Project

General

Profile

Activity

From 05/12/2014 to 06/10/2014

06/10/2014

11:43 PM Revision ecbb0056 (ceph): OSD: The thread pool variable name changed
The variable names are more meaningful now. op_tp -> osd_tp and
op_sharded_tp -> osd_op_tp
Signed-off-by: Somnath Ro...
Somnath Roy
11:19 PM rgw Feature #7774 (Fix Under Review): rgw: cache decoded user and bucket info
Yehuda Sadeh
09:06 AM rgw Feature #7774 (In Progress): rgw: cache decoded user and bucket info
Yehuda Sadeh
11:02 PM Revision 2a5d83d2 (ceph): ShardedTP: The config option changed
The config option for sharded threadpool is changed to
osd_op_num_threads_per_shard instead of osd_op_num_sharded_poo...
Somnath Roy
10:24 PM Revision a0e48b8b (ceph): ShardedTP: Changes related to conforming to Ceph coding guidelines
Signed-off-by: Somnath Roy <somnath.roy@sandisk.com> Somnath Roy
09:16 PM Bug #8578 (Can't reproduce): teuthology: OSD thrasher killing OSDs which hold the sole copies of PGs
ubuntu@teuthology:/a/gregf-2014-06-09_22:28:53-rados-wip-xattr-spillout-testing-basic-plana/303486
For example:
<...
Greg Farnum
07:40 PM Revision f69f2dff (ceph): Automatically watch 'multi' tube if tube_name
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
07:20 PM Revision 2e3f4bc6 (ceph): doc: Fixed typo.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
06:56 PM CephFS Bug #8542: kcephfs: fsx failure on read (expected 0's)
/teuthology-2014-05-30_23:02:02-kcephfs-master-testing-basic-plana/283020/ Greg Farnum
06:46 PM CephFS Bug #8542: kcephfs: fsx failure on read (expected 0's)
/a/teuthology-2014-06-01_23:02:07-kcephfs-next-testing-basic-plana/285055 Greg Farnum
06:10 PM CephFS Bug #8542: kcephfs: fsx failure on read (expected 0's)
teuthology-2014-06-06_23:01:56-kcephfs-master-testing-basic-plana/297431/ Greg Farnum
06:42 PM Revision 67339470 (ceph): Fix for bug #6700
When preparing OSD disks with colocated journals, the intialization process
fails when using dmcrypt. The kernel fail...
Steve Taylor
06:22 PM CephFS Bug #8576 (Resolved): teuthology: nfs tests failing on umount
Just a sample:
http://qa-proxy.ceph.com/teuthology/teuthology-2014-06-02_23:02:20-knfs-master-testing-basic-plana/28...
Greg Farnum
06:19 PM CephFS Bug #8575 (Rejected): linux kernel: possible circular locking dependency detected
http://qa-proxy.ceph.com/teuthology/teuthology-2014-06-02_23:02:20-knfs-master-testing-basic-plana/289526/
Not sur...
Greg Farnum
06:00 PM CephFS Bug #8574 (Resolved): teuthology: NFS mounts on trusty are failing
It looks like there are some changes to the settings which make Trusty's default NFS server settings incompatible wit... Greg Farnum
03:24 PM Feature #8343: please enable data integrity checking (by default) / silent data corruption
Sorry for dropping this...it sounds like you used incorrect replicas as the authoritative sources for the repairs you... Greg Farnum
01:55 PM devops Bug #7627: ceph-disk: does not start daemons properly under systemd
I am having a lot of trouble trying to relate the ssh connection with how we start the daemon and the subsequent fail... Alfredo Deza
12:18 PM Revision 15ae765e (ceph): Merge pull request #270 from ceph/wip-schedule-suite
Refactor teuthology.schedule... Alfredo Deza
09:51 AM devops Bug #8572 (Resolved): ceph-release missing from new FC20 repo
... Alfredo Deza
09:22 AM rbd Feature #8569: RBD Mirroring::
Neil Levine
09:22 AM rbd Feature #8569 (Duplicate): RBD Mirroring::
http://wiki.ceph.com/Planning/Blueprints/Giant/rbd%3A_journaling Neil Levine
09:22 AM rbd Bug #8416: Client Crash when try to map a volume (ubuntu)
FWIW, I'm pretty sure this is a known bug in kernels ~3.5, which manifests as a crash on 'rbd map' if the kernel miss... Ilya Dryomov
09:21 AM rbd Feature #7923: Openstack: backup from in-use volume instead of from detached volume
Josh to write up more detailed blueprint for implementation Neil Levine
09:06 AM Bug #8241 (Pending Backport): XfsFileStoreBackend tries to set extsize but may get EINVAL
Ilya Dryomov
08:31 AM Revision 6c4c1b71 (ceph): Merge pull request #1945 from dachary/wip-warnings
tests: fix signess warning
Reviewed-By: Christophe Courtaut <christophe.courtaut@gmail.com>
Loïc Dachary
06:30 AM Feature #7288: Deep-scrub throttle
Would it be possible to add an nscrubs limit which works across the cluster?, for example at the pool level: osd_pool... Dan van der Ster
02:39 AM Feature #8475: erasure-code: oversized objects when using the Cauchy technique
"rados suite":http://pulpito.ceph.com/loic-2014-06-10_11:39:56-rados-wip-8475-testing-basic-plana/ Loïc Dachary
01:34 AM Feature #8475: erasure-code: oversized objects when using the Cauchy technique
Loïc Dachary
01:37 AM Linux kernel client Bug #8568 (Closed): libceph: kernel BUG at net/ceph/osd_client.c:885
Kernel panic after kill all OSD deamons at the same time.
All kernels.
Jun 10 10:12:34 client01 kernel: [ 1113.39...
Gut Wielki

06/09/2014

11:51 PM rbd Bug #8564: osd cannot be restarted when leveldb is used as backend
Are you mean PR 1941 for bug fix and 1649 for performance purpose?
If so, yes
Haomai Wang
10:48 PM rbd Bug #8564: osd cannot be restarted when leveldb is used as backend
hi haomai , is https://github.com/ceph/ceph/pull/1941 PR the fix for this bug , while https://github.com/ceph/ceph/pu... Xinxin Shu
04:30 AM rbd Bug #8564 (Fix Under Review): osd cannot be restarted when leveldb is used as backend
Haomai Wang
04:29 AM rbd Bug #8564: osd cannot be restarted when leveldb is used as backend
Hi xinxin,
Thanks for your report, you hint a known bug which will solved in (https://github.com/ceph/ceph/pull/16...
Haomai Wang
02:47 AM rbd Bug #8564 (Resolved): osd cannot be restarted when leveldb is used as backend
hi all , recently i enabled leveldb as filestore backend, after restarting my cluster , an osd is crashed, from the l... Xinxin Shu
11:31 PM Revision 78610b03 (ceph): Remote 'delete' and 'show' features
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
11:31 PM Revision 78dbc369 (ceph): Refactor teuthology.schedule...
... to separate the assembling of the job config from the scheduling of
the job. Also port its argument parsing to do...
Zack Cerza
11:30 PM Revision b2f0ddad (ceph): Remove unused import
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
11:06 PM Revision a4923f5b (ceph): fix compilation warnings
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
10:22 PM CephFS Feature #8563: mds: permit storing all metadata in metadata pool
Yeah. I'm actually not sure why it would be sending out rados xattr requests on restart unless you'd lost clients, bu... Greg Farnum
05:53 PM CephFS Feature #8563: mds: permit storing all metadata in metadata pool
I observe the slow requests with ceph -w, or watching osd log files.
(i) The mds getxattr parent requests during r...
Alexandre Oliva
03:55 PM CephFS Feature #8563: mds: permit storing all metadata in metadata pool
Can you talk about cases (i) and (ii) in a little more detail? And how you're observing the delayed xattr requests? Greg Farnum
03:40 PM CephFS Feature #8563: mds: permit storing all metadata in metadata pool
I like the notion of being able to recover at least part of the tree structure from data pools alone. Maybe the opti... Alexandre Oliva
10:19 AM CephFS Feature #8563: mds: permit storing all metadata in metadata pool
My concern with this is that part of the point is to provide a way to recover data into the hierarchy even if we've l... Greg Farnum
09:41 PM Revision a405c228 (ceph): Merge pull request #268 from ceph/wip-5400
Support installing ceph kernels from gitbuilder on rpm machines. Zack Cerza
08:43 PM Revision 3d1df3aa (ceph): Better variable names.
As suggested by review.
Signed-off-by: Sandon Van Ness <sandon@inktank.com>
Sandon Van Ness
05:41 PM Revision 558249c5 (ceph): rgw: cut short object read if a chunk returns error
Fixes: #8289
Backport: firefly, dumpling
When reading an object, if we hit an error when trying to read one of
the ra...
Yehuda Sadeh
04:46 PM Revision dbf984e9 (ceph): Disable "fix" for corrupt repo detection
It's causing more problems than it's solving at this point
Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
Zack Cerza
03:37 PM Bug #7804: backfill racing with a hitset object remove
ubuntu@teuthology:/a/gregf-2014-06-02_14:44:16-rados-wip-sharded-threadpool-testing-basic-plana/287095 Greg Farnum
02:43 PM Documentation #8536 (Resolved): Ceph-deploy install on Centos and RHEL
There were comments on the master branch, but not on Firefly or Dumpling. I've added the comments there too. John Wilkins
02:22 PM Revision 2e5c697f (ceph): Use correct cwd
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
01:31 PM Bug #7891: osd: leaked pg refs on shutdown
http://qa-proxy.ceph.com/teuthology/gregf-2014-06-02_14:44:16-rados-wip-sharded-threadpool-testing-basic-plana/287448/ Greg Farnum
12:29 PM Revision 47028874 (ceph): Merge pull request #1872 from ceph/wip-reweight-tree
mon: add 'osd crush reweight-tree ...'
Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com>
Sage Weil
11:30 AM Revision 63cc7f97 (ceph): Add test for objectstore
Add remove then write object in a transaction.
Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
Haomai Wang
11:20 AM Revision 50c8fee8 (ceph): Fix write operation on a deleted object in the same transaction
If the following op happened:
touch obj
delete obj
write obj
KeyValueStore will fail at "write" operation.
Signed-o...
Haomai Wang
11:20 AM Revision 737c13e1 (ceph): Remove SequencerPosition from KeyValueStore
Now KeyValueStore expects kv backend to ensure consistency and there is unusable
for KeyValueStore to store a Sequenc...
Haomai Wang
11:16 AM Revision 2fe1a057 (ceph): Merge pull request #1940 from vshankar/doc-fix
doc: fix typo in erasure coding section
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
09:34 AM Revision ef065156 (ceph): doc: fix typo in erasure coding section
Signed-off-by: Venky Shankar <vshankar@redhat.com> Venky Shankar
09:10 AM Revision 25200521 (ceph): Merge pull request #1936 from ceph/wip-8554
init-ceph: continue after failure doing osd data mount
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
09:08 AM devops Bug #7598: ceph-disk-activate error with ceph-deploy
One of the differences here is that you cannot call 'activate' without doing a 'prepare' first.
This behavior woul...
Alfredo Deza
05:46 AM Revision 1080e7a3 (ceph): Add upper limit to the write size of set_alloc_hint in KeyValueStore
Signed-off-by: Haomai Wang <haomaiwang@gmail.com> Haomai Wang
05:15 AM Bug #8554 (Pending Backport): sysvinit script stops if it hits a mount error
Sage Weil
04:12 AM Revision a948ce2f (ceph): Merge pull request #1931 from ceph/wip-7068-dumpling
Wip 7068 dumpling
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
03:18 AM Revision 6a7e2014 (ceph): init-ceph: continue after failure doing osd data mount
If we are starting many daemons and hit an error, we normally note it and
move on. Do the same when doing the pre-mo...
Sage Weil
02:24 AM Feature #8071: erasure-code: simplify jerasure initialization
"jerasure merged in master":https://bitbucket.org/jimplank/jerasure/commits/1b30a37c9f75df371cf4deaedfde6b843933b4f0 ... Loïc Dachary

06/08/2014

10:36 PM CephFS Feature #8563: mds: permit storing all metadata in metadata pool
I like this idea too, but currently have no time to implement it. Zheng Yan
09:57 PM CephFS Feature #8563 (New): mds: permit storing all metadata in metadata pool
(this had originally been filed as issue #8230, which was hijacked into an unrelated issue)
I'm speaking specifica...
Alexandre Oliva
10:02 PM Revision 24853e0d (ceph): Merge pull request #1913 from sahid/dist
Populate AUTHORS and ChangeLog for a distributed package
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
09:13 PM rgw Feature #8562 (Resolved): rgw: Conditional PUT on ETag
The object versioning has not been implemented yet in RGW. But there are some use cases that different sources concur... Xiangyu Lv
09:11 PM Bug #7068 (Resolved): os/FileStore.cc: 4035: FAILED assert(omap_attrs.size() == omap_aset.size())...
Sage Weil
03:56 PM Revision 1f99cdae (ceph): mon: gather perf stats on elections
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
03:56 PM Revision 8f36d96f (ceph): mon: gather perf stats on session adds, removes, counts
I found myself wanting to know this on a running cluster.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
10:24 AM Feature #8071 (Fix Under Review): erasure-code: simplify jerasure initialization
"need review":https://github.com/ceph/ceph/pull/1938 Loïc Dachary
10:03 AM Feature #8071: erasure-code: simplify jerasure initialization
"pending pull request on jerasure":https://bitbucket.org/jimplank/jerasure/pull-request/22/add-galois_init_default_fi... Loïc Dachary
09:07 AM Feature #8071 (In Progress): erasure-code: simplify jerasure initialization
Loïc Dachary
07:11 AM Feature #7238 (Fix Under Review): erasure code : implement LRC plugin
Loïc Dachary

06/07/2014

06:03 PM Revision ecda2fef (ceph): OSD: move waiting_for_pg into Session
This way, we can handle pg splits atomically with ms_fast_dispatch.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Samuel Just
04:22 PM Revision db19040e (ceph): Merge pull request #1934 from sahid/patch-1
Fix two typos with "dependencies" Sage Weil
02:48 PM Revision e217488f (ceph): Merge pull request #269 from ceph/wip-radosfix
rados: Add hit_set_list missing from possible op weights Zack Cerza
02:45 PM Revision b7c1e351 (ceph): Update to reflect relocated get_jobs()
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
10:57 AM Revision 1f40c356 (ceph): Add set_alloc_hint test to ceph_test_objectstore
Signed-off-by: Haomai Wang <haomaiwang@gmail.com> Haomai Wang
10:57 AM Revision 5dd9b2ae (ceph): Make KeyValueStore support set_alloc_hint op
Add a new config let KeyValueStore support configurable strip size.
set_alloc_hint op can affect the strip size of th...
Haomai Wang
06:25 AM Revision b0c66a71 (ceph): doc: Fixes spelling errors on README
Signed-off-by: Sahid Orentino Ferdjaoui <sahid.ferdjaoui@cloudwatt.com> Sahid Orentino Ferdjaoui
06:16 AM Revision c5841fac (ceph): Merge pull request #1899 from yuyuyu101/kvstore-osd-perf
Make KeyValueStore support "ceph osd perf" command
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
06:12 AM Revision dacc0acf (ceph): Merge pull request #1912 from yuyuyu101/keyvaluestore-sparse-read-bug
Fix keyvaluestore fiemap bug
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
04:56 AM rbd Bug #8529 (Resolved): vm get killed when leveldb backend enable
Haomai Wang
01:13 AM Revision f5d3f3ea (ceph): Merge branch 'wip-libradosstriper'
https://github.com/ceph/ceph/pull/1186
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
01:12 AM Revision d160ce23 (ceph): Implementation of the radosstriper interface.
The user facing API is implemented in libradosstriper.cc and the backend in RadosStriperImpl.cc.
Details on how the c...
Sebastien Ponce
01:12 AM Revision fa01ca65 (ceph): Added unit test suite for the Rados striping API.
This includes tests for standard io and asynchronous io, similar to what is tested in the rados tests.
In addition, i...
Sebastien Ponce
01:12 AM Revision 910d73c4 (ceph): Added RPM and debian packaging for libradosstriper, creating a new pack...
Signed-off-by: Sebastien Ponce <sebastien.ponce@cern.ch> Sebastien Ponce

06/06/2014

11:05 PM Bug #8554 (Fix Under Review): sysvinit script stops if it hits a mount error
tested fix on the original node and it works. Sage Weil
06:53 AM Bug #8554 (Resolved): sysvinit script stops if it hits a mount error
... Sage Weil
10:30 PM Revision 875e2d24 (ceph): rados: Add hit_set_list missing from possible op weights
For the future help keep in sync with test/osd/TestRados.cc by listing
in order of op_types in that code.
Signed-off...
David Zafman
10:05 PM Feature #8560 (Resolved): mon: instrument paxos
Add perf counters for paxos internals, like
- restarts
- commit
- commit txn size in bytes (average)
- commit t...
Sage Weil
09:33 PM Revision e057f38a (ceph): Merge pull request #1926 from nereocystis/librados-write-full
doc: Use write_full for C++ example John Wilkins
09:16 PM Revision 91045942 (ceph): Merge pull request #1935 from ceph/wip-flag-known-redirs
osdc/Objecter: mark all ops as known-if-redirected
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
09:15 PM Revision 2a7de824 (ceph): Move teuthology-ls's implementation to ls.py
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
08:45 PM Revision a6c34e47 (ceph): osdc/Objecter: mark all ops as known-if-redirected
Signed-off-by: Greg Farnum <greg@inktank.com> Greg Farnum
08:20 PM Revision 036e8072 (ceph): Merge pull request #1891 from ceph/wip-8380-b
osdc/Objecter: flag ops that have been redirected Sage Weil
05:50 PM Revision 9db846b7 (ceph): Merge pull request #1930 from ceph/wip-set-extsize-fix
Wip set extsize fix
Backport: firefly
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
05:01 PM Revision 87e7ff0e (ceph): Fix corrupt repo detection
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
04:19 PM Revision 3b44f08f (ceph): Merge pull request #1922 from ceph/wip-os-rename
os: rename get_*() -> decode_*()
Reviewed-by: Sahid Orentino Ferdjaoui <sahid.ferdjaoui@cloudwatt.com>
Sage Weil
03:55 PM Revision 273c502b (ceph): tasks: Use '_' instead of '-' in names
Python modules are not allowed to contain hyphens
in the name. Using hyphens only works here because
we're using low...
John Spray
03:45 PM Revision 95a464a4 (ceph): Merge remote-tracking branch 'origin/wip-8269-dumpling' into dumpling
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> Yehuda Sadeh
01:30 PM Bug #8556 (Fix Under Review): CEPH_FEATURE_OSD_ERASURE_CODES feature bit
Sage Weil
10:52 AM Bug #8556: CEPH_FEATURE_OSD_ERASURE_CODES feature bit
Hmm, so splitting OSDMap::get_features() into a get_features_client() and get_features_server()? Greg Farnum
10:44 AM Bug #8556: CEPH_FEATURE_OSD_ERASURE_CODES feature bit
We talked about this in core standup and the consensus is currently to just drop the EC requirement for clients entir... Sage Weil
10:40 AM Bug #8556: CEPH_FEATURE_OSD_ERASURE_CODES feature bit
Not really. The course of action I had in mind was to make an exception for the EC bit in the kernel (i.e. ignore th... Ilya Dryomov
10:18 AM Bug #8556: CEPH_FEATURE_OSD_ERASURE_CODES feature bit
RBD is the *only* storage system which can really identify a given client instance as communicating with only a singl... Greg Farnum
10:13 AM Bug #8556 (Resolved): CEPH_FEATURE_OSD_ERASURE_CODES feature bit
Currently OSDs require CEPH_FEATURE_OSD_ERASURE_CODES feature for the clients to be able to connect to the OSDs. Thi... Ilya Dryomov
11:51 AM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
I ran a grep through all the osd logfiles searching for failed asserts and attached the output. The oldest logfiles a... Markus Blank-Burian
11:38 AM Revision bc3b30ed (ceph): XfsFileStoreBackend: call ioctl(XFS_IOC_FSSETXATTR) less often
No need to call ioctl(XFS_IOC_FSSETXATTR) if extsize is already set to
the value we want or if any extents are alloca...
Ilya Dryomov
11:32 AM Revision 750b1db6 (ceph): XfsFileStoreBackend: nuke redundant goto in set_extsize()
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com> Ilya Dryomov
09:15 AM Bug #8555 (Can't reproduce): PUT osd/pool/rename?srcpool=data2&destpool=data3: 400 (dumpling)
2014-06-05T23:00:20.443 INFO:teuthology.task.workunit.client.0.out:[10.214.132.18]: PUT osd/pool/rename?srcpool=data2... Sage Weil
08:08 AM Revision 479a67e0 (ceph): creating smoke-tests suite to include basic tests from all modules
Signed-off-by: tamil <tamil.muthamizhan@inktank.com> Tamilarasi muthamizhan
08:06 AM rbd Bug #6385 (Resolved): run_xfstests.sh: doesn't work with newly-organized upstream tests
Ilya Dryomov
08:05 AM Bug #8241 (Fix Under Review): XfsFileStoreBackend tries to set extsize but may get EINVAL
With the fix applied we shouldn't see this message anymore, so I've
left the priority intact. The other known cause...
Ilya Dryomov
07:52 AM Revision e25e33e2 (ceph): Merge pull request #1928 from nereocystis/pool-pg-fix-spelling
doc: Descrption => Description
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
04:23 AM devops Bug #7598: ceph-disk-activate error with ceph-deploy
My issue (as stated above in comment #4) is resolved.
I discovered that i was using a broken ceph.rpm coming from th...
Thanassis Parathyras
01:53 AM Revision 524a155e (ceph): rgw: reduce calls to rgw_obj.set_obj()
default manifest iterator construction did unneeded init of rgw_obj.
Also add a smarter copy constructor to RGWObjSta...
Yehuda Sadeh
01:24 AM Revision 3dae48c7 (ceph): rados: Fix cache_flush, cache_try_flush, cache_evict and add hit_set_list
Signed-off-by: David Zafman <david.zafman@inktank.com> David Zafman
12:10 AM Revision e31d3fe0 (ceph): doc: Descrption => Description
Correct spelling error.
Signed-off-by: Kevin Dalley <kevin@kelphead.org>
Kevin Dalley

06/05/2014

11:48 PM Revision 0ca43d83 (ceph): doc: Use write_full for C++ example
Latest version of librados uses write_full when writing entire object.
Should the documentation mention that the met...
Kevin Dalley
09:39 PM rgw Bug #8553 (Resolved): rgw: log show broken
... Yehuda Sadeh
08:03 PM Revision 0bd6f679 (ceph): OSD::calc_priors_during: handle CRUSH_ITEM_NONE correctly
Fixes: #8507
Signed-off-by: Samuel Just <sam.just@inktank.com>
Samuel Just
07:11 PM Bug #8551 (Resolved): init-ceph: small drives can end up with defaultweight = 0.00
init-ceph calculates the default weight of an OSD with... Dan Mick
07:08 PM Revision 624ae218 (ceph): Merge pull request #1859 from ceph/wip-multipartition
ceph-disk: Enable creating multiple osds per dev
Reviewed-by: Dan Mick <dan.mick@inktank.com>
Sage Weil
07:07 PM Revision 825db505 (ceph): Added requested comment.
Signed-off-by: Sandon Van Ness <sandon@inktank.com> Sandon Van Ness
07:04 PM Revision 242d9de4 (ceph): Remove unused variables and functions.
Signed-off-by: Sandon Van Ness <sandon@inktank.com> Sandon Van Ness
07:03 PM Revision a63ee379 (ceph): Merge remote-tracking branch 'gh/next'
Sage Weil
07:02 PM Revision 16e2ed0d (ceph): Merge pull request #1925 from ceph/wip-i386-atomic
fix i386 atomic64_t
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
06:59 PM Revision 2081c992 (ceph): include/atomic: make 32-bit atomic64_t unsigned
This fixes
In file included from test/perf_counters.cc:19:0:
./common/perf_counters.h: In member function ‘std::pair...
Sage Weil
06:58 PM Revision 64e99d81 (ceph): ceph-objectstore-test: fix warning in collect_metadata test
In file included from test/objectstore/store_test.cc:33:0:
../src/gtest/include/gtest/gtest.h: In function ‘testing::...
Sage Weil
06:48 PM rgw Bug #8550 (Resolved): rgw: need to reduce calls to rgw_obj.set_obj()
This is a placeholder bug for pull request #1863. Yehuda Sadeh
06:41 PM Revision a79a787b (ceph): doc: Added requiretty comment to preflight checklist.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
06:34 PM Revision 5c5eb21f (ceph): doc: Added Disable requiretty to quick start.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
06:29 PM Revision 34d9b2bc (ceph): doc: Added Disable requiretty commentary.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
05:44 PM Documentation #8549 (Closed): Ceph installation pre-flight problem with ssh-copy-id
http://ceph.com/docs/master/start/quick-start-preflight/#ceph-node-setup
The use of ssh-copy-id in the above docum...
Anonymous
05:07 PM Revision f92d7bb6 (ceph): Fixed missing include of assert.h in RWLock interface
Signed-off-by: Sebastien Ponce <sebastien.ponce@cern.ch> Sebastien Ponce
05:07 PM Revision 8c124910 (ceph): Fixed usage of rhel_version in spec file : it should have been rhel
Signed-off-by: Sebastien Ponce <sebastien.ponce@cern.ch> Sebastien Ponce
05:07 PM Revision 62064f38 (ceph): Added const qualifiers for the layout argument of file_to_extents and o...
Signed-off-by: Sebastien Ponce <sebastien.ponce@cern.ch> Sebastien Ponce
05:07 PM Revision 7ceeb9f8 (ceph): Completed librados documentation of rados_write_op_cmpxattr and rados_r...
Signed-off-by: Sebastien Ponce <sebastien.ponce@cern.ch> Sebastien Ponce
05:07 PM Revision e1ad0bf6 (ceph): Added a striper interface on top of rados called radosstriper.
This interface allows to manipulate striped objects stored in a rados cluster with a standard open/read/write/stat/cl... Sebastien Ponce
05:07 PM Revision b9a35b3c (ceph): Extracted RadosXattrIter from librados.cc into independent .h anc .cc f...
This makes this interface usable by clients of librados.
Signed-off-by: Sebastien Ponce <sebastien.ponce@cern.ch>
Sebastien Ponce
04:19 PM Revision aede8328 (ceph): os: rename get_*() -> decode_*()
These methods have side-effects: they move the decode iterator *and*
return a value. Rename them to avoid confusion ...
Sage Weil
03:33 PM Revision 5bb078f9 (ceph): Populate ChangeLog when distribute
Generates ChangeLog from git history for a
distributed package.
Signed-off-by: Sahid Orentino Ferdjaoui <sahid.ferdj...
Sahid Orentino Ferdjaoui
03:33 PM Revision b22cc850 (ceph): Populate AUTHORS when distribute
Adds a complete list of contributors in a
distributed package.
Signed-off-by: Sahid Orentino Ferdjaoui <sahid.ferdja...
Sahid Orentino Ferdjaoui
01:48 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
attempting to reproduce Samuel Just
12:06 PM rgw Bug #8539: civetweb backend responds with a body when a HEAD request yields an error
Here you go. Sylvain Munaut
10:57 AM rgw Bug #8539: civetweb backend responds with a body when a HEAD request yields an error
This fix seems fine to me. Can you resend it with a proper description and a Signed-off-by line? Yehuda Sadeh
08:05 AM rgw Bug #8539 (Resolved): civetweb backend responds with a body when a HEAD request yields an error
So for example, doing a HEAD on a non-existent key will yield a 404, along with a Content-Length: 79 (which is allowe... Sylvain Munaut
10:55 AM rgw Bug #8540 (Rejected): ERROR: s3tests.functional.test_s3.test_multipart_upload
It fails on the new test for 8452, but the version running here is before the fix for it was merged. Closing. Yehuda Sadeh
08:55 AM rgw Bug #8540 (Rejected): ERROR: s3tests.functional.test_s3.test_multipart_upload
all (?) of the s3-tests tests on master failed like so:... Sage Weil
10:35 AM Feature #8546 (New): add nonce/sequence number associating the several pg metadata keys which are...
If they don't match, we have a indication that the pg metadata is inconsistent. At that point, there are some option... Samuel Just
09:48 AM CephFS Bug #8541 (Duplicate): [WRN] client.4131 released lease on dn 1000000016c/l72 which dne" in clust...
oh, i missed that this was dumpling. won't fix! Sage Weil
09:44 AM CephFS Bug #8541: [WRN] client.4131 released lease on dn 1000000016c/l72 which dne" in cluster log
So do we want to try and backport, or can we close this? Greg Farnum
09:03 AM CephFS Bug #8541 (Duplicate): [WRN] client.4131 released lease on dn 1000000016c/l72 which dne" in clust...
http://pulpito.ceph.com/teuthology-2014-06-02_19:03:01-fs-dumpling-testing-basic-plana/288497/
same symptom as #5871
Sage Weil
09:35 AM Documentation #8536: Ceph-deploy install on Centos and RHEL
Actually, these comments are reasonably prominent in the master branch in both the preflight and in the quick start, ... John Wilkins
09:30 AM rbd Bug #8543 (Rejected): run_xfstests: mkdir: cannot create directory `//var/lib/xfstests/common': F...
cool thanks! Sage Weil
09:26 AM rbd Bug #8543: run_xfstests: mkdir: cannot create directory `//var/lib/xfstests/common': File exists
The problem is this... Ilya Dryomov
09:14 AM rbd Bug #8543 (Rejected): run_xfstests: mkdir: cannot create directory `//var/lib/xfstests/common': F...
... Sage Weil
09:13 AM CephFS Bug #8542 (Resolved): kcephfs: fsx failure on read (expected 0's)
... Sage Weil
07:24 AM rgw Bug #8517 (Duplicate): Can't initiate multipart upload in firefly branch.
Ian Colle
12:37 AM rgw Bug #8517: Can't initiate multipart upload in firefly branch.
Duplicates #8311 ... I can't close it myself. Sylvain Munaut
04:53 AM Bug #8254 (Resolved): Not logging missing_on_shards properly
Loïc Dachary
12:49 AM rbd Bug #8184 (Resolved): krbd: make sure we have latest osdmap on 'rbd map'
Ilya Dryomov
12:49 AM Bug #8530 (Resolved): make tags broken by dummy.cc
Ilya Dryomov
12:36 AM rgw Bug #8311: No pool name error in ubuntu-2014-05-06_21:02:54-upgrade:dumpling-dumpling-testing-bas...
Oh damn, I reported #8517 which turns out to be a duplicate of this.
However I used the bucket.get_data_extra_pool...
Sylvain Munaut
12:32 AM Revision 7c2c8da2 (ceph): Merge pull request #1918 from ceph/wip-da-SCA-20140604
Some new fixes for gcc and cppcheck warnings/issues
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil

06/04/2014

11:56 PM Revision d76936b9 (ceph): OSD::calc_priors_during: fix confusing for loop bracing (cosmetic)
Confusing lack of braces is confusing.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Samuel Just
11:03 PM Revision 3261caad (ceph): Merge pull request #1582 from yuyuyu101/omap-cache
Add header cache to DBObjectMap
Reveiewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
11:01 PM Revision f3b8ebed (ceph): Merge pull request #1865 from dachary/wip-8254
osd: convert shard_id_t from uint8_t to struct + merge shard_t into it
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
10:42 PM Revision 8464a0c6 (ceph): Merge pull request #1917 from ceph/wip-da-update-libs3
libs3: update to latest git master of ceph/libs3
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
10:34 PM Revision 59d727de (ceph): kv_flat_btree_async.cc: remove consecutive break after return
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de> Danny Al-Gaaf
10:30 PM Revision 82f5df45 (ceph): JournalScanner.cc: catch exception by reference
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de> Danny Al-Gaaf
10:13 PM Revision 53533ee9 (ceph): KeyValueStore.cc: silence gcc -Wunused-variable
Don't assign return value of i.get_length() to a variable until
they are really used. Added comment on what is the in...
Danny Al-Gaaf
10:12 PM Revision e24213ed (ceph): MemStore.cc: silence gcc -Wunused-variable
Don't assign return value of i.get_length() to a variable until
they are really used. Added comment on what is the in...
Danny Al-Gaaf
10:07 PM Revision 51abf200 (ceph): Revert "Remove unused variables in MemStore.cc"
This reverts commit 307ba481a2705338560ecb557741d459706bdeb6.
The get_* methods have side effects!
Signed-off-by: S...
Sage Weil
10:06 PM Revision a325e3ec (ceph): Revert "Remove unused variables in KeyValueStore.cc"
This reverts commit 23b657c704865b33aa2e0e39995712c1f583c24d.
The get_* methods have side-effects!
Sage Weil
09:36 PM Revision cac902e3 (ceph): os/KeyValueStore.cc: fix possible null pointer deref warning
Add assert check for header->header and remove redundant 'if'
check for header again later.
Signed-off-by: Danny Al-...
Danny Al-Gaaf
09:29 PM Revision cadcbf39 (ceph): Remove unused import
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
09:22 PM Revision 3ee3e66a (ceph): librbd/internal.cc: check earlier for null pointer
Fix potential null ponter deref, move check for 'order != NULL'
to the beginning of the function to prevent a) deref ...
Danny Al-Gaaf
09:10 PM Revision f17a9633 (ceph): test/librbd/fsx.c: fix gcc warning
Initialize pointer with NULL before call posix_memalign() to fix
gcc warning:
test/librbd/fsx.c:1402:13: warning: ‘t...
Danny Al-Gaaf
08:58 PM Feature #8538 (Resolved): Functionality to have rbdmap also mount after mapping an image
In RE: https://inktank.zendesk.com/agent/#/tickets/1462
Summary: Over 200 rbd images are being mapped to a RHEL7 m...
JuanJose Galvez
08:38 PM Revision f31e4c83 (ceph): libs3: update to latest git master of ceph/libs3
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de> Danny Al-Gaaf
08:33 PM Revision 18c07ec5 (ceph): common/addr_parsing.c: fix realloc memory leak
Fix handling of realloc. If realloc() fails it returns NULL, assigning
the return value of realloc() directly to the ...
Danny Al-Gaaf
08:03 PM Revision 0b35e61b (ceph): Merge pull request #1916 from sahid/unused
Remove unused variables on MemStore.cc and KeyValueStore.cc
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
07:44 PM Revision 5f866526 (ceph): daemon_config.cc: add some more asserts
Add assert checks for some more set_val() calls to fix cppcheck issue
"Variable 'ret' is reassigned a value before th...
Danny Al-Gaaf
06:32 PM Documentation #8536: Ceph-deploy install on Centos and RHEL
That is mentioned in the preflight checklist, which is referred to by the quickstart docs. It's easy to miss, but it... Dan Mick
03:30 PM Documentation #8536 (Resolved): Ceph-deploy install on Centos and RHEL
When one first runs ceph-deploy by following either of the following documents:
http://ceph.com/docs/firefly/start...
Anonymous
06:28 PM Revision 7f35532b (ceph): Merge pull request #1914 from ceph/wip-autotools-dummy
automake: add dummy.cc to fix 'make tags'
Reviewed-by: Dan Mick <dan.mick@inktank.com>
Dan Mick
05:42 PM Revision 549b52b3 (ceph): Update --help output
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
04:42 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
Kernel is 3.14.4 (reboot was update from 3.14.3)
osd.0-62: Controller: AMD SATA onboard (few IDE compat, but most AH...
Markus Blank-Burian
04:15 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
The broken fs/controller/disk theory is the best I've got at the moment, but it's pretty much impossible to falsify m... Samuel Just
04:13 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
Were you able to localize all of the affected pgs to one or two osds at some point in the past? Also, what controlle... Samuel Just
03:57 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
Thanks for your help! After deleting some empty directories and removing a half full one, now all OSDs came up. I am ... Markus Blank-Burian
02:34 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
You can identify the broken pgs by looking at the logs on the crashing osds and tracing the thread which asserts back... Samuel Just
02:33 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
Ok, it looks like these pgs had an osd (or osds) power cycled in a way that caused the pg metadata to go significantl... Samuel Just
02:28 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
i made a list with the directory content on the osds:
hostname: osd
---------------
kaa-63: osd.40, contains dir...
Markus Blank-Burian
02:11 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
Is there a non-empty 504_head on any osd (perhaps 12 or 17?) Samuel Just
02:08 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
both osd.12 and osd.17 crashed, logs are attached. Markus Blank-Burian
01:56 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
on osd.64, the Directory current/0.504_head is empty. Markus Blank-Burian
01:48 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
Also, can you restart osd.12 and osd.17 and post the logs from that (same log levels). Samuel Just
01:45 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
Can you post a recursive ls on osd.64's current/0.504_head directory? Samuel Just
01:36 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
attached log from osd.64 for the same run Markus Blank-Burian
01:30 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
This time it crashed on a message from osd 64, did you happen to have the logging on for that one as well? If not, c... Samuel Just
01:22 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
ok, i now started other osds as well and reproduced the crash on osd.6 and attached the corresponding logs. i have sa... Markus Blank-Burian
12:54 PM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
Yeah, if it didn't crash you'll probably also have to start the other osds. I asked for osd.6 and osd.46 because osd... Samuel Just
11:39 AM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
i think, i had to powercycle osd.12 and osd.17, but the other ones were rebooted normally (i rechecked that for osd.6... Markus Blank-Burian
10:55 AM Bug #8532: 0.80.1: OSD crash (domino effect), same as BUG #8229
Can you restart osd.6 and osd.46 with
debug osd = 20
debug filestore = 20
debug ms = 1
and attach both logs?
...
Samuel Just
08:44 AM Bug #8532 (Can't reproduce): 0.80.1: OSD crash (domino effect), same as BUG #8229
I triggered the same issue as described in the bug http://tracker.ceph.com/issues/8229 on our HPC storage cluster. Fo... Markus Blank-Burian
04:24 PM Revision 703d0eb9 (ceph): rgw: set meta object in extra flag when initializing it
As part of the fix for 8452 we moved the meta object initialization.
Missed moving the extra flag initialization that...
Yehuda Sadeh
03:41 PM Revision 23b657c7 (ceph): Remove unused variables in KeyValueStore.cc
Removes two unused variables expected_object_size and
expected_write_size in KeyValueStore.cc
Signed-off-by: Sahid O...
Sahid Orentino Ferdjaoui
03:41 PM Revision 307ba481 (ceph): Remove unused variables in MemStore.cc
Removes two unused variables expected_object_size and
expected_write_size in MemStore.cc
Signed-off-by: Sahid Orenti...
Sahid Orentino Ferdjaoui
01:52 PM Revision 5185a368 (ceph): automake: add dummy.cc to fix 'make tags'
Commit 421e6c561704 ("test_librbd_fsx: add krbd mode support") added
a requirement for ceph_test_librbd_fsx to be lin...
Ilya Dryomov
12:23 PM Bug #8495: osd: bad state machine event on backfill request
I, see... Nice command by the way, thanks.
All cluster components are v0.80.1.
Dmitry Smirnov
10:27 AM Bug #8495: osd: bad state machine event on backfill request
Dmitry Smirnov wrote:
> Samuel Just wrote:
> > Also, is there any chance that this is a mixed cluster?
>
> I don...
Sage Weil
10:22 AM Bug #8495: osd: bad state machine event on backfill request
Samuel Just wrote:
> Also, is there any chance that this is a mixed cluster?
I don't know what "mixed cluster" is...
Dmitry Smirnov
12:19 PM Revision d4415cb8 (ceph): Merge pull request #1889 from sahid/master
trim dead code and add tests in bloom_filters
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
10:54 AM CephFS Tasks #8535 (Closed): audit the journaler send-to-OSD ordering
I've seen several cases recently where users had a journal header indicating that the journal ended one object later ... Greg Farnum
10:40 AM Bug #8519: msgr: deadlock, blocked on SimpleMessenger::lock
Sage was hoping we could just rearrange things a bit and drop the SimpleMessenger lock while waiting for the Connecti... Greg Farnum
10:11 AM Revision 35509d27 (ceph): bloom_filter, add test to validate assignement operator
Adds a test to ensure the good behavior of the
operator assignement.
Signed-off-by: Sahid Orentino Ferdjaoui <sahid....
Sahid Orentino Ferdjaoui
10:11 AM Revision 90cc6dd6 (ceph): bloom_filter, add assertion to test validate element_count()
Adds to TEST(BloomFilter, Basic) a new assertion
to verify the good behavior of the method 'element_count()'.
Signed...
Sahid Orentino Ferdjaoui
10:11 AM Revision c50f85ea (ceph): bloom_filter, remove unecessary operators
Removes 3 not used operators (intersect, union,
and diff) since after use those methods fpp and is_full are
corrupted...
Sahid Orentino Ferdjaoui
09:22 AM rgw Bug #8452: S3 multipart upload ignores request headers
As reported on the mailing list, there's an issue with the fix. I pushed a fix on top (there's a pull request for it)... Yehuda Sadeh
09:19 AM Revision c323c5b5 (ceph): Fix keyvaluestore fiemap bug
The result of fiemap is wrong and the offset get from
"StripObjectMap::file_to_extents" need to multiply by sequence ...
Haomai Wang
09:06 AM Bug #8530 (Fix Under Review): make tags broken by dummy.cc
Ilya Dryomov
08:24 AM Bug #8476 (Resolved): bloom_filter operators do not handle insert_count_
Loïc Dachary
07:03 AM Bug #8515 (Rejected): rbd ls or rbd info hangs for ever on specific client
So it turns out this was not a Ceph problem after all.
The cluster in this case runs over IPv6 and the network adm...
Wido den Hollander
04:57 AM Bug #8515: rbd ls or rbd info hangs for ever on specific client
Sage Weil wrote:
> Hrm, I'm not seeing what is going wrong here. ANy chance you can repeat this with debug ms = 20 ...
Wido den Hollander
06:56 AM CephFS Feature #8230: mds: new requests are not throttled by backend rados progress
This sounds like a general problem of throttling new work based on the MDS's ability to make progress against the rad... Sage Weil
05:38 AM devops Bug #7598 (In Progress): ceph-disk-activate error with ceph-deploy
Alfredo Deza
02:20 AM devops Bug #7598: ceph-disk-activate error with ceph-deploy
I came across the same issue.
In my case Centos 6.5 is the operating system. Running Ceph 0.80.1.
What i figured ou...
Thanassis Parathyras
05:13 AM devops Feature #5956 (Rejected): Implement a radosgw command in ceph-deploy
It is not feasible to do this anymore. Most (if not all) of the rados gateway install would be at the most a single l... Alfredo Deza
04:31 AM Bug #8498: udev rules not used on RHEL6/CentOS6
Yes, it is present as /lib/udev/rules.d/60-ceph-partuuid-workaround.rules. The rules seem to work, it's just that the... Stefan Walter
02:16 AM rbd Bug #8529 (Fix Under Review): vm get killed when leveldb backend enable
Hi xinxin,
The fix is here(https://github.com/ceph/ceph/pull/1912)
Haomai Wang
01:36 AM Revision ed3ec397 (ceph): Support installing ceph kernels from gitbuilder on rpm machines.
Signed-off-by: Sandon Van Ness <sandon@inktank.com> Sandon Van Ness

06/03/2014

11:22 PM Revision 6d5ab094 (ceph): Merge pull request #1907 from ceph/wip-8311
rgw: if extra data pool name is empty, use data pool name instead
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
11:21 PM Bug #8530 (Resolved): make tags broken by dummy.cc
... Dan Mick
11:06 PM Revision 8d002e9f (ceph): Merge pull request #1896 from ceph/wip-8452
rgw: fetch object attrs on multipart completion
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
11:02 PM Bug #8381: osd crash when osd use leveldb as filestore
haomai , i create a new bug report 8529 for segment fault of librados, it seem to be the problem of buffer class. Xinxin Shu
01:18 PM Bug #8381 (Resolved): osd crash when osd use leveldb as filestore
Sage Weil
09:10 AM Bug #8381: osd crash when osd use leveldb as filestore
Thanks xinxin, I would work on it Haomai Wang
01:30 AM Bug #8381: osd crash when osd use leveldb as filestore
i get the following error from dmesg, how can i identify this error :
[1117577.605328] virbr0: port 1(vnet0) enter...
Xinxin Shu
12:16 AM Bug #8381: osd crash when osd use leveldb as filestore
Hi Shaun,
The fix commit is https://github.com/ceph/ceph/pull/1840. I'm not familiar to ceph-deploy, hope the link c...
Haomai Wang
11:00 PM rbd Bug #8529 (Resolved): vm get killed when leveldb backend enable
hi all , i enabled leveldb backend , but i use 'attach-device' to attach rbd to vm , the vm get killed , i get a segm... Xinxin Shu
11:00 PM Revision 55fb5595 (ceph): Merge pull request #40 from ceph/wip-runxfstests
wip-runxfstests
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
10:58 PM Revision 4c2bee76 (ceph): Merge pull request #38 from ceph/wip-krbd-fsx
krbd: add librbd_fsx -K job
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
10:56 PM Revision 3ae54ead (ceph): Merge pull request #267 from ceph/wip-runxfstests
wip-runxfstests
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
10:53 PM Revision d52239d0 (ceph): Merge pull request #1906 from ceph/wip-runxfstests
wip-runxfstests
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
10:32 PM Revision 765ab804 (ceph): Merge pull request #1869 from ceph/wip-pybind-timeout
pybind/rados: Fix timeouts for small t
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
07:27 PM Revision 1bbebb54 (ceph): Merge pull request #1910 from ceph/wip-8447
librados: simplify/fix rados_pool_list bounds checks
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
07:12 PM devops Bug #8509: ceph-deploy always complaints "sys.excepthook is missing"
I applied the patch you reported on execnet's tracker and it worked.
Thank you.
Ke-fei Lin
11:51 AM devops Bug #8509 (Duplicate): ceph-deploy always complaints "sys.excepthook is missing"
This is a well known issue with Python threads and there is little we can do to mitigate them.
See issue #7594.
Alfredo Deza
06:45 PM Revision 3ec32a6b (ceph): librados: simplify/fix rados_pool_list bounds checks
We were not breaking out of the loop when we filled up the buffer unless
we happened to do so on a pool name boundary...
Sage Weil
05:39 PM Revision 0f04af07 (ceph): Add a timeout for unmounting
Ten minutes seems like it might be longer than necessary, but I doubt
it's too short.
Signed-off-by: Zack Cerza <zac...
Zack Cerza
04:43 PM Revision 3d0c2498 (ceph): Merge pull request #1909 from dachary/wip-8514-documentation
doc: fix osd pool create and add osd erasure-code-profile John Wilkins
04:31 PM Bug #8515: rbd ls or rbd info hangs for ever on specific client
Hrm, I'm not seeing what is going wrong here. ANy chance you can repeat this with debug ms = 20 on the osd and clien... Sage Weil
03:21 PM Bug #8515: rbd ls or rbd info hangs for ever on specific client
I also tried to upgrade the client to Firefly 0.80.1 but that didn't change anything. The end result is the same. Wido den Hollander
02:50 PM Bug #8515: rbd ls or rbd info hangs for ever on specific client
Sage Weil wrote:
> can you confirm that librados2 and ceph-common and librbd1 package versions all match?
>
Yes...
Wido den Hollander
01:20 PM Bug #8515 (Need More Info): rbd ls or rbd info hangs for ever on specific client
can you confirm that librados2 and ceph-common and librbd1 package versions all match?
if so, can you turn up logg...
Sage Weil
04:48 AM Bug #8515: rbd ls or rbd info hangs for ever on specific client
What I forgot to add is that stopping osd.56 doesn't work either,
osd.40 then becomes primary for that PG and the ...
Wido den Hollander
04:37 AM Bug #8515 (Rejected): rbd ls or rbd info hangs for ever on specific client
I'm observing something on a cluster which I can't explain and I think it's a bug in the OSD.
On a freshly install...
Wido den Hollander
04:28 PM Revision eb63015b (ceph): Merge pull request #1908 from jdurgin/wip-librbd-license-header
librbd: clarify license header to say LGPL
Everybody who has contributed has already explicitly agreed to relicense ...
Gregory Farnum
04:25 PM rgw Bug #8428 (Pending Backport): rgw: can't set bucket acls via swift api
commit:2ceb13af34bc38b418bb88d81131d770a71159bf Josh Durgin
04:21 PM rgw Bug #8311 (Pending Backport): No pool name error in ubuntu-2014-05-06_21:02:54-upgrade:dumpling-d...
commit:b300318113b162522759d4794b1cfa3d1d9398e4 Josh Durgin
04:13 PM Bug #8495: osd: bad state machine event on backfill request
Also, is there any chance that this is a mixed cluster? Samuel Just
01:10 PM Bug #8495: osd: bad state machine event on backfill request
Can you reproduce this with
debug osd = 20
debug ms = 1
debug filestore = 20
on all osds and attach all of the ...
Samuel Just
04:12 PM Bug #8507: OSD: failing to create PG (when splitting?) in rados/test.sh
Samuel Just
04:07 PM Bug #8507 (In Progress): OSD: failing to create PG (when splitting?) in rados/test.sh
Samuel Just
04:07 PM Bug #8507: OSD: failing to create PG (when splitting?) in rados/test.sh
Hmm, waiting for priors probably shouldn't be waiting on OSD -1 (Greg: yep, you were right). Looks like the priors c... Samuel Just
04:09 PM rgw Bug #8452 (Pending Backport): S3 multipart upload ignores request headers
Josh Durgin
04:03 PM Bug #8505: OSD osd/OSD.cc: 6222: FAILED assert(p->second.empty())
Samuel Just
04:02 PM Bug #8505: OSD osd/OSD.cc: 6222: FAILED assert(p->second.empty())
Actually, this should be fixed by the patch fixing 8396 Samuel Just
03:58 PM Bug #8361: osd/osd_types.h: 2868: FAILED assert(rwstate.empty())
That log does not have logging turned on. Add
debug osd = 20
debug ms = 1
to the [osd] section of the conf fi...
Samuel Just
03:42 PM Bug #8229: 0.80~rc1: OSD crash (domino effect)
I have the same issue as Dmitry with ceph 0.80.1. It happened after I restarted a some groups of OSDs, each one simul... Markus Blank-Burian
02:17 PM Bug #8512 (Duplicate): osd crashed at Pipe::reader in firefly
Samuel Just
02:15 PM devops Bug #8456 (Won't Fix): download links for RPMs do not show src packages
seems that this was just broken for emporor and not firefly/dumpling so we are closing this. Sandon Van Ness
02:15 PM Bug #8520 (Need More Info): osd: segv in PushOp::print()
I can't seem to actually find this thread in the core, which is odd. Marking as Need More Info for the moment. Samuel Just
09:37 AM Bug #8520 (Can't reproduce): osd: segv in PushOp::print()
... Sage Weil
02:07 PM devops Bug #7053 (Won't Fix): libcurl3-gnutls > 7.28 dependency backport for emperor
Sage Weil
02:03 PM devops Bug #7417 (Resolved): libleveldb1 missing in debian-testing
This was done like almost two weeks ago I believe, forgot to update. Sandon Van Ness
01:29 PM Bug #8315: osd: watch callback vs callback funky
Sage Weil
01:26 PM Bug #8301 (Resolved): Suicide Timeout on Cache Tier OSDs
Samuel Just
01:25 PM Bug #8375 (Can't reproduce): ceph_rest_api test (workunits/rest/test.py) failure
Sage Weil
01:24 PM Bug #8420 (Can't reproduce): iogen workunit not stopping properly
this looks like a hang in the fs client itself. no logs for this run, so we just need to wait for it to come up again Sage Weil
01:22 PM Bug #8498 (Need More Info): udev rules not used on RHEL6/CentOS6
There should be a rhel udev rule set up that hacks around the old version of blkid... see 60-ceph-partuuid-workaround... Sage Weil
01:13 PM devops Bug #7627: ceph-disk: does not start daemons properly under systemd
Alfredo - where are we at on the F20 packages? Ian Colle
01:12 PM Revision 60e6788d (ceph): Merge pull request #265 from ceph/wip-orchestra
Refactor teuthology.orchestra.run Alfredo Deza
12:53 PM Revision 5569d40b (ceph): documentation: add osd erasure-code-profile {set,rm,get,ls}
And a link from rados/operations.
http://tracker.ceph.com/issues/8514 Fixes: #8514
Reported-by: Kenneth Waegeman <K...
Loïc Dachary
12:52 PM Bug #8447 (Pending Backport): librados: buffer overflow in rados_pool_list
Sage Weil
11:45 AM Bug #8447 (Fix Under Review): librados: buffer overflow in rados_pool_list
Sage Weil
12:12 PM Bug #8465 (Resolved): osd/OSD.cc: 4768: FAILED assert(osd->is_active())
this was introduced with the fast dispatch stuff, which isn't (and won't be) backported Sage Weil
12:07 PM Bug #8522 (Duplicate): ENOENT on read from ceph_test_rados + osd thrashing
Samuel Just
12:06 PM Bug #8522: ENOENT on read from ceph_test_rados + osd thrashing
2014-06-02 17:24:13.162057 7fd39a414700 10 filestore(/var/lib/ceph/osd/ceph-2) remove 3.4e_head/96d896ce/plana505832-... Samuel Just
09:58 AM Bug #8522 (Duplicate): ENOENT on read from ceph_test_rados + osd thrashing
2014-06-02T17:24:13.234 INFO:teuthology.task.rados.rados.0.plana50.stderr:2494: Error: oid 42 read returned error cod... Sage Weil
11:47 AM Revision 8ff4edda (ceph): documentation: update osd pool create erasure
The properties are replaced with erasure code profiles. Remove the
reference to properties and the documentation of e...
Loïc Dachary
11:19 AM Feature #8523 (Resolved): pool quota: Should have get-quota
You can set a pool quota with ceph osd pool set-quota and specify max-objects or max-bytes. You cannot, however, get ... John Wilkins
10:32 AM Revision 22bc886a (ceph): doc: fix 'rbd map' example
Looks like 'foo' is a leftover, see commit ea9fc87d8996 ("doc: Removed
foo. Apparently myimage was added and foo not...
Ilya Dryomov
10:30 AM Bug #8519: msgr: deadlock, blocked on SimpleMessenger::lock
Okay, the blocking thread (LWP 23351) is waiting for the "existing" pipe to finish dispatching (LWP 23231).
The exis...
Greg Farnum
09:40 AM Bug #8519: msgr: deadlock, blocked on SimpleMessenger::lock
... Sage Weil
09:35 AM Bug #8519 (Resolved): msgr: deadlock, blocked on SimpleMessenger::lock
lots of threads stuck here:... Sage Weil
10:21 AM Bug #8514 (Resolved): ceph osd pool create documentation is obsolete
Sage says : no need to backport because the web is generated from master anyway. Loïc Dachary
04:49 AM Bug #8514 (Fix Under Review): ceph osd pool create documentation is obsolete
"need review":https://github.com/ceph/ceph/pull/1909 Loïc Dachary
03:38 AM Bug #8514 (Resolved): ceph osd pool create documentation is obsolete
the parameters introduced to support erasure code have been obsoleted by the erasure code profiles and the "documenta... Loïc Dachary
09:56 AM Bug #8521 (Can't reproduce): osd: hung LibRadosListECPP.ListObjectsPP
ubuntu@teuthology:/a/teuthology-2014-06-02_02:30:05-rados-master-testing-basic-plana/285552 Sage Weil
09:52 AM Bug #7891: osd: leaked pg refs on shutdown
ubuntu@teuthology:/a/teuthology-2014-06-02_02:30:05-rados-master-testing-basic-plana/285882 Sage Weil
09:16 AM Bug #7891: osd: leaked pg refs on shutdown
ubuntu@teuthology:/a/teuthology-2014-06-02_02:30:05-rados-master-testing-basic-plana/285880 Sage Weil
09:15 AM rbd Bug #8416: Client Crash when try to map a volume (ubuntu)
Hi Miguel,
Which kernel is this?
Ilya Dryomov
07:07 AM rgw Bug #8517 (Duplicate): Can't initiate multipart upload in firefly branch.
If you don't have any special placement info setup, you get a bucket.data_extra_pool that's empty. There is a get_dat... Sylvain Munaut
05:59 AM Bug #8476: bloom_filter operators do not handle insert_count_
According to the discuss on IRC, I'm going to fix the bug by removing the methods (intersect, union and diff)
PS: ...
Sahid Orentino Ferdjaoui
05:24 AM Revision 49512447 (ceph): Merge remote-tracking branch 'gh/next'
Sage Weil
05:24 AM Revision 4f834fa8 (ceph): doc/release-notes: v0.81
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
04:37 AM rgw Bug #8516 (Resolved): rgw: Can not get object correctly
For ec pool could work correctly, we adjusted the rgw_max_chunk_size to 640KB, in default it's 512KB. [you can see th... Jingjing Zhao
12:34 AM Revision f4e81d32 (ceph): librbd: clarify license header
These were meant to say LGPL, but a typo was propagated and it
referred to the non-existent GPL2.1, and also to COPYI...
Josh Durgin
12:01 AM Revision af73e308 (ceph): Remove some duplicate code.
Well, not exactly duplicate, but functionally compatible. No sense in
maintaining two "walk the filesystem to assembl...
Zack Cerza

06/02/2014

11:29 PM Revision 884a6b37 (ceph): RadosClient: Avoid the DispatchQueue for OSD responses.
Prior to this change, reader thread puts the response on the DispatchQueue
and Dispatcher thread would call librados ...
sushma
11:03 PM Revision 0cc9ade8 (ceph): doc: Updated monitor output and added usage calc explanations.
Fixes: #4948
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
10:31 PM Revision 9c32cb26 (ceph): doc: Added usage for pool quotas.
Fixes: #5653
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
10:17 PM devops Bug #8513 (Can't reproduce): s3tests failed at bootstrap in the nightlies
log: ubuntu@teuthology:/a/teuthology-2014-06-01_23:00:57-rgw-next-testing-basic-plana/284896... Tamilarasi muthamizhan
10:08 PM Bug #8512 (Duplicate): osd crashed at Pipe::reader in firefly
logs:ubuntu@teuthology:/a/teuthology-2014-06-01_23:00:57-rgw-next-testing-basic-plana/284939... Tamilarasi muthamizhan
10:05 PM Revision f6c4d006 (ceph): krbd: add librbd_fsx -K job
Our fsx fork has recently gained a support for the krbd mode - add it
to the test suite. Disable discard ops and ran...
Ilya Dryomov
09:53 PM rbd Bug #8409: "Segmentation fault" in rbd in upgrade:dumpling-dumpling-testing-basic-vps
recent log:
ubuntu@teuthology:/a/teuthology-2014-06-01_19:15:10-upgrade:dumpling-dumpling-testing-basic-vps/284023/
Tamilarasi muthamizhan
09:51 PM rgw Bug #8311: No pool name error in ubuntu-2014-05-06_21:02:54-upgrade:dumpling-dumpling-testing-bas...
recent log [just for reference]:
ubuntu@teuthology:/a/teuthology-2014-06-01_19:15:10-upgrade:dumpling-dumpling-testi...
Tamilarasi muthamizhan
02:19 PM rgw Bug #8311 (Fix Under Review): No pool name error in ubuntu-2014-05-06_21:02:54-upgrade:dumpling-d...
Yehuda Sadeh
09:40 PM Revision 86754cc7 (ceph): doc: Added more discussion of new CRUSH types.
Fixes: #7727
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
09:17 PM devops Bug #8509: ceph-deploy always complaints "sys.excepthook is missing"
I was about to report this for Ice installations as well. Anonymous
08:34 PM devops Bug #8509 (Duplicate): ceph-deploy always complaints "sys.excepthook is missing"
... Ke-fei Lin
08:36 PM devops Fix #8508: packaging: deb repository key should be @redhat.com
This was uncovered while doing complete tests on ICE install. It has no effect on development or testing, but I susp... Anonymous
08:32 PM devops Fix #8508 (Rejected): packaging: deb repository key should be @redhat.com
The following message appears while updating the repository and installing ceph-deploy... Anonymous
07:21 PM Revision cabb8f06 (ceph): doc: Added a section for ceph df.
Fixes: #8281
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
06:37 PM Revision 8de9501d (ceph): 0.81
Jenkins
06:28 PM Revision 4bc5aef9 (ceph): doc: Updated packages to reference firefly.
Fixes: #8466
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
06:06 PM Revision 4763be0c (ceph): Merge pull request #1887 from rossturk/wip-rturkfonts
doc: remove Apex Sans from generated doc HTML John Wilkins
06:05 PM Revision 920bc493 (ceph): Merge pull request #1885 from nereocystis/rados-bench-man
doc: Improve man page for bench John Wilkins
05:45 PM Revision 2d47ff9a (ceph): Merge pull request #1857 from ceph/wip-8265
#8265: config: make int parser accept K, M, G, T, P suffix and scale value accordingly
Reviewed-by: Loic Dachary <lo...
Sage Weil
05:00 PM Bug #8381: osd crash when osd use leveldb as filestore
We saw this issue as well. Is there a way we can pull this fix with ceph-deploy install --dev <Branch or Commit> to g... Shaun McDowell
04:26 PM Bug #8507 (Resolved): OSD: failing to create PG (when splitting?) in rados/test.sh
ubuntu@teuthology:/a/gregf-2014-05-29_16:19:17-rados-wip-client-fast-dispatch-testing-basic-plana/280674/remote/plana... Greg Farnum
04:10 PM Revision 4abb1018 (ceph): Merge pull request #1902 from dachary/wip-copyright-notices
erasure-code: Ceph distributed storage system
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
04:04 PM Documentation #7725 (Resolved): osd pool default size need updating
John Wilkins
04:02 PM Documentation #4948 (Resolved): Document how 'ceph -s' arrives at space numbers
John Wilkins
03:51 PM Documentation #4948 (In Progress): Document how 'ceph -s' arrives at space numbers
John Wilkins
03:48 PM Bug #8505 (Resolved): OSD osd/OSD.cc: 6222: FAILED assert(p->second.empty())
ubuntu@teuthology:/a/gregf-2014-05-29_16:19:17-rados-wip-client-fast-dispatch-testing-basic-plana/280831/remote/plana... Greg Farnum
03:44 PM Messengers Bug #8504: msgr: FAILED assert(0 == "old msgs despite reconnect_seq feature")
Probably a bug I and Guang introduced in commit:bdee119076dd0eb65334840d141ccdf06091e3c9, but perhaps not? Greg Farnum
03:43 PM Messengers Bug #8504 (Resolved): msgr: FAILED assert(0 == "old msgs despite reconnect_seq feature")
ubuntu@teuthology:/a/gregf-2014-05-29_16:19:17-rados-wip-client-fast-dispatch-testing-basic-plana/280874/remote/plana... Greg Farnum
03:41 PM Revision c18cbefe (ceph): qa: add run_xfstests_krbd.sh wrapper
As run_xfstests.sh does no longer accept test lists, wrap it to provide
an expunge file with a list of tests we want ...
Ilya Dryomov
03:41 PM Revision 703166c8 (ceph): qa: cp run_xfstests.sh run_xfstests-obsolete.sh
run_xfstests.sh is going to be updated in the next commit to be able to
drive newer xfstests. Among other things, th...
Ilya Dryomov
03:41 PM Revision cd652461 (ceph): qa: catch up with xfstests changes
Back in 2013 xfstests were rearranged, which also changed the way
./check parses test lists. Catch up with those cha...
Ilya Dryomov
03:31 PM Documentation #5653 (Resolved): Pool quota needs documentation created
Added usage. This feature doesn't have a corresponding "get-quota" method. John Wilkins
02:42 PM Documentation #5653 (In Progress): Pool quota needs documentation created
John Wilkins
03:13 PM Bug #8497 (Duplicate): OSDs stats getting DOWN , OUT , when i starts putting data to cluster
This is a known issue and will be fixed in the next firefly release. :) Greg Farnum
02:59 AM Bug #8497: OSDs stats getting DOWN , OUT , when i starts putting data to cluster

Logs in readable format
osd.123 - http://pastebin.com/QiAP5qGt
osd.118 - http://pastebin.com/GBdBAWg8
...
karan singh
02:54 AM Bug #8497 (Duplicate): OSDs stats getting DOWN , OUT , when i starts putting data to cluster
Hi
I have build a fresh Ceph cluster , stable firefly release 0.80.1, CentOS6.4
15 Node Cluster , 3 monitors , 1...
karan singh
03:11 PM Feature #8389: osd: clean up old ec objects more aggressively
Samuel Just
03:11 PM Feature #8480: modify scrub to detect/repair obsolete rollback objects
Samuel Just
03:08 PM Bug #8438: erasure code: object are not cleanup
Samuel Just
02:47 PM Bug #8333: ceph_test_rados_delete_pools_parallel: Received fewer notifies than expected: 0 < 1
ubuntu@teuthology:/a/gregf-2014-05-29_16:19:17-rados-wip-client-fast-dispatch-testing-basic-plana/280879 Greg Farnum
02:39 PM Documentation #7727 (Resolved): CRUSH "chassis" bucket type
John Wilkins
02:38 PM Documentation #7727 (In Progress): CRUSH "chassis" bucket type
John Wilkins
02:23 PM Revision 3968f6f7 (ceph): radosgw-admin: adjust bucket link interface
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
02:23 PM Revision ae0d26ae (ceph): Try a different approach to zombie prevention
The previous method I had committed had the unfortunate side-effect of
preventing teuthology job processes from givin...
Zack Cerza
02:12 PM Documentation #7726 (Resolved): Quick Start Requires Updating
John Wilkins
12:27 PM Documentation #8281 (Resolved): Documentation: Detailed explanation of ceph df output is non-exis...
http://ceph.com/docs/master/rados/operations/monitoring/#checking-a-cluster-s-usage-stats John Wilkins
11:42 AM Documentation #8281 (In Progress): Documentation: Detailed explanation of ceph df output is non-e...
John Wilkins
11:29 AM Documentation #8466 (Resolved): get-packages still references emperor on code examples and states...
John Wilkins
11:13 AM Documentation #8466 (In Progress): get-packages still references emperor on code examples and sta...
John Wilkins
10:44 AM Feature #8265 (Resolved): config: make int parser accept K, M, G, T, P suffix and scale value acc...
Sage Weil
09:44 AM Fix #8205 (In Progress): FileStore: properly fill in XATTR_NO_SPILL_OUT tag
I ran this branch through testing and it wasn't passing; it appears to be some kind of issue with the spillout logic ... Greg Farnum
09:09 AM rbd Bug #8184 (Fix Under Review): krbd: make sure we have latest osdmap on 'rbd map'
Ilya Dryomov
08:58 AM rbd Bug #6385 (Fix Under Review): run_xfstests.sh: doesn't work with newly-organized upstream tests
Ilya Dryomov
05:38 AM Feature #8499 (Closed): add python 3 support for python scripts
As more distributions start supporting python 3 (and in some cases Python 3 only) it is becoming quite problematic to... Alfredo Deza
04:40 AM Bug #8498 (Can't reproduce): udev rules not used on RHEL6/CentOS6
I am running OSDs with a journal on a dedicated, raw SSD
and let ceph-disk create the journal partitions. When
I bo...
Stefan Walter
02:37 AM Bug #8387: osd: skipping missing maps broken
Unfortunately i was running development release of ceph, sage helped me to give 2 possible fixes.
1) Apply patch w...
karan singh

06/01/2014

08:51 PM Bug #8381: osd crash when osd use leveldb as filestore
It seemed that no useful info can get from log.
My ceph cluster is compiled from master branch.
(qemu-kvm-1.2.0)...
Haomai Wang
08:32 AM Revision 601e25eb (ceph): erasure-code: Ceph distributed storage system
instead of Ceph - scalable distributed file system in the copyright notices
Signed-off-by: Loic Dachary <loic@dachar...
Loïc Dachary
03:33 AM Bug #8438: erasure code: object are not cleanup
Why is the behaviour different with replicated pools? The object's deletion is instantaneous.
Ok so let's imagine ...
Sébastien Han

05/31/2014

09:54 PM Bug #8381: osd crash when osd use leveldb as filestore
hi haimao , i can reproduce this error from time to time, the attached file is detail log for client, btw, can you li... Xinxin Shu
03:23 PM Feature #8496 (Fix Under Review): erasure-code: ErasureCode base class
"need review":https://github.com/ceph/ceph/pull/1903 Loïc Dachary
11:51 AM Feature #8496 (Resolved): erasure-code: ErasureCode base class
Create an ErasureCode base class from which ErasureCodeJerasure can derive. Move non jerasure specific logic to Erasu... Loïc Dachary
03:13 PM Revision 9bac31be (ceph): scripts/run-coverity: don't explicitly specify tool version
Rely on a symlink instead.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
11:58 AM Revision beaf4525 (ceph): Merge pull request #1901 from dachary/wip-erasure-code-concat
erasure-code: make decode_concat virtual
Reviewed-By: Christophe Courtaut <christophe.courtaut@gmail.com>
Loïc Dachary
11:57 AM Revision 0a2aa5e1 (ceph): Merge pull request #1900 from dachary/wip-ostream
common: s/stringstream/ostream/ in str_map
Reviewed-By: Christophe Courtaut <christophe.courtaut@gmail.com>
Loïc Dachary
11:38 AM Revision e158ad96 (ceph): erasure-code: make decode_concat virtual
So that a plugin can provide a more efficient implementation.
Reviewed-By: Christophe Courtaut <christophe.courtaut@...
Loïc Dachary
11:20 AM Revision 35fb7419 (ceph): Merge pull request #1898 from dhxgit/master
Update INSTALL to mention the submodules/recursive
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
11:16 AM Revision 86c60f7d (ceph): rbd: change test_script to run_xfstests_krbd.sh
Invoke run_xfstests.sh through a wrapper, which supplies an exclusion
list, in order to skip tests we don't want to r...
Ilya Dryomov
11:13 AM Revision 6aa45b13 (ceph): common: s/stringstream/ostream/ in str_map
There is no need to specialize more than ostream : it only makes it
impossible to use cerr or cout as a parameter to ...
Loïc Dachary
11:11 AM Revision 2551c528 (ceph): rbd: recognize 'randomize' parameter
randomize (default: false) randomizes xfstests test order.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Ilya Dryomov
10:51 AM Revision 670735f7 (ceph): krbd: catch up with run_xfstests.sh changes
In the current xfstests, tests cannot be listed in ranges. Because of
this, and because our distribution of tests be...
Ilya Dryomov
09:31 AM Revision 319cb504 (ceph): Make KeyValueStore support "ceph osd perf" command
Signed-off-by: Haomai Wang <haomaiwang@gmail.com> Haomai Wang
09:22 AM Revision 06c0a427 (ceph): Update INSTALL to mention the submodules/recursive
Signed-off-by: Dominik Hannen <cantares1+github@gmail.com> Dominik Hannen
08:27 AM Revision 1c91e65d (ceph): Merge pull request #1895 from ceph/wip-sock-cmp
msg: remove comparison operators for sockaddr_storage
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
02:52 AM Bug #8495 (Duplicate): osd: bad state machine event on backfill request
Three OSDs crashed together for no apparent reason during routine backfilling/remapping.
Situation: 12 OSDs on 5 hos...
Dmitry Smirnov
02:07 AM Revision 2dbd85c9 (ceph): WorkQueue: The variable name is corrected.
Modified the variable name from shardedpol_cond->shardedpool_cond
Signed-off-by: Somnath Roy <somnath.roy@sandisk.com>
Somnath Roy
02:03 AM Revision 3e3632ee (ceph): WorkQueue: Removed the unused in_process variable
Signed-off-by: Somnath Roy <somnath.roy@sandisk.com> Somnath Roy
01:46 AM Revision b05da1ca (ceph): WorkQueue: Taking care of potential race condition during pause()
Introduced two variables to keep track of number of threads paused
and drained during threadpool pause/drain. The pau...
Somnath Roy
01:45 AM Revision b15bf6bb (ceph): OSD:Derived sharded queue implementation is changed
All the threadpool related stuff like stop/pause/drain etc. are not
handled by sharded queue anymore. All it is imple...
Somnath Roy
01:45 AM Revision c24ef007 (ceph): ceph-common: The Sharded threadpool worker logic changed
Now, the _process() of the derived queue is processing one request
at a time and the outer loop is controlled by the ...
Somnath Roy
01:44 AM Revision 06845d0b (ceph): OSD: Sharded Op worker queue implementation for handling OSD ops
This is the implementation for the client of the
sharded thread pool/sharded workQ. Removed the op_wq class and now O...
Somnath Roy
01:44 AM Revision 8369c08e (ceph): ceph-common: Implementation of the sharded threadpool.
Threadpool will only be having a single work queue and internally the
work queue will be having multiple storage data...
Somnath Roy

05/30/2014

11:28 PM Revision 241a1025 (ceph): Merge pull request #266 from ceph/wip-8489
Give daemons a five-minute timeout for stopping
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
11:26 PM Revision 83ccba32 (ceph): msg: remove comparison operators for sockaddr_storage
We don't need these at all, it turns out.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
11:20 PM Revision 70afaaab (ceph): rgw: fetch object attrs on multipart completion
Fixes: #8452
Backport: firefly
This fixes a regression following a code cleanup.
Signed-off-by: Yehuda Sadeh <yehuda...
Yehuda Sadeh
10:18 PM Revision 669b6057 (ceph): PGLog: initialize complete_to
Signed-off-by: Samuel Just <sam.just@inktank.com> Samuel Just
09:49 PM Revision b3003181 (ceph): rgw: if extra data pool name is empty, use data pool name instead
Fixes: #8311
An empty pool name could be used as the extra data pool.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh
09:39 PM Revision 205c455f (ceph): Give daemons a five-minute timeout for stopping
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
09:36 PM Revision 7035c17e (ceph): Add RemoteProcess.__repr__()
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
09:09 PM Revision 67dc5751 (ceph): common/strtol.cc: strict_strtosi() converts str with SI units to uint64_t
Accepts values with a suffix (B, K, M, G, T, P, E) and returns the
appropriate byte value.
E.g., 10B = 10, while 10K...
Joao Eduardo Luis
09:09 PM Revision 40587d47 (ceph): test/strtol.cc: Test 'strict_strtosi()'
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com> Joao Eduardo Luis
09:09 PM Revision 5500437e (ceph): common/config.cc: allow integer values to be parsed as SI units
We are allowing this for all and any integer values; that is, OPT_INT,
OPT_LONGLONG, OPT_U32 and OPT_U64.
It's on th...
Joao Eduardo Luis
09:09 PM Revision 38405d35 (ceph): qa/workunits/cephtool: test setting options using SI units
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com> Joao Eduardo Luis
09:05 PM Revision b386f5e5 (ceph): Refactor teuthology.orchestra.run
RemoteProcess behaves more like subprocess.Popen, with some important
differences.
A summary of the API changes:
* R...
Zack Cerza
09:05 PM Revision eac2c2ab (ceph): Update users of the teuthology.orchestra.run APIs
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
08:32 PM Revision 8d04a7ec (ceph): HashIndex: in cleanup, interpret missing dir as completed merge
If we stop between unlinking the empty subdir and removing the root
merge marker, we get ENOENT on the get_info. Tha...
Samuel Just
08:32 PM Revision 308e2ef5 (ceph): ReplicatedPG::start_flush: send delete even if there are no snaps
Even if all snaps for the clone have been removed, we still have to
send the delete to ensure that when the object is...
Samuel Just
08:32 PM Revision 54a89bab (ceph): ReplicatedPG::start_flush: fix clone deletion case
dsnapc.snaps will be non-empty most of the time if there
have been snaps before prev_snapc. What we really want to
k...
Samuel Just
05:13 PM Revision 3b884bec (ceph): Merge remote-tracking branch 'gh/next'
Sage Weil
04:51 PM Revision 9c56c86b (ceph): rgw: calc md5 and compare if user provided appropriate header
Fixes: #8436
Backport: firefly
This was broken in ddc2e1a8e39a5c6b9b224c3eebd1c0e762ca5782. The fix
resurrects and o...
Yehuda Sadeh
04:29 PM rgw Bug #8452: S3 multipart upload ignores request headers
Also updated the s3 test suite to trigger this issue. When pulling this in, also merge that test (wip-8452). Yehuda Sadeh
04:28 PM rgw Bug #8452 (Fix Under Review): S3 multipart upload ignores request headers
Yehuda Sadeh
04:15 PM Revision 72851f13 (ceph): Merge pull request #39 from ceph/wip-multimds-syntax
multimds: fix fuse_default_permissions syntax Zack Cerza
04:00 PM Revision 5ca3c723 (ceph): multimds: fix fuse_default_permissions syntax
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
03:50 PM Bug #8404 (Duplicate): OSD: stopped doing useful work until scrub timeouts nudged PGs
This is a duplicate of #8338. That took too long to figure out. :( Greg Farnum
02:32 PM Bug #8404 (New): OSD: stopped doing useful work until scrub timeouts nudged PGs
Wow, this is crazy:
The MDS is trying to log the ESession(close) event to the OSD, but the OSD connection isn't work...
Greg Farnum
03:44 PM Revision b2bf8cb2 (ceph): add backport of collections.Counter for python2.6
Using Raymond Hettinger's MIT backport
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from co...
Alfredo Deza
02:59 PM Revision 59ee17dc (ceph): Avoid piles of zombies
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
02:56 PM Revision 0fcfaee7 (ceph): Add comment explaining 'preexec_fn=os.setpgrp'
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
02:23 PM devops Feature #7947 (Duplicate): Create separate ceph and ceph-common packages for EL6 and EL7 builds
Ian Colle
02:21 PM devops Tasks #8240 (Resolved): Build 0.67.8 & 0.80 on RHEL7-RC
This has been done for a while. Sandon Van Ness
02:17 PM devops Tasks #8240 (Rejected): Build 0.67.8 & 0.80 on RHEL7-RC
Neil Levine
01:42 PM rbd Feature #7746: Capacity Management: rbd df
Rather than show RBD pool usage via ceph df, the suggestion was a dedicated rbd df command which shows provisioned an... Neil Levine
01:40 PM Feature #7515 (Fix Under Review): mon: 'osd crush reweight-subtree <name> <weight>'
Sage Weil
01:35 PM Feature #8480 (Resolved): modify scrub to detect/repair obsolete rollback objects
Samuel Just
01:32 PM Bug #8334 (Resolved): osd: bug with pool snaps, ceph_test_rados
Samuel Just
01:32 PM Bug #8332 (Resolved): ceph-test-objectstore: bad return value in unlink
Samuel Just
01:31 PM Feature #8389 (In Progress): osd: clean up old ec objects more aggressively
Sage Weil
12:55 PM devops Bug #8384 (Resolved): broken tgt makes teuthology tests fail
This was resolved last week in ceph-qa-chef commit: 5cd0516cf080211ee822bb33f8b0e05fdbc8d5f8 Sandon Van Ness
09:56 AM rgw Bug #8169 (Pending Backport): rgw: swift user manifest does not compute etag
Sage Weil
09:54 AM rgw Bug #8436 (Pending Backport): rgw: ignores provided md5 for object upload
Sage Weil
09:40 AM Feature #7928 (Rejected): erasure-code : no SSE3 specific code
It is not worth it. Loïc Dachary
09:31 AM Bug #8465 (Pending Backport): osd/OSD.cc: 4768: FAILED assert(osd->is_active())
Sage Weil
09:12 AM Bug #8381: osd crash when osd use leveldb as filestore
Hi,xinxin.
I boot the vm with keyvaluestore backend, and attach new disk. There nothing happened.
Haomai Wang
03:28 AM Bug #8381: osd crash when osd use leveldb as filestore
You can add these lines to /etc/ceph/ceph.conf which run qemu
[client]
log_file = /var/log/ceph/ceph-rbd.log
admin...
Haomai Wang
02:28 AM Bug #8381: osd crash when osd use leveldb as filestore
hi haomai , after apply your patch, after i run 'virsh attach-device' to attach rbd, the vm is killed, however , when... Xinxin Shu
08:50 AM devops Bug #8321 (Resolved): ceph-brag missing a python dependency on EL6
Backported to Firefly b2bf8cb21761d88557fb2d5c0935c4a3678ee01f Alfredo Deza
08:46 AM Bug #8083 (In Progress): erasure-code: fix static code analysis errors found in gf-complete
Loïc Dachary
08:03 AM Revision 8691773e (ceph): Merge pull request #1888 from ceph/wip-8472
msg_types.h: Don't use strncmp for comparing sockaddr_in.
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
07:34 AM Bug #8476: bloom_filter operators do not handle insert_count_
Indeed, it is kind of difficult to compute the insert_count_ since the operations are doing by manipulate the bit_tab... Sahid Orentino Ferdjaoui
07:18 AM Bug #8476: bloom_filter operators do not handle insert_count_
"related pull request":https://github.com/ceph/ceph/pull/1889/files Loïc Dachary
06:57 AM Bug #8476: bloom_filter operators do not handle insert_count_
I don't understand, what's the bug here? I don't think we actually use those operators to begin with, and operations ... Greg Farnum
06:44 AM Bug #8476 (Resolved): bloom_filter operators do not handle insert_count_
According to the source code it seems that 'insert_count_' is not taken into account when manipulating objects using ... Sahid Orentino Ferdjaoui
06:56 AM Feature #8475: erasure-code: oversized objects when using the Cauchy technique
"need review":https://github.com/ceph/ceph/pull/1890 Loïc Dachary
06:30 AM Feature #8475 (Resolved): erasure-code: oversized objects when using the Cauchy technique
When the Cauchy technique is used, the alignment constraint and the default packetsize will inflate a 4KB object to 16KB Loïc Dachary
04:04 AM CephFS Bug #8010 (In Progress): It's impossible to remove unused filesystem pools from a cluster
John Spray
01:03 AM Bug #8472 (Resolved): Using strncmp to compare sockaddr_in.
Loïc Dachary

05/29/2014

11:20 PM Revision cae085b6 (ceph): msg_types.h: Don't use strncmp for comparing sockaddr_in.
Fixes: #8472
Signed-off-by: Dan Mick <dan.mick@inktank.com>
Dan Mick
11:03 PM Revision 00b92119 (ceph): doc: Improve man page for rados bench
Add information about seq and rand for bench. Also mention
--no-cleanup option.
Signed-off-by: Kevin Dalley <kevin@k...
Kevin Dalley
10:24 PM Bug #8472 (Fix Under Review): Using strncmp to compare sockaddr_in.
Dan Mick
03:39 PM Bug #8472: Using strncmp to compare sockaddr_in.
Dan Mick
02:30 PM Bug #8472 (Resolved): Using strncmp to compare sockaddr_in.
In /src/msg/msg_types.h strncmp is being used to compare sockaddr_in structures. This means that it will stop at the... Kevin Cox
10:08 PM Revision 2bfb8d48 (ceph): Merge pull request #1886 from nereocystis/pools-doc
doc: update pools documentation
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
09:56 PM Revision 2da26999 (ceph): doc: update pools documentation
set=>get in "Get Pool Values"
grep for 'replicated size', not 'rep size'
Signed-off-by: Kevin Dalley <kevin@kelphead...
Kevin Dalley
09:28 PM Revision ab59a100 (ceph): doc: Improve man page for bench
Add information about seq and rand for bench. Also mention
--no-cleanup option.
Signed-off-by: Kevin Dalley <kevin@k...
Kevin Dalley
09:20 PM Revision dedd8e26 (ceph): osd: explicit shard_id_t() and NO_SHARD
Whenever an integer needs to be converted in a shard_id_t,
replace the implicit conversion with an explicit call to t...
Loïc Dachary
09:20 PM Revision 0fd6e12b (ceph): osd: replace shard_t usage with shard_id_t
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
09:19 PM Revision 20e66d16 (ceph): osd: loop over uint8_t instead of shard_id_t
When looping in the range [0,number of shards[ using a uint8_t variable
which is an integer instead of shard_id_t. Th...
Loïc Dachary
09:12 PM Revision 47e6fc2b (ceph): Merge pull request #1884 from nereocystis/glossary-rados
doc: Define RADOS in glossary John Wilkins
09:11 PM Revision d723cac0 (ceph): osd: factorize shard_id_t/shard_t into a struct
The struct contains a single uint8_t and is preferred to control
conversions that would otherwise be implicit. There ...
Loïc Dachary
09:03 PM Revision dcf9568b (ceph): doc: Define RADOS in glossary
Signed-off-by: Kevin Dalley <kevin@kelphead.org> Kevin Dalley
08:22 PM Revision e52b9c64 (ceph): doc: Added osd pool default size setting example.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
08:22 PM Revision 20a04c6c (ceph): doc: Moved redundant text out of quick-common.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
07:53 PM Revision 6786d609 (ceph): common: WRITE_{EQ,CMP}_OPERATORS_1
For when a struct is defined to encapsulate a single value and get
control over implicit conversions that would other...
Loïc Dachary
07:27 PM Revision e5f9d40b (ceph): Merge pull request #1882 from ceph/wip-8465
osd: fix bad is_active() assert in share_map()
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
07:25 PM Revision 7ac05dc0 (ceph): Merge pull request #1840 from yuyuyu101/wip-8381
Fix set_alloc_hint op cause KeyValueStore crash problem
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
07:10 PM Revision 8679cdbe (ceph): osd_types: add pg_log_t::rollback_info_trimmed_to
Signed-off-by: Samuel Just <sam.just@inktank.com> Samuel Just
06:50 PM Revision afaa7a58 (ceph): Merge pull request #37 from ceph/wip-heartbeat
osd_heartbeat_grace from 20 -> 40 when running valgrind Zack Cerza
06:24 PM Revision 1db4e2cf (ceph): Merge pull request #1874 from ceph/wip-8292
better error reporting on incompatible device requirements
Reviewed-by: Dan Mick <dan.mick@inktank.com>
Sage Weil
06:23 PM rgw Bug #8442: rgw: does not detect/adapt to erasure pool stripe size
Hi sage,
Thanks, it's work!
In addition? I also really want to know design consideration of comment #1.
Thanks...
Jingjing Zhao
05:16 PM rgw Bug #8442: rgw: does not detect/adapt to erasure pool stripe size
Hi Sage,
Can you comment on the design consideration of comment #1? With erasure coding, I think it might be more ef...
Guang Yang
09:47 AM rgw Bug #8442: rgw: does not detect/adapt to erasure pool stripe size
you can manually adjust rgw_max_chunk_size to 640k to work around this for now. we need to make rgw automagically de... Sage Weil
09:46 AM rgw Bug #8442: rgw: does not detect/adapt to erasure pool stripe size
Sage Weil
06:23 PM Revision b26ba605 (ceph): Check for broken repos before updating
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
06:10 PM Revision f1b890ec (ceph): osd: fix bad is_active() assert in share_map()
We may be is_stopping() here if we are racing with shutdown().
Fixes: #8465
Backport: firefly, dumpling
Signed-off-b...
Sage Weil
05:43 PM Revision 0b00443e (ceph): rest-api: key missing for per "rx" and "rwx"
commit 85a1cf31e6 and db266a3fb2 introduce new per "rx" and "rwx", but key missing for per "rx" and "rwx" in permmap
...
Ailing
05:42 PM Revision 18d56c27 (ceph): Merge pull request #1880 from ailzhang/patch-1
rest-api: key missing for per "rx" and "rwx"
Backport: firefly
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
05:40 PM Revision 72fbff6f (ceph): Merge pull request #1847 from ceph/wip-8259
README, Makefile: make install-deps easier
Reviewed-by: Loic Dachary <loic@dachary.org>
Sage Weil
05:37 PM Revision e54f5e3a (ceph): Merge pull request #1860 from ceph/wip-filestore
osd: filestore backend improvements
Reviewed-by: Loic Dachary <loic@dachary.org>
Sage Weil
05:31 PM rgw Feature #8473 (Resolved): rgw: Shard bucket index objects to improve single bucket PUT throughput
There was a blueprint talking about the bucket index scalability issue - https://wiki.ceph.com/Planning/Sideboard/rgw... Guang Yang
05:26 PM Revision 49ac0c0e (ceph): Merge pull request #1864 from dachary/wip-filestore
unit tests for #1860 Sage Weil
04:30 PM devops Feature #8306 (Fix Under Review): separate ceph.rpm into ceph and ceph-common
this is comitted to branch wip-8306 but it needs review. Sandon Van Ness
02:30 PM CephFS Fix #8471 (New): mds: use Connection-based messenger interfaces instead of addr-based ones
We're trying to ditch the address-based messenger interfaces throughout our codebase. We're trying harder now that we... Greg Farnum
12:44 PM Bug #8445: osd not starting anymore
Thank you Sage.
Tyler pointed me to this patch, we are testing it.
Pierre Grandin
09:42 AM Bug #8445: osd not starting anymore
This looks like a bug we fixed in firefly. If you don't need these OSDs to recover, I suggest upgrading and replacin... Sage Weil
11:58 AM Bug #8396: osd: message delayed in Session misdirected after split
Samuel Just
11:42 AM Revision bc26a555 (ceph): Merge pull request #1879 from ceph/wip-8321
fix incompatible `Counter` usage on Python 2.6
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
11:23 AM devops Bug #8292 (Resolved): ceph-disk prepare output not explicit on too small disk
Sage Weil
11:09 AM Bug #8465 (Fix Under Review): osd/OSD.cc: 4768: FAILED assert(osd->is_active())
https://github.com/ceph/ceph/pull/1882
Sage Weil
10:05 AM Bug #8465 (Resolved): osd/OSD.cc: 4768: FAILED assert(osd->is_active())
ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-05-28_02:30:04-rados-next-testing-basic-plana/278237<p... Sage Weil
10:39 AM Bug #8259 (Resolved): add a target to ensure new dependencies are installed before builds
Sage Weil
10:08 AM Documentation #8466 (Resolved): get-packages still references emperor on code examples and states...
http://ceph.com/docs/master/install/get-packages/
- instructs to add http://ceph.com/debian-emperor/ as a repo on de...
Duarte Alencastre
09:56 AM devops Bug #8321 (Pending Backport): ceph-brag missing a python dependency on EL6
Loïc Dachary
04:43 AM devops Bug #8321 (Resolved): ceph-brag missing a python dependency on EL6
Loïc Dachary
09:36 AM Bug #8439 (Need More Info): ceph-osd crashing often
Can you try upgrading to 0.80 and see if this is still present? There were so many things fixed between 78 and 80 it... Sage Weil
09:31 AM Linux kernel client Bug #8464: krbd: deadlock
yeah, looks like it's just because the block io is hung. also, this:... Sage Weil
07:57 AM Linux kernel client Bug #8464 (Resolved): krbd: deadlock
http://pulpito.ceph.com/teuthology-2014-05-27_23:09:10-krbd-firefly-testing-basic-plana/277770/... Zack Cerza
08:57 AM Bug #7891: osd: leaked pg refs on shutdown
ubuntu@teuthology:/a/teuthology-2014-05-28_02:30:04-rados-next-testing-basic-plana/278239 Sage Weil
08:32 AM devops Bug #8441 (Resolved): remove `-q` from pkg manager's install calls
Merged into ceph-deploy's master branch with hash e5aef1b
Alfredo Deza
07:40 AM devops Bug #8441 (Fix Under Review): remove `-q` from pkg manager's install calls
pull request opened https://github.com/ceph/ceph-deploy/pull/193 Alfredo Deza
07:51 AM rgw Bug #8436 (Fix Under Review): rgw: ignores provided md5 for object upload
pull request #1862 Yehuda Sadeh
07:48 AM rgw Bug #8436: rgw: ignores provided md5 for object upload
doh, commit:ddc2e1a8e39a5c6b9b224c3eebd1c0e762ca5782 (#8169), the other commit is for #8170 which is quite harmless. Yehuda Sadeh
05:37 AM Revision abc722c8 (ceph): More verbose logging for daemon restarts
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
04:40 AM Revision 2363a17c (ceph): Bump log level to info
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
12:49 AM Revision f153bc18 (ceph): doc: Made additional changes s/CRUSH/Ceph for pull request #1855
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
12:46 AM Revision cbd75987 (ceph): Merge pull request #1855 from kazhang/patch-1
doc: Update architecture.rst:Calculating PG IDs John Wilkins
12:33 AM Revision 1edb8821 (ceph): Merge pull request #1866 from cmattson/radosgwdoc
doc: Improvements to radosgw docs John Wilkins
12:31 AM Revision dd85e2d6 (ceph): Merge pull request #1881 from hsdhillon/master
doc: Fixes broken hyperlinks John Wilkins
12:13 AM Support #8462 (New): related to integrity of objects
1)how should i ensure my object is altered or not while am using or download that object.
2)how do i check integrity...
vootla srinivas

05/28/2014

10:55 PM Revision c08f4810 (ceph): doc: alter doc CSS for the removal of Apex Sans
Signed-off-by: Ross Turk <rturk@redhat.com> Ross Turk
10:52 PM Revision 7f46b7b5 (ceph): doc: removed Apex Sans font, replace with font from GFE
Signed-off-by: Ross Turk <rturk@redhat.com> Ross Turk
10:12 PM Revision 692f9989 (ceph): doc: Fixes broken hyperlinks
Harpreet Dhillon
09:55 PM Revision 93a61dfc (ceph): Makefile: make install-deps rule
This only works if autoconf is happy, but it's better than nothing.
Fixes: #8259
Signed-off-by: Sage Weil <sage@inkt...
Sage Weil
09:55 PM Revision 6fe75642 (ceph): debian: improve package dep list
This is based on the Build-Depends in debian/control.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
09:51 PM Revision 2fcdbf9c (ceph): Merge pull request #1812 from ceph/wip-smart-df
mon: show writeable space in 'ceph df'
Reviewed-by: Samuel Just <sam.just@inktank.com>
Sage Weil
08:59 PM Revision e06c58c9 (ceph): mon: set min_size to data chunk count for erasure pools
Make the min_size value meaningful for erasure pools.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
08:57 PM Revision 7a9652b5 (ceph): mon: include 'max avail' in df output
Include an estimate of the maximum writeable space for each pool. Note
that this value is a conservative estimate fo...
Sage Weil
08:52 PM Revision 2339d4a5 (ceph): vstart.sh: -e to create an ec pool called 'ec'
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
08:52 PM Revision 2f63a309 (ceph): mon: right justify df values
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
08:49 PM Revision 297f6169 (ceph): crush: add get_rule_weight_map
Calculate a weight map of OSDs for a given rule.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
07:56 PM Revision c4783543 (ceph): Don't block on teuthology-results processes
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
07:37 PM Revision 0b5a6741 (ceph): rest-api: key missing for per "rx" and "rwx"
commit 85a1cf31e6 and db266a3fb2 introduce new per "rx" and "rwx", but key missing for per "rx" and "rwx" in permmap
...
Ailing
05:26 PM Revision 26cd37cf (ceph): Merge pull request #1878 from ceph/wip-map-advance
osd: fix map advance limit to handle map gaps
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
05:11 PM Revision c427af07 (ceph): osd_heartbeat_grace from 20 -> 40 when running valgrind
This will mask occasional 'wrongly marked me down' errors when valgrind
is running (it makes things generally slow).
...
Sage Weil
04:56 PM RADOS Feature #8461: Weighted monitors
What kind of weighting system are you actually interested in? Would the weights be a per-client setting or global? If... Greg Farnum
04:40 PM RADOS Feature #8461 (New): Weighted monitors
For geographically disperse monitors, or when monitors are asymmetrical in performance characteristics, it may be ben... Kyle Bader
04:54 PM Bug #8453 (Rejected): OSD crash "error (28) No space left on device not handled on operation 10 (...
The OSD is definitely getting back ENOSPC from the local filesystem; many of them keep quite a lot of reserved space ... Greg Farnum
03:11 AM Bug #8453 (Rejected): OSD crash "error (28) No space left on device not handled on operation 10 (...
hello,
tonight one of OSDs crashed with "error (28) No space left on device not handled on operation 10 (31417988.0....
Henrik Korkuc
04:43 PM Revision a4a9a1f2 (ceph): Correct logging when running teuthology-results
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
03:54 PM Revision 634780a2 (ceph): remove unused variables, gets all tox envs passing
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
03:48 PM Revision 23b75b55 (ceph): add backport of collections.Counter for python2.6
Using Raymond Hettinger's MIT backport
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
Alfredo Deza
03:27 PM Bug #8332 (Pending Backport): ceph-test-objectstore: bad return value in unlink
Sage Weil
03:26 PM Bug #8332 (Resolved): ceph-test-objectstore: bad return value in unlink
Sage Weil
03:24 PM rgw Bug #8288 (Duplicate): rgw tests cause osd wrongly marked down
dup is #8319, now fixed. Sage Weil
02:55 PM Revision 59b7113c (ceph): intial take on ceph-brag-client and tox. Python 2.7 passes
Signed-off-by: Alfredo Deza <alfredo@deza.pe> Alfredo Deza
02:34 PM Revision baeeddbe (ceph): Merge pull request #263 from qakapil/patch-two
updated the get_system_type function Zack Cerza
02:31 PM Revision 0b97eba3 (ceph): updated the get_system_type function
updated the get_system_type function
to add the rpm condition for SUSE.
Signed-off-by: ksharma@suse.com
Kapil Sharma
02:10 PM rgw Bug #8436: rgw: ignores provided md5 for object upload
Was actually broken in commit:5cc5686039a882ad345681133c9c5a4a2c2fd86b Yehuda Sadeh
01:43 PM Revision 3bf5338c (ceph): Merge pull request #259 from ceph/wip-journal-tool
mds_journal_migration task John Spray
01:39 PM Revision 79dfe2cb (ceph): tasks: Use '_' instead of '-' in names
Python modules are not allowed to contain hyphens
in the name. Using hyphens only works here because
we're using low...
John Spray
01:39 PM Revision 554f8c81 (ceph): task/ceph: separate ceph.conf into function
This is so that other tasks can modify ceph.conf
as needed.
Signed-off-by: John Spray <john.spray@inktank.com>
John Spray
01:39 PM Revision 6f801513 (ceph): tasks/mds_journal_migration: create.
This is for testing the changes to CephFS
journal format, especially the migration
of journals from old to new format...
John Spray
01:00 PM devops Bug #8458 (Resolved): `make check` should call tox tests for ceph-brag, ceph-disk, and ceph
There is a clear need for better testing in multiple Python environments (Ceph supports both 2.6 and 2.7) but
there ...
Alfredo Deza
12:31 PM devops Bug #8456 (Won't Fix): download links for RPMs do not show src packages
e.g. this link http://ceph.com/rpm/rhel6/SRPMS/
Does not have src packages for Ceph since v 0.72.2 which does not ...
Alfredo Deza
12:01 PM Revision f7f7659b (ceph): Merge pull request #1853 from dachary/wip-mailmap
mailmap updates
each commit has been submitted to the author for review
Loïc Dachary
11:58 AM Revision 7582fa98 (ceph): mailmap: Christian Theune affiliation
Reviewed-by: Christian Theune <ct@gocept.com>
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
11:58 AM Revision 0b886598 (ceph): mailmap: Chris Glass affiliation
Reviewed-by: Chris Glass <tribaal@gmail.com>
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
11:58 AM Revision bc85b5d1 (ceph): mailmap: Aristoteles Neto affiliation
and name normalization
Reviewed-by: Aristoteles Neto <aristoteles.neto@webdrive.co.nz>
Signed-off-by: Loic Dachary <...
Loïc Dachary
11:58 AM Revision c7585846 (ceph): mailmap: Kevin Dalley affiliation
Reviewed-by: Kevin Dalley <kevin@kelphead.org>
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
11:58 AM Revision be9f7437 (ceph): mailmap: Daniel J. Hofmann affiliation
Reviewed-by: Daniel J. Hofmann <daniel@trvx.org>
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
11:58 AM Revision 1f225906 (ceph): mailmap: Stephen F Taylor affiliation
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
11:58 AM Revision 831b2c92 (ceph): mailmap: Stefan Eilemann affiliation
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
11:58 AM Revision 63385429 (ceph): mailmap: Michael Riederer affiliation
and name normalization
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
11:58 AM Revision dbc66d7b (ceph): mailmap: Walter Huf affiliation
Reviewed-by: Walter J. Huf <hufman@gmail.com>
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
11:58 AM Revision e4928c50 (ceph): mailmap: Stuart Longland affiliation
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
11:58 AM Revision e65b7e25 (ceph): mailmap: Ilya Dryomov name normalization
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
11:58 AM Revision b2cb36f2 (ceph): mailmap: Shawn Edwards affiliation
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
11:57 AM Revision 7d1a4937 (ceph): mailmap: Florent Flament affiliation
and name normalization
Reviewed-by: Florent Flament <florent.flament@cloudwatt.com>
Signed-off-by: Loic Dachary <loi...
Loïc Dachary
11:45 AM Revision 9bfdc689 (ceph): Merge pull request #1877 from ceph/wip-tier-doc
doc: Fix cache tier docs
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
11:21 AM Revision 2815e4d1 (ceph): doc: Fix cache tier docs
The instructions for 'osd tier add' were plain
wrong, claiming that the cache mode could
be passed as a third argumen...
John Spray
10:32 AM Bug #8345: PG::repair_object() should check for and return errors
Based on the discussion we had today during the standup, my understanding is that the resolution of this problem shou... Loïc Dachary
08:57 AM Bug #8345: PG::repair_object() should check for and return errors
Loïc Dachary
10:24 AM Bug #8387 (Resolved): osd: skipping missing maps broken
Sage Weil
09:43 AM Bug #8387 (Fix Under Review): osd: skipping missing maps broken
see wip-map-advance Sage Weil
10:21 AM Bug #8395 (Resolved): ceph-test-objectstore doesn't clean up
Sage Weil
10:19 AM Revision d14dfac2 (ceph): Merge pull request #1856 from dachary/wip-pg-documentation
documentation: adapt PG formula for erasure coded pools
Reviewed-By: Christophe Courtaut <christophe.courtaut@gmail....
Loïc Dachary
10:15 AM Revision ffbb69fd (ceph): Merge pull request #1867 from dachary/wip-8324
documentation: update pid file description
Reviewed-by: Florent Flament <florent.flament@cloudwatt.com>
Loïc Dachary
09:43 AM Revision 08b3cff5 (ceph): os: FileStore::create_backend unit tests
Assert that the fs_type is mapped to a backend that claims to have the
matching name.
Signed-off-by: Loic Dachary <l...
Loïc Dachary
09:43 AM Revision 30816521 (ceph): os: ObjectStore::collect_metadata unit tests
And an emacs-executable reminder about how to run a specific test.
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
08:59 AM devops Bug #8321 (Fix Under Review): ceph-brag missing a python dependency on EL6
Pull request opened https://github.com/ceph/ceph/pull/1879 Alfredo Deza
06:50 AM Bug #8324 (Resolved): pid files are not created
Loïc Dachary
03:12 AM Revision c995a1c1 (ceph): lib: implemement parsecmdlineargs
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
01:17 AM Tasks #8437 (Won't Fix): erasure code: use cauchy_good instead of reed_sol_van
"reed solomon is best":http://dachary.org/?p=3042 Loïc Dachary
01:14 AM Bug #8440 (Resolved): cache tier: cache_target_full_ratio can't be changed
Loïc Dachary
01:01 AM rgw Bug #8442: rgw: does not detect/adapt to erasure pool stripe size
Hello,
May I ask what is the design consideration behind to use multiple AIO writes (default size is 512KB) for a si...
Guang Yang

05/27/2014

10:42 PM Revision 2916148f (ceph): crush: add adjust_subtree_weight()
Adjust the weight for all (leaf) items beneath a subtree.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
10:42 PM Revision 576315ac (ceph): mon/OSDMonitor: add 'osd crush reweight-subtree ...'
Reweight all items within a subtree. For example, you might reweigh tall
OSDs within a rack to a new value.
Signed-...
Sage Weil
10:42 PM Revision 97023dcc (ceph): mon/OSDMonitor: simplify 'osd crush reweight ...' flow
Use the goto labels; drop the do/while.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
10:06 PM Revision 84e6a61f (ceph): Merge pull request #1813 from ceph/wip-old-out
mon: restore osd weight after auto-mark out and then osd restart
Reviewed-by: Samuel Just <sam.just@inktank.com>
Rev...
Samuel Just
10:04 PM Revision f0316411 (ceph): Merge pull request #1811 from ceph/wip-perf-atomic
perf_counter: use atomics instead of mutex
Reviewed-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Samuel Just <sam...
Samuel Just
10:00 PM Revision e5a17212 (ceph): Merge pull request #1873 from dachary/wip-8440
mon: fix set cache_target_full_ratio
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
08:20 PM rgw Bug #8452 (Resolved): S3 multipart upload ignores request headers
rgw 0.80.1 ignores at least the @Content-Type@, @x-amz-acl@, and @x-amz-meta-*@ request headers in an S3 Initiate Mul... Benjamin Gilbert
08:16 PM Revision 4021fb69 (ceph): osd/: move split_list and split_request into OSD
Signed-off-by: Samuel Just <sam.just@inktank.com> Samuel Just
08:04 PM Revision d113965a (ceph): Merge pull request #1876 from atwardowski/patch-4
Fix Documentation Typo
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
07:41 PM Revision 26862caf (ceph): Fix Documentation Typo
Signed-off-by: Adam Twardowski <adam.twardowski@gmail.com> Adam Twardowski
06:17 PM Revision d4e81198 (ceph): doc: Added primary affinity. Some minor rewrap edits.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
04:54 PM Revision 1ac3a503 (ceph): better error reporting on incompatible device requirements
Signed-off-by: Alfredo Deza <alfredo@deza.pe> Alfredo Deza
01:51 PM Revision 78313b7d (ceph): Merge pull request #1739 from ceph/wip-journal-tool
cephfs-journal-tool John Spray
01:21 PM Bug #8438 (In Progress): erasure code: object are not cleanup
Oops, int Samuel Just
10:43 AM Bug #8438: erasure code: object are not cleanup
Samuel Just
10:42 AM Bug #8438: erasure code: object are not cleanup
This may be correct-ish behavior. The ec__head_F79165DA__4_80_1 file/object will be removed when the log entry at ve... Samuel Just
12:08 PM Bug #8447 (Resolved): librados: buffer overflow in rados_pool_list
When input `len` is small and non-zero `strncat` will correctly avoid overflowing the input buffer, but then `len -= ... Noah Watkins
09:59 AM Bug #8241: XfsFileStoreBackend tries to set extsize but may get EINVAL
Hi Pierre,
I don't think your crash (#8439) has anything to do with this. set_extsize is simply an allocation hin...
Ilya Dryomov
09:29 AM Bug #8241: XfsFileStoreBackend tries to set extsize but may get EINVAL
I am probably facing the same issue. Initially reported as #8439
Did any of you fix the issue on your own cluster?...
Pierre Grandin
09:56 AM devops Bug #8292 (Fix Under Review): ceph-disk prepare output not explicit on too small disk
Pull request opened: https://github.com/ceph/ceph/pull/1874 Alfredo Deza
09:50 AM Bug #8445 (Won't Fix): osd not starting anymore
Beside #8439 we seem to have another issue on some on our OSDs.
I'm attaching the full log of this session ( Durat...
Pierre Grandin
09:41 AM Revision db90d86e (ceph): Merge pull request #1824 from dachary/wip-documentation
erasure-code: fix URL in developer notes
Reviewed-by: koleosfuscus <koleosfuscus@yahoo.com>
Loïc Dachary
08:53 AM Bug #8439: ceph-osd crashing often
As noticed by Vacuity on IRC, it's possibly a dup of #8241 Pierre Grandin
08:38 AM Bug #8439: ceph-osd crashing often
Adding my ceph.conf:... Pierre Grandin
08:24 AM Revision c2225f87 (ceph): mon: fix set cache_target_full_ratio
It was a noop because it was incorrectly using the variable n. Add a
test to protect against regression.
http://trac...
Loïc Dachary
05:58 AM devops Bug #8401 (Resolved): ceph-deploy must warn against custom cluster names when system is sysvinit
Merged into ceph-deploy's master branch with hash e7c5b6e Alfredo Deza
05:40 AM rgw Bug #8442 (Resolved): rgw: does not detect/adapt to erasure pool stripe size
*Step:*
1. Put object which size is 4MB by rados gateway (S3).
2. An error returned. The error is HTTP/1.1 500 I...
Jingjing Zhao
05:39 AM devops Bug #8441 (Resolved): remove `-q` from pkg manager's install calls
ceph-deploy has some internal checks that will make sure that there is some output within 300 seconds otherwise it wi... Alfredo Deza
02:40 AM Documentation #8388 (Resolved): Broken link in developer_notes for erasure_coding
Loïc Dachary
02:28 AM Documentation #8388: Broken link in developer_notes for erasure_coding
I built the documents locally. Problem solved. koleos fuscus
01:21 AM Bug #8440 (Fix Under Review): cache tier: cache_target_full_ratio can't be changed
"need review":https://github.com/ceph/ceph/pull/1873 Loïc Dachary
01:00 AM Bug #8440 (Resolved): cache tier: cache_target_full_ratio can't be changed
h3. Workaround
Set "osd_pool_default_cache_target_full_ratio":https://github.com/ceph/ceph/blob/firefly/src/common...
Geoffrey Hartz

05/26/2014

07:43 PM Revision bb67ba13 (ceph): lib: implement GetClusterStats
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
06:38 PM Revision fc036b6e (ceph): lib: implement WaitForLatestOSDMap
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
06:14 PM Revision 8659a523 (ceph): doc: update readme with example
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
06:11 PM Revision c4f0d8cb (ceph): lib: implement SetConfigOption
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
05:53 PM Revision eb1efa6e (ceph): lib: add conn.ListPools
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
05:24 PM Bug #8439 (Won't Fix): ceph-osd crashing often
We have most of our OSDs crashing at least twice a day since a couple of days.
Here's the logs of such a crash.
...
Pierre Grandin
09:21 AM Bug #8438: erasure code: object are not cleanup
Just tried to run a deep-scrub but this didn't trigger any deletion...
$ sudo ceph pg deep-scrub 4.1da
instructin...
Sébastien Han
07:50 AM Bug #8438 (Resolved): erasure code: object are not cleanup
While playing with the erasure code, I noticed that erasure coded objects were not cleaned up.
My setup:
* Debian...
Sébastien Han
05:57 AM Feature #8265 (Fix Under Review): config: make int parser accept K, M, G, T, P suffix and scale v...
forgot to update the ticket last week
https://github.com/ceph/ceph/pull/1857
Joao Eduardo Luis
05:52 AM Bug #8301: Suicide Timeout on Cache Tier OSDs
After creating and digging through many more gcore, valgrind, and collectl logs, it appears that this issue is relate... Mark Nelson

05/25/2014

04:34 PM Revision 4c22c6f9 (ceph): pybind/rados: Fix timeouts for small t
Previously, if passing a number less than 0.5 for
the timeout, operations would always return
a failure status code.
...
John Spray
03:38 PM Revision 1e0a82fd (ceph): osd: fix map advance limit to handle map gaps
The recent change in cf25bdf6b0090379903981fe8cee5ea75efd7ba0 would stop
advancing after some number of epochs, but d...
Sage Weil
09:05 AM Revision 641732d5 (ceph): documentation: update pid file description
The pid file behavior changed over time and the description is misleading.
http://tracker.ceph.com/issues/8324 Fixes...
Loïc Dachary
09:03 AM Tasks #8437 (Won't Fix): erasure code: use cauchy_good instead of reed_sol_van
The cauchy_good technique is faster than reed_sol_van. It should be used instead. It was not used initially because i... Loïc Dachary
06:05 AM Bug #8387: osd: skipping missing maps broken
Hi
This problem occurred somewhere between 16 - 18th May . Before this cluster was healthy.
# As far as i reme...
karan singh
03:31 AM Bug #8345 (Fix Under Review): PG::repair_object() should check for and return errors
"need review":https://github.com/ceph/ceph/pull/1868 Loïc Dachary
03:15 AM Revision c9ff4818 (ceph): Improvements to radosgw docs
- Correct a typo (`rados-admin`) in admin docs
- Reorder sections in config so "above example" is in expected position
Colin Mattson
02:06 AM Bug #8324 (Fix Under Review): pid files are not created
"need review":https://github.com/ceph/ceph/pull/1867 Loïc Dachary

05/24/2014

10:58 PM Revision e741ea4e (ceph): os/FileStore: include filestore backend, f_type in osd metadata
Report the FileStore backend and underlying fs type as part of the osd
metadata shared with the monitor.
Signed-off-...
Sage Weil
06:42 PM Revision 8ba5ad18 (ceph): lib: rename Open to NewConn
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
06:41 PM Revision 56038c26 (ceph): doc: update method doc
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
06:28 PM Revision af6b92d0 (ceph): doc: update method docs
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
06:13 PM Revision e9bf7323 (ceph): doc: add readme
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
05:36 PM Revision 3a6efa08 (ceph): initial commit: starting go-rados library
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
05:00 PM Bug #8324 (In Progress): pid files are not created
The "default for pid_file":https://github.com/ceph/ceph/blob/firefly/src/common/config_opts.h#L30 in firefly is indee... Loïc Dachary
04:15 PM Bug #8254 (Fix Under Review): Not logging missing_on_shards properly
"need review":https://github.com/ceph/ceph/pull/1865 Loïc Dachary
02:57 PM Feature #7571 (Resolved): erasure code: implicit ruleset creation when a pool is created
Loïc Dachary
02:50 PM Bug #8387: osd: skipping missing maps broken
Hello Sage
As per our discussion on mailing list and IRC , i tried your procedure for this problem. Its been over...
karan singh
08:57 AM Revision 6bb0a048 (ceph): Merge pull request #1861 from atwardowski/patch-3
Fix Doc Typo
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
07:03 AM Revision 82e40bcf (ceph): Merge pull request #1846 from ceph/wip-8342
init-ceph: keep trying other OSDs if we fail to set CRUSH position on one OSD
Reviewed-by: Loic Dachary <loic@dachar...
Loïc Dachary
01:43 AM Bug #6043 (Won't Fix): upstart does not reflect running ceph-osd daemons (ubuntu 13.04 only)
Loïc Dachary
01:36 AM Bug #7188 (Won't Fix): Admin socket files are lost on log rotation calling initctl reload (ubuntu...
Document the proposed workaround in the description for reference. Loïc Dachary
12:05 AM Bug #8342 (Resolved): init script may not start all OSDs
Loïc Dachary

05/23/2014

11:35 PM Bug #7188: Admin socket files are lost on log rotation calling initctl reload (ubuntu 13.04 only)
We had the same issue a few months back (http://tracker.ceph.com/issues/6043). We cheated and updated ceph's logrotat... Hunter Nield
09:52 AM Bug #7188: Admin socket files are lost on log rotation calling initctl reload (ubuntu 13.04 only)
Mail sent asking for advice on the best workaround to James Page and James Hunt cc'd ceph-devel. Loïc Dachary
08:44 AM Bug #7188: Admin socket files are lost on log rotation calling initctl reload (ubuntu 13.04 only)
The problem is confirmed to not happen on ubuntu *trusty* Loïc Dachary
07:43 AM Bug #7188: Admin socket files are lost on log rotation calling initctl reload (ubuntu 13.04 only)
On a ceph 0.80.1 on a freshly installed ubuntu *raring*, when running... Loïc Dachary
11:17 PM Bug #8342: init script may not start all OSDs
Sage, your fix works perfectly. Thank you.
Dmitry Smirnov
08:37 PM Revision b4a55d76 (ceph): Merge pull request #1858 from ceph/wip-8428
rgw: check appropriate entity permission on put_metadata
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
07:35 PM Revision 827df7d5 (ceph): Fix Doc Typo
Signed-off-by: Adam Twardowski <adam.twardowski@gmail.com> Adam Twardowski
06:47 PM Revision 3cf723c5 (ceph): os/FileStore: fix journal-less operation
A few different fixes here:
- make the op_seq correct on start when there is no replay
- make an explicit 'no journa...
Sage Weil
06:47 PM Revision 9312c5ea (ceph): os/FileStore: refactor backend instantiation
Use a factory method for creating the backend so that the logic is not
duplicated in mkfs() and _detect_fs(). Make t...
Sage Weil
05:38 PM Revision 200d0ae9 (ceph): ceph-disk: Enable creating multiple osds per dev
This is useful for demos and testing. Enables
creation of lots of OSDs on a single block device
by simply running ce...
John Spray
04:40 PM Revision 2ceb13af (ceph): rgw: check appropriate entity permission on put_metadata
Fixes: #8428
Backport: firefly
Cannot use verify_object_permission() to test acls, as the operation
here might eithe...
Yehuda Sadeh
04:10 PM Bug #8396 (In Progress): osd: message delayed in Session misdirected after split
Samuel Just
11:20 AM Bug #8396: osd: message delayed in Session misdirected after split
It's the waiting_for_pg queue. It used to get reprocessed through the op queue, but now gets directly enqueue'd into... Samuel Just
11:04 AM Bug #8396: osd: message delayed in Session misdirected after split
Hmm... op misdirected to the pre-split pgid (3.7 not 3.47)... I suspect the new queuing code. Samuel Just
02:52 PM rgw Bug #8436 (Resolved): rgw: ignores provided md5 for object upload
broken in commit:01235f4365d00771ac4f8c1ce33f1bccdea77d3c Yehuda Sadeh
12:38 PM CephFS Bug #8432: ceph-fuse process not dying
I bet these are duplicates of #8427, but I haven't checked yet. Greg Farnum
08:46 AM CephFS Bug #8432: ceph-fuse process not dying
Also:
http://pulpito.ceph.com/teuthology-2014-05-20_19:02:54-fs-dumpling-testing-basic-plana/267406/
http://qa-prox...
Zack Cerza
08:41 AM CephFS Bug #8432 (Closed): ceph-fuse process not dying
http://pulpito.ceph.com/ubuntu-2014-05-19_20:36:16-multimds-wip-multimds-testing-basic-plana/264360/
http://qa-proxy...
Zack Cerza
11:34 AM rbd Bug #8425: osd crashed in rados-dumpling-testing-basic-plana suite
Same problem in http://qa-proxy.ceph.com/teuthology/teuthology-2014-05-21_19:55:29-upgrade:dumpling-x:stress-split-fi... Yuri Weinstein
09:43 AM rgw Bug #8428 (Fix Under Review): rgw: can't set bucket acls via swift api
Yehuda Sadeh
04:54 AM rgw Bug #8428: rgw: can't set bucket acls via swift api
Ian Colle
08:52 AM Bug #8433: SSHException: Key-exchange timed out waiting for key negotiation
Also:
http://pulpito.ceph.com/teuthology-2014-05-20_19:55:02-upgrade:dumpling-x:stress-split-firefly---basic-plana/2...
Zack Cerza
08:51 AM Bug #8433: SSHException: Key-exchange timed out waiting for key negotiation
The last host mentioned is 10.214.132.4... Zack Cerza
08:49 AM Bug #8433 (Duplicate): SSHException: Key-exchange timed out waiting for key negotiation
http://pulpito.ceph.com/teuthology-2014-05-20_19:55:02-upgrade:dumpling-x:stress-split-firefly---basic-plana/268335/
...
Zack Cerza
08:24 AM Bug #8429 (Resolved): admin socket: commands always return 0 even if erroneous
It appears that the admin socket interface will only propagate error codes to the user if the command itself outputs ... Joao Eduardo Luis
05:25 AM Revision ab0db34e (ceph): documentation: adapt PG formula for erasure coded pools
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
12:37 AM Revision d3af8fa8 (ceph): Update architecture.rst:Calculating PG IDs
To Calculate PG ID, if I didn't get it wrong, CRUSH calculates the hash modulo
the number of PGs instead of OSDs, acc...
Kai Zhang

05/22/2014

10:37 PM Revision a898ee56 (ceph): cephfs-java: build against older jni headers
Older versions of the JNI interface expected non-const parameters
to their memory move functions. It's unpleasant, bu...
Greg Farnum
09:34 PM Revision 04f8e0ef (ceph): Merge pull request #1854 from ceph/wip-java-build
cephfs-java: build against older jni headers
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
09:01 PM Revision 7ad2fbf3 (ceph): ReplicatedPG: lock snapdir obc during write
Otherwise, we won't block properly in prep_push_backfill_object.
Signed-off-by: Samuel Just <sam.just@inktank.com>
(...
Samuel Just
06:20 PM rgw Bug #8428 (Resolved): rgw: can't set bucket acls via swift api
The swift POST on bucket is broken. Yehuda Sadeh
05:26 PM Revision d3c23007 (ceph): Don't crash when an invalid branch is passed
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
05:20 PM Revision 74f736d3 (ceph): Merge pull request #258 from ceph/wip-fix-format-samba
Fixed formatting and added a better variable for time out
Conflicts:
teuthology/task/samba.py
Zack Cerza
05:07 PM Revision 8a12a940 (ceph): OSDMonitor: set next commit in mon primary-affinity reply
Commit 8c5c55c8b47e ("mon: set next commit in mon command replies")
fixed MMonCommand replies to include the right ve...
Ilya Dryomov
05:06 PM rbd Bug #8425 (Duplicate): osd crashed in rados-dumpling-testing-basic-plana suite
Tamilarasi muthamizhan
08:56 AM rbd Bug #8425 (Duplicate): osd crashed in rados-dumpling-testing-basic-plana suite
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-05-20_19:00:34-rados-dumpling-testing-basic-plana/267... Yuri Weinstein
05:05 PM rgw Bug #8213 (Duplicate): RGW is creating empty pool names
Samuel Just
05:05 PM rgw Bug #8311: No pool name error in ubuntu-2014-05-06_21:02:54-upgrade:dumpling-dumpling-testing-bas...
looks like something that needs to be backported to dumpling? Samuel Just
05:04 PM rgw Bug #8311: No pool name error in ubuntu-2014-05-06_21:02:54-upgrade:dumpling-dumpling-testing-bas...
(05:04:35 PM) sjusthm: yehudasa: ping
(05:04:45 PM) sjusthm: in the mon log
(05:04:46 PM) sjusthm: I see
(05:04:48...
Samuel Just
04:30 PM Bug #8346: OSD crashes on master (FAILED assert(ip_op.waiting_for_commit.count(from)))
Hmm, ceph-osd.0.log appears to have a bunch of 0s in the middle of the file. Samuel Just
04:17 PM Bug #7999: osd: pgs share info that hasn't been persisted
Samuel Just
04:15 PM Bug #8361 (Need More Info): osd/osd_types.h: 2868: FAILED assert(rwstate.empty())
Samuel Just
04:15 PM Bug #8361: osd/osd_types.h: 2868: FAILED assert(rwstate.empty())
Any chance of being able to reproduce this with debug osd = 20, debug ms = 1? Samuel Just
04:13 PM Bug #8315: osd: watch callback vs callback funky
Also, the bug manifested in dumpling, but appears to exist in the same form in master as well. Samuel Just
04:13 PM Bug #8315: osd: watch callback vs callback funky
Tricky... check_blacklisted_obc_watchers does not clear watch->cb before calling handle_watch_timeout. Watch.cc sugg... Samuel Just
03:36 PM Bug #8359 (Resolved): FTBFS with gcj-jdk v4.9.0
In firefly branch, commit:a898ee56e01f971c12c0f7bebc22e59f850f27c0 Greg Farnum
02:34 PM Bug #8359 (Pending Backport): FTBFS with gcj-jdk v4.9.0
This got merged in to master by commit:04f8e0effaae54d1cc4c4b34d30db48b85039595; I guess we need to backport it to Fi... Greg Farnum
09:35 AM Bug #8359 (Fix Under Review): FTBFS with gcj-jdk v4.9.0
https://github.com/ceph/ceph/pull/1854
It passes on our gitbuilders too; I'll get somebody to merge at standup.
Greg Farnum
09:34 AM Bug #8359 (In Progress): FTBFS with gcj-jdk v4.9.0
With attached patch (provided by Gregory) I built Ceph_0.80.1 successfully with gcj-jdk and with openjdk-7.
Thank ...
Dmitry Smirnov
02:59 PM CephFS Bug #8427 (Won't Fix): ceph-fuse: Dumpling "cache still has 0+1 items, waiting (for caps to relea...
A similar (same?) issue was previously tracked at #4451, but that's resolved now:... Greg Farnum
02:19 PM Revision ddb580b1 (ceph): Merge pull request #1851 from FlorentFlament/master
documentation: update osd pool default size from 2 to 3.
Reviewed-By: Christophe Courtaut <christophe.courtaut@gmail...
Loïc Dachary
02:14 PM Bug #7068: os/FileStore.cc: 4035: FAILED assert(omap_attrs.size() == omap_aset.size()) (dumpling)
The bug is that the snap trimmer in dumpling does not take a lock on the object. The fix is probably backporting b87... Samuel Just
09:09 AM Bug #7068: os/FileStore.cc: 4035: FAILED assert(omap_attrs.size() == omap_aset.size()) (dumpling)
ubuntu@teuthology:/a/teuthology-2014-05-20_19:00:34-rados-dumpling-testing-basic-plana/267269 Sage Weil
01:49 PM Revision a6aebe2b (ceph): Set results_server by default
So that queued jobs get pushed to paddles
Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
Zack Cerza
12:14 PM Revision 99b96827 (ceph): documentation: update osd pool default size from 2 to 3.
Fixes: #7725
Signed-off-by: Florent Flament <florent.flament@cloudwatt.com>
Florent Flament
09:28 AM devops Bug #8292 (In Progress): ceph-disk prepare output not explicit on too small disk
This is going to be tricky.
I have a working solution, with nice/informational output, but the calls keep going an...
Alfredo Deza
08:58 AM Revision 4a29c6a1 (ceph): Merge pull request #1815 from ceph/osd-metadata
osd: include osd_objectstore in metadata reported to mon
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
05:53 AM Revision 7b3457d5 (ceph): Merge pull request #1793 from ceph/wip-multimds
Speed up several cross-MDS operations by reducing the number of two-phase commit disk accesses we have to go through.... Gregory Farnum
05:22 AM Documentation #7725: osd pool default size need updating
"Need review":https://github.com/ceph/ceph/pull/1851
Florent Flament
04:41 AM Revision 4d4b77e5 (ceph): cephfs-java: build against older jni headers
Older versions of the JNI interface expected non-const parameters
to their memory move functions. It's unpleasant, bu...
Greg Farnum
02:38 AM Fix #6278: osd: throttle snap trimming
Sorry for noise, one precision please : if I want to throttle to N io per OSD, the correct formula is : 1 * osd_disk_... Olivier Bonvalet
02:17 AM Fix #6278: osd: throttle snap trimming
Ok, I found the commit : https://github.com/ceph/ceph/commit/4e5e41deeaf91c885773d90e6f94da60f6d4efd3
We have to s...
Olivier Bonvalet
02:12 AM Fix #6278: osd: throttle snap trimming
Hi,
I see that a patch for this is present
* in dumpling (0.67.9) :
> osd: allow snap trim throttling with simpl...
Olivier Bonvalet
12:29 AM Revision f81c53a7 (ceph): Merge pull request #1838 from ceph/wip-client-time
fs: include client timestamp in all mds requests
Reviewed-by: Greg Farnum <greg@inktank.com>
Sage Weil
12:09 AM Revision e02a773a (ceph): Merge pull request #36 from ceph/wip-objectstore
fix objecstore test Sage Weil

05/21/2014

10:56 PM CephFS Bug #8405: multimds: FAILED assert(dir->is_frozen_tree_root())
Drat, teuthology hung on this run instead of finishing and gathering up the logs. :( I'll make sure to save them when... Greg Farnum
09:33 PM Revision 19af7635 (ceph): Merge pull request #35 from ceph/wip-add-v0.67.9
Added dumpling v0.67.9 version, removed older versions v0.67.2, v0.67.4,... Warren Usui
09:18 PM Revision 3b382b70 (ceph): Process queued jobs synchronously
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
09:06 PM Revision 50dc3eac (ceph): Added dumpling v0.67.9 version, removed older versions v0.67.2, v0.67.4...
This will reduce number of tests to run, some may failed on old versions, see http://tracker.ceph.com/issues/8409
S...
Yuri Weinstein
08:20 PM Revision 596d4118 (ceph): Merge pull request #1835 from ceph/wip-8332
HashIndex: in cleanup, interpret missing dir as completed merge
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
08:19 PM Revision 5d8442a1 (ceph): Merge pull request #1834 from ceph/wip-8334
Wip 8334
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
08:00 PM Revision 6069ff08 (ceph): doc/release-notes: v0.67.9
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
07:37 PM Bug #7593: Disk saturation during PG folder splitting
Sage Weil wrote:
> At a high level, sure, if you know ahead of time how many objects per PG you expect you can pre-h...
Guang Yang
07:11 PM rbd Bug #8413 (Rejected): test_librbd_fsx: krbd size check races with disk size revalidation
In this case it was in fact the kernel client taking more than 10 seconds to respond to a notify, not a bug. This may... Josh Durgin
05:24 PM Revision 9a376d71 (ceph): Merge pull request #260 from ceph/wip-8422
Add exception hook to teuthology-worker Warren Usui
05:22 PM Revision df0086c0 (ceph): Merge pull request #1848 from yuyuyu101/librbd-flush
Avoid extra check for clean object
Backport: firefly, dumpling
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
05:09 PM Bug #8395 (Pending Backport): ceph-test-objectstore doesn't clean up
fix in master of ceph-qa-suite. to 'backport', we jsut need to advance the firefly branch up to master in ceph-qa-su... Sage Weil
04:57 PM Revision ba340a97 (ceph): 0.67.9
Jenkins
04:34 PM Revision 08ace546 (ceph): Add exception hook to teuthology-worker
Workers processes are dying occasionally, and this should cause the
exceptions to be logged.
Signed-off-by: Zack Cer...
Zack Cerza
03:54 PM Bug #8379: cephx: clock skew causes verify_reply decode error
Greg Farnum wrote:
> So this is just the OSDs (or, possibly, the monitors) being out of sync with their clocks and e...
Sage Weil
03:52 PM Bug #8379: cephx: clock skew causes verify_reply decode error
So this is just the OSDs (or, possibly, the monitors) being out of sync with their clocks and expiring or advancing t... Greg Farnum
08:41 AM Bug #8379: cephx: clock skew causes verify_reply decode error
Thanks, now that we know what causes this I think we should improve the error message (at a minimum). Sage Weil
08:40 AM Bug #8379: cephx: clock skew causes verify_reply decode error
Sage Weil
04:30 AM Bug #8379: cephx: clock skew causes verify_reply decode error
@Sage Weil, I can't close that issue, please help to close, thanks!
BTW, if add the above limitation to CEPH install...
jian li
04:25 AM Bug #8379: cephx: clock skew causes verify_reply decode error
I fix that issue when all the MON and OSDs nodes SYNC the same system date. jian li
03:40 PM Revision 7784ef17 (ceph): Merge pull request #1849 from ceph/wip-mon-get-version
mon: set MMonGetVersionReply tid
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
10:18 AM Feature #8265 (In Progress): config: make int parser accept K, M, G, T, P suffix and scale value ...
Joao Eduardo Luis
10:14 AM Revision f51e33bd (ceph): Avoid extra check for clean object
We needn't to check clean object via buffer state, skip the clean object.
Signed-off-by: Haomai Wang <haomaiwang@gma...
Haomai Wang
09:09 AM Bug #8420 (Can't reproduce): iogen workunit not stopping properly
http://pulpito.front.sepia.ceph.com/ubuntu-2014-05-19_20:36:16-multimds-wip-multimds-testing-basic-plana/264360/
<pr...
Zack Cerza
08:57 AM Bug #8241: XfsFileStoreBackend tries to set extsize but may get EINVAL
I'm also seeing this error. Will try removing my RBD cache tier and see if it stops happening.... Michael Potter
08:42 AM rbd Bug #8416 (Closed): Client Crash when try to map a volume (ubuntu)
Hi,
I'm following the install guide, and when I do the following command in the client:
sudo rbd map foo --poo...
Miguel Barajas
08:20 AM Revision 9235dcb1 (ceph): mon: set MMonGetVersionReply tid
Currently we don't set MMonGetVersionReply tid even if the original
MMonGetVersion message had a non-zero tid. This ...
Ilya Dryomov
06:39 AM devops Bug #8401 (Fix Under Review): ceph-deploy must warn against custom cluster names when system is s...
pull request opened https://github.com/ceph/ceph-deploy/pull/191 Alfredo Deza
05:09 AM Revision 5be30f0d (ceph): rados/singleton-nomsgr/all/objectstore: clean up
ceph_test_objectstore scribbles in $CWD. Keep it inside $TESTDIR and
clean up when we are done.
Fixes: #8395
Signed...
Sage Weil
05:09 AM Revision a4f00aba (ceph): rados/singleton-nomsgr/all/objectstore: skip ceph cluster setup
Not necessary; we just need it installed.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
04:56 AM Revision ba53889a (ceph): README: move package dependencies into separate files
Sorted, with a short, simple command to paste. Avoid the duplication of
the lists.
Signed-off-by: Sage Weil <sage@i...
Sage Weil
04:55 AM Revision 77066a23 (ceph): README: clean up some whitespace
Remove $ prefix, use tab consistently.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
04:53 AM Linux kernel client Bug #8415 (Closed): run_xfstests.sh: install quota package
xfstests 234 failed on a jessie/sid box... Ilya Dryomov
04:51 AM Revision c08adbc9 (ceph): Fix set_alloc_hint op cause KeyValueStore crash problem
Now KeyValueStore doesn't support set_alloc_hit op, the implementation of
_do_transaction need to consider decoding t...
Haomai Wang
04:42 AM rbd Bug #6385: run_xfstests.sh: doesn't work with newly-organized upstream tests
We need to do this transition sooner rather than later. xfstests 103 failed
with the following...
Ilya Dryomov
03:22 AM Revision e5fb75fc (ceph): Merge pull request #1723 from ctheune/patch-1
Fix formatting of header Sage Weil
12:40 AM Revision e9eb30c0 (ceph): Merge pull request #1845 from wdneto/master
doc: Update manual-deployment.rst John Wilkins
12:38 AM Revision 08f72d56 (ceph): Merge pull request #1828 from nereocystis/faq-fix
doc: fix link to FAQ John Wilkins
12:37 AM Revision a147b46a (ceph): Merge pull request #1829 from nereocystis/preflight-cleanup
doc: Clean up pre-flight documentation John Wilkins
12:37 AM Revision 7564df85 (ceph): Merge pull request #1833 from nereocystis/admin-node-doc
doc: mention admin-node in common documentation John Wilkins
12:35 AM Revision aca9b1ed (ceph): Merge pull request #1837 from nereocystis/quick-ceph-deploy-cleanup
doc: quick-ceph-deploy cleanup John Wilkins
12:10 AM Revision ec9a9dac (ceph): Merge pull request #1841 from ceph/wip-8373
Fixes for crush ruleset selection during pool creation (Wip 8373)
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
12:06 AM Revision 4523e581 (ceph): Merge pull request #1836 from ceph/wip-doc-openstack
doc: Clarifications to Ceph Block Device and OpenStack documentation. John Wilkins

05/20/2014

11:29 PM Revision d04b386b (ceph): init-ceph: if we fail to set an OSD's crush position, continue
If we fail to set the CRUSH position for one OSD, continue on to try
starting others, just as we do when we fail to s...
Sage Weil
11:28 PM Revision 46f1eb68 (ceph): init-ceph: continue loop immediately on daemon start failure
If we fail to start the daemon, skip the post_start and other steps.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
11:08 PM Revision 3bbe29eb (ceph): common/Finisher: add queue(list<Context*>&)
Signed-off-by: Samuel Just <sam.just@inktank.com> Samuel Just
10:27 PM Revision f305b8a3 (ceph): Merge pull request #1842 from ceph/wip-buildroot
ceph.spec.in: remove BuildRoot Sage Weil
10:26 PM Revision 7e695318 (ceph): Merge pull request #1844 from onlyjob/hppa
Define AO_REQUIRE_CAS (fixes FTBFS on 'hppa')
Backport: firefly
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
10:22 PM Revision 6ba57da9 (ceph): prioritise use of `javac` executable (gcj provides it through alternati...
On Debian this fixes FTBFS when gcj-jdk and openjdk-7-jdk are installed at
the same time because build system will u...
Dmitry Smirnov
10:22 PM Revision 9e335a49 (ceph): pass '-classpath' option (gcj/javah ignores CLASSPATH environment varia...
This should not affect OpenJDK which understands '-classpath' as well.
With gcj-jdk we still get FTBFS later:
~~~...
Dmitry Smirnov
10:22 PM Revision fcd30500 (ceph): look for "jni.h" in gcj-jdk path, needed to find "jni.h" with gcj-jdk_4...
Signed-off-by: Dmitry Smirnov <onlyjob@member.fsf.org>
(cherry picked from commit 0f4120c0115e7977ae7c03458addcc2b291...
Dmitry Smirnov
10:20 PM Revision c2644b29 (ceph): Update manual-deployment.rst
- When creating the OSD data, specify osd-uuid so that it matches when the osd is first created.
- Modify caps when a...
Aristoteles Neto
10:18 PM Revision 74218f3d (ceph): Define AO_REQUIRE_CAS (fixes FTBFS on 'hppa')
to fix FTBFS due to undeclared atomic functions.
As reported
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=...
Dmitry Smirnov
10:16 PM Revision 1d9e4ac2 (ceph): mon: Fix default replicated pool ruleset choice
Specifically, in the case where the configured
default ruleset is CEPH_DEFAULT_CRUSH_REPLICATED_RULESET,
instead of a...
John Spray
10:16 PM Revision fb504bae (ceph): mon: Fix check of ruleset id on pool update
This code was using CrushWrapper::rule_exists, which
checks for a *rule* existing, whereas the value being
set is a *...
John Spray
10:07 PM Revision 799f76aa (ceph): mds: remove unused Mutation::now
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
10:07 PM Revision 87688570 (ceph): mds: use mds_stamp for mksnap
Use the server timestamp for the snapshot timestamp. This could arguably
be the client timestamp, but I think snapsh...
Sage Weil
10:05 PM Revision 9c86a110 (ceph): mds: use client-provided time stamp for user-visible file metadata
Use the op_stamp from the MDRequest, populated by the MClientRequest when
possible, for setting timestamps on user-vi...
Sage Weil
10:05 PM Revision a09547bb (ceph): mds: use real timestamp for readdir leases
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
10:05 PM Revision d4bfa39f (ceph): mds: reset mds_stamp for readdir, rename, link
These ops to complicated work prior to starting the real operation, like
fetching missing directories, or opening rem...
Sage Weil
10:05 PM Revision 4a6596ec (ceph): mds: do the balancer and propagation timers based on mds_stmap
Use the (more) stable time stamp for doing the balancer's temperature
tracking and associated decay.
Signed-off-by: ...
Sage Weil
09:55 PM Bug #8259 (Fix Under Review): add a target to ensure new dependencies are installed before builds
Sage Weil
09:55 PM Revision fd1f9bda (ceph): mds: do rstat timestamps (rctime, fragstat mtime) in terms of op stamp
Use the op (client) timestamp for the recursive stats, for santity's sake.
Note that since this is monotonically incr...
Sage Weil
09:52 PM Revision 29c68444 (ceph): mds: make sure mds_stamp is set when we journal
This is a catch-all that we are carrying over from before. It may not
be strictly necessary, but I'm not inclined to...
Sage Weil
09:51 PM Revision 3569e809 (ceph): mds: set mds_stamp on lock acquisition
Normally acquire_locks() will set this for us.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
09:50 PM Revision e4c9c341 (ceph): mds: add {mds,op}_stamp to Mutation
With getters, setters.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
08:47 PM Revision de43f6d2 (ceph): ceph-disk: partprobe before settle when preparing dev
Two users have reported this fixes a problem with using --dmcrypt.
Fixes: #6966
Tested-by: Eric Eastman <eric0e@aol....
Sage Weil
08:41 PM Revision 401319a1 (ceph): ceph.spec.in: remove BuildRoot
Deprecated
Fixes: #8143
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
07:28 PM Bug #8404: OSD: stopped doing useful work until scrub timeouts nudged PGs
Zheng Yan
07:27 PM Bug #8404: OSD: stopped doing useful work until scrub timeouts nudged PGs
client is waiting for session close message from mds.8. ... Zheng Yan
02:37 PM Bug #8404: OSD: stopped doing useful work until scrub timeouts nudged PGs
teuthology-2014-05-18_23:02:25-multimds-master-testing-basic-plana/262204/ Greg Farnum
02:28 PM Bug #8404 (Duplicate): OSD: stopped doing useful work until scrub timeouts nudged PGs
http://qa-proxy.ceph.com/teuthology/teuthology-2014-05-18_23:02:25-multimds-master-testing-basic-plana/262217/
I d...
Greg Farnum
07:20 PM rbd Bug #8413 (Rejected): test_librbd_fsx: krbd size check races with disk size revalidation
Running test_librbd_fsx in kernel rbd mode against wip-discard in ceph-client.git, this error was reported:... Josh Durgin
06:50 PM rbd Bug #8409 (Duplicate): "Segmentation fault" in rbd in upgrade:dumpling-dumpling-testing-basic-vps
This is #5426, which was already backported and released in 0.67.6. Since the test starts with an older version of du... Josh Durgin
02:57 PM rbd Bug #8409: "Segmentation fault" in rbd in upgrade:dumpling-dumpling-testing-basic-vps
#7997 Zack Cerza
02:54 PM rbd Bug #8409 (Duplicate): "Segmentation fault" in rbd in upgrade:dumpling-dumpling-testing-basic-vps
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-05-19_19:15:03-upgrade:dumpling-dumpling-testing-basi... Yuri Weinstein
06:32 PM CephFS Bug #8405: multimds: FAILED assert(dir->is_frozen_tree_root())
where is the mds log? Zheng Yan
02:31 PM CephFS Bug #8405 (Duplicate): multimds: FAILED assert(dir->is_frozen_tree_root())
... Greg Farnum
05:52 PM Revision b638d19d (ceph): msg: Fix inconsistent message sequence negotiation during connection reset
Backport: firefly, emperor, dumpling
Signed-off-by: Guang Yang (yguang@yahoo-inc.com)
Reviewed-by: Greg Farnum <greg...
Guang Yang
05:52 PM Revision 804fb969 (ceph): msg: Fix inconsistent message sequence negotiation during connection reset
Backport: firefly, emperor, dumpling
Signed-off-by: Guang Yang (yguang@yahoo-inc.com)
Reviewed-by: Greg Farnum <greg...
Guang Yang
05:46 PM Revision 55b11128 (ceph): OSD::handle_pg_query: on dne pg, send lb=hobject_t() if deleting
We will set lb=hobject_t() if we resurrect the pg. In that case,
we need to have sent that to the primary before han...
Sage Weil
05:35 PM Revision 3b8ab41e (ceph): mon/MonClient: remove stray _finish_hunting() calls
Callig _finish_hunting() clears out the bool hunting flag, which means we
don't retry by connection to another mon pe...
Sage Weil
05:19 PM Revision a9153504 (ceph): Merge pull request #1826 from ceph/wip-8162-dumpling
Wip 8162 dumpling
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
05:09 PM Bug #8373 (Pending Backport): Pool creation defaults to ruleset 0 even if it doesn't exist
Sage Weil
01:21 PM Bug #8373 (Fix Under Review): Pool creation defaults to ruleset 0 even if it doesn't exist
Sage Weil
09:34 AM Bug #8373 (In Progress): Pool creation defaults to ruleset 0 even if it doesn't exist
John Spray
04:29 PM Bug #8342 (Fix Under Review): init script may not start all OSDs
https://github.com/ceph/ceph/pull/1846
Can you take a look at this patch? I think I understand now what you were ...
Sage Weil
04:23 PM Bug #8359: FTBFS with gcj-jdk v4.9.0
There was a switch from expecting non-const to expecting const in that function:
http://docs.oracle.com/javase/1.5.0...
Greg Farnum
03:34 PM Bug #8359: FTBFS with gcj-jdk v4.9.0
No it wasn't resolved, at least not in Firefly.
Merged patches fixed other problems so build can progress until it s...
Dmitry Smirnov
03:20 PM Bug #8359 (Resolved): FTBFS with gcj-jdk v4.9.0
merged these to masters a few days ago, now in firefly branch too. Sage Weil
04:01 PM Fix #8205: FileStore: properly fill in XATTR_NO_SPILL_OUT tag
Significantly revised the branch (it now removes the user_only stuff in the ObjectStore interface), and have schedule... Greg Farnum
03:28 PM rgw Bug #8411 (Can't reproduce): "Segmentation fault" in radosgw in upgrade:dumpling-dumpling-testing...
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-05-19_19:15:03-upgrade:dumpling-dumpling-testing-basi... Yuri Weinstein
03:26 PM Bug #8143 (Resolved): BuildRoot is now silently ignored in .spec files
Sage Weil
01:40 PM Bug #8143 (Fix Under Review): BuildRoot is now silently ignored in .spec files
Sage Weil
02:28 PM Feature #8231: ceph filestore dump improvements
Adding:
ceph-filestore-dump ... <object> (get-omaphdr)
ceph-filestore-dump ... <object> (set-omaphdr) <value>
David Zafman
02:06 PM devops Bug #8326 (Rejected): Issues with ceph deploy - disk issue
it sounds like the vm disk is too small. the default journal 5 or 10 gb ... so the volume needs to be at least 20 pr... Sage Weil
02:02 PM devops Bug #5306 (Can't reproduce): Xen based OSDs fail to start ceph-osd process
Sage Weil
01:58 PM Revision 7418d062 (ceph): Merge pull request #34 from ceph/wip-fuse
fuse_defautl_permissions = 0;
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
01:50 PM Bug #8256: unhelpful ceph cli command --help
Yeah, I debated about "parametername" vs "type" being more helpful. Of course, always, everyone wants both, and then... Dan Mick
01:32 PM Bug #8256: unhelpful ceph cli command --help
Sage Weil
01:50 PM Revision e207e903 (ceph): multimds: fuse_default_permissions = 0
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
01:48 PM Revision bf99313d (ceph): multimds: factor install+ceph tasks out of other collections
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
01:46 PM Bug #6806 (Resolved): mon: audit cmd_getval() calls to make sure they handle failures correctly
no backport, too big. Sage Weil
01:46 PM devops Bug #8403 (Resolved): Fedora 20 needs to be supported with repos
There are a few tickets (like #7627) that are now in need of Fedora 20 repos to start testing and fixing some issues.... Alfredo Deza
01:45 PM Bug #7210 (Resolved): mon: does not validate snapshot removal commands
no backport. Sage Weil
01:44 PM Bug #7371 (Resolved): libceph: tid 36856 reply has 196608 bytes we had only 147456 bytes ready
not worried about backporting this now that firefly is out. Sage Weil
01:43 PM Revision 93a5b88f (ceph): tools/cephfs: error handling in EventOutput
Print error instead of hitting assertion if
we e.g. can't create directory.
Signed-off-by: John Spray <john.spray@in...
John Spray
01:43 PM Revision 3bef125f (ceph): qa: Add smoke test for cephfs-journal-tool
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
01:43 PM Revision 708c4eda (ceph): tools/JournalTool: update usage()
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
01:43 PM Revision d66fa53d (ceph): tools/MDSUtility: fix crash on bad config
Was calling messenger destructor before
calling shutdown in case where config
was bad, e.g. in vstart without "-c" fl...
John Spray
01:43 PM Revision 2621b5d9 (ceph): tools/cephfs-journal-tool: handle --help cleanly
Handle --help before main initialization so that
it works with or without a running mon cluster.
Signed-off-by: John...
John Spray
01:43 PM Revision 3207c505 (ceph): osdc/Journaler: fix obj count in Journaler:erase
This was off by one (too few) in the case of a
trimmedpos->write_pos range that had length
layout_period+2 and starti...
John Spray
01:41 PM Bug #7159 (Resolved): ceph status --format=json num_in_osds and num_up_osds formatting not consis...
in firefly, no need for dumpling. Sage Weil
01:38 PM Bug #8214 (Can't reproduce): Crash in Thread.cc "common/Thread.cc: 110: FAILED assert(ret == 0)" ...
Sage Weil
01:36 PM Bug #8250 (Resolved): osd crashed "pthread lock: Invalid argument" in upgrade:dumpling-x:stress-s...
Sage Weil
01:31 PM Bug #8294 (Rejected): "Error EINVAL" in upgrade:dumpling-dumpling-testing-basic-vps
wrong qa-suite branch Sage Weil
01:30 PM rbd Bug #8318: "rbd: create error" in upgrade:dumpling-dumpling-testing-basic-plana suite
this a bug in the test? ENOSYS, probably suing rbd_create3 or something? Sage Weil
01:28 PM Bug #8320: heartbeat timeouts too low for vps machines
Make this increase teh timeouts when running on vps. Sage Weil
01:24 PM Bug #8335 (Rejected): Crash while recovering from XFS corruption
Recreate this OSD (after you confirm the cluster has other copies of the data :) Sage Weil
01:23 PM Bug #8367 (Duplicate): osd crashed in upgrade:dumpling-x:stress-split-firefly---basic-plana
Samuel Just
01:22 PM Bug #7991 (Rejected): ceph-mon crash
Sage Weil
01:21 PM Bug #7991 (Can't reproduce): ceph-mon crash
Sage Weil
01:16 PM Bug #6565 (Resolved): stuck in recovery_wait
Sage Weil
01:15 PM Bug #8114 (Can't reproduce): "osd/RadosModel.h: 1055: FAILED assert" in upgrade:dumpling-x:stress...
Sage Weil
01:14 PM Bug #8229 (Can't reproduce): 0.80~rc1: OSD crash (domino effect)
Sage Weil
01:12 PM Bug #8162 (Resolved): osd: dumpling advances last_backfill prematurely
Sage Weil
01:12 PM Bug #8011 (Resolved): osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || soid >= s...
Samuel Just
01:12 PM Bug #8270 (Can't reproduce): 0.80~rc1: OSD crash during replication after repair
Sage Weil
01:11 PM Bug #8319 (Resolved): osd: spurious 'wrongly marked me down' message on shutdown
Sage Weil
01:11 PM Bug #8328 (Resolved): osd: null op on dup_ops list
Sage Weil
01:11 PM Bug #8305 (Resolved): objecter, osd: pool overlay change should trigger op resend
Sage Weil
01:11 PM Bug #7588 (Resolved): OSD Seg fault in string assign ObjectOperation::C_ObjectOperation_copyget::...
Sage Weil
01:09 PM Bug #8332 (Pending Backport): ceph-test-objectstore: bad return value in unlink
Samuel Just
01:09 PM Bug #8334 (Pending Backport): osd: bug with pool snaps, ceph_test_rados
Samuel Just
01:08 PM Bug #8380 (Resolved): osd/ReplicatedPG.cc: 1849: FAILED assert(0 == "out of order op")
Sage Weil
01:07 PM Revision 1c2bb374 (ceph): osdc: new style encoding for Journal::Header
Switch to the ENCODE_START, DECODE_START macros
for Journal::Header. As well as being nice generally,
this has the i...
John Spray
01:07 PM Revision bac434de (ceph): mds: Make EMetaBlob::dirlump::_decode_bits const
This requires making the late-decoded attributes mutable.
The motivation is to allow get_paths and all the code downs...
John Spray
01:07 PM Revision 5d913e2b (ceph): mds: Create JournalPointer in MDLog::create
Otherwise it is created at next startup when
we see that it is missing: this is cleaner.
Signed-off-by: John Spray <...
John Spray
01:07 PM Revision b76e64b2 (ceph): tools/JournalTool: Fix 'header set' corner case
If the write_pos was behind the other pointers, we
would flag the header as invalid. However, that
would prevent doi...
John Spray
01:07 PM Revision 68b20094 (ceph): mds: Introduce mds_journal_format config setting
This is useful for testing, so that we can create an
old-style journal and then test the version migration
by changin...
John Spray
01:07 PM Revision 5c2a6401 (ceph): osdc/Journaler: include stream_format in dump()
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
01:07 PM Revision d52eeaba (ceph): Fix JournalStream::read length assertion
This was checking for new-format length even
when reading from an old-format journal.
Signed-off-by: John Spray <joh...
John Spray
01:07 PM Revision fc1d6da7 (ceph): tools/cephfs: Update for JournalPointer
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
01:07 PM Revision 423dafe0 (ceph): osdc/Journaler: use pointers for out params
Fixes a style violation in the new JournalStream class.
Signed-off-by: John Spray <john.spray@inktank.com>
John Spray
01:07 PM Revision 45c319ef (ceph): doc: Update ceph-mds manpage
Add missing --hot-standby and --journal-check
arguments.
Fixes: #8258
John Spray
01:07 PM Revision 778b7851 (ceph): osdc/JournalStream: DRY envelope size calc
Replace repetitive sizeof() calculation for the
envelope overhead per log event with #defines.
Signed-off-by: John S...
John Spray
01:07 PM Revision f12b9e8b (ceph): osdc: make JournalStream::readable const
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
01:07 PM Revision f609a0af (ceph): osdc: Simplify JournalStream::read
It was doing an unnecessary series of splices() after
reading header, replace with a single pass through
the data and...
John Spray
01:07 PM Revision 79a1e587 (ceph): mds/CDir: remove redundant inode enc/dec
Use the new CInode::encode_bare/decode_bare fns
in CDir, so that we only have one implementation
of the inode encode/...
John Spray
01:07 PM Revision e7fa5be6 (ceph): osdc: Clean up journalstream readable check
Fix redundant (and subtly incorrect) calculation of
the number of bytes needed. It worked because waiting
for a few ...
John Spray
01:07 PM Revision 405cb857 (ceph): osdc: fix redundant branch
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
01:07 PM Revision c73902b1 (ceph): osdc/Journaler: privatize members
Many members were unnecessarily public.
Signed-off-by: John Spray <john.spray@inktank.com>
John Spray
01:07 PM Revision dba721f1 (ceph): osdc: Add Journaler.erase
This is used in subsequent commits to delete journals
that are no longer needed, such as after rewriting
a journal in...
John Spray
01:07 PM Revision a4ae168a (ceph): mds: add atomic log rewrite on format change
Two main pieces to this:
* A new JournalPointer object that stores two journal
inodes so that we can do a double-...
John Spray
01:07 PM Revision 8b0383ed (ceph): dencoder: add JournalPointer to types.h
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
01:07 PM Revision 1612677b (ceph): mds: Generalize JournalPointer functionality
...so that we can use it places like Resetter too. And use it
there.
Signed-off-by: John Spray <john.spray@inktank....
John Spray
01:07 PM Revision aec4c939 (ceph): tools: update Dumper to use JournalPointer
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
01:07 PM Revision f7e9ff17 (ceph): tools: Create cephfs-journal-tool
This is for debugging/repairing CephFS journals.
Signed-off-by: John Spray <john.spray@inktank.com>
John Spray
01:07 PM Revision 403e0288 (ceph): rpm: add cephfs-journal-tool to RPM build
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
01:07 PM Revision 7e4da00a (ceph): debian: add cephfs-journal-tool to ceph-mds
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
01:07 PM Revision 5011945f (ceph): gitignore: Add cephfs-journal-tool
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
01:07 PM Revision 49add983 (ceph): docs: Add cephfs-journal-tool
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
01:05 PM Bug #8393 (Duplicate): osd crashed in rbd-master-testing-basic-plana suite
dup of #8380 Sage Weil
01:03 PM Bug #8395: ceph-test-objectstore doesn't clean up
We need to kill the test working dir, test bug. Samuel Just
01:02 PM Bug #7986: 3.1s0 scrub stat mismatch, got 2041/2044 objects, 0/0 clones, 2041/2044 dirty, 0/0
ubuntu@teuthology:/a/teuthology-2014-05-18_02:30:21-rados-master-testing-basic-plana/260674 Sage Weil
01:02 PM Bug #8398 (Duplicate): scrub stat mismatch errors in nightlies
#7986 Sage Weil
01:01 PM Bug #8397 (Duplicate): radosbench test reported error in cluster log
Samuel Just
11:59 AM devops Bug #8401 (Resolved): ceph-deploy must warn against custom cluster names when system is sysvinit
sysvinit systems do not support custom cluster names, so ceph-deploy should refuse deployment when that is the case
...
Alfredo Deza
11:55 AM Feature #8391: sysvinit does not support custom cluster names
This is just an issue with sysvinit, so nothing in that realm will support custom cluster names. Alfredo Deza
10:04 AM Feature #8391: sysvinit does not support custom cluster names
I believe there is no cluster name support at all in the sysvinit script. We only added it to the upstart stuff. I ... Sage Weil
11:38 AM rgw Bug #8311: No pool name error in ubuntu-2014-05-06_21:02:54-upgrade:dumpling-dumpling-testing-bas...
Here is the latest run results:
http://pulpito.front.sepia.ceph.com/teuthology-2014-05-19_19:15:03-upgrade:dumpling-...
Yuri Weinstein
10:52 AM Bug #8232 (Resolved): Race condition during messenger rebind
This is in firefly as commit:6078726e0b89d0202c3a4bc594be11fc346bed9d, emperor as commit:804fb9694f34e1685c787c851e06... Greg Farnum
09:07 AM rgw Bug #8269 (Pending Backport): rgw: corrupted multipart object
Josh Durgin
08:38 AM Revision 107821fe (ceph): Fix formatting of header
The header was displaying the newline as a simple "n" due to quoting in the doc.
Signed-off-by: Christian Theune <ct...
Christian Theune
06:24 AM devops Bug #7627: ceph-disk: does not start daemons properly under systemd
...and we don't have packages for F20
Going to try and build for F20, but it might take some effort to get there a...
Alfredo Deza
05:58 AM Bug #8379: cephx: clock skew causes verify_reply decode error
The settings of ceph.conf is
[global]
fsid = 49a9d6de-0583-4941-b6ee-7007482ddbcf
mon_initial_members = estack4-HP...
jian li
03:58 AM Revision fe1fbfb7 (ceph): test: fix some templates to match new output code
Signed-off-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Samuel Just <sam.just@inktank.com>
(cherry picked from com...
Greg Farnum
01:43 AM Bug #8387: osd: skipping missing maps broken
Hi Sage
Here is the complete output from another OSD. This problem is occurring with multiple OSDS (74 osd) across...
karan singh

05/19/2014

11:56 PM Revision bd5d6f11 (ceph): OSD: fix an osdmap_subscribe interface misuse
When calling osdmap_subscribe, you have to pass an epoch newer than the
current map's. _maybe_boot() was not doing th...
Greg Farnum
11:53 PM Revision 1383b649 (ceph): Merge pull request #1839 from ceph/wip-8338
OSD: verify that client ops are targeted correctly in the current epoch
Reviewed-by: Samuel Just <sam.just@inktank.com>
Gregory Farnum
11:40 PM Bug #8381 (Fix Under Review): osd crash when osd use leveldb as filestore
Haomai Wang
11:25 PM Bug #8381: osd crash when osd use leveldb as filestore
Thanks to xinxin!
The bug is resulted from set_alloc_hint op.
case Transaction::OP_SETALLOCHINT:
//...
Haomai Wang
07:20 PM Bug #8381: osd crash when osd use leveldb as filestore
log with 'debug keyvaluestore = 20/20' Xinxin Shu
12:49 AM Bug #8381: osd crash when osd use leveldb as filestore
you can add "debug_keyvaluestore = 20/20" to ceph.conf Haomai Wang
12:39 AM Bug #8381: osd crash when osd use leveldb as filestore

hi haomai , this issue occurs once i used 'dd', in order to help you root the cause, what kind of info should i pro...
Xinxin Shu
11:39 PM Revision 5c8afaa8 (ceph): OSD: fix an osdmap_subscribe interface misuse
When calling osdmap_subscribe, you have to pass an epoch newer than the
current map's. _maybe_boot() was not doing th...
Greg Farnum
11:33 PM Revision d58d2c15 (ceph): Merge pull request #258 from ceph/wip-fix-format-samba
Fixed formatting and added a better variable for time out Zack Cerza
11:31 PM Revision eb98f4ab (ceph): osd: skip out of order op checks on tiered pools
When we send redirected ops, we do not assign a new tid, which means that
a given client's ops for a pool may not hav...
Sage Weil
11:23 PM Revision ad2e20f2 (ceph): client: set timestamp in MDS requests
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
11:23 PM Revision a91e072b (ceph): mds: include timestamp in MClientRequest
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
11:18 PM Revision c8881bb7 (ceph): Fixed broken quotations
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
11:05 PM Revision 2f023b5d (ceph): Merge pull request #1825 from ceph/wip-8380
osd: skip out of order op checks on tiered pools
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
11:03 PM Revision 6e893ad2 (ceph): Fixed formatting and added a better variable for time out
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
10:31 PM Revision de3564bc (ceph): Merge pull request #1830 from salehe/docfix-1
rados.py: clarify documentation of recent change on write return value
Reviewed-by: Josh Durgin <josh.durgin@inktank...
Josh Durgin
10:25 PM Revision d71839a6 (ceph): doc: clarify openstack cow sentence
Missing subject, and now applies to disks in general.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
10:24 PM Revision 0f7f1e84 (ceph): doc: note cinder multi-backend restriction
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> Josh Durgin
10:14 PM Revision e92f2d9f (ceph): doc: link to ephemeral rbd patches
These are necessary for using ephemeral rbd with Havana and Icehouse.
Signed-off-by: Josh Durgin <josh.durgin@inktan...
Josh Durgin
10:04 PM Revision 216e5c9a (ceph): task/: add populate_rbd_pool for quickly spinning up a pool with rbd stuff
Signed-off-by: Samuel Just <sam.just@inktank.com> Samuel Just
10:03 PM Revision 13d6c3cf (ceph): doc: quick-ceph-deploy cleanup
Improve documentation in quick-ceph-deploy.rst
Use admin-node consistently.
ceph should be installed on admin-node fo...
Kevin Dalley
09:41 PM Revision 50f6641a (ceph): Merge pull request #1804 from ceph/wip-8104
Wip 8104
Reviewed-by: Sage Weil <sage@inktank.com>
Samuel Just
09:36 PM Revision d40ba053 (ceph): doc: mention admin-node in common documentation
This change uses admin-node in the quick-common.rst file, including
the graph. This name is already used in
quick-cep...
Kevin Dalley
09:27 PM Revision 29f615b7 (ceph): ReplicatedPG::start_flush: fix clone deletion case
dsnapc.snaps will be non-empty most of the time if there
have been snaps before prev_snapc. What we really want to
k...
Samuel Just
09:24 PM Revision 5ff95dbd (ceph): HashIndex: in cleanup, interpret missing dir as completed merge
If we stop between unlinking the empty subdir and removing the root
merge marker, we get ENOENT on the get_info. Tha...
Samuel Just
09:18 PM Revision bc897b84 (ceph): rados.py: clarify recent change on write return value
Signed-off-by: Mohammad Salehe <salehe+dev@gmail.com> Mohammad Salehe
09:10 PM Revision 8d9d724b (ceph): ceph_manager: reset timeout if we are making progress in wait_for_recovery
Signed-off-by: Samuel Just <sam.just@inktank.com> Samuel Just
08:55 PM Revision 29ee6fae (ceph): ReplicatedPG: block scrub on blocked object contexts
Fixes: #8011
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 741147715...
Samuel Just
08:54 PM Revision 6078726e (ceph): msg: Fix inconsistent message sequence negotiation during connection reset
Backport: firefly, emperor, dumpling
Signed-off-by: Guang Yang (yguang@yahoo-inc.com)
Reviewed-by: Greg Farnum <greg...
Guang Yang
08:54 PM Revision 02557f8b (ceph): Objecter::_op_submit: only replace the tid if it's 0
Otherwise, redirected ops will suddenly have a different tid
and will become uncancelable.
Fixes: #7588
Signed-off-b...
Samuel Just
08:54 PM Revision 788b5fad (ceph): mon/OSDMonitor: force op resend when pool overlay changes
If a client is sending a sequence of ops (say, a, b, c, d) and partway
through that sequence it receives an OSDMap up...
Sage Weil
08:54 PM Revision 268df486 (ceph): osd: discard client ops sent before last_force_op_resend
If an op is sent before last_force_op_resend, and the client's feature is
present, drop the op because we know they w...
Sage Weil
08:53 PM Revision 8bc4992f (ceph): osdc/Objecter: resend ops in the last_force_op_resend epoch
If we are a client, and process a map that sets last_force_op_resend to
the current epoch, force a resend of this op....
Sage Weil
08:53 PM Revision 83d03f0e (ceph): osd/osd_types: add last_force_op_resend to pg_pool_t
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked from commit 3152faf79f498a723ae0fe44301ccb21b15a96ab)
Sage Weil
08:53 PM Revision c94ac44a (ceph): osd: handle race between osdmap and prepare_to_stop
If we get a MOSDMarkMeDown message and set service.state == STOPPING, we
kick the prepare_to_stop() thread. Normally...
Sage Weil
08:53 PM Revision 22e2e7cf (ceph): osd/ReplicatedPG: do not queue NULL dup_op
We call start_flush() with a NULL op in a couple different places. Do not
put a NULL pointer on the dup_ops list or ...
Sage Weil
08:45 PM Revision c5ae8ffc (ceph): Merge pull request #1827 from ceph/wip-6565-dumpling
Wip 6565 dumpling
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
08:38 PM Revision 63721185 (ceph): doc: Clean up pre-flight documentation
Mention recent Ceph releases.
Move important message about sudo and ceph-deploy closer to the use of
ceph-deploy.
Men...
Kevin Dalley
07:02 PM Bug #8398 (Duplicate): scrub stat mismatch errors in nightlies
logs: ubuntu@teuthology:/a/teuthology-2014-05-18_02:30:21-rados-master-testing-basic-plana/260674
ceph version: v0...
Tamilarasi muthamizhan
06:59 PM Revision 06d05fd6 (ceph): doc: fix link to FAQ
The location of the ceph wiki FAQ has changed.
Now, the link from ceph documentation matches the current FAQ location...
Kevin Dalley
06:54 PM Bug #8397 (Duplicate): radosbench test reported error in cluster log
logs: ubuntu@teuthology:/a/teuthology-2014-05-18_02:30:21-rados-master-testing-basic-plana/260589
ceph version: v0...
Tamilarasi muthamizhan
06:30 PM Bug #7891: osd: leaked pg refs on shutdown
recent logs from the nightlies:ubuntu@teuthology:/a/teuthology-2014-05-18_02:30:21-rados-master-testing-basic-plana/2... Tamilarasi muthamizhan
02:17 PM Bug #7891: osd: leaked pg refs on shutdown
two instances:
ubuntu@teuthology:/a/samuelj-2014-05-16_23:11:10-rados-wip-sam-testing-testing-basic-plana/259903
Samuel Just
06:24 PM Bug #8396 (Resolved): osd: message delayed in Session misdirected after split
logs:ubuntu@teuthology:/a/teuthology-2014-05-18_02:30:21-rados-master-testing-basic-plana/260734
ceph version 0.80...
Tamilarasi muthamizhan
06:24 PM Revision f3925b20 (ceph): Merge pull request #1820 from ceph/wip-osd-refs
osd: fix a session ref leak, and an unlikely race
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
06:23 PM Revision 8d991fa2 (ceph): OSD: check for splitting when processing recover/backfill reservations
Fixes: 6565
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
(cherry picke...
Samuel Just
06:18 PM Revision 03e3ccdd (ceph): doc: Restored glance_api_version=2 setting.
Fixes: #8347
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
06:18 PM rbd Bug #8378: krbd: Kernel oops in rbd_img_obj_callback
Hi Ilya,
I would like to know some details abount this patch, is this patch will applied to RBD Kernel Module or O...
Rain Li
10:20 AM rbd Bug #8378: krbd: Kernel oops in rbd_img_obj_callback
Hi Rain,
We have a patch pending that I hope will fix this, but it needs more testing.
I'll notify you when it's ...
Ilya Dryomov
05:48 PM Bug #8395 (Resolved): ceph-test-objectstore doesn't clean up
logs: ubuntu@teuthology:/a/teuthology-2014-05-18_02:30:21-rados-master-testing-basic-plana/260352
ceph version 0.8...
Tamilarasi muthamizhan
05:46 PM Revision e8756be2 (ceph): osdc/Objecter: flag ops that have been redirected
In the future it may be helpful to know whether an op has been redirected
by the OSD.
We don't need it bad enough no...
Sage Weil
05:32 PM Revision cf2b172c (ceph): osd: skip out of order op checks on tiered pools
When we send redirected ops, we do not assign a new tid, which means that
a given client's ops for a pool may not hav...
Sage Weil
05:10 PM Bug #8393: osd crashed in rbd-master-testing-basic-plana suite
coredump from ceph-osd.4.log @ ubuntu@teuthology:/a/teuthology-2014-05-18_23:00:04-rbd-master-testing-basic-plana/261... Yuri Weinstein
04:13 PM Bug #8393: osd crashed in rbd-master-testing-basic-plana suite
I was able to reproduce this issue on manual re-run.
Logs can be found on 'yw' box (ssh from teuthology) in /home/ub...
Yuri Weinstein
03:34 PM Bug #8393: osd crashed in rbd-master-testing-basic-plana suite
Moved to ceph Yuri Weinstein
02:54 PM Bug #8393 (Duplicate): osd crashed in rbd-master-testing-basic-plana suite
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-05-18_23:00:04-rbd-master-testing-basic-plana/261772/... Yuri Weinstein
05:05 PM RADOS Fix #8394 (New): Monitor: allow specifying "begin" and "target" values on map subscriptions
Right now, the monitor subscription mechanism lets you specify the map you want to start receiving updates as of. Unf... Greg Farnum
04:58 PM Bug #8338 (Resolved): OSD: no longer checking that ops on older maps are correctly targeted
Merged to master by commit 1383b649d7ae97c99e9840c42bef0c0db5a0f65e as commit 9f0825ca13320187ee9d763160ea2f49738f83f2 Greg Farnum
04:57 PM Bug #8371 (Resolved): osd not booting
This is in the dumpling branch as bd5d6f116416d1b410d57ce00cb3e2abf6de102b, and Sage has it in firefly-next as 5c8afa... Greg Farnum
04:49 PM Revision 9e80c29c (ceph): Merge remote-tracking branch 'gh/next'
Sage Weil
04:41 PM Revision e47049ba (ceph): erasure-code: fix URL in developer notes
add missing ..
http://tracker.ceph.com/issues/8388 Fixes: #8388
Reported-by: koleosfuscus <koleosfuscus@yahoo.com>
...
Loïc Dachary
04:29 PM Bug #8380 (Pending Backport): osd/ReplicatedPG.cc: 1849: FAILED assert(0 == "out of order op")
Sage Weil
04:28 PM Bug #8380 (Resolved): osd/ReplicatedPG.cc: 1849: FAILED assert(0 == "out of order op")
Sage Weil
01:26 PM Bug #8380 (Fix Under Review): osd/ReplicatedPG.cc: 1849: FAILED assert(0 == "out of order op")
Sage Weil
10:01 AM Bug #8380: osd/ReplicatedPG.cc: 1849: FAILED assert(0 == "out of order op")
ah, this is a problem with the out of order debug checks vs the client behavior of keeping the same tid when doing re... Sage Weil
09:08 AM Bug #8380: osd/ReplicatedPG.cc: 1849: FAILED assert(0 == "out of order op")
ubuntu@teuthology:/a/teuthology-2014-05-18_23:00:04-rbd-master-testing-basic-plana/261772 Sage Weil
04:12 PM Revision debf84ed (ceph): Split progress indicator out to separate functions
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
04:12 PM Revision 33dbffff (ceph): Avoid calling yaml.safe_load() twice on each job
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
04:03 PM Bug #8239 (Resolved): log [WRN] : slow request 30.404834 seconds old, received at 2014-04-26 04:0...
Samuel Just
04:02 PM Bug #8104 (Resolved): OSD: changing min_size to larger than the acting set does not make the PG g...
Samuel Just
03:11 PM Revision a2421846 (ceph): increased sleep to 100 as 90 seemed insufficient
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
02:27 PM CephFS Feature #4354 (Resolved): mds: add an equivalent to the OSD OpTracker
5ace4d1af712b6852b27b5046b7df2a66c0f4e51 Greg Farnum
12:58 PM Revision d3488202 (ceph): Merge pull request #1823 from dachary/wip-mailmap
mailmap: Sahid Orentino Ferdjaoui affiliation
Reviewed-by: Sahid Orentino Ferdjaoui <sahid.ferdjaoui@cloudwatt.com>
Loïc Dachary
12:49 PM Revision 23787ba0 (ceph): mailmap: Sahid Orentino Ferdjaoui affiliation
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
12:40 PM Revision 9bdc3f2a (ceph): Merge pull request #1822 from sahid/master
Ensure autogen.sh to be executed at the top-level
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
12:33 PM Revision 30ae96a7 (ceph): Ensure autogen.sh to be executed at the top-level
The commit adds a test to ensure that user executes
'autogen.sh' at the top-level of the directory.
Signed-off-by: S...
Sahid Orentino Ferdjaoui
12:00 PM Feature #8391 (Resolved): sysvinit does not support custom cluster names
It looks like custom cluster names are not being handled correctly by the init script.
This is reproducible with...
Alfredo Deza
11:31 AM rbd Documentation #8347 (Resolved): Openstack / Cinder Config
Restored glance_api_version=2 John Wilkins
11:13 AM Bug #6565 (Pending Backport): stuck in recovery_wait
saw in dumpling, not sure if we should bother backporting it Samuel Just
09:50 AM Bug #8379 (Need More Info): cephx: clock skew causes verify_reply decode error
Sage Weil
09:25 AM Bug #8379: cephx: clock skew causes verify_reply decode error
can you reproduce this with debug auth = 20 and debug monc = 20? Sage Weil
09:36 AM Bug #8387 (Need More Info): osd: skipping missing maps broken
Ian Colle
09:19 AM Bug #8387: osd: skipping missing maps broken
Can you find one line like
2014-05-19 17:09:52.374646 7fcbb109e700 0 -- 192.168.1.110:6819/58592 >> 192.168.1.107...
Sage Weil
07:31 AM Bug #8387: osd: skipping missing maps broken
More logs :: http://pastebin.com/zx6W4TY9 karan singh
07:21 AM Bug #8387 (Resolved): osd: skipping missing maps broken
Problem :: 50+ OSD are getting marked out of cluster and are down. The cluster is degraded. On checking logs of faile... karan singh
09:35 AM Bug #8382 (Won't Fix): Existed object in EC pool will not be overwritten or deleted by new writes
it will get cleaned up later, although right now garbage colletion is not very aggressive. see ticket #8389 Sage Weil
09:34 AM Feature #8389 (Resolved): osd: clean up old ec objects more aggressively
Sage Weil
08:17 AM Documentation #8388 (In Progress): Broken link in developer_notes for erasure_coding
"work in progress":https://github.com/ceph/ceph/pull/1824 Loïc Dachary
08:04 AM Documentation #8388: Broken link in developer_notes for erasure_coding
Probably the page was moved to:http://ceph.com/docs/master/dev/erasure-coded-pool/ koleos fuscus
07:31 AM Documentation #8388 (Resolved): Broken link in developer_notes for erasure_coding

The page describing "erasure code developer notes" has a broken link:
http://ceph.com/docs/master/dev/osd_internal...
koleos fuscus
06:55 AM CephFS Feature #8386 (Resolved): mds: use client (instead of mds) ctime/mtime
Sage Weil
06:18 AM rbd Bug #8385 (Closed): RBD / QEMU Crash: Invalid fastbin entry (free)
After the fix for Issue #6480, RBD and Qemu have been much more stable on our workload. With 0.67.8 we just saw anoth... Mike Dawson
05:21 AM devops Bug #8384 (Resolved): broken tgt makes teuthology tests fail
... Alfredo Deza
12:50 AM rgw Bug #8383 (Rejected): Upload part of one object passed with incorrect upload id or incorrect obje...
Description?
Test the case which upload parts of object orderly with radios gateway S3 API. When upload part of obje...
Jingjing Zhao

05/18/2014

10:31 PM Bug #8381: osd crash when osd use leveldb as filestore
I can't reproduce the crash. It seemed ok running "dd" or "fio" in vm.
The bad op "2307" is confusing. As I know, ...
Haomai Wang
08:05 PM Bug #8381 (Resolved): osd crash when osd use leveldb as filestore
when osd use leveldb as filestore(add configur options osd_objectstore = keyvaluestore-dev in ceph.conf), i used qem... Xinxin Shu
09:19 PM Bug #8382 (Won't Fix): Existed object in EC pool will not be overwritten or deleted by new writes
As listed below, object named "testobj9" is already existed in this EC pool, but every new writes to this object afte... Zhi Zhang
10:21 AM Revision e56b88fe (ceph): mds: Add ENoOp for padding journals
This allows us to implement journal splicing
without moving blocks around, and without modifying
the outer journal sy...
John Spray
10:21 AM Revision bf7b84c9 (ceph): mds: Add get_metablob to LogEvent
Previously the only way to get at the payload
of things like EUpdate and EOpen was to replay() them
(required a full ...
John Spray
10:21 AM Revision 2282739b (ceph): mds: Refactor CINode encoding into CInodeStore
CInode itself combined the on-disk format and
encode/decode logic with lots of other complex
behaviours. This separa...
John Spray
10:21 AM Revision b3e57ba5 (ceph): mds: Publicize some EMetaBlob members
This makes them accessible for JournalTool.
Signed-off-by: John Spray <john.spray@inktank.com>
John Spray
10:21 AM Revision 96720b6d (ceph): mds: Add EMetaBlob::get_inodes
Used for filtering in journal tool
Signed-off-by: John Spray <john.spray@inktank.com>
John Spray
10:21 AM Revision e7f4b9e2 (ceph): mds: Fix Dumper::undump (missing lock)
Two problems were causing undump to fail:
* Objecter lock was not being taken around call to
.write() and .write_fu...
John Spray
10:21 AM Revision c1b185bd (ceph): mds: Add LogEvent::str_to_type
For cephfs-journal-tool to resolve user input to a EVENT_* constant
Signed-off-by: John Spray <john.spray@inktank.com>
John Spray
10:21 AM Revision f4927f00 (ceph): mds: switch __u32 with LogEvent::EventType typedef
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
10:21 AM Revision 220f9c96 (ceph): mds: Add EMetaBlob::get_dentries
For tools that would like to know which dentries are
touched by a metablob, without understanding its
internal format...
John Spray
10:21 AM Revision 730a0115 (ceph): mds: Add getter for EMetablob.client_name
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
10:21 AM Revision fb8f4698 (ceph): mds: add getter for ESession.client_inst
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
10:21 AM Revision 5b20debd (ceph): mds: Add LogEvent::get_type_str()
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
10:21 AM Revision d4a250ca (ceph): common: Add write_stream(ostream) to bufferlist
Enable C++ style file I/O when writing binary output
from bufferlists.
Signed-off-by: John Spray <john.spray@inktank...
John Spray
10:21 AM Revision e3a9f669 (ceph): objecter: Don't warn on multiple admin sockets
Suppress messages about failure to register admin sockets
if they are EEXIST, because this is a case that can occur
n...
John Spray
10:21 AM Revision 928d0f58 (ceph): mds: Fix typo 'Sesion'
Signed-off-by: John Spray <john.spray@inktank.com> John Spray
10:21 AM Revision 3fa825c9 (ceph): osdc: Revise Journaler format
* Separate journal encoding/envelope format
code (JournalStream) from I/O code (Journaler)
* Add new sentinel and s...
John Spray
10:21 AM Revision 74ac6e4e (ceph): mds: Add get_paths method to EMetaBlob
Used by cephfs-journal-tool to filter events
by path fragment.
Signed-off-by: John Spray <john.spray@inktank.com>
John Spray
07:06 AM Revision d776ec43 (ceph): mds: queue waiters in MDCache::force_dir_fragment()
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> Yan, Zheng
06:57 AM Revision b223055e (ceph): mds: fix remote auth pin race
When removing auth unpinned objects from mdr->remote_auth_pins,
Server::handle_slave_auth_pin() checks object's autho...
Yan, Zheng
05:42 AM Revision 991f7f15 (ceph): Merge pull request #1821 from ceph/mds-optracker-fixes
mds: fix Server::submit_mdlog_entry() Zheng Yan
05:38 AM Revision 57c89bf8 (ceph): mds: fix Server::submit_mdlog_entry()
mdr can be NULL when rolling back slave update
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Yan, Zheng
05:13 AM Revision 0c148788 (ceph): mds: allow early reply when there is no journaled slave update
Early reply should be Ok when there were slave requests, but none
of them modifies any object.
Signed-off-by: Yan, Z...
Yan, Zheng
05:13 AM Revision affce7a9 (ceph): mds: journal rename source inode when rollback rename
rename changes the source inode's ctime. rollback changes ctime back
to its original value.
Signed-off-by: Yan, Zhen...
Yan, Zheng
05:13 AM Revision 0bd4fa2e (ceph): mds: try trimming exported objects
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> Yan, Zheng
05:13 AM Revision f19cd4f7 (ceph): mds: drop locks after removing import/export state
dropping locks can dispatch requests, which may export the same
dirfrag again.
Signed-off-by: Yan, Zheng <zheng.z.ya...
Yan, Zheng
05:13 AM Revision 6fe7d175 (ceph): mds: skip journaling slave rename when possible
Rename operation can affect three dentries and two inodes. For MDS
who receives rename slave request, but isn't autho...
Yan, Zheng
05:12 AM Revision 37836532 (ceph): mds: include all of directory inode's replicas in rmdir witnesses
If a MDS crashed after journaling a rmdir operation, but before sending
MDentryUnlink messages. Survivor MDS may have...
Yan, Zheng
04:44 AM Revision 08b79ea8 (ceph): mds: journal EImportFinish after adjusting subtree auth
Submitting EImportFinish may journal an extra ESubtreeMap as side
effect. The ESubtreeMap should reflects subtree sta...
Yan, Zheng
01:33 AM Revision c18da048 (ceph): osd: fix narrow session refcount race
The set_priv() call consumes a ref. Technically the local 's' is not
actually owned by use by the time we get to the...
Sage Weil
01:32 AM Revision 2c4391b2 (ceph): osd: fix session leak in ms_handle_fast_connect
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil

05/17/2014

07:06 PM Bug #8380 (Resolved): osd/ReplicatedPG.cc: 1849: FAILED assert(0 == "out of order op")
... Sage Weil
05:15 PM Revision 6eac0baf (ceph): Added more logging
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
05:07 PM Revision 686c01b5 (ceph): Added sleep log info line
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
05:41 AM Revision 15350de7 (ceph): Add header cache to DBObjectMap
OMap is used by PG to store PGLog, so it's called by each write operations.
Because leveldb don't play well in random...
Haomai Wang

05/16/2014

11:49 PM Bug #8379: cephx: clock skew causes verify_reply decode error
the pervios test cli 'ceph health' is wrong
In actually the result alway be 'HEALTH_OK'
I collect the osd log:
...
jian li
08:54 PM Bug #8379 (Can't reproduce): cephx: clock skew causes verify_reply decode error
linux: 3.11.0-12-generic #19-Ubuntu SMP Wed Oct 9 16:20:46 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux
4 osd and 1 mon
...
jian li
10:17 PM Revision 5ace4d1a (ceph): Merge pull request #1809 from ceph/wip-4354-mds-optracker
MDS OpTracker
Reviewed-by: Sage Weil <sage@inktank.com>
Gregory Farnum
10:15 PM Revision ee92a39b (ceph): MDS: add admin socket cleanup on shutdown
Signed-off-by: Greg Farnum <greg@inktank.com> Greg Farnum
07:36 PM Documentation #7725: osd pool default size need updating
Please leave this to Florent as a low hanging fruit ;-) Loïc Dachary
06:52 PM rbd Bug #8378 (Resolved): krbd: Kernel oops in rbd_img_obj_callback
... Rain Li
06:31 PM rbd Bug #6494: High memory consumption of qemu/librbd with enabled cache
During RBD performance testing using the librbd fio engine, I appear to be able to reproduce something that looks qui... Mark Nelson
04:44 PM Bug #8334: osd: bug with pool snaps, ceph_test_rados
found another related flaw, testing more Samuel Just
04:37 PM Revision 7a153818 (ceph): Merge pull request #1817 from ceph/wip-osdmon-pa-wait
OSDMonitor: set next commit in mon primary-affinity reply
Reviewed-by: Sage Weil <sage@inktank.com>
Ilya Dryomov
04:25 PM Revision a78b14ec (ceph): OSDMonitor: set next commit in mon primary-affinity reply
Commit 8c5c55c8b47e ("mon: set next commit in mon command replies")
fixed MMonCommand replies to include the right ve...
Ilya Dryomov
02:08 PM Revision a877d12f (ceph): Merge pull request #1816 from onlyjob/docs
sample.ceph.conf: minor update
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
11:40 AM devops Bug #7627: ceph-disk: does not start daemons properly under systemd
Alfredo Deza wrote:
> Fedora 19 does not have (nor it provides) systemd-run:
>
> [...]
>
> According to the Fe...
Sage Weil
11:29 AM devops Bug #7627: ceph-disk: does not start daemons properly under systemd
Fedora 19 does not have (nor it provides) systemd-run:... Alfredo Deza
10:30 AM Revision 6dfc544e (ceph): sample.ceph.conf: minor update
* Moved filestore settings above [osd.*] declarations otherwise
(if uncommented) those settings might be applied o...
Dmitry Smirnov
10:25 AM Bug #8375 (Can't reproduce): ceph_rest_api test (workunits/rest/test.py) failure
http://pulpito.ceph.com/gregf-2014-05-15_17:40:43-rados-wip-osdmap-sub-bug-testing-basic-plana/255543/
This was on...
Greg Farnum
09:26 AM Bug #8371 (Pending Backport): osd not booting
Sage put this in master (commit:290ac818696414758978b78517b137c226110bb4), and it passed the suite run overnight. Greg Farnum
07:42 AM devops Bug #8374 (Won't Fix): redhat-lsb is not recognized as a dependency in FC19
In issue #8028 redhat-lsb-core was added in the spec file, and although it works in CentOS it most certainly is not
...
Alfredo Deza
02:15 AM Bug #8373 (Resolved): Pool creation defaults to ruleset 0 even if it doesn't exist
... John Spray

05/15/2014

11:54 PM Revision 290ac818 (ceph): OSD: fix an osdmap_subscribe interface misuse
When calling osdmap_subscribe, you have to pass an epoch newer than the
current map's. _maybe_boot() was not doing th...
Greg Farnum
09:44 PM Revision 6ec3c460 (ceph): osd: include osd_objectstore in metadata reported to mon
Provide visibility into the OSD backend.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
08:48 PM Bug #8332: ceph-test-objectstore: bad return value in unlink
Samuel Just
08:48 PM Bug #8332: ceph-test-objectstore: bad return value in unlink
Ahah, the ENOENT is on the subdir which had been merged. The bug occurs when cleanup() happens after the subdir is m... Samuel Just
02:57 PM Bug #8332: ceph-test-objectstore: bad return value in unlink
get_info is returning -ENOENT for a clearly extant collection (checked the current/ directory manually). I've set up... Samuel Just
10:16 AM Bug #8332 (In Progress): ceph-test-objectstore: bad return value in unlink
Ian Colle
08:42 PM Bug #8301: Suicide Timeout on Cache Tier OSDs
Sadly this happened again today when running RBD tests with 18 OSDs using SSD journals. Not seeing much new in the c... Mark Nelson
05:56 PM rgw Bug #8372 (Won't Fix): Can't abort multipart uploads created with Emperor
We have buckets containing abandoned multipart uploads created with rgw 0.72.2. After upgrading to rgw 0.80.1, these... Benjamin Gilbert
05:44 PM Bug #8371: osd not booting
The master version passed my local test (I got master to hang, and then ran the same test and it did not hang). Sched... Greg Farnum
05:07 PM Bug #8371: osd not booting
wip-osdmap-sub-bug
wip-dumpling-osdmap-sub-bug
It'll also need a firefly backport, but maybe we should just remov...
Greg Farnum
05:05 PM Bug #8371 (Resolved): osd not booting
I'm not entirely clear on how the monmap subscriptions work, but it
looks from the log as if
OSD::_maybe_boot wit...
Samuel Just
04:48 PM Revision 333c7ad7 (ceph): Merge pull request #257 from ceph/wip-samba-fix
increased time wait to 90 sec so tests does not fail Zack Cerza
04:33 PM Revision 1a503aff (ceph): increased time wait to 90 sec so tests does not fail
Yuri Weinstein
04:08 PM Revision d8c74392 (ceph): Add Pulpito links to teuthology emails
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
04:06 PM CephFS Bug #6608 (Can't reproduce): samba teuthology dbench failure
Zheng Yan
03:17 PM CephFS Feature #8369 (New): ceph-fuse: handle suspend and resume more gracefully
Right now, suspend and resume are treated as "mds disconnect" events and we basically lose everything or risk data lo... Greg Farnum
03:06 PM CephFS Feature #8368 (Resolved): kernel: Notify users of mds disconnect and allow them to react to it
See the discussion at http://comments.gmane.org/gmane.comp.file-systems.ceph.devel/19594, and the userspace version a... Greg Farnum
02:19 PM Bug #8367 (Duplicate): osd crashed in upgrade:dumpling-x:stress-split-firefly---basic-plana
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-05-14_19:55:03-upgrade:dumpling-x:stress-split-firefl... Yuri Weinstein
02:10 PM devops Tasks #8366 (Resolved): Update ceph.com/docs to default to the latest major release (0.80)
When users go to our docs, it should show the branch for the most recent major release (Firefly).
Users should be ab...
Neil Levine
01:44 PM Support #7609 (Resolved): http://tracker.ceph.com/account/register returns 500 Internal error
Applied patch and verified by someone that they were able to register with the problem browser after originally not b... Sandon Van Ness
01:10 PM devops Bug #7486: python-backports needs fixing for rhel
Alfredo Deza
01:08 PM devops Bug #7486: python-backports needs fixing for rhel
Sheldon: could you give us a bit more information as to how you are getting these warnings? Sandon and I could not re... Alfredo Deza
09:33 AM Bug #8361 (Can't reproduce): osd/osd_types.h: 2868: FAILED assert(rwstate.empty())

OSD crash with error
osd/osd_types.h: 2868: FAILED assert(rwstate.empty())
ceph version 0.80.1 (a38fe1169b6d...
Sergey Korolev
05:21 AM Bug #8359 (Resolved): FTBFS with gcj-jdk v4.9.0
Ceph FTBFS with gcj-jdk v4.9.0:... Dmitry Smirnov
03:43 AM Revision c3dccbf2 (ceph): Downgrade paramiko to work around an SFTP bug
See http://tracker.ceph.com/issues/8327
Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
Zack Cerza
02:53 AM Feature #8343: please enable data integrity checking (by default) / silent data corruption
Greg Farnum wrote:
> Have you ever run a scrub repair on the cluster?
Yes. Since release of 0.80 I instructed all...
Dmitry Smirnov

05/14/2014

09:20 PM Revision 405063b1 (ceph): workunits: provide some output in the dirfrag.sh test
Signed-off-by: Greg Farnum <greg@inktank.com> Greg Farnum
08:26 PM rbd Documentation #8347: Openstack / Cinder Config
Thanks for the logs! This is (https://bugs.launchpad.net/cinder/+bug/1308058), which is fixed in cinder's master bran... Josh Durgin
09:34 AM rbd Documentation #8347: Openstack / Cinder Config
OK, I did that, but didn't see anything interesting come up in the cinder-volume.log. I did get this traceback in the... Quentin Hartman
08:01 AM rbd Documentation #8347: Openstack / Cinder Config
Sure. I'll have to get my other stuff to a stopping point before I can do that though.
fwiw, I did do some poking ...
Quentin Hartman
07:53 AM rbd Documentation #8347 (In Progress): Openstack / Cinder Config
That setting is necessary for copy-on-write cloning to work when creating a volume from an image. It sounds like ther... Josh Durgin
08:24 PM Revision fe19a1db (ceph): Merge pull request #1803 from onlyjob/java-gcj
Java GCJ fixes
Reviewed-by: Greg Farnum <greg@inktank.com>
Acked-by: Noah Watkins <noahwatkins@gmail.com>
Gregory Farnum
08:17 PM Revision f47c160e (ceph): PG: replace is_split, acting_up_affected with should_restart_peering
This way, we restart peering using the same criteria as
check_new_interval.
Fixes: #8104
Signed-off-by: Samuel Just ...
Samuel Just
08:17 PM Revision aec5634e (ceph): osd_types: remove the pool_id argument from (is|check)_new_interval
Signed-off-by: Samuel Just <sam.just@inktank.com> Samuel Just
08:17 PM Revision 2ee35511 (ceph): osd_types: factor out is_new_interval from check_new_interval
Signed-off-by: Samuel Just <sam.just@inktank.com> Samuel Just
08:09 PM Revision 4391ce53 (ceph): Merge pull request #31 from ceph/wip-fuse-permission
marginal/multimds: fuse_default_permissions = 0 for ceph-fuse Gregory Farnum
07:12 PM Revision e6e3cec3 (ceph): Merge pull request #1806 from ceph/wip-8011
ReplicatedPG: block scrub on blocked object contexts
Reviewed-by: Sage Weil <sage@inktank.com>
Samuel Just
03:58 PM Revision c48a4ef9 (ceph): common/perf_counters: use second atomic to make counters safe to read
Add a second counter so that we can detect a race with an add/inc during
read, and retry.
Signed-off-by: Sage Weil <...
Sage Weil
03:23 PM CephFS Feature #8358 (New): client: opportunistically update backtraces on files
I've seen a few reports that backtrace updates are causing issues due to the increased write load (and the bursty nat... Greg Farnum
02:44 PM CephFS Feature #4354 (Fix Under Review): mds: add an equivalent to the OSD OpTracker
https://github.com/ceph/ceph/pull/1809 Greg Farnum
01:35 PM devops Tasks #8240 (In Progress): Build 0.67.8 & 0.80 on RHEL7-RC
Firefly is built/pushed out. Dumpling still needs to be built/pushed. Sandon Van Ness
12:54 PM Bug #7804: backfill racing with a hitset object remove
I don't think this bug is fixed, picking this one as the root bug (there was a duplicates loop) Samuel Just
11:27 AM Feature #8343: please enable data integrity checking (by default) / silent data corruption
Have you ever run a scrub repair on the cluster?
Was your cluster going through any backfilling?
The corrupt data...
Greg Farnum
10:52 AM Bug #7765: Bogus bandwidth statistics during pg creation
Ian Colle
08:37 AM Bug #7765: Bogus bandwidth statistics during pg creation
Confirmed this is still an issue on master, completely breaks the calamari iops chart with the spike generated :-/ John Spray
10:51 AM RADOS Feature #8355 (New): extend rados import to support recovering from archived erasure-coded shards
This is a lot trickier: we'll need to be able to specify the erasure coding library and paramaters in order to recons... Samuel Just
09:16 AM rgw Bug #8233: Installation & Documentation broken for Ubuntu Trusty 14.04 - rgw
You could get around this by apt pinning or via aptitude (which is effectively doing the same thing)
This just fee...
Joel Merrick
09:14 AM rgw Bug #8233: Installation & Documentation broken for Ubuntu Trusty 14.04 - rgw
... Joel Merrick
09:11 AM rgw Bug #8233: Installation & Documentation broken for Ubuntu Trusty 14.04 - rgw
Sure, the issue is that the 100-continue packages you provide are not installable. This is due to apache 2.4 being th... Joel Merrick
07:23 AM Revision ab907c5a (ceph): doc: Clarified Debian uses sysvinit.
Fixes: #7182
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
07:14 AM Revision c71c2921 (ceph): doc: Added rgw print continue guidance.
Fixes: #7731
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
07:13 AM Revision b082fd68 (ceph): doc: Minor edit.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
07:02 AM Revision ca833bd5 (ceph): doc: Added clarifying text to CRUSH add command.
Fixes: #8322
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
06:20 AM Revision 48337e0c (ceph): doc: Omitted glance_api_version=2 to fix creating images from volumes.
Fixes: #8347 John Wilkins
06:18 AM Revision 17930a1e (ceph): doc: Changed example to use virtio and put key usage into one line.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
01:49 AM rbd Bug #7790 (Resolved): Kernel panic when creating ZFS pools on CEPH RBD devices
... Ilya Dryomov
01:46 AM Revision c7540cb6 (ceph): Merge pull request #1802 from ceph/wip-mds-misc
Wip mds misc
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
01:24 AM Bug #8335: Crash while recovering from XFS corruption
After successful XFS recovery Ceph is crashing. Or Ceph cannot recover itself? Do I need to delete this OSD and creat... Pavel Veretennikov
01:11 AM Revision b4b128b2 (ceph): Merge pull request #1810 from ceph/wip-fedora
doc: update instructions for RPM distros Sage Weil
12:39 AM Documentation #2994 (Resolved): doc: expand/complete librados API doc
Created an intro to librados. John Wilkins
12:38 AM Documentation #2624 (Resolved): OpenStack creation instructions should recommend non-default numb...
John Wilkins
12:36 AM Documentation #1814 (Resolved): doc: openstack + ceph install howto
John Wilkins
12:26 AM Documentation #6774 (Resolved): Documentation: osd scrub load threshold incorrect.
Was resolved some time ago. John Wilkins
12:23 AM devops Documentation #7182 (Resolved): documentation miscategorizes debian as upstart based. debian use...
John Wilkins
12:20 AM devops Documentation #7182 (In Progress): documentation miscategorizes debian as upstart based. debian ...
John Wilkins
12:13 AM rgw Documentation #7731 (Resolved): Warning about "rgw print continue" should be added to radosgw con...
John Wilkins
12:07 AM rgw Documentation #7731 (In Progress): Warning about "rgw print continue" should be added to radosgw ...
John Wilkins
12:01 AM Documentation #8322 (Resolved): make "manually add OSD" documents to make CRUSH command needs cle...
John Wilkins

05/13/2014

11:57 PM Documentation #8322 (In Progress): make "manually add OSD" documents to make CRUSH command needs ...
John Wilkins
11:36 PM rgw Bug #8233 (Need More Info): Installation & Documentation broken for Ubuntu Trusty 14.04 - rgw
I haven't installed on Trusty yet, but will upgrade my machines soon. Can you provide more specifics of the problems ... John Wilkins
11:22 PM rbd Documentation #8347 (Resolved): Openstack / Cinder Config
I removed the reference entirely. I also removed it from http://openstack.redhat.com/Using_Ceph_for_Cinder_with_RDO_H... John Wilkins
11:13 PM rbd Documentation #8347 (In Progress): Openstack / Cinder Config
John Wilkins
03:39 PM rbd Documentation #8347 (Resolved): Openstack / Cinder Config
At http://ceph.com/docs/master/rbd/rbd-openstack/#configuring-cinder the setting "glance_api_version=2" is recommende... Quentin Hartman
09:48 PM Revision 8dd1190d (ceph): Improve Bash completion for various tools
Wido den Hollander
09:26 PM Revision c7d7abae (ceph): Merge pull request #256 from ceph/wip-6542-wusui
Add missng docstrings to repair_test.py Zack Cerza
09:02 PM Revision 5dfc5700 (ceph): Add missng docstrings to repair_test.py
Fixes: 6542
Signed-off-by: Warren Usui <warren.usui@inktank.com>
Warren Usui
08:59 PM Feature #8343: please enable data integrity checking (by default) / silent data corruption
Sure but instead of going deep into probably irrelevant details of how OSD was lost I would appreciate more focus on ... Dmitry Smirnov
07:31 PM Feature #8343: please enable data integrity checking (by default) / silent data corruption
You're going to need to provide more details; as described this doesn't quite make sense.
Did you witness filesystem...
Greg Farnum
04:50 PM Feature #8343: please enable data integrity checking (by default) / silent data corruption
P.S. Naturally corruption was not limited to test data file -- at least one MySQL database was lost on RBD device (it... Dmitry Smirnov
06:54 AM Feature #8343 (Closed): please enable data integrity checking (by default) / silent data corruption
I'm scrubbing CephFS with @fsprobe@ which detected data corruption around the time when one OSD died due to Btrfs fil... Dmitry Smirnov
08:38 PM Revision 00225d73 (ceph): test: fix some templates to match new output code
Signed-off-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Samuel Just <sam.just@inktank.com>
Greg Farnum
08:01 PM Bug #8349 (Resolved): env-vs-args unittest is racy
The env-vs-args unit test frequently fails non-deterministically on the gitbuilders. I'm pretty sure it's an issue wi... Greg Farnum
06:22 PM Revision 20aad8ff (ceph): doc: update instructions for RPM distros
Fix RPM building instructions: this has been broken since
libs3 was included inline in the ceph repo as a submodule.
...
John Spray
06:09 PM Revision 010f83f1 (ceph): Fix unit tests under Jenkins
os.getlogin() was throwing:
OSError: [Errno 25] Inappropriate ioctl for device
Signed-off-by: Zack Cerza <zack.cer...
Zack Cerza
05:58 PM Bug #8335: Crash while recovering from XFS corruption
How is this related to Ceph?
Corruption on xfs may be manifestation of hardware errors or a kernel bug.
I would c...
Dmitry Smirnov
05:54 PM Revision 04c280e7 (ceph): Merge remote-tracking branch 'origin/master' into wip-4354-mds-optracker
Conflicts:
src/mds/Locker.cc
src/osd/OpRequest.cc
src/osd/OpRequest.h
Signed-off-by: Greg Farnum <greg@inktank.com>
Greg Farnum
05:46 PM Feature #8348 (New): include "ChangeLog" and/or "NEWS" files to release tarball
Please include meaningful "ChangeLog" and/or "NEWS" files to release tarball.
It is nice when those files contain in...
Dmitry Smirnov
04:36 PM CephFS Bug #8291 (Resolved): 0.80: fuse client hangs after wake-up from suspend until restart of active MDS
Finally with following commits from "master":
* commit:b8aa58af793bea4ed1a150ac5bf554fc894774f1
* commit:70ab07...
Dmitry Smirnov
09:36 AM CephFS Bug #8291: 0.80: fuse client hangs after wake-up from suspend until restart of active MDS
Zheng Yan wrote:
> The attached (one line) patch can make fuse-client work as you expected.
Thanks, but I still...
Dmitry Smirnov
04:22 PM Bug #8332: ceph-test-objectstore: bad return value in unlink
reproducible, working on it Samuel Just
11:23 AM Bug #8332: ceph-test-objectstore: bad return value in unlink
Hmm, the assert is after a failure injection and cleanup sequence, trying to reproduce. Samuel Just
04:15 PM Bug #8333: ceph_test_rados_delete_pools_parallel: Received fewer notifies than expected: 0 < 1
It appears that the semaphores in st_rados_(watch|notify).cc and rados_watch_notify.cc should have prevented this ord... Samuel Just
04:06 PM Bug #8333: ceph_test_rados_delete_pools_parallel: Received fewer notifies than expected: 0 < 1
Odd... the watch in the test happened out of order. Samuel Just
03:21 PM Revision de321790 (ceph): Use VersionNotFoundError packages are missing
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
03:11 PM devops Bug #8330: repodata on rpm repos do not list latest ceph-deploy (1.5.2)
Yes, exactly:... Simon Ironside
07:21 AM devops Bug #8330: repodata on rpm repos do not list latest ceph-deploy (1.5.2)
You mean, you cannot install any other package in that repository? Alfredo Deza
02:50 PM Bug #8346 (Can't reproduce): OSD crashes on master (FAILED assert(ip_op.waiting_for_commit.count(...

Revision bfce3d4, home-built packages on Fedora 20: wasn't actually trying to test Ceph, but since I saw some crash...
John Spray
02:44 PM Bug #8345 (Rejected): PG::repair_object() should check for and return errors

The function should make sure that the OI_ATTR exists.
The decode of object_info_t should should be called explici...
David Zafman
01:05 PM Feature #8231: ceph filestore dump improvements
Adding:
ceph-filestore-dump ... <pgid> <object> list-omap
David Zafman
01:04 PM Feature #8231 (In Progress): ceph filestore dump improvements
David Zafman
12:42 PM Bug #8344 (Can't reproduce): Upstart scripts silently fail when asok missing
In situations like Issue 7188, the admin socket can be lost from /var/run/ceph/ceph-<daemon>.<name>.asok. When this h... Mike Dawson
12:21 PM Bug #7188: Admin socket files are lost on log rotation calling initctl reload (ubuntu 13.04 only)
I think this fix was incomplete: https://www.mail-archive.com/ceph-users@lists.ceph.com/msg09754.html
We probably ...
Greg Farnum
11:13 AM Bug #8334: osd: bug with pool snaps, ceph_test_rados
Samuel Just
09:11 AM rgw Bug #8251 (Fix Under Review): radosgw-agent does not sync objects uploaded to recreated buckets
Yehuda Sadeh
09:07 AM Linux kernel client Feature #3837 (In Progress): krbd: support format 2 striping
Ian Colle
09:06 AM Linux kernel client Feature #190: krbd: DISCARD support
Ian Colle
06:31 AM Bug #8342 (Resolved): init script may not start all OSDs
On server with multiple OSDs init script aborts and do not attempt to start other OSDs after failure to start one of ... Dmitry Smirnov
05:52 AM Revision 26151ec6 (ceph): mds: lower IO priority of storing backtrace
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> Yan, Zheng
04:39 AM Revision bfce3d4f (ceph): Merge pull request #1771 from ceph/wip-5021
Wip 5021
Reviewed-by: Sage Weil <sage@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
04:31 AM Linux kernel client Bug #8341 (New): improve falover to next available MON
I suspend one computer with RBD mapped device fairly often.
Every time it comes out of suspend gracefully but today ...
Dmitry Smirnov
01:31 AM Revision 20814de9 (ceph): Merge pull request #1807 from ceph/wip-mds-flock
mds: reduce verbosity of handle_client_file_{readlock,setlock} Zheng Yan
01:29 AM Revision 019483fd (ceph): mds: reduce verbosity of handle_client_file_{readlock,setlock}
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> Yan, Zheng
01:27 AM Revision ca313c20 (ceph): mds: add a Server::submit_mdlog_entry() to provide event marking
Signed-off-by: Greg Farnum <greg@inktank.com> Greg Farnum
12:22 AM Revision 9f0825ca (ceph): OSD: verify that client ops are targeted correctly in the current epoch
We were previously only looking at the epoch the op was sent in, which meant
that if we had dropped responsibility so...
Greg Farnum
12:09 AM CephFS Bug #5021 (Resolved): ceph-fuse: crash on traceless reply
Zheng Yan
12:07 AM CephFS Bug #2863 (Resolved): client: does not tolerate traceless replies from mds
Zheng Yan

05/12/2014

11:04 PM rgw Bug #8340 (Rejected): The name of request headers in radios gateway is not match with Amazon S3 w...
Error Message?
1. Create a Amazon s3 client to test the function of radosgw.
2. Copy object with optional constrain...
Jingjing Zhao
10:31 PM Revision 74114771 (ceph): ReplicatedPG: block scrub on blocked object contexts
Fixes: #8011
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com>
Samuel Just
10:29 PM Revision c32c56b7 (ceph): Merge pull request #1779 from ceph/wip-7553
Wip 7553
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
10:08 PM Revision 2ec21827 (ceph): ReplicatedPG::start_flush: send delete even if there are no snaps
Even if all snaps for the clone have been removed, we still have to
send the delete to ensure that when the object is...
Samuel Just
09:44 PM Revision a6aa8121 (ceph): MDCache: mark ops at various finish points
Signed-off-by: Greg Farnum <greg@inktank.com> Greg Farnum
09:44 PM Revision 2df68b6b (ceph): Server: mark events when journaling and replying
Signed-off-by: Greg Farnum <greg@inktank.com> Greg Farnum
09:44 PM Revision ca917430 (ceph): Locker: mark_event in acquire_locks() when blocking or succeeding
Signed-off-by: Greg Farnum <greg@inktank.com> Greg Farnum
09:44 PM Revision 87f6cd49 (ceph): MDS: add an OpTracker and use it
Signed-off-by: Greg Farnum <greg@inktank.com> Greg Farnum
09:41 PM Revision b2778029 (ceph): Mutation: add an MDRequestParams struct and use that when building MDRe...
We now have a single constructor and one path to build MDRequests with.
Signed-off-by: Greg Farnum <greg@inktank.com>
Greg Farnum
09:39 PM Revision 06d6d32b (ceph): mds: remove a couple leftover declarations of MDRequest
Signed-off-by: Greg Farnum <greg@inktank.com> Greg Farnum
09:39 PM Revision 0d89e5ca (ceph): MDCache: pass the causative message to request_start_slave()
We were passing the causative MDS (as an int), but pushing down the
actual Message will help us as we set up an OpTra...
Greg Farnum
09:39 PM Revision ae80a1f3 (ceph): MDS: add stubs for an AdminSocketHook
Signed-off-by: Greg Farnum <greg@inktank.com> Greg Farnum
09:25 PM Revision 428319e8 (ceph): doc/release-notes: v0.80.1
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
09:25 PM Revision 971c0652 (ceph): Use config.archive_base if one isn't passed
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
09:09 PM Revision d945e564 (ceph): Add retries to orchestra.connection.connect()
This is an attempt to fix: http://tracker.ceph.com/issues/8314
Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
Zack Cerza
09:03 PM Revision 2b8232a3 (ceph): Better logging
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
09:03 PM Revision dfb2352d (ceph): Fix typo
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
08:33 PM Revision b99244e5 (ceph): Merge pull request #1799 from ceph/wip-8305
osd: fix op ordering with pool overlay set/removed
Reviewed-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Josh...
Sage Weil
08:20 PM Revision 19f8849a (ceph): doc: Improvements to qemu installation.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
07:42 PM CephFS Bug #8291: 0.80: fuse client hangs after wake-up from suspend until restart of active MDS
Dmitry Smirnov wrote:
> Greg Farnum wrote:
> > A suspended client isn't participating in the cluster and gets boote...
Zheng Yan
07:08 PM CephFS Bug #8291: 0.80: fuse client hangs after wake-up from suspend until restart of active MDS
Greg Farnum wrote:
> A suspended client isn't participating in the cluster and gets booted out; if it has stale data...
Dmitry Smirnov
09:19 AM CephFS Bug #8291: 0.80: fuse client hangs after wake-up from suspend until restart of active MDS
A suspended client isn't participating in the cluster and gets booted out; if it has stale data it *cannot* rejoin th... Greg Farnum
03:08 AM CephFS Bug #8291: 0.80: fuse client hangs after wake-up from suspend until restart of active MDS
Zheng Yan wrote:
> The patch doesn't make ceph-fuse recover automatically. you need to use following command to reco...
Dmitry Smirnov
01:08 AM CephFS Bug #8291: 0.80: fuse client hangs after wake-up from suspend until restart of active MDS
Dmitry Smirnov wrote:
> Perhaps my initial confusion was because only the above-mentioned commit was cherry-picked t...
Zheng Yan
12:26 AM CephFS Bug #8291: 0.80: fuse client hangs after wake-up from suspend until restart of active MDS
Perhaps my initial confusion was because only the above-mentioned commit was cherry-picked to "firefly" branch.
Zh...
Dmitry Smirnov
06:33 PM Revision 6e4455d6 (ceph): doc: Added note on Default requiretty for CentOS and others.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
06:31 PM CephFS Bug #8337 (Resolved): Logging too verbose in handle_client_file_setlock
fixed by commit 019483fdaa Zheng Yan
01:08 PM CephFS Bug #8337 (Resolved): Logging too verbose in handle_client_file_setlock
Each time, a client uses fcntl locking, the mds server writes a log message with loglevel 0. This results in quite a ... Markus Blank-Burian
05:57 PM Revision 756a6bfc (ceph): Move "no results server" warning
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
05:57 PM Revision ad012469 (ceph): Move list of exceptions to catch
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
05:57 PM Revision 456a1148 (ceph): Add try_mark_run_dead()
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
05:57 PM Revision 47f5d835 (ceph): Use try_mark_run_dead()
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
05:57 PM Revision e0e01265 (ceph): Fix name parsing
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
05:50 PM Revision b1859c79 (ceph): Merge pull request #255 from ceph/wip-6921-wusui
Allow .teuthology.yaml to set downburst path Zack Cerza
05:22 PM Bug #8338: OSD: no longer checking that ops on older maps are correctly targeted
wip-8338. It passes trivial tests (local cluster, rados bench); Sam said he'd run it through testing with some change... Greg Farnum
04:12 PM Bug #8338 (Resolved): OSD: no longer checking that ops on older maps are correctly targeted
I probably did this, but I'm not sure how. Found a hung kclient run today, blocked on OSD ops. This was in the OSD lo... Greg Farnum
03:28 PM Feature #7553 (Resolved): Remove classic scrub
Samuel Just
03:15 PM Bug #7891: osd: leaked pg refs on shutdown
ubuntu@teuthology:/a/samuelj-2014-05-09_14:11:19-rados-wip-sam-testing-testing-basic-plana/245865 Samuel Just
06:08 AM Bug #7891: osd: leaked pg refs on shutdown
ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-05-11_09:34:50-rados-firefly-testing-basic-plana/249163 Sage Weil
03:13 PM Bug #6756: journal full hang on startup
ubuntu@teuthology:/a/samuelj-2014-05-09_14:11:19-rados-wip-sam-testing-testing-basic-plana/245820 Samuel Just
03:12 PM Revision a38fe116 (ceph): 0.80.1
Jenkins
03:11 PM Revision ab873622 (ceph): 0.80.1
Jenkins
03:10 PM Revision 03e7d1d5 (ceph): 0.80.1
Jenkins
03:09 PM Revision 7a4919c7 (ceph): 0.80.1
Jenkins
01:36 PM devops Bug #8330: repodata on rpm repos do not list latest ceph-deploy (1.5.2)
Behaviour seems to have changed here since yesterday.
ceph-deploy 1.5.2 is now the *only* package that appears in ...
Simon Ironside
01:32 PM Bug #8328 (Pending Backport): osd: null op on dup_ops list
Sage Weil
01:32 PM Bug #8305 (Pending Backport): objecter, osd: pool overlay change should trigger op resend
Sage Weil
01:10 PM Revision 59e2381f (ceph): Merge pull request #1801 from ceph/wip-update-gitignore
Update gitignore entries for master
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
12:51 PM Revision b4ffd661 (ceph): Merge pull request #1800 from ceph/wip-da-SCA-20140510
fixes from SCA
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil
08:59 AM Bug #8301: Suicide Timeout on Cache Tier OSDs
Over the weekend I was able to reproduce this under valgrind. Sadly valgrind didn't report any errors for the OSDs t... Mark Nelson
07:38 AM Bug #8335 (Rejected): Crash while recovering from XFS corruption
There was a FS corruption:... Pavel Veretennikov
06:04 AM Bug #8334 (Resolved): osd: bug with pool snaps, ceph_test_rados
ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-05-11_09:34:50-rados-firefly-testing-basic-plana/249051
f...
Sage Weil
05:57 AM Bug #8333 (Can't reproduce): ceph_test_rados_delete_pools_parallel: Received fewer notifies than ...
... Sage Weil
05:56 AM Bug #8332 (Resolved): ceph-test-objectstore: bad return value in unlink
ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-05-11_09:34:50-rados-firefly-testing-basic-plana/248775... Sage Weil
05:19 AM rbd Feature #7895 (Resolved): krbd: test cloning, discard, plus regular I/O via fsx
https://github.com/ceph/ceph/pull/1766 Ilya Dryomov
05:18 AM rbd Bug #8184 (In Progress): krbd: make sure we have latest osdmap on 'rbd map'
Ilya Dryomov
05:16 AM Linux kernel client Bug #8226 (Resolved): 0.80~rc1: RBD read errors (ENXIO)
... Ilya Dryomov
04:08 AM Revision 8b682d16 (ceph): prioritise use of `javac` executable (gcj provides it through alternati...
On Debian this fixes FTBFS when gcj-jdk and openjdk-7-jdk are installed at
the same time because build system will u...
Dmitry Smirnov
04:06 AM Revision 89fe0353 (ceph): pass '-classpath' option (gcj/javah ignores CLASSPATH environment varia...
This should not affect OpenJDK which understands '-classpath' as well.
With gcj-jdk we still get FTBFS later:
~~~...
Dmitry Smirnov
03:57 AM Revision 0f4120c0 (ceph): look for "jni.h" in gcj-jdk path, needed to find "jni.h" with gcj-jdk_4...
Signed-off-by: Dmitry Smirnov <onlyjob@member.fsf.org> Dmitry Smirnov
03:38 AM Revision 20015726 (ceph): mds: deny reconnect for closed session
The client that tries reconnect may have dirty caps and unsafe requests.
Allowing the reconnect attempt may compromis...
Yan, Zheng
03:26 AM Revision 59f539c1 (ceph): mds: revert EMetaBlob::{fullbit,remotebit,nullbit} encoding optimization
Revert commit 40d56a97 (mds: optimize EMetaBlob::fullbit, remotebit,
nullbit encoding). This optimization creates sma...
Yan, Zheng
03:14 AM Revision 1f92f558 (ceph): mds: cleanup usage of MDCache::predirty_journal_parent()
The sixth parameter of MDCache::predirty_journal_parent() is 'int'
with default value 0.
Signed-off-by: Yan, Zheng <...
Yan, Zheng
03:14 AM Revision 54a90376 (ceph): mds: avoid journaling unnecessary dir context
If base inode is reached, try clearing the 'maybe' list, then stop.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Yan, Zheng
03:14 AM Revision 58ee5560 (ceph): mds: propagate inode rstat if it has never been propagated
Otherwise the 'last_dirstat_prop' of directory inode keeps in 'never'
state.
Signed-off-by: Yan, Zheng <zheng.z.yan@...
Yan, Zheng
03:14 AM Revision f35648bf (ceph): mds: properly clear new flag for stale client cap
CInode::encode_inodestat() should clear the 'new' flag of client
cap even when session is stale, because the 'new' fl...
Yan, Zheng
12:33 AM Revision 3d7f527c (ceph): BtrfsFileStoreBackend.cc: fix ::unlinkat() result handling
Don't check for 'fd' but for the return value of the ::unlinkat() call.
Fix for:
[src/os/BtrfsFileStoreBackend.cc:72...
Danny Al-Gaaf
12:17 AM Revision 5f89128f (ceph): TestLFNIndex.cc: remove unused variable 'mangled_name'
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de> Danny Al-Gaaf
12:07 AM Revision a4455299 (ceph): rgw_user.cc: remove dead assignment in generate_key()
Fix for:
[src/rgw/rgw_user.cc:778]: (style) Variable 'subuser' is
assigned a value that is never used.
Signed-off-b...
Danny Al-Gaaf
12:01 AM Revision b1196795 (ceph): rgw_user.cc: cleanup RGWAccessKeyPool::check_op()
Remove dead assignment and unsued variable 'secret_key'. Check
op_state.get_access_key() directly for emptiness witho...
Danny Al-Gaaf
 

Also available in: Atom