RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2024-12-25 04:43:17 +00:00
Code Issues Packages Projects Releases Wiki Activity
73,660 Commits 211 Branches 394 Tags 1.2 GiB
41a90b99aa
Commit Graph

73660 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Yuri Weinstein
41a90b99aa Merge pull request #15223 from badone/wip-ceph_test_admin_socket_output 
tests: Add integration tests for admin socket output

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 2017-06-09 09:01:52 -07:00
Yuri Weinstein
b38272a7d8 Merge pull request #15573 from liewegas/wip-20208 
mon/PGMap: fix osd_epoch update when removing osd_stat

Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Loic Dachary <loic@dachary.org>
 2017-06-09 09:00:10 -07:00
Jason Dillaman
3ae72a001b Merge pull request #15594 from idryomov/wip-krbd-rados-ls 
qa: update krbd_data_pool.sh to match the new rados ls behavior

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2017-06-09 10:50:20 -04:00
Sage Weil
f3904aad0c Merge pull request #15600 from ceph/luminous 
12.0.3
 2017-06-09 09:13:59 -05:00
Sage Weil
b6a5b946f8 Merge pull request #15180 from theanalyst/doc/rgw-dev-s3tests 
doc: dev improve the s3tests doc to reflect current scripts
 2017-06-09 08:32:10 -05:00
Jos Collin
9eda2bb9ff Merge pull request #15570 from joscollin/wip-cleanup-testmsgr-warning 
test/msgr: silence warnings from -Wsign-compare

Reviewed-by: Haomai Wang <haomaiwang@gmail.com>
 2017-06-09 12:30:35 +05:30
Sage Weil
f7187559af Merge pull request #15100 from kestrels/wip-crc32c-fastzero2 
crc32c: Add ppc64le fast zero optimized assembly.

Reviewed-By: Piotr Dałek <piotr.dalek@corp.ovh.com>
 2017-06-08 20:39:59 -05:00
Sage Weil
dc3b053fd5 Merge pull request #15387 from jecluis/wip-20123 
mon/PGMap: show %used in formatted output

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-06-08 20:39:33 -05:00
Sage Weil
642fe3bc3e Merge pull request #15508 from tchaikov/wip-ceph.in-cleanup 
ceph.in: drop the compatiiblity to handle non json commands
 2017-06-08 20:39:15 -05:00
Sage Weil
6c776f5d8a Merge pull request #15560 from dmick/master 
mgr/DaemonServer.cc: log daemon type string as well as id

Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: John Spray <john.spray@redhat.com>
 2017-06-08 20:38:58 -05:00
Sage Weil
c2e346cf12 Merge pull request #15562 from tchaikov/wip-no-assert.h 
common,osdc: remove atomic_t completely

Reviewed-by: Sage Weil <sage@redhat.com>
 2017-06-08 20:38:23 -05:00
Sage Weil
6f561c8c77 Merge pull request #15567 from Fabian-Gruenbichler/pull/logrotatesync 
debian: sync logrotate packaging with downstream

Reviewed-by: Sage Weil <sage@redhat.com>
 2017-06-08 20:38:08 -05:00
Sage Weil
860d5cee28 Merge pull request #15578 from jcsp/wip-mgr-cleanup 
mgr: cleanup, stop clients sending in perf counters

Reviewed-by: Sage Weil <sage@redhat.com>
 2017-06-08 20:37:53 -05:00
Sage Weil
840eb7fa88 Merge pull request #15582 from liewegas/wip-20226 
osd: take PGRef for recovery sleep wakeup event

Reviewed-by: Gregory Farnum <gfarnum@redhat.com>
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 2017-06-08 20:37:27 -05:00
Sage Weil
1101661a2b Merge pull request #15405 from liewegas/wip-rest-test 
pybind/mgr/restful: improve cert handling; work with vstart

Reviewed-by: Boris Ranto <branto@redhat.com>
 2017-06-08 17:07:22 -05:00
Sage Weil
0f23f35210 debian/control: Build-Depends on python-pecan 
This is needed for make check (so that restful mgr module can load).

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 17:45:47 -04:00
Sage Weil
f05a34aff2 Merge pull request #15557 from liewegas/wip-mgr-respawn 
mgr/MgrStandby: respawn when deactivated

Reviewed-by: John Spray <john.spray@redhat.com>
 2017-06-08 15:45:52 -05:00
Casey Bodley
babd25fa36 Merge pull request #15301 from shashalu/rgw-ssl 
rgw: add "rgw_verify_ssl" config

Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2017-06-08 16:01:47 -04:00
Casey Bodley
734083b0ea Merge pull request #15257 from shashalu/fix-rgw-admin-retcode 
rgw: fix radosgw-admin retcode

Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2017-06-08 16:00:02 -04:00
Casey Bodley
43cad44cb3 Merge pull request #14950 from fangyuxiangGL/set-obj-accounted-size 
rgw: set object accounted size correctly

Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2017-06-08 15:59:20 -04:00
Sage Weil
30ae537a55 mgr/Mgr: fix deadlock in load_config 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:30:10 -04:00
Sage Weil
390ad9031f pybind/mgr/dashboard: get_localized_config for server_{addr,port} 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:30:10 -04:00
Sage Weil
1bf53c1fc8 qa/workunits/rest/test_mgr_rest_api.py: improvements 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:37 -04:00
Sage Weil
455318ef11 pybind/mgr/restful: store each key+pass in a separate key 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:37 -04:00
Sage Weil
de6af1b2f4 pybind/mgr/restful: prefix commands with 'restful' 
And use - instead of _, following the ceph convention.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:37 -04:00
Sage Weil
1ebc3f12aa mgr: add get_config_prefix 
Fetch a dict of all config options with a given prefix.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:37 -04:00
Sage Weil
2535c2699e pybind/mgr/restful: localize key/crt keys 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:37 -04:00
Sage Weil
c7cfbc2bd5 mgr/PyState: add get_mgr_id() to module interface 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:37 -04:00
Sage Weil
b5994f3ee0 mgr/PyModules: prefix by mgr/, not mgr/$id/ 
If modules want per-instance state, they can include
the mgr id in their portion of the key name.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:37 -04:00
Sage Weil
66e2c3f784 mon/ConfigKeyService: more useful status message 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:37 -04:00
Sage Weil
a06000312b vstart.sh: start up mgr restful API 
We try to stagger the ports that mgr modules bind to so that
concurrent vstart instances can run with consecutive ports but
the services will not collide.  Yes, this is awkward.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:37 -04:00
Sage Weil
38ef1b5a89 pybind/mgr/restful: all crt/key or filename in config-key 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:36 -04:00
Sage Weil
6bf68ce29a ceph.spec.in: add python-pecan as BuildDepends too 
For install-deps.sh's benefit.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:36 -04:00
Sage Weil
d48a29a938 vstart.sh: load 'restful' (not 'rest') mgr module 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:36 -04:00
Sage Weil
aec50b906c test/vstart_wrapper.sh: fix MGR_PYTHON_PATH 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:36 -04:00
Sage Weil
84e13070a6 vstart.sh: start mgr dashboard, echo URL(s) 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:29:36 -04:00
Yehuda Sadeh
2076175e0d Merge pull request #15507 from zhangsw/fix-rgw-logtrim-bug 
rgw: datalog trim and mdlog trim handles the result returned by osd incorrectly.

Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2017-06-08 11:26:44 -07:00
Sage Weil
b0e9f3ed51 Merge pull request #15102 from majianpeng/blockdevice-pmem 
os/bluestore/BlockDevice: support pmem device as bluestore backend

Reviewed-by: Sage Weil <sage@redhat.com>
 2017-06-08 13:15:52 -05:00
Sage Weil
a40f987120 osd: take PGRef for recovery sleep wakeup event 
The raw PG* may not remain valid when the event fires.

Fixes: http://tracker.ceph.com/issues/20226
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-08 14:08:52 -04:00
Ilya Dryomov
a3e23a7c3c qa: update krbd_data_pool.sh to match the new rados ls behavior 
See commit e9973adf1f ("PendingReleaseNotes: notes on whiteouts vs
pgnls").

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
 2017-06-08 20:01:13 +02:00
Jason Dillaman
108b38ef5f Merge pull request #15577 from jcsp/wip-dashboard-cleanup 
mgr: dashboard code cleanup

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2017-06-08 13:56:19 -04:00
John Spray
b48011d60b mgr/dashboard: remove confusing SyncObject stuff 
This was a hangover from when these wrapper classes
were borrowed from Calamari, which used these
versions/equality functions to work out when
to go fetch data from the ceph cluster.

Signed-off-by: John Spray <john.spray@redhat.com>
 2017-06-08 11:43:20 -04:00
John Spray
f175090596 mgr/dashboard: fix server view on funny version format 
Signed-off-by: John Spray <john.spray@redhat.com>
 2017-06-08 11:43:20 -04:00
John Spray
f23c68f168 mgr/dashboard: fix error handling in remote_view_cache 
Signed-off-by: John Spray <john.spray@redhat.com>
 2017-06-08 11:43:20 -04:00
John Spray
4296962edc mgr/dashboard: keep a global librados instance 
...so that classes that need one aren't creating
their own all the time.

Signed-off-by: John Spray <john.spray@redhat.com>
 2017-06-08 11:43:15 -04:00
Kefu Chai
630010fdba Merge pull request #15566 from baruza/wip-doc-20199 
doc: Change the default values of some OSD options

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-06-08 23:10:51 +08:00
Sage Weil
5bfd0b4594 Merge pull request #15485 from ceph/wip-hdd-throttle-cost 
common/config_opts.h: Lower HDD throttle cost.

Reviewed-by: Sage Weil <sage@redhat.com>
 2017-06-08 10:02:54 -05:00
Kefu Chai
d010f6a1de Merge pull request #15203 from Liuchang0812/wip-kill-broken-links 
doc: kill some broken links

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-06-08 22:59:11 +08:00
Fabian Grünbichler
6ce28c406a debian: sync logrotate packaging with downstream 
both Ubuntu and Debian put the logrotate script into
ceph-common, to ensure that radosgw logs are rotated as
well.

to prevent duplicate logrotate scripts handling the same log
files, and to minimize the delta between upstream and
downstream packaging, sync this change back upstream.

Fixes: http://tracker.ceph.com/issues/19938

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
 2017-06-08 16:12:01 +02:00
John Spray
040e8cf306 mgr: drop PyModules::lock around module shutdown calls 
This deadlocked if the module's shutdown call was trying
to take a pythonland lock that was held by a thread
that was in turn trying to call into C++ land and take
PyModules::lock.

Signed-off-by: John Spray <john.spray@redhat.com>
 2017-06-08 10:03:55 -04:00