Sage Weil
3f0ca4668e
Merge remote-tracking branch 'gh/giant'
2014-09-13 21:20:33 -07:00
Sage Weil
b285788c56
Merge pull request #2481 from sjahl/master
...
doc: fixes a formatting error on ops/crush-map
2014-09-13 12:46:24 -07:00
Stephen Jahl
b8a1ec08a1
doc: fixes a formatting error on ops/crush-map
...
Signed-off-by: Stephen Jahl <stephenjahl@gmail.com>
2014-09-13 15:31:53 -04:00
Loic Dachary
8d066732db
Merge pull request #2467 from majianpeng/fix3
...
buffer: In rebuild_page_aligned for the last ptr is page aligned, no need call rebuild().
Reviewed-by: Loic Dachary <loic-201408@dachary.org>
2014-09-13 17:56:16 +02:00
Loic Dachary
04e40737f1
Merge pull request #2478 from ceph/wip-9445
...
global: fix hang when segv happens inside logging code
Reviewed-by: Loic Dachary <loic-201408@dachary.org>
2014-09-13 17:32:57 +02:00
Yan, Zheng
499a73b3c3
Merge pull request #2477 from ceph/wip-client-msg-leak
...
client: fix a message leak
2014-09-13 08:42:15 +08:00
John Spray
c3c6468cad
mds: update segment references during journal rewrite
...
... to avoid leaving log events that reference log
segments by offsets which no longer exist.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 386f2d7c829422695a1b1f41bd3f17ca3eef1f61)
Reviewed-by: Greg Farnum <greg@inktank.com>
2014-09-12 17:39:53 -07:00
Gregory Farnum
e06f4251ac
Merge pull request #2469 from ceph/wip-9427-rewrite
...
mds: update segment references during journal rewrite
Reviewed-by: Greg Farnum <greg@inktank.com>
2014-09-12 17:35:54 -07:00
Sage Weil
a8c943a0e4
log: add simple test to verify an internal SEGV doesn't hang
...
Test that the segv injection works.
Test that a segv while logging something doesn't hang when the signal
handlers are installed. Note that this fails/hangs without the previous
fix.
Signed-off-by: Sage Weil <sage@redhat.com>
2014-09-12 17:18:01 -07:00
John Spray
2313ce1d02
client: fix a message leak
...
Signed-off-by: John Spray <john.spray@redhat.com>
2014-09-13 00:14:36 +01:00
Sage Weil
e3fe18aabe
global/signal_handler: do not log if SEGV originated inside log code
...
Signed-off-by: Sage Weil <sage@redhat.com>
2014-09-12 15:25:03 -07:00
Sage Weil
558463e815
log: add Log::is_inside_log_lock()
...
Signed-off-by: Sage Weil <sage@redhat.com>
2014-09-12 15:24:50 -07:00
John Spray
386f2d7c82
mds: update segment references during journal rewrite
...
... to avoid leaving log events that reference log
segments by offsets which no longer exist.
Signed-off-by: John Spray <john.spray@redhat.com>
2014-09-12 23:05:38 +01:00
John Spray
9f4c687288
Merge pull request #2471 from ceph/wip-9446
...
mon: fix MDS health detail output
Reviewed-by: Sage Weil <sage@redhat.com>
2014-09-12 16:47:52 +01:00
John Spray
9ba4e78f00
mon: fix MDS health detail output
...
I fat fingered a couple of things here.
Signed-off-by: John Spray <john.spray@redhat.com>
2014-09-12 16:43:20 +01:00
Ma Jianpeng
2b45bfa236
buffer: Add a test for bufferlist::rebuild_page_aligned
...
Make the last prt of bufferlist which is page-aligned don't change anything
after rebuild_page_aligned.
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com>
2014-09-12 22:18:08 +08:00
Ma Jianpeng
c268400bcb
buffer: In rebuild_page_aligned for the last ptr is page aligned, no need call rebuild().
...
This only happen for the last ptr. Because rebuild() don't change the len
of ptr, so if last ptr isn't page-size aligned but is page aligned, the
rebuild() don't change anything.
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com>
2014-09-12 22:14:49 +08:00
Loic Dachary
5283460382
Merge pull request #2468 from dachary/wip-always-create-pidfile
...
daemons: write pid file even when told not to daemonize
Reviewed-by: Loic Dachary <loic-201408@dachary.org>
Reviewed-by: Sage Weil <sage@redhat.com>
2014-09-12 16:00:27 +02:00
Alexandre Oliva
bccb0eb648
daemons: write pid file even when told not to daemonize
...
systemd wants to run daemons in foreground, but daemons wouldn't write
out the pid file with -f. Fixed.
Signed-off-by: Alexandre Oliva <oliva@gnu.org>
2014-09-12 12:05:56 +02:00
Loic Dachary
62ee5fc1f4
Merge pull request #2464 from dachary/wip-9429-bench
...
erasure-code: fix erasure_code_benchmark goop
Reviewed-by: Janne Grunau <j@jannau.net>
2014-09-12 11:30:31 +02:00
Sage Weil
e01e0c274e
Merge pull request #2416 from xiaoxichen/make_crush_private
...
Change CrushWrapper::crush to private
Reviewed-by: Sage Weil <sage@redhat.com>
2014-09-11 20:41:07 -07:00
Sage Weil
69d3594f9e
Merge pull request #2450 from dachary/wip-9413-erasure-code-version-check
...
erasure-code: mon, osd etc. depend on the plugins
2014-09-11 19:56:04 -07:00
Loic Dachary
b7bdb93ff7
erasure-code: fix erasure_code_benchmark goop
...
Using a stringstream that is only displayed on error when calling the
erasure code factory, instead of cerr. The user expects the output to be
clean when there is no error.
http://tracker.ceph.com/issues/9429 Fixes : #9429
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
2014-09-11 22:24:53 +02:00
Loic Dachary
60d65daa96
Merge pull request #2409 from apeters1971/wip-ec-isa-table-cache-refac-master
...
EC-ISA: add intelligent table cache
Reviewed-by: Loic Dachary <loic-201408@dachary.org>
2014-09-11 21:56:19 +02:00
John Wilkins
2fbe82dfeb
doc: Fixed syntax error.
...
Signed-off-by: John Wilkins <jowilki@redhat.com>
2014-09-11 10:50:42 -07:00
John Wilkins
7dbf750fc5
doc: Updated authentication notes. Fixed syntax error.
...
Signed-off-by: John Wilkins <jowilki@redhat.com>
2014-09-11 10:50:22 -07:00
Loic Dachary
9e6d05afb4
Merge pull request #2459 from ceph/wip-7934
...
test: Fix ceph_test_rados_watch_notify to delete the pools it creates
Reviewed-by: Loic Dachary <loic-201408@dachary.org>
2014-09-11 19:33:46 +02:00
John Spray
dcc608d5d3
Merge pull request #2463 from ceph/wip-mds-beacon
...
mds: a couple fixes for the beacons
Reviewed-by: John Spray <john.spray@redhat.com>
2014-09-11 16:45:21 +01:00
Sage Weil
7281638d4a
mds: sleep in progress thread if laggy and waiting_for_nolaggy waiters
...
If we have nolaggy waiters but are laggy we want to sleep. Otherwise,
we will just spin and spam the log ...
Signed-off-by: Sage Weil <sage@redhat.com>
2014-09-11 08:35:49 -07:00
Sage Weil
6fb5769aee
mds/Beacon: do not reconnect to mon in quick succession
...
Wait at least one beacon interval between mon session resets.
Fixes : #9428
Signed-off-by: Sage Weil <sage@redhat.com>
2014-09-11 08:35:49 -07:00
Yan, Zheng
c930624163
Merge pull request #2460 from ceph/wip-client-ll-ref
...
client: include ll_ref when printing inode
2014-09-11 17:06:39 +08:00
Yan, Zheng
4ad5db0050
client: include ll_ref when printing inode
...
Signed-off-by: Yan, Zheng <zyan@redhat.com>
2014-09-11 17:03:55 +08:00
Sage Weil
2235aa990a
Merge pull request #2444 from wonzhq/read-recency
...
osd: set min_read_recency_for_promote to default 1 when doing upgrade
Reviewed-by: Sage Weil <sage@redhat.com>
2014-09-10 20:44:13 -07:00
Sage Weil
59f1a49b3e
Merge pull request #2449 from majianpeng/fix3
...
fix two bugs about perfcounter
Reviewed-by: Sage Weil <sage@redhat.com>
2014-09-10 20:37:03 -07:00
David Zafman
cf70b906ae
test: Fix ceph_test_rados_watch_notify to delete the pools it creates
...
Fixes : #7934
Signed-off-by: David Zafman <dzafman@redhat.com>
2014-09-10 19:19:08 -07:00
Ma Jianpeng
2e4e98bd08
ReplicatedPG: Make perfcounter record the read-size for
...
async-read.
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com>
2014-09-11 08:32:06 +08:00
Ma Jianpeng
cf34e00d49
ReplicatedPG: record correctly subop for perfcounter.
...
In log_subop_stats, it omit to record the counter of subop.
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com>
2014-09-11 08:09:47 +08:00
Sage Weil
e50c426bef
Merge pull request #2454 from athanatos/wip-9269
...
FileStore: report l_os_j_lat as commit latency
Reviewed-by: Sage Weil <sage@redhat.com>
2014-09-10 12:15:19 -07:00
Sage Weil
6846811ed3
Merge pull request #2453 from athanatos/wip-9220
...
Objecter::_recalc_linger_op: resend for any acting set change
Reviewed-by: Sage Weil <sage@redhat.com>
2014-09-10 12:09:34 -07:00
Samuel Just
dc5e98d3e4
Merge pull request #2443 from ceph/wip-9241
...
osdc/Objecter: drop bad session nref assert
Reviewed-by: Samuel Just <sam.just@inktank.com>
2014-09-10 12:09:04 -07:00
Samuel Just
d0feeaa150
Merge pull request #2293 from ceph/wip-hitset-bytes
...
osd: improve agent calculation by factoring out hit_set bytes used properly
Reviewed-by: Samuel Just <sam.just@inktank.com>
2014-09-10 12:02:56 -07:00
Loic Dachary
117de007ef
erasure-code: mon, osd etc. depend on the plugins
...
Since the erasure code plugin version check has been introduced,
whenever a library/binary that can load plugin needs to be recompiled,
the erasure code plugins must also be considered. If the reason for
recompiling the library/binary is a new commit, the plugins will fail to
load.
The dependency is not based on source compilation and a shared library
dependency on liberasure-code.la is added instead. This library is
uniformly used whenever a plugin is to be loaded and therefore covers
all library/binaries that need it.
http://tracker.ceph.com/issues/9413 Fixes : #9413
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
2014-09-10 20:49:00 +02:00
Sage Weil
c6678eb57a
Merge pull request #2451 from ceph/wip-osdc-leak
...
osdc/Objecter: fix leak of MStatfsReply
Reviewed-by: John Spray <john.spray@redhat.com>
2014-09-10 11:31:27 -07:00
Yehuda Sadeh
bead44cd98
Merge pull request #2447 from reclosedev/s3_colon_in_access_key
...
[rgw][s3] Allow colon ':' in access key
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
2014-09-10 09:59:53 -07:00
Roman Haritonov
5a05e6b8b9
[rgw][s3] Allow colon ':' in access key
...
When access key contains ':', e.g. `some_info:for_user',
authorization header looks like:
"AWS some_info:for_user:request_signature"
so `auth_str.find(':')` result in auth_id = "some_info",
auth_sign = "for_user:request_signature".
auth_str.rfind(':') solves this issue.
Signed-off-by: Roman Haritonov <reclosedev@gmail.com>
2014-09-10 20:45:52 +04:00
Sage Weil
46bbe30e68
osdc/Objecter: fix leak of MStatfsReply
...
Signed-off-by: Sage Weil <sage@redhat.com>
2014-09-10 06:57:12 -07:00
John Spray
136c5e1104
Merge pull request #2448 from ceph/wip-replay-locking
...
mds: fix replay locking
Reviewed-by: John Spray <john.spray@redhat.com>
2014-09-10 14:56:31 +01:00
Andreas-Joachim Peters
1ff94ccf89
EC-ISA: avoid usage of table cache lock outside the class implementation by introducing the setEncodingTable/setEncodingCoefficient methods
2014-09-10 11:25:05 +02:00
Andreas-Joachim Peters
5b41dacfd9
EC-ISA: add intelligent table cache
2014-09-10 11:25:05 +02:00
Yan, Zheng
83bd3430e3
mds: fix replay locking
...
When replaying EImportFinish/EFragment event, the replay thread may call
MDS::queue_waiters. MDS::queue_waiters() requires its caller to hold the
mds_lock. Otherwise assert(waiter_mutex == __null || waiter_mutex->is_locked())
in Cond::Signal() will be tiggered.
Signed-off-by: Yan, Zheng <zyan@redhat.com>
2014-09-10 16:48:47 +08:00