Commit Graph

54773 Commits

Author SHA1 Message Date
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
John Spray
3c8b5f6308 Merge pull request #9588 from renhwztetecs/renhw-wip-mdsboot
mds: Ignore create_root_inode's access value

Reviewed-by: John Spray <john.spray@redhat.com>
2016-06-13 15:08:06 +01:00
Kefu Chai
b8ed93ac4e Merge pull request #9502 from tchaikov/wip-buffer-noexcept
common/buffer: add noexcept to ensure move ctor is used

Reviewed-by: Casey Bodley <cbodley@redhat.com>
2016-06-13 21:49:41 +08:00
Jason Dillaman
da11f6d1f7 Merge pull request #9658 from runsisi/ss-wip-fix-rbd-mirror-doc
doc: fix typo for rbd-mirroring

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
2016-06-13 09:29:12 -04:00
Mykola Golub
9f56d39cd2 Merge pull request #9657 from dillaman/wip-16235
librbd: avoid holding owner_lock waiting for exclusive lock

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
2016-06-13 13:42:52 +03:00
Mykola Golub
7748d2e9ca Merge pull request #9661 from dillaman/wip-16248
qa/workunits: ensure replay has started before checking position

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
2016-06-13 13:41:56 +03:00
Boris Ranto
2a6c738abd selinux: allow chown for self and setattr for /var/run/ceph
Fixes: http://tracker.ceph.com/issues/16126

Signed-off-by: Boris Ranto <branto@redhat.com>
2016-06-13 12:35:19 +02:00
xie xingguo
93d0fc9d93 mds/Mutation: remove redundant _dump method
The original passed in parameter "now" is useless, so
_dump method is identical to dump and shall be considered
as redundant.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
2016-06-13 17:00:36 +08:00
Kefu Chai
bfed5fd5ef rocksdb: remove rdb source files from dist tarball
rdb is based on NodeJS, and we don't build rdb to use rocksdb as
a keyvaluestore backend.

Fixes: http://tracker.ceph.com/issues/13554
Signed-off-by: Kefu Chai <kchai@redhat.com>
2016-06-13 16:41:52 +08:00
xie xingguo
b686785704 mds: remove useless code
As scrub_dentry_next() never returns a negative result code,
it always returns 0, EAGAIN, ENOENT.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
2016-06-13 16:38:07 +08:00
Jason Dillaman
1948047acb rbd-mirror: reset local image id when creating new image
Fixes: http://tracker.ceph.com/issues/16253
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2016-06-13 02:34:00 -04:00
Jason Dillaman
45dda67b3e rbd-mirror: handle an attempt to delete a non-mirrored image
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2016-06-13 02:33:59 -04:00
Jason Dillaman
7b2f99a97f cls_rbd: disallow setting an empty mirror image entry to disabling
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2016-06-13 02:03:29 -04:00
Haomai Wang
e7aff3da8b Merge pull request #9642 from yuyuyu101/wip-16237
msg/async/AsyncConnection: make verify auth callback without connection lock

Reviewed-by: Kefu Chai <kefu@redhat.com>
2016-06-13 13:35:59 +08:00
Kefu Chai
f7ddec3c58 Merge pull request #9213 from stiopaa1/replicatedpg_movetoforward
osd/ReplicatedPG.h: remove incorrect use of move

Reviewed-by: Adam Emerson <aemerson@redhat.com>
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
2016-06-13 13:09:21 +08:00
Kefu Chai
f9c48df52c Merge pull request #6967 from majianpeng/osd-full-skip-readop
osd: bypass readonly ops when osd full.

Reviewed-by: Samuel Just <sjust@redhat.com>
2016-06-13 13:05:49 +08:00
Kefu Chai
3ab134363e Merge pull request #8913 from vumrao/wip-vumrao-15704
rados: Add cleanup message with time to rados bench output

Reviewed-by: David Zafman <dzafman@redhat.com>
2016-06-13 13:01:44 +08:00
Jason Dillaman
6a68b68e0d qa/workunits: ensure replay has started before checking position
Fixes: http://tracker.ceph.com/issues/16248
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2016-06-13 00:19:37 -04:00
runsisi
9ec13bb344 doc: fix typo for rbd-mirroring
Signed-off-by: runsisi <runsisi@zte.com.cn>
2016-06-13 10:05:11 +08:00
Jason Dillaman
b96ab805ee librbd: avoid holding owner_lock waiting for exclusive lock
Fixes: http://tracker.ceph.com/issues/16235
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2016-06-12 21:42:59 -04:00
Gregory Farnum
e223d2eaff Merge pull request #9374 from batrick/i16024
client: use fh min readahead to enable readahead

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
2016-06-12 14:36:51 -07:00
Gregory Farnum
b3364be590 Merge pull request #8739 from ukernel/wip-14271
mds,client: sort dentries in hash order

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
2016-06-12 14:21:39 -07:00
Mykola Golub
9dd0487af9 Merge pull request #9640 from dillaman/wip-16235
librbd: recursive lock possible when disabling journaling

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
2016-06-12 20:39:25 +03:00
Mykola Golub
b8bfc0f2ef rbd-mirror: FAILED assert(!m_status_watcher)
Fixes: http://tracker.ceph.com/issues/16245
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
2016-06-12 20:36:09 +03:00
Jason Dillaman
67f1370b84 Merge pull request #9645 from trociny/rbd-mirror-teuthology-fix
qa/workunits/rbd: specify source path for rbd-mirror helper functions

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
2016-06-12 08:06:19 -04:00
huanwen ren
8a13c2ba6e mds/mdsrank: fix MDSRank::get_metadata_pool return value
Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn>
2016-06-12 18:00:58 +08:00
huanwen ren
6fb499f155 mds/tools: fix mds->mdsmap->get_metadata_pool() return int->int64_t
mds->mdsmap->get_metadata_pool() return should be int64_t not int,
fix up

Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn>
2016-06-12 17:56:14 +08:00
huanwen ren
17f24fb84d mds/mdcache: fix get poolid in open_remote_dentry
if dnl->get_remote_d_type() == DT_DIR not establish, then pool = -1
so the type of pool should be int64_t not uint64_t

Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn>
2016-06-12 16:54:48 +08:00
huanwen ren
8dc02f9957 mds/cdir: fix the type of dentry in the judgment
Perfect is_primary/is_remote/is_null type of dentry in the judgment

Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn>
2016-06-12 15:13:52 +08:00
huanwen ren
30bda3ddee mds/MDSRank: Ignore create_root_inode's access value
add (void), Ignore create_root_inode's access value

Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn>
2016-06-12 14:58:29 +08:00
xie xingguo
d5b4c4da16 server: don't get mdr hang for unknown lock type
It's better to respond and do the cleanup jobs anyway.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
2016-06-12 14:42:51 +08:00
xie xingguo
463feb272c server: return after respond_to_request() for EROFS
Respond_to_request will do cleanup jobs, thus we shall
stall and exit under this case.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
2016-06-12 14:21:25 +08:00
Mykola Golub
698242e743 qa/workunits/rbd: specify source path
otherwise it looks in $PATH for the script

Signed-off-by: Mykola Golub <mgolub@mirantis.com>
2016-06-11 20:30:32 +03:00
Kefu Chai
fb2667f3e1 Merge pull request #8436 from dreamhost/wip-python-prefix
python: Pass prefix/sbindir from autoconf to distutils.

Reviewed-by: Kefu Chai <kchai@redhat.com>
2016-06-11 18:07:54 +08:00
Kefu Chai
9db46b0cb5 Merge pull request #9364 from H3C/H3C
osd: sparse_read offset may not be zero for ecpool

Reviewed-by: Kefu Chai <kchai@redhat.com>
2016-06-11 18:06:24 +08:00
Kefu Chai
73a90f02f8 Merge pull request #9387 from stiopaa1/osd_osd_constiter
osd/OSD.h: change iterator to const_iterator

Reviewed-by: Kefu Chai <kchai@redhat.com>
2016-06-11 18:05:08 +08:00
Kefu Chai
bbf72c254e Merge pull request #9508 from x11562/auth_rm
mon: fix typo of 'ceph auth rm'

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
2016-06-11 18:04:04 +08:00
Kefu Chai
8d95f364af Merge pull request #9395 from tchaikov/wip-16010
buffer: handle integer underflow in iterator::copy(large_int, dest)

Reviewed-by: Sage Weil <sage@redhat.com>
2016-06-11 17:59:23 +08:00
Haomai Wang
2b7776545c msg/async/AsyncConnection: make verify auth callback without connection lock
Fixes: http://tracker.ceph.com/issues/16237
Signed-off-by: Haomai Wang <haomai@xsky.com>
2016-06-11 13:39:23 +08:00
Patrick Donnelly
b5a0773259
client: move readahead ref get/put to obj constr
The readahead context was incrementing the file handle reference count but only
decreasing in finish which is not called if readahead is a no-op.

This fixes a bug caught in testing where readahead was causing inode to become
disconnected:

    2016-06-10 19:46:48.953018 7f2a4351be80  1 client.4125 dump_inode: DISCONNECTED inode 10000000502 #10000000502 ref 110000000502.head(faked_ino=2307 ref=1 ll_ref=0 cap_refs={1024=0,2048=0,4096=0,8192=0} open={1=0,2=0} mode=100666 size=4194304/0 mtime=2016-06-10 19:29:45.107417 caps=-(0=pAsLsXsFscr) objectset[10000000502 ts 2/4012653 objects 0 dirty_or_tx 0] 0x7f2a24300d00)
    2016-06-10 19:46:48.953032 7f2a4351be80  2 client.4125 cache still has 0+155 items, waiting (for caps to release?)

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
2016-06-10 18:02:31 -04:00
Sage Weil
efc9bf1dc0 Merge pull request #9637 from stiopaa1/osdc_ObjectCacher_deleteCopyAssign
osdc/ObjectCacher.h: delete copy constr and assign

Reviewed-by: Sage Weil <sage@redhat.com>
2016-06-10 17:36:11 -04:00
Patrick Donnelly
01c179bcce
client: use fh min readahead to enable readahead
The client was using the configured maximum readahead of 0 which was changed in
95ee699891. This prevented readahead from ever
running (without setting a different default from 0).

Fixes: http://tracker.ceph.com/issues/16024

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
2016-06-10 17:33:13 -04:00
Patrick Donnelly
bee02a073f
client: update comment
Default was changed to unlimited in 11e4680e73.

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
2016-06-10 17:33:13 -04:00
Patrick Donnelly
b331d12a43
client: use layout.get_period() helper
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
2016-06-10 17:33:13 -04:00
Jianjian Huo
91db68c178 os/bluestore: Add an option to choose bluefs direct or buffered io
For now, set default to buffered io; after rocksdb block cache
has ability to cache writes, we will set default to direct io.

Signed-off-by: Jianjian Huo <jianjian.huo@ssi.samsung.com>
2016-06-10 14:33:01 -07:00
Matt Benjamin
8b5c1234c4 Merge pull request #9593 from Yan-waller/yj-hwip-rgwbasictype
rgw/basic_type: use the standard usage of string::npos

verified
2016-06-10 16:59:06 -04:00