RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2025-01-02 00:52:22 +00:00
Code Issues Packages Projects Releases Wiki Activity
73,899 Commits 212 Branches 394 Tags 1.5 GiB
5daf94ce72
Commit Graph

73899 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Casey Bodley
9d7b477c70 Merge pull request #15613 from cbodley/wip-20244 
rgw: only log metadata on metadata master zone

Reviewed-by: Orit Wasserman <owasserm@redhat.com>
 2017-06-13 10:37:27 -04:00
Casey Bodley
40d8bc7324 Merge pull request #15280 from ZVampirEM77/wip-bucket-policy-cleanup 
rgw: fix get bucket policy s3 compatible issue

Reviewed-by: Adam C. Emerson <aemerson@redhat.com>
 2017-06-13 10:27:47 -04:00
Kefu Chai
85293e7633 Merge pull request #15663 from zealoussnow/wip-fix-build 
cmake: exclude *.css while generating ctags

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-06-13 21:07:24 +08:00
Matt Benjamin
1031f90219 Merge pull request #15655 from guihecheng/rgw_file-fix-mkdir 
rgw_file: prevent conflict of mkdir between restarts
 2017-06-13 08:40:48 -04:00
Igor Fedotov
697dea6086 Merge pull request #15659 from lixiaoy1/fio_ceph 
test/fio: Fix assert in set_cache_shards in bluestore fio
 2017-06-13 16:24:50 +04:00
lisali
21fc54b11b Fix assert in set_cache_shards in bluestore fio 
The config osd_op_num_shards is changed so it needs to change in
fio_ceph_objectstore.cc.

Signed-off-by: Xiaoyan Li <xiaoyan.li@intel.com>
 2017-06-13 12:41:04 -04:00
Mykola Golub
b7d657d45e Merge pull request #15593 from chenfangxian/rbd-nbd-fix-unmap 
rbd-nbd: ensure unmap returns error code

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
Reviewed-by: Pan Liu <wanjun.lp@alibaba-inc.com>
 2017-06-13 15:10:25 +03:00
Leo Zhang
3c1eb63cb9 doc: remove some non-existent and fix the default value according to source code 
Signed-off-by: Leo Zhang <nguzcf@gmail.com>
 2017-06-13 19:57:07 +08:00
Leo Zhang
7681fd8e4a build: add css file to excludes while generate tags 
Signed-off-by: Leo Zhang <nguzcf@gmail.com>
 2017-06-13 19:43:54 +08:00
Jing Wenjun
ce88212582 rgw: fix the signature mismatch of FormPost in swift API 
Fixes: http://tracker.ceph.com/issues/20220
Signed-off-by: Jing Wenjun <jingwenjun@cmss.chinamobile.com>
 2017-06-13 18:05:52 +08:00
Kefu Chai
b7f59b6437 qa/suites/upgrade: remove duplicated upgrade task 
Signed-off-by: Kefu Chai <kchai@redhat.com>
 2017-06-13 17:00:06 +08:00
Kefu Chai
3734280522 qa/suites/upgrade: set "sortbitwise" for jewel clusters 
so ceph.healthy or wait-for-healthy won't be blocked by this warning.

Signed-off-by: Kefu Chai <kchai@redhat.com>
 2017-06-13 17:00:00 +08:00
Gui Hecheng
87cec5ff9a rgw_file: prevent conflict of mkdir between restarts 
Fixes: http://tracker.ceph.com/issues/20275

Signed-off-by: Gui Hecheng <guihecheng@cmss.chinamobile.com>
 2017-06-13 15:28:37 +08:00
Gregory Farnum
52f0deb9a3 Merge pull request #15648 from joscollin/wip-CID1412575-incorrect-expression 
mon: Incorrect expression in PGMap::get_health()

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
 2017-06-13 00:20:18 -04:00
Kefu Chai
3048d94ee3 Merge pull request #15638 from liewegas/wip-19948 
ceph_test_rados_api_*: wait for snap trim on ENOENT during cleanup

Reviewed-by: Brad Hubbard <bhubbard@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-06-13 11:40:53 +08:00
chenfangxian
890adc605b rbd-nbd: fix nbd do_unmap error handling 
when nbd disconnect failed, ensure do_unmap return error logic.

Signed-off-by: chenfangxian <chenfangxian@cmss.chinamobile.com>
Signed-off-by: guojiannan <guojiannan@cmss.chinamobile.com>
 2017-06-13 11:05:51 +08:00
Sage Weil
aeb402988a Merge pull request #15519 from liewegas/wip-19595 
osd/PrimaryLogPG: clear oi from trim_object()

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 2017-06-12 22:01:05 -05:00
Jos Collin
7c2ab49cdf mon: Incorrect expression in PGMap::get_health() 
Fixes:

CID 1412575:  Incorrect expression  (COPY_PASTE_ERROR)
ceph/src/mon/PGMap.cc: 2764 in PGMap::get_health()
"num_err" in "ss2 << num_err" looks like a copy-paste error.

Signed-off-by: Jos Collin <jcollin@redhat.com>
 2017-06-13 08:24:23 +05:30
Haomai Wang
6d23ae59c1 Merge pull request #15629 from Linbing/wip-msg-async-op-threads 
msg/async: add assert of ms_async_op_threads > 0

Reviewed-by: Haomai Wang <haomai@xsky.com>
 2017-06-13 10:26:35 +08:00
Haomai Wang
bf98babb32 msg/async: fix accept_conn not remove entry in conns when lazy delete 
Signed-off-by: Haomai Wang <haomai@xsky.com>
 2017-06-13 10:16:47 +08:00
Yehuda Sadeh
7ad2ae27a4 Merge pull request #15426 from zhangsw/fix-rgw-gc-loglevel 
rgw: lower some log's level in gc process.

Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2017-06-12 13:07:43 -07:00
Casey Bodley
78c6e6e8d0 Merge pull request #15571 from ceph/wip-fix-fio-shard-hint 
test/fio: fix lack of setting for Sequencer::shard_hint

Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2017-06-12 16:03:21 -04:00
Sage Weil
ad5a4363cc Merge pull request #15603 from idryomov/wip-rados-df-typo 
rados: fix typo in 'df' column name
 2017-06-12 14:13:21 -05:00
Sage Weil
c7f7500b88 Merge pull request #15615 from joscollin/wip-CID717210-Uninitialized-members 
common: initialize _hash in LogEntryKey()

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
 2017-06-12 14:12:40 -05:00
Sage Weil
04ab255175 Merge pull request #15627 from theanalyst/12.1.0/release-notes-2 
doc: 12.1.0/release notes 2
 2017-06-12 14:06:40 -05:00
Sage Weil
4302237637 ceph_test_rados_api_*: wait for snap trim on ENOENT during cleanup 
Assume that an ENOENT removing an object during cleanup means the
head is a whiteout because there are clones.  When that happens
just sleep for a bit and then retry in the hopes that snap
trimming has happened.

Time out (and fail) after retrying for 10 minutes' worth of tries.

Fixes: http://tracker.ceph.com/issues/19948
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-12 14:57:22 -04:00
Sage Weil
cd1271b432 osd/PrimaryLogPG: clear oi from trim_object() 
trim_object() is unique in that its resulting OpContxtUPtr is fed to
simple_opc_submit() but does not first pass through finish_ctx().  Almost
none of what finish_ctx is helpful for us because we are the only place
where we delete a clone, and for that reason it is not terribly useful to
put what we do into finish_ctx.  However, *one* thing that finish_ctx()
does do that we don't (didn't) is clear out the object_info_t io on
delete.  This would leave things like version with a value that would
bleed into future instantiations of the object and break things like
the prior_version in pg_log_entry_t's and, much later, break asserts
when loading the missing set off of disk.

Fix by clearing oi.

Fixes: http://tracker.ceph.com/issues/19947
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-12 14:05:41 -04:00
Sage Weil
b98fad9b1c Merge pull request #15604 from liewegas/wip-rest-test-qa 
qa/suites/rados/rest: test restful mgr module

Reviewed-by: Boris Ranto <branto@redhat.com>
 2017-06-12 12:47:32 -05:00
Sage Weil
be88220965 qa/suites/rados/rest/mgr-restful: test on centos 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-12 13:46:58 -04:00
Sage Weil
0c2e02e4c8 Merge pull request #15584 from liewegas/wip-20219 
mon/MgrStatMonitor: keep mgrstat version ahead of pgmon

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
 2017-06-12 11:42:39 -05:00
Kefu Chai
ed249becf0 Merge pull request #15620 from tchaikov/wip-20249 
test/vstart_wrapper.sh: display_log on test failure

Reviewed-by: Loic Dachary <ldachary@redhat.com>
Reviewed-by: Sage Weil <sage@redhat.com>
 2017-06-13 00:19:11 +08:00
Sage Weil
7df54f2f24 qa/workunits/rest/test_mgr_rest_api.py: tolerate old distros 
Work with older requests and/or missing urllib3.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-12 11:56:16 -04:00
Sage Weil
408e78e5f2 qa/suites/rados/rest/mgr-restful: test restful module 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-12 11:55:10 -04:00
Kefu Chai
efb1c30174 Merge pull request #15634 from liewegas/wip-pyformatter 
mgr/PyFormatter: implement dump_format_va

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-06-12 23:10:40 +08:00
Sage Weil
e56f8273e9 Merge pull request #15580 from liewegas/wip-bluestore-memory 
os/bluestore: configure rocksdb cache via bluestore_cache_kv_ratio

Reviewed-by: Mark Nelson <mnelson@redhat.com>
 2017-06-12 09:57:07 -05:00
Sage Weil
86c851e138 mgr/PyFormatter: implement dump_format_va 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-12 09:57:00 -04:00
Haomai Wang
6d0b4ce160 Merge pull request #15624 from Adirl/wip 
msg/async/rdma: check if fin message completed

Reviewed-by: Haomai Wang <haomai@xsky.com>
 2017-06-12 21:56:49 +08:00
Adir Lev
f4af675f68 msg/async/rdma: check if fin message completed 
Change-Id: I9cb1e523be7b21127503a9252b09331a37bea204
Signed-off-by: Alexander Mikheev <alexm@mellanox.com>
Signed-off-by: Adir Lev <adirl@mellanox.com>
 2017-06-12 13:52:17 +00:00
Sage Weil
5325460b39 mon/PGMonitor: assert PGMonitor stops proposing after luminous 
One hacky exception here for the rebuild-mondb case.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-12 09:48:48 -04:00
Sage Weil
62703782ce mon/MgrStatMonitor: keep mgrstat version ahead of pgmon 
Pre-luminous, statfs and getpoolstats come from pgmon, and clients will
see and track the paxos version for that.  After the switchover, those
move to mgrstatmon.  If a clients sends a version ahead of the current
mgrstat version, it will block until the paxos version catches up, which
may be years for an aged cluster.

Fix by keeping mgrstat version ahead of pgmon's every time it updates.
This only really matters right at the transition point (we don't update
before that, and after that pgmon doesn't update).

Fixes: http://tracker.ceph.com/issues/20219
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-12 09:48:16 -04:00
Loic Dachary
93c29f1a43 Merge pull request #14753 from theanalyst/v12.0.2/mailmap 
doc: mailmap for v12.0.2 

Reviewed-by: Loic Dachary <ldachary@redhat.com>
 2017-06-12 15:44:53 +02:00
Radoslaw Zarzynski
f51771b517 Merge pull request #15328 from jimifm/wip-RGWPutObj 
rgw/rgw_op:Prevents memory leaks when calling func swift_versioning_copy() fails

Reviewed-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
 2017-06-12 14:50:45 +02:00
Radoslaw Zarzynski
fe9fedf12e Merge pull request #15323 from fangyuxiangGL/compress_bug_fix 
rgw: get wrong content when download object with specific range with compression

Reviewed-by: Adam Kupczyk <akupczyk@mirantis.com>
Reviewed-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
 2017-06-12 14:47:38 +02:00
linbing
db20e853be msg/async: add assert of ms_async_op_threads > 0 
Signed-off-by: linbing <linbing@t2cloud.net>
 2017-06-12 19:38:37 +08:00
Kefu Chai
63840ffaba qa: timeout if flush_pg_stats() takes too long 
a "timeout" which defaults to 300 seconds is added to flush_pg_stats()

Signed-off-by: Kefu Chai <kchai@redhat.com>
 2017-06-12 19:32:11 +08:00
Abhishek Lekshmanan
ba88ca7f69 mailmap: Yunfei Guan affiliation 
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
 2017-06-12 13:16:56 +02:00
Abhishek Lekshmanan
0c5cd613c0 mailmap: Andrey Parfenov affiliation 
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
 2017-06-12 13:16:56 +02:00
Abhishek Lekshmanan
2ee666c72e mailmap: Yao Zongyou affiliation 
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
 2017-06-12 13:16:56 +02:00
Abhishek Lekshmanan
ef96856f5e mailmap: Patrick Dinnen affiliation 
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
 2017-06-12 13:16:56 +02:00
Abhishek Lekshmanan
8fca92b281 mailmap: Henrik Korkuc affiliation 
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
 2017-06-12 13:16:48 +02:00