RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2025-01-01 08:32:24 +00:00
Code Issues Packages Projects Releases Wiki Activity
54,823 Commits 212 Branches 394 Tags 1.5 GiB
e5cbb450b6
Commit Graph

54823 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Zhi Zhang
92154c80a4 Ceph status outputs mds slow request for better monitoring 
Signed-off-by: Zhi Zhang <zhangz.david@outlook.com>
 2016-06-15 11:28:34 +08:00
Kefu Chai
cbc3089aed script/make_rocksdb_makefile.sh: do not include rocksdb/tools/rdb/* 
Fixes: http://tracker.ceph.com/issues/13554
Signed-off-by: Kefu Chai <kchai@redhat.com>
 2016-06-15 11:26:41 +08:00
xie xingguo
fc6d0656de server: avoid side-effect of assert 
The more() method will implicitly allocate a new internal _more memeber
if it does not have and we shall avoid this.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-15 10:16:13 +08:00
Yan, Zheng
339176e40e Merge pull request #9697 from jcsp/wip-16298 
mds: fix MDLog recovery rare case
 2016-06-14 21:02:54 -05:00
Josh Durgin
b9fd29f267 Merge pull request #9705 from dillaman/wip-librados-compat 
librados: restore ability to compile against librados

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 2016-06-14 18:58:26 -07:00
Jason Dillaman
edbc47222d librados: restore ability to compile against librados 
Partial revert of 62be9268de5e9c9a08bdb977a7dab1ab9c55b2be which
added a new header dependency which was is not part of the librados
API.

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2016-06-14 20:56:39 -04:00
Jason Dillaman
9988ad7018 Merge pull request #9093 from zhouyuan/journaling_flush_skip_cache 
librbd: do not flush rbd cache if journaling is enabled

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2016-06-14 20:30:43 -04:00
Yuan Zhou
f2def83b7a rbd: Skip rbd cache flush if journaling is enabled under aio_flush 
With journaling rbd writes will be persisteted on rbd journal objects.
The journal will be replayed if crash happen. So it's not necessary to
flush rbd_cache in this case. This will improve the flush latency.

This patch adds checking on handling aio_flush: if journaling is
enabled, rbd cache flushing is skipped.
In a system flush(ImageCtx::flush) the cache is flushed even w/ journaling
where we truly do need to flush all IO out to disk.

Fixes: http://tracker.ceph.com/issues/15761

Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>
 2016-06-15 08:02:03 +08:00
Matt Benjamin
54b6c99f8f Merge pull request #7741 from rjfd/wip-14527 
Lookup monitors through DNS, feature #7741 

verified in teuthology (we had 8 failures that appear unrelated to this change)
 2016-06-14 15:38:03 -04:00
Kefu Chai
281e74daa0 Merge pull request #9602 from tchaikov/wip-ceph_test_cls_hello 
test: fix ceph_test_cls_hello test

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2016-06-15 01:15:47 +08:00
John Spray
40547be1ef Merge pull request #9572 from xiexingguo/xxg-wip-fix-ll_read_block 
client: fix wrong lock logic of ll_read_block() method

Reviewed-by: John Spray <john.spray@redhat.com>
 2016-06-14 17:44:23 +01:00
John Spray
acec1f1fd2 Merge pull request #9586 from xiexingguo/xxg-wip-fix_write 
client: add check for _lseek() during _write() process

Reviewed-by: John Spray <john.spray@redhat.com>
 2016-06-14 17:43:34 +01:00
John Spray
5f22e00050 Merge pull request #9583 from xiexingguo/xxg-wip-fix_preadv_pwritev 
client: fix potential access violation

Reviewed-by: John Spray <john.spray@redhat.com>
 2016-06-14 17:42:50 +01:00
John Spray
67dd8e7cc6 Merge pull request #5521 from ceph/wip-12653 
ceph-fuse: fuse_disable_pagecache

Reviewed-by: John Spray <john.spray@redhat.com>
 2016-06-14 17:29:49 +01:00
John Spray
ed2928dc08 Merge pull request #9590 from xiexingguo/xxg-wip-fix-make_request 
client: fix error process if we fail to choose a target mds

Reviewed-by: John Spray <john.spray@redhat.com>
 2016-06-14 17:26:44 +01:00
John Spray
c232840321 Merge pull request #9537 from xiexingguo/xxg-wip-fix-clientmemoryleak 
client: fix potential memory leak

Reviewed-by: Yan, Zheng <zyan@redhat.com>
 2016-06-14 17:26:01 +01:00
John Spray
7051321fc6 Merge pull request #9447 from ukernel/wip-mds-snap-fix 
mds: fix SnapRealm::have_past_parents_open()

Reviewed-by: John Spray <john.spray@redhat.com>
 2016-06-14 17:23:33 +01:00
John Spray
84aa28aaa0 mds: fix MDLog recovery rare case 
This handled the case where one daemon
saw a log that was partially rewritten
by another.  It was calling the context
incorrectly (without mds_lock held)

Fixes: http://tracker.ceph.com/issues/16298
Signed-off-by: John Spray <john.spray@redhat.com>
 2016-06-14 17:21:31 +01:00
Matt Benjamin
34734ced69 Merge pull request #9540 from Yan-waller/yj-wip-rgwsystemobj 
rgw/rados: remove meaningless assignment operation

since the removed variable (manifest_bl) is indeed unused, this change cannot have any side effect
 2016-06-14 10:28:47 -04:00
Matt Benjamin
4bb98f578f Merge pull request #9689 from jmunhoz/aws4-streaming-fix-bs 
rgw: aws4: fix buffer sharing issue with chunked uploads

verfified (by hand)
 2016-06-14 10:06:11 -04:00
xie xingguo
4c4da77275 mds/MDSMap: change test_flag() return type from int to bool 
which better matches its behaviour.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-14 21:10:24 +08:00
xie xingguo
89f74c98ea server: fix potential memory leak 
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-14 21:10:24 +08:00
xie xingguo
4828b2f205 mds: cancel context correctly if we already have required osdmap in hand 
By simply deleting the fin, the callbacks will be requeued into
the specified finisher and recalled later, which is not the expected
behaviour.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-14 20:40:14 +08:00
John Spray
6d6828fe52 Merge pull request #9669 from ceph/wip-selinux 
selinux: allow chown for self and setattr for /var/run/ceph

Reviewed-by: John Spray <john.spray@redhat.com>
 2016-06-14 13:34:56 +01:00
John Spray
7925d1d29d Merge pull request #9346 from jcsp/wip-enable-quota 
client: enable client_quota by default

Reviewed-by: John Spray <john.spray@redhat.com>
 2016-06-14 13:31:13 +01:00
John Spray
8ef90af3ba Merge pull request #9591 from ukernel/wip-16066 
client: kill QuotaTree

Reviewed-by: John Spray <john.spray@redhat.com>
 2016-06-14 13:29:52 +01:00
John Spray
b714f3004c Merge pull request #9651 from renhwztetecs/renhw-wip-mdcache 
mds: fix mdsmap->get_metadata_pool() return to int64_t

Reviewed-by: John Spray <john.spray@redhat.com>
 2016-06-14 13:26:43 +01:00
xie xingguo
510d0cc92f server: avoid duplicated call to utime_t 
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-14 19:59:41 +08:00
xie xingguo
26931f888c server: negative error code when responding to client 
As the comment suggests. Also a zero or positive return code
shall indicates a success, which does not match our intention here.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-14 19:32:01 +08:00
Mykola Golub
ba8ead5a4e Merge pull request #9672 from dillaman/wip-16260 
librbd: do not shut down exclusive lock while acquiring

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
 2016-06-14 14:21:54 +03:00
Jason Dillaman
c49f70a6d3 Merge pull request #9653 from trociny/wip-16245 
rbd-mirror: FAILED assert(!m_status_watcher)

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2016-06-14 06:55:10 -04:00
Mykola Golub
6ea199221c Merge pull request #9675 from dillaman/wip-16268 
librbd: remove should ignore mirror errors from older OSDs

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
 2016-06-14 13:27:16 +03:00
Javier M. Mellid
dd9f53115a rgw: aws4: fix buffer sharing issue with chunked uploads 
Signed-off-by: Javier M. Mellid <jmunhoz@igalia.com>
 2016-06-14 11:15:24 +02:00
Sage Weil
780925c4a2 Merge pull request #9684 from chhabaramesh/master 
os/bluestore: Fix for bit_alloc unit test case stack size and handle device that has less than block size at end.
 2016-06-14 05:08:15 -04:00
Yan, Zheng
161954bbfe client: kill QuotaTree 
Multiple clients can modify cephfs at the same time. It is
very tricky to keep QuotaTree consistant with the global FS
hiberarchy. This patch kills the quota tree.

After removing the quota tree, we traverse inode's path to
find quota root.

Fixes: http://tracker.ceph.com/issues/16066
Fixes: http://tracker.ceph.com/issues/16067
Signed-off-by: Yan, Zheng <zyan@redhat.com>
 2016-06-14 17:06:01 +08:00
Kefu Chai
9e28fe9953 Merge pull request #9668 from ceph/wip-rocksdb-wo-rdb 
rocksdb: remove rdb source files from dist tarball

Reviewed-by: Haomai Wang <haomai@xsky.com>
 2016-06-14 16:48:20 +08:00
Kefu Chai
64271f0345 test: fix ceph_test_cls_hello test 
cls_register_cxx_filter() is exercised by cls_hello, so add the missing
stub for cls_register_cxx_filter() call.

Signed-off-by: Kefu Chai <kchai@redhat.com>
 2016-06-14 11:51:16 +08:00
xie xingguo
4c52bf9ed4 mds/SessionMap: avoid potential memory leak 
This is safer and clean.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-14 11:34:58 +08:00
Ramesh Chander
32b3a8ac56 os/bluestore: Reduce stack size for bit alloc test case 
Signed-off-by: Ramesh Chander <Ramesh.Chander@sandisk.com>
 2016-06-13 20:33:09 -07:00
Ramesh Chander
6a2f04df5f os/bluestore: fix few calculations in bit alloc 
Signed-off-by: Ramesh Chander <Ramesh.Chander@sandisk.com>
 2016-06-13 20:33:09 -07:00
Kefu Chai
c01e3ebb6c Merge pull request #9678 from dzafman/wip-16273 
test: Explicitly use tempfile delete= feature instead of os.unlink()

Reviewed-by: Kefu Chai <kefu@redhat.com>
 2016-06-14 11:02:13 +08:00
xie xingguo
de8bc250b9 mds/SessionMap: explicitly require importing_count to be postive 
The importing_count is of type int, thus it can go underflow and
assert(importing_count) can still return false positive while it
should not supposed to do so.

Also this can make the assert itself more sensitive, e.g., can
catch underflow errors.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-14 10:31:20 +08:00
xie xingguo
eabd0b7b7b server: remove redundant semicolon 
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-14 09:43:58 +08:00
xie xingguo
bbbc326345 server: set terminating_sessions flag correctly 
So caller won't do this multiple times.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2016-06-14 09:11:15 +08:00
Jason Dillaman
a1ea1edce9 librbd: track in-flight flush events for journaling 
The tracking for flush events was accidentally dropped. This
can result in the journal being closed too early since it
might miss in-flight flush ops.

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2016-06-13 20:21:13 -04:00
David Zafman
431871eae8 test: Explicitly use tempfile delete= feature instead of os.unlink() 
The warning messages are caused by new delete= tempfile feature
trying to unlink a file after test has used os.unlink().

Signed-off-by: David Zafman <dzafman@redhat.com>
 2016-06-13 15:59:55 -07:00
Matt Benjamin
2378709134 Merge pull request #9484 from jmunhoz/aws4-streaming 
rgw: aws4: add STREAMING-AWS4-HMAC-SHA256-PAYLOAD support

verified (by hand)
 2016-06-13 16:47:27 -04:00
Jason Dillaman
5a97a9b7ef librbd: remove should ignore mirror errors from older OSDs 
Fixes: http://tracker.ceph.com/issues/16268
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2016-06-13 15:29:43 -04:00
Jason Dillaman
c5694fc676 librbd: do not shut down exclusive lock while acquiring 
Fixes: http://tracker.ceph.com/issues/16260
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2016-06-13 12:06:40 -04:00
Mykola Golub
082f918a99 Merge pull request #9664 from dillaman/wip-16253 
rbd-mirror: do not re-use image id from mirror directory if creating image

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
 2016-06-13 18:58:30 +03:00