Yehuda Sadeh
f79d8f24e9
Merge remote-tracking branch 'origin/wip-bi-sharding-3' into next
2015-01-19 09:14:32 -08:00
Josh Durgin
ac93c28522
Merge pull request #3396 from leseb/doc-openstack-fix-glance
...
doc: Fix OpenStack Glance configuration
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2015-01-19 08:38:37 -08:00
Sage Weil
ce1e1fe4b2
Merge pull request #3401 from FrankYu/master
...
Doc: rbd-snapshot: Fix the typo
2015-01-19 08:34:32 -08:00
Loic Dachary
b5e79a8842
Merge pull request #3374 from dachary/wip-mailmap
...
mailmap updates
Reviewed-by: Loic Dachary <ldachary@redhat.com>
2015-01-19 17:06:46 +01:00
John Spray
66f3bd6278
tools: output per-event errors from recover dentries
2015-01-19 15:16:46 +00:00
John Spray
3e15fd5f06
tools: handle hardlinks in recover_dentries
...
Signed-off-by: Johh Spray <john.spray@redhat.com>
2015-01-19 14:09:34 +00:00
John Spray
9d91757559
tools: recover_dentries efficiency
...
Avoid a redundant stat, and gather updates to a frag
into a single OMAP get/set.
Still could be heaps more efficient in the case of
many updates to the same dirs by adding in a little
cache and batching the updates.
Signed-off-by: John Spray <john.spray@redhat.com>
2015-01-19 14:09:34 +00:00
John Spray
e2ae91fc70
tweak comment wording in recover_dentries
2015-01-19 14:09:34 +00:00
John Spray
5f348313a5
fixup some oversized lines
2015-01-19 14:09:34 +00:00
John Spray
1dc023b36a
fix handling of io.read retval
...
(it returns length read, which was falling through as
a spurious nonzero "error")
2015-01-19 14:09:34 +00:00
John Spray
d94a7a8fdd
tools: remove duplicated InoTable encoding
...
...and add a method to InoTable so that we can
artifically acquire inodes.
Signed-off-by: John Spray <john.spray@redhat.com>
2015-01-19 14:09:34 +00:00
John Spray
478659f34a
JournalTool: handle corrupt fnodes
2015-01-19 14:09:34 +00:00
John Spray
36830ebc3e
tools/cephfs: add recover_dentries to journaltool
...
This is intended as a comparatively safe recovery
operation, where we compare the versions
of journalled dentries with backing store dentries,
and write into the backing store only when the
existing contents are older than the journal
or invalid.
Fixes : #9883
Signed-off-by: John Spray <john.spray@redhat.com>
2015-01-19 14:09:34 +00:00
John Spray
46326ffb43
mds: perfcounters for RecoveryQueue
...
Define these at MDCache scope, as RecoveryQueue is
not an official subsystem to be made visible to
external parties.
Signed-off-by: John Spray <john.spray@redhat.com>
2015-01-19 13:52:34 +00:00
John Spray
62b4e43916
mds/MDCache: stray/purge perfcounters
...
Fixes : #10388
Signed-off-by: John Spray <john.spray@redhat.com>
2015-01-19 13:06:45 +00:00
Frank Yu
4bd0894cc7
Doc: rbd-snapshot: Fix the typo
...
Signed-off-by: Frank Yu <flyxiaoyu@gmail.com>
2015-01-19 20:19:25 +08:00
Yunchuan Wen
07e1d18f4b
rbd: fix bug about rbd watch command
...
the header oid should be prefix+image_id, rather than prefix+image_name
Signed-off-by: Yunchuan Wen <yunchuanwen@ubuntukylin.com>
2015-01-19 09:04:47 +00:00
Sage Weil
751e504275
Merge pull request #3397 from liewegas/wip-prealloc
...
mon: fix globalid when prealloc value is larger than max
2015-01-18 20:46:31 -08:00
Sage Weil
bbaf582e5b
mon: handle case where mon_globalid_prealloc > max_global_id
...
This triggers with the new larger mon_globalid_prealloc value. It didn't
trigger on the existing cluster I tested on because it already had a very
large max.
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-18 16:49:20 -08:00
Sébastien Han
d29aef5eb9
doc: Fix OpenStack Glance configuration
...
Glance has not completely moved to 'store' yet so we need to configure
the store in the DEFAULT section as well.
Fixes : #10478
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-18 22:55:57 +01:00
Sage Weil
8379b454b9
Merge pull request #3361 from wonzhq/watch-notify
...
osd/ReplicatedPG: force promotion for watch/notify ops
Reviewed-by: Sage Weil <sage@redhat.com>
2015-01-18 10:44:35 -08:00
Sage Weil
f078a429fd
Merge pull request #3395 from liewegas/wip-cache-none
...
osd: skip all of maybe_handle_cache if cachemode is none
2015-01-18 10:43:58 -08:00
Sage Weil
e7f9bffa50
Merge pull request #3315 from majianpeng/fix6
...
bug fix
Reviewed-by: Sage Weil <sage@redhat.com>
2015-01-18 10:42:31 -08:00
Sage Weil
ba73ef9378
Merge pull request #3385 from majianpeng/misc
...
Misc
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2015-01-18 10:41:55 -08:00
Sage Weil
31eb4c615a
Merge pull request #3194 from dachary/wip-10350-erasure-code-choose-tries
...
resolve and document most common erasure coded pool pain points
Documentation-Reviewed-by: Italo Santos <okdokk@gmail.com>
2015-01-18 10:41:31 -08:00
Sage Weil
1d1215fe5f
mon: change mon_globalid_prealloc to 10000 (from 100)
...
100 ids (session 100 authentications) can be consumed quite quickly if
the monitor is being queried by the CLI via scripts or on a large cluster,
especially if the propose interval is long (many seconds). These live in
a 64-bit value and are only "lost" if we have a mon election before they
are consumed, so there's no real risk here.
Backport: giant, firefly
Reviewed-by: Joao Eduardo Luis <joao@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-18 10:39:25 -08:00
Sage Weil
be1416a044
mon: silently ignore mark_down, mark_disposable on AnonConnection
...
This mirrors 0a49db8e6f
but was not captured
by 9fff0c53bd
.
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-18 10:37:51 -08:00
Sage Weil
87b3e9600a
Merge remote-tracking branch 'gh/wip-xio'
...
Reviewed-by: Sage Weil <sage@redhat.com>
2015-01-18 10:37:19 -08:00
Sage Weil
78fd69f1aa
Merge pull request #3391 from liewegas/wip-pool-delete
...
mon: global option to prevent pool deletion
Reviewed-by: John Spray <john.spray@redhat.com>
2015-01-18 10:34:59 -08:00
Sage Weil
b5e1bd1fab
mon/OSDMonitor: pull totals into the table format
...
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-17 17:14:48 -08:00
Sage Weil
169d6805ff
mon/OSDMonitor: don't -nan for variance when no osds
...
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-17 17:14:48 -08:00
Sage Weil
9aa0006343
common/TextTable: do padding during final step
...
We don't know the full width until we get all table entries. Do not do
any padding before that. In particular, setw may left-pad on a column
that is left-justified.
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-17 17:14:48 -08:00
Sage Weil
ea5166ef2a
mon/OSDMonitor: drop unnecessary stringify
...
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-17 17:14:48 -08:00
Sage Weil
8a0d124124
crushtool: fix cli tests for new tree output
...
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-17 17:14:48 -08:00
Sage Weil
4698882382
osd/ReplicatedPG: skip all of maybe_handle_cache if caching is off
...
Return quickly and avoid all of the checks.
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-17 10:31:10 -08:00
Sage Weil
04c09ed276
crushtool: add --tree option
...
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-17 08:58:40 -08:00
Sage Weil
871c3ec597
crush/CrushWrapper: drop weights from crush dump_tree
...
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-17 08:58:40 -08:00
Sage Weil
86729e7970
osd, crush: caps for headers
...
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-17 08:57:26 -08:00
Sage Weil
022950ecbf
crush/CrushTreeDumper: add missing close_section
...
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-17 08:57:26 -08:00
Sage Weil
7a04518498
crush, osdmap: use weightf_t helper for all weights
...
Signed-off-by: Sage Weil <sage@redhat.com>
2015-01-17 08:57:26 -08:00
Mykola Golub
9e22f6179b
osd: use TextTable for osd tree command output
...
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
2015-01-17 08:57:26 -08:00
Mykola Golub
1d984a5239
osd: use weightf_t type when outputing weights by osd tree command
...
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
2015-01-17 08:57:26 -08:00
Mykola Golub
4bae3a8db3
mon: ceph osd df -- a new command to show OSD utilization
...
Fixes : #10452
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
2015-01-17 08:57:26 -08:00
Mykola Golub
b42c7e28f0
Add a type (weightf_t) to output float weights.
...
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
2015-01-17 08:57:26 -08:00
Mykola Golub
3e2eaa4ba0
osd: use CrushTreeDumper for printing osd tree.
...
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
2015-01-17 08:57:26 -08:00
Mykola Golub
3f9b4d1f33
crush: use CrushTreeDumper for dumping crush tree.
...
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
2015-01-17 08:57:26 -08:00
Mykola Golub
26921f43cb
crush: introduce a helper class for dumping crush tree.
...
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
2015-01-17 08:57:26 -08:00
Haomai Wang
4d0e0ae3b5
AsyncConnection: Fix memory leak for AsyncConnection
...
*_handler will store a reference to AsyncConnection, it need to explicit reset
it.
Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
2015-01-17 19:52:19 +08:00
Josh Durgin
a76e537630
Merge remote-tracking branch 'origin/next'
2015-01-16 14:40:27 -08:00
Josh Durgin
896c8899ac
Merge remote-tracking branch 'origin/wip-10271' into next
...
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2015-01-16 14:33:59 -08:00