Brad Hubbard
eeeed6497b
qa/suites/rados: Disable scrub backoff
...
A long run of lost coin flips can lead to a timeout in
test_large_omap_detection.py.
Fixes: http://tracker.ceph.com/issues/23578
Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
2018-04-09 17:21:01 +10:00
Mykola Golub
46df695b05
Merge pull request #21248 from dillaman/wip-23548
...
librbd: potential race between discard and writeback
Reviewed-by: Mykola Golub <mgolub@suse.com>
2018-04-08 23:22:38 +03:00
Nathan Cutler
fa5e8f9376
Merge pull request #17531 from smithfarm/wip-calamari-tests
...
tests: cleanup: drop calamari tasks
Reviewed-by: John Spray <john.spray@redhat.com>
Reviewed-by: Lenz Grimmer <lenz@grimmer.com>
2018-04-08 08:54:27 +02:00
Xie Xingguo
1cdc9161ff
Merge pull request #21245 from ifed01/wip-ifed-shard-bug-logging
...
os/bluestore: be more verbose when hitting unloaded shard in extent map
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
2018-04-08 13:37:00 +08:00
Kefu Chai
8d87f1c055
Merge pull request #20866 from jan--f/mgr-fix-metadata-return
...
mon/MonCommands: mgr metadata - improve parameter naming consistency
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Joao Eduardo Luis <joao@suse.de>
Reviewed-by: John Spray <john.spray@redhat.com>
2018-04-07 17:31:04 +08:00
Kefu Chai
4cfc09d1b5
Merge pull request #21217 from tchaikov/wip-cmp-hobject
...
common/hobject: preserve the order of hobject
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
Reviewed-by: Sage Weil <sage@redhat.com>
2018-04-07 17:29:57 +08:00
Kefu Chai
8c7f8bd433
Merge pull request #21211 from tchaikov/wip-clang-without-map-merge
...
cmake,mds: detect std::map::merge() before using it.
Reviewed-by: Zheng Yan <zyan@redhat.com>
2018-04-07 17:29:15 +08:00
Kefu Chai
9bfb1f6246
Merge pull request #21161 from cbodley/wip-mon-config-obs
...
config: notify config observers on set_mon_vals()
Reviewed-by: Sage Weil <sage@redhat.com>
2018-04-07 17:27:15 +08:00
Kefu Chai
9fda4d2b5e
Merge pull request #21151 from xiexingguo/wip-prune-olr
...
osd/PGBackend: drop input "snapid_t" from objects_list_range()
Reviewed-by: Kefu Chai <kchai@redhat.com>
2018-04-07 17:25:48 +08:00
Kefu Chai
6f6b70025b
Merge pull request #21150 from xiexingguo/wip-large-omap-obj-counting
...
osd/PGBackend: fix large_omap_objects checking
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
Reviewed-by: David Zafman <dzafman@redhat.com>
2018-04-07 17:24:56 +08:00
Kefu Chai
52012a069c
Merge pull request #21119 from Liuchang0812/refine-optracker-history
...
common, osd: duplicated "start" event in OpTracker, improve OpTracker::dump_ops
Reviewed-by: Sage Weil <sage@redhat.com>
2018-04-07 17:23:35 +08:00
Kefu Chai
008aefd723
Merge pull request #21095 from wido/ceph-metadata-version
...
common: Add metadata with only Ceph version number and release
Reviewed-by: John Spray <john.spray@redhat.com>
2018-04-07 17:22:01 +08:00
Kefu Chai
7e8d490f7d
Merge pull request #21055 from tchaikov/wip-pr-20994
...
filestore,rgw: fix types/casts making clang on 32-Bit working
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2018-04-07 17:21:06 +08:00
Willem Jan Withagen
3263b90e1f
mds: Clang does not (yet) support std::map::merge
...
@tchaikov found:
libc++ still does not support "Splicing Maps and Sets",
see https://libcxx.llvm.org/cxx1z_status.html ,
search for "p0083r3" .
Make this case the more explicit coded.
Signed-off-by: Willem Jan Withagen <wjw@digiware.nl>
2018-04-07 13:25:22 +08:00
Kefu Chai
2e45598d8a
cmake: detect if std::map::merge() is supported
...
Signed-off-by: Kefu Chai <kchai@redhat.com>
2018-04-07 13:25:19 +08:00
Kefu Chai
9e840c4382
Merge pull request #21274 from dzafman/wip-cot-config
...
tools: Use --no-mon-config so ceph_objectstore_tool.py test doesn't hang
Reviewed-by: Kefu Chai <kchai@redhat.com>
2018-04-07 11:59:28 +08:00
Sage Weil
43978fc68d
Merge PR #20760 into wip-sage-testing-20180407.023110
...
* refs/pull/20760/head:
mon/ConfigMonitor: add 'config reset' command
mon/ConfigMonitor: add 'config log' command
mon/ConfigMap: add ConfigChangeSet
mon/ConfigMonitor: avoid no-op updates
mon/ConfigMonitor: maintain history of all config changes
2018-04-06 21:31:19 -05:00
Sage Weil
5d0c06d5fe
Merge PR #20516 into wip-sage-testing-20180407.023003
...
* refs/pull/20516/head:
os/filestore: advertise vdo in filestore metadata
os/filestore: implement thin provisioning utilization stats for VDO
os/bluestore/KernelDevice: advertise vdo and physical_size in metadata
os/bluestore/KernelDevice: implement thin utilization stats for VDO
os/bluestore: hook to expose utilization of thinly-provisioned device
common/blkdev: identify VDO devices beneath LVM
common/blkdev: implement VDO helper functions
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2018-04-06 21:30:18 -05:00
Sage Weil
d7bca5d19f
Merge PR #20512 into wip-sage-testing-20180407.022843
...
* refs/pull/20512/head:
osd: osd_mon_report_interval_min -> osd_mon_report_interval, kill _max
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2018-04-06 21:29:32 -05:00
Sage Weil
aec47bc652
githubmap: update contributors
...
Signed-off-by: Sage Weil <sage@redhat.com>
2018-04-06 21:28:58 -05:00
Sage Weil
f67f3f87b7
Merge PR #21033 into wip-sage-testing-20180407.022623
...
* refs/pull/21033/head:
buffer: kill push_front
tools/RadosDump: avoid bufferlist::push_front
os/filestore/FileJournal: avoid push_front
erasure-code: avoid bufferlist::push_front
buffer: kill claim_prepend
unittest_compression: avoid claim_prepend
os/bluestore: avoid claim_prepend
osdc/Striper: rewrite assemble_result without push_front or claim_prepend
Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Jesse Williamson <jwilliamson@suse.de>
2018-04-06 21:27:20 -05:00
Sage Weil
f38bb6577d
githubmap: update contributors
...
Signed-off-by: Sage Weil <sage@redhat.com>
2018-04-06 21:27:19 -05:00
Nathan Cutler
bdb674639d
Merge pull request #19594 from smithfarm/wip-doc-make-check
...
doc: expand developer documentation of unit tests
Reviewed-by: Lenz Grimmer <lgrimmer@suse.com>
2018-04-06 22:19:03 +02:00
David Zafman
a8d26122dc
tools: Use --no-mon-config so ceph_objectstore_tool.py test doesn't hang
...
Signed-off-by: David Zafman <dzafman@redhat.com>
2018-04-06 11:52:10 -07:00
Gregory Farnum
6d2e4c9b7b
Merge pull request #19973 from liewegas/wip-peering-fast-dispatch
...
osd: fast dispatch of peering events and pg_map + osd sharded wq refactor
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
2018-04-06 11:48:11 -07:00
Sage Weil
577737d007
osd: osd_mon_report_interval_min -> osd_mon_report_interval, kill _max
...
The _max isn't used. Drop the _min suffix.
Signed-off-by: Sage Weil <sage@redhat.com>
2018-04-06 11:00:14 -05:00
Sage Weil
8f8f65594b
Merge pull request #21250 from liewegas/wip-recovery-slow-ops
...
qa/suites/rados/singleton/all/recovery_preemption: whitelist SLOW_OPS
2018-04-06 10:59:23 -05:00
Sage Weil
b235a3f62a
qa/suites/rados/singleton/all/ec-lost-unfound: whitelist SLOW_OPS
...
Signed-off-by: Sage Weil <sage@redhat.com>
2018-04-06 10:38:45 -05:00
Lenz Grimmer
95ee4e4142
Merge pull request #21239 from rjfd/wip-dashboard-routes
...
mgr/dashboard: using RoutesDispatcher as HTTP request dispatcher
Reviewed-by: Sebastian Wagner <sebastian.wagner@suse.com>
2018-04-06 16:36:34 +02:00
Joao Eduardo Luis
940dd941ef
Merge pull request #19331 from jecluis/wip-mon-osdmap-prune
...
mon: osdmap prune
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
2018-04-06 15:22:28 +01:00
Casey Bodley
072a9be2e0
Merge pull request #20464 from cbodley/wip-rgw-beast-ssl
...
rgw: add ssl support to beast frontend
Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
2018-04-06 09:52:55 -04:00
Nathan Cutler
a9ba2011f6
doc: dev: improve headings within Testing chapter
...
Signed-off-by: Nathan Cutler <ncutler@suse.com>
2018-04-06 14:18:37 +02:00
Lenz Grimmer
658e5d5a50
Merge pull request #21128 from ricardoasmarques/wip-usage-bar
...
mgr/dashboard: Add usage bar component
Reviewed-by: Tiago Melo <tmelo@suse.com>
Reviewed-by: Volker Theile <vtheile@suse.com>
2018-04-06 11:07:01 +02:00
Lenz Grimmer
6def989ade
Merge pull request #21240 from rjfd/wip-dashboard-task-fixes
...
mgr/dashboard: TaskManager bug fixes
Reviewed-by: Sebastian Wagner <sebastian.wagner@suse.com>
Reviewed-by: Ricardo Marques <rimarques@suse.com>
2018-04-06 10:50:45 +02:00
Joao Eduardo Luis
3997eed4db
qa: enable mon osdmap pruning on 'rados/' suites
...
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
2018-04-06 04:18:23 +01:00
Joao Eduardo Luis
65e7e4a38f
qa: enable mon osdmap pruning on rados/monthrash
...
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
2018-04-06 04:18:23 +01:00
Joao Eduardo Luis
2ffed4c98f
qa: mon: osdmap pruning standalone/workunit
...
Keep a standalone wrapper for the workunit, so we can test it locally,
leveraging the ceph-helpers to do the setup. Keep a workunit to be
exercised by teuthology.
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
2018-04-06 04:18:23 +01:00
Joao Eduardo Luis
6393c7ddeb
qa/standalone: change PATH to allow finding sysctl
...
If, by default, a user's PATH does not contain /sbin and /usr/sbin, we
may have a hard time finding sysctl to adjust the kernel core pattern.
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
2018-04-06 04:18:22 +01:00
Joao Eduardo Luis
f501458967
qa/standalone: deduplicate cleanup code
...
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
2018-04-06 04:18:22 +01:00
Joao Eduardo Luis
5ba6286834
qa/standalone: discard pipe in sysctl kernel.core_pattern
...
At least verified in openSUSE tumbleweed and ubuntu 16.04,
`sysctl -n kernel.core_pattern` returns a pipe at the start of the
value. We reset core_pattern to its original form once the script is
about to end, but if we do not discard the pipe the new value will
contain an extra pipe (apparently, the pipe is added automatically at
some point, possibly simply on read). We are simply stripping it on
read, as to prevent this behavior.
Additionally, we are also enclosing the reset of kernel.core_pattern in
quotes, so as to make sure patterns that include spaces (e.g., ubuntu's
apport, or tumbleweed's systemd-coredump) are properly reset.
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
2018-04-06 04:18:22 +01:00
Joao Eduardo Luis
5cd2b057f3
mon/OSDMonitor: allow triggering prunes from tick()
...
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
2018-04-06 04:18:22 +01:00
Joao Eduardo Luis
029807b3e9
mon/OSDMonitor: allow blocking trim if option is enabled
...
We need to allow blocking osdmap trimming to test osdmap pruning.
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
2018-04-06 04:18:22 +01:00
Joao Eduardo Luis
b64122ac37
mon/OSDMonitor: dump osdmap manifest alongside other infos
...
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
2018-04-06 04:18:22 +01:00
Joao Eduardo Luis
bc5df2b449
mon/OSDMonitor: prune full maps on encode_pending
...
If we have gone over a safety threshold for number of versions, start
pruning full osdmaps, poking holes in the sequence. To make up for the
missing maps, we will rebuild full maps based on the incremental maps
and the closest pinned full map we have available.
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
2018-04-06 04:18:22 +01:00
Joao Eduardo Luis
06f3ab684f
doc: add dev notes for mon's full osdmap pruning
...
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
2018-04-06 04:18:22 +01:00
Sage Weil
922a855817
osd: legacy pg_create creates pg in current epoch
...
For the legacy create path, we pregenerate the history and instantiate
the PG at the current epoch.
Signed-off-by: Sage Weil <sage@redhat.com>
2018-04-05 21:47:48 -05:00
Nathan Cutler
507b872858
Merge pull request #21127 from smithfarm/wip-make-dist-py3
...
build/ops: make-dist: fall back to python3
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Ricardo Dias <rdias@suse.com>
2018-04-06 04:00:46 +02:00
Patrick Donnelly
2156fd071c
Merge PR #21238 into master
...
* refs/pull/21238/head:
doc: fix typos and other minor errors in cephfs-nfs doc
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
2018-04-05 17:03:18 -07:00
Patrick Donnelly
fa430543d9
Merge PR #21204 into master
...
* refs/pull/21204/head:
mds: PendingReleaseNotes and dev doc for open file table
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
2018-04-05 17:02:20 -07:00
Casey Bodley
fe5fb47ba5
rgw: beast frontend calls shutdown before closing connections
...
Signed-off-by: Casey Bodley <cbodley@redhat.com>
2018-04-05 17:38:41 -04:00