Commit Graph

29360 Commits

Author SHA1 Message Date
Yan, Zheng
34d09417b7 client: fix invalid iterator dereference in Client::trim_caps()
trimming inode drops a reference to the inode's parent, it may cause
the inode's parent also be trimmed. If the cap iterator 'p' happens to
point to the inode's parent and the inode's parent is trimmed, the cap
iterator 'p' become invalid.

Fix the issue by delaying removing cap from the seesion cap list.
(similar to what the kclient does)

Fixes #6599
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
2013-10-20 21:34:02 +08:00
Sage Weil
4d3502ba9b Merge pull request #734 from kri5/master
ceph: Remove unavailable option with clang

Reviewed-by: Sage Weil <sage@inktank.com>
2013-10-18 18:41:13 -07:00
Loic Dachary
1203a9a6e9 Merge pull request #741 from hjwsm1989/master
ceph.spec.in:  Add erasure-code related files.
2013-10-18 13:44:56 -07:00
John Wilkins
f06647eb0f Merge pull request #742 from atwardowski/patch-1
Update ceph-authtool.rst - Fix typo
2013-10-18 11:21:57 -07:00
John Wilkins
45d184602b doc: Removed references to Chef.
fixes: #6464

Signed-off-by: John Wilkins <john.wilkins@inktank.com>
2013-10-18 11:12:43 -07:00
Sage Weil
89995efe54 doc/release-notes: missed mds snaps
Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-18 10:35:31 -07:00
Sage Weil
5905227495 doc/release-notes: v0.71
Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-18 10:33:23 -07:00
Adam Twardowski
5d2cf46c1d Update ceph-authtool.rst
Fix typo
2013-10-18 13:14:25 -04:00
huangjun
7bcfe0967f ceph.spec.in: Add erasure-code related files.
Signed-off-by: huangjun  <hjwsm1989@gmail.com>
2013-10-18 18:21:22 +08:00
Sage Weil
ac533bf670 Merge pull request #722 from ceph/wip-6475
radosgw: create /var/log/radosgw in package, not init script

Reviewed-and-tested-by: Gary Lowell <gary.lowell@inktank.com>
2013-10-17 21:11:57 -07:00
Gary Lowell
09bd6bbe3b Merge branch 'next' 2013-10-18 01:51:16 +00:00
Josh Durgin
9af9c50cf1 Merge pull request #737 from xarses/6127
Add Redhat init script option
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
2013-10-17 17:27:34 -07:00
Andrew Woodward
533626c7e0 Add Redhat init script option
Resolves: 6127
Signed-off-by: Andrew Woodward <awoodward@mirantis.com>
2013-10-17 17:23:04 -07:00
Sage Weil
93f1befdd7 Merge pull request #738 from ceph/wip-cache-crc
fix cached crc, bug #6583

Reviewed-by: Samuel Just <sam.just@inktank.com>
2013-10-17 16:52:29 -07:00
Sage Weil
3cfe9f602b common/buffer: invalidate crc on zero, copy_in
This does not capture users who

 - calc a crc
 - use c_str() to modify the buffer content
 - (re)calc a crc

Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-17 16:47:29 -07:00
Sage Weil
8ec3aed516 common/buffer: fix crc_map types
Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-17 16:46:10 -07:00
Sage Weil
394ec17540 common/buffer: drop unused fields
Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-17 16:36:20 -07:00
Gary Lowell
2edc04cde5 ceph.spec.in: Need to cread radosgw log directory.
Signed-off-by: Gary Lowell  <gary.lowell@inktank.com>
2013-10-17 15:40:12 -07:00
Sage Weil
1f291f5c89 qa/workunits/rest/test.py: fix mds {add,remove}_data_pool test
Arg name changed from poolid to pool in e2602c54.

Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-17 14:38:37 -07:00
Sage Weil
11fc80df13 doc/release-notes: link ot the changelog
Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-17 13:25:59 -07:00
Sage Weil
eb0a3b7e94 doc/release-notes: v0.61.9
Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-17 13:15:23 -07:00
Sage Weil
d3f0c0b2a6 Makefile: fix /sbin vs /usr/sbin behavior
Instead of telling configure to put things in /sbin, explicitly put the
two important items (mkcephfs and mount.fuse.ceph) in /sbin via an
automake rule.  This unbreaks FreeBSD 9.1 and probably others.

Based on patches originally from Alan Somers <asomers@gmail.com>, modified
for the current Makefile structure and applied to the specfile too.

Fixes: #6456
Signed-off-by: Sage Weil <sage@inktank.com>
Tested-by: Alan Somers <asomers@gmail.com>
2013-10-17 12:06:49 -07:00
Samuel Just
15ec5332ba OSD: check for splitting when processing recover/backfill reservations
Fixes: 6565
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
2013-10-17 09:58:42 -07:00
Christophe Courtaut
08177f24df ceph: Remove unavailable option with clang
Some options are unavailable in clang. (-Wstrict-null-sentinel,
-rdynamic)

They were reintroduced by this commit
6949d221ad

This patch fixes it.

Signed-off-by: Christophe Courtaut <christophe.courtaut@gmail.com>
2013-10-17 18:00:39 +02:00
Gregory Farnum
ff8e6a763a Merge pull request #691 from ceph/wip-dirfrag
Reviewed-by: Greg Farnum <greg@inktank.com>
Partly-Reviewed-by: Sage Weil <sage@inktank.com>
2013-10-17 08:35:38 -07:00
Gary Lowell
e509cb1e69 v0.71 2013-10-17 09:19:36 +00:00
Sage Weil
e1e6408403 Merge remote-tracking branch 'gh/next' 2013-10-16 21:04:01 -07:00
Sage Weil
10b466e945 radosgw: create /var/log/radosgw in package, not init script
Fixes: #6475
Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-16 16:16:16 -07:00
Sage Weil
5c280a242b .gitignore: ceph-kvstore-tool
Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-16 14:23:13 -07:00
Sage Weil
5c476277a6 Merge remote-tracking branch 'gh/next' 2013-10-16 14:13:03 -07:00
Sage Weil
14e91bf124 debian, specfile: fix ceph-kvstore-tool packaging
See da69fa09c8

Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-16 13:59:13 -07:00
Sage Weil
d1547d2223 Merge pull request #732 from ceph/wip-kvstore-tool
tools: ceph-kvstore-tool: Access & manipulate leveldb store using the KeyValueDB interface

Reviewed-by: Sage Weil <sage@inktank.com>
2013-10-16 13:35:07 -07:00
Joao Eduardo Luis
fd6e2b8618 ceph-kvstore-tool: copy one leveldb store to some other place
Iterates over the provided source store's keys and copies them to the
provided destination store.

Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
2013-10-16 21:25:06 +01:00
Joao Eduardo Luis
85914b27e6 ceph-kvstore-tool: calc store crc
Reuse 'list()' function to traverse the store and calc not only version's
crcs, but also calc the store's crc.

Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
2013-10-16 21:25:06 +01:00
Sage Weil
a6a3f50dce Merge pull request #655 from ceph/wip-cache-crc
buffer: cache crc values in buffer::raw

Reviewed-by: Yehuda Sadeh <yehuda@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
2013-10-16 12:54:41 -07:00
Joao Eduardo Luis
da69fa09c8 tools: move 'test_store_tool' to 'tools/ceph-kvstore-tool'
ceph-kvstore-tool allows for lower-level access to leveldb stores.

Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
2013-10-16 20:49:49 +01:00
Gregory Farnum
9fdf42a7fc Merge pull request #709 from ceph/wip-filerecover
This patch prevents us from inadvertently reducing sparse file sizes during recovery.
We also reduce some code duplication by using eval() directly in do_file_recover()
instead of reproducing the parts we care about.

Reviewed-by: Greg Farnum <greg@inktank.com>
2013-10-16 11:13:35 -07:00
Sage Weil
eafdc92fc9 common/buffer: behave when cached crc stats don't start at 0
In particular, this helps when we set CEPH_BUFFER_TRACK=true

Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-16 10:58:31 -07:00
Josh Durgin
81c8ce2cee Merge pull request #724 from ceph/wip-4047
cls_rbd: do not make noise in osd log on rbd removal

Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
2013-10-16 09:56:35 -07:00
Yehuda Sadeh
c5cdf4ec60 crc32c: expand unit test
Verify crc on a buffer range, and on a NULL buffer (with various
sizes).

Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
2013-10-16 09:29:39 -07:00
Yehuda Sadeh
0620eead3d sctp_crc32c: fix crc calculation in the NULL buffer case
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
2013-10-16 09:28:14 -07:00
Yehuda Sadeh
b96ee5ccf5 crc32c: add a special case crc32c NULL buffer handling
Adding a special case crc32c NULL buffer handling for the intel assembly
code.

Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
2013-10-16 09:28:14 -07:00
Sage Weil
d9e30775c0 common/buffer: pass NULL buffer instead of a zero-filled one
This is about 2x faster on my box using the sctp code.

Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-16 09:28:14 -07:00
Sage Weil
9626aaf43b common/crc32c: document interface
Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-16 09:28:14 -07:00
Sage Weil
36e18ae997 common/crc32c_intel_baseline: accept null data buffer
Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-16 09:28:14 -07:00
Sage Weil
c545fc2f4c common/sctp_crc32: accept NULL buffer pointer
If we get NULL we behave as if it were a zero-filled buffer.

Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-16 09:28:14 -07:00
Sage Weil
a3e9344103 common/buffer: explicitly init zbuf to zeros
This might not be necessary?

Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-16 09:28:14 -07:00
Sage Weil
88ae8f6805 common/buffer: add crc caching performance test
On my old box:
 - matching cached values is a big win (free), obviously
 - the adjustment is the same speed as redoing the calculation.  this
   is probably because the data is already in L1/L2 cache; we still
   save memory bandwidth.

Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-16 09:28:14 -07:00
Sage Weil
6464516d9c common/buffer: instrument utilization of cached crcs
This is similar to the alloc tracking, but I've added a method to let
you enable it easily (in this case, from the unit test).

Signed-off-by: Sage Weil <sage@inktank.com>
2013-10-16 09:28:13 -07:00
Sage Weil
0c23a5624a common/buffer: cache crcs in buffer::raw
Signed-off-by: Samuel Just <sam.just@inktank.com>
2013-10-16 09:28:13 -07:00