]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Dan Mick [Fri, 4 Apr 2014 22:26:42 +0000 (15:26 -0700)]
ceph-post-file: use getopt for multiple options, add longopts to help
Fixes: #7993
Signed-off-by: Dan Mick <dan.mick@inktank.com>
Sage Weil [Fri, 4 Apr 2014 01:28:12 +0000 (18:28 -0700)]
doc/release-notes: v0.79 release notes
Signed-off-by: Sage Weil <sage@inktank.com>
Dan Mick [Thu, 3 Apr 2014 20:59:59 +0000 (13:59 -0700)]
Fix byte-order dependency in calculation of initial challenge
Fixes: #7977
Signed-off-by: Dan Mick <dan.mick@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Fri, 4 Apr 2014 00:44:13 +0000 (17:44 -0700)]
Merge pull request #1599 from ceph/wip-7978
rgw: only look at next placement rule if we're not at the last rule
Reviewed-by: Sage Weil <sage@inktank.com>
Yehuda Sadeh [Thu, 3 Apr 2014 22:15:41 +0000 (15:15 -0700)]
rgw: only look at next placement rule if we're not at the last rule
Fixes: #7978
We tried to move to the next placement rule, but we were already at the
last one, so we ended up looping forever.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Sage Weil [Thu, 3 Apr 2014 18:45:21 +0000 (11:45 -0700)]
Merge pull request #1598 from ceph/wip-test-alloc-hint-ec-fix
qa: test_alloc_hint: set ec ruleset-failure-domain to osd
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 3 Apr 2014 18:44:29 +0000 (11:44 -0700)]
Merge pull request #1581 from ceph/wip-init
a few deb changes
Ilya Dryomov [Thu, 3 Apr 2014 17:03:53 +0000 (21:03 +0400)]
qa: test_alloc_hint: set ec ruleset-failure-domain to osd
Create a custom profile with ruleset-failure-domain=osd. (The default
ruleset-failure-domain=host won't do because this script assumes and
works only if all osds are on the same host.) While at it, set k and m
explicitly to avoid troubles in the future.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Sage Weil [Thu, 3 Apr 2014 16:57:49 +0000 (09:57 -0700)]
Merge pull request #1593 from dachary/wip-vstart-erasure-code-default
vstart: set a sensible default for ruleset-failure-domain
Reviewed-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Sage Weil [Thu, 3 Apr 2014 16:57:04 +0000 (09:57 -0700)]
Merge pull request #1596 from ceph/wip-vstop-unmap
Unmap rbd images when stopping the whole cluster
Reviewed-by: Sage Weil <sage@inktank.com>
Ilya Dryomov [Tue, 1 Apr 2014 06:18:23 +0000 (10:18 +0400)]
stop.sh: unmap rbd images when stopping the whole cluster
Unmap rbd images when stopping the whole cluster. Not doing so results
in images that cannot be unmapped until the same cluster is brought
back up. Issue a warning if we failed to unmap all images.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Ilya Dryomov [Thu, 3 Apr 2014 14:11:49 +0000 (18:11 +0400)]
stop.sh: do not trace commands
Command tracing here doesn't bring any value and simply pollutes the
terminal, as the script always runs to completion.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Ilya Dryomov [Tue, 1 Apr 2014 06:18:23 +0000 (10:18 +0400)]
stop.sh: indent 4 spaces universally
Currently there is a mix between tabs and 4 spaces indent. Switch to
4 spaces indent.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Loic Dachary [Thu, 3 Apr 2014 12:07:19 +0000 (14:07 +0200)]
vstart: set a sensible default for ruleset-failure-domain
Set ruleset-failure-domain=osd so that
./ceph osd pool create ecpool 12 12 erasure
./rados --pool ecpool put SOMETHING /etc/group
works by default. When using a vstart cluster the default failure
domain (host) won't work because all OSDs are in "localhost".
Signed-off-by: Loic Dachary <loic@dachary.org>
Josh Durgin [Thu, 3 Apr 2014 00:03:09 +0000 (17:03 -0700)]
Merge pull request #1592 from ceph/wip-7965
lockdep: fix when instantiated multiple times (bug 7965)
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Sage Weil [Wed, 2 Apr 2014 23:43:10 +0000 (16:43 -0700)]
lockdep: reset state on shutdown
If we shut down, clear out all of the lockdep state. This ensures that if
we start up again on another cct, we will not be confused by old type ids
and dependency state.
Possibly contributed to #7965.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Wed, 2 Apr 2014 23:46:30 +0000 (16:46 -0700)]
lockdep: do not initialize if already started
If we have already registered a cct for lockdep, do not accept another one.
We already check that the cct matches when we shut down. This we will run
for the life span of a single cct and no longer.
Fixes: #7965
Signed-off-by: Sage Weil <sage@inktank.com>
Samuel Just [Wed, 2 Apr 2014 23:13:59 +0000 (16:13 -0700)]
Merge pull request #1591 from ceph/wip-7915
mon: bump snap_epoch when adding a tier (fixes 7915)
Reviewed-by: Samuel Just <sam.just@inktank.com>
Sage Weil [Wed, 2 Apr 2014 23:03:37 +0000 (16:03 -0700)]
OSDMap: bump snap_epoch when adding a tier
When we make an existing pool a tier, we start copying the snap metadata
from the base tier. That includes removed_snaps. In order for the OSD
to recognize that this value is changing for the first time, we need to
set snap_epoch, or else the OSD doesn't update it's in-memory PGPool
with removed snaps and we eventually hit an assertion failure because
PGPool::cached_remove_snaps is incorrect (e.g., empty).
Fix this by bumping snap_epoch when we add the new tier.
Fixes: #7915
Signed-off-by: Sage Weil <sage@inktank.com>
Samuel Just [Wed, 2 Apr 2014 22:15:56 +0000 (15:15 -0700)]
Merge pull request #1580 from ceph/wip-7937
osd: fix scrub logic for snapdir object
Reviewed-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Samuel Just [Wed, 2 Apr 2014 17:11:02 +0000 (10:11 -0700)]
ReplicatedPG::_scrub: don't bail early for snapdir
Fixes: #7937
Signed-off-by: Samuel Just <sam.just@inktank.com>
Mohammad Salehe [Wed, 2 Apr 2014 15:25:37 +0000 (19:55 +0430)]
debian: fix control to allow upgrades
Signed-off-by: Mohammad Salehe <salehe+dev@gmail.com>
Sage Weil [Wed, 2 Apr 2014 17:52:11 +0000 (10:52 -0700)]
Merge pull request #1590 from ceph/wip-7939
PG: set role for replicated even if role != shard
Reviewed-by: Sage Weil <sage@inktank.com>
Samuel Just [Wed, 2 Apr 2014 17:35:38 +0000 (10:35 -0700)]
Merge pull request #1579 from ceph/wip-7907
osd/ReplicatedPG: mark_unrollbackable when _rollback_to head
Reviewed-by: Samuel Just <sam.just@inktank.com>
Sage Weil [Tue, 1 Apr 2014 21:39:02 +0000 (14:39 -0700)]
debian: move rbdmap config and sysvinit/upstart scripts into ceph-common
Fixes: #7171
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Wed, 2 Apr 2014 15:44:33 +0000 (08:44 -0700)]
Merge pull request #1586 from ceph/wip-dirfrag
mds: fix check for merging/spliting dirfrag
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Wed, 2 Apr 2014 15:43:02 +0000 (08:43 -0700)]
Merge pull request #1587 from onlyjob/debian
init.d: correcting rbdmap LSB header / init order:
Reviewed-by: Sage Weil <sage@inktank.com>
Dmitry Smirnov [Wed, 2 Apr 2014 14:23:19 +0000 (01:23 +1100)]
init.d: correcting rbdmap init order:
* Require "$remote_fs" since it guarantees /usr availability
(rbd executable is in /usr/bin/rbd)
* Speed-up init.d rbd mapping on machines acting as MON/OSD
by starting rbdmap after /init.d/ceph (when possible) and
shutting down rbd before ceph.
* Map rbd devices before starting X (helpful when /home is mounted from rbd).
Yan, Zheng [Wed, 2 Apr 2014 05:24:37 +0000 (13:24 +0800)]
mds: fix check for merging/spliting dirfrag
check actual number of items instead of number of cached items
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Sage Weil [Wed, 2 Apr 2014 04:47:51 +0000 (21:47 -0700)]
Merge pull request #1583 from ceph/wip-largedir
Wip largedir
Reviewed-by: Sage Weil <sage@inktank.com>
Yan, Zheng [Tue, 1 Apr 2014 14:42:14 +0000 (22:42 +0800)]
mds: ignore CDir::check_rstats() when debug_scatterstat is off
It uses lots of CPU when dirfrag is large
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Yan, Zheng [Wed, 2 Apr 2014 01:33:37 +0000 (09:33 +0800)]
mds: initialize bloom filter according to dirfrag size
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Yan, Zheng [Tue, 1 Apr 2014 13:08:02 +0000 (21:08 +0800)]
mds: add dentries in dirfrag to LRU in reverse order
Files in a dirfrag are usually processed in the order of readdir
results. Files at the beginning of are more likely to be used in
the future than files at the last.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Sage Weil [Wed, 2 Apr 2014 04:01:07 +0000 (21:01 -0700)]
Merge pull request #1584 from ceph/wip-multimds
Wip multimds
Reviewed-by: Sage Weil <sage@inktank.com>
Yan, Zheng [Tue, 1 Apr 2014 03:19:10 +0000 (11:19 +0800)]
mds: handle freeze authpin race
For across authority rename, the MDS first freezes the source inode's
authpin. It happens while the source dentry isn't locked. So when the
inode's authpin become frozen, the source dentry may have changed and
be linked to a different inode.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Yan, Zheng [Tue, 1 Apr 2014 02:10:28 +0000 (10:10 +0800)]
mds: treat cluster as degraded when there is clientreplay MDS
This forbids exporting subtrees and fragmenting dirfrags when there
is MDS in clientreplay state. During replaying client requests, the
MDS may need to authpin some remote objects. Exporting subtrees and
fragmenting dirfrags slow down replaying client requests.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Yan, Zheng [Mon, 31 Mar 2014 01:46:58 +0000 (09:46 +0800)]
mds: don't start new segment while finishing disambiguate imports
This avoid inserting ESubtreeMap among EImportFinish events that
finish disambiguate imports. Because the ESubtreeMap reflects the
subtree state when all EImportFinish events are replayed.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Yan, Zheng [Mon, 31 Mar 2014 01:20:30 +0000 (09:20 +0800)]
mds: trim non-auth subtree more aggressively
When a non-auth dirfrag is pinned by uncommitted slave update,
there still can be non-auth child dirfrags that are trimmable.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Sage Weil [Tue, 1 Apr 2014 21:38:05 +0000 (14:38 -0700)]
debian: make ceph-common own etc/ceph, var/log/ceph
Clients can make use of these directories, and ceph-common is required by
ceph, so nothing should break here.
Change the purge postrm script to be for ceph-common (it does nothing else).
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Tue, 1 Apr 2014 21:27:31 +0000 (14:27 -0700)]
osd/ReplicatedPG: mark_unrollbackable when _rollback_to head
We fell into the case in _rollback_to where we just set ctx->modify = true
and don't explicitly mark the ctx and unrollbackable. Later, we screw up
in proc_replica_log as a result because we think we can rollback this
update to the head when in reality we cannot.
Fixes: #7907
Signed-off-by: Sage Weil <sage@inktank.com>
Samuel Just [Tue, 1 Apr 2014 20:58:10 +0000 (13:58 -0700)]
Merge pull request #1578 from ceph/wip-7437
wip-7437
Reviewed-by: Samuel Just <sam.just@inktank.com>
Sage Weil [Sat, 1 Mar 2014 20:07:34 +0000 (12:07 -0800)]
debian: move ceph-rest-api from ceph-common to ceph
ceph-rest-api is a 'server' thing that runs persistently; 'ceph' is a
better match.
Signed-off-by: Sage Weil <sage@inktank.com>
Josh Durgin [Tue, 1 Apr 2014 19:38:46 +0000 (12:38 -0700)]
Merge pull request #1577 from ceph/wip-7935
rgw: reverse logic to identify next part
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Samuel Just [Tue, 1 Apr 2014 19:19:52 +0000 (12:19 -0700)]
PG: set role for replicated even if role != shard
Fixes: #7939
Signed-off-by: Samuel Just <sam.just@inktank.com>
David Zafman [Mon, 31 Mar 2014 21:09:26 +0000 (14:09 -0700)]
test: Add EC testing to ceph_test_rados_api_snaphots
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman [Sat, 29 Mar 2014 00:48:15 +0000 (17:48 -0700)]
test: Add EC testing to ceph_test_rados_api_lock
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman [Fri, 28 Mar 2014 22:07:08 +0000 (15:07 -0700)]
test: Add EC testing to ceph_test_rados_api_tier
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman [Fri, 28 Mar 2014 21:15:43 +0000 (14:15 -0700)]
test: Add EC to ceph_test_rados_api_watch_notify
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman [Fri, 28 Mar 2014 21:12:15 +0000 (14:12 -0700)]
test: Add EC testing to ceph_test_rados_api_stat
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman [Fri, 28 Mar 2014 21:05:00 +0000 (14:05 -0700)]
test: Add EC testing to ceph_test_rados_api_list
Fixes: #7437
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman [Fri, 28 Mar 2014 22:35:29 +0000 (15:35 -0700)]
test: Add EC testing to ceph_test_rados_api_io
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman [Fri, 28 Mar 2014 22:35:46 +0000 (15:35 -0700)]
test: Add EC test classes RadosTestEC and RadosTestECPP
Fixes: #7437
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman [Tue, 1 Apr 2014 06:51:04 +0000 (23:51 -0700)]
test: Fix ceph_test_rados_list_parallel to clean up test pool
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman [Tue, 1 Apr 2014 06:03:53 +0000 (23:03 -0700)]
test: Fix ceph_test_rados_api_aio to clean up all test pools
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman [Sat, 29 Mar 2014 00:22:47 +0000 (17:22 -0700)]
test: Cleanup pools when FlushWriteRaces is done
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman [Thu, 27 Mar 2014 23:35:41 +0000 (16:35 -0700)]
librados: Add missing C variants for alignment
Add rados_ioctx_pool_requires_alignment()
Add rados_ioctx_pool_required_alignment()
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman [Thu, 27 Mar 2014 22:10:38 +0000 (15:10 -0700)]
librados: Add missing rados_wait_for_latest_osdmap()
Signed-off-by: David Zafman <david.zafman@inktank.com>
Sage Weil [Tue, 1 Apr 2014 18:02:42 +0000 (11:02 -0700)]
osd/ReplicatedPG: continue scrub logic when snapset.head_exists doesn't match
The 'continue' will cause more damange/noise than continuing because the
next_clone value won't be updated properly.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Tue, 1 Apr 2014 18:04:47 +0000 (11:04 -0700)]
osd/ReplicatedPG: handle snapdir properly during scrub
Handle snapdir similarly to how head is treated when updating the
next_clone info. Also, add a warning when we have a snapdir object and
head_exists == true (the converse of the existing check).
Fixes: #7937
Signed-off-by: Sage Weil <sage@inktank.com>
Yehuda Sadeh [Tue, 1 Apr 2014 17:51:20 +0000 (10:51 -0700)]
rgw: reverse logic to identify next part
Fixes: #7935
The check that identifies whether we need to move to the next part when
iterating over an object was reversed.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Sage Weil [Tue, 1 Apr 2014 01:15:12 +0000 (18:15 -0700)]
Merge pull request #1575 from jdurgin/wip-librbd-cp-sparse
librbd: skip zeroes when copying an image
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Mon, 31 Mar 2014 22:29:00 +0000 (15:29 -0700)]
osd/PG: debug cached_removed_snaps changes
See #7915.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Mon, 31 Mar 2014 22:19:57 +0000 (15:19 -0700)]
Merge pull request #1574 from mikenel/master
Add ceph-client-debug and jerasure shared objects to RPM spec file.
Reviewed-by: Sage Weil <sage@inktank.com>
Josh Durgin [Mon, 31 Mar 2014 21:53:31 +0000 (14:53 -0700)]
librbd: skip zeroes when copying an image
This is the simple coarse-grained solution, but it works well in
common cases like a small base image resized with a bunch of empty
space at the end. Finer-grained sparseness can be copied by using rbd
{export,import}-diff.
Fixes: #6257
Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
Sage Weil [Mon, 31 Mar 2014 21:59:38 +0000 (14:59 -0700)]
Merge pull request #1573 from ceph/wip-7912
mon/PGMap: clear pool sum when last pg is deleted
Reviewed-by: Greg Farnum <greg@inktank.com>
Sage Weil [Mon, 31 Mar 2014 21:57:57 +0000 (14:57 -0700)]
Merge pull request #1556 from ceph/wip-7888
msgr: add new ping/ping reply to use in place of keepalive
Reviewed-by: Greg Farnum <greg@inktank.com>
Sage Weil [Mon, 31 Mar 2014 21:56:13 +0000 (14:56 -0700)]
Merge pull request #1566 from ceph/wip-fuse-access
fuse: implement 'access' low level function
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 29 Mar 2014 17:46:16 +0000 (10:46 -0700)]
qa/workunits/cephtool/test.sh: test 'osd pg-temp ...'
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 29 Mar 2014 00:54:17 +0000 (17:54 -0700)]
mon/OSDMonitor: clear primary_temp on osd pg_temp updates
Until the OSD and the MOSDPGTemp messages encode primary_temp updates,
assume that any pg_temp update will clear primary_temp.
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Ilya Dryomov [Fri, 28 Mar 2014 16:28:44 +0000 (18:28 +0200)]
OSDMonitor: add 'mon osd allow primary temp' bool option
By default, we don't send out maps with primary_temp mappings because
there is no infrastructure in place that would make sure that the
entire cluster knows about primary_temp. Add an option to allow
primary_temp mappings, for development purposes.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Ilya Dryomov [Fri, 28 Mar 2014 16:28:44 +0000 (18:28 +0200)]
OSDMonitor: add 'osd primary-temp ...' command
ceph osd primary-temp <pgid> [<osd>]
Examples:
ceph osd primary-temp 0.2 4 # set primary_temp mapping for 0.2 to osd4
ceph osd primary-temp 0.2 # remove primary_temp mapping for 0.2
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Ilya Dryomov [Fri, 28 Mar 2014 16:28:44 +0000 (18:28 +0200)]
OSDMonitor: add 'osd pg-temp ...' command
ceph osd pg-temp <pgid> [<osd1> [<osd2> ...]]
Examples:
ceph osd pg-temp 0.2 0 1 2 # set pg_temp mapping for 0.2 to osds [0,1,2]
ceph osd pg-temp 0.2 3 # set pg_temp mapping for 0.2 to osds [3]
ceph osd pg-temp 0.2 # remove pg_temp mapping for 0.2
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Greg Farnum [Mon, 31 Mar 2014 20:17:22 +0000 (13:17 -0700)]
java/test: ceph.file.layout xattr is still not there now
b8ea65694faf59f12f285a65dc21753dab20ba11 tried to fix this, but
missed a spot.
Signed-off-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Josh Durgin [Mon, 31 Mar 2014 19:40:35 +0000 (12:40 -0700)]
Merge pull request #1567 from ceph/wip-7849
ceph-conf: don't create log files
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin [Mon, 31 Mar 2014 19:36:07 +0000 (12:36 -0700)]
Merge pull request #1570 from dachary/wip-gitignore
.gitignore: add examples/librados files
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Michael Nelson [Mon, 31 Mar 2014 19:35:56 +0000 (12:35 -0700)]
Add ceph-client-debug and jerasure shared objects to RPM spec file.
Sage Weil [Mon, 31 Mar 2014 17:42:23 +0000 (10:42 -0700)]
mon/PGMap: clear pool sum when last pg is deleted
Use the x.0 pg as a sentinel for the existence of the pool. Note that we
have to clean in up two paths: apply_incrmenetal (which is actually
deprecated) and the normal PGMonitor refresh.
Fixes: #7912
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Mon, 31 Mar 2014 17:18:58 +0000 (10:18 -0700)]
Merge pull request #1572 from ceph/wip-ec-profile-idempotent
mon: make 'ceph osd erasure-code-profile set ...' idempotent
Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com>
Sage Weil [Mon, 31 Mar 2014 17:01:43 +0000 (10:01 -0700)]
mon: make 'ceph osd erasure-code-profile set ...' idempotent
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Mon, 31 Mar 2014 16:14:36 +0000 (09:14 -0700)]
qa/workunits/rados/test_alloc_hint: fix erasure syntax
This changed recently.
Signed-off-by: Sage Weil <sage@inktank.com>
Loic Dachary [Mon, 31 Mar 2014 14:27:43 +0000 (16:27 +0200)]
Merge pull request #1571 from kotnik/docufix
Small glossary typo fix
Reviewed-by: Loic Dachary <loic@dachary.org>
Nikola Kotur [Mon, 31 Mar 2014 14:24:17 +0000 (16:24 +0200)]
doc: fix typos in glossary
Loic Dachary [Mon, 31 Mar 2014 09:30:10 +0000 (11:30 +0200)]
.gitignore: add examples/librados files
Signed-off-by: Loic Dachary <loic@dachary.org>
Sage Weil [Sun, 30 Mar 2014 16:20:58 +0000 (09:20 -0700)]
Merge pull request #1564 from dachary/wip-erasure-code-doc
doc: updates to the erasure code development docs
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Sun, 30 Mar 2014 16:17:21 +0000 (09:17 -0700)]
Merge pull request #1568 from dachary/wip-jerasure-warnings
erasure-code: update jerasure / gf-complete submodules
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Sun, 30 Mar 2014 16:16:44 +0000 (09:16 -0700)]
Merge pull request #1569 from dachary/wip-ssse3
autotools: s/ssse3/sse3/ typo
Reviewed-by: Sage Weil <sage@inktank.com>
Loic Dachary [Sun, 30 Mar 2014 15:57:22 +0000 (17:57 +0200)]
autotools: s/ssse3/sse3/ typo
Reported-by: Justin Erenkrantz <justin@erenkrantz.com>
Signed-off-by: Loic Dachary <loic@dachary.org>
Loic Dachary [Sun, 30 Mar 2014 09:07:46 +0000 (11:07 +0200)]
erasure-code: update jerasure / gf-complete submodules
For compilation warning patches.
http://tracker.ceph.com/issues/7909 Fixes #7909
Signed-off-by: Loic Dachary <loic@dachary.org>
Loic Dachary [Sat, 29 Mar 2014 10:30:42 +0000 (11:30 +0100)]
doc: pgbackend dev doc outdated notice
* Warn the reader that the implementation is ahead and may differ
* Update the links to the Firefly branch
* Remove links to issues used during development to avoid confusion
Signed-off-by: Loic Dachary <loic@dachary.org>
Loic Dachary [Sat, 29 Mar 2014 10:29:22 +0000 (11:29 +0100)]
doc: update jerasure plugin
* The parameters come from the erasure code profile
* Add a note about the upstream submodules gf-complete / jerasure
Signed-off-by: Loic Dachary <loic@dachary.org>
Loic Dachary [Sat, 29 Mar 2014 10:27:00 +0000 (11:27 +0100)]
doc: erasure code developer notes updates
Update the introduction to explain erasure code profiles. Remove
obsolete explanations about partial writes etc. Remove links to tickets
used during development. Update permalinks to be closer to
Firefly (v0.78).
Signed-off-by: Loic Dachary <loic@dachary.org>
Yan, Zheng [Sun, 30 Mar 2014 01:21:57 +0000 (09:21 +0800)]
fuse: implement 'access' low level function
Add an empty 'access' function to fuse low level functions. This
allow us to use ceph-fuse with fuse_default_permissions = false.
'fuse_default_permissions = false' can significantly improve the
speed of create/removing large number of files.
When fuse_default_permissions is true, the fuse kernel module sends
a getattr request whenever the kernel needs to check a directory's
permission. getattr (STAT_CAP_INODE_ALL) can be very slow if the
directory was just modified.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Sage Weil [Sun, 30 Mar 2014 05:28:13 +0000 (22:28 -0700)]
osd/ReplicatedPG: fix cache tier scrub again
This condition was flipped from commit
eb71924ea27e78d97bd45674ef5e6a7f
and the test case in
c3292e48483d861148322590ea1f05afd28cc2d3 still didn't
catch it. (It does now.)
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sun, 30 Mar 2014 05:27:04 +0000 (22:27 -0700)]
ceph_test_rados_api_tier: improve promote+scrub test
We need to have multiple clones with some different patterns of
missing-ness.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sun, 30 Mar 2014 04:52:09 +0000 (21:52 -0700)]
ceph-conf: use global_pre_init to avoid starting logging
This avoids starting up logging, which is not appropriate when we are
examining the config state and not actually starting up the entity in
question.
Fixes: #7849
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sun, 30 Mar 2014 04:51:20 +0000 (21:51 -0700)]
global: separate first half of global_init into global_pre_init
The pre_init now captures enough to create the g_ceph_context and parse
and initialize the in-memory config. However, we don't
- fiddle with signal handlers
- init lockdep
- call config observers (which starts up logging)
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sun, 30 Mar 2014 04:27:08 +0000 (21:27 -0700)]
Merge pull request #1520 from ceph/wip-multimds
Wip multimds
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Sun, 30 Mar 2014 00:28:36 +0000 (17:28 -0700)]
Merge pull request #1563 from dachary/wip-sse-i386
erasure-code: do not attempt to compile SSE4 on i386
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 29 Mar 2014 23:58:38 +0000 (16:58 -0700)]
ceph_test_rados_api_tier: improve cache tier + scrub test
Create lots of objects and make *some* of them be missing clones but not
all.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 29 Mar 2014 23:57:48 +0000 (16:57 -0700)]
osd/ReplicatedPG: tolerate trailing missing clones on cache tiers
I missed this case in
eb71924ea27e78d97bd45674ef5e6a7fce30932f .
Signed-off-by: Sage Weil <sage@inktank.com>