Commit Graph

37469 Commits

Author SHA1 Message Date
Ray Lv
7925b82c20 rgw: Conditional PUT on ETag
Fixes: #8562

Signed-off-by: Ray Lv <raylv@yahoo-inc.com>
2014-12-12 00:31:16 -08:00
Sage Weil
cfa54bbbc7 Merge pull request #3155 from athanatos/wip-8011
ReplicatedPG: fail a non-blocking flush if the object is being scrubbed

Reviewed-by: Sage Weil <sage@redhat.com>
2014-12-11 13:42:09 -08:00
Samuel Just
9b26de3f36 ReplicatedPG: fail a non-blocking flush if the object is being scrubbed
Fixes: #8011
Backport: firefly, giant
Signed-off-by: Samuel Just <sjust@redhat.com>
2014-12-11 13:30:11 -08:00
Sage Weil
9a04e7515d Merge pull request #3154 from athanatos/wip-10059
Wip 10059

Reviewed-by: Sage Weil <sage@redhat.com>
2014-12-11 13:28:03 -08:00
Sage Weil
69507e51b5 Merge pull request #3156 from athanatos/wip-10150
ReplicatedPG::scan_range: an object can disappear between the list and t...

Reviewed-by: Sage Weil <sage@redhat.com>
2014-12-11 13:23:28 -08:00
Sage Weil
f6a02697b1 Merge pull request #3157 from athanatos/wip-10157
PGLog: include rollback_info_trimmed_to in (read|write)_log

Reviewed-by: Sage Weil <sage@redhat.com>
2014-12-11 13:22:05 -08:00
Sage Weil
2667eed93c Merge pull request #3158 from athanatos/wip-10259
osd_types: op_queue_age_hist and fs_perf_stat should be in osd_stat_t::o...

Reviewed-by: Sage Weil <sage@redhat.com>
2014-12-11 13:16:13 -08:00
Samuel Just
dce6f288ad ReplicatedPG::scan_range: an object can disappear between the list and the attr get
The first item in the range is often last_backfill, upon which writes
can be occuring.  It's trimmed off on the primary side anyway.

Fixes: 10150
Backport: dumpling, firefly, giant
Signed-off-by: Samuel Just <sjust@redhat.com>
2014-12-11 13:05:54 -08:00
Samuel Just
449d7d47e4 Merge pull request #2579 from yuandong1222/cache-hboject-filestore
Cache hobject filestore_key to avoid massive _reverse_nibbles calls

Reviewed-by: Samuel Just <sjust@redhat.com>
2014-12-11 13:02:22 -08:00
Samuel Just
b03fc61466 Merge pull request #3018 from tchaikov/wip-fix-safe_read_file
safe_io: do not set ending \0 in safe_read_file()

Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Loic Dachary <ldachary@redhat.com>
2014-12-11 13:00:45 -08:00
Samuel Just
81af475318 Merge pull request #3070 from ceph/wip-10080
SimpleMessenger: allow RESETSESSION whenever we forget an endpoint

Reviewed-by: Sage Weil <sage@redhat.com>
2014-12-11 12:59:36 -08:00
Gregory Farnum
a09a87ad14 Merge pull request #3150 from ceph/wip-10288
mon: fix `fs ls` on peons

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
2014-12-11 10:47:11 -08:00
Loic Dachary
c64b3c5371 Merge pull request #3152 from dachary/wip-10296-journal-discard
os: do not attempt to detect discard by default

Reviewed-by: Sage Weil <sage@redhat.com>
2014-12-11 19:15:05 +01:00
Loic Dachary
2ad55cd601 tests: do not ignore ceph-disk.sh::test_activate_dev failure
The status of the test commands must be tested instead of being
silently ignored.

Signed-off-by: Loic Dachary <ldachary@redhat.com>
2014-12-11 18:35:13 +01:00
Loic Dachary
9c5fd8c20a os: do not attempt to detect discard by default
If the journal_discard option is false, the journal should no try to
guess if the device has discard capabilities.

http://tracker.ceph.com/issues/10296 Refs: #10296

Signed-off-by: Loic Dachary <ldachary@redhat.com>
2014-12-11 18:35:06 +01:00
John Spray
5559e6aea9 mon: fix fs ls on peons
This was incorrectly using pending_mdsmap instead
of mdsmap.  We didn't notice in test because of
single-mon configurations.

Fixes: #10288
Backport: giant

Signed-off-by: John Spray <john.spray@redhat.com>
2014-12-11 14:00:57 +00:00
Loic Dachary
44a4f9578f Merge pull request #3149 from ceph/wip-typo
mds: fix log typo

Reviewed-by: Loic Dachary <ldachary@redhat.com>
2014-12-11 14:25:50 +01:00
John Spray
b025f56e72 mds: fix log typo
s/trvserse/traverse/

Signed-off-by: John Spray <john.spray@redhat.com>
2014-12-11 12:40:08 +00:00
Sage Weil
1dc064a8d4 Merge pull request #3137 from thesues/fix-strtol
Fix libstriprados::remove, use strtoll insdead of strtol

Reviewed-by: Sage Weil <sage@redhat.com>
2014-12-10 21:10:57 -08:00
Gregory Farnum
a1eb443ee5 Merge pull request #3038 from ceph/wip-mds-readonly
Wip mds readonly

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
2014-12-10 20:39:37 -08:00
Sage Weil
69c1f4869d Merge pull request #3145 from XinzeChi/wip-finisher
Finisher: call cond signal if necessary in function queue

Reviewed-by: Sage Weil <sage@redhat.com>
2014-12-10 20:34:34 -08:00
Xinze Chi
2e5ee3080b Finisher: call signal if necessary in function queue
Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
2014-12-11 04:33:34 +00:00
Loic Dachary
d3f2ec34d2 Merge pull request #3119 from ceph/wip-8797
rados.py: remove Rados.__del__(); it just causes problems

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Loic Dachary <ldachary@redhat.com>
2014-12-11 00:21:46 +01:00
Dan Mick
b038e8fbf9 Call Rados.shutdown() explicitly before exit
This is mostly a demonstration of good behavior, as the resources will
be reclaimed on exit anyway.

Signed-off-by: Dan Mick <dan.mick@redhat.com>
2014-12-10 15:09:32 -08:00
Dan Mick
5ba9b8f21f rados.py: remove Rados.__del__(); it just causes problems
Recent versions of Python contain a change to thread shutdown that
causes ceph to hang on exit; see http://bugs.python.org/issue21963.
As it turns out, this is relatively easy to avoid by not spawning
threads on exit, as Rados.__del__() will certainly do by calling
shutdown(); I suspect, but haven't proven, that the problem is
that shutdown() tries to start() a threading.Thread() that never
makes it all the way back to signal start().

Also add a PendingReleaseNote and extra doc comments to clarify.

Fixes: #8797
Signed-off-by: Dan Mick <dan.mick@redhat.com>
2014-12-10 13:19:16 -08:00
Gregory Farnum
616c46e521 Merge pull request #3129 from ceph/wip-gfcomplete-dirty
gf-complete: update submodule to newest upstream

Reviewed-by: Loic Dachary <ldachary@redhat.com>
2014-12-10 13:14:05 -08:00
Gregory Farnum
24ca9f1c25 Merge pull request #3071 from ceph/wip-10164
mds: drop dirty dentries in deleted directory

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
2014-12-10 11:04:04 -08:00
Sage Weil
f02d2d721a Merge pull request #3135 from XinzeChi/wip-3
os/FileStore.cc: insert not empty list<Context*> to op_finisher/ondisk_finisher

Reviewed-by: Sage Weil <sage@redhat.com>
2014-12-10 08:39:32 -08:00
Xinze Chi
b34e54521c os/FileStore.cc: insert not empty list<Context*> to op_finisher/ondisk_finisher.
Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
2014-12-10 17:00:43 +00:00
Sage Weil
d56eccdc7f Merge pull request #3140 from leseb/sample-ceph.conf
ceph.conf: update sample

Reviewed-by: Sage Weil <sage@redhat.com>
2014-12-10 06:43:07 -08:00
Sébastien Han
7ab4a3931c ceph.conf: update sample
This commit adds:

* client admin socket
* change some default
* re-indent

Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2014-12-10 15:32:17 +01:00
Dongmao Zhang
78a15ee4c6 Fix libstriprados::remove, use strtoll insdead of strtol
Signed-off-by: Dongmao Zhang <deanraccoon@gmail.com>
2014-12-10 18:55:55 +08:00
Loic Dachary
a3912d2308 Merge pull request #3120 from ceph/wip-10010
ceph-osd: don't close stderr if not daemonizing

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Loic Dachary <ldachary@redhat.com>
2014-12-10 11:12:45 +01:00
Yan, Zheng
6b51a9f509 mds: set dirfrag version when fetching dirfrag is skipped
Signed-off-by: Yan, Zheng <zyan@redhat.com>
2014-12-10 11:32:43 +08:00
Dan Mick
17c72f591f ceph-osd: remove extra close of stderr
Otherwise, one loses log messages when running with -f or -d.  When
daemonizing, stderr is already closed in global_init_postfork_finish.

Fixes: #10010, #10113, #9810

Signed-off-by: Dan Mick <dan.mick@redhat.com>
2014-12-09 19:28:49 -08:00
Greg Farnum
606b004d94 gf-complete: update submodule to newest upstream
Keep the tree cleaner!

Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2014-12-09 15:28:47 -08:00
Loic Dachary
cdb110539f Merge pull request #3073 from cholcombe973/get_pool_values_docs
Updating docs for Get Pool Value 

Reviewed-by: Loic Dachary <ldachary@redhat.com>
2014-12-09 11:41:17 +01:00
cholcombe973
a198deea5f Modifying the docs to add the Get pool commands to match the CLI.
Signed-off-by: Chris Holcombe <chris.holcombe@nebula.com>
2014-12-08 16:10:38 -08:00
Loic Dachary
bc2b9f6bf5 Merge pull request #3106 from danderson/asan-fixes
Asan fixes

Reviewed-by: Loic Dachary <ldachary@redhat.com>
2014-12-08 08:33:26 +01:00
Loic Dachary
7d381a60c8 Merge pull request #3113 from wonzhq/name-norm
mailmap: Zhiqiang Wang name normalization

Reviewed-by: Loic Dachary <ldachary@redhat.com>
2014-12-08 08:09:50 +01:00
Zhiqiang Wang
e8b412cbcd mailmap: Zhiqiang Wang name normalization
Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com>
2014-12-08 12:33:00 +08:00
David Anderson
12d85c64ed Fix stack buffer overflow reported by ASan.
Signed-off-by: David Anderson <dave@natulte.net>
2014-12-07 14:40:09 -08:00
David Anderson
f6f6ea2982 Fix alloc-dealloc mismatch reported by ASan (new[] vs. delete).
Signed-off-by: David Anderson <dave@natulte.net>
2014-12-07 12:02:48 -08:00
David Anderson
5bb86664b8 Fix alloc-dealloc mismatch reported by ASan (malloc vs. C++ delete).
Signed-off-by: David Anderson <dave@natulte.net>
2014-12-07 11:52:54 -08:00
Loic Dachary
6010f1784b Merge pull request #3103 from danderson/ceph-dev-mode-python-envvar
ceph: respect the PYTHON environment variable for dev mode.

Reviewed-by: Loic Dachary <ldachary@redhat.com>
2014-12-07 11:12:06 +01:00
David Anderson
812ff7aac5 ceph: respect the PYTHON environment variable for dev mode.
On OSes where `python` is python3, dev mode's re-exec makes the
ceph tool fail. The standard way to fix this is by exporting
the PYTHON envvar pointing to the python2 interpreter.

Signed-off-by: Dave Anderson <dave@natulte.net>
2014-12-07 02:02:05 -08:00
Loic Dachary
a368533e78 Merge pull request #3099 from sileht/sileht/python-rados-object-lock
python-rados: Add object lock support

Reviewed-by: Loic Dachary <ldachary@redhat.com>
2014-12-06 16:56:47 +01:00
Mehdi Abaakouk
f5bf75fa41 python-rados: Add object lock support
This change adds to the python binding the support of:
 - rados_lock_exclusive
 - rados_lock_shared
 - rados_unlock

http://tracker.ceph.com/issues/6114 Refs: #6114

Signed-off-by: Mehdi Abaakouk <sileht@redhat.com>
2014-12-06 16:28:01 +01:00
Sage Weil
53fe29611d Merge pull request #3035 from dachary/wip-10197-arch-arm
arch: add support for HW_CAP based neon runtime detection

Reviewed-by: Sage Weil <sage@redhat.com>
2014-12-05 17:31:44 -08:00
Samuel Just
1ac17c0a66 osd_types: op_queue_age_hist and fs_perf_stat should be in osd_stat_t::operator==
Fixes: 10259
Backport: giant, firefly, dumpling
Signed-off-by: Samuel Just <sjust@redhat.com>
2014-12-05 16:49:08 -08:00