]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Tim Serong [Mon, 10 Sep 2018 10:27:11 +0000 (20:27 +1000)]
ceph-disk: compatibility fix for python 3
In python 3, dev is a string, but mounts_dev is bytes (because
/proc/mounts was opened with mode 'rb') so they can't compare
equal, resulting in is_mounted() returning None for mounted OSDs.
The safest fix for this we could come up with was to normalize
dev to a str using _bytes2str() (just in case), and open
/proc/mounts in mode 'r', so its lines are interpreted as strs.
Fixes: https://tracker.ceph.com/issues/35906
Signed-off-by: Tim Serong <tserong@suse.com>
(cherry picked from commit
8e3d948ff4c3b19dc55b9a5ae63f5b63f0c476a3 )
Patrick Donnelly [Fri, 28 Sep 2018 21:28:51 +0000 (14:28 -0700)]
Merge PR #23553 into luminous
* refs/pull/23553/head:
mds: report lagginess at lower debug
MDSMonitor: note beacons and cluster changes at low dbg level
Reviewed-by: Yuri Weinstein <yweins@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Conflicts:
src/mds/Beacon.cc
Yuri Weinstein [Fri, 28 Sep 2018 20:06:52 +0000 (13:06 -0700)]
Merge pull request #24083 from batrick/i26851
luminous: ceph_volume_client: py3 compatible
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Fri, 28 Sep 2018 20:05:53 +0000 (13:05 -0700)]
Merge pull request #24313 from batrick/i36198
luminous: ceph-fuse: add SELinux policy
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Fri, 28 Sep 2018 20:04:28 +0000 (13:04 -0700)]
Merge pull request #24119 from ukernel/luminous-22504
luminous: client: drop null child dentries before try pruning inode's alias
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Fri, 28 Sep 2018 20:04:01 +0000 (13:04 -0700)]
Merge pull request #24138 from pdvian/wip-35718-luminous
luminous: mds: print is_laggy message once
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Fri, 28 Sep 2018 20:03:28 +0000 (13:03 -0700)]
Merge pull request #24268 from smithfarm/wip-36101-luminous
luminous: qa: remove knfs site from future releases
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Yuri Weinstein <yweins@redhat.com>
Yuri Weinstein [Fri, 28 Sep 2018 20:02:45 +0000 (13:02 -0700)]
Merge pull request #24285 from iridescent-rsy/luminous-35958
luminous: librbd: fixed assert when flattening clone with zero overlap
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Yuri Weinstein [Fri, 28 Sep 2018 20:02:06 +0000 (13:02 -0700)]
Merge pull request #24303 from vshankar/wip-35931
luminous: client: retry remount on dcache invalidation failure
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Andrew Schoen [Fri, 28 Sep 2018 19:18:48 +0000 (14:18 -0500)]
Merge pull request #24338 from alfredodeza/luminous-rm36251
luminous ceph-volume add new ceph-handlers role from ceph-ansible
Reviewed-by: Andrew Schoen <aschoen@redhat.com>
Alfredo Deza [Fri, 28 Sep 2018 16:40:09 +0000 (12:40 -0400)]
ceph-volume tests.functional add new ceph-handlers role from ceph-ansible
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
956dc5e65c6e35e27ab6524d4fb9fdd05d616cf6 )
Patrick Donnelly [Thu, 20 Sep 2018 17:27:43 +0000 (10:27 -0700)]
ceph-fuse: add to selinux profile
Fixes: http://tracker.ceph.com/issues/36103
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit
2bfaac762d283a0e000074f79f704033afd6dead )
Patrick Donnelly [Sat, 11 Aug 2018 18:21:43 +0000 (11:21 -0700)]
mds: report lagginess at lower debug
To help diagnose laggy daemons with performance tuned debug settings.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit
c8aff620c273fd4579b23e1647450b6f00c6bac7 )
Conflicts:
src/mds/Beacon.cc
Patrick Donnelly [Sat, 11 Aug 2018 17:40:03 +0000 (10:40 -0700)]
MDSMonitor: note beacons and cluster changes at low dbg level
These messages are essential for diagnosing the reason why the MDSMonitor is
kicking MDSs out of the MDSMap. They should also be rare enough that the extra
verbosity is not noticable.
Fixes: http://tracker.ceph.com/issues/26898
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit
d6f6b4f72aae3c869ed7ee40dc737bb2e2323d3c )
Conflicts:
src/mon/MDSMonitor.cc
Patrick Donnelly [Fri, 27 Jul 2018 21:05:42 +0000 (14:05 -0700)]
ceph_volume_client: use integer division for pg_num
Otherwise a float is sent to the manager which is not the invalid format.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit
33910303cb4940ca90cd76a5b4adb4de9c88d04c )
Rishabh Dave [Mon, 23 Jul 2018 10:31:42 +0000 (10:31 +0000)]
packages.yaml: install py3 compatible cephfs and rados bindings...
on Ubuntu and CentOS for upstream testing.
Signed-off-by: Rishabh Dave <ridave@redhat.com>
(cherry picked from commit
c1efd59f618e24cf060d564ac0f21d5b0b57fd4a )
Patrick Donnelly [Wed, 6 Jun 2018 23:51:05 +0000 (16:51 -0700)]
qa: test py2 and py3 for volume client
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit
a5e3b0512c8c4e5ffbd960b903f40871d1867601 )
Rishabh Dave [Tue, 15 May 2018 06:06:39 +0000 (06:06 +0000)]
qa: make test_volume_client.py py3 compatible
Signed-off-by: Rishabh Dave <ridave@redhat.com>
(cherry picked from commit
f28274dc70aa102e3c4523059a65e5da8c8a0426 )
Rishabh Dave [Tue, 24 Jul 2018 07:59:55 +0000 (07:59 +0000)]
qa/ceph-volume: allow executing python pyloads using python3
Signed-off-by: Rishabh Dave <ridave@redhat.com>
(cherry picked from commit
2f35ade05f9910ef9bdaaa24de9593bd584073fa )
Conflicts:
qa/tasks/vstart_runner.py
Rishabh Dave [Wed, 16 May 2018 19:06:42 +0000 (19:06 +0000)]
cephfs.pyx: add py3 compatibility
Signed-off-by: Rishabh Dave <ridave@redhat.com>
(cherry picked from commit
5106582fc7edae7f39161cf89e566c020fcfa0ce )
Rishabh Dave [Tue, 15 May 2018 05:58:25 +0000 (05:58 +0000)]
ceph-volume-client: add py3 compatibility
Fixes: http://tracker.ceph.com/issues/17230
Signed-off-by: Rishabh Dave <ridave@redhat.com>
(cherry picked from commit
efdf1af41d198565c2993598b8e345683396b169 )
Conflicts:
src/pybind/ceph_volume_client.py
Yuri Weinstein [Thu, 27 Sep 2018 23:19:49 +0000 (16:19 -0700)]
Merge pull request #24084 from batrick/i24862
luminous: ceph_volume_client: allow atomic update of RADOS objects
Reviewed-by: Nathan Cutler <ncutler@suse.com>
Yuri Weinstein [Thu, 27 Sep 2018 23:19:01 +0000 (16:19 -0700)]
Merge pull request #24282 from pdvian/wip-35933-luminous
luminous:client:two ceph-fuse client, one can not list out files created by an…
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Thu, 27 Sep 2018 21:07:06 +0000 (14:07 -0700)]
Merge pull request #24188 from pdvian/wip-35844-luminous
luminous: osdc/Objecter: fix split vs reconnect race
Reviewed-by: Kefu Chai <kchai@redhat.com>
Yuri Weinstein [Thu, 27 Sep 2018 21:06:18 +0000 (14:06 -0700)]
Merge pull request #24245 from dillaman/wip-36116-luminous
luminous: test/librbd: not valid to have different parents between image snapshots
Reviewed-by: Mykola Golub <mgolub@mirantis.com>
Yuri Weinstein [Thu, 27 Sep 2018 21:05:38 +0000 (14:05 -0700)]
Merge pull request #24257 from pdvian/wip-35929-luminous
luminous: mon/OSDMonitor: invalidate max_failed_since on cancel_report
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
Yuri Weinstein [Thu, 27 Sep 2018 21:04:40 +0000 (14:04 -0700)]
Merge pull request #24259 from tchaikov/luminous-35969
luminous: rpm: use updated gperftools
Reviewed-by: Brad Hubbard <bhubbard@redhat.com>
Reviewed-by: Nathan Cutler <ncutler@suse.com>
Andrew Schoen [Thu, 27 Sep 2018 18:47:30 +0000 (13:47 -0500)]
Merge pull request #24307 from alfredodeza/luminous-batch-report-fix
luminous ceph-volume: fix journal and filestore data size in `lvm batch --report`
Reviewed-by: Andrew Schoen <aschoen@redhat.com>
Andrew Schoen [Tue, 25 Sep 2018 17:09:15 +0000 (12:09 -0500)]
ceph-volume: only run ceph-common once during tests
Using import_role we can have just one play and avoid rerunning
ceph-common over and over again in daemon specific plays.
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit
a81386123e66e2d61391371cec4485b3148ff5a7 )
Andrew Schoen [Mon, 24 Sep 2018 21:41:37 +0000 (16:41 -0500)]
ceph-volume: fail fast during tests
This also rsyncs the ceph-volume code to the testing vms before
a ceph.conf is generated because ceph-volume is needed now for this
to figure out the number of osds when you're using 'lvm batch'.
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit
eed244473d11a153cd7a37e0e9bfaf804a20db89 )
Andrew Schoen [Mon, 24 Sep 2018 21:37:35 +0000 (16:37 -0500)]
ceph-volume: when using batch --report return journal/data size as ints
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit
a5e8bc97e70195cee54c4b66d10684d6cd3f5e38 )
Andrew Schoen [Thu, 27 Sep 2018 16:11:06 +0000 (11:11 -0500)]
Merge pull request #24280 from alfredodeza/luminous-rm27210
luminous ceph-volume custom cluster names fail on filestore trigger
Reviewed-by: Andrew Schoen <aschoen@redhat.com>
Nathan Cutler [Thu, 27 Sep 2018 14:29:47 +0000 (16:29 +0200)]
Merge pull request #24211 from pdvian/wip-35854-luminous
luminous: doc: remove deprecated 'scrubq' from ceph(8)
Reviewed-by: Nathan Cutler <ncutler@suse.com>
Venky Shankar [Tue, 28 Aug 2018 10:35:18 +0000 (06:35 -0400)]
client: retry remount on dcache invalidation failure
For some (unknown) reason, there have been reports of
ceph-fuse crash due to failure is remounting at the
time of invalidating kernel dentry cache. This issue
is not also reproducible yet.
Therefore, as suggested by Patrick and Zheng, for a
temporary workaround, client would ignore the failure
as the invalidation would be retried again. There is
a max cap on the number of consecutive remount failures
after which client would abort.
Fixes: http://tracker.ceph.com/issues/35931
Signed-off-by: Venky Shankar <vshankar@redhat.com>
(cherry picked from commit
d1471f070cd1ad9c0f773e00d2552161d1ad5955 )
Conflicts:
src/client/Client.cc
src/client/Client.h
src/common/options.cc
Yuri Weinstein [Wed, 26 Sep 2018 16:30:16 +0000 (09:30 -0700)]
Merge pull request #24190 from joke-lee/luminous
luminous:rgw: return x-amz-version-id: null when delete obj in versio…
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Wed, 26 Sep 2018 16:29:46 +0000 (09:29 -0700)]
Merge pull request #24231 from pdvian/wip-35856-luminous
luminous: multisite: segfault on shutdown/realm reload
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Wed, 26 Sep 2018 16:29:03 +0000 (09:29 -0700)]
Merge pull request #24171 from ukernel/luminous-25046
luminous: mds: health warning for slow metadata IO
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Jason Dillaman [Wed, 12 Sep 2018 01:52:05 +0000 (21:52 -0400)]
luminous: librbd: fixed assert when flattening clone with zero overlap
Fixes: http://tracker.ceph.com/issues/35702
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit
8e90354 )
Conflicts:
src/librbd/Operations.cc
src/test/librbd/test_librbd.cc
Peng Xie [Sun, 26 Aug 2018 14:02:28 +0000 (22:02 +0800)]
client:two ceph-fuse client, one can not list out files created by another.
the problem happend after mds restored the stale client session,
right after one normal client created files under one specific
dir,while the restored client can not list out the just created
files.
Fixes: http://tracker.ceph.com/issues/27051
Signed-off-by: Peng Xie <peng.hse@xtaotech.com>
(cherry picked from commit
fa1c2be4a93da46252a0b53776696baf80840700 )
Conflicts:
src/client/Client.cc : Resolved for cap.issued
Alfredo Deza [Mon, 24 Sep 2018 18:44:31 +0000 (14:44 -0400)]
ceph-volume tests/functional update dmcrypt test playbooks to use --cluster
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
96e29adf72804cb4fca62c04416b8fe2d23587bb )
Alfredo Deza [Mon, 24 Sep 2018 16:52:50 +0000 (12:52 -0400)]
ceph-volume tests/functional/batch update tests to use 'test' cluster name
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
5f72d438da94323c0b04f261b1a60b13eaf03966 )
Alfredo Deza [Mon, 24 Sep 2018 16:52:33 +0000 (12:52 -0400)]
ceph-volume tests/functional/lvm update tests to use 'test' cluster name
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
7b0d2dc66968f81192f152fa166fbc2cb4dcb438 )
Alfredo Deza [Mon, 24 Sep 2018 16:51:47 +0000 (12:51 -0400)]
ceph-volume tests/functional update playbooks to use --cluster
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
ba10b6820fd7f352ed6ba78246a6976c54783cbd )
Alfredo Deza [Mon, 24 Sep 2018 16:27:04 +0000 (12:27 -0400)]
ceph-volume main move out the load_ceph_conf_path to the configuration module
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
532908099eb63bd805db8aecae1ef12301076410 )
Alfredo Deza [Mon, 24 Sep 2018 14:34:55 +0000 (10:34 -0400)]
ceph-volume tests.devices patch configuration.load to avoid raising non-existing ceph.conf
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
fe80e5399e9dd1f6b9f94404d732a76f961d9eb4 )
Alfredo Deza [Mon, 24 Sep 2018 14:29:12 +0000 (10:29 -0400)]
ceph-volume tests reset the cluster name after every unit test
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
877ee1a27e0651355812e27fbc6ac0a87ea56a68 )
Alfredo Deza [Wed, 19 Sep 2018 17:23:19 +0000 (13:23 -0400)]
ceph-volume tests.functional add custom cluster name to centos7-filestore-create
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
e61848dc97b638c6b8e60f0896d325f436142864 )
Alfredo Deza [Wed, 19 Sep 2018 16:04:41 +0000 (12:04 -0400)]
ceph-volume lvm.activate load the config always by reading the cluster name from the LV
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
a88cbfc73a258b51b20e060e269234962444ce63 )
Alfredo Deza [Wed, 19 Sep 2018 16:04:17 +0000 (12:04 -0400)]
ceph-volume configuration be able to set conf.ceph
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
4f8226f359d94a86a39492ed653d9a9e5d8dd8f9 )
Yuri Weinstein [Tue, 25 Sep 2018 20:28:56 +0000 (13:28 -0700)]
Merge pull request #24135 from linuxbox2/luminous-rgw-sync-loglevel
luminous: rgw: raise debug level on redundant data sync error messages
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Tue, 18 Sep 2018 21:55:27 +0000 (14:55 -0700)]
qa/tests: removed knfs suite
Fixes: http://tracker.ceph.com/issues/36075
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
(cherry picked from commit
eaca18873350bbef45a695f29569fab269d238da )
Conflicts:
qa/suites/knfs/basic/clusters/extra-client.yaml
Kefu Chai [Mon, 17 Sep 2018 06:09:17 +0000 (14:09 +0800)]
rpm: use updated gperftools
make sure we only build with the higher version of gperftools on
distros where both 2.4 and 2.6.1 are packaged. see
https://git.centos.org/summary/rpms!gperftools.git . at the time of
writing, gperftools 2.6.1 is packaged for CentOS/RHEL 7, if gperftools
(>= 2.4) is required by Ceph, and user already has this version
installed, when new Ceph packages are installed, the updated gperftools
2.6.1 version won't be installed as a dependency. when launching
Ceph compiled with tcmalloc enabled, we will have
symbol lookup error: ceph-osd: undefined symbol: _ZdaPvm
so, by bumping up the required version of gperftools, the updated
gperftools will be installed.
see https://software.opensuse.org/package/gperftools, openSUSE/SLE offer
2.5. so they are safe at this moment.
Fixes: http://tracker.ceph.com/issues/35969
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit
e69e50a859acebe0bf854c12a71bfe745e0c1cf6 )
xie xingguo [Thu, 6 Sep 2018 02:44:26 +0000 (10:44 +0800)]
mon/OSDMonitor: invalidate max_failed_since on cancel_report
max_failed_since might reference the very failure-report which is to be
cancelled. We can simply invalidate it here to make **get_failed_since()**
recalculate if necessary.
Fixes: http://tracker.ceph.com/issues/35860
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
(cherry picked from commit
91a2d408a9349db08e916e2145040df8b1f17e14 )
Yuri Weinstein [Mon, 24 Sep 2018 19:28:22 +0000 (12:28 -0700)]
Merge pull request #24065 from jdurgin/wip-luminous-cache-autotune
luminous: os/bluestore: cache autotuning and memory limit
Reviewed-by: Sage Weil <sage@redhat.com>
Yuri Weinstein [Mon, 24 Sep 2018 17:09:12 +0000 (10:09 -0700)]
Merge pull request #24115 from dzafman/wip-35991
luminous: tools: ceph-objectstore-tool: Allow target level as first positional …
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Jason Dillaman [Tue, 4 Sep 2018 12:31:07 +0000 (08:31 -0400)]
test/librbd: not valid to have different parents between image snapshots
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit
3b8f4f136b52d591dc39e6c4f8bd59c1e63c158a )
Fixes: http://tracker.ceph.com/issues/36118
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
Casey Bodley [Tue, 4 Sep 2018 20:50:22 +0000 (16:50 -0400)]
rgw: RGWAsyncPutSystemObjAttrs does not access coroutine memory
also exposes optional objv_tracker to RGWSimpleRadosWriteAttrsCR
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit
b01b3952db4c45feb98dbc7e224ca1982c224c72 )
Casey Bodley [Tue, 4 Sep 2018 20:33:48 +0000 (16:33 -0400)]
rgw: RGWAsyncPutSystemObj does not access coroutine memory
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit
85396674811be8f7a25dc9d25fd2db68c93561c5 )
Casey Bodley [Tue, 4 Sep 2018 20:14:18 +0000 (16:14 -0400)]
rgw: RGWAsyncGetSystemObj does not access coroutine memory
now reads the buffer/attrs into local memory, and calling coroutines
copy out the result during request_complete()
Fixes: http://tracker.ceph.com/issues/35543
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit
089db9f74266073eb1d45900d982f1831c72fe59 )
Conflicts:
src/rgw/rgw_cr_rados.h : Resolved in request_complete
Andrew Schoen [Fri, 21 Sep 2018 14:12:13 +0000 (09:12 -0500)]
Merge pull request #24206 from alfredodeza/luminous-rm24794
luminous ceph-volume: earlier detection for --journal and --filestore flag requirements
Reviewed-by: Andrew Schoen <aschoen@redhat.com>
Andrew Schoen [Fri, 21 Sep 2018 14:02:18 +0000 (09:02 -0500)]
Merge pull request #24209 from alfredodeza/luminous-rm36088
luminous ceph-volume batch: allow journal+block.db sizing on the CLI
Reviewed-by: Andrew Schoen <aschoen@redhat.com>
Yan, Zheng [Fri, 13 Jul 2018 08:10:39 +0000 (16:10 +0800)]
mds: health warning for slow metadata IO
Fixes: http://tracker.ceph.com/issues/24879
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
(cherry picked from commit
0f735f40315448560fde049ed3ea019a7d30d868 )
Conflicts:
src/mds/Beacon.cc
src/mds/OpenFileTable.cc
src/mds/MDSContext.h
src/mds/MDSContext.cc
Ruben Kerkhof [Thu, 6 Sep 2018 12:17:23 +0000 (14:17 +0200)]
doc: remove deprecated 'scrubq' from ceph(8)
The option was removed in
a30cbe9adea34782be8699c010d4281d245feb6b
Fixes: http://tracker.ceph.com/issues/35813
Signed-off-by: Ruben Kerkhof <ruben@rubenkerkhof.com>
(cherry picked from commit
d05cd1a3e612f52eab9bde42c184022205f65384 )
Alfredo Deza [Thu, 20 Sep 2018 15:01:21 +0000 (11:01 -0400)]
doc/man/ceph-volume update batch command with new sizing flags
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
042c2cbe475023cd3c34102b710b93c60d92077e )
Alfredo Deza [Thu, 20 Sep 2018 14:54:12 +0000 (10:54 -0400)]
ceph-volume tests.devices update fixture to include new size flags for strategies
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
5df56a09c7b445c36d2b9d2874962c40b681a65c )
Alfredo Deza [Wed, 19 Sep 2018 21:40:24 +0000 (17:40 -0400)]
ceph-volume lvm.batch.filestore consume --journal-size
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
2378c282c78f49f4d911316a7bc85e49a3f8e7f9 )
Alfredo Deza [Wed, 19 Sep 2018 21:39:59 +0000 (17:39 -0400)]
ceph-volume lvm.batch.bluestore consume --block-db-size
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
7e52bc559b21b18a06ff1dffd764dbe8aefe96a2 )
Alfredo Deza [Wed, 19 Sep 2018 21:08:16 +0000 (17:08 -0400)]
ceph-volume lvm.batch add sizing flags for journal and block.db LVs
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
0a16dbd7a8f597cde6debbdcd968cdaf44f6ce47 )
Yuri Weinstein [Thu, 20 Sep 2018 22:27:55 +0000 (15:27 -0700)]
Merge pull request #23773 from ifed01/wip-ifed-perf-reset-luminous
luminous: os/bluestore: bluestore_buffer_hit_bytes perf counter doesn't reset
Reviewed-by: Neha Ojha <nojha@redhat.com>
Yuri Weinstein [Thu, 20 Sep 2018 22:10:26 +0000 (15:10 -0700)]
Merge pull request #23817 from pdvian/wip-26992-luminous
luminous: discover_all_missing() not always called during activating
Reviewed-by: David Zafman <dzafman@redhat.com>
Yuri Weinstein [Thu, 20 Sep 2018 22:09:49 +0000 (15:09 -0700)]
Merge pull request #23832 from pdvian/wip-27058-luminous
luminous: debian/rules: fix ceph-mgr .pyc files left behind
Reviewed-by: Nathan Cutler <ncutler@suse.com>
Yuri Weinstein [Thu, 20 Sep 2018 22:09:26 +0000 (15:09 -0700)]
Merge pull request #23915 from liewegas/wip-pod-hostname-luminous
luminous: common: get real hostname from container/pod environment
Reviewed-by: Kefu Chai <kchai@redhat.com>
Alfredo Deza [Tue, 18 Sep 2018 18:07:06 +0000 (14:07 -0400)]
ceph-volume tests.lvm verify that it is reported when --journal is not used with --filestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
ae657d2fe012b0e78959ead9d8d3baa1095a31ff )
Alfredo Deza [Tue, 18 Sep 2018 18:06:19 +0000 (14:06 -0400)]
ceph-volume lvm.prepare report back when --journal is not used when --filestore is set
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
2fc1ee48ac10bd8008ff5d54382db1d83e57ec1b )
yuliyang [Wed, 5 Sep 2018 06:18:44 +0000 (14:18 +0800)]
luminous:rgw: return x-amz-version-id: null when delete obj in versioning
suspended bucket
fix http://tracker.ceph.com/issues/35814
Signed-off-by: yuliyang <yuliyang@cmss.chinamobile.com>
(cherry picked from commit
9c39147 )
Sage Weil [Sat, 1 Sep 2018 20:48:37 +0000 (15:48 -0500)]
osdc/Objecter: fix split vs reconnect race
If we are reconnecting to an OSD, we may not have a connection established
so this feature check is completely unreliable. If there happens to be a
split, that could mean we don't resend an op that should have been resent.
Instead, we can infer the OSD has the new behavior if require_osd_release
is luminous or later. If not, we can look at the per-osd features in the
OSDMap.
Fixes: http://tracker.ceph.com/issues/22544
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit
917fdbe5422d099902aaa08b3406efc0247ef155 )
Yuri Weinstein [Wed, 19 Sep 2018 16:06:51 +0000 (09:06 -0700)]
Merge pull request #24054 from smithfarm/wip-24983-luminous
luminous: 'radosgw-admin sync error trim' only trims partially
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Wed, 19 Sep 2018 16:05:14 +0000 (09:05 -0700)]
Merge pull request #24056 from smithfarm/wip-24985-luminous
luminous: multisite: object metadata operations are skipped by sync
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Wed, 19 Sep 2018 16:04:39 +0000 (09:04 -0700)]
Merge pull request #24069 from pdvian/wip-35709-luminous
luminous: rgw: fix deadlock on RGWIndexCompletionManager::stop
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yan, Zheng [Wed, 19 Sep 2018 01:42:41 +0000 (09:42 +0800)]
Merge pull request #23884 from smithfarm/wip-32084-luminous
luminous: mds: MDBalancer::try_rebalance() may stop prematurely
Yan, Zheng [Fri, 3 Aug 2018 10:43:41 +0000 (18:43 +0800)]
mds: fix premature stop of MDBalancer::try_rebalance()
When mds wants to export loads to multiple targets. The amount of single
export can easily be less than (0.2 * target_loads). For example:
6 active mds, mds.0's load is 11, other ranks' loads are 5. target_load
is 6. Export amount for each target is 1.
Fixes: http://tracker.ceph.com/issues/26973
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
(cherry picked from commit
e68cbec6a647f0252d848620ce6e51b4f3eea1e3 )
Yan, Zheng [Wed, 1 Aug 2018 05:16:10 +0000 (13:16 +0800)]
mds: don't queue straydir for split
straydir can not get splitted (for now at least)
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
(cherry picked from commit
959e63c1f88b6ce6b2492cfb98e5875d928558e4 )
Conflicts:
src/mds/MDBalancer.cc
- finesse split/merge conditional expression
Yan, Zheng [Wed, 19 Sep 2018 01:13:02 +0000 (09:13 +0800)]
Merge pull request #23789 from ukernel/luminous-mds-export-size
luminous: mds: optimize the way how max export size is enforced
Yan, Zheng [Wed, 19 Sep 2018 01:12:45 +0000 (09:12 +0800)]
Merge branch 'luminous' into luminous-mds-export-size
Yuri Weinstein [Tue, 18 Sep 2018 20:00:57 +0000 (13:00 -0700)]
Merge pull request #23594 from dillaman/wip-23604-luminous
luminous: librbd: discard should wait for in-flight cache writeback to complete
Reviewed-by: Mykola Golub <mgolub@mirantis.com>
Yuri Weinstein [Tue, 18 Sep 2018 20:00:24 +0000 (13:00 -0700)]
Merge pull request #24100 from pdvian/wip-35711-luminous
luminous: librbd: journaling unable request can not be sent to remote lock owner
Reviewed-by: Mykola Golub <mgolub@mirantis.com>
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Yuri Weinstein [Tue, 18 Sep 2018 19:58:32 +0000 (12:58 -0700)]
Merge pull request #23793 from pdvian/wip-26983-luminous
luminous : client: fix choose_target_mds for requests that do name lookup
Reviewed-by: Yan, Zheng <zyan@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Tue, 18 Sep 2018 19:57:51 +0000 (12:57 -0700)]
Merge pull request #23881 from smithfarm/wip-26977-luminous
luminous: cephfs-data-scan: print the max used ino
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Yan, Zheng <zyan@redhat.com>
Yuri Weinstein [Tue, 18 Sep 2018 19:56:31 +0000 (12:56 -0700)]
Merge pull request #23946 from ukernel/luminous-35721
luminous: mds: prevent MDSRank::evict_client from blocking finisher thread
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Tue, 18 Sep 2018 19:56:01 +0000 (12:56 -0700)]
Merge pull request #23990 from batrick/i35852
luminous: MDSMonitor: lookup of gid in prepare_beacon that has been removed will cause exception
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
Yuri Weinstein [Tue, 18 Sep 2018 19:55:23 +0000 (12:55 -0700)]
Merge pull request #24033 from smithfarm/wip-24934-luminous
luminous: cephfs-journal-tool: wrong layout info used
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Tue, 18 Sep 2018 19:55:04 +0000 (12:55 -0700)]
Merge pull request #24134 from batrick/qa-.qa-symlink
luminous: qa: add .qa helper link
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Sun, 2 Sep 2018 01:47:18 +0000 (18:47 -0700)]
mds: print is_laggy message once
Beacon::is_laggy gets called frequently which causes the debug log to get
spammed with messages.
Steps to reproduce:
- setup vstart cluster
- start IO from client (e.g. `cp -av /usr /mnt`)
- find connection to mon by the mds
$ ss -tnp | grep ceph-mds
...
ESTAB 0 0 127.0.0.1:58988 127.0.0.1:40928 users:(("ceph-mds",pid=13095,fd=37))
- block the connection using iptables on the MDS node:
$ sudo iptables -I INPUT -p tcp --sport 58988 --dport 40928 -j DROP
- verify MDS log using:
$ tail -f mds.a.log | grep beacon
...
2018-09-05 19:21:05.672
7f2908a9d700 1 mds.beacon.a is_laggy 19.8876 > 15 since last acked beacon
2018-09-05 19:21:05.672
7f2908a9d700 1 mds.beacon.a is_laggy 19.8876 > 15 since last acked beacon
2018-09-05 19:21:05.672
7f2908a9d700 1 mds.beacon.a is_laggy 19.8876 > 15 since last acked beacon
2018-09-05 19:21:05.672
7f2908a9d700 1 mds.beacon.a is_laggy 19.8876 > 15 since last acked beacon
2018-09-05 19:21:05.784
7f2905a97700 5 mds.beacon.a Sending beacon up:active seq 114
2018-09-05 19:21:05.784
7f2905a97700 1 -- 127.0.0.1:6813/
2277776624 --> 127.0.0.1:40929/0 -- mdsbeacon(14162/a up:active seq 114 v7) v7 -- 0x7d1e7e2a80 con 0
2018-09-05 19:21:05.784
7f2905a97700 20 mds.beacon.a sender thread waiting interval 4s
Fixes: http://tracker.ceph.com/issues/35250
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit
439902311910f9c239f48c0958fd95fa5839e309 )
Conflicts:
src/mds/Beacon.cc: Resovled for laggy, changed to was_laggy
Casey Bodley [Fri, 7 Sep 2018 13:51:56 +0000 (09:51 -0400)]
rgw: raise debug level on redundant data sync error messages
each of these errors have already been logged at a lower level with a
more detailed error message. by logging them as ERRORs at level 0 here,
the messages could be easily confused as separate failures
Fixes: http://tracker.ceph.com/issues/36037
See-also: http://tracker.ceph.com/issues/35830
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit
a6e392f2b8cb885f6fede5dfa9b584659740e161 )
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
Yuri Weinstein [Mon, 17 Sep 2018 19:48:11 +0000 (12:48 -0700)]
Merge pull request #23408 from ukernel/luminous-24820
luminous: mds: avoid using g_conf->get_val<...>(...) in hot path
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Mon, 17 Sep 2018 19:47:39 +0000 (12:47 -0700)]
Merge pull request #23505 from ukernel/luminous-26834
luminous: mds: calculate load by checking self CPU usage
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Mon, 17 Sep 2018 19:47:10 +0000 (12:47 -0700)]
Merge pull request #23507 from ukernel/luminous-26858
luminous: mds: reset heartbeat map at potential time-consuming places
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Mon, 17 Sep 2018 19:46:44 +0000 (12:46 -0700)]
Merge pull request #23554 from batrick/i26906
luminous: MDSMonitor: consider raising priority of MMDSBeacons from MDS so they are processed before other client messages
Reviewed-by: Nathan Cutler <ncutler@suse.com>
Yuri Weinstein [Mon, 17 Sep 2018 19:46:09 +0000 (12:46 -0700)]
Merge pull request #23556 from batrick/i26924
luminous: mds: mds got laggy because of MDSBeacon stuck in mqueue
Reviewed-by: Yan, Zheng <zyan@redhat.com>
Yuri Weinstein [Mon, 17 Sep 2018 19:45:37 +0000 (12:45 -0700)]
Merge pull request #23617 from pdvian/wip-26915-luminous
luminous: client: check for unmounted condition before printing debug output
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>