Sage Weil
bb99b51eff
osd/osd_types: fix PastIntervals::update_type()
...
Signed-off-by: Sage Weil <sage@redhat.com>
2017-04-28 11:30:39 -04:00
Sage Weil
2fd345188b
osd/osd_types: fix pi_compact_rep empty() condition
...
We may have no intervals but still be non-empty (have a first and
last), because during that period there were no osds. This easily
happens when a pool is created before osds are up.
Signed-off-by: Sage Weil <sage@redhat.com>
2017-04-28 11:30:39 -04:00
Sage Weil
1c15c18896
osd/osd_types: fix pi_compact_rep bounds; rename fields
...
The second bound is non-inclusive. Use 'first' and 'last' instead of
'start' and 'end' to reinforce that.
Signed-off-by: Sage Weil <sage@redhat.com>
2017-04-28 11:30:39 -04:00
Sage Weil
f2a13dc0d7
osd/PG: put interval bounds [) in debug/log output
...
Signed-off-by: Sage Weil <sage@redhat.com>
2017-04-28 11:30:39 -04:00
Sage Weil
0e94e17740
osd/PG: log peering intervals at high debug level
...
Because we are no longer explicitly tracking all past intervals
I think it is a good idea to log these more aggressively. This will
give us potentially vital information when debugging peering problems
(says someone who just debugged a peering problem in which the past
interval information provided a vital clue).
Signed-off-by: Sage Weil <sage@redhat.com>
2017-04-28 11:30:39 -04:00
Sage Weil
a9d7911913
unittest_osd_types: minor fix to past_intervals tests
...
Set [up_]primary to valid values (not that it is actually
relevant to the test!).
Signed-off-by: Sage Weil <sage@redhat.com>
2017-04-28 11:30:39 -04:00
Samuel Just
6505175684
test/osd/types: add tests specifically for prior set generation
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:39 -04:00
Samuel Just
44264de239
test/osd/types.cc: fix check_new_interval test
...
Removes the asserts relating to past_intervals state.
They don't really make sense with the new representation,
we'll add new tests for that next.
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:39 -04:00
Samuel Just
5a9c3be3c7
osd_types: refactor PriorSet so it can be tested without an OSDMap
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:39 -04:00
Samuel Just
6f1783d9ba
osd_types: just pass last_epoch_started to PriorSet
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:38 -04:00
Samuel Just
acb713b48c
osd_types: fill in generate_past_intervals for new types
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:38 -04:00
Samuel Just
867fd4e790
osd_types: implement PastIntervals pi_compact_rep
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:38 -04:00
Samuel Just
3ed85f0f2e
PG: accommodate new past_intervals encoding
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:38 -04:00
Samuel Just
aeee4bc8ce
RadosDump: accommodate new past_intervals encoding
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:38 -04:00
Samuel Just
8e4856041d
messages/: accomodate the new past_intervals encoding
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:38 -04:00
Samuel Just
f52313c461
osd_types: generalize PastIntervals to support multiple representations
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:38 -04:00
Samuel Just
0baebe166b
osd/: hide pg_interval_t encoder
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:38 -04:00
Samuel Just
065bb89ca6
osd/: refactor past_intervals logic into PastIntervals
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:38 -04:00
Samuel Just
e4c7ecea4a
PG::GetInfo: remove logic for detecting intervals without complete peers
...
c7d92d1d3f
introduced this back when
the acting set could contain incomplete peers during backfill. That
hasn't been true since dumpling. Now, any interval where the acting
set contains an incomplete peer cannot possibly go active. Thus, it
can't change last_epoch_started or history.last_epoch_started. Thus,
even though choose_acting omits incomplete peers, the answer can't
change.
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:38 -04:00
Samuel Just
7485f9e7cc
test/osd_types.cc: fix emacs line at the top
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-04-28 11:30:38 -04:00
Sage Weil
3ab8dff07f
qa/suites/upgrade/jewel-x: add cache tiering + snaps workload
...
Signed-off-by: Sage Weil <sage@redhat.com>
2017-04-28 11:30:38 -04:00
Mykola Golub
59a48daec3
Merge pull request #14848 from dillaman/wip-19798
...
test: remove hard-coded image name from RBD metadata test
Reviewed-by: Mykola Golub <mgolub@mirantis.com>
2017-04-28 17:26:39 +03:00
Sage Weil
a07cacb9e3
Merge pull request #13315 from dang/wip-build-no-nss
...
cmake: Allow tests to build without NSS
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-04-28 08:47:46 -05:00
Sage Weil
bce9c53144
Merge pull request #14367 from mslovy/wip-fix-infinitloops
...
os/filestore: fix infinit loops in fiemap()
Reviewed-by: Sage Weil <sage@redhat.com>
2017-04-28 08:47:05 -05:00
Sage Weil
e9fb6a06e0
Merge pull request #14391 from LiumxNL/wip-170407
...
mon/PGMonitor: fix wrongly report "pg stuck in inactive"
Reviewed-by: Sage Weil <sage@redhat.com>
2017-04-28 08:46:47 -05:00
Sage Weil
2944c48d3f
Merge pull request #14675 from ifed01/wip-bluestore-nolock
...
os/bluestore: eliminate some excessive stuff
Reviewed-by: Sage Weil <sage@redhat.com>
2017-04-28 08:46:23 -05:00
Sage Weil
0530a24bb6
Merge pull request #14475 from liewegas/wip-message-compat-cruft
...
messages: remove compat cruft
Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-04-28 08:30:19 -05:00
Igor Fedotov
6b457d168b
os/bluestore: move TransContext finalization out of osr_lock to avoid potential contention.
...
Signed-off-by: Igor Fedotov <ifedotov@mirantis.com>
2017-04-28 04:30:25 -07:00
Igor Fedotov
ad9d87c87a
os/bluestore: get rid off Onode::flush_txn set.
...
It looks like flushing_count counter is enough.
Signed-off-by: Igor Fedotov <ifedotov@mirantis.com>
2017-04-28 04:29:52 -07:00
Kefu Chai
1d642f248b
Merge pull request #14840 from drunkard/master
...
doc: style fix for doc/cephfs/client-config-ref.rst
Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-04-28 10:57:53 +08:00
Drunkard Zhang
ab0310883e
doc: typo fixes in doc/cephfs/client-config-ref.rst
...
fixed:
Reference symbol, i.e. s/`/``/
Hyperlink typo at ``client_trace``
Signed-off-by: Drunkard Zhang <gongfan193@gmail.com>
2017-04-28 09:53:17 +08:00
Sage Weil
ebabb5fe77
Merge pull request #14302 from liewegas/wip-19452
...
osdc/Objecter: handle command target that goes down
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2017-04-27 20:02:38 -05:00
Kefu Chai
e15fda3e84
Merge pull request #14850 from tchaikov/wip-cmake
...
cmake: add "container" to required boost components
Reviewed-by: Casey Bodley <cbodley@redhat.com>
2017-04-28 08:57:58 +08:00
Brad Hubbard
01625a91b5
Merge pull request #14794 from joscollin/wip-warning-simplespin
...
test: fixing assert that creates warning: comparison between signed and unsigned integer expressions
Reviewed-by: Brad Hubbard <bhubbard@redhat.com>
2017-04-28 10:22:39 +10:00
Kefu Chai
b32f1813bb
cmake: add "container" to required boost components
...
it is used by our denc
Signed-off-by: Kefu Chai <kchai@redhat.com>
2017-04-28 05:09:08 +08:00
Sage Weil
a001e0bfa4
Merge pull request #14843 from wjwithagen/wip-wjw-fix-pipe
...
src/msg/simple/Pipe.cc: Fix the inclusion of '}'
Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-04-27 15:51:25 -05:00
Jason Dillaman
8f72e745e1
test: remove hard-coded image name from RBD metadata test
...
Fixes: http://tracker.ceph.com/issues/19798
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2017-04-27 16:24:02 -04:00
Willem Jan Withagen
cdc067e06c
src/msg/simple/Pipe.cc: Fix the inclusion of '}'
...
Signed-off-by: Willem Jan Withagen <wjw@digiware.nl>
2017-04-27 20:53:49 +02:00
Jason Dillaman
b67c5c7aa4
Merge pull request #14754 from trociny/wip-19128
...
rbd: import needs to sanity check auto-generated image name
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
2017-04-27 14:40:51 -04:00
Yehuda Sadeh
f929d4f798
Merge pull request #14592 from prallabh/wip-manifest
...
rgw: custom user data header
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
2017-04-27 11:25:00 -07:00
Casey Bodley
bac49f33ce
Merge pull request #14816 from cbodley/wip-rgw-test-multi-list-fix
...
test/rgw: fix for empty lists as default arguments
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
2017-04-27 13:38:15 -04:00
Sage Weil
9fa79ba205
Merge pull request #14473 from ShiqiCooperation/master
...
os/bluestore: assert blob map returns success
Reviewed-by: Sage Weil <sage@redhat.com>
2017-04-27 12:33:06 -05:00
Sage Weil
930d2151c6
Merge pull request #14491 from liewegas/wip-bluestore-deferred-rewrite
...
os/bluestore: rewrite deferred write handling
Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>
2017-04-27 12:32:52 -05:00
Sage Weil
05bb3b5360
Merge pull request #14695 from dachary/wip-choosleaf-stable
...
crush: builder: legacy has chooseleaf_stable = 0
Reviewed-by: Sage Weil <sage@redhat.com>
2017-04-27 12:32:31 -05:00
Sage Weil
528c6e067a
Merge pull request #14696 from dachary/wip-crush-workspace
...
crush: crush_init_workspace starts with struct crush_work
Reviewed-by: Sage Weil <sage@redhat.com>
2017-04-27 12:32:21 -05:00
Sage Weil
559a7a83ee
Merge pull request #14814 from neha-ojha/wip-osd-filestore-default
...
osd: switch filestore to default to rocksdb
Reviewed-by: Sage Weil <sage@redhat.com>
2017-04-27 12:32:09 -05:00
Kefu Chai
8fb1af0aba
Merge pull request #14795 from tchaikov/wip-so-priority-cleanup
...
msg/async: return right away in NetHandler::set_priority() if not supported
Reviewed-by: Haomai Wang <haomai@xsky.com>
2017-04-28 00:48:23 +08:00
Sage Weil
40699a61b2
Merge pull request #14802 from tchaikov/wip-mgr-cleanup
...
mgr: various cleanups
Reviewed-by: Sage Weil <sage@redhat.com>
2017-04-27 11:43:32 -05:00
Sage Weil
add2015fe3
Merge pull request #14807 from liewegas/wip-jewel-x
...
ceph_test_rados_api_tier: tolerate ENOENT from 'pg scrub'
Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-04-27 11:43:12 -05:00
Yuri Weinstein
b9d9eb9933
Merge pull request #14717 from liewegas/wip-bluestore-remove-max-ops
...
os/bluestore: rename/fix throttle options
Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>
2017-04-27 09:35:09 -07:00