Commit Graph

73894 Commits

Author SHA1 Message Date
Sage Weil
de6b679faf Merge pull request #15309 from ifed01/wip-bluestore-boost-statfs
os/bluestore: keep statfs replica in RAM to avoid expensive KV retrieval

Reviewed-by: Sage Weil <sage@redhat.com>
2017-06-08 08:52:34 -05:00
Sage Weil
e9973adf1f PendingReleaseNotes: notes on whiteouts vs pgnls
Signed-off-by: Sage Weil <sage@redhat.com>
2017-06-08 09:41:25 -04:00
Jason Dillaman
c884cc01d9 pybind/rbd: OSError should be picklable
Fixes: http://tracker.ceph.com/issues/20223
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2017-06-08 09:30:31 -04:00
Sage Weil
a883bac077 mon/PGMap: erase osd_epochs entry too when we erase osd_stat entry
Signed-off-by: Sage Weil <sage@redhat.com>
2017-06-08 09:18:05 -04:00
Sage Weil
c225899139 mgr/ClusterState: dump pgmap and inc at dout 30
Signed-off-by: Sage Weil <sage@redhat.com>
2017-06-08 09:17:43 -04:00
Mykola Golub
578468f38c Merge pull request #15532 from dillaman/wip-rbd-mirror-stress
rbd-mirror: lock loss during sync should wait for in-flight copies

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
2017-06-08 14:42:29 +03:00
Mykola Golub
eac34cc5e2 Merge pull request #15518 from dillaman/wip-rbd-missing-default-pool
rbd: removed hardcoded default pool

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
2017-06-08 14:41:51 +03:00
Igor Fedotov
e081091a65 os/bluestore: introduce a debug switch to bypass kv update.
Signed-off-by: Igor Fedotov <ifedotov@mirantis.com>
2017-06-08 04:38:47 -07:00
Igor Fedotov
07761f2a3a os/bluestore: introduce debug parameter to bypass bdev verification
E.g. this allows to backup WAL/DB volumes after preconditioning and quickly rollback to that state when needed.

Signed-off-by: Igor Fedotov <ifedotov@mirantis.com>
2017-06-08 04:38:47 -07:00
Igor Fedotov
b14ced7bce os/bluestore: keep statfs replica in memory to avoid expensive KV access
Signed-off-by: Igor Fedotov <ifedotov@mirantis.com>
2017-06-08 04:38:47 -07:00
Igor Fedotov
2e47f36a11 os/bluestore: move volatile_statfs struct out of TransContext
Signed-off-by: Igor Fedotov <ifedotov@mirantis.com>
2017-06-08 04:37:51 -07:00
Igor Fedotov
41305afeb5 test/fio: fix lack of setting for Sequencer::shard_hint
Signed-off-by: Igor Fedotov <ifedotov@mirantis.com>
2017-06-08 04:35:22 -07:00
Jos Collin
e64b4c864c test/msgr: silence warnings from -Wsign-compare
The following warning appears during make in the latest code:

ceph/src/test/msgr/test_msgr.cc:1474:46: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
     CHECK_AND_WAIT_TRUE(srv_dispatcher.count > last + 1);
                         ~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~
ceph/src/test/msgr/test_msgr.cc:54:9: note: in definition of macro ‘CHECK_AND_WAIT_TRUE’
     if (expr)                           \
         ^~~~
ceph/src/test/msgr/test_msgr.cc:1475:30: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
     if (srv_dispatcher.count == last) {

Signed-off-by: Jos Collin <jcollin@redhat.com>
2017-06-08 16:48:24 +05:30
Loic Dachary
40e84a131d Merge pull request #15433 from wjwithagen/wip-wjw-prettytable
ceph-disk: add setting for external py-modules for tox-testing

Reviewed-by: Loic Dachary <ldachary@redhat.com>
2017-06-08 12:52:31 +02:00
Gui Hecheng
449b8b7fc5 rgw_file: use predefined format_child_name to replace raw string concatenation
Signed-off-by: Gui Hecheng <guihecheng@cmss.chinamobile.com>
2017-06-08 18:25:50 +08:00
Barbora Ančincová
706de54d63 doc: Change the default values of some OSD options
These values were changed to reflect this change a3f9cba

Fixes: http://tracker.ceph.com/issues/20199
Signed-off-by: Bara Ancincova <bara@redhat.com>
2017-06-08 11:49:37 +02:00
Willem Jan Withagen
91d36b64f3 ceph-disk: use system modules if needed
- prettytable usage was introduced in:
	https://github.com/ceph/ceph/commit/3fa8bb1
   It is in the install-deps.sh file to be installed, but
   it is not per default installed in the testenvironment
   So allow tox tests to use "external" modules

Signed-off-by: Willem Jan Withagen <wjw@digiware.nl>
2017-06-08 11:33:04 +02:00
Gui Hecheng
0e2223cffc rgw_file: pass bucketname instead of uri for bucket create/delete
Signed-off-by: Gui Hecheng <guihecheng@cmss.chinamobile.com>
2017-06-08 15:05:21 +08:00
Kefu Chai
2b427198de tools/rbd_mirror: do not include ununsed header
Signed-off-by: Kefu Chai <kchai@redhat.com>
2017-06-08 12:23:01 +08:00
Kefu Chai
5e06938bc1 client: s/atomic_t/std::atomic<>
Signed-off-by: Kefu Chai <kchai@redhat.com>
2017-06-08 12:21:00 +08:00
Kefu Chai
a87fa37e46 cmake,debian,rpm: remove atomic_t completely
Signed-off-by: Kefu Chai <kchai@redhat.com>
2017-06-08 12:21:00 +08:00
Kefu Chai
e702564538 msg,client,os,journal: do not include unused header
Signed-off-by: Kefu Chai <kchai@redhat.com>
2017-06-08 12:20:56 +08:00
Kefu Chai
f6d6759983 msg/async: s/atomic_t/atomic<>/
Signed-off-by: Kefu Chai <kchai@redhat.com>
2017-06-08 11:51:06 +08:00
Kefu Chai
a3a958fa43 rgw: rgw_rados: s/atomic_t/atomic<>/
Signed-off-by: Kefu Chai <kchai@redhat.com>
2017-06-08 11:46:26 +08:00
Kefu Chai
28c8373734 msg,common: do not include assert.h everywhere to clobber <assert.h>
* reverts c93daaf92b
* add "#include "include/assert.h" to where <cassert> or <assert.h> are
  included, or avoid including them if possible.

Signed-off-by: Kefu Chai <kchai@redhat.com>
2017-06-08 11:46:26 +08:00
Kefu Chai
161ea38fd5 osdc/Objecter: s/atomic_t/std::atomic<>/
Signed-off-by: Kefu Chai <kchai@redhat.com>
2017-06-08 11:46:26 +08:00
Kefu Chai
fd48d550c0 common/ceph_context: s/atomic_t/std::atomic<>/
Signed-off-by: Kefu Chai <kchai@redhat.com>
2017-06-08 11:23:13 +08:00
Dan Mick
0384acea9d mgr/DaemonServer.cc: log daemon type string as well as id
Signed-off-by: Dan Mick <dan.mick@redhat.com>
2017-06-07 19:52:13 -07:00
Kefu Chai
c7034595b8 Merge pull request #14866 from chardan/jfw-wip-halflife_atomic_t-varan
common,test: migrate atomic_t to std::atomic

Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-06-08 09:57:26 +08:00
linbing
1eae74ec64 osd: when osd is not in failure_pending, we don't need to get osd inst from osdmap.
Signed-off-by: linbing <linbing@t2cloud.net>
2017-06-08 08:43:47 +08:00
Sage Weil
8a0d6720d9 Merge pull request #15520 from liewegas/wip-max-object-size
osd: reduce rados_max_object_size from 100 GB -> 128 MB

Reviewed-by: Gregory Farnum <gfarnum@redhat.com>
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
Reviewed-by: John Spray <john.spray@redhat.com>
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
2017-06-07 17:08:32 -05:00
Sage Weil
ca3d9e6a70 osd: reduce rados_max_object_size from 100 GB -> 128 MB
I'm not sure why we had this so huge!  128 MB is a pretty reasonable size.

Signed-off-by: Sage Weil <sage@redhat.com>
2017-06-07 18:07:23 -04:00
Sage Weil
8ded50ed62 Merge pull request #15230 from chardan/jfw-wip-halflife_atomic_t-xio
xio: migrate atomic_t to std::atomic<>

Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-06-07 16:00:30 -05:00
Sage Weil
c011314b77 Merge pull request #15516 from javacruft/i386-cmake-fixes
cmake: misc fixes for build on i386

Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-06-07 16:00:10 -05:00
Sage Weil
91c5af152e Merge pull request #15525 from wjwithagen/wip-wjw-clang-testosdmap
test: osd/TestOSDMap.cc: fix Clang complain about promotion

Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-06-07 15:59:54 -05:00
Sage Weil
2fd49a1a61 mgr/MgrStandby: respawn when deactivated
- It is ugly to unwind all of the Mgr state so that we can reactivate
  later.
- It is perhaps impossible to do shut down the python state reliably.
- Respawning provides a clean state and is reliable.

This mostly just copies MDSServer::respawn().

Fixes: http://tracker.ceph.com/issues/19595
Fixes: http://tracker.ceph.com/issues/19549
Signed-off-by: Sage Weil <sage@redhat.com>
2017-06-07 16:54:19 -04:00
Jason Dillaman
e1cfaedc43 rbd: removed hardcoded default rbd pool name
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2017-06-07 16:26:25 -04:00
Jason Dillaman
708d7bb0c5 rbd-nbd: removed default rbd pool name
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2017-06-07 16:26:24 -04:00
Jason Dillaman
c6cd5a9418 rbd-replay: removed default rbd pool name
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2017-06-07 16:26:24 -04:00
Jason Dillaman
d9a9a07c44 pybind/rbd: module should depend on librbd not rbd CLI
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2017-06-07 16:26:24 -04:00
Jason Dillaman
af615a5933 common: new 'rbd_default_pool' configuration option
The 'rbd' pool will no longer be automatically created. Allow
the user to specify a custom default RBD pool name.

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2017-06-07 16:26:22 -04:00
Jason Dillaman
21d19325f3 rbd-mirror: narrowed scope of exclusive lock op tracking
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2017-06-07 16:09:10 -04:00
Jason Dillaman
02f9ea4b3f rbd-mirror: partial revert of 649b29d
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2017-06-07 15:33:51 -04:00
Daniel Gryniewicz
4327d1bd2d Document RGW quota cache options
Quota cache options were undocumented.

Fixes: http://tracker.ceph.com/issues/18747
Signed-off-by: Daniel Gryniewicz <dang@redhat.com>
2017-06-07 14:47:04 -04:00
Adam C. Emerson
7059b9b201 Merge pull request #15498 from adamemerson/wip-fixation
test: Fix narrowing conversion
rgw: Support certain archaic and antiquated distributions

Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-06-07 13:29:28 -04:00
Yuri Weinstein
17376dbff6 Merge pull request #15436 from liewegas/wip-versions
mon: {mon,osd,mds} {versions,count-metadata}

Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-06-07 09:36:59 -07:00
Sage Weil
2cbcec1876 Merge pull request #15501 from liewegas/wip-blocked-is-err
mon/PGMap: call blocked requests ERR not WARN

Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
2017-06-07 11:17:39 -05:00
Sage Weil
7cacb701ad mon/PGMap: call requests blocked for 128x as long ERR not WARN
- rename the option (max -> warn)
- add an err_..._ratio multiplier
- switch to HEALTH_ERR once requests are blocked long enough
- make the error ratio high (default is 32*128s -> about an hour) so that
we don't trigger on a heavily loaded cluster.

Signed-off-by: Sage Weil <sage@redhat.com>
2017-06-07 12:16:58 -04:00
Adam C. Emerson
af8d9bdf07 rgw: Support certain archaic and antiquated distributions
Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
2017-06-07 12:11:40 -04:00
Sameer Tiwari
9e37d893f0 Documentation: Fixed my mistake of describing the journal allowed vs desired.
Update to Bug http://tracker.ceph.com/issues/19879

Signed-off-by: Sameer Tiwari <stiwari@salesforce.com>
2017-06-07 11:53:22 -04:00