconsistent operand con... - 08:40 PM Revision 56690777 (ceph): buffer: move inline memory ops to inline_memory.h; gcc + x86_64 only
- Keep the architecture-sensitive code in a separate header.
Avoid duplicating the unrolled memcpy in each buffer.cc
me... - 08:30 PM Revision cdaf9974 (ceph): unittest_bufferlist: benchmark buffer::ptr::append, copy_in, copy_out
- Signed-off-by: Sage Weil <sage@redhat.com>
- 07:20 PM Revision 5686d9ba (ceph): Merge branch 'dont-cache-recovery' of git://github.com/yuyuyu101/ceph
- Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Sage Weil <sage@redhat.com>
# Conflicts:
# src/osd/Replicat... - 07:11 PM rgw Feature #12714 (Fix Under Review): RGW : setting max number of buckets for user via ceph.conf opt...
- PR : https://github.com/ceph/ceph/pull/5624
- 07:09 PM Revision 779466e3 (ceph): Merge pull request #5259 from branch-predictor/bp-tuned-bufferlists
- Bufferlist tuning
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Haomai Wang <haomaiwang@gmail.com>
Reviewed-... - 06:59 PM Revision 8bb76f29 (ceph): Merge pull request #5575 from ceph/wip-12370
- rgw: don't preserve acls when copying object
- 06:53 PM Bug #12652 (Pending Backport): osd/PGLog.cc: 732: FAILED assert(log.log.size() == log_keys_debug....
- 06:53 PM Revision 181dc288 (ceph): Merge pull request #5518 from ceph/wip-12652
- osd/PGLog: dirty_to is inclusive
Reviewed-by: Samuel Just <sjust@redhat.com> - 06:47 PM Revision 6468ffb6 (ceph): Merge pull request #5539 from wonzhq/flush-mode-stat
- mon: show # of PGs at high flush mode in the command output
Reviewed-by: Sage Weil <sage@redhat.com> - 06:46 PM Bug #12737 (Can't reproduce): pg stuck in WaitLocalBackfillReserved, asok unresponsive
- several PGs were stuck in backfilling and wait_backfill state. They appeared blocked by osd.1, which had 2 osds in wa...
- 06:38 PM Bug #12736 (Resolved): boost uuid makes valgrind complain
- the boost random number generator uses uninitialized memory as an extra source of randomness:
https://svn.boost.or... - 04:15 PM Revision aeb80d03 (ceph): Merge pull request #5556 from zaitcev/cleanup-dead-code
- rgw: cleanup dead init_bucket
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 04:13 PM Revision f04322a8 (ceph): Merge pull request #5598 from jasuarez/remove-useless-code
- rgw: Remove useless code in calc_hmac_sha1()
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 03:50 PM Bug #12734 (Resolved): uuid-runtime no longer installed by build depends
- 02:26 PM Bug #12734: uuid-runtime no longer installed by build depends
- 01:47 PM Bug #12734: uuid-runtime no longer installed by build depends
- The commit that caused the problem is commit:62bfc7a1ab1587e81ed3bff0ddfbb1aa69d1c299
- 01:46 PM Bug #12734 (Resolved): uuid-runtime no longer installed by build depends
- It looks like something happened in the chain of dependencies in Ubuntu 14.04 recently and the package uuid-runtime w...
- 03:49 PM Revision bf108a34 (ceph): Merge pull request #5620 from dachary/wip-12734-uuid
- build/ops: add uuid-runtime to debian/control
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Reviewed-by: Kefu Chai ... - 02:28 PM Bug #9221: LibRadosTwoPoolsPP.PromoteOn2ndRead
- This can still happen if there is a connection reset... and the client can reconnect quickly so the previous workarou...
- 02:19 PM Revision d2d540c1 (ceph): build/ops: add uuid-runtime to debian/control
- uuid-runtime was incorrectly removed by
62bfc7a1ab1587e81ed3bff0ddfbb1aa69d1c299, restore it.
http://tracker.ceph.co... - 02:18 PM Revision 4e88268d (ceph): Merge pull request #5621 from ceph/revert-5613-wip-12729
- Revert "Fix compile warning unused-result"
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 02:17 PM Revision 951a3d6d (ceph): Revert "Fix compile warning unused-result"
- 01:51 PM Bug #12665: osd/ReplicatedPG.cc: 2706: FAILED assert(p != snapset.clones.end())
- Hi Bram,
Can you generate a full log (debug osd = 20, debug ms = 1) for a scrub that leads to a crash (if it does ... - 01:41 PM Bug #12665: osd/ReplicatedPG.cc: 2706: FAILED assert(p != snapset.clones.end())
- Hi Sage,
Any ideas on this problem ?
Can we provide more logs or should we run some extra tests ?
The problem is... - 01:39 PM Bug #12614: recursive lock of md_config_t (0)
- /a/sage-2015-08-19_19:13:53-rados-wip-sage-testing-distro-basic-multi/1023145
- 01:08 PM Bug #12729 (Resolved): ms/async: build warning
- 01:08 PM Revision 910cc17b (ceph): Merge pull request #5613 from ceph/wip-12729
- Fix compile warning unused-result
Reviewed-by: Sage Weil <sage@redhat.com> - 12:52 PM Revision 278d732e (ceph): os/FileJournal: Fix journal write fail, align for direct io
- when config journal_zero_on_create true, osd mkfs will fail when zeroing journal.
journal open with O_DIRECT, buf sho... - 12:43 PM RADOS Bug #12733 (New): cli: error connecting to cluster does not explain the problem
- Trying to check the health status on a Ceph node:...
- 11:34 AM rgw Bug #12363 (Pending Backport): testGetContentType and testHead failed
- 11:33 AM rgw Bug #12363: testGetContentType and testHead failed
- https://github.com/ceph/ceph/pull/5577
- 11:03 AM Revision 753f93dd (ceph): Merge pull request #526 from ceph/wip-stashed-layout
- tasks/cephfs: implement TestDataScan.test_stashed_layout
- 10:13 AM Revision bb70ac8e (ceph): Merge pull request #5523 from ceph/wip-scan-use-layout
- Striped layouts in cephfs-data-scan
- 10:10 AM Revision aef84913 (ceph): Merge pull request #5272 from chenji-kael/work-branch
- rados.py: wrap omap operation for librados
Reviewed-by: Kefu Chai <kchai@redhat.com> - 10:07 AM CephFS Bug #12732 (Resolved): very slow read when a file has holes.
- When a file in cephfs has holes, reading this file is very slow.
This problem can be reproduced by the commands bel... - 09:47 AM Revision ea103798 (ceph): Merge pull request #5614 from liewegas/wip-logtomonitors
- mds/MDSRank: less noisy about log_to_monitors
- 09:40 AM Revision 2050d08c (ceph): rbd:remove the local file when rbd export-diff fail
- Signed-off-by: Bo Cai <cai.bo@h3c.com>
- 09:28 AM rgw Bug #12728 (Fix Under Review): rgw: value of Swift API's X-Object-Manifest header is not url_deco...
- Pull request: https://github.com/ceph/ceph/pull/5617.
- 09:25 AM Revision 69cf0895 (ceph): rgw: url_decode values from X-Object-Manifest during GET on Swift DLO.
- Fixes: #12728
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 09:19 AM Revision d353dbc5 (ceph): rados.py: This module now supports omap operations
- you can operate omap in python
test_rados.py: add unit-test cases
Signed-off-by: Ji Chen <insomnia@139.com> - 03:01 AM Revision 050e1c5c (ceph): Merge pull request #5601 from wonzhq/doc-peering
- doc: fix the format of peering.rst
Reviewed-by: Kefu Chai <kchai@redhat.com> - 02:58 AM Revision d5a56c49 (ceph): doc: update some of the outdated cache tiering doc
- Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com>
- 02:49 AM Revision de00f6d0 (ceph): doc: add the description for min_read_recency_for_promote
- This cache tiering option is introduced in proxy read.
Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com> - 02:30 AM Revision b297e6d2 (ceph): ceph-objectstore-tool: drop support for pre-pgmeta PGs
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:30 AM Revision 71deb4b9 (ceph): test/objectstore/FilestoreDiff: don't diff coll attrs
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:30 AM Revision 893e00bc (ceph): os: drop deprecated collection_* attr methods
- We no longer need these for upgrades.
Signed-off-by: Sage Weil <sage@redhat.com> - 02:30 AM Revision ffa224a2 (ceph): mon: disallow post-hammer OSDs if there are up pre-hammer OSDs
- Force *all* OSDs to upgrade to hammer before allowing post-hammer
OSDs to join. This prevents any pre-hammer OSDs fr... - 02:30 AM Revision e7e4aefb (ceph): osd: some debug output during store version upgrades
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:30 AM Revision 14d3a2b8 (ceph): os/FileStore: fix version check
- We need to do this when we first read the version, before we
proceed with the mount. By the time we get to upgrade()... - 02:30 AM Revision 10cb507d (ceph): os/FileStore: require upgrade to hammer before moving beyond
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:30 AM Revision c0f83df4 (ceph): osd: require an upgrade to hammer first
- We must prevent a user from upgrading past hammer and then being both
unable to join the cluster and unable to downgr... - 02:30 AM Revision b474991d (ceph): ceph-objectstore-tool: require hammer upgrade
- Do not work on a store that has not been upgraded to hammer.
Signed-off-by: Sage Weil <sage@redhat.com> - 02:30 AM Revision cd4e676e (ceph): osd: drop support for pre-hammer pg metadata
- Users need to upgrade to hammer before moving beyond.
Signed-off-by: Sage Weil <sage@redhat.com> - 02:16 AM Revision 4b75baa4 (ceph): mds/MDSRank: less noisy about log_to_monitors
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:10 AM Revision 5ae3e36b (ceph): Fix compile warning unused-result
- Fix #12729
Signed-off-by: Haomai Wang <haomaiwang@gmail.com> - 01:48 AM Revision 73d4b7db (ceph): doc: fix the format of peering.rst
- Fix an incorrent number in the ordered list and some indention issue.
Make the ordered list to use '1' or 'a' for the... - 01:31 AM Bug #12717: pool's statistic data not updated after doing Cache evict operation
- hi,kefu
I think https://github.com/ceph/ceph/commit/78b1fb5e9440dcd4ef99301c3ac857385e870cf3 have resolved this.
In... - 12:56 AM Bug #12717: pool's statistic data not updated after doing Cache evict operation
- hmm, but the agent_cache_evict and agent_cache_flush is used only in cache pool, the base pool will not do this.
08/19/2015
- 11:07 PM Revision afb4db73 (ceph): Merge remote-tracking branch 'origin/wip-sam-proxy-write'
- Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Sage Weil <sage@redhat.com>
Conflicts:
src/include/ceph_fe... - 10:39 PM RADOS Bug #12687: osd thrashing + pg import/export can cause maybe_went_rw intervals to be missed
- ubuntu@teuthology:/a/samuelj-2015-08-19_00:45:42-rados-wip-sam-working-distro-basic-multi/1022696/remote
Similar p... - 09:25 PM Bug #12731 (Resolved): jerasure/gf-complete: build warnings
- ...
- 09:13 PM Revision f0ca14df (ceph): osd/PGLog: dirty_to is inclusive
- There are only two callers of mark_dirty_to who do not pass max,
and they are both in the merge_log extending tail pa... - 09:11 PM Bug #12730 (Resolved): common/Subprocess: build warnings
- ...
- 09:10 PM Bug #12729 (Resolved): ms/async: build warning
- ...
- 09:04 PM Revision f021e91e (ceph): os/RocksDBStore: log any options we pass to rocksdb
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision d94a52c9 (ceph): common/hobject: rename get_filestore_key* -> get_nibblewise_key*
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision 11946129 (ceph): os/GenericObjectMap: use bitwise hash, not nibblewise
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision 4afb0841 (ceph): common/hobject: rename get_bitreverse_key* -> get_bitwise_key*
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision c7d281b8 (ceph): os/FileStore: ensure sync() doesn't wait forever
- Avoid
2015-08-19 01:46:37.586450 7f89a53ca980 10 filestore(b.10) start_sync
2015-08-19 01:46:37.586451 7f89a53ca980 ... - 09:03 PM Revision 32bccedc (ceph): ceph_test_objectstore: test omap clone
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision 1c488370 (ceph): ceph_test_objectstore: strengthen clone test w/ new assert
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision 79da9d66 (ceph): ceph_test_objectstore: test many small writes
- This exercises the overlay code better.
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision 7ffb7b07 (ceph): test_objectstore_memstore.sh on make check
- This'll give us some coverage from ceph_test_objectstore, albeit
on the memstore backend. Upside is that this takes ... - 09:03 PM Revision 11a069db (ceph): common/hobject: make hobject_t::set_key avoid dups
- If the key matches the oid name, clear it.
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision c6b4f08f (ceph): ceph-kvstore-tool: take a db type too
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision 333f3a01 (ceph): rbd: fix bench-write
- Make each IO get a unique offset!
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision 3516a285 (ceph): os/ObjectStore: drop sync_and_flush()
- The semantics of this call were FileStore-specific; and there are no more
callers.
Signed-off-by: Sage Weil <sage@re... - 09:03 PM Revision 902557e9 (ceph): os/ObjectStore: drop flush()
- No more callers.
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision 1bb9bd9a (ceph): os/ObjectStore: drop sync()
- No callers.
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision d6b2e681 (ceph): os/ObjectStore: drop sync(Context *onsync)
- No users.
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision c33b6fbc (ceph): os/fs: add simple FS abstraction layer
- - open by handle
- xfs allocation hint
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision 5cde57b6 (ceph): os/fs: add zero / hole punch support
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision 4b9523f7 (ceph): os/fs: fix open_handle when name_to_handle not present
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision cf42c017 (ceph): ceph_test_objectstore: much cleanup, new tests
- These changes are necessary to make newstore (which is more picky
about unique collection names and objects that land... - 09:03 PM Revision 5837ec69 (ceph): osd: drop explicit sync/flush calls before umount
- Do it implicity on umount.
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision 202aa9c2 (ceph): osd: sync (no flush) on osd suicide
- Not that we even need to sync here!
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision de60159a (ceph): ceph_test_objectstore: drop useless sync_and_flush
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision 06153798 (ceph): ceph-objectstore-tool: drop explicit sync_and_flush() calls
- We implicitly sync on shutdown.
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision ff4c4cc7 (ceph): osd: drop dead suicide() code
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision b339eff0 (ceph): buffer: add prepare_iov() method
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision 6b6039fa (ceph): osd: improve osd bench
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision ec60a2cc (ceph): os/KeyValueDB: return -ENOENT if key doesn't exist
- Previous code return 0 in all cases, caller doesn't
know if the key is exists in the db.
Signed-off-by: Xiaoxi Chen ... - 09:03 PM Revision c76e3d6f (ceph): os/ObjectStore: ref count Sequencer_impl
- This allows the implementation to have a lifecycle that extends beyond
the Sequencer interface object we expose to th... - 09:03 PM Revision 8ac59239 (ceph): os/ObjectStore: do not let Sequencer::flush_commit delete the context
- Let the caller manage the lifecycle; this allows us to potentially put it
on the stack, for instance.
Signed-off-by:... - 09:03 PM Revision be6c4761 (ceph): osd: use osr for scan_snaps
- Updates should be ordered with respect to PG metadata. Not that it matters
that much since we are doing cleanup/repa... - 09:03 PM Revision e7bbafa3 (ceph): osd: explicitly use a 'meta' Sequencer
- Use an explicit Sequencer for all transactions.
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision 30e49780 (ceph): osd: fix flush_journal command
- We need to sync_and_flush *and* sync() again in order to retire any items
in the FileStore journal.
Signed-off-by: S... - 09:03 PM Revision 1ca0e7c2 (ceph): osd/osd_types: make coll_t operator== less picky
- For meta coll's do not compare pg.
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision 8b416189 (ceph): os/RocksDbStore: init pointers in ctor
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision 54421fef (ceph): vstart.sh: allow rocksdb
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision dbaf837f (ceph): common/WorkQueue: add wait()
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision 1523265d (ceph): os/ObjectStore: return const refs for oid, cid
- Should be a bit faster.
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision 8cb48147 (ceph): osd/osd_types: pg_t: add contains() helper to check objects is in a pg
- This calls the ghobject_t method that does the same, along with ps().
Signed-off-by: Sage Weil <sage@redhat.com> - 09:03 PM Revision ed5a8d76 (ceph): os/ObjectStore: pass bits to create_collection
- This informs the backend how many bits of the hash range will be covered
by the collection. Those bits are implicitl... - 09:03 PM Revision 6c54d576 (ceph): os/FileStore: clear cached fd for renamed-over object
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision ed8d3c00 (ceph): common/Formatter: add dump_object helper
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:03 PM Revision e69ac3a7 (ceph): os/KeyValueDB: add sipler get() variant
- Signed-off-by: Sage Weil <sage@redhat.com>
- 04:50 PM rgw Bug #12728 (In Progress): rgw: value of Swift API's X-Object-Manifest header is not url_decoded d...
- 04:47 PM rgw Bug #12728 (Resolved): rgw: value of Swift API's X-Object-Manifest header is not url_decoded duri...
- A try to get a dynamic large object defined with special characters in its _X-Object-Manifest_ header leads to an err...
- 03:31 PM Bug #9732: ReplicatedPG::hit_set_trim osd/ReplicatedPG.cc: 11006: FAILED assert(obc)
- sam, just want to make sure i understand your suggestion correctly, so the test would be something like:
# setup a... - 03:27 PM CephFS Bug #12727 (Duplicate): fsstress.sh failed in powercycle suite
- dup of #12710
- 03:03 PM CephFS Bug #12727 (Duplicate): fsstress.sh failed in powercycle suite
- Run: http://pulpito.ceph.com/teuthology-2015-08-18_09:06:57-powercycle-hammer-testing-basic-multi/
Job: 1020696
Log... - 03:07 PM Revision b4a4bc14 (ceph): Merge pull request #5470 from ceph/wip-newstore-sort
- osd: change sort order for hobject_t
Reviewed-by: Samuel Just <sjust@redhat.com> - 02:52 PM Bug #12717: pool's statistic data not updated after doing Cache evict operation
- huangjun,
i think if we are using the "fast" pool as the cache tier of the "slow" pool. it is supposed to be a tra... - 02:24 PM Bug #12717: pool's statistic data not updated after doing Cache evict operation
- I thinks you should put objects to pool 'fast' not 'slow', bc 'fast' is a cache pool
- 02:05 PM Bug #12717: pool's statistic data not updated after doing Cache evict operation
- hi,kefu
My ceph version is 0.80.7 maybe a bit older,
I will try it in ceph newest master branch. - 11:23 AM Bug #12717: pool's statistic data not updated after doing Cache evict operation
- huangjun,
i am trying to repeat your test:... - 10:39 AM Bug #12717 (Fix Under Review): pool's statistic data not updated after doing Cache evict operation
- https://github.com/ceph/ceph/pull/5605
- 02:47 PM Bug #12012 (In Progress): IO error on kvm/rbd with an erasure coded pool tier (after an upgrade f...
- 01:52 PM Revision 65b380f7 (ceph): Merge remote-tracking branch 'gh/wip-osd-compat-hammer' into wip-12536-...
- 01:21 PM Revision c9ec2c08 (ceph): Merge pull request #5424 from lebauce/fix_swift_user_command_line
- doc: fix command line for swift user creation
Reviewed-by: Sage Weil <sage@redhat.com> - 01:21 PM Revision 4a850b57 (ceph): Merge pull request #5048 from SUSE/wip-objecter-comment-cleanup
- Objecter.cc: fix name of _take_op_budget() function in comment
Reviewed-by: Sage Weil <sage@redhat.com> - 11:55 AM Revision d2407f29 (ceph): Merge pull request #5606 from liewegas/wip-scrub-errormsg
- osd/ReplicatedPG: fix missing whitespace in scrub error message
Reviewed-by: Kefu Chai <kchai@redhat.com> - 11:51 AM rgw Bug #12724 (Fix Under Review): rgw: wrong error handling in RGWGetObj_ObjStore_SWIFT
- Pull request: https://github.com/ceph/ceph/pull/5607.
- 11:43 AM Revision 5d873878 (ceph): rgw: fix error handling during GET on object through Swift API.
- Fixes: #12724
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 11:14 AM Stable releases Tasks #12701: hammer v0.94.4
- Joao Luis wrote:
> The stack trace appears to get stuck while dealloc'ing a string. I don't really know how this can... - 09:31 AM Stable releases Tasks #12701: hammer v0.94.4
- The stack trace appears to get stuck while dealloc'ing a string. I don't really know how this can happen.
- 10:48 AM Revision 1abf1d65 (ceph): Merge pull request #5604 from oritwas/wip-cmake-fix-rbd
- cmake: add DiffIterate.cc to librbd
Reviewed-by: Kefu Chai <kchai@redhat.com> - 10:46 AM Revision 62bfc7a1 (ceph): moved to use boost uuid implementation, based on commit 4fe89a7b14c97b2...
- Signed-off-by: Rohan Mars <code@rohanmars.com>
Reviewed-by: Casey Bodley <cbodley@redhat.com> - 10:44 AM Revision 49c3119a (ceph): Merge pull request #5433 from zqkkqz/master
- common/Thread: pthread_attr_destroy(thread_attr) when done with it
Reviewed-by: Kefu Chai <kchai@redhat.com> - 10:40 AM Revision 041642e8 (ceph): osd/ReplicatedPG: fix missing whitespace in scrub error message
- Signed-off-by: Sage Weil <sage@redhat.com>
- 10:29 AM Revision a0ae5bcf (ceph): Merge pull request #5538 from ceph/wip-selinux-improve-update
- selinux: Relabel files if and only if the policy version changed
Reviewed-by: Milan Broz <mbroz@redhat.com> - 09:26 AM Revision e4cff034 (ceph): ceph.spec.in: Require awk for selinux post script
- Signed-off-by: Boris Ranto <branto@redhat.com>
- 09:17 AM Revision 9038488b (ceph): selinux: Relabel files if and only if the policy version changed
- Currently, the ceph files are being relabelled every time the package is
rebuilt. Fix this by checking the policy ver... - 09:00 AM Revision a4d5ae28 (ceph): Merge pull request #5596 from mbroz/wip-selinux-systemd
- Update ceph-selinux rpm spec
Reviewed-by: Boris Ranto <branto@redhat.com> - 08:56 AM Revision aa50321a (ceph): cmake: add DiffIterate.cc to librbd
- Signed-off-by: Orit Wasserman <owasserm@redhat.com>
- 05:52 AM Bug #10399: kvm die with assert(m_seed < old_pg_num)
- Sage Weil wrote:
> is this teh same as #10543 ?
yes, I think so.
It looks like very similar on stack trace and...
08/18/2015
- 11:28 PM Stable releases Tasks #12701: hammer v0.94.4
- http://pulpito.ceph.com/loic-2015-08-15_22:14:11-upgrade:hammer-hammer-backports---basic-multi/
Sage: every single... - 09:55 PM Bug #12718 (Resolved): ceph-disk always calls restorecon
- 12:52 PM Bug #12718: ceph-disk always calls restorecon
- Interesting, I have if which('restorecon') here locally.
Some wrong patch was merged then... - 12:46 PM Bug #12718 (Fix Under Review): ceph-disk always calls restorecon
- https://github.com/ceph/ceph/pull/5597
- 11:42 AM Bug #12718: ceph-disk always calls restorecon
- 11:42 AM Bug #12718 (Resolved): ceph-disk always calls restorecon
- restorecon must be called only if it's available. It fails all ceph-deploy run on operating systems that do not have ...
- 09:01 PM Bug #12469: RadosModel.h: read returned error code -2 (hammer)
- It's a bug with ceph-objectstore-tool that is not a blocker for a hammer release.
- 08:51 PM Bug #12469: RadosModel.h: read returned error code -2 (hammer)
- This appears to be caused by bad past_intervals on PG 1.16, resulting from the ceph-objectstore-tool import/export:
... - 08:55 PM Bug #12536 (Pending Backport): "FAILED assert(!log.null() || olog.tail == eversion_t())"
- 08:55 PM Backport #12571 (In Progress): "FAILED assert(!log.null() || olog.tail == eversion_t())"
- 08:03 PM Revision a998b466 (ceph): Merge pull request #536 from ceph/wip-11881-multipath
- ceph-deploy: hints for OpenStack
- 07:46 PM Revision 53a714ee (ceph): update the list of supported distributions
- CentOS 6 and Ubuntu 12.04 are not support after Infernalis.
CentOS 7 is supported starting with Infernalis.
Signed-o... - 07:35 PM Revision 21c01c78 (ceph): Merge pull request #537 from dachary/wip-supported
- update the list of supported distributions
Reviewed-by: Sage Weil <sage@redhat.com> - 07:29 PM Revision e252c47f (ceph): update the list of supported distributions
- CentOS 6 and Ubuntu 12.04 are not support after Infernalis.
CentOS 7 is supported starting with Infernalis.
Signed-o... - 07:28 PM rgw Bug #12712 (In Progress): civetweb frontend: sent bytes are not properly accounted
- 07:27 PM rgw Bug #12713 (In Progress): civetweb frontend: response is buffered in memory if content length is ...
- 07:26 PM rgw Bug #12724 (In Progress): rgw: wrong error handling in RGWGetObj_ObjStore_SWIFT
- 07:26 PM rgw Bug #12724 (Resolved): rgw: wrong error handling in RGWGetObj_ObjStore_SWIFT
- _RGWGetObj_ObjStore_SWIFT::send_response_data()_ method wrongly handles errors. It generates HTTP response with heade...
- 07:19 PM Revision cf8a360c (ceph): Merge pull request #5600 from liewegas/wip-osd-bind
- doc: ceph-osd binds to up to 4 ports, not 3
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 06:33 PM Revision 3da86724 (ceph): ceph-deploy: hints for OpenStack
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 06:31 PM Revision 5d109e97 (ceph): common: support perf counter (for unhealthy workers) on CephContext
- Signed-off-by: Guang Yang <yguang@yahoo-inc.com>
- 06:31 PM Revision de0b66ab (ceph): test: add test for the perf counter of CephContext
- Signed-off-by: Guang Yang <yguang@yahoo-inc.com>
- 06:31 PM Revision 15a3e866 (ceph): rgw: enable perf counter for unhealthy workers
- Fixes: #12666
Signed-off-by: Guang Yang <yguang@yahoo-inc.com> - 06:31 PM Revision 5fa03e99 (ceph): osd: expose the number of unhealthy threads from heartbeat map
- Signed-off-by: Guang Yang <yguang@yahoo-inc.com>
- 06:25 PM Revision 5390072c (ceph): ReplicatedPG: block writes on promote of rollback snap promotion
- Signed-off-by: Samuel Just <sjust@redhat.com>
- 06:25 PM Revision 79f310ac (ceph): ReplicatedPG: treat object as degraded until on_global_recover
- Otherwise, an op might slip by between on_local_recover and
on_global_recover.
Signed-off-by: Samuel Just <sjust@red... - 06:25 PM Revision 9b233929 (ceph): ReplicatedPG::cancel_pull: also finish_degraded_object
- We might have degraded object waiters on a pull.
Signed-off-by: Samuel Just <sjust@redhat.com> - 06:25 PM Revision da68bb37 (ceph): osd: purge the object from the cache when proxying and not promoting th...
- When proxying the write/cache op, if it is decided to not promote the
object, need to purge it from the object_contex... - 06:25 PM Revision b7225a48 (ceph): osd: copy the reqids even if the object is deleted during promotion
- If the object is deleted on the base tier, and the reqids are not copied
during promotion, this again leads to the 'o... - 06:25 PM Revision 61c86a12 (ceph): test: change the test_tiering test case accordingly for proxy write
- When an object is first created, it's proxied to base tier, need to
change the behavior of the test_tiering test case... - 06:25 PM Revision e7107ce0 (ceph): Revert "osd: set the blocked_by relationship when rolling back to a deg...
- This reverts commit 48dacd75566f64cb89f991ac4d05e2e50c4e73e0.
- 06:25 PM Revision 7d4f695b (ceph): RadosModel: send writes before and after rollback to verify ordering
- Signed-off-by: Samuel Just <sjust@redhat.com>
- 06:25 PM Revision 35af63b9 (ceph): ReplicatedPG: enforce write ordering on rollback
- Previously, rollback ops could reorder w.r.t other writes due to waiting
on degraded snaps other than head. To fix t... - 06:25 PM Revision 58dd21c0 (ceph): osd: add reqid in MOSDOp
- The cache tier needs to set the reqid explicitly to the original reqid
from the client when proxying the write op to ... - 06:25 PM Revision ff658bd8 (ceph): Objecter: optionally setting the reqid in the mutate interface
- The cache tier needs to set the reqid to the original reqid from client
when proxying the write op.
Signed-off-by: Z... - 06:25 PM Revision bf5d51e2 (ceph): osd: explicitly set the reqid when proxying the write op
- This is needed as in the following scenario:
- Client sends 3 writes and a read on the same object to base tier
- Set... - 06:25 PM Revision 626c5692 (ceph): osd: force promote for object overwrites on a ec base pool
- When doing object overwrites on the ec base pool, the write op can't be
proxied. Always force promotion in this case.... - 06:25 PM Revision 98fff96b (ceph): osd: rename SKIP_PROMOTE to SKIP_HANDLE_CACHE
- To comply with what it really means.
Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com> - 06:25 PM Revision fa5b6e6d (ceph): osd: skip promotion when proxying a delete op
- When the object is deleted, there is no need to promote it.
Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com> - 06:25 PM Revision aff8aa1a (ceph): osd: set the blocked_by relationship when rolling back to a degraded
- object
In a scenario like below:
- A rollback op comes in, and is enqueued.
- Several other ops on the same object c... - 06:25 PM Revision d836a645 (ceph): osd/ReplicatedPG: minor updates on proxy write
- Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com>
- 06:25 PM Revision 7e27e615 (ceph): osd/ReplicatedPG: add helper function check_for_promote
- This function is used to check if we need to initiate a promotion in
maybe_handle_cache.
Signed-off-by: Zhiqiang Wan... - 06:25 PM Revision ab39e03d (ceph): osd/ReplicatedPG: don't check order in finish_proxy_write
- When proxy write comes back from base tier, the write op may or may not
sit at the front of the list.
Signed-off-by:... - 06:25 PM Revision cb9390dc (ceph): osd/ReplicatedPG: add the proxy write feature bit support
- For older versions which doesn't support proxy write, doing promote.
Otherwise, we can proxy the write.
Signed-off-b... - 06:25 PM Revision b9ec7e64 (ceph): osd: add proxy write perf counter
- Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com>
- 06:25 PM Revision 0d7759f3 (ceph): osd: turn on proxy write feature bit by default
- Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com>
- 06:25 PM Revision c01d20b6 (ceph): osd/ReplicatedPG: set the RWORDERED flag for the promote copy-from op
- Set the RWORDERED flag when doing promote copy-from op. This is in case
there are proxy writes in flight.
Signed-off... - 06:25 PM Revision 1099ec23 (ceph): osd/osd_types: add min_write_recency_for_promote in pg_pool_t
- This field stands for the minimum number of hit sets to check before
promote on write.
Signed-off-by: Zhiqiang Wang ... - 06:25 PM Revision 3d5300bd (ceph): osd/ReplicatedPG: promote on 2nd write
- If min_write_recency_for_promote is
- 0: Promote when there is a write.
- 1: Check if the object is in current hit se... - 06:25 PM Revision 257c851e (ceph): mon: add osd pool set/get for min_write_recency_for_promote
- Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com>
Conflicts:
src/mon/MonCommands.h
src/mon/OSDMonitor.cc - 06:25 PM Revision f8b3a409 (ceph): osd: tiering: add proxy write support
- Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com>
Conflicts:
src/osd/ReplicatedPG.cc - 06:25 PM Revision 772617bf (ceph): osd/ReplicatedPG: remove the peer_type assertion in eval_repop
- When moving to proxy write, this assertion doesn't hold any more.
Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.... - 06:25 PM Revision 90e5f410 (ceph): osd: tiering: use proxy write in writeback mode
- Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com>
Conflicts:
src/osd/ReplicatedPG.cc - 06:18 PM Bug #12681: tcmalloc doesn't release memory back to system
- No recovery ongoing, I just started the service and put load for 1~2 hours, the run into this situation. There is no ...
- 06:00 PM Bug #12681 (Need More Info): tcmalloc doesn't release memory back to system
- 06:00 PM Bug #12681: tcmalloc doesn't release memory back to system
- Was any recovery taking place? This could be this bug: https://github.com/ceph/ceph/pull/5451
- 06:02 PM Bug #10399: kvm die with assert(m_seed < old_pg_num)
- is this teh same as #10543 ?
- 05:59 PM Bug #12655 (Rejected): unable to bind to IP:PORT on any port in range 6800-7300: (98) Address alr...
- https://github.com/ceph/ceph/pull/5600
- 05:57 PM Bug #12655: unable to bind to IP:PORT on any port in range 6800-7300: (98) Address already in us
- karan singh wrote:
> Thanks sage , setting a larger port range fixes this.
>
> Before closing this ticket could y... - 05:57 PM Revision 792e24bd (ceph): doc: ceph-osd binds to up to 4 ports, not 3
- Signed-off-by: Sage Weil <sage@redhat.com>
- 05:17 PM Bug #12722 (Resolved): osd: OSD::do_mon_report stuck at acquiring osd_lock (more than 10mins) and...
- Copy the discussion from mailing list as following:...
- 05:11 PM rgw Feature #12608 (In Progress): rgw: multisite: per-shard full sync stage
- 05:11 PM rgw Feature #12607 (Fix Under Review): rgw:multisite: build metadata indexes for full sync
- 05:11 PM rgw Feature #12124 (Fix Under Review): radosgw-admin command to configure realm, period
- 05:11 PM rgw Feature #11935 (Fix Under Review): rgw: create immutable metadata objects
- 03:54 PM Bug #12638 (Fix Under Review): build_incremental() could take 40678 ms to finish
- https://github.com/ceph/ceph/pull/5524
- 03:52 PM Revision f2cbbdd7 (ceph): Merge remote-tracking branch 'gh/next'
- 03:36 PM Bug #12720 (Can't reproduce): probably import/export related: osd/PG.cc: 6893: FAILED assert(pg->...
- 2015-08-17 15:39:51.750439 7facf0b6c700 5 osd.4 pg_epoch: 309 pg[0.30(unlocked)] enter Initial
2015-08-17 15:39:51.... - 03:08 PM Revision ec4e22b7 (ceph): rgw: Remove useless code in calc_hmac_sha1()
- Hex conversion from buf is not used at all.
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com> - 02:38 PM Feature #12601 (In Progress): cmake
- 02:36 PM Feature #12719 (In Progress): osd: improved memstore
- 02:36 PM Feature #12719 (Resolved): osd: improved memstore
- 01:05 PM Revision 7e02fb1f (ceph): Merge pull request #5597 from dachary/wip-12718-restorecon
- ceph-disk: only call restorecon when available
Reviewed-by: Boris Ranto <branto@redhat.com> - 12:43 PM Revision 3aab146b (ceph): ceph-disk: only call restorecon when available
- 9db80da12803d42bb676d67f37442c0c54d83448 added an unconditional call to
restorecon after mounting the filesystem. It ... - 12:25 PM Revision 28d3c4c9 (ceph): Do not use hardcoded paths in spec post section for selinux commands.
- Signed-off-by: Milan Broz <mbroz@redhat.com>
- 12:25 PM Revision 14da7e20 (ceph): Update rpm spec to properly restart systemd ceph.target.
- If we are using systemd, ceph target should be used.
Also fail to restart service should not stop policy to load.
Si... - 12:25 PM Revision fa2203e9 (ceph): Add dependence to selinux-policy-base to fix installation in kickstart.
- Signed-off-by: Milan Broz <mbroz@redhat.com>
- 12:07 PM CephFS Bug #12710: fsstress.sh fails
- commit:47519365484056e1731cac54cce835332d258121
- 07:28 AM CephFS Bug #12710 (Fix Under Review): fsstress.sh fails
- https://github.com/ceph/ceph/pull/5595
- 12:07 PM Revision 47519365 (ceph): Merge pull request #5595 from ceph/wip-12710
- qa/fsstress.sh: fix 'cp not writing through dangling symlink'
Reviewed-by: Greg Farnum <gfarnum@redhat.com> - 12:01 PM CephFS Bug #12711: mds get damaged
- merged in commit:3cfb7e4ccc08a67ceec73ee684049320c75e9bb2
- 06:59 AM CephFS Bug #12711 (Fix Under Review): mds get damaged
- https://github.com/ceph/ceph/pull/5594
- 12:01 PM Revision 3cfb7e4c (ceph): Merge pull request #5594 from ceph/wip-12711
- mds: properly set client incarnation
Reviewed-by: Greg Farnum <gfarnum@redhat.com> - 11:43 AM Revision 697101e4 (ceph): Merge pull request #5589 from ceph/hammer-12709
- Workunits : fs/misc/chmod.sh : Include ACL characters in permission check
Reviewed-by: Greg Farnum <gfarnum@redhat.com> - 10:46 AM rgw Feature #12714: RGW : setting max number of buckets for user via ceph.conf option
- Warren Wang wrote:
> I would suggest that this go one step further, and all default RGW quotas be settable in ceph.c... - 06:54 AM rgw Feature #12714: RGW : setting max number of buckets for user via ceph.conf option
- I would suggest that this go one step further, and all default RGW quotas be settable in ceph.conf.It is difficult to...
- 09:57 AM Bug #12717 (Duplicate): pool's statistic data not updated after doing Cache evict operation
- We setup a cache tier and set target_max_bytes to 10240,
After put 2 objects(each 40MB) by 'rados put' to the cache ... - 09:38 AM Revision 641169f2 (ceph): Merge pull request #531 from ceph/wip-mds-debug
- tasks/kcephfs: enable MDS debug
- 09:03 AM devops Feature #11881: ceph-disk support for multipath
- > Once I am done with the changes, I will commit it in my github. Probably you can have a look and advice.
Extreme... - 09:00 AM devops Feature #11881: ceph-disk support for multipath
- > the parttuid links are not very reliable for multipath device as these are usually hotplug. We have seen the partty...
- 08:55 AM devops Feature #11881: ceph-disk support for multipath
- > The holders based partition detection may not hold true in case of dm-crypt on mpath (or any other layering)
Yes... - 08:38 AM devops Feature #11881: ceph-disk support for multipath
- Loic,
couple of issue we have seen with multipath experiments.
1. the parttuid links are not very reliable for m... - 08:57 AM Revision f92b27e1 (ceph): tools/cephfs: account for striping in size estimation
- Signed-off-by: John Spray <john.spray@redhat.com>
- 08:49 AM Revision 4400aa85 (ceph): client: track cap flushing TIDs for each session
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 08:49 AM Revision 3a0c14e8 (ceph): client: make wait_sync_caps(uint64_t) check per-session flush TIDs
- Now we have data structure to track per-session flush TIDs. It's
easy to check if all cap flushes before a TID are co... - 08:49 AM Revision 36b0e3aa (ceph): client: include oldest pending caps flushing in cap message
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 08:49 AM Revision 0ca7a49b (ceph): mds: record completed cap flushes in session map
- This information is for detecting duplicated cap flushes when MDS
failovers. We only record completed cap flushes for... - 08:49 AM Revision 2d6ab6e5 (ceph): mds: record cap flush tid in log event
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 08:49 AM Revision aa7e7965 (ceph): mds: detect completed cap flush
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 08:49 AM Revision a1d14b74 (ceph): client: re-send flushing caps (which are revoked) in reconnect stage
- if flushing caps were revoked, we should re-send the cap flush in
client reconnect stage. This guarantees that MDS pr... - 08:49 AM Revision 976b449b (ceph): mds: re-issue caps when recovering MDS becomes active
- this makes sure the recovering MDS handle cap flushes re-sent by
clients before re-issuing caps.
Signed-off-by: Yan,... - 08:49 AM Revision 00c56c6c (ceph): mds: warn if client does not advance its oldest flush tid
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 08:49 AM Revision dafa7d27 (ceph): client: don't change flush TID when re-send cap flush
- Client::wait_sync_caps requires stable TIDs for flushing caps.
Signed-off-by: Yan, Zheng <zyan@redhat.com> - 08:49 AM Revision 06cf2243 (ceph): client: globally unique cap flush TID
- Globally unique TID will simplify MDS code that dectects duplicated
cap flushes.
Signed-off-by: Yan, Zheng <zyan@red... - 08:12 AM Revision c8a83cd1 (ceph): client: track flush TIDs for all pending flushing caps
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 07:26 AM Revision 479f2a76 (ceph): qa/fsstress.sh: fix 'cp not writing through dangling symlink'
- On some test machines, /usr/lib/ltp/testcases/bin/fsstress is
dangling symlink. 'cp -f' is impotent in this case.
Fi... - 06:57 AM Revision 6387ec9f (ceph): mds: properly set client incarnation
- Fixes: #12711
Signed-off-by: Yan, Zheng <zyan@redhat.com>
08/17/2015
- 11:52 PM devops Feature #11881: ceph-disk support for multipath
- Running the ceph-deploy suite will use ceph-disk on various distributions...
- 10:31 PM Revision 2811ef21 (ceph): Adds radosbenchsweep task
- This tasks executes a parameter sweep of radosbench by varying the size
of the pool (# of replicas), the number of os... - 09:30 PM Bug #12716 (Can't reproduce): Cluster health_warn stuck on active+remapped
- I ran a ceph osd reweight-by-utilization last week and partway through had a network interruption. After the network ...
- 09:11 PM rgw Feature #12714: RGW : setting max number of buckets for user via ceph.conf option
- Ideally we would backport this to Hammer as well.
- 07:00 PM rgw Feature #12714 (Resolved): RGW : setting max number of buckets for user via ceph.conf option
- RGW : setting max number of buckets for user via ceph.conf option
We can change the user's total max buckets by u... - 08:39 PM rgw Backport #12591: rgw: create a tool for orphaned objects cleanup
- For what it's worth, the upstream master PR (https://github.com/ceph/ceph/pull/5109) cherry-picks cleanly onto v0.94....
- 08:23 PM Revision 87f6b73e (ceph): Add new cls_numops class for numeric operations
- Signed-off-by: Joaquim Rocha <joaquim.rocha@cern.ch>
- 08:23 PM Revision d17f158e (ceph): cls_numops: Add cls_numops client
- Signed-off-by: Joaquim Rocha <joaquim.rocha@cern.ch>
- 08:23 PM Revision d742e797 (ceph): tests: Add unit tests for CLS numops class
- Signed-off-by: Joaquim Rocha <joaquim.rocha@cern.ch>
- 08:23 PM Revision 63a0cf29 (ceph): qa/workunits/cls: add workunit for cls_numops tests
- Signed-off-by: Joaquim Rocha <joaquim.rocha@cern.ch>
- 08:14 PM CephFS Bug #12715 (Resolved): "[ERR] bad backtrace on dir ino 600" in cluster log"
- Run: http://pulpito.ceph.com/teuthology-2015-08-14_16:56:20-upgrade:firefly-x-hammer-distro-basic-multi/
Job: 101471... - 07:59 PM Revision c3de0aff (ceph): Merge pull request #5593 from ceph/revert-4927-snapset-obc
- Revert "osd/ReplicatedPG: snapset is not persisted"
Reviewed-by: Samuel Just <sjust@redhat.com> - 07:58 PM Revision 0ba2e145 (ceph): Revert "osd/ReplicatedPG: snapset is not persisted"
- 07:00 PM rbd Feature #12698 (In Progress): librbd: write maintenance operations to journal
- 06:19 PM Revision 989820a9 (ceph): Merge pull request #5577 from oritwas/wip-next-12363
- rgw: we should not overide Swift sent content type
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 05:28 PM Bug #12536: "FAILED assert(!log.null() || olog.tail == eversion_t())"
- Note: https://github.com/ceph/ceph-qa-suite/pull/525 needs to be merged after this ticked is fixed/backported.
See #... - 05:27 PM Bug #12410: OSDMonitor::preprocess_get_osdmap: must send the last map as well
- Note: https://github.com/ceph/ceph-qa-suite/pull/525 needs to be merged after this ticket is fixed/backported.
See #... - 05:05 PM Revision 8f66ecb3 (ceph): Added singleton hammer-x test to address #12625
- Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
- 04:56 PM Revision 48a8b235 (ceph): Merge pull request #533 from ceph/revert-525-wip-12625-master
- Revert "Added singleton hammer-x test to address #12625"
- 04:56 PM Revision 8bfb1c91 (ceph): Revert "Added singleton hammer-x test to address #12625"
- 04:50 PM Revision 4764e7f8 (ceph): Merge pull request #525 from ceph/wip-12625-master
- Added singleton hammer-x test to address #12625
Reviewed-by: Samuel Just <sjust@redhat.com> - 03:47 PM rbd Bug #12708 (Fix Under Review): Even if RBD export-diff failure, will still create a local file
- 02:38 PM rbd Bug #12708 (Pending Backport): Even if RBD export-diff failure, will still create a local file
- https://github.com/ceph/ceph/pull/5587
- 02:36 AM rbd Bug #12708 (Resolved): Even if RBD export-diff failure, will still create a local file
- root@wfnd95:/caibo# clear
#there is no file in here
root@wfnd95:/caibo# ls
#i try to use export-diff,it must be fa... - 03:42 PM Revision d43d6844 (ceph): Merge pull request #5558 from s09816/rbd-fix
- rbd:improve the error handle of rbd,check the return value.
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Review... - 03:24 PM rgw Bug #12713 (Resolved): civetweb frontend: response is buffered in memory if content length is not...
- If content length is not explicitly specified by upper layer through a call to _RGWMongoose::send_content_length()_, ...
- 03:15 PM rgw Bug #12712 (Closed): civetweb frontend: sent bytes are not properly accounted
- _RGWClientIO_ class provides mechanism for accounting bytes transmitted to a client. Each call to _RGWClientIO::write...
- 02:30 PM Revision 3540fb9f (ceph): osdc/Objecter: restart listing this PG if sort order changes
- If the cluster sort order changes mid-way through our listing, our
cursor within this pg is meaningless and we need t... - 02:30 PM Revision b18558b7 (ceph): osd/OSDMap: test_flag returns bool
- Otherwise comparing the result to a bool will promote the bool to
an int and fail the comparison.
Signed-off-by: Sag... - 01:50 PM Revision fc61fd79 (ceph): osd/osd_types: add pg_missing_t::resort() method
- Set/change sort order as needed.
Signed-off-by: Sage Weil <sage@redhat.com> - 01:50 PM Revision 35c19700 (ceph): osd/ReplicatedPG: fix missing set sort order on [N]PGLS
- This is the only thing that cares about the missing set sort order,
so only adjust it here as needed.
Signed-off-by:... - 01:50 PM Revision a5e27de5 (ceph): osd: refuse to boot if SORTBITWISE not set but backend cannot sort nibb...
- This prevenents OSDs with new backends from starting if the
cluster is still using the legacy sort order.
Signed-off... - 01:50 PM Revision 517921f9 (ceph): osd/osd_types: make pg_missing_t sort order dynamic
- Signed-off-by: Sage Weil <sage@redhat.com>
- 01:06 PM CephFS Bug #12711 (Resolved): mds get damaged
- http://pulpito.ceph.com/teuthology-2015-08-10_23:08:02-kcephfs-master-testing-basic-multi/1010323/
It's easy to re... - 01:00 PM Revision a5e0d8a3 (ceph): Merge pull request #5586 from ceph/wip-jni-loader
- wip-jni-loader
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> - 11:18 AM Linux kernel client Bug #12697: [RBD] rbd_dev_id_put causes system crash
- Ilya Dryomov wrote:
> I think the problem is that the reference dropped by put_device() caused by delayed fput was t... - 10:44 AM Linux kernel client Bug #12697: [RBD] rbd_dev_id_put causes system crash
- And the rest suggests a use-after-free on struct rbd_device.
- 10:10 AM Linux kernel client Bug #12697 (Need More Info): [RBD] rbd_dev_id_put causes system crash
- 10:10 AM Linux kernel client Bug #12697: [RBD] rbd_dev_id_put causes system crash
- I think the problem is that the reference dropped by put_device() caused by delayed fput was the last one. Unless I'...
- 09:57 AM Cleanup #12706: ceph-disk should use /sys/dev/block/maj:min/partition
- https://github.com/ceph/ceph/pull/5591
- 08:06 AM CephFS Bug #12710 (Resolved): fsstress.sh fails
- see quite a lot fsstress failures. one of them is http://qa-proxy.ceph.com/teuthology/teuthology-2015-08-10_23:04:02-...
- 07:50 AM CephFS Bug #11783: protocol: flushing caps on MDS restart can go bad
- see this again http://pulpito.ceph.com/teuthology-2015-08-11_23:04:02-fs-next---basic-multi/1011375
- 06:42 AM Revision 81a311a7 (ceph): Workunits : fs/misc/chmod.sh : Include ACL characters in permission check.
- Signed-off-by: Yazen Ghannam <yazen.ghannam@linaro.org>
(cherry picked from commit d3dbfffefb0ae53583350f53258dc90267... - 05:35 AM Revision 97c66e3f (ceph): erasure-code: Update ISA-L to 2.14
- Ceph requires all .s files to annotated to assert that stack
should not be executable. see bug #10114, f60da6b and 06... - 05:20 AM Revision 7cf4b3a8 (ceph): Merge pull request #5588 from zhouyuan/isal_yasm_fix
- configure: Fix checking for yasm compatibility
Reviewed-by: Kefu Chai <kchai@redhat.com> - 03:49 AM CephFS Bug #12709 (Resolved): hammer chmod.sh fails
- http://magna002.ceph.redhat.com/teuthology-2015-08-13_18:04:02-fs-hammer---basic-magna/173814/teuthology.log...
- 02:48 AM Revision 0bb57f10 (ceph): configure: Fix checking for yasm compability
- Fix typo when checking yasm
Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>
08/16/2015
- 10:15 PM Revision 5afa21d6 (ceph): java: search for JNI bits in common dirs
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 10:15 PM Revision 2743cc40 (ceph): java: add libcommon to deps
- Why do indirect dependencies seem to work randomly:
undefined symbol: _ZN4ceph3log3Log12create_entryEii
Signed-of... - 04:53 PM Bug #12707 (Resolved): ceph-detect-init fails on centos7
- After a *ceph-deploy install --testing* ceph-detect-init fails with:...
- 04:41 PM devops Feature #11881: ceph-disk support for multipath
- To manually try from sources:...
- 12:50 AM devops Feature #11881 (Fix Under Review): ceph-disk support for multipath
- 04:28 PM devops Bug #12559: centos7 packages are not installable for v0.94.1
- Also with *ceph-deploy install --testing try* manually:...
- 10:19 AM Bug #9732 (Fix Under Review): ReplicatedPG::hit_set_trim osd/ReplicatedPG.cc: 11006: FAILED asser...
- https://github.com/ceph/ceph/pull/5585
08/15/2015
- 08:40 PM Cleanup #12706 (Resolved): ceph-disk should use /sys/dev/block/maj:min/partition
- ceph-disk should use /sys/dev/block/maj:min/partition to figure out if a device is a partition instead of iterating o...
- 07:17 PM devops Bug #12702 (Rejected): multiple bugs in rcceph script
- dupe!
- 06:07 PM devops Bug #12702 (Pending Backport): multiple bugs in rcceph script
- 06:07 PM devops Bug #12702: multiple bugs in rcceph script
- Master PR: https://github.com/ceph/ceph/pull/5002 (merged)
- 06:06 PM devops Bug #12702 (Rejected): multiple bugs in rcceph script
- Owen Synge reports that he fixed the following bugs in rcceph script:
* starts OSDs even if mon daemons are not pr... - 07:07 PM devops Backport #12704 (Rejected): multiple bugs in rcceph script
- systemd/ceph is the sysvinit compatibility script for systemd enablement -> the script is present in hammer, but reje...
- 06:09 PM devops Backport #12704 (Rejected): multiple bugs in rcceph script
- 07:06 PM devops Backport #12703 (Rejected): multiple bugs in rcceph script
- rcceph is the sysvinit compatibility script for systemd enablement: it is not present in firefly
- 06:08 PM devops Backport #12703 (Rejected): multiple bugs in rcceph script
- 05:15 PM devops Feature #11881: ceph-disk support for multipath
- The scsi_debug kernel module on Ubuntu 14.04 is with the linux-image-extra-3.13.0-61-generic package
- 05:07 PM devops Bug #12043 (Resolved): rgw/logrotate.conf calls service with wrong init script name
- 05:04 PM devops Backport #12503 (Resolved): rgw/logrotate.conf calls service with wrong init script name
- 05:03 PM devops Backport #12045 (Rejected): rgw/logrotate.conf calls service with wrong init script name
- The PR from #12503 has been merged.
- 11:04 AM Stable releases Tasks #12701 (Resolved): hammer v0.94.4
- h3. Workflow
* "Preparing the release":http://ceph.com/docs/master/dev/development-workflow/#preparing-a-new-relea... - 03:56 AM Revision 128f5a25 (ceph): Merge pull request #5572 from xinxinsh/wip-rados-bench-error
- fix print error of rados bench
Reviewed-by: Kefu Chai <kchai@redhat.com> - 03:19 AM rbd Bug #12700: incomplete rbd delete(command:rbd rm) damaged the rbd block
- This is a bug or was it so ?
And if this is a bug,i want to fix it myself
because there is many way to prevent this... - 03:15 AM rbd Bug #12700 (Duplicate): incomplete rbd delete(command:rbd rm) damaged the rbd block
- #When I try to delete a rbd block(the size is 20G),I use the command
rbd -p rbd rm myblock
#When this command is ... - 02:32 AM Revision af0ebeeb (ceph): rbd:improve the error handle of rbd,check the return value.
- Signed-off-by: s09816 <shi.lu@h3c.com>
08/14/2015
- 10:56 PM Revision 513cf0ae (ceph): Adds task to wrap the aver utility
- Aver is a utility that allows to test high-level assertions over
benchmarking output. This task allows to specify ass... - 10:50 PM Revision ae54a9fa (ceph): Merge pull request #5443 from ceph/wip-wrlock
- cleanup: remove all traces of rados 'lock' operations
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 10:07 PM devops Feature #11881: ceph-disk support for multipath
- On an OpenStack instance with a /dev/vdb disk did something like:...
- 02:35 PM devops Feature #11881: ceph-disk support for multipath
- To find out if mpatha as found in /dev/mapper is managed by multipath:...
- 01:32 PM devops Feature #11881: ceph-disk support for multipath
- Would it be an option to have ceph-disk prepare /dev/mapper/mpath create a file /var/lib/ceph/osd/ceph-0/multipath to...
- 09:18 PM Revision 87fdba9c (ceph): Merge remote-tracking branch 'origin/next'
- 09:16 PM Revision ee909dcc (ceph): Merge pull request #5537 from s09816/master
- rbd:modify the log of purging snaps so that it is more appropriate
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 09:03 PM Revision 1f2f27f1 (ceph): Merge pull request #5560 from solesoul1127/master
- rbd:Check the dest image name, if it is empty string, return error and give a message
Reviewed-by: Josh Durgin <jdur... - 06:24 PM Revision dceae70e (ceph): Merge pull request #514 from ceph/wip-cd-add-branch-install
- - Use branch from config for ceph-deploy testing
Reviewed-by: Travis Rhoden <trhoden@redhat.com> - 06:18 PM Bug #9732: ReplicatedPG::hit_set_trim osd/ReplicatedPG.cc: 11006: FAILED assert(obc)
- I think it needs to be cluster wide, and it just means that new and old can't coexist once you have flipped the switc...
- 06:10 PM Bug #9732: ReplicatedPG::hit_set_trim osd/ReplicatedPG.cc: 11006: FAILED assert(obc)
- Kefu Chai wrote:
> sam reminded me:
>
> * we don't want to convert an existing OSD automatically, from localtime ... - 04:10 PM Bug #12665: osd/ReplicatedPG.cc: 2706: FAILED assert(p != snapset.clones.end())
- Hi Sage,
No, untill now we've never used cache tiering.
This cluster is pretty "old" and went thru upgrades since... - 03:29 PM rbd Feature #12698 (Resolved): librbd: write maintenance operations to journal
- All operations that mutate the image (resize, snap create/remove, rollback, flatten, etc) should be recorded in the j...
- 03:01 PM Revision d0d67277 (ceph): common/syncfs: fall back to sync(2) if syncfs(2) not available
- Fixes: #12512
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 08210d6a85527b0759f6dc4411f742bc... - 03:01 PM Revision 40494c6e (ceph): sync_filesystem.h: fix unreachable code
- Fix for:
CID 1264460 (#1 of 1): Structurally dead code (UNREACHABLE)
unreachable: This code cannot be reached: sync(... - 02:54 PM Revision 3dbbc86a (ceph): mon, os: check the result of sync_filesystem.
- Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
(cherry picked from commit 27cb78bd3b2dde40c70de245769d1c0ec252533b) - 02:53 PM Revision 7fa6fdc6 (ceph): common: Don't call ioctl(BTRFS_IOC_SYNC) in sync_filesystem.
- From the manual of syncfs, it first appeared in Linux 2.6.39. At this
point, btrfs didn't as a stable production. So ... - 02:53 PM Revision 9f15ed5b (ceph): common: Directly return the result of syncfs().
- In commit 808c644248e4, it will try sync() if syncfs() return error.
No evidence prove this way can work. And sync() ... - 02:23 PM Bug #12638: build_incremental() could take 40678 ms to finish
- > Do we have any more details about what's happening here
per Tupper, this happens when an OSD host restarts.
>... - 01:28 PM Bug #12638: build_incremental() could take 40678 ms to finish
- These numbers look very strange to me. Do we have any more details about what's happening here, or why things might b...
- 02:22 PM Revision 4eaa9ea1 (ceph): fix print error of rados bench
- Total time run: 12.279167
Total writes made: 92
Write size: 4194304
Bandwidth (MB/sec): ... - 12:31 PM Revision 79242319 (ceph): tests: tiering agent and proxy read
- Verify that an object promoted to a cache tier because of a proxy read
is evicted as expected.
http://tracker.ceph.c... - 12:31 PM Revision e1f58feb (ceph): osd: trigger the cache agent after a promotion
- When a proxy read happens, the object promotion is done in parallel. The
agent_choose_mode function must be called to... - 10:59 AM Linux kernel client Bug #12697 (Resolved): [RBD] rbd_dev_id_put causes system crash
- Today we randomly meet system crash two times. Both crashes are related to rbd_dev_id_put. We are still not sure how ...
- 10:07 AM Bug #10399: kvm die with assert(m_seed < old_pg_num)
- I have the same issue with ceph-0.94.2
Ubuntu version: 12.04.5
kernel version: 3.13.0-35
Qemu version: 2.0.0
8 ... - 08:48 AM CephFS Bug #12676 (Fix Under Review): MDSMap assertion in MDCache::trim (multimds)
- 08:47 AM CephFS Bug #12676: MDSMap assertion in MDCache::trim (multimds)
- https://github.com/ceph/ceph/pull/5583
- 08:46 AM Revision 333cd7d6 (ceph): mds: check mds up before calling MDSMap::get_mds_info
- Fixes: #12676
Signed-off-by: Yan, Zheng <zyan@redhat.com> - 08:30 AM Bug #12696: ceph-osd starts before network, fails
- Submit:
https://github.com/ceph/ceph/pull/5582 - 07:33 AM Bug #12696 (Can't reproduce): ceph-osd starts before network, fails
- Our environment at startup, found that part of the OSD boot failure, log print is as follows:
OSD log:... - 07:59 AM Bug #12655: unable to bind to IP:PORT on any port in range 6800-7300: (98) Address already in us
- Thanks sage , setting a larger port range fixes this.
Before closing this ticket could you help me understanding t... - 07:45 AM Revision c577ea2e (ceph): Merge pull request #5569 from ceph/wip-unused-var
- client: fix unused var warning
- 07:18 AM Linux kernel client Bug #12695 (Closed): rbd flatten makes clone's snap crash
- ceph version:0.94.2
cluster:3 nodes,each node has 10 OSDs and 1 mon.
Operation?
(1)rbd create img1 --size 25... - 03:34 AM Revision fa37198d (ceph): Merge pull request #529 from ceph/wip-calamari-cli
- calamari_setup: install "cli" utils on Calamari node
Reviewed-by: Dan Mick <dmick@redhat.com> - 03:24 AM Revision 7f32a3de (ceph): rbd:modify the log of purging snaps so that it is more appropriate.
- Signed-off-by: s09816 <shi.lu@h3c.com>
- 02:14 AM Revision fabd6357 (ceph): ReplicatedPG: Don't cache recovery and scrub data
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
08/13/2015
- 08:26 PM Bug #12551: osd: copy-from doesn't preserve truncate_{seq,size}
- 08:25 PM Bug #12536: "FAILED assert(!log.null() || olog.tail == eversion_t())"
- commit:1ca6bf60e92734b5f34b2d29eea64eec5864800a
commit:1357ed1bd0e250b942bcba0346c97c24bb79a5d1 (hammer) - 08:24 PM Bug #12536 (Resolved): "FAILED assert(!log.null() || olog.tail == eversion_t())"
- 08:23 PM Bug #12536: "FAILED assert(!log.null() || olog.tail == eversion_t())"
- 08:22 PM Bug #12655: unable to bind to IP:PORT on any port in range 6800-7300: (98) Address already in us
- YOu can increase the range of ports we try to bind to with ms_bind_port_min (6800) and ms_bind_port_max (7300). You ...
- 08:21 PM Bug #12665 (Need More Info): osd/ReplicatedPG.cc: 2706: FAILED assert(p != snapset.clones.end())
- Hi Brad,
Are you using cache tiering? This may be a result of an old bug, #8629, that was present in your older ve... - 12:26 PM Bug #12665: osd/ReplicatedPG.cc: 2706: FAILED assert(p != snapset.clones.end())
- So far the problem keeps persisting, so here is a small part of the log from an crashing osd...
2015-08-08 23:31... - 08:08 PM rbd Bug #11593 (Resolved): diff_iterate API call should use callback's return code
- 08:08 PM rbd Feature #11625 (Resolved): diff_iterate should utilize concurrent RADOS operations
- 08:04 PM Revision 7083ed5e (ceph): Merge pull request #4744 from ceph/wip-11625
- librbd: diff_iterate should issue concurrent ops
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 07:27 PM Linux kernel client Bug #12691 (Resolved): ffsb osd thrash test - osd/ReplicatedPG.cc: 2348: FAILED assert(0 == "out ...
- Seen duing rbd:thrah Can reprodouce 1/3 runs...
- 07:11 PM Revision 84007cf3 (ceph): Merge pull request #5555 from dachary/wip-mon-test-timeouts
- tests: be more generous with mon tests timeouts
Reviewed-by: Kefu Chai <kchai@redhat.com> - 07:03 PM Revision c229f7f6 (ceph): Merge pull request #5576 from liewegas/wip-dencoder
- simplify handling for objects w/ nondeterministic encoding
Reviewed-by: John Spray <john.spray@redhat.com> - 06:50 PM Revision 423cf136 (ceph): rgw: we should not overide Swift sent content type
- Fixes: #12363
backport: hammer
Signed-off-by: Orit Wasserman <owasserm@redhat.com> - 06:49 PM Revision 623a8e7c (ceph): test/encoding/check-generated: test sorted json dumps for nondeterminis...
- Nondeterministic objects dump nondeterministically (usually due
to unordered_map or _set). Compare their sorted outp... - 06:37 PM Revision 23119ffa (ceph): Revert "osd/HitSet: make subclasses dump deterministically"
- This reverts commit 34c048b3cb2b16658e692ac5110115612031510d.
- 06:37 PM Revision c8bb8a2a (ceph): Revert "mon/PGMap: dump osd_epochs in deterministic order"
- This reverts commit 3df6438cd2c9a77a06b7a8d011bafbed00f1fc5e.
- 06:17 PM Bug #10399: kvm die with assert(m_seed < old_pg_num)
- This was Ceph 0.94.1 on Linux/x86_64, so whatever version of librados came with that release.
About 2 of 10 VMs di... - 05:48 PM Bug #10399 (Need More Info): kvm die with assert(m_seed < old_pg_num)
- Which version of librados was this Roy? How many VMs saw this crash, and how many total are there (trying to figure o...
- 05:33 PM Bug #10399: kvm die with assert(m_seed < old_pg_num)
- I just ran into this issue as well, while growing pg_num and pgp_num....
- 06:08 PM Revision fb020247 (ceph): rgw: don't preserve acls when copying object
- Fixes: #12370
When copying an object we need to use the acls we calculated earlier,
and not the source acls.
This wa... - 06:08 PM Revision aa84941c (ceph): tasks/kcephfs: enable MDS debug
- To help us debug #11482
Signed-off-by: John Spray <john.spray@redhat.com> - 06:03 PM Feature #12602 (Resolved): c++11
- 06:03 PM Feature #11033 (Resolved): selinux policy for ceph-osd
- 06:03 PM Feature #11032 (Resolved): selinux policy for ceph-mon
- 05:48 PM rgw Bug #12690 (Resolved): rgw: test_multipart_upload_incorrect_etag fails
- a new test that was added, not a regression, never worked probably.
- 04:29 PM Revision 42a8f7c2 (ceph): rgw: cleanup dead init_bucket
- Signed-off-by: Pete Zaitcev <zaitcev@redhat.com>
- 04:24 PM devops Bug #12689 (Rejected): Error "Packages for distro 'precise' and ceph hash 'bda233c477bd3dbae52419...
- Run: http://pulpito.ceph.com/teuthology-2015-08-11_17:02:02-upgrade:hammer-x-next-distro-basic-multi/
Jobs: ['101068... - 04:23 PM Bug #11661: "FAILED assert(0 == "Running incompatible OSD")"
- I'm running from the hammer branch and one of my OSDs is hitting this frequently. Will this be back-ported to the ham...
- 04:22 PM Bug #12410: OSDMonitor::preprocess_get_osdmap: must send the last map as well
- Run: http://pulpito.ceph.com/teuthology-2015-08-11_17:02:02-upgrade:hammer-x-next-distro-basic-multi/
Jobs: ['101068... - 04:00 PM Bug #10725 (Resolved): missing man pages for ceph-create-keys, ceph-disk-*
- backport merged, consider it done.
- 03:35 PM Revision 46c422f8 (ceph): Merge pull request #4864 from stevecapperlinaro/for-ceph/journal-page-s...
- FileJournal: Remove CEPH_PAGE_SIZE assumptions
Reviewed-by: Sage Weil <sage@redhat.com> - 03:34 PM Revision 9558d594 (ceph): Merge pull request #5100 from stevecapperlinaro/for-ceph/tweak-buffer-a...
- Common: Do not use CEPH_PAGE_SIZE when appending buffers in Ceph
Reviewed-by: Sage Weil <sage@redhat.com> - 03:34 PM Revision 6cb8bc7f (ceph): Merge pull request #5339 from xinxinsh/wip-use-throttle
- use throttle framework to throttle objectstore ops/bytes
Reviewed-by: Li Wang <liwang@ubuntukylin.com>
Reviewed-by: ... - 03:29 PM Bug #12688 (Duplicate): pg import can cause bogus interval (first > last)
- ...
- 02:48 PM CephFS Bug #12598 (Resolved): LibCephFS.GetPoolId failure
- commit:4d4fe9dbc0eb0d0eaa9a608474fecc892626f542
- 02:46 PM Revision 4d4fe9db (ceph): crypto: fix unbalanced ceph::crypto::init/ceph::crypto:shutdown
- we may create a CephContext without calling common_init_finish(), then
delete the CephContext. In this case, ceph::cr... - 02:44 PM RADOS Bug #12687 (New): osd thrashing + pg import/export can cause maybe_went_rw intervals to be missed
- symptom is...
- 02:21 PM devops Bug #12685 (Resolved): %preun script is running systemctl disable/stop on package upgrade
- 01:50 PM devops Bug #12685: %preun script is running systemctl disable/stop on package upgrade
- master PR: https://github.com/ceph/ceph/pull/5573
- 01:41 PM devops Bug #12685 (Resolved): %preun script is running systemctl disable/stop on package upgrade
- The %preun section now contains logic for disabling and stopping all the Ceph systemd units when the ceph package is ...
- 02:21 PM Revision 6c6d62fa (ceph): Merge pull request #5573 from SUSE/wip-restart-on-rpm-removal-only
- ceph.spec.in: test %preun argument is zero for removal-only operations
- 01:36 PM Revision 6c803cdf (ceph): ceph.spec.in: test %preun argument is zero for removal-only operations
- The %preun section now contains logic for disabling and stopping all the
Ceph systemd units when the ceph package is ... - 01:29 PM Revision c1ca95cd (ceph): tasks/cephfs: implement TestDataScan.test_stashed_layout
- This is for testing how cephfs-data-scan handles
the new 'layout' xattr when found during recovery.
Signed-off-by: J... - 01:28 PM Revision bde85b71 (ceph): tools/cephfs: pass layouts around in DataScan
- ...instead of chunk_size+data_pool arguments. This
is a precursor to using the 0th-object xattr
stashed layouts duri... - 01:28 PM Revision 773b4312 (ceph): tools/cephfs: use xattr layout if present+valid
- So that in cases where files had an unguessable (e.g.
striped) layout, but they were lucky enough to
have the layout ... - 01:09 PM Bug #9732: ReplicatedPG::hit_set_trim osd/ReplicatedPG.cc: 11006: FAILED assert(obc)
- another occurrence: /a/sage-2015-08-12_14:04:07-rados-wip-sage-testing---basic-multi/1012342/remote/plana91/log/ceph-...
- 12:51 PM Revision bf0a614f (ceph): Merge pull request #5562 from varadakari/wip-kvstore-statfs
- Adding statfs api to KeyValueDB
Reviewed-by: Sage Weil <sage@redhat.com> - 12:44 PM Revision 930927bc (ceph): Merge pull request #5565 from varadakari/wip-kvstore-makefile
- Remove redundant line from Makefile
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 12:14 PM Revision 5f41e8d2 (ceph): tests: do not test timeout mon add
- Strictly speaking it's not a feature, but an expected hahaviour. After
discussion it does not seem useful to have tha... - 11:55 AM Bug #12673 (Fix Under Review): cache agent is idle although one object is left in the cache
- https://github.com/ceph/ceph/pull/5570
- 10:33 AM Bug #12673: cache agent is idle although one object is left in the cache
- Running the script in the description, here is what I find in the logs: when the agent_choose_mode function runs at 2...
- 06:50 AM Bug #12673: cache agent is idle although one object is left in the cache
- Even if this is not fixed, it should be documented in a place where the developer / tester / user facing this situati...
- 05:44 AM Bug #12673: cache agent is idle although one object is left in the cache
- Loïc, just want to rectify some points here.
> agent_choose_mode flush_mode: idle evict_mode: idle *num_objects: 1... - 11:02 AM Cleanup #12684 (Resolved): Remove no use code in pipe.cc.
- In pipe.cc the below function
int Pipe::do_sendmsg(struct msghdr *msg, int len, bool more)
{
while (len > 0... - 10:39 AM rgw Bug #12683 (Closed): rgw/rgw_civetweb:Conditional always is invalid
- "if (0 && data.length() == 0)" should be replaced by "if ( data.length() == 0)"
- 10:18 AM Revision f2a38bfa (ceph): client: fix unused var warning
- Signed-off-by: John Spray <john.spray@redhat.com>
- 10:01 AM Revision ce03de1b (ceph): mon/MDSMonitor: fix fs reset map init
- `fs reset` was failing to fully initialize the
mdsmap. Especially, failing to initialize
session_timeout was breakin... - 10:01 AM Revision 553bc3ff (ceph): cls: load layout xattr in cephfs cls
- Do it with FAILOK because, like the backtrace,
its presence is optional.
Signed-off-by: John Spray <john.spray@redha... - 10:00 AM Revision 593fcc1a (ceph): Merge remote-tracking branch 'origin/next'
- 09:43 AM Revision ad8aaae8 (ceph): Merge pull request #5564 from wonzhq/test-inode
- mds: initialize 'divergent' to false when comparing inode_t
Reviewed-by: John Spray <john.spray@redhat.com> - 09:36 AM Revision ca7780eb (ceph): Merge pull request #5553 from majianpeng/mds
- mds (fix to flags_to_mode and cleanup in handle_client_open)
Reviewed-by: John Spray <john.spray@redhat.com> - 09:04 AM Revision 408ada4e (ceph): mds/Server: clean up code for handle_client_open.
- Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
- 09:03 AM Revision 43529a58 (ceph): common/ceph_fs: set mode default is -1 for func ceph_flags_to_mode.
- Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
- 06:53 AM Revision 6eea7941 (ceph): Remove redundant line from Makefile
- Signed-off-by: Varada Kari <varada.kari@sandisk.com>
- 06:28 AM Revision e1ff09ce (ceph): mds: initialize 'divergent' to false when comparing inode_t
- 'divergent' is not set in the case when all the fields of 'inode_t' are
equal.
Signed-off-by: Zhiqiang Wang <zhiqian... - 05:45 AM Revision bf21555c (ceph): Change key format to preserve order of keys in backend db.
- Changing the key format to add additional digits to preserve the order
for the read aheads and lookups to be faster. ... - 05:35 AM Revision d33ad154 (ceph): Adding statfs api to KeyValueDB
- If any backend supports and implements statfs will extract the
stats from backend to show correct utilization in the ... - 03:26 AM rbd Feature #11287 (Fix Under Review): librbd: write to journal
- *master PR*: https://github.com/ceph/ceph/pull/5278
- 03:22 AM rbd Backport #12682 (Fix Under Review): object_map_update fails with -EINVAL return code
- 03:14 AM rbd Backport #12682 (Resolved): object_map_update fails with -EINVAL return code
- https://github.com/ceph/ceph/pull/5559
- 03:18 AM Revision 7add6468 (ceph): rbd:Check the dest image name, if it is empty string, refuse to execute...
- Signed-off-by: solesoul1127 <chen.yehua@h3c.com>
- 03:13 AM Revision 153744d7 (ceph): tests: increase test coverage for partial encodes/decodes
- Multiple combinations of offsets/lengths are now tested when
performing partial encodes/decodes of the bit vector.
S... - 03:13 AM Revision fca78765 (ceph): common: bit_vector extent calculation incorrect for last page
- It's highly probable that the last page in the bit vector will not
be a full page size. As a result, the computed ex... - 02:47 AM Revision 9e58c628 (ceph): mon: show number of PGs in flush/evict mode in command line
- Show the number of PGs in the high/low flush mode and some/full evict
mode in the output of 'ceph -s' and 'ceph osd p... - 02:34 AM Revision d57d36da (ceph): osd: add flush/evict mode in pg stats
- These statistic data are used to show the PGs in high/low flush mode and
some/full evict mode in the command line.
S... - 02:01 AM Revision af2a38bf (ceph): mon: fix the output of cache_io_rate_summary
- If there is no flush but evict/promote ongoing, the output doesn't look
good.
Signed-off-by: Zhiqiang Wang <zhiqiang... - 01:09 AM Bug #12681 (Won't Fix): tcmalloc doesn't release memory back to system
- We observed high memory consumption by OSD daemons when we evaluating Hammer (0.94.2) in Yahoo! environment. Initiall...
08/12/2015
- 11:53 PM rbd Bug #12664 (Pending Backport): Crash during TestInternal.MultipleResize
- 12:06 AM rbd Bug #12664: Crash during TestInternal.MultipleResize
- For the record http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/6991/consoleText
- 11:22 PM Revision 6a1dbf30 (ceph): Merge pull request #5544 from ceph/wip-12664-next
- librbd: prevent race condition between resize requests
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 10:17 PM rgw Bug #12623: rgw: all worker threads hung with peering PG
- https://github.com/ceph/ceph/pull/5501
- 08:59 PM Bug #12673 (Won't Fix): cache agent is idle although one object is left in the cache
- When using 50 objects instead of just one as in the tracker description, the cache eviction happens as expected.
<pr... - 05:20 PM Bug #12673: cache agent is idle although one object is left in the cache
- The hit_set_period value of 5 is super low.. you probably want 600 at a minimum (10 minutes). But I'm not sure that'...
- 04:13 PM Bug #12673: cache agent is idle although one object is left in the cache
- The cache agent does not wake up when a proxy read happens and promotes the object in ReplicatedPG::maybe_handle_cach...
- 02:24 PM Bug #12673: cache agent is idle although one object is left in the cache
- Confirmed on master as well.
- 08:25 AM Bug #12673 (Resolved): cache agent is idle although one object is left in the cache
- Steps to reproduce:...
- 08:52 PM Revision 0cf220ad (ceph): calamari_setup: install "cli" utils on Calamari node
- In our RHCS 1.3 ceph-deploy docs, we tell users to run "ceph-deploy
install --cli" on their calamari admin node, but ... - 07:45 PM rgw Bug #12680 (Resolved): release version 1.2.4 of the agent
- merged commit e354ff8 into master
- 07:32 PM rgw Bug #12680 (Resolved): release version 1.2.4 of the agent
- It contains fixes for installation and attribute errors.
- 07:28 PM rgw Bug #12677 (Resolved): agent has AttributeErrors on wrong exception references
- merged commit 5d0d60b into master
- 07:25 PM rgw Bug #12677: agent has AttributeErrors on wrong exception references
- It has fixed the attribute error previously seen during issue 12661...
- 05:04 PM rgw Bug #12677 (Fix Under Review): agent has AttributeErrors on wrong exception references
- PR opened: https://github.com/ceph/radosgw-agent/pull/36
- 04:52 PM rgw Bug #12677 (Resolved): agent has AttributeErrors on wrong exception references
- The agent is causing errors like:...
- 07:21 PM rbd Feature #12678: Support the ability to rename snapshots
- +1
- 05:21 PM rbd Feature #12678 (Resolved): Support the ability to rename snapshots
- 04:30 PM rgw Bug #12661: radosgw-agent datasync S3ResponseError: 403 Forbidden dest_connection.get_bucket(buck...
- I see the following crash on Ubuntu not seen in RHEL run, this is on next branch and fails during the same tiny_file ...
- 01:05 AM rgw Bug #12661: radosgw-agent datasync S3ResponseError: 403 Forbidden dest_connection.get_bucket(buck...
- looks like some config issue , the key gets listed if one gets bucket first and iterates over key...
- 02:59 PM Revision b78883b1 (ceph): tests: be more generous with mon tests timeouts
- Change the timeouts of the TEST_mon_add_to_single_mon tests to be 120
seconds. They have value to guard against block... - 02:45 PM Bug #12654 (Duplicate): unable to bind to IP:PORT on any port in range 6800-7300: (98) Address al...
- 01:42 PM Documentation #12620 (Resolved): note the behaviour that "ceph mon add <mon-id>" takes forever in...
- 01:22 PM CephFS Bug #12674: Semi-reproducible crash of ceph-fuse
- Some additional info which might be useful:
* We are no longer 100% certain that the crashes relate to renames. The ... - 01:19 PM CephFS Bug #12674: Semi-reproducible crash of ceph-fuse
- Zheng Yan wrote:
> the log seems like memory allocation failure. how many memory does your machine have?
The VM h... - 12:45 PM CephFS Bug #12674: Semi-reproducible crash of ceph-fuse
- the log seems like memory allocation failure. how many memory does your machine have?
- 10:22 AM CephFS Bug #12674 (Resolved): Semi-reproducible crash of ceph-fuse
- we are running ceph version 0.94.2 with a cephfs mounted using ceph-fuse on
Ubuntu 14.04 LTS. I think we have found ... - 01:18 PM CephFS Bug #12676 (Resolved): MDSMap assertion in MDCache::trim (multimds)
- ...
- 12:27 PM CephFS Bug #12612: fuse jobs fail to start on centos7
- Works for me. Let's see if we see more instances I guess.
- 11:12 AM CephFS Bug #12612: fuse jobs fail to start on centos7
- Hmm. Just ran a job using fuse and it passed. Trying again with this exact config.
- 10:45 AM Linux kernel client Support #12648: [old topic] OSD will stuck if running kernel rbd or rbd-fuse on the same node wit...
- Hi Ilya,
It's really hard to catch the timing to collect dmesg logs under our test scenario because of low memory ... - 09:25 AM Revision d8d6bb89 (ceph): Merge pull request #5550 from schuzhoy/schuzhoy-patch-1
- doc: Correcting the default number of copies.
Reviewed-by: Travis Rhoden <trhoden@redhat.com>
Reviewed-by: Kefu Chai... - 09:21 AM Revision 62d3f459 (ceph): Merge pull request #5072 from zhouyuan/doc_rgw_rados_handlers
- doc: update rgw configuration on multiple rgw rados handlers feature
Reviewed-by: Kefu Chai <kchai@redhat.com> - 07:49 AM Revision fe28e19b (ceph): Merge pull request #5495 from tchaikov/wip-12620
- doc/rados/operations/add-or-rm-mons: simplify the steps to add a mon
Reviewed-by: Joao Eduardo Luis <joao@suse.de>
R... - 07:43 AM Revision ea601efc (ceph): Merge pull request #5390 from SUSE/wip-12503-firefly
- rgw/logrotate.conf calls service with wrong init script name
Reviewed-by: Ken Dreyer <kdreyer@redhat.com> - 06:23 AM CephFS Bug #12617 (Resolved): Missing handling of supplementary groups in ceph-fuse -> rendering it usel...
- 02:42 AM CephFS Bug #12617: Missing handling of supplementary groups in ceph-fuse -> rendering it useless for sha...
- ignore client check patch makes ceph-fuse work, the get_grouplist patch makes programs that directly use libcephfs wo...
- 03:30 AM Revision 606e858a (ceph): Merge pull request #5528 from Be-El/master
- Correct handling of supplementary groups in ceph-fuse (Bug 12617)
- 01:42 AM Revision 8ebab808 (ceph): Merge pull request #527 from dmick/master
- calamari_setup: change to use ceph-deploy repo command
Reviewed-by: Ken Dreyer <kdreyer@redhat.com> - 01:09 AM Revision 7e6f8190 (ceph): doc: update rgw configuration on multiple rgw rados handlers feature
- Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>
08/11/2015
- 11:14 PM Revision 532a4e4a (ceph): calamari_setup: change to use ceph-deploy repo command
- Fixes: #12643
Signed-off-by: Dan Mick <dan.mick@redhat.com> - 10:40 PM Revision efc8969e (ceph): Doc: Correcting the default number of copies.
- According to the "quick ceph deploy" page, the default number of replicas in the Ceph configuration file is 3
Signed... - 10:17 PM Stable releases Tasks #11644: firefly v0.80.11
- rgw is green
- 09:51 PM CephFS Feature #12671: Enforce cache limit during dirfrag load during open_ino (during rejoin)
- The source of this observation was https://www.mail-archive.com/ceph-users@lists.ceph.com/msg22235.html
In this in... - 09:49 PM CephFS Feature #12671 (New): Enforce cache limit during dirfrag load during open_ino (during rejoin)
When clients replay requests referring to inodes not found in cache, the inode numbers are stashed for loading late...- 09:02 PM rgw Bug #12108 (Pending Backport): RGW Swift API: X-Trans-Id header is wrongly formatted
- 09:01 PM Revision 79e77686 (ceph): Merge pull request #5047 from rzarzynski/wip-12108
- rgw: rework X-Trans-Id header to be conform with Swift API.
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 07:45 PM rgw Bug #12574: Firefly to Hammer RGW Upgrade
- You are correct, a small number of them had now been updated due to a apt repo error - Thanks for spotting!
- 06:47 PM rgw Bug #12574: Firefly to Hammer RGW Upgrade
- It looks as if you might not have restarted your osds after the upgrade.
- 06:55 PM Bug #12665: osd/ReplicatedPG.cc: 2706: FAILED assert(p != snapset.clones.end())
- An update:
We were able to finish the rebalancing by setting:
osd pg max concurrent snap trims = 0
We also nee... - 06:40 PM rgw Bug #12630 (Pending Backport): segmentation fault when rgw_gc_max_objs > HASH_PRIME
- 06:39 PM Revision 0328bc68 (ceph): Merge pull request #5449 from yangruifeng/wip-rgw-bugfix1
- rgw:segmentation fault when rgw_gc_max_objs > HASH_PRIME
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 06:35 PM rgw Bug #12640: radosgw + civetweb latency issue on Hammer
- any chance you could provide a corresponding strace?
- 06:33 PM rgw Bug #12629 (Pending Backport): the arguments 'domain' should not be assigned when return false
- 06:32 PM Revision 7b5f9b05 (ceph): Merge pull request #5447 from yangruifeng/master
- rgw:the arguments 'domain' should not be assigned when return false
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 06:28 PM rgw Bug #11367 (Resolved): Keystone PKI token expiration is not enforced
- 06:27 PM rgw Backport #11721 (Resolved): Keystone PKI token expiration is not enforced
- 06:27 PM Revision 50fa9636 (ceph): Merge pull request #4765 from SUSE/wip-11721-firefly
- rgw: Keystone PKI token expiration is not enforced
- 06:24 PM rgw Bug #7647 (Pending Backport): Ensure that swift keys don't include backslashes
- 06:24 PM rgw Bug #7647 (Resolved): Ensure that swift keys don't include backslashes
- Fixed, commit:3f04a6126fdbfb93304f798da3775c0eec9b7d44
- 06:20 PM rgw Backport #12635 (In Progress): swift smoke test fails on TestAccountUTF8
- 06:18 PM rgw Backport #12635 (Resolved): swift smoke test fails on TestAccountUTF8
- 06:20 PM rgw Backport #12634 (In Progress): swift smoke test fails on TestAccountUTF8
- 06:18 PM rgw Backport #12634 (Resolved): swift smoke test fails on TestAccountUTF8
- 06:19 PM rgw Bug #11438 (Pending Backport): swift smoke test fails on TestAccountUTF8
- 06:19 PM rgw Bug #11438 (Resolved): swift smoke test fails on TestAccountUTF8
- 06:19 PM Revision 28a0a9af (ceph): Merge pull request #5543 from ceph/wip-cmake-nlist
- CMake: add missing librados nlist test
Reviewed-by: Kefu Chai <kchai@redhat.com> - 06:16 PM rgw Backport #12042 (Resolved): DragonDisk fails to create directories via S3: MissingContentLength
- 06:13 PM rgw Bug #12208: GWWatcher::handle_error -> common/Mutex.cc: 95: FAILED assert(r == 0)
- We delete RGWRados::finisher prior to unregistering the watch, the watch error path calls finisher.
- 06:06 PM rgw Bug #11851 (Resolved): Civetweb RGW appears to report full size of object as downloaded when only...
- 06:06 PM rgw Backport #12244 (Resolved): Civetweb RGW appears to report full size of object as downloaded when...
- 06:03 PM rgw Bug #11974 (Resolved): Fix tool for #11442 does not correctly fix objects created via multipart u...
- 06:03 PM rgw Backport #12242 (Resolved): Fix tool for #11442 does not correctly fix objects created via multip...
- 06:02 PM rgw Bug #12079 (Resolved): rgw: need conversion tool to handle fixes following #11974
- 05:49 PM Documentation #11082 (Resolved): Crushmap documentation still references raid4
- Fixed by Aron Gunn.
- 05:48 PM Revision 9b8c518b (ceph): Merge pull request #5547 from ritz303/wip-doc-11082
- doc: Removed reference to RAID-4
Reviewed-by: John Wilkins <jowilkin@redhat.com> - 05:46 PM Revision 6f768a7c (ceph): doc: Removed reference to RAID-4
- Fixes: #11082
Signed-off-by: ritz303 <ritz_303@yahoo.com> - 05:14 PM rbd Documentation #11897 (Resolved): Documentation on ceph auth caps is not specific enough.
- Fixed by Aron Gunn.
- 05:11 PM Revision f3330a18 (ceph): Merge pull request #5546 from ritz303/wip-doc-11897
- doc: Clarify how 'ceph auth caps' works.
Reviewed-by: John Wilkins <jowilkin@redhat.com> - 04:30 PM devops Bug #12667 (Duplicate): Missing http://gitbuilder.ceph.com/ceph-deb--x86_64-basic/ref/firefly/ver...
- Duplicate of #12668, closing this one.
- 04:21 PM devops Bug #12667 (Duplicate): Missing http://gitbuilder.ceph.com/ceph-deb--x86_64-basic/ref/firefly/ver...
- hammer v0.94.3 QE validation
run: http://pulpito.ceph.com/teuthology-2015-08-10_14:12:42-upgrade:firefly-x-hammer-... - 03:35 PM Revision c6cf558c (ceph): CMake: cut down unnecessary linkage on rados tests
- Signed-off-by: John Spray <john.spray@redhat.com>
- 03:00 PM Revision 6ee955d1 (ceph): CMake: add missing librados nlist test
- Signed-off-by: John Spray <john.spray@redhat.com>
- 02:40 PM Revision db55527e (ceph): Merge pull request #5354 from ceph/wip-pgls-filter-librados
- librados: partially expose PGLS filter interface
Reviewed-by: David Zafman <dzafman@redhat.com> - 02:18 PM Revision b199ac6c (ceph): doc/rados/operations/add-or-rm-mons: simplify the steps to add a mon
- this change removes the step to "ceph mon add" before starting a new
monitor. because the existing leader will start ... - 02:01 PM CephFS Bug #12617: Missing handling of supplementary groups in ceph-fuse -> rendering it useless for sha...
- Do we need the Client.cc patch here, or just the pull request (https://github.com/ceph/ceph/pull/5480), or both?
- 02:00 PM CephFS Bug #12321: MDS crash when try to connect clients
- Hi, do you have any updates for us on this? If the system is unavailable for any more debug then we'll close this.
- 01:56 PM Fix #11666 (Resolved): erasure-code: add lrc to rados/thrash-erasure-code
- 01:55 PM Tasks #11831: erasure-code: add chance_test_backfill_full to ceph-qa-suite tests
- 01:55 PM CephFS Bug #12619 (Resolved): valgrind errors in fs suite
- 01:52 PM Bug #12213 (Rejected): cannot specify an alternate chef repository in firefly
- using ansible now
- 01:32 PM CephFS Bug #11482: kclient: intermittent log warnings "client.XXXX isn't responding to mclientcaps(revoke)"
- We need to turn on MDS logs to get a more useful bit of information
- 01:29 PM rbd Bug #12664 (Fix Under Review): Crash during TestInternal.MultipleResize
- *next PR*: https://github.com/ceph/ceph/pull/5544
- 01:26 PM Revision 6ec431ba (ceph): librbd: prevent race condition between resize requests
- It was possible that the same resize request could be sent twice
if a completed resize op started a newly created res... - 01:18 PM Bug #12655: unable to bind to IP:PORT on any port in range 6800-7300: (98) Address already in us
- HI
Today also i have encountered similar OSD error and OSD crashed with this error.
Need help to fix this probl... - 12:52 PM Revision 21464435 (ceph): Merge pull request #5535 from ceph/wip-gentest
- mon: Improve PGMap::generate_test_instances(), build map per incremental
Reviewed-by: Sage Weil <sage@redhat.com> - 12:49 PM Revision 55b34904 (ceph): Merge pull request #5457 from ceph/wip-cmake-vstart
- Make vstart work with cmake out of tree builds
Reviewed-by: Kefu Chai <kchai@redhat.com> - 11:41 AM Revision af18d479 (ceph): Merge remote-tracking branch 'upstream/master'
- 10:17 AM Revision 4649ba50 (ceph): librados: partially expose PGLS filter interface
- Not bothering with this for non-namespaced object iterator
or the C interface. Expect to replace this with
the new-s... - 10:17 AM Revision e8a53d23 (ceph): test: add case for pgls filter arg in librados
- Signed-off-by: John Spray <john.spray@redhat.com>
- 09:07 AM CephFS Bug #12209: CephFS should have a complete timeout mechanism to avoid endless waiting or unpredict...
- This issue is covered in http://tracker.ceph.com/issues/10944, and we have tested the two patches, which works well. ...
- 08:57 AM Revision a6ba518f (ceph): Merge pull request #5542 from l11625/master
- The listen_sd of thread also can be zero when we call the processor::start function..
Reviewed-by: Kefu Chai <kchai@... - 08:26 AM Revision 7ffcfdb6 (ceph): Merge pull request #5533 from ceph/wip-fix-mon-join.joao
- mon: Monitor: set MMonJoin's op type as service
Reviewed-by: Kefu Chai <kchai@redhat.com> - 08:23 AM Backport #12418 (In Progress): ceph-disk: sometimes the journal symlink is not created
- 08:19 AM Revision e57e6f5d (ceph): ceph-disk: set guid if reusing a journal partition
- When reusing a journal partition (e.g. /dev/sda2) we should set a
new partition guid and link it correctly with the O... - 08:19 AM Revision 04b2a878 (ceph): ceph-disk: don't change the journal partition uuid
- We observe that the new /dev/disk/by-partuuid/<journal_uuid>
symlink is not always created by udev when reusing a jou... - 08:12 AM Revision cabfd5c4 (ceph): when the processor start ,listen_sd also can be zero.
- 07:12 AM Revision fec459de (ceph): mon: refactor OSDMonitor::send_incremental()
- * remove the duplicated part of the two implementations of
OSDMonitor::send_incremental()
Signed-off-by: Kefu Chai... - 06:49 AM Revision c05753ea (ceph): mon: track osd_epoch in MonSession
- * remove osd_epoch<osd, epoch> from OSDMonitor
* add osd_epoch to MonSession to track the latest osdmap epoch
OSDMo... - 06:49 AM Revision de43a02e (ceph): mon: track osd_epoch of all MonSessions
- previously, we only track the osd_epoch for OSD peers. but other
MonClients who receives osdmaps can also benefit fro... - 03:01 AM Revision 91337bdd (ceph): Merge pull request #5480 from ceph/wip-12617
- client: ignore permission check when fuse_default_permissions is on
- 02:21 AM Revision f9368333 (ceph): test/mon: add test for "mon add"
- Signed-off-by: Kefu Chai <kchai@redhat.com>
- 01:03 AM Revision a3b29475 (ceph): Merge pull request #524 from ceph/wip-kdreyer-calamari-1.2
- calamari_setup: handle RHCS 1.2 (missing /mnt/MON)
Reviewed-by: Dan Mick <dmick@redhat.com> - 12:46 AM rgw Feature #12666 (Resolved): rgw: expose the number of *stuck threads* via admin socket
- With our Ceph cluster, we came across a couple of times that rgw only returned HTTP 500, which was due to the fact th...
08/10/2015
- 11:23 PM Revision bda233c4 (ceph): Merge pull request #5536 from ceph/wip-12662-next
- WorkQueue: add/remove_work_queue methods now thread safe
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 11:12 PM Bug #12662 (Fix Under Review): ThreadPool add/remove work queue methods not thread safe
- *next PR*: https://github.com/ceph/ceph/pull/5536
- 10:39 PM Bug #12662 (In Progress): ThreadPool add/remove work queue methods not thread safe
- 09:42 PM Bug #12662 (Resolved): ThreadPool add/remove work queue methods not thread safe
- librbd uses a singleton ThreadPool for asynchronous operations for each image. As images are opened and closed, Work...
- 11:12 PM Bug #12665 (Resolved): osd/ReplicatedPG.cc: 2706: FAILED assert(p != snapset.clones.end())
- After upgrading our Ceph cluster from 0.80.4 to 0.94.1 we have intermittent crashes on multiple osd's.
Marking those... - 11:10 PM Revision 3e18449b (ceph): WorkQueue: add/remove_work_queue methods now thread safe
- These methods were not acquiring the ThreadPool lock when
manipulating the work_queue collection. This was causing
o... - 11:02 PM Revision af336313 (ceph): Merge pull request #5344 from trhoden/update_debain_maintainer
- debian: Update maintainers and uploaders
Reviewed-by: Ken Dreyer <kdreyer@redhat.com> - 11:01 PM rbd Bug #12611 (Pending Backport): object_map_update fails with -EINVAL return code
- 01:48 PM rbd Bug #12611 (Fix Under Review): object_map_update fails with -EINVAL return code
- *next PR*: https://github.com/ceph/ceph/pull/5525
- 10:58 PM Revision d8a157cf (ceph): Merge pull request #5525 from ceph/wip-12611-next
- bit_vector extent calculation incorrect for last page
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 10:39 PM rbd Bug #12664 (In Progress): Crash during TestInternal.MultipleResize
- 10:26 PM rbd Bug #12664 (Resolved): Crash during TestInternal.MultipleResize
- A race condition is possible where the same request can be issued twice, leading to a crash:...
- 09:07 PM CephFS Bug #11482: kclient: intermittent log warnings "client.XXXX isn't responding to mclientcaps(revoke)"
- Reopening this because we're still seeing it on kcephfs, unknown whether underlying cause is in the same area:
http:... - 09:05 PM Revision ff6e9758 (ceph): Fixed mistaken reference to mon in osd section. Changed release name to...
- Fixes: #7807
Signed-off-by: John Wilkins <jowilkin@redhat.com> - 08:38 PM Revision 65e8f195 (ceph): doc: Clarify how 'ceph auth caps' works.
- Fixes: #11897
Signed-off-by: ritz303 <ritz_303@yahoo.com> - 08:37 PM rgw Bug #12661 (Duplicate): radosgw-agent datasync S3ResponseError: 403 Forbidden dest_connection.get...
- frontend is apache...
- 08:31 PM Revision efb6220b (ceph): mon: Improve PGMap::generate_test_instances(), build map per incremental
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 08:01 PM RADOS Bug #12659 (Closed): Can't delete cache pool
- Hi,
this is kind of a follow-up from issue #10138 where we encountered some strange issues with cache tiers.
On... - 07:03 PM Bug #8752: firefly: scrub/repair stat mismatch
- I have 0.80.9. I've set up caching pools and now I see same errors on the caching osds.
I have all xfs osds with the... - 06:49 PM rgw Bug #12470: radosgw-agent bootstrap failure (ubuntu 12.04)
- Thanks Alfredo, this is working fine, Partial logs from test run...
- 06:13 PM rgw Bug #12470 (Resolved): radosgw-agent bootstrap failure (ubuntu 12.04)
- merged commit fc3985a into master
- 05:53 PM rgw Bug #12470 (Fix Under Review): radosgw-agent bootstrap failure (ubuntu 12.04)
- Pull request opened https://github.com/ceph/radosgw-agent/pull/35
- 05:22 PM rgw Bug #12470: radosgw-agent bootstrap failure (ubuntu 12.04)
- Alfredo found the root cause in setup.py, Assigning to him.
- 04:50 PM rgw Bug #12470: radosgw-agent bootstrap failure (ubuntu 12.04)
- It's weird because I don't see "setuptools==0.9.8" @ https://github.com/boto/boto/blob/2.38.0/requirements.txt
- 04:43 PM rgw Bug #12470: radosgw-agent bootstrap failure (ubuntu 12.04)
- This is due to boto's egg dependency on setuptools 0.9.8 and mock 1.3.0 requires 17.1...
- 06:40 PM CephFS Bug #11255: nfs: mount failures on ceph-backed NFS share
- http://pulpito.ceph.com/teuthology-2015-08-07_23:10:02-knfs-master-testing-basic-multi/1006003/
- 06:31 PM CephFS Bug #12657: Failure in TestStrays.test_ops_throttle
- This purged 1600 strays in 6 seconds (and therefore wasn't busy doing so for long enough to hit the required high wat...
- 06:30 PM CephFS Bug #12657 (Can't reproduce): Failure in TestStrays.test_ops_throttle
http://pulpito.ceph.com/teuthology-2015-08-07_23:04:03-fs-master---basic-multi/1005757/- 05:53 PM Backport #12488 (Fix Under Review): pg_interval_t::check_new_interval - for ec pool, should not r...
- 05:21 PM Revision c8713055 (ceph): mon: Monitor: set MMonJoin's op type as service
- Otherwise, when the MonmapMonitor's PaxosService::dispatch() handles the
message, it will assert (as it should) becau... - 05:20 PM Bug #12012: IO error on kvm/rbd with an erasure coded pool tier (after an upgrade from 0.87.1 to ...
- 05:16 PM Bug #12012: IO error on kvm/rbd with an erasure coded pool tier (after an upgrade from 0.87.1 to ...
- to fix this issue,
* we need sparse reads in @ReplicatedPG::do_op@ to be treated as normal reads on ec pools
* it... - 04:01 PM Bug #12012: IO error on kvm/rbd with an erasure coded pool tier (after an upgrade from 0.87.1 to ...
- Actually, this might be as simple as the fact that when not using the librbd cache, the read ops are issued as a spar...
- 03:32 PM Bug #12012: IO error on kvm/rbd with an erasure coded pool tier (after an upgrade from 0.87.1 to ...
- Mehdi Abaakouk wrote:
> I guess all our pool setup is described into the previously send ceph-report. Lets me known ... - 04:14 PM rgw Backport #12635 (In Progress): swift smoke test fails on TestAccountUTF8
- 04:10 PM rgw Bug #11438: swift smoke test fails on TestAccountUTF8
- master PR: https://github.com/ceph/ceph/pull/4418
- 04:09 PM rgw Backport #12633 (In Progress): rgw: shouldn't return content-type: application/xml if content len...
- 04:06 PM Backport #12587 (In Progress): FileStore calls syncfs(2) even it is not supported
- 03:59 PM Backport #12586 (In Progress): FileStore calls syncfs(2) even it is not supported
- 03:43 PM Revision f04c8da5 (ceph): added permission check based on getgrouplist
- 03:26 PM Revision a99e7332 (ceph): rbd: export-diff now issues concurrent AIO read requests
- Combined with the improvements to the diff_iterate API, this should
noticeably increase the speed of a large image ex... - 03:24 PM Revision e9c78c6c (ceph): librbd: move diff_iterate logic to its own class
- In preparation for parallelizing the diff_iterate logic,
move all existing logic to its own class. No functional
cha... - 03:24 PM Revision f13bdaeb (ceph): librbd: execute multiple object diffs in parallel
- Issue up to 'concurrent_management_ops' list_snap ops to the
OSDs concurrently. The iterate callbacks will continue ... - 03:24 PM Revision 368f6320 (ceph): librbd: diff_iterate should handle callback errors
- If the diff_iterate callback returns an error, it should
result in diff_iterate being aborted with the error code
pro... - 03:24 PM Revision 404cab31 (ceph): tests: verify that diff_iterate callbacks can abort diff
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 03:17 PM rbd Bug #12479 (Pending Backport): [ FAILED ] TestLibRBD.BlockingAIO
- 02:57 PM Bug #12654: unable to bind to IP:PORT on any port in range 6800-7300: (98) Address already in us
- By mistake the issue got created twice. You can close this.
Bug #12655 needs your attention - 02:51 PM Bug #12654 (Duplicate): unable to bind to IP:PORT on any port in range 6800-7300: (98) Address al...
- Hi There
In my production ceph cluster several OSD's ( ~ 20 OSD ) crashed throwing this error... - 02:53 PM Bug #12655 (Rejected): unable to bind to IP:PORT on any port in range 6800-7300: (98) Address alr...
- Hi There
In my production ceph cluster several OSD's ( ~ 20 OSD ) crashed throwing this error... - 02:35 PM Backport #12585 (In Progress): OSD crash creating/deleting pools
- 02:34 PM Revision 3396a969 (ceph): osd/OSDMap: handle incrementals that modify+del pool
- Because new_pools was processed after old_pools, if something
was modified and then deleted in the same incremental
m... - 02:33 PM Backport #12584 (In Progress): OSD crash creating/deleting pools
- 02:33 PM Revision ebba1d59 (ceph): osd/OSDMap: handle incrementals that modify+del pool
- Because new_pools was processed after old_pools, if something
was modified and then deleted in the same incremental
m... - 02:31 PM devops Fix #10959 (Rejected): create a 'configurator' web service for configuration generation
- 02:29 PM rbd Bug #12176 (Resolved): librbd/internal.cc: 1967: FAILED assert(watchers.size() == 1)
- 02:28 PM Bug #10911 (Resolved): mon: osd featuers aren't updated correctly
- 02:07 PM rbd Bug #12596 (Resolved): Prevent object map errors from failing IO operations
- Already included in master and backported to v0.94.2
- 02:04 PM Stable releases Tasks #12639 (Rejected): radosgw + civetweb latency issue on Hammer
- 01:46 PM Revision 3e145f71 (ceph): tests: increase test coverage for partial encodes/decodes
- Multiple combinations of offsets/lengths are now tested when
performing partial encodes/decodes of the bit vector.
S... - 01:35 PM Revision c6d98992 (ceph): common: bit_vector extent calculation incorrect for last page
- It's highly probable that the last page in the bit vector will not
be a full page size. As a result, the computed ex... - 01:33 PM rbd Documentation #12642: snap_unprotect() requires access to all pools
- Josh Durgin wrote:
> The reason we didn't store the children with the parent image is to let a user without write ac... - 01:31 PM Revision 16b59c6c (ceph): added autoconf check for getgrouplist
- 01:17 PM Revision ab31f14c (ceph): Merge pull request #5353 from ceph/wip-pgls-filter-osd
- Make PGLSFilter interface more flexible (for cephfs)
Reviewed-by: Sage Weil <sage@redhat.com> - 01:09 PM Revision 0695d179 (ceph): vstart: detect and handle cmake environ
- Signed-off-by: John Spray <john.spray@redhat.com>
- 01:09 PM Revision 44ed436c (ceph): ceph.in: detect paths in out of tree build
- a la what we currently do for PYTHONPATH
and LD_LIBRARY_PATH, but for cmake out
of tree builds.
Signed-off-by: John ... - 01:09 PM Revision c747e29a (ceph): vstart: detect CEPH_BIN in stop.sh too
- ...for the benefit of rbd.
Signed-off-by: John Spray <john.spray@redhat.com> - 01:05 PM Revision dac84a7a (ceph): vstart: enable more path customization
- ...for working with out-of-tree builds.
Signed-off-by: John Spray <john.spray@redhat.com> - 01:05 PM Revision 1f5ddbf7 (ceph): CMake: build dencoder at src/ceph-dencoder
- ...instead of in src/test/dencoder.
While the source lives in test/, the dencoder tool is built
and distributed as a... - 01:05 PM Revision af31afa8 (ceph): init-ceph.in: set executable bit
- Scripts expect the generated init-ceph script
to be +x, and CMake does that if the file you
feed into it is +x. This... - 01:05 PM Revision 6545981b (ceph): ceph.in: use cmake-style substitution
- Avoid need for separate ceph.in and ceph.in.cmake files.
Signed-off-by: John Spray <john.spray@redhat.com> - 01:05 PM Revision 64255b96 (ceph): Makefile: build ceph.in the cmake way
- A little bit more of a tongue-twisting command line
to do string replacement, but saves us from having
separate ceph.... - 10:01 AM Revision 9d17a65d (ceph): Merge pull request #5522 from oritwas/wip-cmake-fix-mds
- cmake: fix mds compilation and link
Reviewed-by: John Spray <john.spray@redhat.com> - 08:50 AM Revision 570887f1 (ceph): cmake: fix mds compilation and link
- Rename MDS.cc MDSDaemon.cc
Add missing files and match order to the Makefile
Signed-off-by: Orit Wasserman <owasserm... - 08:03 AM CephFS Bug #12653: fuse mounted file systems fails SAMBA CTDB ping_pong rw test with v9.0.2
- This is a ceph-fuse bug, running the test directly on ceph-fuse mount can reproduce the bug. The bug is because that ...
- 03:37 AM CephFS Bug #12653 (Resolved): fuse mounted file systems fails SAMBA CTDB ping_pong rw test with v9.0.2
- While working with the SAMBA CTDB ping_pong test and the Ceph file system, I noticed that file systems mounted using ...
- 07:35 AM rbd Feature #12111 (Resolved): rbd: support size suffixes for all size-based options
- 06:01 AM Revision b049c949 (ceph): Merge branch 'wip-vumrao-12111' of https://github.com/vumrao/ceph
- Reviewed-by: Josh Durgin <jdurgin@redhat.com>
- 05:20 AM Revision dcf20b35 (ceph): rbd: creating refused, if feature is specified when the format is equal...
- Signed-off-by: x11562 <xie.na@h3c.com>
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 03:26 AM Revision 281a3dde (ceph): Merge pull request #5520 from badone/master
- doc: remove duplicate word in Motivation section.
Reviewed-by: Kefu Chai <kchai@redhat.com> - 03:00 AM Revision 8023f1db (ceph): Merge pull request #4927 from wonzhq/snapset-obc
- osd/ReplicatedPG: snapset is not persisted
- 02:52 AM Revision 585b49f4 (ceph): Merge pull request #5212 from ceph/wip-mds-refactor
- mds: big refactor of MDS into MDSDaemon, MDSRank, MDSRank Dispatcher
- 02:03 AM Bug #12645: chunky_scrub assert fail
- According to my point,the reason for this bug is :
1.3 scrub request to repica
-1936> 2015-07-30 16:21:31.804154... - 01:51 AM Revision 370e4a71 (ceph): doc: remove duplicate word in Motivation section.
- Remove to\nto typo in doc/dev/osd_internals/erasure_coding/pgbackend.rst
Signed-off-by: Brad Hubbard <bhubbard@redha...
08/09/2015
- 02:49 PM Bug #12652 (Fix Under Review): osd/PGLog.cc: 732: FAILED assert(log.log.size() == log_keys_debug....
- https://github.com/ceph/ceph/pull/5518
- 02:46 PM Bug #12652: osd/PGLog.cc: 732: FAILED assert(log.log.size() == log_keys_debug.size())
- merge_log, just prior:...
- 02:45 PM Bug #12652 (Resolved): osd/PGLog.cc: 732: FAILED assert(log.log.size() == log_keys_debug.size())
- ...
- 11:47 AM Revision dc6d90cc (ceph): Merge pull request #5517 from tchaikov/wip-cmake-remove-flat-index
- cmake: remove Flat_index stuff
Reviewed-by: Sage Weil <sage@redhat.com> - 11:45 AM Revision a85b921c (ceph): Merge pull request #5515 from ceph/wip-make-check-fixes
- several fixes for make check
Reviewed-by: Loic Dachary <ldachary@redhat.com>
Reviewed-by: David Zafman <dzafman@redh... - 11:44 AM Revision 6dd143d8 (ceph): Merge pull request #4683 from ceph/wip-siddharth-security-fixes
- Enable security hardening flags globally
Testedj-by: Kefu Chai <kchai@redhat.com> - 08:28 AM Revision 14045020 (ceph): cmake: remove Flat_index stuff
- the Flat_index stuff was removed in 5df688a
Signed-off-by: Kefu Chai <kchai@redhat.com>
08/08/2015
- 03:40 PM Revision c74461e9 (ceph): Merge pull request #5504 from tchaikov/wip-check-good-yasm
- cmake: check for good yasm
- 03:21 PM Revision 8f06d3ee (ceph): Enable security hardening flags globally
- 03:21 PM Revision e5569d4c (ceph): Removing unwanted -grecord-gcc-switches , relro in cflags and adding -fPIC
- 03:21 PM Revision 2c6acfc9 (ceph): Removing unwanted -fPIC in cflags
- 01:42 PM Revision 1b94f1c9 (ceph): Merge pull request #5293 from majianpeng/oprequest-markevent
- osd: Oprequest markevent
Reviewed-by: Huang Jun <hjwsm1989@gmail.com>
Reviewed-by: Sage Weil <sage@redhat.com> - 12:38 PM Bug #12356 (Resolved): test/erasure-code/test-erasure-code.sh:227: TEST_rados_put_get_jerasure: ...
- 12:38 PM Revision 67b001d8 (ceph): Merge pull request #5267 from dachary/wip-12356-rados-osds-out-in
- tests: test-erasure-code.sh/rados_osds_out_in must wait_for_clean
Reviewed-by: Kefu Chai <kchai@redhat.com> - 12:27 PM Revision e04f324b (ceph): Merge pull request #4909 from zhouyuan/rgw_doc_bucket_index
- doc: update rgw configuration doc on bucket index sharing feature
Reviewed-by: Abhishek Lekshmanan <abhishek.lekshma... - 11:45 AM Revision d3727185 (ceph): osd/HitSet: mark subclasses that encode nondeterministically
- Signed-off-by: Sage Weil <sage@redhat.com>
- 11:45 AM Revision 34c048b3 (ceph): osd/HitSet: make subclasses dump deterministically
- Signed-off-by: Sage Weil <sage@redhat.com>
- 11:45 AM Revision 6b97addc (ceph): test/encoding/check-generated: make error msg cut and pasteable
- Signed-off-by: Sage Weil <sage@redhat.com>
- 11:36 AM Revision 3df6438c (ceph): mon/PGMap: dump osd_epochs in deterministic order
- Signed-off-by: Sage Weil <sage@redhat.com>
- 11:30 AM Revision 62f94221 (ceph): test/encoding/check-generated: skip some tests if non-determinstic enco...
- Signed-off-by: Sage Weil <sage@redhat.com>
- 11:30 AM Revision 2bcab031 (ceph): ceph-dencoder: add 'is_deterministic' command
- Signed-off-by: Sage Weil <sage@redhat.com>
- 11:30 AM Revision 9ea05223 (ceph): ceph-dencoder: clean up macro name
- Signed-off-by: Sage Weil <sage@redhat.com>
- 11:30 AM Revision aaa5b754 (ceph): ceph-dencoder: mark PGMap with nondeterministic encoding
- Signed-off-by: Sage Weil <sage@redhat.com>
- 11:29 AM Revision 82533aa4 (ceph): test/encoding: c++11 STL appears to make fewer copies, yay
- The failure was
test/encoding.cc:191: Failure
Value of: 10
Expected: my_key_t::get_copy_ctor()
Which is: 5
test/enco... - 10:47 AM Bug #12651 (Won't Fix): osd crashes at startup if current/commit_op_seq is not readable
- on master 281d45d2ad449d38a0ec3be668d9bba7735b9d21...
- 04:49 AM Revision e06046c3 (ceph): cmake: check for good yasm
- this enables faster crc32 calculation on x64 intel platforms,
see common/crc32c_intel_fast.c
Signed-off-by: Kefu Cha... - 04:23 AM Revision 281d45d2 (ceph): Merge pull request #5500 from oritwas/wip-cmake-isa
- cmake: check for better yasm and fix isa
Reviewed-by: Kefu Chai <kchai@redhat.com> - 12:13 AM Revision 317559df (ceph): Added singleton hammer-x test to address #12625
- Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
08/07/2015
- 10:14 PM Revision 016a5d5d (ceph): mon/PGMap: make PGMap dump osd_epochs too
- Signed-off-by: Sage Weil <sage@redhat.com>
- 08:25 PM Revision 841ae130 (ceph): mon/OSDMonitor: prevent old OSDs from starting if 'sortbitwise' is enabled
- Signed-off-by: Sage Weil <sage@redhat.com>
- 08:23 PM Revision bf49bb3e (ceph): osd/osd_types: pg_interval_t: a change in sort order triggers a new int...
- Signed-off-by: Sage Weil <sage@redhat.com>
- 08:23 PM Revision 0031e0e6 (ceph): osd/PG: select interval sort order based on OSDMap
- If the OSDMap says sort bitwise, sort bitwise... unless we have
enabled osd_debug_randomize_hobject_sort_order, in wh... - 08:23 PM Revision f02d7e7a (ceph): mon/OSDMonitor: sort bitwise on new clusters
- Signed-off-by: Sage Weil <sage@redhat.com>
- 08:23 PM Revision 383185bf (ceph): mon/OSDMonitor: osd set/unset sortbitwise
- Add monitor command to flip the switch on the OSD hobject_t sort
order.
Signed-off-by: Sage Weil <sage@redhat.com> - 08:02 PM Revision 3092a18f (ceph): cmake: check for better yasm and fix isa
- Signed-off-by: Orit Wasserman <owasserm@redhat.com>
- 08:01 PM Revision 968261b1 (ceph): osd/OSDMap: add a SORTBITWISE OSDMap flag
- This flag will indicate that hobject_t's shall hence-forth be
sorted in a bitwise fashion.
Signed-off-by: Sage Weil ... - 07:58 PM Revision c2fc74d1 (ceph): osd/ReplicatedPG: fix filter lifecycle in do_osd_ops
- If we take an error path in [N]PGLS (objects_list_partial error
or bad handle) and FAILOK is set we may exec a second... - 07:58 PM Revision 62214832 (ceph): fix 'rados ls' issue if using keyvaluestore as objectstore backend
- Fixes: #12637
Signed-off-by: xinxin shu <xinxin.shu@intel.com> - 07:32 PM Revision dcd6e964 (ceph): ceph-object-corpus: do not try to decode pre-hammer ObjectStore::Transa...
- We don't understand collection_attr any more.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:44 PM Bug #12570 (Fix Under Review): common/Thread:pthread_attr_destroy(thread_attr) when done with it
- https://github.com/ceph/ceph/pull/5433
- 06:17 PM Revision 22e367b4 (ceph): Merge pull request #5510 from yuyuyu101/fix-async-fix-22
- AsyncConnection: Fix wrong order of local message delivering
Reviewed-by: Sage Weil <sage@redhat.com> - 06:14 PM Bug #12638: build_incremental() could take 40678 ms to finish
- sam suggested,
> we can also change the
>
> - mon_leveldb_cache_size // increase this value to see if it remed... - 08:10 AM Bug #12638: build_incremental() could take 40678 ms to finish
- ...
- 05:49 PM Revision 49aec61f (ceph): Merge pull request #5512 from ddiss/rbd_initrd_no_tr
- rbdnamer: drop unneccessary tr usage
Reviewed-by: Sage Weil <sage@redhat.com> - 05:41 PM Revision b98b1048 (ceph): rbdnamer: drop unneccessary tr usage
- tr is not present by default in Dracut generated initramfs images. The
same logic can anyhow be performed via the exi... - 04:59 PM Revision 144792ef (ceph): Merge pull request #5310 from ceph/wip-pipefd-leak
- Wip pipefd leak
Reviewed-by: Jianpeng Ma <jianpeng.ma@intel.com> - 04:14 PM devops Bug #12382: Package python-ceph is obsoleted by python-rados, but obsoleting package does not pro...
- Until I can make time to implement and test the plan I described @ http://www.spinics.net/lists/ceph-devel/msg24497.h...
- 03:56 PM Revision 180a500a (ceph): Merge pull request #5511 from ceph/wip-cxx11
- c++11 ftw
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Sage Weil <sage@redhat.com> - 03:42 PM Revision 3260adf1 (ceph): c++11: xlist needs a value_type for back_inserter
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:42 PM Revision 11a5d292 (ceph): c++11: fixes for preprocessor string concatenation
- c++11 adds user-defined literals through type suffixes. for example,
"foo"s is a literal of type std::basic_string. t... - 03:42 PM Revision f4d2032c (ceph): c++11: disambiguate std::isnan
- c++11's isnan() is a constexpr function, while c's is a macro
Signed-off-by: Casey Bodley <cbodley@redhat.com> - 03:42 PM Revision 3a680804 (ceph): c++11: remove hash_namespace.h and default to std
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:42 PM Revision 511106e3 (ceph): c++11: fix shared_ptr conversions to bool
- operator bool() is marked as explicit, which prevents it from being used
directly in gtest macros
Signed-off-by: Cas... - 03:42 PM Revision f7c478c6 (ceph): c++11: replace boost::assign with initializer list
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:38 PM Revision ec619c7d (ceph): Merge pull request #4832 from ceph/wip-11511
- osd: Even in objects_read_sync() case don't read past oi.size
Reviewed-by: Samuel Just <sjust@redhat.com> - 03:16 PM Linux kernel client Support #12648: [old topic] OSD will stuck if running kernel rbd or rbd-fuse on the same node wit...
- This looks like a bunch of processes stuck trying to acquire mmap_sem, while another bunch is waiting for page writeb...
- 10:32 AM Linux kernel client Support #12648 (New): [old topic] OSD will stuck if running kernel rbd or rbd-fuse on the same no...
- I know this is an old topic and not recommended all the time, but we are limited to the hardware quantity and this wi...
- 03:13 PM Revision 9cdc3278 (ceph): AsyncConnection: Fix wrong order of local message delivering
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 02:59 PM Revision 97882853 (ceph): Merge pull request #445 from dachary/wip-11666-erasure-code-lrc
- erasure-code: lrc plugin workload
Reviewed-by: Samuel Just <sjust@redhat.com> - 02:17 PM Revision eaa2312e (ceph): osd: re-sort object_contects when the sort order changes
- Scrub uses the object_contexts sort order to check whether it is
safe to extend the scrub interval (verifying there a... - 02:17 PM Revision a78cef4f (ceph): common/shared_cache: allow comparator to be adjusted
- Allow the use of a comparator that has a value, and for that
comparator to be adjusted at runtime for the purposes fo... - 02:16 PM Revision 3c507667 (ceph): osd/ReplicatedPG: dynamically vary sort order for backfill sets and maps
- Reset the map and set so that the comparator matches the sort order at the
start of each backfill.
Signed-off-by: Sa... - 02:16 PM Revision 1f151468 (ceph): osd/ReplicatedPG: fix comparator resets
- We need to take care not to clear() any of the maps or sets or else we'll lose
the comparator.
Signed-off-by: Sage W... - 02:16 PM Revision 5ea8290c (ceph): osd/ReplicatedPG: better debug on recover_backfill
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 8cc861f2 (ceph): osd/ReplicatedPG: assert map and set sort order is correct during backfill
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 03de3b16 (ceph): osd/PG: do some new interval work on init() too
- Some work we need to do on any new interval, *including* when the PG is
first created and there is no "interval chang... - 02:16 PM Revision c5a1709e (ceph): osd/PG: starting or continuing backfill .. be more precise
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 526b63fc (ceph): osd/PG: fix forced backfill when last_backfill sort order doesn't match
- We need to force a backfill restart even if last_update is up to date.
Signed-off-by: Sage Weil <sage@redhat.com> - 02:16 PM Revision 51a02db3 (ceph): osd/ReplicatedPG: drop old hobject_t encoding compatibility
- This was introduced in 0.47 ... we don't need it any more!
Signed-off-by: Sage Weil <sage@redhat.com> - 02:16 PM Revision d90cc26b (ceph): osd: add and use operator<< for BackfillInterval
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 35b521a8 (ceph): osd/ReplicatedPG: fix a bit of whitespace
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 1f9e8b85 (ceph): osd/PG: use hash instead of filestore key for scrub
- It's doesn't matter, since it's a deterministic permutation of hte value
and we're checking for equality, but avoid t... - 02:16 PM Revision bee3ffef (ceph): osd/PG: note NIBBLEWISE in PG debug string
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 579d4f42 (ceph): include/encoding: add decode_noclear() variant for map<> decode with co...
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 1d66bbf8 (ceph): osd: dynamically vary sort order of BackfillInterval map
- We initialize all BackfillIntervals when backfill starts. At that point we
set the Comparator order via reset().
Si... - 02:16 PM Revision a208832c (ceph): explicitly specify comparator for all hobject_t maps and sets
- This is necessary now that operator< is not defined.
Signed-off-by: Sage Weil <sage@redhat.com> - 02:16 PM Revision 7624125d (ceph): explicitly specify comparator for all ghobject_t maps and sets
- We no longer have operator<
Signed-off-by: Sage Weil <sage@redhat.com> - 02:16 PM Revision cdffea95 (ceph): use explicit hobject_t comparators
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 64e00ba9 (ceph): use explicit ghobject_t comparators
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 49539ab8 (ceph): hobject_t: conditional Comparator object
- This one can go either bitwise or nibblewise, depending on what you give
pass to the STL container constructor. E.g.... - 02:16 PM Revision 29e140d8 (ceph): include/encoding: map<> encoders when comparator is specified
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision c56edf8e (ceph): include/types: map operator<< with comparator
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 9b8e536f (ceph): include/types: operator<< for set with comparator
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision a33895e0 (ceph): include/encoding: handle set<> with comparator
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 332e1a48 (ceph): common/shared_cache: parameterize the map<> comparator
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 456540f8 (ceph): common/sharedptr_registry: parameterize map<> comparator
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 35f90d98 (ceph): common/simple_cache: parameterize map<> comparator
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 6038aeff (ceph): hobject_t: typed MIN_* helpers
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision fd6cfa34 (ceph): osd/osd_types: clean up pg_info_t comments a bit
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 897d98f7 (ceph): osd: set acting, upacting features from osdmap, not peers
- Set in start_peering_interval based on the osdmap, so that it is defined
deterministically on primary, replica, and s... - 02:16 PM Revision 8230375b (ceph): osd: use accessor to set last_backfill
- Always will in last_backfill and last_backfill_bitwise together.
Signed-off-by: Sage Weil <sage@redhat.com> - 02:16 PM Revision 26d5d7a1 (ceph): osd: restart backfill on peers if last_backfill sort is off
- If a peer has a last_backfill value but the wrong sort order, we need to
restart backfill at the beginning.
Signed-o... - 02:16 PM Revision a36cae1a (ceph): osd/PG: do not trust last_backfill for purposes of missing objects if s...
- If the peer has the wrong sort order, we cannot trust lsst_backfill and we
do not know if the object is there.
Signe... - 02:16 PM Revision 25f0afc8 (ceph): osd: allow sort order to randomly revert to nibblewise via debug config
- Note that this option is misused and applied to some but not all OSDs they
will disagree on the sort order and get co... - 02:16 PM Revision 8063a9d3 (ceph): osd: ignore CRUSH_ITEM_NONE when calculating interval features
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 5e23f04c (ceph): common/hobject_t: remove comparison operators for [g]hobject_t
- *All* comparisions, including everything in the OSD, needs to
conditionally use either bitwise or nibblewise sort ord... - 02:16 PM Revision 466afca0 (ceph): os/FileStore: use bitwise sort when we don't care (e.g., split)
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision bbd10446 (ceph): os/FileStore: pass sort_bitwise down to CollectionIndex
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision c087d414 (ceph): os/LFNIndex: pass through sort_bitwise
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision ea4fe762 (ceph): os/HashIndex: handle legacy nibblewise sort
- We need to separate the bitwise and nibblewise implementations because
they diverge too much. Start by converting th... - 02:16 PM Revision 48503d84 (ceph): os/HashIndex: handle new bitwise sorting
- Handle listing by hash with bitwise sorting. This involves futzing with
the hex digit sort order by reversing the di... - 02:16 PM Revision 24fdbd11 (ceph): ceph_test_objectstore: call+test new collection_list variants
- In particular, make SimpleListTest check both sorting methods and asssert
that the result is sorted.
Signed-off-by: ... - 02:16 PM Revision e50d2a7b (ceph): hobject_t: pad hash to 8 digits
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision d5ca7e9d (ceph): osd/osd_types: add last_backfill_bitwise flag to pg_info_t
- We need to know what sort order last_backfill is based on.
Signed-off-by: Sage Weil <sage@redhat.com> - 02:16 PM Revision f72ba6e8 (ceph): hobject_t: restore default comparators ... bitwise!
- Note that this breaks compatibility for the moment.
Signed-off-by: Sage Weil <sage@redhat.com> - 02:16 PM Revision 73b3ed87 (ceph): osd: add OSD_BITWISE_HOBJ_SORT feature
- We will sort [g]hobject_t's bitwise (instead of nibblewise) if all
OSDs who have participated in peering support the ... - 02:16 PM Revision 8aaa437f (ceph): os: add sort type to ObjectStore interface (incomplete)
- Note that this doesn't build, but it will be easier to review this way.
Signed-off-by: Sage Weil <sage@redhat.com> - 02:16 PM Revision f0f87da9 (ceph): ceph-objecstore-tool: use new api
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision f459bdc5 (ceph): qa: fix misc collection_list callers
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision c5bd0347 (ceph): osd: use bitwise collection_list sort when we don't care
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 464bd6b9 (ceph): osd/PGBackend: ask PG which sort order to use
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 744508ff (ceph): os/MemStore: only support bitwise sort order
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:16 PM Revision 7d627398 (ceph): os/KeyValueStore: only support bitwise sort
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:15 PM Revision 89cb2bba (ceph): os/memstore: fix omap_rmkey_range
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:15 PM Revision c4128f13 (ceph): os/HashIndex: drop unused lower_bound arg
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:15 PM Revision c5893284 (ceph): os/LFNIndex: return vector from list_subdirs
- We'll need this later when we sort the subdirs bitwise instead of
nibblewise.
Signed-off-by: Sage Weil <sage@redhat.... - 02:15 PM Revision 8d8eb2b4 (ceph): hobject_t: nibblewise and bitwise comparators
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:15 PM Revision 7811c776 (ceph): Merge pull request #5331 from liewegas/wip-collection-list
- os: clean up collection_list
Reviewed-by: Samuel Just <sjust@redhat.com> - 02:04 PM Bug #12373 (Resolved): sgdisk hangs on zap command
- Closing this as the fix went already into master (and backported to hammer and firefly):
https://github.com/osynge... - 01:35 PM Revision 1beab53c (ceph): Merge pull request #5271 from wonzhq/cache-rate
- mon: add cache tier IO rate in the 'osd pool stats' command
Reviewed-by: Sage Weil <sage@redhat.com> - 12:56 PM Bug #12436 (Resolved): blkdev.cc::get_device_by_uuid() leak
- 12:55 PM Revision 2a760538 (ceph): Merge pull request #5507 from athanatos/wip-12436
- blkdev.cc::get_device_by_uuid: do not leak cache
Reviewed-by: Sage Weil <sage@redhat.com> - 12:55 PM Bug #12338 (Pending Backport): long standing slow requests: connection->session->waiting_for_map-...
- 12:55 PM Revision 01d3aa7c (ceph): Merge pull request #5506 from athanatos/wip-12338
- OSD: break connection->session->waiting message->connection cycle
Reviewed-by: Sage Weil <sage@redhat.com> - 12:45 PM Revision 7ee3eeff (ceph): Kill collection_list in CollectionIndex
- Now we use CollectionIndex.collection_list_partial.
Signed-off-by: Xiaoxi Chen <xiaoxi.chen@intel.com> - 12:45 PM Revision e46a855e (ceph): s/collection_list_impl/collection_list/
- Signed-off-by: Sage Weil <sage@redhat.com>
- 12:45 PM Revision cdb1fc01 (ceph): os: drop snapid_t arg to collection_list
- All this does is filter out objects with old snapids, but this can be
just as easily done in the one caller that spec... - 12:45 PM Revision 654c16ed (ceph): os/ObjectStore: drop get_ideal_list_min()
- Unused. Fix comment on get_ideal_list_max while we are here.
Signed-off-by: Sage Weil <sage@redhat.com> - 12:45 PM Revision 5df688a5 (ceph): Kill Flat_index.
- Too old, remove it entirely.
Signed-off-by: Xiaoxi Chen <xiaoxi.chen@intel.com> - 12:45 PM Revision c5dc4047 (ceph): Kill collection_list
- use collection_list_impl as well.
Signed-off-by: Xiaoxi Chen <xiaoxi.chen@intel.com> - 12:45 PM Revision 7bf999cf (ceph): os/KeyValuestore:Refactor collection_list_range and collection_list_par...
- Add collection_list_impl which abstract the common process
of collection list behavior.
Signed-off-by: Xiaoxi Chen <... - 12:45 PM Revision 71655109 (ceph): Kill collection_list_range
- We can directly use collection_list_impl.
Signed-off-by: Xiaoxi Chen <xiaoxi.chen@intel.com> - 12:45 PM Revision 2d5ed30f (ceph): Kill collection_list_partial
- Use collection_list_impl directly.
Signed-off-by: Xiaoxi Chen <xiaoxi.chen@intel.com> - 12:45 PM Revision 26668d6f (ceph): os/CollectionIndex: use const ref for end
- Signed-off-by: Sage Weil <sage@redhat.com>
- 12:44 PM Revision 80b6961d (ceph): Merge pull request #5509 from ceph/wip-12536
- common/hobject_t: correctly decode pre-infernalis hobject_t min
Reviewed-by: Samuel Just <sjust@redhat.com> - 12:43 PM Revision d171537b (ceph): os/Memstore:Refactor collection_list_range and collection_list_partial
- Add collection_list_impl which abstract the common process
of collection list behavior.
Signed-off-by: Xiaoxi Chen <... - 12:43 PM Revision 921c4586 (ceph): os/Filestore:Refactor collection_list_range and collection_list_partial
- Add collection_list_impl which abstract the common process
of coleection list behavior.
Todo: Refactor Index::collec... - 12:18 PM Bug #12410 (Pending Backport): OSDMonitor::preprocess_get_osdmap: must send the last map as well
- https://github.com/ceph/ceph/pull/5508
- 12:17 PM Revision ddc83b6b (ceph): Merge pull request #5508 from athanatos/wip-12410
- OSDMonitor::preprocess_get_osdmap: send the last map as well
Reviewed-by: Kefu Chai <kchai@redhat.com> - 11:51 AM Bug #12523 (Pending Backport): osd suicide timeout during peering - search for missing objects
- 11:50 AM Revision 8e0a4445 (ceph): Merge pull request #5413 from guangyy/wip-12523
- osd: copy the RecoveryCtx::handle for when creating RecoveryCtx instance from another one
Reviewed-by: Samuel Just <... - 11:26 AM Bug #12194: osd crash FAILED assert(!parent->get_log().get_missing().is_missing(soid))
- I can confirm what looks like this same exact bug. While still on 0.93, filesystem on OSDs had to be converted from ...
- 08:58 AM Bug #12340: ceph-helpers.sh: test_get_last_scrub_stamp: wait_for_clean fails
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/6928/consoleFull
http://jenkins.ceph.dachary.org/jo... - 08:42 AM CephFS Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
- please provide backtrace of the crash
- 07:33 AM CephFS Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
- Zheng Yan wrote:
> will do
HI,i use your patch in ceph dokan client. and seems cause to client assert crash.
D... - 07:23 AM Bug #12647 (Can't reproduce): Ceph log rotation backup strategy leads to log stop recording
- hi
I have a ceph logrotate problem:
Use ceph default logrotate. Conf strategy log backup regularly polling, would ... - 05:47 AM rgw Backport #12632 (In Progress): rgw: shouldn't return content-type: application/xml if content len...
- 05:47 AM rgw Backport #12634 (In Progress): swift smoke test fails on TestAccountUTF8
- 02:55 AM Revision 9471bb83 (ceph): Common/Thread: pthread_attr_destroy(thread_attr) when done with it
- When a thread attributes object is no longer required, it should be destroyed using the
pthread_attr_destroy() functi... - 02:26 AM CephFS Bug #12189 (Won't Fix): Editing / Creating files fails for NFS-over-CephFS on EC pool with cache ...
- 02:04 AM Revision e35d27fc (ceph): Merge pull request #5492 from ceph/wip-vxattr-setall
- mds: fix setting whole layout in one vxattr
- 01:47 AM rgw Bug #12629: the arguments 'domain' should not be assigned when return false
- master PR https://github.com/ceph/ceph/pull/5447
- 01:30 AM Bug #12645 (Can't reproduce): chunky_scrub assert fail
- Cluster:5 nodes, 4 mons ,each node has 10 osds
Ceph version:0.87
Operation:
(1)reboot node.1
(2)then unmount all... - 12:27 AM Revision e3147b8d (ceph): rgw:segmentation fault when rgw_gc_max_objs > HASH_PRIME
- obj_names[] is index out of bounds when rgw_gc_max_objs > HASH_PRIME
Fixes: #12630
Signed-off-by: Ruifeng Yang <1492... - 12:24 AM Revision 9420d24c (ceph): rgw:the arguments 'domain' should not be assigned when return false
- Hostnames: [B.A]
Inputs: [X.BB.A]
Return: [false]
Output: [B.A] it is wrong.
Fixes: #12629
Signed-off-by... - 12:21 AM Revision 2938fdd5 (ceph): radosgw-admin: Create --secret-key alias for --secret
- Fixes: #5821
Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>
08/06/2015
- 10:43 PM rbd Documentation #12642: snap_unprotect() requires access to all pools
- The reason we didn't store the children with the parent image is to let a user without write access to the parent poo...
- 09:04 PM rbd Documentation #12642 (Resolved): snap_unprotect() requires access to all pools
- Hi,
we have a setup with multiple users and pools which prevents the users from unprotecting their snapshots.
S... - 08:45 PM Bug #10138 (Can't reproduce): osd: crash in SnapSet::from_snap_set
- 08:41 PM Bug #10138: osd: crash in SnapSet::from_snap_set
- We did not encounter this again since upgrading to 0.9x so this can be closed.
- 05:21 PM ceph-dokan Bug #12513: TCMALLOC_MAX_TOTAL_THREAD_CACHE_BYTES environment variable defualt value 32 MB is eno...
- OK, got it.
- 04:28 PM Bug #12637: bug of rados ls with keyvaluestore
- https://github.com/ceph/ceph/pull/5499
- 04:08 PM Bug #12637 (Resolved): bug of rados ls with keyvaluestore
- if leveldb is usedas keyvaluestore backend , run 'rados -p rbd ls', then cluster crash , log shows as follow :
... - 04:25 PM CephFS Bug #12189: Editing / Creating files fails for NFS-over-CephFS on EC pool with cache tier
- I was able to update the kernel on the NFS server to version 4.1.4 today, which also allows me to use the kernel clie...
- 04:22 PM Stable releases Tasks #12639: radosgw + civetweb latency issue on Hammer
- Wrong queue, Please close this tracker,
created new one in the correct queue http://tracker.ceph.com/issues/12640 - 04:16 PM Stable releases Tasks #12639 (Rejected): radosgw + civetweb latency issue on Hammer
- After upgrading to Hammer and moving from apache to civetweb. We
started seeing high PUT latency in the order of 2 s... - 04:20 PM rgw Bug #12640 (Closed): radosgw + civetweb latency issue on Hammer
- After upgrading to Hammer and moving from apache to civetweb. We
started seeing high PUT latency in the order of 2 s... - 04:17 PM Revision b3f1b3ac (ceph): Merge pull request #5488 from oritwas/wip-cmake-fix-rbd
- cmake: add global lib to rbd
Reviewed-by: John Spray <john.spray@redhat.com> - 04:12 PM Bug #12638 (Resolved): build_incremental() could take 40678 ms to finish
- ceph version 0.80.7 (6c0127fcb58008793d3c8b62d925bc91963672a3)
it is reported that a monitor with over 800 osdmap ... - 04:11 PM Revision 3ab5d82a (ceph): rgw: set http status in civetweb
- Need to set the http status in civetweb so that we report it correctly.
Fixes: #12432
Signed-off-by: Yehuda Sadeh <y... - 04:11 PM Revision 10a03839 (ceph): civetweb: update submodule to support setting of http status
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 04:09 PM Revision 427f48bf (ceph): erasure-code: CEPH_FEATURE_ERASURE_CODE_PLUGINS_V3 integration tests
- http://tracker.ceph.com/issues/10887 Refs: #10887
Signed-off-by: Loic Dachary <ldachary@redhat.com> - 03:25 PM Revision 18bf415d (ceph): fix indentation error in docstring
- Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit af973259eff9ffbbbe9aa209c3b3dd2d80d7219b) - 03:25 PM Revision 51122026 (ceph): do not ommit syntax errors in linter checks
- Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit db69142b9078fd9ed35fb3a542fc340dae9be665) - 03:24 PM Revision 30f10ee1 (ceph): tasks: fix syntax error in ceph.py
- From e195f9fa.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 9f530092e2aa3bdc73fcddc0... - 03:10 PM Revision cd4ac1ca (ceph): rbd: support size suffixes for size-based options
- This patch will provide support for size suffixes for
below given options:
--stripe-unit <size in B/K/M>
--io-size <... - 02:59 PM rgw Backport #12635 (Resolved): swift smoke test fails on TestAccountUTF8
- https://github.com/ceph/ceph/pull/5532
- 02:58 PM rgw Backport #12634 (Resolved): swift smoke test fails on TestAccountUTF8
- https://github.com/ceph/ceph/pull/5718
- 02:56 PM rgw Bug #11438 (Pending Backport): swift smoke test fails on TestAccountUTF8
- 02:45 PM rgw Backport #12633 (Resolved): rgw: shouldn't return content-type: application/xml if content length...
- https://github.com/ceph/ceph/pull/5532
- 02:38 PM rgw Backport #12632 (Resolved): rgw: shouldn't return content-type: application/xml if content length...
- https://github.com/ceph/ceph/pull/5718
- 02:35 PM Documentation #12620 (In Progress): note the behaviour that "ceph mon add <mon-id>" takes forever...
- https://github.com/ceph/ceph/pull/5495
- 02:33 PM rgw Bug #11091 (Pending Backport): rgw: shouldn't return content-type: application/xml if content len...
- This introduced a regression as reported by this issue : http://tracker.ceph.com/issues/11438
- 02:24 PM rgw Bug #11091: rgw: shouldn't return content-type: application/xml if content length is 0
- PR : https://github.com/ceph/ceph/pull/4361
- 02:33 PM Bug #11266: data digest mismatch 11177
- Hi!
My ceph version 0.94.2
I have the same issues with 193 pgs inconsistent of 656 pgs total.
Will it be resolved ... - 02:27 PM Revision d1735a4e (ceph): rgw: rework X-Trans-Id header to be conform with Swift API.
- Fixes: #12108
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 01:59 PM Revision 278a6ae0 (ceph): qa: add fs layout case for stripe_size decrease
- The case fixed by commit:
mds: fix setting whole layout in one vxattr
Signed-off-by: John Spray <john.spray@redhat.... - 01:58 PM Revision 63e6d87d (ceph): calamari_setup: handle RHCS 1.2 (missing /mnt/MON)
- In RHCS 1.2 we don't have a /mnt/MON directory. The intention of
35c6363a1eaf5011e26be8561ffcd4e9324d13a2 was to hand... - 01:56 PM rgw Backport #12246 (In Progress): RGW Swift API: responses for several request types don't contain m...
- 01:56 PM Feature #10887 (Fix Under Review): erasure-code: allow upgrades for shec plugins
- 01:55 PM Feature #10887 (In Progress): erasure-code: allow upgrades for shec plugins
- https://github.com/ceph/ceph-qa-suite/pull/523
- 01:06 PM Feature #10887 (Fix Under Review): erasure-code: allow upgrades for shec plugins
- https://github.com/ceph/ceph/pull/5493
- 12:50 PM Feature #10887 (In Progress): erasure-code: allow upgrades for shec plugins
- 01:51 PM rgw Bug #12630 (In Progress): segmentation fault when rgw_gc_max_objs > HASH_PRIME
- master PR https://github.com/ceph/ceph/pull/5449
- 10:34 AM rgw Bug #12630 (Resolved): segmentation fault when rgw_gc_max_objs > HASH_PRIME
- 01:33 PM Revision 12f4dffc (ceph): erasure-code: add write_append_excl: false
- Add write_append_excl: false to all erasure-code workloads and use the
same workload in all upgrade tests. write_appe... - 01:29 PM Revision b63f2afa (ceph): Merge pull request #5475 from cxwshawn/com-fix
- rgw_user.h: modify interface comments.
Reviewed-by: Sage Weil <sage@redhat.com> - 01:25 PM Revision e61976b2 (ceph): Merge pull request #5483 from oddomatik/patch-1
- Fix location of --access=full in SWIFT user creation
Reviewed-by: Sage Weil <sage@redhat.com> - 01:25 PM Revision 0ab7926c (ceph): Merge pull request #5484 from cxwshawn/de-fix
- ceph-dencoder: add RGWRegion, RGWZoneParams, RGWOLHInfo support.
Reviewed-by: Sage Weil <sage@redhat.com> - 01:25 PM Revision 995cc922 (ceph): Merge pull request #522 from ceph/syntax-check
- Include `SyntaxError` checks in linter
Reviewed-by: John Spray <john.spray@redhat.com> - 01:12 PM Revision b436a35f (ceph): Merge pull request #5421 from ceph/wip-selinux-policy-no-user
- SELinux support for ceph
Reviewed-by: Kefu Chai <kchai@redhat.com>
Tested-by: Yuri Weinstein <yweinstei@redhat.com> - 12:41 PM Revision af973259 (ceph): fix indentation error in docstring
- Signed-off-by: Alfredo Deza <adeza@redhat.com>
- 12:38 PM Revision db69142b (ceph): do not ommit syntax errors in linter checks
- Signed-off-by: Alfredo Deza <adeza@redhat.com>
- 12:14 PM Revision 880ffe9d (ceph): mds: fix setting whole layout in one vxattr
- Previously we were validating the layout
after setting each field, so even when
setting multiple fields in one go (to... - 11:16 AM CephFS Bug #12619: valgrind errors in fs suite
- 08:15 AM CephFS Bug #12619 (Fix Under Review): valgrind errors in fs suite
- https://github.com/ceph/ceph/pull/5487
- 10:17 AM rgw Bug #12629 (Resolved): the arguments 'domain' should not be assigned when return false
- in the function RGWREST::preprocess, if the rgw_find_host_in_domains return false, the domain may be have a wrong val...
- 10:09 AM Revision 57f2e060 (ceph): Merge pull request #521 from ceph/wip-fix-ceph
- tasks: fix syntax error in ceph.py
Reviewed-by: Kefu Chai <kchai@redhat.com> - 09:49 AM Revision b34363ac (ceph): cleanup: fix the eol dumping in JSONFormatter flush & close_section
- Signed-off-by: Aran85 zhangzengran@h3c.com
- 09:25 AM Revision 9f530092 (ceph): tasks: fix syntax error in ceph.py
- From e195f9fa.
Signed-off-by: John Spray <john.spray@redhat.com> - 09:14 AM CephFS Bug #12617: Missing handling of supplementary groups in ceph-fuse -> rendering it useless for sha...
- Burkhard Linke wrote:
> The following patch adds correct supplementary group retrieval:
>
> If no callback is def... - 09:10 AM Revision 0de2ea65 (ceph): Merge pull request #5487 from ceph/wip-12619
- mds: initialize InodeStoreBase::damage_flags
Reviewed-by: John Spray <john.spray@redhat.com> - 08:20 AM Revision 1559d5e9 (ceph): cmake: add global lib to rbd
- Signed-off-by: Orit Wasserman <owasserm@redhat.com>
- 08:14 AM Revision 6b292330 (ceph): mds: initialize InodeStoreBase::damage_flags
- Fixes: #12619
Signed-off-by: Yan, Zheng <zyan@redhat.com> - 02:48 AM Revision 5d7cb4c2 (ceph): ceph-dencoder: add RGWRegion, RGWZoneParams, RGWOLHInfo support.
- since ceph-dencoder does not support the up-list structures, so
I add these three supports to analyze data more q...
08/05/2015
- 11:25 PM Revision e67539e3 (ceph): Fix location of --access=full in SWIFT user creation
- --access=full was incorrectly placed in the output of the command, while it was supposed to have been a flag run with...
- 09:38 PM Revision 3489e5cd (ceph): Set the SELinux context of the logrotate config
- Signed-off-by: Zack Cerza <zack@redhat.com>
(cherry picked from commit e195f9fa1c3415157ce7be0bb50aead9794f49f1) - 09:30 PM Revision fd39a7e0 (ceph): Merge pull request #520 from ceph/wip-logrotate-chcon
- Set the SELinux context of the logrotate config
- 08:44 PM RADOS Bug #12405: filestore: syncfs causes high cpu load due to kernel implementation in high-memory boxes
- Try this: reboot the nodes, and
echo 10000 > /proc/sys/vm/vfs_cache_pressure
That should favor data (page ca... - 08:13 PM RADOS Bug #12405: filestore: syncfs causes high cpu load due to kernel implementation in high-memory boxes
- nevermind, new kernels do the same. i think the reason why it is problematic on these nodes is because they have gob...
- 08:02 PM RADOS Bug #12405: filestore: syncfs causes high cpu load due to kernel implementation in high-memory boxes
- 'perf record -ag -F 99 -- sleep 10' sez...
- 07:58 PM rgw Feature #12606 (In Progress): rgw: multisite: keep metadata zone sync state
- 07:38 PM Bug #12614: recursive lock of md_config_t (0)
- ubuntu@teuthology:/a/samuelj-2015-08-04_19:30:50-rados-wip-sam-testing-distro-basic-multi
- 07:34 PM Revision d8516694 (ceph): Merge pull request #519 from ceph/wip-silent-siganl
- tasks/ceph_manager: be silent about sending SIGHUPs
- 07:12 PM Feature #12600: wip-osd-compat: ensure upgrades stop at hammer
- 07:12 PM Feature #12599 (Resolved): wip-systemd
- 06:18 PM Revision 00d802d9 (ceph): hobject_t: fix get_boundary to work with new sorting regime
- The hash is no longer the most significant field; set everything that is
more significant, too.
Signed-off-by: Sage ... - 05:28 PM rbd Feature #12111 (Fix Under Review): rbd: support size suffixes for all size-based options
- https://github.com/ceph/ceph/pull/5481
- 05:06 PM rgw Bug #12623 (Closed): rgw: all worker threads hung with peering PG
- The discussion around the feature could be found here:
http://thread.gmane.org/gmane.comp.file-systems.ceph.devel/26407 - 04:14 PM Bug #9806: Objecter: resend linger ops on split
- That's a relief! Thanks for the explanation, I hope other people stumbling over this bug will find this helpful, too. :)
- 03:52 PM Bug #9806: Objecter: resend linger ops on split
- Yes, restarting qemu will fix it. The trigger for the issue is pg split, so it would only affect pools where you had ...
- 08:56 AM Bug #9806: Objecter: resend linger ops on split
- Ok, so I restarted one of the VMs exiting Qemu and starting afresh. Took a snapshot immediately after that and it's b...
- 08:40 AM Bug #9806: Objecter: resend linger ops on split
- Ah. So in that case restarting Qemu would be the specific action for that, right?
I'm currently trying this out. I... - 03:37 AM Bug #9806: Objecter: resend linger ops on split
- A workaround is to detach and reattach your images. This reopens them and reestablishes the watch.
- 04:14 PM Documentation #11082: Crushmap documentation still references raid4
- Checked the "doc/rados/operations/crush-map.rst" file, and it looks like all the RAID 4 references have already been ...
- 04:08 PM Revision e195f9fa (ceph): Set the SELinux context of the logrotate config
- Signed-off-by: Zack Cerza <zack@redhat.com>
- 03:55 PM Revision 8c53a582 (ceph): ceph.spec.in: Make SELinux opt-out, not opt-in
- We should enable SELinux support by default on rpm-based systems that
build with SELinux support to get better QE cov... - 03:16 PM Revision 93cb5472 (ceph): Merge pull request #518 from ceph/fix-async-test
- Fix experimental feature override makes user config useless
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 03:06 PM Revision f4853d88 (ceph): Merge pull request #5478 from dachary/wip-test-rados-striper
- tests: rados striper tests use 7116 instead of 7113
Reviewed-by: Sebastien Ponce <sebastien.ponce@cern.ch> - 02:48 PM Documentation #12620 (Resolved): note the behaviour that "ceph mon add <mon-id>" takes forever in...
- in a single-monitor cluster, if user tries to add another monitor using (take the vstart cluster as an example)
<pre... - 02:34 PM Revision 1092ed95 (ceph): Merge pull request #5459 from cxwshawn/master
- radosgw-admin: use cout instead of cerr to print help message.
Reviewed-by: Sage Weil <sage@redhat.com> - 02:29 PM CephFS Bug #12617: Missing handling of supplementary groups in ceph-fuse -> rendering it useless for sha...
- see #3301, and commit:328bf482349485b90911fe8953d8ce5f0769c69e for when this was introduced... hopefully we aren't re...
- 02:27 PM CephFS Bug #12617: Missing handling of supplementary groups in ceph-fuse -> rendering it useless for sha...
- The following patch adds correct supplementary group retrieval:
If no callback is defined, the getgrouplist() func... - 02:25 PM CephFS Bug #12617 (Fix Under Review): Missing handling of supplementary groups in ceph-fuse -> rendering...
- https://github.com/ceph/ceph/pull/5480
- 11:30 AM CephFS Bug #12617: Missing handling of supplementary groups in ceph-fuse -> rendering it useless for sha...
- I checked the kernel code, It seem that kernel also does the permission check. can we just remove the permission in c...
- 08:06 AM CephFS Bug #12617 (Resolved): Missing handling of supplementary groups in ceph-fuse -> rendering it usel...
- During investigating a user's bug report I've stumbled accross
https://github.com/ceph/ceph/blob/master/src/client... - 02:10 PM Revision 51aae7b5 (ceph): client: ignore permission check when fuse_default_permissions is on
- When fuse_default_permissions is on, linux kernel does permission check.
So there is no need to do extra permission c... - 01:33 PM Revision 4866d89c (ceph): osd: make PGLSFilter xattr read optional
- The cephfs filter needs to be able to express
interest in an xattr, but not reject objects
which do not have it.
Sig... - 01:30 PM Revision 60e903f3 (ceph): osd: enable PGLS filters to see hobject_t
- We will want this for CephFS, where the name's suffix
(i.e. .00000000) is important to us.
Signed-off-by: John Spray... - 01:21 PM Revision 9db80da1 (ceph): Fix selinux context after intitial OSD mount.
- The SELinux context is not preserved across file-systems, we need to
manually set the context of the root of the file... - 01:21 PM Revision d0fd8ffa (ceph): Update selinux policy (after local test).
- Changes enerated with ceph-test package.
Signed-off-by: Milan Broz <mbroz@redhat.com> - 01:21 PM Revision bed57033 (ceph): selinux: Allow setuid and setgid to ceph-mon and ceph-osd
- Signed-off-by: Boris Ranto <branto@redhat.com>
- 01:21 PM Revision c014f2b6 (ceph): ceph-disk: set selinux context
- Ensure that SELinux context is properly set according to policy.We
need this as SELinux context is not preserved acro... - 01:21 PM Revision 03d7a65b (ceph): SELinux Makefile can't work in parallel
- We need to force single-core compilation of SELinux policy files in the
sub-make target as SELinux Makefile does not ... - 01:21 PM Revision 73bf34d9 (ceph): selinux: Update the SELinux policy rules
- Few new denials were found while testing the policy. Updating the policy
rules to refelct that.
Signed-off-by: Boris... - 01:21 PM Revision c6d6c783 (ceph): ceph.spec.in: stop/start service on policy upgrade/removal
- Signed-off-by: Boris Ranto <branto@redhat.com>
- 01:21 PM Revision 736fe062 (ceph): selinux: Add .gitignore file
- The gitbuilders release script needs this. Otherwise, the ceph-release
build will fail because there were some untrac... - 01:21 PM Revision c52eb995 (ceph): Add initial SELinux support
- This patch modifies the build system and spec file to provide a support
for SELinux enforcing in an opt-in matter via... - 01:06 PM Revision 01b38737 (ceph): Merge pull request #4860 from ceph/wip-nobjectiterator-crash
- librados: fix empty NObjectIterator crash on comparisons
Reviewed-by: Sage Weil <sage@redhat.com> - 01:03 PM Revision c40df505 (ceph): librados: fix empty NObjectIterator crash on comparisons
- This would segfault if you did a
NObjectIterator i;
i == ...
Fix comparisons by allowing pointer-wise comparisons
fo... - 12:46 PM Revision b04bafc0 (ceph): tests: rados striper tests use 7116 instead of 7113
- Change src/test/libradosstriper/rados-striper.sh monitor address to 7116
because src/test/mon/osd-crush.sh uses 7113 ... - 12:35 PM CephFS Bug #12619 (Resolved): valgrind errors in fs suite
- Unclear if these are really CephFS bugs (the SyscallParam thing especially), but it's our tests they're coming up in....
- 12:18 PM rbd Bug #12618: incomplete rbd delete(command:rbd rm) damaged the rbd block
- Also, I found the RBD resize command can repair it
- 12:16 PM rbd Bug #12618 (Closed): incomplete rbd delete(command:rbd rm) damaged the rbd block
- #When I try to delete a rbd block(the size is 20G),I use the command
rbd -p rbd rm myblock
#When this command is ... - 11:32 AM CephFS Bug #12578 (Resolved): multimds: mds/SnapRealm.cc: 238: FAILED assert(open)
- 10:54 AM CephFS Bug #12612: fuse jobs fail to start on centos7
- Have these centos7 nodes run other tests successfully? There's a bunch of spurious failures from missing centos7 pac...
- 10:14 AM CephFS Bug #12612 (Need More Info): fuse jobs fail to start on centos7
- ceph-fuse fails with -ETIMEOUT, no information why timeout happened. In this test, fuse_mount.py kept checking /sys/f...
- 08:35 AM Revision 21581a49 (ceph): Merge pull request #5474 from kylinstorage/wip-sparse_read_bugfix
- osd/ReplicatedPG: sparse read should return (extent_map, data_bufferlist)
Reviewed-by: Kefu Chai <kchai@redhat.com>
... - 07:50 AM CephFS Bug #12598 (Fix Under Review): LibCephFS.GetPoolId failure
- https://github.com/ceph/ceph/pull/5477
- 05:46 AM Revision 28bc30c7 (ceph): osd/ReplicatedPG: sparse read should return (extent_map, data_bufferlist)
- Signed-off-by: Yunchuan Wen <yunchuanwen@ubuntukylin.com>
- 03:40 AM Revision 0cdd77d6 (ceph): rgw_user.h: modify interface comments.
- Signed-off-by: shawn chen <cxwshawn@gmail.com>
- 01:13 AM Revision 7d11b6b9 (ceph): Merge remote-tracking branch 'origin/next'
- Conflicts:
src/Makefile-client.am (link rbd to libglobal again) - 01:10 AM Revision add30145 (ceph): Revert "rbd: remove dependency on non-ABI controlled CephContext"
- This reverts commit fa78739933d7b9501116a12a4fbfd58b5ca99cdb.
Original approach broke krbd map/unmap functions and di... - 12:37 AM Bug #12200: assert(hinfo.get_total_chunk_size() == (uint64_t)st.st_size)
Changing get_hash_info() to return ECUtil::HashInfoRef() (NULL HashInfoRef) can cause this assert in submit_transac...
08/04/2015
- 11:25 PM RADOS Bug #12615 (New): Repair of Erasure Coded pool with an unrepairable object causes pg state to los...
After an erasure coded pool 2 + 1 with 2 chunks of a single object corrupted, doing a repair which can't succeed ca...- 11:12 PM Bug #12614: recursive lock of md_config_t (0)
- sjust@teuthology:/a/samuelj-2015-08-03_15:54:39-rados-wip-sam-testing-distro-basic-multi/996686
wip-sam-testing ru... - 11:12 PM Bug #12614 (Resolved): recursive lock of md_config_t (0)
- 2015-08-03T18:14:43.494 INFO:tasks.mon_recovery.ceph_manager:quorum is size 3
2015-08-03T18:14:43.494 INFO:tasks.mon... - 10:35 PM Bug #12613: hammer needs to set pool in hobject_t::get_boundary
- pushed as part of wip-12536-hammer
- 10:35 PM Bug #12613: hammer needs to set pool in hobject_t::get_boundary
- 45281efad1ae6b2b5ddd5549c56dbdd46671e5c5 will fix it if we backport it to hammer.
- 10:34 PM Bug #12613: hammer needs to set pool in hobject_t::get_boundary
- 10:33 PM Bug #12613 (Resolved): hammer needs to set pool in hobject_t::get_boundary
- 09:49 PM CephFS Bug #12612 (Can't reproduce): fuse jobs fail to start on centos7
- ...
- 08:08 PM Bug #9806: Objecter: resend linger ops on split
- As far as I understand this hurts snapshots. I'm on Firefly and getting bitten by this. Is there a work-around to get...
- 07:27 PM Bug #12581: osd: pg stuck at peering
- This may just be related to whatever messed up the osd directory (which seems unlikely to be the osd process).
- 07:17 PM Bug #12581: osd: pg stuck at peering
- /home/sam/git-checkouts/ceph2/src/osdmaptool: osdmap file 'osdmap.51341__0_4F1413C9__none__osd247'
parsed '3.8c7' -... - 06:53 PM Bug #12581: osd: pg stuck at peering
- Got some logs from OSD.43:...
- 04:07 PM Bug #12581: osd: pg stuck at peering
- Further digging at the peering log, I found the problem came from OSD.43, which did not response to a GetInfo in a ti...
- 06:02 AM Bug #12581: osd: pg stuck at peering
- Log file - https://www.dropbox.com/sh/g8d0hc0kqwt9yig/AABy9UqOnD7wWDAlpGvJuAC0a?dl=0
(I had problem to directly at... - 05:56 AM Bug #12581 (Can't reproduce): osd: pg stuck at peering
- Ceph version: v0.87
PG stuck at peering even though there is no write to the PG, looks like it stuck at GetInfo, a... - 07:20 PM rbd Bug #12611 (Resolved): object_map_update fails with -EINVAL return code
- The bit vector implementation incorrectly computes the number of bytes to read when decoding a data chunk. This resu...
- 07:14 PM rbd Bug #12604: rbd map failure in tasks/rbd_workunit_suites_iozone.yaml
- Should be addressed by https://github.com/ceph/ceph/pull/5444
- 04:03 PM rbd Bug #12604: rbd map failure in tasks/rbd_workunit_suites_iozone.yaml
- ...
- 03:15 PM rbd Bug #12604 (Resolved): rbd map failure in tasks/rbd_workunit_suites_iozone.yaml
- ...
- 06:23 PM Bug #12610: upgrade firefly->hammer results in object with empty name being reported as missing o...
- Ok, hammer actually does let you create these objects as well. We need a patch that allows them to be deleted, but n...
- 05:56 PM Bug #12610: upgrade firefly->hammer results in object with empty name being reported as missing o...
- http://tracker.ceph.com/issues/8587
Probably the above bug created the object. - 05:55 PM Bug #12610 (Won't Fix): upgrade firefly->hammer results in object with empty name being reported ...
- 1) hammer and firefly need to not allow objects with empty names
2) hammer during upgrade needs to detect such objec... - 06:05 PM Revision 13ccfb4e (ceph): Merge pull request #5446 from ceph/wip-systemd
- systemd support
Reviewed-by: Owen Synge <osynge@suse.com> - 05:36 PM rgw Feature #12608 (Resolved): rgw: multisite: per-shard full sync stage
- - read zone sync status
- if 'syncing', per shard:
- read shard status, if doing full sync, iterate through loc... - 05:30 PM rgw Feature #12607 (Fix Under Review): rgw:multisite: build metadata indexes for full sync
- - try to lock state object
- if cannot lock, wait, retry
- read state, if already 'building indexes', try to rem... - 05:23 PM Revision aebb9e79 (ceph): tools: ceph-release-notes unicode handling
- Prefix the print string with u so that it handles unicode gracefully.
Add the cli prefix (for ceph.in related pull r... - 05:21 PM Revision 9b91adc5 (ceph): mon: disallow post-hammer OSDs if there are up pre-hammer OSDs
- Force *all* OSDs to upgrade to hammer before allowing post-hammer
OSDs to join. This prevents any pre-hammer OSDs fr... - 05:16 PM Revision 8a559c1e (ceph): include/ceph_features: define MON_METADATA feature
- This was added in master in 40307748b9b8f84c98354a7141da645a3c556ba9. We
define it but notably do not support it!
S... - 05:13 PM Revision cdd5f770 (ceph): Merge pull request #5469 from theanalyst/doc/v0.94.3-release
- doc: release notes for v0.94.3
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 05:08 PM rgw Feature #12606 (Resolved): rgw: multisite: keep metadata zone sync state
- The metadata state consists of:
1. current state of the initiating process
- init
- building special index for... - 05:08 PM Revision c93fe1f1 (ceph): tasks/ceph_manager: be silent about sending SIGHUPs
- At the default interval this generates tons of log noise.
Signed-off-by: Sage Weil <sage@redhat.com> - 03:58 PM Backport #12583: Inconsistent PGs that ceph pg repair does not fix
- Sam marked 12577 Urgent, so I'm marking this one too. The pull request in the description here is the new pull reque...
- 10:19 AM Backport #12583: Inconsistent PGs that ceph pg repair does not fix
- David : if you think this is a backport that we need before releasing v0.94.3, please mark it as urgent and we'll loo...
- 08:02 AM Backport #12583 (In Progress): Inconsistent PGs that ceph pg repair does not fix
- Backport done by author, but targeting "hammer-backports-next" branch instead of "hammer".
- 08:01 AM Backport #12583 (Resolved): Inconsistent PGs that ceph pg repair does not fix
- https://github.com/ceph/ceph/pull/5468
- 03:57 PM Linux kernel client Bug #12579 (Duplicate): rbd map leads to system crash!
- Duplicate of #12538
- 03:56 PM Linux kernel client Bug #12527 (Duplicate): rbd map leads to system crash!
- Duplicate of #12538
- 03:53 PM Revision 793fe520 (ceph): doc: release notes for v0.94.3
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@ril.com>
- 03:51 PM Documentation #12441 (Won't Fix): There is no clear information regarding to replicas and higher ...
- I believe the documentation already contains the necessary information. If you think it does not, would you be so kin...
- 02:49 PM Feature #2127: Save kernel core dumps on all of our test machines
- 02:45 PM Feature #12602 (Resolved): c++11
- 02:45 PM Feature #12601 (In Progress): cmake
- 02:39 PM Feature #12600 (Resolved): wip-osd-compat: ensure upgrades stop at hammer
- 02:39 PM Feature #12599 (Resolved): wip-systemd
- 02:39 PM Revision 4e5d146a (ceph): osd: record digest if object is clean (vs entire scrub chunk)
- If an individual object is clean, record it in the missing_digest map.
If not, leave it out. This lets us drop the o... - 02:39 PM Revision f2002b7f (ceph): osd: be precise about "known" vs "best guess"
- We cannot assume that the auth info has the "known" digest; all replicas
may have a matching digest that does not mat... - 02:39 PM Revision 674029bc (ceph): osd: be slightly paranoid about value of okseed
- The scrubber.seed value is set based on the peer feature; check for that
too explicitly, and assert the scrubber.seed... - 02:39 PM Revision acfed6b5 (ceph): osd: move recorded vs on disk digest warning into be_compare_scrubmaps
- This is a better place for it. While we are here, change the wording to
clearly call out the 'recorded' digest (from... - 02:39 PM Revision d4f4c5c4 (ceph): osd: repair record digest if all replicas match but do not match
- If the recorded digest does not match the replicas, and all replicas
match, then (on repair) fix the recorded digest.... - 02:39 PM Revision 0f57c701 (ceph): osd: log a scrub error when we can't pick an auth object
- If we really can't find a suitable auth, log an error.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked fro... - 02:39 PM Revision 67e7946c (ceph): osd: fix fallback logic; move into be_select_auth_object
- The fallback behavior was printing a promising message but then doing
a 'continue' and not actually scrubbing the obj... - 02:39 PM Revision 455eb2a8 (ceph): osd: fix condition for loggin scrub errors
- We may have an error to log even though the authoritative map is empty:
when we have a digest that doesn't match all ... - 02:39 PM Revision 4faa8e04 (ceph): osd: include newlines in scrub errors
- We may log more than one of these in a scrubmap; make sure they are
not concatenated.
Signed-off-by: Sage Weil <sage... - 02:25 PM Linux kernel client Bug #12538: An image with a long clone chain blows up kernel stack
- 01:57 PM Revision de40c409 (ceph): client/Makefile: ship InodeRef
- Signed-off-by: Sage Weil <sage@redhat.com>
- 01:54 PM CephFS Bug #12598 (Resolved): LibCephFS.GetPoolId failure
- ...
- 01:13 PM Revision 653dbecd (ceph): Merge pull request #4753 from ceph/wip-client-inode-refs
- client: hold reference for returned inode
Reviewed-by: John Spray <john.spray@redhat.com> - 01:12 PM Revision 694990ff (ceph): Merge pull request #5466 from ceph/wip-12578
- mds: open base inode's snaprealm after decoding snapblob
Reviewed-by: John Spray <john.spray@redhat.com> - 01:03 PM rbd Backport #12597: Crash during shutdown after writeback blocked by IO errors
- *master PR*: https://github.com/ceph/ceph/pull/3625
- 01:02 PM rbd Backport #12597 (Resolved): Crash during shutdown after writeback blocked by IO errors
- https://github.com/ceph/ceph/pull/5767
- 12:52 PM rbd Bug #12596 (Resolved): Prevent object map errors from failing IO operations
- All errors encountered when manipulating the object map should be masked from the calling IO state machines. The "ob...
- 12:42 PM Backport #12595 (Resolved): hammer-x upgrade test.sh fails with ENOTEMPTY
- https://github.com/ceph/ceph/pull/5252
- 11:51 AM CephFS Bug #12578 (Fix Under Review): multimds: mds/SnapRealm.cc: 238: FAILED assert(open)
- https://github.com/ceph/ceph/pull/5466
- 11:28 AM Revision 0e69527f (ceph): mds: open base inode's snaprealm after decoding snapblob
- base inode has no parent, so can we mark its snaprealm open
immediately. This change makes sure replica mdsdir's snap... - 11:08 AM Revision 68e4d620 (ceph): Merge pull request #5140 from tchaikov/wip-pg-ls-update-pending-rel-notes
- doc: add changes to "pg ls*" commands to pending release notes
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 11:06 AM Revision 963c5241 (ceph): doc: add changes to "pg ls*" commands to pending release notes
- "pg ls*" commands now accept recovering instead of recovery as
one of its states.
Fixes: #11569
Signed-off-by: Kefu ... - 11:02 AM Revision 88e7ee71 (ceph): Merge pull request #5160 from theanalyst/wip-11910-hammer
- mon: pg ls is broken
Reviewed-by: Kefu Chai <kchai@redhat.com> - 10:40 AM rgw Backport #12502 (Resolved): rgw: need conversion tool to handle fixes following #11974
- 10:34 AM Revision bb12f925 (ceph): Merge pull request #5384 from dachary/wip-12502-hammer
- rgw: need conversion tool to handle fixes following #11974
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com> - 10:33 AM Bug #12577: Inconsistent PGs that ceph pg repair does not fix
- Hi David,
Thanks very much for the quick response/resolution. A couple of questions if you don't mind regarding t... - 12:59 AM Bug #12577 (Pending Backport): Inconsistent PGs that ceph pg repair does not fix
- 12:33 AM Bug #12577: Inconsistent PGs that ceph pg repair does not fix
- 10:19 AM Bug #11695: keyvaluestore: can not overwrite object after snapshot is created
- plz let me know if anything bother you :-)
- 09:59 AM rgw Documentation #12594: RGW pools naming convention
- https://github.com/ceph/ceph/pull/5464
- 09:37 AM rgw Documentation #12594 (Closed): RGW pools naming convention
- http://ceph.com/docs/master/radosgw/config/#create-pools
Current documentation for RGW pools naming convention is ... - 09:23 AM Backport #12520 (Resolved): Log::reopen_log_file() must take the flusher lock to avoid closing an...
- 09:22 AM Revision e801d4c9 (ceph): Merge pull request #5405 from ceph/wip-12465-hammer
- Log::reopen_log_file: take m_flush_mutex
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 09:14 AM rgw Backport #12593 (Resolved): HTTP return code is not being logged by CivetWeb
- https://github.com/ceph/ceph/pull/5498
- 09:12 AM rgw Feature #9604: rgw: create a tool for orphaned objects cleanup
- ...
- 09:10 AM rgw Feature #9604: rgw: create a tool for orphaned objects cleanup
- This is a very important tool for fixing problems, hence we make an exception to our "bugfixes only" backporting policy.
- 09:12 AM rgw Backport #12592 (Resolved): RGW returns requested bucket name raw in "Bucket" response header
- https://github.com/ceph/ceph/pull/5715
- 09:10 AM Revision 09839d00 (ceph): Allow evict operations to be throttled
- In agent_work(), it did not touch agent_ops when
it started an evict operation, so evict operations
were not throttle... - 09:10 AM rgw Backport #12591 (Resolved): rgw: create a tool for orphaned objects cleanup
- https://github.com/ceph/ceph/pull/5717
- 09:10 AM Revision caf98c83 (ceph): add agent_start/finish_evict_op to control agent_ops
- Signed-off-by: Yunchuan Wen <yunchuanwen@ubuntukylin.com>
Reviewed-by: Li Wang <liwang@ubuntukylin.com>
Reviewed-by: ... - 08:52 AM CephFS Backport #12590 (Resolved): "ceph mds add_data_pool" check for EC pool is wrong
- https://github.com/ceph/ceph/pull/5766
- 08:51 AM Backport #12589 (Resolved): ceph-disk zap should ensure block device
- https://github.com/ceph/ceph/pull/5755
- 08:50 AM Backport #12588 (Resolved): Change radosgw pools default crush ruleset
- https://github.com/ceph/ceph/pull/5754
- 08:49 AM Backport #12587 (Resolved): FileStore calls syncfs(2) even it is not supported
- https://github.com/ceph/ceph/pull/5530
- 08:49 AM Backport #12586 (Resolved): FileStore calls syncfs(2) even it is not supported
- https://github.com/ceph/ceph/pull/5529
- 08:47 AM Backport #12585 (Resolved): OSD crash creating/deleting pools
- https://github.com/ceph/ceph/pull/5527
- 08:46 AM Backport #12584 (Resolved): OSD crash creating/deleting pools
- https://github.com/ceph/ceph/pull/5526
- 07:59 AM Revision 311ef972 (ceph): buffer.cc: short-circuit copy_in for small lengths
- For small lengths (below 64 bytes) use shorter code paths, this
increases performance of function by up to 40%.
Sign... - 07:59 AM Revision ae2396ec (ceph): buffer.cc: postpone crc cache invalidation in bufferlist::rebuild()
- Postpone CRC cache invalidation. Do it only once after all data have
been moved.
Signed-off-by: Piotr Dałek <piotr.d... - 07:59 AM Revision d34e0415 (ceph): buffer.cc: add possibility to omit crc cache invalidation
- In some cases, it might be worthwile to either not invalidate crc
cache at all, or postpone it until we're done with ... - 07:59 AM Revision 472f7322 (ceph): buffer.cc: short-circuit in append for single char
- This increases performance of function by around 20-40% by reducing
call count during single-char appends.
Signed-of... - 07:56 AM Revision 33e1f17d (ceph): buffer.cc: short-circuit in copy_out for small lengths
- For small lengths (1, 2, 4 and 8 bytes) use shorter code paths, this
increases performance of function by around 40%.... - 07:56 AM Revision e79ee92e (ceph): buffer.cc: is_zero() optimization
- Instead of checking byte-by-byte, check by 16 bytes first (only on x64 and
only when compiled with GCC), then by 4 by... - 07:56 AM Revision 2219200c (ceph): src/include/encoding.h: reduce amount of 0-byte appends
- Some encoding functions (in particular, encoders for strings) right now
use two encode() calls, one for data length a... - 07:56 AM Revision 03877d68 (ceph): bufferlist: tuning small appends
- This change increases performance of small buffer::list::append and
buffer::ptr::append. In particular, the most comm... - 07:26 AM Revision 35e45694 (ceph): FileJournal: reduce time wasted by bufferptr::zero
- bufferptr::zero(), apart from actual memset() call, also invalidates its
internal CRC cache, which is empty at the ti... - 07:26 AM Revision 10a336fe (ceph): ObjectStore: partially remove op_ptr.zero()
- Partially remove zeroing of op_ptr buffer, it's not necessary because it
is overwritten later anyway, and bufferptr::... - 07:25 AM Bug #12570: common/Thread:pthread_attr_destroy(thread_attr) when done with it
- https://github.com/ceph/ceph/pull/5433
- 07:04 AM Revision a400b48b (ceph): Merge pull request #5337 from branch-predictor/bp-reduce-mallocs
- Thread.cc: remove malloc/free pair
Reviewed-by: Kefu Chai <kchai@redhat.com> - 07:03 AM Revision eeb660d0 (ceph): Merge pull request #5000 from ceph/wip-da-SCA-20150601
- Various SCA and Coverity fixes
Reviewed-by: Kefu Chai <kchai@redhat.com> - 06:34 AM Revision f68553e7 (ceph): osd/osd_types.cc: get rid of str concat when making hash key
- When namespaces are used, during hash key calculation there's temporary
string involved which is generated by concate... - 06:01 AM Bug #12429 (Pending Backport): OSD crash creating/deleting pools
- 06:00 AM Bug #12429 (Fix Under Review): OSD crash creating/deleting pools
- https://github.com/ceph/ceph/pull/5420
- 05:58 AM Revision 16385250 (ceph): Merge pull request #5312 from xinxinsh/wip-sparse-copy
- use SEEK_HOLE/SEEK_DATA for sparse copy
Reviewed-by: Haomai Wang <haomaiwang@gmail.com>
Reviewed-by: Sage Weil <sage... - 05:54 AM Bug #12580: Enclosure power failure pausing client IO
- Steps lead to hit the issue:
Step 1:
Populated some data
Cluster is in clean state & all OSD’s u... - 04:48 AM Bug #12580: Enclosure power failure pausing client IO
- BTW, crush failure domain was set to chassis level. Till the other chassis is powered up, we are not able to complete...
- 04:46 AM Bug #12580: Enclosure power failure pausing client IO
- To add more context to the problem:
Min_size was set to 1 and replication size is 2.
There was a flaky power co... - 04:39 AM Bug #12580 (Can't reproduce): Enclosure power failure pausing client IO
- I have active client IO running on cluster. (Random write profile with 4M block size & 64 Queue depth).
One of st... - 05:52 AM Revision 8f0028a9 (ceph): Merge pull request #5420 from ceph/wip-12429
- osd/OSDMap: handle incrementals that modify+del pool
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
Reviewed-by: Kefu... - 03:14 AM Revision 111ecf85 (ceph): radosgw-admin: use cout not cerr to print help message.
- everytime I print radosgw-admin help message, it's too long, and I wanna add | more,
but the help message prints to c... - 02:26 AM Revision 911c6865 (ceph): Fix experimental feature override makes user config useless
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 01:49 AM Revision 89c99db0 (ceph): Merge pull request #5431 from kylinstorage/test_cache_pool_script_fix
- fix: qa/workunits/rados/test_cache_pool.sh
Reviewed-by: Sage Weil <sage@redhat.com> - 01:48 AM rgw Bug #12537 (Pending Backport): [CVE-2015-5245] RGW returns requested bucket name raw in "Bucket" ...
- 01:47 AM Revision 75d80aa5 (ceph): Merge pull request #5430 from ceph/wip-12537
- rgw: url encode exposed bucket
Reviewed-by: Sage Weil <sage@redhat.com>
08/03/2015
- 11:54 PM Linux kernel client Bug #12579 (Duplicate): rbd map leads to system crash!
- #I create a rbd block named r1
rbd -p rbd create r1 --size 1024
#then create a snapshot of r1
rbd -p rbd snap cr... - 10:58 PM Bug #12577: Inconsistent PGs that ceph pg repair does not fix
- I'm pretty sure we have a fix for this in the development branch already (pull request 4365), but it will need to be ...
- 10:03 PM Bug #12577: Inconsistent PGs that ceph pg repair does not fix
- ...
- 09:16 PM Bug #12577: Inconsistent PGs that ceph pg repair does not fix
- I need the logs with increased log level, as I had updated by comment above.
- 08:53 PM Bug #12577: Inconsistent PGs that ceph pg repair does not fix
- ...
- 08:43 PM Bug #12577: Inconsistent PGs that ceph pg repair does not fix
- Could you attach a portion of the log of osd.78 around the time of a repair?
You probably need to bump the osd log... - 08:07 PM Bug #12577 (Resolved): Inconsistent PGs that ceph pg repair does not fix
- ...
- 09:54 PM Revision 6d01d6b7 (ceph): OSDMonitor::preprocess_get_osdmap: send the last map as well
- Fixes: #12410
Backport: hammer
Signed-off-by: Samuel Just <sjust@redhat.com>
(cherry picked from commit 1320e29dfaee9... - 09:54 PM Revision 1357ed1b (ceph): hobject_t: decode future hobject_t::get_min() properly
- The post-hammer wip-temp branch changed hobject_t::get_min() so that pool
is INT64_MIN instead of -1 and neglected to... - 09:03 PM CephFS Bug #12578 (Resolved): multimds: mds/SnapRealm.cc: 238: FAILED assert(open)
- 2015-08-02T07:54:05.552 INFO:tasks.ceph.mds.d.mira041.stderr:mds/SnapRealm.cc: In function 'void SnapRealm::check_cac...
- 07:16 PM CephFS Bug #12222 (Resolved): MDSMonitor: set max_mds doesn't respect MAX_MDS
- 12:56 PM CephFS Bug #12222 (Fix Under Review): MDSMonitor: set max_mds doesn't respect MAX_MDS
- https://github.com/ceph/ceph/pull/5397
- 07:15 PM Revision 3cbec352 (ceph): Merge pull request #5397 from ceph/wip-12222
- mon: reject over-large values of max_mds
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Joao Eduardo Luis <j... - 07:14 PM Revision 06d15464 (ceph): Merge pull request #5400 from tchaikov/wip-12451
- tools/rados: update the help message
Reviewed-by: Sage Weil <sage@redhat.com> - 06:41 PM Revision 145364bb (ceph): logrotate: fix log rotation with systemd
- systemctl does not have a nice way to enumerate (active) units so we can
reload them. On centos7, the is-active wild... - 06:05 PM Revision fb4e3321 (ceph): Merge pull request #5440 from vumrao/wip-vumrao-12458-doc
- doc: change --keyfile description in man page of rbd help
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 06:03 PM Bug #12575 (Duplicate): "Mutex.cc: 95: FAILED assert(r == 0)"
- Pull request #5333 has the fix, but is yet to merge.
- 04:44 PM Bug #12575 (Duplicate): "Mutex.cc: 95: FAILED assert(r == 0)"
- Run: http://pulpito.ceph.com/teuthology-2015-07-28_23:20:02-upgrade:client-upgrade-next-distro-basic-multi/
Jobs: ['... - 05:47 PM Bug #12428: garbage data in osd data dir crashes ceph-objectstore-tool
- You could try manually moving the pg directory somewhere for safe keeping, and then restart the osd.
- 05:36 PM Bug #12338: long standing slow requests: connection->session->waiting_for_map->connection ref cycle
- Guang: no, this request was sort of lingering because the client (radosgw) which started the request restarted.. so t...
- 04:46 PM Bug #12338: long standing slow requests: connection->session->waiting_for_map->connection ref cycle
- Hi Abhishek,
When you saw this issue occurred, did you see the upper layer (e.g. radosgw) hung at the same time? - 12:07 PM Bug #12338: long standing slow requests: connection->session->waiting_for_map->connection ref cycle
- Tried the patch.. seemed to solve the issue.. can we have a master pr & a hammer backport
- 05:27 PM Revision 85cb86d0 (ceph): doc: change "--keyfile" description in man page of rbd help
- "--keyring" has higher priority than "--keyfile".
"--keyfile" option is overridden by "--keyring"
if latter is also s... - 05:18 PM Revision 12c4e7da (ceph): Merge pull request #5455 from ceph/wip-cephfs-cmake
- CMake fixes for CephFS
Reviewed-by: Kefu Chai <kchai@redhat.com> - 04:54 PM Backport #12520: Log::reopen_log_file() must take the flusher lock to avoid closing an fd ::_flus...
- ...
- 04:34 PM Bug #12536: "FAILED assert(!log.null() || olog.tail == eversion_t())"
- Tentative fixes http://github.com/liewegas/ceph/commit/wip-12536 and http://github.com/liewegas/ceph/commit/wip-12536...
- 04:09 PM Revision b93b6726 (ceph): Merge pull request #5456 from kylinstorage/wip-do_autogen.sh
- Allow do_autogen.sh pass configure parameters
Reviewed-by: Sage Weil <sage@redhat.com> - 03:53 PM Revision 1ca6bf60 (ceph): common/hobject_t: correctly decode pre-infernalis hobject_t min
- In commit ff99af38df830e215359bfb8837bf310a2023a4d we changed
the value of hobject_t() (the min value) but did not ha... - 03:52 PM rgw Backport #12502: rgw: need conversion tool to handle fixes following #11974
#12564 is resolved, re-scheduling the last 19 jobs....- 03:40 PM rgw Bug #12574 (Closed): Firefly to Hammer RGW Upgrade
- Following an upgrade from Firefly to Hammer I noticed a number of buckets can no longer be listed using s3cmd as show...
- 03:39 PM Revision fe970bc1 (ceph): CMake: fix librados build
- Signed-off-by: John Spray <john.spray@redhat.com>
- 03:39 PM Revision bc0d9427 (ceph): CMake: add crushtool
- Signed-off-by: John Spray <john.spray@redhat.com>
- 03:39 PM Revision 2761b296 (ceph): CMake: add cephfs tools
- cephfs-table-tool, cephfs-data-scan, cephfs-journal-tool
Signed-off-by: John Spray <john.spray@redhat.com> - 03:39 PM Revision cc35d9b4 (ceph): CMake: fix cephfs test targets
- These had bad lib names.
Signed-off-by: John Spray <john.spray@redhat.com> - 03:39 PM Revision a4a81c33 (ceph): CMake: fix cephfs cls build
- Signed-off-by: John Spray <john.spray@redhat.com>
- 03:39 PM Revision 1f541b2e (ceph): CMake: set WITH_CEPHFS by default
- Option was referenced in CMakeLists but
not actually set.
Signed-off-by: John Spray <john.spray@redhat.com> - 03:39 PM Revision 57cd851d (ceph): CMake: smalliobenchrbd should dep on rbd
- Signed-off-by: John Spray <john.spray@redhat.com>
- 03:39 PM Revision 22e8a291 (ceph): CMake: libblkid not only needed with rbd
- It's linked in libcommon.
Signed-off-by: John Spray <john.spray@redhat.com> - 03:39 PM Revision 2355c454 (ceph): CMake: fix rbd build
- Don't link with 'global' from rbd (to match
autotools update fa78739). Fixes "g_ceph_context
defined twice" error.
... - 03:16 PM Revision b2e484d9 (ceph): cmake: build as c++11
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:16 PM Revision 51a218c8 (ceph): autoconf: build as c++11
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:16 PM Revision 2d18941f (ceph): c++11: remove references to the std::tr1 namespace
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:16 PM Revision 5b15ea2b (ceph): c++11: replace auto_ptr with unique_ptr
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:16 PM Revision 3f525834 (ceph): c++11: stream output operators for stream types
- stream output operators for the stream types are not part of the standard,
so this statement no longer works:
std::s... - 03:16 PM Revision 5416b1fb (ceph): c++11: fixes for std::pair type conversions
- mds_rank_t is no longer a BOOST_STRONG_TYPEDEF, because that was
preventing the conversion between pairs of int and p... - 03:15 PM devops Bug #12572 (Rejected): "VersionNotFoundError..ceph-deb-trusty-x86_64-basic"
- **Run: http://pulpito.ceph.com/teuthology-2015-08-02_23:20:02-upgrade:client-upgrade-next-distro-basic-multi/
Job: [... - 03:05 PM Backport #12571 (Resolved): "FAILED assert(!log.null() || olog.tail == eversion_t())"
- commit:1357ed1bd0e250b942bcba0346c97c24bb79a5d1
- 02:57 PM Revision ec8433df (ceph): Makefile: include systemd udev rules in tarball
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:37 PM Revision fdf0e030 (ceph): Merge pull request #5422 from ceph/wip-fix-filestore-getattr
- tests: fix the ObjectStore/StoreTest.SimpleCloneTest in "ceph_test_objectstore"
Reviewed-by: Loic Dachary <ldachary@... - 01:58 PM Bug #12570 (Resolved): common/Thread:pthread_attr_destroy(thread_attr) when done with it
- when a thread attributes object is no longer required, it should be destroyed using the pthread_attr_destroy() functi...
- 01:28 PM Revision 02590c1d (ceph): Merge pull request #5375 from ceph/wip-config-root-ino
- Make UID of new root inos configurable
- 01:16 PM Revision e4c6922c (ceph): add help to do_autogen.sh
- Signed-off-by: Yunchuan Wen <yunchuanwen@ubuntukylin.com>
Reviewed-by: Li Wang <liwang@ubuntukylin.com> - 01:03 PM Revision e28b3d11 (ceph): Allow do_autogen.sh to pass configure parameters
- Currently do_autogen.sh does not allow to pass parameters
of configure to disable/enable components during building, ... - 12:47 PM Bug #12569 (Won't Fix): "ceph mon add <mon-id>" takes forever in a one-monitor cluster
- This is expected behavior. There is little we can do without resorting to hackish behavior on update_from_paxos() and...
- 12:24 PM Bug #12569: "ceph mon add <mon-id>" takes forever in a one-monitor cluster
- probably we can live with this, but instead, add a note to the document that the "ceph mon add" command will not retu...
- 12:12 PM Bug #12569 (Won't Fix): "ceph mon add <mon-id>" takes forever in a one-monitor cluster
- in a single-monitor cluster, if user tries to add another monitor using (take the vstart cluster as an example)
<pre... - 10:21 AM rgw Documentation #12568 (Closed): update Install Ceph Object Gateway documentation to reflect Civetweb
- Although RGW now uses Civetweb instead of apache, the documentation has not caught up to this:
http://ceph.com/doc... - 09:21 AM Bug #12271 (Rejected): bug in ec pool chunk mapping
- not a bug, closing per xinze's request.
- 08:19 AM Bug #12567 (Can't reproduce): OSD:ENOENT on clone suggests osd bug
- This occurred in my osd.49 when 1 mon down.
Ceph version:0.87
Cluster:5 nodes(4 mons , each node has 11 OSDS)
He... - 08:12 AM Revision e7920c93 (ceph): client: convert CapSnap::in to smart pointer
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 08:12 AM Revision 89648e7e (ceph): client: convert Inode::snapdir_parent to smart pointer
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 08:12 AM Revision 8c45ae12 (ceph): client: use smart pointer to track 'cwd' and 'root_parents'
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 08:12 AM Revision 217837b8 (ceph): client: use smart pointer to track temporary inode reference
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 08:12 AM Revision dac11e6b (ceph): client: convert Fh::inode to smart pointer
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 08:02 AM Revision fd02f0f2 (ceph): client: use smart pointers in MetaRequest
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 07:42 AM Revision 07f58097 (ceph): client: convert Dentry::inode to smart pointer
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 07:42 AM Revision ad9c22a4 (ceph): client: hold reference for returned inode
- CLient::make_request() returns pointer of target inode, but it does not
increase reference count of the returned inod... - 07:41 AM Revision 6b17e210 (ceph): osd/recover_backfill: assert(obc) when adding pg stat for backfill objects
- The obc should always be on the primary.
Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com> - 07:35 AM Revision 199352df (ceph): osd/recover_primary: remove the unfound check when recovering an object
- The function recovering_missing checks if the object is unfound, there
is no need to double check it in recover_prima... - 05:39 AM rgw Bug #12542: radosgw deleted log leaks
- It rotate all files in .log format in /var/log/ceph/ dir.
but do not reload ceph-radosgw services in postrotate scri... - 05:32 AM rgw Bug #12542: radosgw deleted log leaks
- Thanks for the info. I see the postrotate only reload ceph service.
It should also reload ceph-radosgw service in ... - 05:09 AM rgw Bug #12542: radosgw deleted log leaks
- .gz file is generated by logrotate service, you can find the script in /etc/logrotate.d/ceph
- 05:03 AM rgw Bug #12542: radosgw deleted log leaks
- Jifeng Yin wrote:
> I know it exactly, but I didn't delete it manually.
> There is a same file named in '.gz' under... - 05:02 AM rgw Bug #12542: radosgw deleted log leaks
- I know it exactly, but I didn't delete it manually.
There is a same file in gzip format under the same directory, so... - 01:41 AM rgw Bug #12542: radosgw deleted log leaks
- If you delete a file that is opened by a process, then df will not changed because the file's inode wasn't deleted an...
- 04:09 AM Bug #12566: assert failed in OSD::ms_fast_dispatch tracepoint()
(gdb) bt
#0 0x0000003b8260f5db in raise () from /lib64/libpthread.so.0
#1 0x0000000000a0ac16 in reraise_fatal (...- 03:56 AM Bug #12566 (Can't reproduce): assert failed in OSD::ms_fast_dispatch tracepoint()
- The log is:
(gdb) bt
#0 0x0000003b8260f5db in raise () from /lib64/libpthread.so.0
#1 0x0000000000a0ac16 in ?? (...
08/02/2015
- 05:00 PM Bug #12565 (Fix Under Review): osd: do not cache unused memory in attrs
- https://github.com/ceph/ceph/pull/5451
- 10:45 AM Bug #12565 (Resolved): osd: do not cache unused memory in attrs
- attrs maybe only a range of bufferlist in recovery, If obc cache it,
this causes the whole memory in bufferlist woul... - 12:24 PM rgw Backport #12502: rgw: need conversion tool to handle fixes following #11974
- 08:25 AM Bug #12012: IO error on kvm/rbd with an erasure coded pool tier (after an upgrade from 0.87.1 to ...
- I guess all our pool setup is described into the previously send ceph-report. Lets me known is you need something els...
08/01/2015
- 06:41 PM Bug #12291: mon: wrong health warning of PGs/OSD for EC pool
- Oh ok, I think Sage already fixed it with this commit: https://github.com/ceph/ceph/commit/b70be56119f18cd86eb2a8aaaf...
- 11:05 AM Bug #12291: mon: wrong health warning of PGs/OSD for EC pool
- This is currently assigned to me but I haven't had a chance to work on it yet.
Will do as soon as I have spare cycle. - 05:56 PM rgw Backport #12502: rgw: need conversion tool to handle fixes following #11974
- ...
- 02:28 PM Revision 8652a370 (ceph): osd: avoid unnecessary calculation in agent_choose_mode()
- Signed-off-by: Li Wang <liwang@ubuntukylin.com>
Reviewed-by: Yunchuan Wen <yunchuanwen@ubuntukylin.com> - 01:58 PM Revision 6cfb4b35 (ceph): ceph-disk: add --log-stdout parameter
- This can be used when ceph-disk is run by systemd services, to ensure
that output is captured in the journal.
Signed... - 01:58 PM Revision 4fd9cf27 (ceph): ceph-disk: split get_dmcrypt_key_path from key creation
- In preparation for mapping dmcrypt cyphertext devs during activation.
Signed-off-by: David Disseldorp <ddiss@suse.de> - 01:58 PM Revision 29431944 (ceph): ceph-disk: map dmcrypt devices prior to activation
- Support mapping of dmcrypt devices during activation via the new
ceph-disk activate[-journal] --dmcrypt and --dmcrypt... - 01:58 PM Revision ac16d839 (ceph): ceph.spec: use sysvinit rbdmap script for now
- Signed-off-by: Sage Weil <sage@redhat.com>
- 01:58 PM Revision 85a89469 (ceph): systemd: activate disks via systemd service instead of udev
- The udev(7) man page states:
RUN
...
This can only be used for very short-running foreground tasks. Running
a... - 01:58 PM Revision e034c2af (ceph): ceph.spec: ship ceph-disk-* units
- Signed-off-by: Sage Weil <sage@redhat.com>
- 01:58 PM Revision 69cdfcb1 (ceph): remove ceph-disk-{activate,prepare} wrappers
- These ancient aliases shouldn't be used. We should have removed them
immediately way back when, honestly.
Signed-of... - 01:58 PM Revision e3e9548b (ceph): ceph.spec.in:fixup bad merge of systemd logic
- 01:58 PM Revision 3d6100eb (ceph): ceph.spec.in:Fixup renamed ceph-rgw.conf
- This is a simple renaming bug.
Signed-off-by: Owen Synge <osynge@suse.com> - 01:58 PM Revision 8a675612 (ceph): ceph.spec: install the new systemd-based udev rules from ddiss
- Signed-off-by: Sage Weil <sage@redhat.com>
- 01:58 PM Revision 4d10dc13 (ceph): systemd: fix ceph-radosgw@ service
- There's no prestart. Fix the instance name. Cleanup.
Signed-off-by: Sage Weil <sage@redhat.com> - 01:58 PM Revision 95db160d (ceph): ceph.spec.in:Update rpm hooks and file content
- Take rpm hooks from SUSE rpm and move upstream.
These are tested on SLE12 to include the correct files.
Signed-off-b... - 01:58 PM Revision 61240190 (ceph): Added tmpfiles.d for rgw: templated user and group.
- tmpfiles.d are part of system.d and define how temporary directories are setup.
rgw needs a socket directory. To do t... - 01:58 PM Revision ef961be5 (ceph): ceph.spec.in: Change redhat defaults for rgw
- Changes
--with-rgw-user=$USER \
--with-rgw-group=$GROUP \
to
USER=root
GROUP=root
from
USER=apache
... - 01:58 PM Revision fbc85e3c (ceph): radosgw systemd support
- Added a radosgw systemd support and associated prestart script.
- With improved checking over first revison.
- ce... - 01:58 PM Revision be4f878a (ceph): radosgw systemd prestart moved to systemd
- prestart scripts are going to be added to systemd to simplify the switching behavior.
Signed-off-by: Owen Synge <osy... - 01:58 PM Revision d6213b61 (ceph): Add unit files to the spec file.
- %{_unitdir}/ceph-mds@.service
%{_unitdir}/ceph-mon@.service
%{_unitdir}/ceph-osd@.service
%{_unitdir}/ceph.target
Si... - 01:58 PM Revision 1ba4694d (ceph): Autotools to install systemd unit files.
- To simplify the spec file we should install as much using autotools
and as little as possible in the spec file.
Sign... - 01:58 PM Revision 43d6b121 (ceph): system: include all unit files
- Signed-off-by: Sage Weil <sage@redhat.com>
- 01:58 PM Revision e283222d (ceph): ceph.spec: include systemd units files
- Signed-off-by: Sage Weil <sage@redhat.com>
- 01:58 PM Revision 6842a209 (ceph): systemd: add ceph-create-keys@ service
- Signed-off-by: Sage Weil <sage@redhat.com>
- 01:58 PM Revision d1ce1782 (ceph): ceph.spec: no rcceph[-radosgw] with systemd
- Signed-off-by: Sage Weil <sage@redhat.com>
- 08:14 AM Bug #12457: tests: run_osd sometime fails
- Has not happened in a week, back to priority normal. Happened again at https://github.com/ceph/ceph/pull/5443#issueco...
07/31/2015
- 10:57 PM rgw Bug #12432 (Pending Backport): HTTP return code is not being logged by CivetWeb
- 06:52 PM rgw Bug #12432: HTTP return code is not being logged by CivetWeb
- PR for master: https://github.com/ceph/ceph/pull/5445
- 06:10 PM rgw Bug #12432 (Fix Under Review): HTTP return code is not being logged by CivetWeb
- 09:01 PM Bug #12291: mon: wrong health warning of PGs/OSD for EC pool
- Looks like you already checked in the fix..
- 08:38 PM Bug #12291: mon: wrong health warning of PGs/OSD for EC pool
- Hi Joao,
Do you mind me taking over this one? - 08:06 PM devops Bug #12559 (Resolved): centos7 packages are not installable for v0.94.1
- The following:...
- 07:39 PM Revision ef86e292 (ceph): Merge pull request #5445 from ceph/wip-12432
- Wip 12432
- 07:18 PM Revision 78a4024c (ceph): Merge pull request #5121 from theanalyst/wip-11983-hammer
- FAILED assert(!old_value.deleted()) in upgrade:giant-x-hammer-distro-basic-multi run
Reviewed-by: Sage Weil <sage@re... - 07:17 PM Revision a451e882 (ceph): Merge pull request #5269 from dachary/wip-12362-hammer
- stuck incomplete
Reviewed-by: Sage Weil <sage@redhat.com> - 07:05 PM rgw Feature #11148 (Resolved): RGW Swift API: support for 202 Accepted response code on container cre...
- 07:04 PM rgw Backport #12299 (Resolved): RGW Swift API: support for 202 Accepted response code on container cr...
- 07:03 PM rgw Bug #11473 (Resolved): rgw: keystone does not support chunked input
- 07:03 PM rgw Backport #12322 (Resolved): rgw: keystone does not support chunked input
- 07:02 PM rgw Bug #11431 (Resolved): RGW Swift API: XML document generated in response for GET on account does ...
- 07:01 PM rgw Backport #12323 (Resolved): RGW Swift API: XML document generated in response for GET on account ...
- 06:59 PM rgw Bug #12298 (Resolved): rgw: Properly respond to the "Connection" header with Civetweb
- 06:58 PM Stable releases Tasks #11990: hammer v0.94.3
- * + "Pull request 5026":https://github.com/ceph/ceph/pull/5026
* + "Pull request 5028":https://github.com/ceph/ceph/... - 06:58 PM rgw Backport #12398 (Resolved): rgw: Properly respond to the "Connection" header with Civetweb
- 06:57 PM rgw Bug #11494 (Resolved): rgw: multipart list part response returns incorrect field
- 06:57 PM rgw Backport #12399 (Resolved): rgw: multipart list part response returns incorrect field
- 06:56 PM rgw Bug #12286 (Resolved): rgw: radosgw-admin dumps user info twice
- 06:56 PM rgw Backport #12400 (Resolved): rgw: radosgw-admin dumps user info twice
- 06:55 PM Revision 218f5374 (ceph): Merge pull request #5117 from theanalyst/wip-12099-hammer
- rgw: rados objects wronly deleted
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com> - 06:55 PM Revision 07fa83ae (ceph): Merge pull request #5118 from theanalyst/wip-12042-hammer
- DragonDisk fails to create directories via S3: MissingContentLength
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com> - 06:55 PM Revision a69d431e (ceph): Merge pull request #5214 from SUSE/wip-12299-hammer
- RGW Swift API: support for 202 Accepted response code on container creation
Reviewed-by: Yehuda Sadeh <ysadehwe@redh... - 06:54 PM Revision 5353480f (ceph): Merge pull request #5226 from SUSE/wip-12322-hammer
- rgw: keystone does not support chunked input
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com> - 06:54 PM Revision fb9156f4 (ceph): Merge pull request #5227 from SUSE/wip-12323-hammer
- RGW Swift API: XML document generated in response for GET on account does not contain account name
Reviewed-by: Yehu... - 06:54 PM Revision 7193c16b (ceph): Merge pull request #5228 from theanalyst/wip-11872-hammer
- RGW does not send Date HTTP header when civetweb frontend is used
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com> - 06:54 PM Revision ac864908 (ceph): Merge pull request #5229 from theanalyst/wip-12242-hammer
- Fix tool for #11442 does not correctly fix objects created via multipart uploads
Reviewed-by: Yehuda Sadeh <ysadehwe... - 06:53 PM Revision 33dbfc69 (ceph): Merge pull request #5237 from theanalyst/wip-12245-hammer
- rgw: empty json response when getting user quota
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com> - 06:53 PM Revision 99ca62f2 (ceph): Merge pull request #5284 from SUSE/wip-12398-hammer
- rgw: Properly respond to the Connection header with Civetweb
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com> - 06:53 PM Revision 1f5f3190 (ceph): Merge pull request #5285 from SUSE/wip-12399-hammer
- rgw: multipart list part response returns incorrect field
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com> - 06:52 PM Revision 5cbb6cfb (ceph): Merge pull request #5286 from SUSE/wip-12400-hammer
- rgw: radosgw-admin dumps user info twice
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com> - 06:25 PM rbd Bug #12383: "Segmentation fault rbd()"
- *next PR*: https://github.com/ceph/ceph/pull/5444
- 06:09 PM Revision 0451f193 (ceph): civetweb: update submodule to support setting of http status
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 06:09 PM Revision b8e28ab9 (ceph): rgw: set http status in civetweb
- Need to set the http status in civetweb so that we report it correctly.
Fixes: #12432
Signed-off-by: Yehuda Sadeh <y... - 05:56 PM rgw Feature #9604 (Pending Backport): rgw: create a tool for orphaned objects cleanup
- 05:56 PM rgw Feature #9604: rgw: create a tool for orphaned objects cleanup
- https://github.com/ceph/ceph/pull/5109
- 05:25 PM Revision c8653b11 (ceph): Merge pull request #5258 from zhouyuan/rgw_skip_first_chunk
- rgw: skip prefetch first chunk if range get falls to shadow objects
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 04:50 PM Revision 3fbcf5e2 (ceph): doc: krbd supports clones since 3.10
- Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
- 03:52 PM Feature #12316: For EC pool, read K+M shards (instead of K) to reduce latency (with IO overhead)
- PR - https://github.com/ceph/ceph/pull/5309
- 03:22 PM Bug #12550: CEPH_QA_SUITE/AARCH64: ceph_test_async_driver fail
- I don't know why AARCH64 will cause "::socket(AF_INET, SOCK_STREAM, 0)" return -1:
2015-07-09 06:46:26.966675 3ff9... - 12:29 PM Bug #12550 (Can't reproduce): CEPH_QA_SUITE/AARCH64: ceph_test_async_driver fail
- The following test fails on AArch64:
rados:singleton-nomsgr/{all/msgr.yaml}... - 02:43 PM Bug #11640 (Pending Backport): Change radosgw pools default crush ruleset
- 02:42 PM Bug #12021 (Resolved): hammer-x upgrade test.sh fails with ENOTEMPTY
- both pull request got merge, so consider it resolved.
- 02:15 PM Bug #12536: "FAILED assert(!log.null() || olog.tail == eversion_t())"
- A simpler fix (with less long-term cruft) would be to patch hammer to correctly interpret INT64_MIN as hobject_t::get...
- 02:09 PM Bug #12536: "FAILED assert(!log.null() || olog.tail == eversion_t())"
- Okay, I think it's not as bad as I thought.. just the min issue after all.
- 02:08 AM Bug #12536: "FAILED assert(!log.null() || olog.tail == eversion_t())"
- Samuel Just wrote:
> Ah, we switched to using INT64_MIN (from -1) for pool for min() with wip-temp. We need to make... - 01:19 PM Bug #9732 (In Progress): ReplicatedPG::hit_set_trim osd/ReplicatedPG.cc: 11006: FAILED assert(obc)
- sam reminded me:
* we don't want to convert an existing OSD automatically, from localtime to GMT after upgrading t... - 01:16 PM Revision 5ea618a9 (ceph): include/rados: remove op definitions for LOCK ops
- These haven't been used in many many years. They can safely be removed
and the ids reused since no deployed code is ... - 01:12 PM Revision 05f70ab6 (ceph): osdc/Objecter: remove unused/obsolete lock() operation
- Signed-off-by: Sage Weil <sage@redhat.com>
- 01:12 PM Revision 11631a50 (ceph): osdc/WritebackHandler: remove obsolete lock operation
- Signed-off-by: Sage Weil <sage@redhat.com>
- 01:10 PM Feature #11815 (Resolved): mon: allow injecting new crushmap
- 01:10 PM Revision 399d8888 (ceph): osd/osd_types: remove unused wrlock_by field
- This is ancient history, back from when the CephFS client would take
exclusive locks over objects. That code is long... - 01:05 PM CephFS Feature #12552: qa: test cephfs over cache tier in fs suite
- see #12551
- 01:04 PM CephFS Feature #12552 (Rejected): qa: test cephfs over cache tier in fs suite
- see
- 01:01 PM Bug #12551: osd: copy-from doesn't preserve truncate_{seq,size}
- https://github.com/ceph/ceph/pull/5442
- 12:47 PM Bug #12551 (Resolved): osd: copy-from doesn't preserve truncate_{seq,size}
- This breaks CephFS when used over cache tiering.
https://github.com/athanatos/ceph/commit/60222da153920286fbcb9cf5... - 12:42 PM Bug #12540: osd: avoid multi set osd_op.outdata in tier pool
- https://github.com/ceph/ceph/pull/4895
- 03:34 AM Bug #12540: osd: avoid multi set osd_op.outdata in tier pool
- PR: https://github.com/ceph/ceph/pull/4895
- 03:32 AM Bug #12540 (Resolved): osd: avoid multi set osd_op.outdata in tier pool
- There are two read op on the same object for ec pool. First op read
miss happend, calling do_proxy_read and promote_... - 12:22 PM Bug #12549 (Rejected): CEPH_QA_SUITE/AARCH64: Command failed (workunit test rados/test_alloc_hint...
- The following test fails on AArch64:
rados:objectstore/{alloc-hint.yaml}
Sample Output:... - 12:14 PM Bug #12548 (Can't reproduce): CEPH_QA_SUITE/AARCH64: ceph_test_filejournal fail
- The follow test fails on AArch64:
rados:objectstore/{filejournal.yaml}
Sample Output:... - 12:05 PM Bug #12458: rbdmap should be changed to map a rbd automatically when starup
- doc fix PR : https://github.com/ceph/ceph/pull/5440
- 06:49 AM Bug #12458 (New): rbdmap should be changed to map a rbd automatically when starup
- 06:49 AM Bug #12458: rbdmap should be changed to map a rbd automatically when starup
- 06:48 AM Bug #12458 (Resolved): rbdmap should be changed to map a rbd automatically when starup
- 11:42 AM CephFS Bug #12547 (New): CEPH_QA_SUITE/AARCH64: marginal:mds_restart fails
- The following test fails on AArch64:
marginal:mds_restart/{clusters/one_mds.yaml tasks/restart-workunit-backtraces.y... - 11:37 AM CephFS Bug #12546 (New): CEPH_QA_SUITE/AARCH64: marginal:fs-misc fail
- The following test fails on AArch64:
marginal:fs-misc/{clusters/two_clients.yaml fs/btrfs.yaml tasks/locktest.yaml}
... - 11:08 AM rbd Bug #12545 (Can't reproduce): CEPH_QA_SUITE/AARCH64: Command failed (workunit test rbd/map-snapsh...
- The following tests fail on AArch64:
krbd:rbd-nomount/{clusters/fixed-3.yaml conf.yaml fs/btrfs.yaml install/ceph.ya... - 11:05 AM Linux kernel client Bug #12544 (New): CEPH_QA_SUITE/AARCH64: Command failed (workunit test rbd/kernel.sh)
- The following test fails on AArch64:
krbd:rbd-nomount/{clusters/fixed-3.yaml conf.yaml fs/btrfs.yaml install/ceph.ya... - 10:29 AM Revision a9cb1b89 (ceph): Merge pull request #5438 from ceph/fix-doc-build
- doc: fix the build of placement-groups.rst
Reviewed-by: Kefu Chai <kchai@redhat.com> - 09:35 AM Revision 1fb9bacd (ceph): os/FileStore: getattr() should return 0 if success
- * fix the test of ObjectStore/StoreTest.SimpleCloneTest, the expected
return value of getattr() should be 0. otherw... - 09:33 AM Revision 57606555 (ceph): KeyValueStore: Fix incorrect rm_keys behaviors
- GenericObjectMap will copyup parents keys when rm keys, but if
caller(KeyValueStore) already buffer "set keys", it sh... - 09:08 AM Revision 81afcc20 (ceph): doc: fix the build of placement-groups.rst
- add a reference target of "mapping pgs to osds"
Signed-off-by: Kefu Chai <kchai@redhat.com> - 08:47 AM Revision f0084407 (ceph): Merge pull request #5432 from ceph/wip-fix-rbd-build
- make: fix the build of "rbd"
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 08:46 AM rgw Feature #12539: skip prefetch first chunk if range get falls to shadow objects
- https://github.com/ceph/ceph/pull/5258
- 12:41 AM rgw Feature #12539 (Resolved): skip prefetch first chunk if range get falls to shadow objects
- Currently the head object will be prefetched in each GET. This is unnecessary if the Range GET falls to shadow object...
- 08:46 AM rgw Bug #12537: [CVE-2015-5245] RGW returns requested bucket name raw in "Bucket" response header
- https://github.com/ceph/ceph/pull/5430
- 08:38 AM Revision 79c7d2e4 (ceph): fix: qa/workunits/rados/test_cache_pool.sh
- Signed-off-by: Mingxin Liu <mingxinliu@ubuntukylin.com>
- 07:55 AM Revision 2d90da86 (ceph): Merge pull request #5434 from tobegit3hub/remove_execute_mode
- Remove execute mode of source files
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 07:52 AM CephFS Bug #12088 (Resolved): cephfs client crash after enable readahead mechanism through setting conf ...
- 07:52 AM CephFS Backport #12319 (Resolved): cephfs client crash after enable readahead mechanism through setting ...
- 07:50 AM Bug #12512 (Pending Backport): FileStore calls syncfs(2) even it is not supported
- 07:47 AM Revision f8554f9b (ceph): Remove execute mode of source files
- 07:41 AM Revision 1a2f2905 (ceph): Merge pull request #5398 from mingyuez/master
- rbd:If the image has any protected snaps,'rbd snap purge image' will return failure without removing any snaps.
Revi... - 06:45 AM Revision b255b4c5 (ceph): make: fix the build of "rbd"
- `CEPH_GLOBAL` was removed in fa78739, but the linked libraries does not
offer all needed symbols for "rbd". so add th... - 06:09 AM Revision 79560bb7 (ceph): rename filestore_ops/bytes to journal_ops/bytes
- Signed-off-by: xinxin shu <xinxin.shu@intel.com>
- 06:09 AM Revision 692289a4 (ceph): reset max if runtime configure change
- Signed-off-by: xinxin shu <xinxin.shu@intel.com>
- 06:09 AM Revision 1cc9cc85 (ceph): use throttle framework to throttle ops/bytes for keyvaluestore
- Signed-off-by: xinxin shu <xinxin.shu@intel.com>
- 06:08 AM Revision e4fd086d (ceph): use throttle framework to throttle ops/bytes for filestore
- Signed-off-by: xinxin shu <xinxin.shu@intel.com>
- 05:40 AM Bug #12523 (Fix Under Review): osd suicide timeout during peering - search for missing objects
- 05:10 AM Bug #12523: osd suicide timeout during peering - search for missing objects
- thank you, i will try it.
- 05:04 AM Bug #12523: osd suicide timeout during peering - search for missing objects
- Update: after applying this patch, we managed to bring back the cluster.
Although this does not happen frequently,... - 05:02 AM Bug #12523: osd suicide timeout during peering - search for missing objects
- That looks like a different issue. 'suicide timeout' is ceph's way to make sure the thread workers are responsive, an...
- 01:13 AM Bug #12523: osd suicide timeout during peering - search for missing objects
- Thanks, GuangYang.
We also encounter this kind of problems, it's osd_op_tp threads timeout, which result in osd suic... - 04:06 AM Revision 8a7a52dd (ceph): rbd:'rbd purge image' will return failure without removing any snaps if...
- Signed-off-by: mingyuez <zhao.mingyue@h3c.com>
- 03:54 AM rgw Bug #12542 (Closed): radosgw deleted log leaks
- The `df` is full, but `du` says no.
I run `lsof +L1` shows... - 02:24 AM Revision d5142ba2 (ceph): Merge pull request #5395 from tchaikov/wip-12512
- common/syncfs: fall back to sync(2) if syncfs(2) not available
Reviewed-by: Haomai Wang <haomaiwang@gmail.com>
Revie... - 02:14 AM Revision f705d8c7 (ceph): Merge pull request #5402 from ceph/wip-claim-sparse-read-buffer
- osd/ReplicatedPG: claim the read buffer instead of copying it
Reviewed-by: David Zafman <dzafman@redhat.com> - 02:13 AM Revision b7e5d2af (ceph): Merge pull request #5428 from ceph/wip-kill-rest-bench
- kill rest-bench
Reviewed-by: Ken Dreyer <kdreyer@redhat.com> - 01:50 AM Revision 1df93e19 (ceph): Merge pull request #5427 from dachary/wip-12088-hammer-part-2
- Fh ref count will leak if readahead does not need to do read from osd
- 12:47 AM Support #12443: The ceph cluster will be stuck after delete a 65535 pgs pool
- @Jason Dillaman
- 12:42 AM Revision 95745550 (ceph): rgw: skip prefetch first chunk if range get falls to shadow objects
- Currently the head object will be prefetched in each GET:
a) This is unnecessary if the Range GET falls to shadow ob...
07/30/2015
- 11:35 PM Revision 03491ae8 (ceph): Merge remote-tracking branch 'origin/next'
- 11:33 PM Linux kernel client Bug #12538 (Resolved): An image with a long clone chain blows up kernel stack
- #I create a rbd block named r1
rbd -p rbd create r1 --size 1024
#then create a snapshot of r1
rbd -p rbd snap cr... - 11:26 PM Revision 27fa210f (ceph): Merge pull request #5425 from ceph/wip-12479-next
- librbd: test crash during client upgrade suite
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 10:38 PM Bug #12536: "FAILED assert(!log.null() || olog.tail == eversion_t())"
- Ah, we switched to using INT64_MIN (from -1) for pool for min() with wip-temp. We need to make sure that messages wh...
- 10:18 PM Bug #12536: "FAILED assert(!log.null() || olog.tail == eversion_t())"
- Not actually related to the above bug, just revealed by the same test. Bah, it's from wip-temp, some difference betw...
- 06:53 PM Bug #12536 (Resolved): "FAILED assert(!log.null() || olog.tail == eversion_t())"
- This is only one test to validate fix for #12410
Command line:
@teuthology-suite -v -c wip-12410 -k distro -m ... - 10:22 PM Bug #12427: OSD dies after a couple of seconds
- The crc errors in the osdmap are actually correct, the new osds encode a version other than what the mons encode sinc...
- 09:47 PM Revision ad5507fe (ceph): rgw: url encode exposed bucket
- Fixes: #12537
Don't send the bucket name back without url encoding it.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 09:32 PM Revision ca40bf0b (ceph): Merge pull request #5423 from oddomatik/patch-1
- doc: No CRUSH involvement in deciding PG target
Reviewed-by: Sage Weil <sage@redhat.com> - 09:08 PM Revision 04d87662 (ceph): Merge pull request #5426 from ceph/wip-12383-next
- rbd: remove dependency on non-ABI controlled CephContext
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 09:00 PM Revision 115f3706 (ceph): Merge pull request #5399 from tchaikov/wip-12458
- rbd: add "--keyring" option to help message
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 08:25 PM rgw Bug #12363: testGetContentType and testHead failed
- Run: http://pulpito.ceph.com/teuthology-2015-07-29_05:00:09-smoke-master-distro-basic-multi/
Job: 991514 - 08:23 PM rgw Bug #12370: test_s3.test_object_copy_canned_acl ... FAIL
- Run: http://pulpito.ceph.com/teuthology-2015-07-29_05:00:09-smoke-master-distro-basic-multi/
Job: 991513 - 08:21 PM Bug #11272 (Pending Backport): ceph-disk zap should ensure block device
- We're cherry-picking this downstream in RHCS 1.3, so it would be great to get it fixed upstream in Hammer too!
- 08:15 PM Revision 2ecb3b7f (ceph): Fh ref count will leak if readahead does not need to do read from osd
- The 3c8cdeacf46ae4031189d2ef6948aa3b6ab4ae43 backport introduced a leak.
http://tracker.ceph.com/issues/12319 Fixes:... - 07:43 PM Revision a48cbc0a (ceph): Merge pull request #5120 from theanalyst/wip-11999-hammer
- cephfs Dumper tries to load whole journal into memory at once
Reviewed-by: Greg Farnum <gfarnum@redhat.com> - 07:43 PM Revision 06c27cdd (ceph): Merge pull request #5119 from theanalyst/wip-12098-hammer
- kernel_untar_build fails on EL7
Reviewed-by: Greg Farnum <gfarnum@redhat.com> - 07:10 PM rgw Bug #12537 (Resolved): [CVE-2015-5245] RGW returns requested bucket name raw in "Bucket" response...
- PCI scans have determined that the RGW is returning whatever string it thought was the name of the bucket requested a...
- 06:53 PM rbd Bug #12535 (Can't reproduce): CEPH_QA_SUITE/AARCH64: segfault in rbd-fuse
- The following test fails on AArch64:
rbd:singleton/{all/merge_diff.yaml}... - 05:58 PM Bug #12194: osd crash FAILED assert(!parent->get_log().get_missing().is_missing(soid))
- Continue to encounter this one periodically, now has happened on a 2nd OSD. btrfs, centos 7.1.
- 05:57 PM rgw Bug #12534 (Closed): rgw: region is not removed from region map
- Bug reported by duanweijun@h3c.com on github:
this is my region:
$rados -p .rgw.root ls... - 05:52 PM Revision d621d987 (ceph): Merge pull request #5414 from cxwshawn/master
- doc: add bucket object version description.
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 05:49 PM Bug #12466: Init script bug with two clusters with the same osd ID on the same host
- Looks like I managed to abandon usage of ceph init-script. I wrote my own systemd-generator script which is very hand...
- 05:37 PM rgw Feature #12533 (New): rgw: upgrade to civetweb 1.7
- 05:31 PM Bug #12532 (Can't reproduce): CEPH_QA_SUITE/AARCH64: Command failed (workunit test snaps/snaptest...
- The following test fails on AArch64:
stress:bench/{clusters/fixed-3-cephfs.yaml tasks/cfuse_workunit_snaps.yaml}
... - 05:24 PM Revision a634ab34 (ceph): remove libs3 submodule
- Signed-off-by: Sage Weil <sage@redhat.com>
- 05:21 PM Revision bbe84578 (ceph): remove rest-bench
- This is a weak tool; users should look to cosbench or others instead.
Signed-off-by: Sage Weil <sage@redhat.com> - 04:45 PM Bug #12531: CEPH_QA_SUITE/AARCH64:knfs suite fails
- Correction:
Tests running v3.yaml have "mount.nfs: requested NFS version or transport protocol is not supported".
... - 04:33 PM Bug #12531 (Can't reproduce): CEPH_QA_SUITE/AARCH64:knfs suite fails
- All tests in the knfs suite fail with the following message:
"mount.nfs: requested NFS version or transport protocol... - 04:42 PM rbd Bug #12383 (Fix Under Review): "Segmentation fault rbd()"
- *next PR*: https://github.com/ceph/ceph/pull/5426
- 04:39 PM rbd Bug #12479 (Fix Under Review): [ FAILED ] TestLibRBD.BlockingAIO
- *next PR*: https://github.com/ceph/ceph/pull/5425
- 04:25 PM Revision 744a71a8 (ceph): Merge pull request #5419 from ceph/wip-kfsx-invalidate
- test_librbd_fsx: invalidate before discard in krbd mode
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 03:58 PM Bug #12523: osd suicide timeout during peering - search for missing objects
- Hi Huang Jun,
I have the issue reproduced with debug_osd = 10, what I observed was, the peering thread stuck at PG::... - 02:43 PM Bug #12523: osd suicide timeout during peering - search for missing objects
- hi,guangyang
how did you see what the thread is doing during timeout? by turn on debug_osd level?
do you used ltt... - 03:22 PM Revision a8d33c9f (ceph): doc: fix command line for swift user creation
- 03:04 PM devops Backport #11998 (Resolved): /usr/bin/ceph from ceph-common is broken without installing ceph
- 02:48 PM devops Backport #11998 (Fix Under Review): /usr/bin/ceph from ceph-common is broken without installing ceph
- 09:55 AM devops Backport #11998: /usr/bin/ceph from ceph-common is broken without installing ceph
- ...
- 07:51 AM devops Backport #11998: /usr/bin/ceph from ceph-common is broken without installing ceph
- https://github.com/ceph/ceph/pull/5417
- 07:34 AM devops Backport #11998 (In Progress): /usr/bin/ceph from ceph-common is broken without installing ceph
- http://www.spinics.net/lists/ceph-devel/msg25259.html
http://ceph.aevoo.fr:8081/ubuntu-2015-07-29_14:35:14-upgrade... - 03:00 PM Revision 19abe5ee (ceph): Merge pull request #5417 from dachary/wip-11998-hammer
- debian/control: ceph-common (>> 0.94.2) must be >= 0.94.2-2
Reviewed-by: Sage Weil <sage@redhat.com> - 02:59 PM Bug #12429: OSD crash creating/deleting pools
- I'm fairly confused about how our tests could have created this scenario. Once we request_proposal() any subsequent a...
- 12:56 PM Bug #12429: OSD crash creating/deleting pools
- I think the problem may be because of the update to osdmap that we do during "fs new", setting crash_replay interval....
- 02:57 PM Backport #12530 (Resolved): OSDMonitor::preprocess_get_osdmap: must send the last map as well
- 02:56 PM Bug #12410: OSDMonitor::preprocess_get_osdmap: must send the last map as well
- 09:02 AM Bug #12410: OSDMonitor::preprocess_get_osdmap: must send the last map as well
- the fix of @e <= MIN@ looks good to me, coz the osd was off the last epoch by only *1*. and we don't reply the OSD wi...
- 02:33 PM Revision 77cdb500 (ceph): No CRUSH involvement in deciding PG target
- Quoting from elsewhere "[we calculate] a placement group using the object name, a hash code, the number of PGs in the...
- 02:19 PM Bug #11586 (Duplicate): RBD Data Corruption with Logging on Giant
- root cause is #12465
- 01:23 PM Revision 8d4932e7 (ceph): osd/OSDMap: handle incrementals that modify+del pool
- Because new_pools was processed after old_pools, if something
was modified and then deleted in the same incremental
m... - 01:01 PM Revision af0cade0 (ceph): lockdep: allow lockdep to be dynamically enabled/disabled
- librbd test cases attempt to enable lockdep coverage via the librados
API. Use a configuration observer to register/... - 01:01 PM Revision 2c51aadb (ceph): tests: librbd API test cannot use private md_config_t struct
- Remove all depencencies on md_config_t and instead use librados API
methods to get/set configuration values.
Fixes: ... - 01:01 PM Revision 4d03c664 (ceph): librados_test_stub: implement conf get/set API methods
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 12:32 PM rgw Feature #12102: Request for increased logging in RGW, for CivetWeb.
- Setting 'debug civetweb = 10' doesn't work as intended. There are no logs with the marker "civetweb:" in the rgw logs...
- 11:26 AM Revision 4c199bf5 (ceph): debian/control: ceph-common (>> 0.94.2) must be >= 0.94.2-2
- The d8733be2ef8874b9a858a7ffddfb81b9b656e9a6 backport introduced a
regression by adding an incorrect Depends / Break ... - 09:24 AM RADOS Bug #12405: filestore: syncfs causes high cpu load due to kernel implementation in high-memory boxes
An update to this issue:
a) This is for the Congress (DreamObjects) cluster.
b) There weren't any changes in ...- 08:49 AM CephFS Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
- will do
- 07:45 AM CephFS Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
- Hi Yan,
Those patches look good after performing test for a long time. Will you commit the patches to both ceph an... - 07:59 AM Bug #12328: Ceph KV OSD crashes sometimes with FAILED assert(0 == "hit suicide timeout") on 0.94.1
- huang jun wrote:
> hi, Kenneth
> We also encounter this suicide error very often when we test the 4K random writes.... - 07:58 AM Bug #12328: Ceph KV OSD crashes sometimes with FAILED assert(0 == "hit suicide timeout") on 0.94.1
- hi, Kenneth
We also encounter this suicide error very often when we test the 4K random writes.
the osd_op_tp thre... - 07:27 AM devops Fix #12529 (Rejected): moving files between packages with deb
- When moving a file from one package to the other, from ceph to ceph-common for instance, the following must be done i...
- 06:12 AM Revision cb51b170 (ceph): mon: reject over-large values of max_mds
- Fixes: #12222
Signed-off-by: John Spray <john.spray@redhat.com> - 06:03 AM Bug #12528: legacy pg log is around after OSD starts
- /a/ubuntu-2015-07-28_20:43:19-rados-wip-kefu-t3sting---basic-multi/990637/remote/plana93/log/ceph-osd.0.log.gz
<pr... - 06:02 AM Bug #12528 (Can't reproduce): legacy pg log is around after OSD starts
- see /a/ubuntu-2015-07-28_21:20:52-rados-master---basic-multi/990944/remote/plana10/log/ceph-osd.5.log.gz...
- 05:41 AM Linux kernel client Bug #12527 (Duplicate): rbd map leads to system crash!
- #I create a rbd block named r1
rbd -p rbd create r1 --size 1024
#then create a snapshot of r1
rbd -p rbd snap cr... - 03:56 AM Revision 258cb348 (ceph): rbd: add "--keyring" option to help message
- Refs: #12458
Signed-off-by: Kefu Chai <kchai@redhat.com> - 03:37 AM Bug #12450 (Resolved): rados load-gen generates no operations if --max-ops is not explicitly set
- 03:36 AM Revision 7fbf013d (ceph): Merge pull request #5401 from ceph/wip-12450
- tools/rados: change the default max_ops to 16
Reviewed-by: David Zafman <dzafman@redhat.com> - 03:18 AM Bug #12517: assert failure in Objecter::_finish_op(Objecter::Op*)
- Sorry,my ceph version is 0.87,but there is not 0.87 in this bug system to choose.Why the bug system does not have 0.8...
- 03:15 AM Feature #12514: PGMonitor::get_health get ops' blocked the detail information, show too much
- 1."ceph -s" show only according to the statistics of "time"...
- 02:09 AM Revision 5c395ff8 (ceph): doc: add bucket object version description.
- bucket object version has been supported, but do not have description in the docs,
so add this part.
Signed-off-by: ...
Also available in: Atom