RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2024-12-27 14:03:25 +00:00
Code Issues Packages Projects Releases Wiki Activity
32,574 Commits 211 Branches 394 Tags 1.4 GiB
1bd575e223
Commit Graph

32574 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Sage Weil
b742f94e4e Merge pull request #1515 from ceph/wip-rgw-less-alloc 
Wip rgw less alloc

Reviewed-by: Sage Weil <sage@inktank.com>
 2014-03-21 16:32:27 -07:00
Sage Weil
185bcc42d5 doc/release-notes: final 0.78 
Signed-off-by: Sage Weil <sage@inktank.com>
 2014-03-21 15:43:39 -07:00
Yehuda Sadeh
1817c23c2c rgw: get rid of a memory allocation 
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
 2014-03-21 15:16:23 -07:00
Yehuda Sadeh
1e7cd10d39 rgw: remove memory allocation 
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
 2014-03-21 15:16:23 -07:00
Jenkins
f6c746c314 0.78 2014-03-21 22:05:12 +00:00
Sage Weil
e4e8dc5081 Merge pull request #1512 from dachary/wip-jerasure-sse 
erasure-code: gf-complete detects SSE at runtime

Reviewed-by: Sage Weil <sage@inktank.com>
 2014-03-21 14:51:30 -07:00
Loic Dachary
fed2ed1acc Merge pull request #1513 from ceph/wip-7438 
Reviewed-by: Loic Dachary <loic@dachary.org>
 2014-03-21 22:06:30 +01:00
Samuel Just
a8f59a383a Merge pull request #1510 from ceph/wip-pq 
osd: some PrioritizedQueue cleanups; dump locking fix; debug pqueue state at high debug level

Reviewed-by: Samuel Just <sam.just@inktank.com>
 2014-03-21 13:39:50 -07:00
Sage Weil
28d8e7f567 Revert "ReplicatedPG: disallow trans which atomically create and remove an object" 
This reverts commit 9ee1084281.

This appears to break rgw in its current state; test and fix soon.
 2014-03-21 11:57:03 -07:00
Sage Weil
49a0190068 doc/release-notes: 0.78 notes 
Signed-off-by: Sage Weil <sage@inktank.com>
 2014-03-21 08:59:10 -07:00
Sage Weil
8012ba4313 Merge remote-tracking branch 'gh/wip-multimds-fixes' 
Reviewed-by: Sage Weil <sage@inktank.com>
 2014-03-21 08:21:21 -07:00
Loic Dachary
dbcf447183 erasure-code: gf-complete detects SSE at runtime 
And the gf-complete code has been updated to rely on the
gf_internal_t::sse flag instead of #if conditionals.

Signed-off-by: Loic Dachary <loic@dachary.org>
 2014-03-21 14:56:30 +01:00
Loic Dachary
8c7f6c11d4 autotools: AX_SSE detects the compiler SSE capabilities 
For each SSE feature supported by the compiler

* add the corresponding -msse* flag
* define HAVE_SSE*

Remove AX_EXT because it decides based on the CPU capabilities of the
machine compiling the binary which may or may not be the one running
them.

Signed-off-by: Loic Dachary <loic@dachary.org>
 2014-03-21 13:07:32 +01:00
David Zafman
5a3f6c7c8a test: Add erasure coding to stress watch test 
Fixes: #7438

Signed-off-by: David Zafman <david.zafman@inktank.com>
 2014-03-20 23:12:47 -07:00
David Zafman
6fb65882a9 test: Reorg multi_stress_watch to prepare for ec option 
If old options are specified maintain backward compatibility

Signed-off-by: David Zafman <david.zafman@inktank.com>
 2014-03-20 23:10:14 -07:00
David Zafman
b110275de2 test: Fix ceph_filestore_dump.sh test for new EC pool creation 
Signed-off-by: David Zafman <david.zafman@inktank.com>
 2014-03-20 20:22:30 -07:00
Sage Weil
bcc254fe50 Merge pull request #1500 from ceph/wip-tid_t 
Change tid_t to ceph_tid_t to avoid conflicts with a popular type

Reviewed-by: Sage Weil <sage@inktank.com>
 2014-03-20 16:07:29 -07:00
Sage Weil
c9b4e7d9be Merge pull request #1465 from ceph/wip-s3pm-wusui 
Remove code duplication from s3 tests.
 2014-03-20 16:05:48 -07:00
Sage Weil
f13c6421a6 Merge pull request #1511 from dachary/wip-exclamation 
tests: use ^ instead of ! as invalid char

Reviewed-by: Sage Weil <sage@inktank.com>
 2014-03-20 16:04:59 -07:00
Loic Dachary
dad0faf1a1 tests: use ^ instead of ! as invalid char 
Because !!! is incorrectly interpreted as an error by gitbuilder.

Signed-off-by: Loic Dachary <loic@dachary.org>
 2014-03-20 23:29:29 +01:00
Sage Weil
d4d77d71d0 doc/release-notes: stop confusing sphinx 
ERROR: /srv/autobuild-ceph/gitbuilder.git/build/doc/release-notes.rst:594: Unknown target name: "debug".

Signed-off-by: Sage Weil <sage@inktank.com>
 2014-03-20 14:43:13 -07:00
Sage Weil
f50d208cc5 Merge pull request #1503 from dachary/wip-jerasure-submodules 
erasure-code: add gf-complete / jerasure submodules

Reviewed-by: Sage Weil <sage@inktank.com>
 2014-03-20 14:11:44 -07:00
Samuel Just
5158f1544f Merge pull request #1505 from ceph/wip-7728 
Wip 7728

Reviewed-by: Sage Weil <sage@inktank.com>
 2014-03-20 14:05:47 -07:00
Sage Weil
513e4ad8ec Merge remote-tracking branch 'gh/firefly' 2014-03-20 10:42:26 -07:00
Samuel Just
048ca96c4d Merge pull request #1509 from guangyy/wip-folder-merge 
Make it possible for FileStore to be in a state which never do merge and keep it backward compatible

Reviewed-by: Samuel Just <sam.just@inktank.com>
 2014-03-20 10:13:05 -07:00
Sage Weil
d1f8742db1 Merge pull request #1506 from ceph/wip-7755 
PG::find_best_info: fix log_tail component

Reviewed-by: Sage Weil <sage@inktank.com>
 2014-03-20 09:24:26 -07:00
Sage Weil
b395cb1951 Merge pull request #1507 from ceph/wip-7733 
PG::start_peering_interval: always send_notify if !primary

Reviewed-by: Sage Weil <sage@inktank.com>
 2014-03-20 09:19:58 -07:00
Sage Weil
bf68dabaaa Merge pull request #1508 from ceph/wip-7777 
ReplicatedPG::do_op: delay if snapdir is unreadable

Reviewed-by: Sage Weil <sage@inktank.com>
 2014-03-20 09:19:10 -07:00
Josh Durgin
cffcf10ff6 Merge pull request #1312 from cernceph/master 
rados.py: add aio_remove

Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
 2014-03-19 11:00:58 -07:00
Josh Durgin
c647a03fff Merge pull request #1504 from ceph/wip-7780 
objecter: waive OSDMAP_FULL check for MDS

Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
 2014-03-19 10:47:05 -07:00
Samuel Just
72093f97a9 Merge pull request #1476 from ceph/wip-scrub-lock 
osd: do not queue scrub for non-primary

Reviewed-by: Samuel Just <sam.just@inktank.com>
 2014-03-19 10:46:32 -07:00
Josh Durgin
cb744ca382 Merge pull request #1502 from ceph/wip-flush-journal-asok 
wip-flush-journal-asok

Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
 2014-03-19 10:40:47 -07:00
John Spray
78ede90c4e objecter: waive OSDMAP_FULL check for MDS 
The MDS expects to be able to perform writes to OSDs even
if the full ratio has been reached, in order to journal
file deletions to free space.

Fixes: #7780

Signed-off-by: John Spray <john.spray@inktank.com>
 2014-03-19 16:22:37 +00:00
Ilya Dryomov
a4849fb914 qa: test_alloc_hint: flush journal before prodding the FS 
OSDs that for some reason get behind on processing their op queue break
expect_alloc_hint_eq(), as it pokes the FS and not the journal.  Fix it
by flushing the journal before proceeding with anything else.

Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
 2014-03-19 10:59:27 +02:00
Ilya Dryomov
9d31c1be1f osd: add flush_journal admin socket command 
Add flush_journal admin socket command to be able to flush journal to
the permanent store for online osds.  (For offline osds we already have
ceph-osd --flush-journal.)

Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
 2014-03-19 10:59:26 +02:00
Yan, Zheng
b444e88650 mds: fix typo in MDCache::handle_cache_rejoin_strong 
It should invent dirfrags according to the dirfragtree

Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
 2014-03-19 16:25:26 +08:00
Loic Dachary
ddbb2f7b0a erasure-code: add gf-complete / jerasure submodules 
Instead of copying the files in the ceph repository, which is less
convenient.

When building the headers are ignored, even though they do
not exist. When creating the tarbal with make dist, it fails because
they cannot be found. I misread src/gf_int.h to be include/gf_int.h and
wrongfully thought the submodules were to blame. This is why they were
removed shortly after being added.

Signed-off-by: Loic Dachary <loic@dachary.org>
 2014-03-19 08:02:12 +01:00
Loic Dachary
5c34a0f01a erasure-code: remove copy of gf-complete / jerasure 
Signed-off-by: Loic Dachary <loic@dachary.org>
 2014-03-19 07:38:41 +01:00
Yan, Zheng
0d167d2cd0 mds: fix NULL pointer dereference in Server::handle_client_rename 
Inode's parent dentry can be NULL, which causes segment fault in
Locker::acquire_locks

Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
 2014-03-19 14:30:54 +08:00
Yan, Zheng
272b53b22c mds: avoid infinite loop in MDLog::submit_entry() 
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
 2014-03-19 14:26:03 +08:00
Yan, Zheng
002c7e7544 Merge pull request #1496 from ceph/wip-7708 
mds/Locker: fix null deref on cap import
 2014-03-19 12:25:05 +08:00
Sage Weil
328941383f Merge pull request #1499 from ceph/wip-no-version 
Add NO_VERSION to avoid rebuilding ceph_ver.h and relinking

Reviewed-by: Sage Weil <sage@inktank.com>
 2014-03-18 21:03:01 -07:00
Sage Weil
5dd270e862 Merge pull request #1498 from ceph/wip-limit-libkeyutils 
Makefiles: remove libkeyutils from every binary except two

Reviewed-by: Sage Weil <sage@inktank.com>
 2014-03-18 21:01:13 -07:00
Yan, Zheng
3cadbfbb11 mds: fix potential invalid pointer dereference 
pin importing inode until subtree importing finishes

Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
 2014-03-19 11:35:57 +08:00
Yan, Zheng
91c88c1ea1 mds: rdlock dirfragtree lock when renaming directory 
This make sure inode importer and inode exporter have consistent
view of dirfrags.

Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
 2014-03-19 11:35:56 +08:00
Yan, Zheng
a86716660c mds: don't mark scatter locks dirty when dirfrag is dirty 
The journal reply code has check that decides which scatter locks
should be marked as dirty. So don't unconditionally mark scatter
locks dirty when dirfrag is dirty

Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
 2014-03-19 11:35:56 +08:00
Yan, Zheng
5150fe63d8 mds: trim other MDS' mdsdir during recovery 
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
 2014-03-19 11:35:56 +08:00
Yan, Zheng
150c6ce454 mds: remove wrong assertion 
when handling resolve message, it's possible that bound dirfrags
are missing.

Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
 2014-03-19 11:35:56 +08:00
Yan, Zheng
de461a9266 fragtree_t: fix get_leaves_under() 
If fragtree is (*^1) and the caller wants leaves under frag 00*.
get_leaves_under() should return empty list.

Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
 2014-03-19 11:35:56 +08:00
Yan, Zheng
cbd1c44759 mds: improve success rate of subtree exporting 
When exporting a subtree, the migrator acquires the required locks,
then freezes the subtree and releases the locks. After subtree is
frozen, it try acquiring the same locks again.

This patch make scatter locks keep in their old states if inode has
exporting dirfrag. It improves the chance that migrator acquires all
required locks when subtree is frozen.

Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
 2014-03-19 11:35:56 +08:00