]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Sage Weil [Thu, 17 Sep 2015 01:44:04 +0000 (21:44 -0400)]
mon: let peon mons send the osdmap replies
Currently the leader mon often replies to OSDs by sending a set of
incremental OSDmaps (e.g., in response to an osd boot or failure).
Instead, send a small message to the proxying peon mon (if any)
with the epoch to start from and let *them* generate a suitable
reply.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 6 Oct 2015 19:37:31 +0000 (15:37 -0400)]
msg/simple/Pipe: show keepalives at level 2
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 6 Oct 2015 19:35:58 +0000 (15:35 -0400)]
mon: set mon_subscribe_interval to a day
This is only needed for legacy clients to avoid confusing them--
we don't actually need the renewals at all. Make them infrequent
to reduce mon load.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 6 Oct 2015 19:25:02 +0000 (15:25 -0400)]
mon: only ack subscriptions (and renew) if client or mon is old
Old client expect an ack so they can schedule renewal; send it for
them only.
Old mons expect renewals.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 6 Oct 2015 19:19:33 +0000 (15:19 -0400)]
mon: remove old subscribe renewal-based timeouts
This is no longer needed/used.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 6 Oct 2015 19:18:21 +0000 (15:18 -0400)]
mon: small cleanup in _ms_dispatch
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 6 Oct 2015 19:11:03 +0000 (15:11 -0400)]
mon: new session_timeout mechanism that is not subscribe-based
Simplify the session liveness detection:
- renew on any message
- renew on keepalive[2] messages (lightweight ping in msgr)
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 6 Oct 2015 19:10:02 +0000 (15:10 -0400)]
msg: make last_keepalive[_ack] lock safe
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 6 Oct 2015 19:08:57 +0000 (15:08 -0400)]
msg: track stamp of last keepalive[2] received
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 6 Oct 2015 18:38:47 +0000 (14:38 -0400)]
common: mirror leveldb default tuning w/ rocksdb
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 6 Oct 2015 18:38:30 +0000 (14:38 -0400)]
mon/MonClient: don't send log if we're reconnecting
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Fri, 2 Oct 2015 13:15:33 +0000 (09:15 -0400)]
mon: disabled rocksdb compression when used as the backend
This significantly reduced CPU utilization on the bigbang scale
testing cluster at CERN. Note that it is already disabled for
leveldb by default (in ceph_mon.cc).
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Fri, 2 Oct 2015 13:06:29 +0000 (09:06 -0400)]
osd: cap adjusted max mon report interval at 2/3 of timeout
This ensures that we don't throttle back mon reports so much that
the mon times out out due to no pg stat reports. Since there is
little value is having a lower max anyway, just set this at an
upper bound (relative to the mon's timeout value).
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Wed, 30 Sep 2015 01:03:53 +0000 (21:03 -0400)]
osd: protect mon reporting with mon_report_lock
We need an exclusive lock over paths that update state related to
mon reports, lest they step on fields like up_thru_*, *stats_ack*,
last_mon_report, and so on. Everybody still needs a read lock
on map_lock too to get a stable OSDMap epoch.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Mon, 23 Nov 2015 13:38:44 +0000 (08:38 -0500)]
osd: fix reconnect behavior from booting state
We don't need to restart the boot process unless we are in preboot;
if we are in booting state we just need to resend the boot
message.
Signed-off-by: Sage Weil <sage@redhat.com>
Guang Yang [Tue, 22 Sep 2015 20:59:28 +0000 (20:59 +0000)]
osd: move the monitor report to OSD::tick_without_osd_lock
Fixes: #12722
Reviewed-by: Guang Yang <yguang@yahoo-inc.com>
Guang Yang [Tue, 29 Sep 2015 22:26:14 +0000 (22:26 +0000)]
osd: _got_mon_epochs - refactor the lock scope to avoid a race (which fail make check)
Reviewed-by: Guang Yang <yguang@yahoo-inc.com>
Sage Weil [Mon, 28 Sep 2015 21:22:01 +0000 (17:22 -0400)]
osd: don't send dup subscribes so much
The subscribe MonClient service is stateful--we don't need to
force a new subscribe send unless sub_want() says we need to.
Keep forcing it for instances where we request an *old* map.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Wed, 23 Sep 2015 21:58:15 +0000 (17:58 -0400)]
osd: introduce explicit preboot stage
We want to separate the stage where we do a bunch of work
prior to booting (but intend to eventually boot), like when we
get maps and wait to be healthy, from the point after we've sent
the boot message while we are just waiting for a response (so that
we can avoid resending that boot message needlessly).
- start at PREBOOT in start_boot()
- transition to BOOTING in _send_boot()
- only call _preboot() while in PREBOOT state
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Wed, 23 Sep 2015 21:31:50 +0000 (17:31 -0400)]
osd: skip osdmap version query if we can
If we get OSDmaps from the mon we *also* learn the oldest/newest
map epochs; no need to query again.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Wed, 23 Sep 2015 21:33:28 +0000 (17:33 -0400)]
osd: make [_]maybe_boot lockless variant
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 22 Sep 2015 15:16:15 +0000 (11:16 -0400)]
osd: only send boot if booting on getversion completion
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Fri, 18 Sep 2015 17:27:49 +0000 (13:27 -0400)]
osd: do not resend pg_temp requests
Send each pg_temp request once (per mon session); no need to
resend everything that is pending every time.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Fri, 18 Sep 2015 01:45:16 +0000 (21:45 -0400)]
osd: do not send dup failure reports
If a failure report is already pending, we do not need to resend
it.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Fri, 18 Sep 2015 01:48:30 +0000 (21:48 -0400)]
osd: resend pending failure reports with a new mon session
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Fri, 18 Sep 2015 01:42:53 +0000 (21:42 -0400)]
osd: fix send_failures() locking
It is unsafe to check failure_queue.empty() without the lock.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Thu, 17 Sep 2015 21:47:54 +0000 (17:47 -0400)]
osd: backoff the max reporting interval, too
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Thu, 17 Sep 2015 21:47:43 +0000 (17:47 -0400)]
osd: no need for regular send_pg_temps
This is done by process_peering_events.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Fri, 18 Sep 2015 18:24:27 +0000 (14:24 -0400)]
osd: just send alive when it is queue
No need to futz with last_mon_report or resend it again later.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Wed, 16 Sep 2015 15:00:57 +0000 (11:00 -0400)]
osd: fix pg stat reporting
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Fri, 18 Sep 2015 18:24:05 +0000 (14:24 -0400)]
osd: inline do_mon_report
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 15 Sep 2015 20:41:03 +0000 (16:41 -0400)]
osd: limit nubmer of pg stat updates in flight
There is no reason to heavily pipeline this. If the mon is slow
committing them we should go slow too.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 15 Sep 2015 20:34:34 +0000 (16:34 -0400)]
osd: fix pg_stats_queue lock protection
We are indirectly relying on osd_lock, but that may no longer
work for us in the future. Use the stats lock instead.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 15 Sep 2015 20:26:54 +0000 (16:26 -0400)]
osd: scale mon report interval with timeout backoff
If we have had to scale the backoff by 3x because the mon is
loaded, scale the min report interval accordingly.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 15 Sep 2015 20:26:06 +0000 (16:26 -0400)]
osd: keep count of outstanding pg stat updates to mon
Count how many stat updates we have in flight.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 15 Sep 2015 20:16:22 +0000 (16:16 -0400)]
osd: no stats outstanding when we reset the session
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 15 Sep 2015 20:15:29 +0000 (16:15 -0400)]
osd: remove old stats backoff mechanism
This would only backoff 2x the configured rate, and is less
robust than the new backoff + decay approach.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 15 Sep 2015 20:08:02 +0000 (16:08 -0400)]
osd: exponential backoff on pg stats ack timeout
If we don't get a timely response to our pg stats update we fail
the mon connection and reconnect to a new mon. If the mons aren't
responding because they are overloaded (for example, because they
are overwhelmed with stats updates) this just makes the problem
worse.
Mitigate the situation by doing an exponential backoff on the
timeout. When we do successfully send an update, slowly decay the
timeout back to the initial value.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 15 Sep 2015 19:38:52 +0000 (15:38 -0400)]
message/MLog: include seq in print
...so we can disambiguate which log message(s) we have.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Mon, 14 Sep 2015 21:11:31 +0000 (17:11 -0400)]
osd/OSDMap: cache values for in, up osds
We already do this for num_osd; do the same for the up and in
counts.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Mon, 14 Sep 2015 21:04:23 +0000 (17:04 -0400)]
mon/PGMonitor: avoid iterating over all pgs to find stale
Instead of iterating over all pgs when an osd goes down, make a
set of all osds that might have gone down, and only check pgs that
it manages. This is more efficient, especially for large clusters
with large numbers of OSDs.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Sat, 14 Nov 2015 02:14:56 +0000 (21:14 -0500)]
Merge pull request #6574 from yuyuyu101/fix-broken-kinects
kv/KineticStore: fix broken split_key
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil [Sat, 14 Nov 2015 00:27:45 +0000 (19:27 -0500)]
Merge pull request #6571 from dachary/wip-test-run-cli
tests: restore run-cli-tests
Loic Dachary [Fri, 13 Nov 2015 21:04:56 +0000 (22:04 +0100)]
Merge pull request #6578 from dachary/wip-13785-debian-rbd-replay
build/ops: rbd-replay moved from ceph-test-dbg to ceph-common-dbg
Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
Sage Weil [Fri, 13 Nov 2015 20:39:54 +0000 (15:39 -0500)]
Merge pull request #4737 from kylinstorage/wip-temp-based-object-eviction
osd: improve temperature calculation for cache tier agent
Reviewed-by: Sage Weil
Sage Weil [Fri, 13 Nov 2015 20:22:08 +0000 (15:22 -0500)]
Merge pull request #6422 from xiexingguo/xxg-wip13639
librados: fix potential null pointer access when do pool_snap_list
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Sage Weil [Fri, 13 Nov 2015 20:20:48 +0000 (15:20 -0500)]
Merge pull request #6486 from XinzeChi/wip-multiple-finisher
osd: FileStore: support multiple ondisk finish and apply finishers
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Samuel Just <sjust@redhat.com>
Sage Weil [Fri, 13 Nov 2015 20:20:13 +0000 (15:20 -0500)]
Merge pull request #6518 from kylinstorage/wip-trivial-optimization
osd: optimize scrub subset_last_update calculation
Reviewed-by: Sage Weil <sage@redhat.com>
Loic Dachary [Fri, 13 Nov 2015 18:10:28 +0000 (19:10 +0100)]
build/ops: rbd-replay moved from ceph-test-dbg to ceph-common-dbg
http://tracker.ceph.com/issues/13785 Fixes: #13785
Signed-off-by: Loic Dachary <loic@dachary.org>
Loic Dachary [Fri, 13 Nov 2015 17:47:31 +0000 (18:47 +0100)]
tests: avoid bashism
The shell used by the cli tests is not always bash. Not using the
here-word is also more readable in this specific case.
Signed-off-by: Loic Dachary <loic@dachary.org>
Jason Dillaman [Fri, 13 Nov 2015 15:20:19 +0000 (10:20 -0500)]
rbd: hardcode application name into help
Avoid dynamically detecting the application name and instead hardcode
the rbd CLI name into the help output.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
Haomai Wang [Fri, 13 Nov 2015 17:04:11 +0000 (01:04 +0800)]
kv/KineticStore: Fix broken split_key
Introduced by PR #6312
Signed-off-by: Haomai Wang <haomai@xsky.com>
Loic Dachary [Fri, 13 Nov 2015 15:23:27 +0000 (16:23 +0100)]
tests: restore run-cli-tests
e4ca468 moved src/test/run-cli-tests from check-local to check_SCRIPTS
but did not add it to the TESTS variable.
Signed-off-by: Loic Dachary <loic@dachary.org>
Sage Weil [Fri, 13 Nov 2015 15:06:18 +0000 (10:06 -0500)]
radosgw-admin: fix cli tests
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Fri, 13 Nov 2015 15:05:53 +0000 (10:05 -0500)]
osdmaptool: fix cli tests
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Fri, 13 Nov 2015 14:53:37 +0000 (09:53 -0500)]
crushtool: fix cli tests
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Fri, 13 Nov 2015 14:27:28 +0000 (09:27 -0500)]
crushtool: fix cli test help
Signed-off-by: Sage Weil <sage@redhat.com>
Loic Dachary [Fri, 13 Nov 2015 14:14:11 +0000 (15:14 +0100)]
Merge pull request #6532 from dachary/wip-mailmap
mailmap: Ubuntu Kylin name changed to Kylin Cloud
Reviewed-by: Li Wang <li.wang@kylin-cloud.com>
Sage Weil [Fri, 13 Nov 2015 14:04:47 +0000 (09:04 -0500)]
Merge pull request #5848 from storage-zuiwanyuan/wip-nonblock-connect
msg/async: support of non-block connect in async messenger
Reviewed-by: Haomai Wang <haomai@xsky.com>
Sage Weil [Fri, 13 Nov 2015 14:03:35 +0000 (09:03 -0500)]
Merge pull request #6478 from yuyuyu101/wip-13666
msg/async: let receiver ack message ASAP
Reviewed-by: Sage Weil <sage@redhat.com>
Jianhui Yuan [Fri, 13 Nov 2015 07:36:36 +0000 (15:36 +0800)]
msg/async: support of non-block connect in async messenger
Fixes: #12802
Signed-off-by: Jianhui Yuan <zuiwanyuan@gmail.com>
Kefu Chai [Fri, 13 Nov 2015 07:28:39 +0000 (15:28 +0800)]
Merge pull request #6534 from kylinstorage/wip-trivial-scrub-cleanup
osd: clarify the scrub result report
Reviewed-by: Kefu Chai <kchai@redhat.com>
Li Wang [Fri, 13 Nov 2015 07:00:09 +0000 (15:00 +0800)]
scrub: clarify the result report
It may happen that the authoritative object
such that auth.size != be_get_ondisk_size(auth_oi.size),
in that case, clarify the error report.
Signed-off-by: Li Wang <li.wang@kylin-cloud.com>
Josh Durgin [Fri, 13 Nov 2015 04:15:48 +0000 (20:15 -0800)]
Merge branch 'wip-py3'
pybind: a few more python 3 fixes for rbd and rados
Reviewed-by: David Coles <dcoles@gaikai.com>
Josh Durgin [Thu, 12 Nov 2015 08:57:36 +0000 (00:57 -0800)]
pybind/rados: return pool_reverse_lookup() result as a string
This makes it symmetric with create_pool() in python 3.
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
Josh Durgin [Thu, 12 Nov 2015 07:59:21 +0000 (23:59 -0800)]
pybind/test_rbd: convert a few more str to bytes for py3
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
Josh Durgin [Thu, 12 Nov 2015 07:57:16 +0000 (23:57 -0800)]
pybind/rbd: encode snap_rename args for py3
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
Josh Durgin [Thu, 12 Nov 2015 07:56:14 +0000 (23:56 -0800)]
pybind/rbd: decode stat() and list_children() results for py3
For stat(), only block_name_prefix is filled in - parent and
parent_pool are always blank.
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
Josh Durgin [Thu, 12 Nov 2015 03:05:59 +0000 (19:05 -0800)]
pybind/rbd: decode parent_info() to str types for py3
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
Josh Durgin [Thu, 12 Nov 2015 03:02:12 +0000 (19:02 -0800)]
pybind/test_rbd: fix map() usage for py3 compat
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
Josh Durgin [Thu, 12 Nov 2015 08:06:14 +0000 (00:06 -0800)]
pybind/test_rbd: use // for division for py3
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
Josh Durgin [Fri, 13 Nov 2015 03:32:42 +0000 (19:32 -0800)]
Merge branch 'pybind3' of https://github.com/dcoles/ceph into wip-pybind3
pybind: Add Python 3 support for rados and rbd modules
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Conflicts:
src/pybind/rbd.py (new create args, minor fix to work with py3)
David Coles [Wed, 11 Nov 2015 22:06:45 +0000 (14:06 -0800)]
ceph: Make stdout/stderr always output Unicode (UTF-8)
If a stream is not interactive, then under Python 2, then the encoding for
stdout/stderr may be None. This means that it's not possible to print Unicode
characters since the encoding will fall back to ASCII.
This explicitly makes sys.stdout/sys.stderr always use UTF-8 encoding for
strings, regardless of the system's local or if the console is interactive or
not.
This matches the existing tests that assume that output of non-ASCII pool names
will be UTF-8 encoded.
When outputting raw binary data (such as the CRUSH-map), we must bypass the
codec and write directly to raw streams (since the new stream will only accept
ASCII byte-strings or Unicode strings).
Signed-off-by: David Coles <dcoles@gaikai.com>
David Coles [Tue, 27 Oct 2015 20:32:44 +0000 (13:32 -0700)]
pybind: Add decode_cstr helper function
This function attempts to decode a C-style string into a Python Unicode string.
It accepts an optional "size" parameter for the string length, otherwise it is
assumed that the string is NUL-terminated.
If the pointer is NULL, then this function returns None.
Signed-off-by: David Coles <dcoles@gaikai.com>
David Coles [Tue, 20 Oct 2015 17:57:46 +0000 (10:57 -0700)]
pybind: Add test for creating pool by raw UTF-8
Some clients try providing non-ASCII pool names by sending raw encoded bytes.
This check ensures that we still support this behaviour for Python 2.
In Python 3, bytestrings will fail since strings are Unicode strings and thus
clients should use Unicode escapes instead.
Signed-off-by: David Coles <dcoles@gaikai.com>
David Coles [Tue, 20 Oct 2015 17:55:44 +0000 (10:55 -0700)]
pybind: Import cstr from the rados module
Since rados is required for rbd, we can avoid duplication of code across these
two modules.
Signed-off-by: David Coles <dcoles@gaikai.com>
David Coles [Tue, 20 Oct 2015 02:42:18 +0000 (19:42 -0700)]
pybind: Don't encode str on Python 2
If you attempt to call encode on a non-ASCII string, then a UnicodeDecodeError
will be raised.
Since str on Python 2 is an 8-bit string, it's possible that it's already UTF-8
encoded. As such we should just pass it through to the C API unmodified.
On Python 3 or if the user explicitly uses unicode, then we'll encode it to
UTF-8 for them.
Signed-off-by: David Coles <dcoles@gaikai.com>
Josh Durgin [Thu, 12 Nov 2015 22:08:31 +0000 (14:08 -0800)]
Merge branch 'wip-13504' of https://github.com/trociny/ceph
rbd: API: options on image create
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Conflicts:
src/test/librbd/test_librbd.cc (trivial, two tests added at end of file)
Mykola Golub [Fri, 23 Oct 2015 14:27:33 +0000 (17:27 +0300)]
librbd: API: options on image create: update tests
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Fri, 23 Oct 2015 09:38:08 +0000 (12:38 +0300)]
librbd: API: options on image create
Pass options as key/value configuration pairs on image create instead
of expanding rbd_create/rbd_clone/rbd_copy for every possible
configuration override.
Fixes: #13504
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Jason Dillaman [Thu, 12 Nov 2015 18:32:31 +0000 (13:32 -0500)]
Merge pull request #6557 from ceph/wip-rbd-cmake-fixes
cmake: librbd and libjournal build fixes
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Casey Bodley [Thu, 12 Nov 2015 17:34:08 +0000 (12:34 -0500)]
Merge pull request #6556 from athanatos/wip-cmake-nss
cmake: add nss as a suffix for pk11pub.h
Samuel Just [Wed, 11 Nov 2015 21:28:08 +0000 (13:28 -0800)]
cmake: add nss as a suffix for pk11pub.h
Signed-off-by: Samuel Just <sjust@redhat.com>
Ilya Dryomov [Thu, 12 Nov 2015 13:38:05 +0000 (14:38 +0100)]
librbd: provide an out-of-class definition for MAX_DESCRIPTION_OFFSET
std::min() takes its parameters by reference, so provide a definition
for OptionPrinter::MAX_DESCRIPTION_OFFSET constant.
Signed-off-by: Ilya Dryomov <idryomov@redhat.com>
Ilya Dryomov [Thu, 12 Nov 2015 13:52:47 +0000 (14:52 +0100)]
cmake: fix librbd and add src/journal
librbd was missing a recently added file, src/journal was missing
entirely - this brings the build back.
Signed-off-by: Ilya Dryomov <idryomov@redhat.com>
Loic Dachary [Wed, 11 Nov 2015 06:10:30 +0000 (07:10 +0100)]
mailmap: Ubuntu Kylin name changed to Kylin Cloud
Reviewed-by: Li Wang <li.wang@kylin-cloud.com>
Signed-off-by: Loic Dachary <loic@dachary.org>
Loic Dachary [Wed, 11 Nov 2015 06:07:07 +0000 (07:07 +0100)]
mailmap: sort files
Signed-off-by: Loic Dachary <loic@dachary.org>
Sage Weil [Thu, 12 Nov 2015 13:40:42 +0000 (08:40 -0500)]
Merge pull request #6517 from kylinstorage/wip-trivial-bug
osd: fix bug in last_* PG state timestamps
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
Jason Dillaman [Thu, 12 Nov 2015 13:21:21 +0000 (08:21 -0500)]
Merge pull request #6552 from trociny/wip-journal
journal: incremental improvements and fixes
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Loic Dachary [Thu, 12 Nov 2015 11:35:20 +0000 (12:35 +0100)]
Merge pull request #6462 from javacruft/wip/ec-modules
Build internal plugins and classes as modules
Reviewed-by: Loic Dachary <ldachary@redhat.com>
Mykola Golub [Thu, 29 Oct 2015 14:07:23 +0000 (16:07 +0200)]
journal: update allocated tid when skipping committed entry in player
Otherwise, if on image open, there are no any uncommitted entries in
journal, allocated tid is not updated to the latest commited and
recording always starts from tid=0.
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Kefu Chai [Thu, 12 Nov 2015 03:53:34 +0000 (11:53 +0800)]
Merge pull request #6539 from yehudasa/wip-cmake-fix
cmake: fix files list
Reviewed-by: Kefu Chai <kchai@redhat.com>
Josh Durgin [Wed, 11 Nov 2015 19:54:56 +0000 (11:54 -0800)]
Merge pull request #6316 from dillaman/wip-13494
rbd: dynamically generated bash completion
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Josh Durgin [Wed, 11 Nov 2015 19:51:49 +0000 (11:51 -0800)]
Merge remote-tracking branch 'origin/jewel'
Yehuda Sadeh [Wed, 11 Nov 2015 16:38:05 +0000 (08:38 -0800)]
cmake: fix files list
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
Mykola Golub [Thu, 8 Oct 2015 06:01:54 +0000 (09:01 +0300)]
cls::journal: fixup: constify dump functions
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Mon, 28 Sep 2015 18:37:05 +0000 (21:37 +0300)]
journal: call metadata shutdown on journal remove
This fixes error observed when disabling journal:
JournalMetadata: journal watch error: (107) Transport endpoint is not connected
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Thu, 17 Sep 2015 07:05:20 +0000 (10:05 +0300)]
journal: don't use object_number when comparing positions
Only tid does matter.
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Tue, 1 Sep 2015 20:09:52 +0000 (23:09 +0300)]
journal: make commit and flush params configurable
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Sun, 6 Sep 2015 14:06:44 +0000 (17:06 +0300)]
journal: allow alternate pool for journal objects
Signed-off-by: Mykola Golub <mgolub@mirantis.com>