Commit Graph

66982 Commits

Author SHA1 Message Date
Yunchuan Wen
b7bfb1f332 mds/snap.h: remove unused declaration
Signed-off-by: Yunchuan Wen <yunchuan.wen@kylin-cloud.com>
2016-12-13 03:15:19 +00:00
Kefu Chai
a3bf341876 Merge pull request #12294 from dx9/test-ecp-uname
test: use uname -m vs. uname -p in test-erasure-code-plugins.sh

Reviewed-by: Kefu Chai <kchai@redhat.com>
2016-12-13 10:56:14 +08:00
Xiaoxi Chen
a88ceff276 librbd/diff_iterator: use proper snap to query parent overlap
If no snapshot has been made, we will use CEPH_NOSNAP instead of 0,
to query the parent overlap.

Fixes bug http://tracker.ceph.com/issues/18200

Signed-off-by: Xiaoxi Chen <xiaoxchen@ebay.com>
2016-12-13 10:53:37 +08:00
John Spray
8fdae47fd3 tasks/cephfs: set norequire_active_mds on kclient
The kernel client's cluster availability test is
more primitive than the fuse client, so we need
to switch it off to avoid client mounts failing
while MDSs are still coming up.

Fixes: http://tracker.ceph.com/issues/18161
Signed-off-by: John Spray <john.spray@redhat.com>
2016-12-13 01:05:25 +00:00
David Zafman
85e0774a7d test: Update for new error message when doing scrub with deep-scrub errors
Signed-off-by: David Zafman <dzafman@redhat.com>
2016-12-12 15:53:25 -08:00
Casey Bodley
73cd8df887 rgw: use explicit flag to cancel RGWCoroutinesManager::run()
RGWCoroutinesManager::run() was setting ret = -ECANCELED to break out of
the loop when it sees going_down. coroutines that failed with -ECANCELED
were confusing this logic and leading to coroutine deadlock assertions
below. when we hit the going_down case, set a 'canceled' flag, and check
that flag when deciding whether to break out of the loop

Fixes: http://tracker.ceph.com/issues/17465

Signed-off-by: Casey Bodley <cbodley@redhat.com>
2016-12-12 16:42:17 -05:00
Jason Dillaman
857dfee932 Merge pull request #1304 from vshankar/wip-15028
cram: support fetching from sha1 branch, tag, commit hash

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
2016-12-12 16:29:54 -05:00
Sage Weil
df052e791c Merge pull request #12450 from athanatos/wip-18180
osd/ReplicatedPG::record_write_error: don't leak orig_reply on cancel

Reviewed-by: Sage Weil <sage@redhat.com>
2016-12-12 14:20:27 -06:00
Samuel Just
3b07638c6e Merge pull request #12402 from liewegas/wip-scrub-recovery
osd: default osd_scrub_during_recovery=false

Reviewed-by: Samuel Just <sjust@redhat.com>
2016-12-12 11:30:18 -08:00
Sage Weil
ec3d62b6c3 Merge pull request #12448 from theanalyst/doc/11.1.0-rc
doc: final additions to 11.1.0-rc release notes

Reviewed-by: Sage Weil <sage@redhat.com>
2016-12-12 13:25:08 -06:00
Abhishek Lekshmanan
1e31aee432 doc: final additions to 11.1.0-rc release notes
Signed-off-by: Abhishek Lekshmanan <alekshmanan@suse.com>
2016-12-12 19:33:13 +01:00
Jenkins Build Slave User
91421d8d04 11.1.0 2016-12-12 18:27:56 +00:00
Sage Weil
63249312c2 Merge pull request #1305 from ceph/wip-h-j-x
upgrade/hammer-jewel-x: whitelist quota
2016-12-12 11:37:30 -06:00
Sage Weil
9c31801188 upgrade/hammer-jewel-x: whitelist quota
This is from teh librados api test.

Signed-off-by: Sage Weil <sage@redhat.com>
2016-12-12 12:29:09 -05:00
Kefu Chai
51342e87ff Merge pull request #10209 from runsisi/sisi-wip-remove-redundant-check
osd/ReplicatedPG: remove redundant check for balance/localize read

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
Reviewed-by: Samuel Just <sjust@redhat.com>
2016-12-13 00:21:46 +08:00
Kefu Chai
2bdf59303b Merge pull request #12190 from liupan1111/wip-speed-up-acting-ods
osd/OSDMap: improve the performance of pg_to_acting_osds

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
2016-12-13 00:18:55 +08:00
Kefu Chai
6af0c8fc04 Merge pull request #12279 from wjwithagen/wip-wjw-freebsd-ENODATA-2
os/filestore/chain_xattr.h:uses ENODATA, so include compat.h

Reviewed-by: Kefu Chai <kchai@redhat.com>
2016-12-13 00:08:02 +08:00
Kefu Chai
d89e2fca63 Merge pull request #12365 from liewegas/wip-mon-flag-warning
mon/OSDMonitor: only show interesting flags in health warning

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
2016-12-13 00:06:56 +08:00
Kefu Chai
1b533f8e26 Merge pull request #12439 from tchaikov/wip-pybind-requires
pybind/rados: add missing "length" requires for aio_execute()

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2016-12-13 00:05:07 +08:00
Kefu Chai
9418e07069 Merge pull request #12390 from tchaikov/wip-start-after-setting-nonce
msg/async: set nonce before starting the workers

Reviewed-by: Haomai Wang <haomai@xsky.com>
2016-12-13 00:03:01 +08:00
Kefu Chai
106115dac3 Merge pull request #12309 from wjwithagen/wip-wjw-freebsd-CLOCK_MONOTONE
rgw/rgw_file.cc: Add compat.h to allow CLOCK_MONOTONE

Reviewed-by: Matt Benjamin <mbenjamin@redhat.com>
2016-12-13 00:02:10 +08:00
Kefu Chai
55e18d2e5f Merge pull request #12257 from kylinstorage/wip-remove-unneeded-forward-declaration
remove unneeded forward declaration

Reviewed-by: Sage Weil <sage@redhat.com>
2016-12-13 00:01:19 +08:00
Kefu Chai
405e94ab8e Merge pull request #12301 from wjwithagen/wip-wjw-freebsd-blkid
src/test/CMakeLists.txt: Exclude test on HAVE_BLKID

Reviewed-by: Kefu Chai <kchai@redhat.com>
2016-12-13 00:00:33 +08:00
David Zafman
836d896868 Merge pull request #12378 from dzafman/wip-18139
test: Don't write to a poolid that this test might not have created

Reviewed-by: Sage Weil <sage@redhat.com>
2016-12-12 07:56:24 -08:00
Mykola Golub
b1181c7088 Merge pull request #12371 from ceph/wip-18149
qa/workunits/rbd: use more recent qemu-iotests that support Xenial

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
2016-12-12 16:17:13 +02:00
John Spray
57ac1af88f mds: make cache size health warning factor configurable
...now that it's used more than once place, copying around
the fraction was bad.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-12-12 12:02:43 +00:00
John Spray
6b578e7f25 mds: release pool allocator memory after exceeding size limit
Usually we like keeping the memory in our pools, but in circumstances
where we have violated our self-imposed inode count cap, we should
shrink our usage of system memory afterwards.

Fixes: http://tracker.ceph.com/issues/18225
Signed-off-by: John Spray <john.spray@redhat.com>
2016-12-12 12:02:40 +00:00
Brad Hubbard
ba421b5725 Merge pull request #12436 from badone/wip-cov1395747
crush: Silence coverity warnings for test/crush/crush.cc

Reviewed-by: xiexingguo <xie.xingguo@zte.com.cn>
2016-12-12 14:41:41 +10:00
Kefu Chai
2bb42d9c7f Merge pull request #12409 from tchaikov/wip-doc-releases
doc/releases: add links to kraken and v10.2.4

Reviewed-by: Brad Hubbard <bhubbard@redhat.com>
2016-12-12 12:39:13 +08:00
Kefu Chai
2126436294 pybind/rados: add missing "length" requires for aio_execute()
Signed-off-by: Kefu Chai <kchai@redhat.com>
2016-12-12 11:24:44 +08:00
songbaisen
a3fda1d721 mon: make it more clearly to debug for paxos state
Signed-off-by:song baisen <song.baisen@zte.com.cn>
2016-12-12 09:10:55 +08:00
Brad Hubbard
efbe45fe47 crush: Silence coverity warnings for test/crush/crush.cc
Coverity 1395747 reports that "c" leaks. Switching to smart pointers should
resolve that.

Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
2016-12-12 09:16:33 +10:00
Venky Shankar
d2f0d74598 cram: support fetching from sha1 branch, tag, commit hash
Signed-off-by: Venky Shankar <vshankar@redhat.com>
2016-12-11 16:46:36 +05:30
Jason Dillaman
4314cb945a qa/workunits/rbd: use more recent qemu-iotests that support Xenial
Fixes: http://tracker.ceph.com/issues/18149
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2016-12-10 20:42:52 -05:00
Loic Dachary
a5c5ba6bb7 Merge pull request #12410 from dachary/wip-10-2-5
doc: release notes for 10.2.5

Reviewed-by: Sage Weil <sage@redhat.com>
2016-12-11 00:29:53 +01:00
Loic Dachary
966d42569a doc: release notes for 10.2.5
Fixes: http://tracker.ceph.com/issues/18207

Signed-off-by: Loic Dachary <loic@dachary.org>
2016-12-10 10:14:46 +01:00
David Zafman
64a7012e98 test: Add test for keeping deep-scrub information
Signed-off-by: David Zafman <dzafman@redhat.com>
2016-12-09 16:51:20 -08:00
David Zafman
c8dc945260 osd: Add "trigger_scrub" admin socket command
This allows testing to fake a schedule scrub (must_scrub not set)

Signed-off-by: David Zafman <dzafman@redhat.com>
2016-12-09 16:51:20 -08:00
Mykola Golub
b40603f60a Merge pull request #12366 from dillaman/wip-10773
qa/workunits/rbd: removed qemu-iotest case 077

Reviewed-by: Loic Dachary <ldachary@redhat.com>
2016-12-09 22:07:20 +02:00
Casey Bodley
6d7b384cdd Merge pull request #12253 from aclamk/fix-copy-object-compression
Fixed problem with PUT with x-amz-copy-source when source object is compressed.

Reviewed-by: Casey Bodley <cbodley@redhat.com>
2016-12-09 12:08:34 -05:00
Casey Bodley
ff152935c6 Merge pull request #12283 from ceph/wip-rgw-18106
rgw: make RGWLocalAuthApplier::is_admin_of() aware about system users.

Reviewed-by: Orit Wasserman <owassrm@redhat.com>
Reviewed-by: Casey Bodley <cbodley@redhat.com>
2016-12-09 12:05:17 -05:00
Patrick Donnelly
79c2db9bea
Merge remote-tracking branch 'upstream/pull/1302/head'
* upstream/pull/1302/head:
  fsx-mpi: fsx-mpi expects a file name not a dir
  fsx-mpi: remove chef task
  Fix mpi calls
2016-12-09 11:09:02 -05:00
Loic Dachary
0a99c92ede Merge pull request #12414 from ceph/wip-16125
ceph-disk: prevent unnecessary tracebacks from subprocess.check_call

Reviewed-by: Loic Dachary <ldachary@redhat.com>
2016-12-09 14:39:32 +01:00
Jason Dillaman
cb798835a9 Merge pull request #12368 from trociny/wip-18146
rbd-mirror: fix sparse read optimization in image sync

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
2016-12-09 07:59:17 -05:00
Alfredo Deza
2cb9bfe6e4 ceph-disk: update tests to use the new exit flag
Signed-off-by: Alfredo Deza <adeza@redhat.com>
2016-12-09 07:52:25 -05:00
Alfredo Deza
9be64fb00f ceph-disk: remove try/except for CalledProcessError to use new exit flag handler
Signed-off-by: Alfredo Deza <adeza@redhat.com>
2016-12-09 07:52:25 -05:00
Alfredo Deza
8e524d142a ceph-disk: allow to exit nicely when using command_check_call
Signed-off-by: Alfredo Deza <adeza@redhat.com>
2016-12-09 07:52:25 -05:00
Bartłomiej Święcki
6a76adcdb1 osd: Increase priority for inactive PGs backfill
This change does prioritize backfill of PGs which don't
have min_size active copies. Such PGs would cause IO stalls
for clients and would increase throttlers usage.

This change also fixes few subtlle out-of-bounds bugs.

Signed-off-by: Bartłomiej Święcki <bartlomiej.swiecki@corp.ovh.com>
2016-12-09 13:30:11 +01:00
Brad Hubbard
483969fcc6 Merge pull request #12408 from badone/wip-mon_osd_down_out_interval-docfix
doc: Modify documentation for mon_osd_down_out_interval

Reviewed-by: xiexingguo <xie.xingguo@zte.com.cn>
2016-12-09 19:20:13 +10:00
Kefu Chai
26c6864667 doc/releases: add links to kraken and v10.2.4
Signed-off-by: Kefu Chai <kchai@redhat.com>
2016-12-09 16:48:09 +08:00