]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Sage Weil [Wed, 24 Feb 2016 18:37:13 +0000 (10:37 -0800)]
Merge pull request #7681 from jjhuo/kstore-updates
osd: kstore: sync up kstore with recent bluestore updates
Reviewed-by: Sage Weil <sage@redhat.com>
Jason Dillaman [Wed, 24 Feb 2016 16:40:43 +0000 (11:40 -0500)]
Merge pull request #7759 from trociny/wip-rbd-mirror-image-replayer-improvements
rbd-mirror: ImageReplayer improvements
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Jason Dillaman [Wed, 24 Feb 2016 16:07:22 +0000 (11:07 -0500)]
Merge pull request #7736 from trociny/fix-librbd-asok-empty-name
librbd: retrieve image name when opening by id
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Jason Dillaman [Wed, 24 Feb 2016 15:35:28 +0000 (10:35 -0500)]
Merge pull request #7620 from jdurgin/wip-14419
cls_rbd: mirroring directory
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Josh Durgin [Fri, 12 Feb 2016 04:38:07 +0000 (20:38 -0800)]
cls_rbd: add methods for keeping track of mirrored images
These will track whether local images should be mirrored, and map them
to a unique global id. There's a state field for safely disabling
mirroring while operating on multiple objects.
Fixes: #14419
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
Josh Durgin [Thu, 4 Feb 2016 20:15:00 +0000 (12:15 -0800)]
librbd: rename rbd_pool_settings object to rbd_mirroring
We'll use this object only for mirroring-related purposes, not generic
settings on a pool.
Refs: #14419
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
Loic Dachary [Wed, 24 Feb 2016 14:42:02 +0000 (21:42 +0700)]
Merge pull request #7762 from ErwanAliasr1/evelu-check
Improving 'make check' for ceph-disk
Reviewed-by: Loic Dachary <ldachary@redhat.com>
Mykola Golub [Tue, 23 Feb 2016 20:20:51 +0000 (22:20 +0200)]
librbd: init asok_hook on open so name is always known
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Tue, 23 Feb 2016 09:28:00 +0000 (11:28 +0200)]
librbd: get image name on open if it is opened by id
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Kefu Chai [Wed, 24 Feb 2016 13:57:05 +0000 (21:57 +0800)]
Merge pull request #7743 from JiYou/open-test-for-bug-2339
test: enable test for bug #2339 which has been resolved.
Reviewed-by: Kefu Chai <kchai@redhat.com>
Haomai Wang [Wed, 24 Feb 2016 13:49:21 +0000 (21:49 +0800)]
Merge pull request #7773 from roidayan/xio_remove_redundant_magic_methods
xio: remove redundant magic methods
Reviewed-by: Haomai Wang<haomai@xsky.com>
Jason Dillaman [Wed, 24 Feb 2016 13:35:06 +0000 (08:35 -0500)]
Merge pull request #7771 from xiexingguo/xxg-wip-fixawatch
librados: do not clear handle for aio_watch()
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Roi Dayan [Mon, 15 Feb 2016 07:42:11 +0000 (09:42 +0200)]
xio: remove redundant magic methods
The same methods are inherited from Messenger.
Signed-off-by: Roi Dayan <roid@mellanox.com>
xie xingguo [Wed, 24 Feb 2016 08:59:33 +0000 (16:59 +0800)]
librados: remove unused local variables
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
xie xingguo [Wed, 24 Feb 2016 08:56:23 +0000 (16:56 +0800)]
librados: do not clear handle for aio_watch()
which is needed for aio_unwatch() to work.
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
Erwan Velu [Tue, 23 Feb 2016 16:54:21 +0000 (17:54 +0100)]
tests/ceph-disk: Let teardown clearing data
When running the ceph-disk check suite, 'test_mark_init' & 'test_activate_dir' had a
permission denied of the freshly created stuff like in :
tests/ceph-disk.sh:237: test_mark_init: /bin/rm -fr /ceph/src/ceph-disk/test-ceph-disk/dir
/bin/rm: cannot remove « /ceph/src/ceph-disk/test-ceph-disk/dir/snap_1 »: Permission denied
/bin/rm: cannot remove « /ceph/src/ceph-disk/test-ceph-disk/dir/current »: Permission denied
/bin/rm: cannot remove « /ceph/src/ceph-disk/test-ceph-disk/dir/snap_2 »: Permission denied
Those two tests are starting a mon with run_mon.
Trying to delete the files while the daemon are still alive is impossible.
This patch removes the explicit 'rm' and let the teardown doing the cleaning
stuff by stopping daemons & cleaning content.
By using this patch, make check is now sucessful on ceph-disk.
Signed-off-by: Erwan Velu <erwan@redhat.com>
Erwan Velu [Tue, 23 Feb 2016 15:14:48 +0000 (16:14 +0100)]
tests/ceph-disk: Using dummy device mappers
When doing a make test, if your local setup was running a "dm-0" or "dm-1",
the make check was failing.
The detection of the local "dm-x" were leading to a wrong comparaison with
the expected simulated values.
The fix is using a dummy name for that "dm" to prevent any collision with the local setup.
Prior that patch, a typical output of that error looked like :
> assert expect == main.list_devices()
E assert [{'partitions...: '/dev/Xda'}] == [{'partitions'...: '/dev/Xda'}]
E At index 0 diff: {'partitions': [{'dmcrypt': {'holders': ['dm-0'], 'type': 'plain'}, 'ptype': '
4fbd7e29 -9d25-41b8-afd0-
5ec00ceff05d ', 'path': '/dev/Xda1', 'is_partition': True, 'mount': None, 'uuid': '
56244cf5 -83ef-4984-888a-
2d8b8e0e04b2 ', 'type': 'data', 'state': 'unprepared', 'fs_type': None}], 'path': '/dev/Xda'} != {'partitions': [{'ptype': '
4fbd7e29 -9d25-41b8-afd0-
5ec00ceff05d ', 'whoami': None, 'path': '/dev/Xda1', 'is_partition': True, 'mount': '/var/cache/ccache', 'uuid': '
56244cf5 -83ef-4984-888a-
2d8b8e0e04b2 ', 'ceph_fsid': None, 'fs_type': 'btrfs', 'dmcrypt': {'holders': ['dm-0'], 'type': 'plain'}, 'type': 'data', 'state': 'active'}], 'path': '/dev/Xda'}
E Full diff:
E - [{'partitions': [{'dmcrypt': {'holders': ['dm-0'], 'type': 'plain'},
E + [{'partitions': [{'ceph_fsid': None,
E + 'dmcrypt': {'holders': ['dm-0'], 'type': 'plain'},
E - 'fs_type': None,
E ? ^^^^
E + 'fs_type': 'btrfs',
E ? ^^^^^^^
E 'is_partition': True,
E - 'mount': None,
E + 'mount': '/var/cache/ccache',
E 'path': '/dev/Xda1',
E 'ptype': '
4fbd7e29 -9d25-41b8-afd0-
5ec00ceff05d ',
E - 'state': 'unprepared',
E ? ^^^^ -----
E + 'state': 'active',
E ? ^^^^^
E 'type': 'data',
E - 'uuid': '
56244cf5 -83ef-4984-888a-
2d8b8e0e04b2 '}],
E ? --
E + 'uuid': '
56244cf5 -83ef-4984-888a-
2d8b8e0e04b2 ',
E + 'whoami': None}],
E 'path': '/dev/Xda'}]
tests/test_main.py:342: AssertionError
Signed-off-by: Erwan Velu <erwan@redhat.com>
Erwan Velu [Tue, 23 Feb 2016 10:36:51 +0000 (11:36 +0100)]
tests/ceph-disk: Creating missing working dir
When running run-tox.sh in a very simple env,
the test will fail if '/var/lib/ceph/tmp' doesn't exist.
This patch adds a check to create this directory if required as mkdtemp doesn't do it for you.
Prior this patch, the following behavior was seen :
tests/test_main.py:342:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
ceph_disk/main.py:3753: in list_devices
fstype=fs_type, options='')
ceph_disk/main.py:1217: in mount
dir=STATEDIR + '/tmp',
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
suffix = '', prefix = 'mnt.', dir = '/var/lib/ceph/tmp'
def mkdtemp(suffix="", prefix=template, dir=None):
"""User-callable function to create and return a unique temporary
directory. The return value is the pathname of the directory.
Arguments are as for mkstemp, except that the 'text' argument is
not accepted.
The directory is readable, writable, and searchable only by the
creating user.
Caller is responsible for deleting the directory when done with it.
"""
if dir is None:
dir = gettempdir()
names = _get_candidate_names()
for seq in xrange(TMP_MAX):
name = names.next()
file = _os.path.join(dir, prefix + name + suffix)
try:
> _os.mkdir(file, 0700)
E OSError: [Errno 2] No such file or directory: '/var/lib/ceph/tmp/mnt.KoAV85'
/usr/lib64/python2.7/tempfile.py:333: OSError
Signed-off-by: Erwan Velu <erwan@redhat.com>
Jason Dillaman [Wed, 24 Feb 2016 04:41:03 +0000 (23:41 -0500)]
Merge pull request #7649 from yuyuyu101/wip-async-watch
librados: implement async watch/unwatch
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
You Ji [Mon, 22 Feb 2016 15:52:38 +0000 (08:52 -0700)]
test: enable test for bug #2339 which has been resolved.
Signed-off-by: You Ji <jiyou09@gmail.com>
Josh Durgin [Wed, 24 Feb 2016 02:07:12 +0000 (18:07 -0800)]
Merge pull request #7761 from dillaman/wip-14847
librbd: fix state machine race conditions during shut down
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Haomai Wang [Tue, 23 Feb 2016 04:17:50 +0000 (12:17 +0800)]
RadosClient: call watch_flush before finisher stop
Signed-off-by: Haomai Wang <haomai@xsky.com>
Mykola Golub [Tue, 23 Feb 2016 09:26:37 +0000 (11:26 +0200)]
cls_rbd: async version of dir_get_name
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Yehuda Sadeh [Tue, 23 Feb 2016 19:44:50 +0000 (11:44 -0800)]
Merge pull request #7632 from zaitcev/cleanup-dead-code-3
rgw: Drop unused usage_exit from rgw_admin.cc
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
Yehuda Sadeh [Tue, 23 Feb 2016 19:25:38 +0000 (11:25 -0800)]
Merge pull request #7285 from magicrobotmonkey/patch-1
Increase verbosity level on RGWObjManifest line
Reviewed-by: Yehuda Saeh <yehuda@redhat.com>
Yehuda Sadeh [Tue, 23 Feb 2016 19:23:16 +0000 (11:23 -0800)]
Merge pull request #7618 from zaitcev/wip-5073-subuser-bis
rgw: Fix subuser harder with tenants
Reviewed-y: Yehuda Sadeh <yehuda@redhat.com>
Sage Weil [Tue, 23 Feb 2016 18:18:00 +0000 (13:18 -0500)]
Merge pull request #7763 from dachary/wip-releases
doc: fix 0.94.4 and 0.94.5 ordering
Jason Dillaman [Tue, 23 Feb 2016 18:17:52 +0000 (13:17 -0500)]
Merge pull request #7364 from javenwu/master
common: default cluster name to config file prefix
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Jason Dillaman [Tue, 23 Feb 2016 18:15:29 +0000 (13:15 -0500)]
Merge pull request #7548 from yuyuyu101/librbd-avoid-memcpy
librbd: reduce mem copies to user-buffer during read
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Jason Dillaman [Tue, 23 Feb 2016 18:14:23 +0000 (13:14 -0500)]
Merge pull request #7667 from jdurgin/wip-14780
Revert "librbd: use task finisher per CephContext"
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Haomai Wang [Tue, 23 Feb 2016 17:30:34 +0000 (01:30 +0800)]
Merge pull request #7625 from branch-predictor/bp-async-more
msg/async: smarter MSG_MORE
Reviewed-by: Haomai Wang <haomai@xsky.com>
Loic Dachary [Tue, 23 Feb 2016 17:13:53 +0000 (00:13 +0700)]
doc: fix 0.94.4 and 0.94.5 ordering
Signed-off-by: Loic Dachary <loic@dachary.org>
Casey Bodley [Tue, 23 Feb 2016 16:02:03 +0000 (11:02 -0500)]
Merge pull request #7760 from theanalyst/fix/rgw-fcgi-build
rgw: fcgi should include acconfig
Jason Dillaman [Tue, 23 Feb 2016 15:59:09 +0000 (10:59 -0500)]
librbd: deterministically unblock writes on snap set
Previously this was handled in the state machine's destructor,
but that can lead to nondeterministic behavior due to interactions
between threads.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
Jason Dillaman [Tue, 23 Feb 2016 15:49:24 +0000 (10:49 -0500)]
librbd: avoid close race-condition within ImageState
The lock was previously unlocked/locked to avoid lock cycles. This
creates an issue when closing the image because the another thread
might have already deleted the ImageState before the lock could be
re-locked.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
Abhishek Lekshmanan [Tue, 23 Feb 2016 15:39:06 +0000 (16:39 +0100)]
rgw: fcgi should include acconfig
As it references FASTCGI_INCLUDE_DIR which is referenced from it
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
Sage Weil [Tue, 23 Feb 2016 15:35:34 +0000 (10:35 -0500)]
doc/release-notes: v0.94.6 final notes
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 23 Feb 2016 15:29:53 +0000 (10:29 -0500)]
Merge pull request #7689 from dachary/wip-release-notes
release-notes: draft v0.94.6 release notes
Piotr Dałek [Thu, 11 Feb 2016 08:38:46 +0000 (09:38 +0100)]
msg/async: cut the middle-man
Get rid of complete_bl and let messenger write directly to outcoming_bl of
a connection. Also, if message bufferlist is small enough, append its
contents to outcoming_bl directly, so we'll use less iovecs and in best
case, pack entire message (together with header and footer added in
write_message()) in single bufferptr.
Signed-off-by: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
Haomai Wang [Tue, 23 Feb 2016 14:19:34 +0000 (22:19 +0800)]
Merge pull request #7725 from ceph/wip-fix-async-valgrind
AsyncMessenger: fix valgrind leak
Kefu Chai [Tue, 23 Feb 2016 13:11:18 +0000 (21:11 +0800)]
Merge pull request #7758 from dachary/wip-ceph-disk-prepare-help
ceph-disk: fix prepare --help
Reviewed-by: Kefu Chai <kchai@redhat.com>
Loic Dachary [Tue, 23 Feb 2016 09:39:48 +0000 (16:39 +0700)]
ceph-disk: fix prepare --help
Signed-off-by: Loic Dachary <loic@dachary.org>
Kefu Chai [Tue, 23 Feb 2016 11:12:08 +0000 (19:12 +0800)]
Merge pull request #7756 from dachary/wip-ceph-disk-args
ceph-disk: deactivate / destroy PATH arg are optional
Reviewed-by: Kefu Chai <kchai@redhat.com>
Loic Dachary [Tue, 23 Feb 2016 08:52:39 +0000 (15:52 +0700)]
ceph-disk: deactivate / destroy PATH arg are optional
And the mandatory positional arguments need not be == 1, it is enough to
leave the nargs parameter out.
Signed-off-by: Loic Dachary <loic@dachary.org>
Mykola Golub [Sun, 21 Feb 2016 17:42:54 +0000 (19:42 +0200)]
test: image_replay: use flush to speedup commit position update
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Sun, 21 Feb 2016 16:59:07 +0000 (18:59 +0200)]
rbd-mirror: add asok commands to get image replyer status and flush
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Sun, 21 Feb 2016 15:03:28 +0000 (17:03 +0200)]
rbd-mirror: store sync snapshot name in registered client
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Sun, 21 Feb 2016 15:02:02 +0000 (17:02 +0200)]
librbd: add field to MirrorPeerClientMeta to store snap name
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Sun, 21 Feb 2016 11:16:04 +0000 (13:16 +0200)]
rbd-mirror: ImageReplayer: pass registered client ID as external param
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Loic Dachary [Tue, 23 Feb 2016 05:08:46 +0000 (12:08 +0700)]
Merge pull request #7687 from yangdongsheng/ceph-disk_nargs
ceph-disk: make some arguments as required if necessary
Reviewed-by: Loic Dachary <ldachary@redhat.com>
Sage Weil [Tue, 23 Feb 2016 02:59:28 +0000 (21:59 -0500)]
Merge pull request #7640 from batrick/vstart-btrfs-missing
vstart.sh: silence a harmless msg where btrfs is not found
Reviewed-by: Kefu Chai <kchai@redhat.com>
Jason Dillaman [Mon, 22 Feb 2016 17:37:33 +0000 (12:37 -0500)]
Merge pull request #7737 from trociny/wip-librados_test_stub-debug
test: more debug logging for TestWatchNotify
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Sage Weil [Mon, 22 Feb 2016 17:36:42 +0000 (12:36 -0500)]
Merge pull request #7714 from SUSE/wip-qa-typo
doc: standardize @param (not @parma, @parmam, @params)
Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
Jason Dillaman [Mon, 22 Feb 2016 17:36:01 +0000 (12:36 -0500)]
Merge pull request #7738 from trociny/test-librbd-poolname
test: create pools for rbd tests with different prefix
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Sage Weil [Mon, 22 Feb 2016 17:29:47 +0000 (12:29 -0500)]
Merge pull request #7727 from jmunhoz/ceph-fix-14829
vstart.sh: avoid race condition starting rgw via vstart.sh
Reviewed-by: Sage Weil <sage@redhat.com>
Jason Dillaman [Mon, 22 Feb 2016 16:47:49 +0000 (11:47 -0500)]
Merge pull request #7614 from trociny/wip-rbd-mirroring-image-reply
rbd-mirror: implement ImageReplayer
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Loic Dachary [Mon, 22 Feb 2016 16:07:35 +0000 (23:07 +0700)]
Merge pull request #7744 from dachary/wip-14838-ceph-disk-dmcrypt
ceph-disk: s/dmcrpyt/dmcrypt/
Reviewed-by: Nathan Cutler <ncutler@suse.com>
Loic Dachary [Mon, 22 Feb 2016 16:02:46 +0000 (23:02 +0700)]
ceph-disk: s/dmcrpyt/dmcrypt/
http://tracker.ceph.com/issues/14838 Fixes: #14838
Signed-off-by: Frode Sandholtbraaten <fsa@basefarm.no>
Signed-off-by: Loic Dachary <loic@dachary.org>
Kefu Chai [Mon, 22 Feb 2016 14:42:41 +0000 (22:42 +0800)]
Merge pull request #7688 from tchaikov/wip-14788
buffer: hide iterator_impl symbols
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Mykola Golub [Mon, 22 Feb 2016 06:47:40 +0000 (08:47 +0200)]
test: create pools for rbd tests with different prefix
This makes easier to find out what test fails to cleanup.
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Loic Dachary [Mon, 22 Feb 2016 05:44:14 +0000 (12:44 +0700)]
Merge pull request #7723 from tchaikov/wip-rgw-test
test/radosgw-admin: update the expected usage outputs
Reviewed-by: Loic Dachary <ldachary@redhat.com>
Haomai Wang [Fri, 19 Feb 2016 04:13:58 +0000 (12:13 +0800)]
test_rados_watch_notify: add async_watch_flush tests
Signed-off-by: Haomai Wang <haomai@xsky.com>
Kefu Chai [Sun, 21 Feb 2016 11:30:25 +0000 (19:30 +0800)]
Merge pull request #7647 from yuyuyu101/wip-spdk
NVMEDevice: refactor probe/attach codes and support zero command
Reviewed-by: Kefu Chai <kchai@redhat.com>
Mykola Golub [Sat, 20 Feb 2016 20:48:48 +0000 (22:48 +0200)]
test: more debug for TestWatchNotify
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Haomai Wang [Sun, 21 Feb 2016 10:43:02 +0000 (18:43 +0800)]
NVMEDevice: accept spdk api name changes
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Sun, 21 Feb 2016 10:38:11 +0000 (18:38 +0800)]
spdk: update spdk submodule to accept new interface changes
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Sun, 21 Feb 2016 10:33:24 +0000 (18:33 +0800)]
NVMEDevice: add default constructor for Task and clean up
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Mon, 15 Feb 2016 05:54:03 +0000 (13:54 +0800)]
NVMEDevice: only activate zero command when backend supported
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Fri, 12 Feb 2016 17:19:58 +0000 (01:19 +0800)]
BlockDevice: detect symbol file basename
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Fri, 12 Feb 2016 15:34:14 +0000 (23:34 +0800)]
BlueStore: fix spdk link file create failed
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Fri, 12 Feb 2016 13:37:52 +0000 (21:37 +0800)]
NVMEDevice: fix atomic and lock changes
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Sun, 7 Feb 2016 10:51:14 +0000 (18:51 +0800)]
NVMEDevice: use nvme_probe to simply init
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Sun, 7 Feb 2016 07:06:39 +0000 (15:06 +0800)]
spdk: update spdk submodule
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Sun, 7 Feb 2016 06:57:13 +0000 (14:57 +0800)]
NVMEDevice: use nvme zero command instead of writing zero
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Sun, 7 Feb 2016 06:46:51 +0000 (14:46 +0800)]
NVMEDevice: cleanup task iterator process
Signed-off-by: Haomai Wang <haomai@xsky.com>
Javier M. Mellid [Sat, 20 Feb 2016 17:33:21 +0000 (17:33 +0000)]
vstart.sh: avoid race condition starting rgw via vstart.sh
In slow environments running vstart.sh hits a race condition setting up
the rgw server. Creating the S3 and Swift users before booting rgw
avoids the issue.
Fixes: #14829
Signed-off-by: Javier M. Mellid <jmunhoz@igalia.com>
Kefu Chai [Sat, 20 Feb 2016 04:10:53 +0000 (12:10 +0800)]
test/radosgw-admin: update the expected usage outputs
to match with the latest radosgw-admin changes.
Signed-off-by: Kefu Chai <kchai@redhat.com>
Mykola Golub [Sat, 13 Feb 2016 07:29:07 +0000 (09:29 +0200)]
test: rbd-mirror: add tool to test ImageReplayer
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Sat, 13 Feb 2016 07:26:06 +0000 (09:26 +0200)]
test: rbd-mirror: add ImageReplayer tests
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Sat, 13 Feb 2016 07:22:28 +0000 (09:22 +0200)]
rbd-mirror: implement ImageReplayer
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Haomai Wang [Fri, 19 Feb 2016 03:39:28 +0000 (11:39 +0800)]
librados: add async_watch_flush api
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Fri, 19 Feb 2016 03:26:45 +0000 (11:26 +0800)]
RadosClient: add async watch_flush version
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Fri, 19 Feb 2016 03:11:26 +0000 (11:11 +0800)]
Objecter: make linger_callback_flush accept context instead of blocking
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Tue, 16 Feb 2016 02:38:52 +0000 (10:38 +0800)]
librados: add C api impl and tracing codes
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Mon, 15 Feb 2016 10:11:37 +0000 (18:11 +0800)]
librados: remove unused variables
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Mon, 15 Feb 2016 09:33:56 +0000 (17:33 +0800)]
test: add aio_unwatch unittest
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Mon, 15 Feb 2016 09:31:12 +0000 (17:31 +0800)]
librados: support aio_unwatch api
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Mon, 15 Feb 2016 09:25:38 +0000 (17:25 +0800)]
librados: add unittest for rados aio watch
Signed-off-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Mon, 15 Feb 2016 09:16:59 +0000 (17:16 +0800)]
librados: add async watch api
Signed-off-by: Haomai Wang <haomai@xsky.com>
Mykola Golub [Sat, 13 Feb 2016 07:14:28 +0000 (09:14 +0200)]
journal: wrappers to get journal metadata
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Sat, 13 Feb 2016 07:10:46 +0000 (09:10 +0200)]
librbd: allow to use journal replay externally
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Mykola Golub [Sun, 7 Feb 2016 10:50:24 +0000 (12:50 +0200)]
cls::journal: don't print client data
It makes `rbd journal status` output look ugly.
It is a temporary solution and is going to be improved in
the future (it could be useful to see data but decoded).
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Jianpeng Ma [Fri, 7 Aug 2015 05:12:36 +0000 (13:12 +0800)]
test/bufferlist:Add test case for bufferlist::is_provided_buffer.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
Jianpeng Ma [Mon, 19 Jan 2015 03:51:40 +0000 (11:51 +0800)]
Striper: Add function 'assemble_result(CephContext *cct, char *buffer, size_t len)' in StripedReadResult.
For sparse read, it can avoid copying zero around.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
Jianpeng Ma [Mon, 19 Jan 2015 02:15:59 +0000 (10:15 +0800)]
Striper: Add total_intended_len in StripedReadResult
Using this field record the total len which we wanted.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
Jianpeng Ma [Fri, 7 Aug 2015 05:18:11 +0000 (13:18 +0800)]
osdc/objectCacher: Remove the unsed field.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
Jianpeng Ma [Tue, 20 Jan 2015 02:40:11 +0000 (10:40 +0800)]
bufferlist: Add new func is_provided_buffer(const char *dst)
Using this function to judge whether copy data from bufferlist to
dst. It mainly used librados C interface to avoid copy data from librados
to caller.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
Josh Durgin [Sat, 20 Feb 2016 00:51:07 +0000 (16:51 -0800)]
Merge pull request #7627 from dillaman/wip-librbd-read-replay
librbd: block read requests until journal replayed
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Josh Durgin [Sat, 20 Feb 2016 00:50:44 +0000 (16:50 -0800)]
Merge pull request #7718 from dillaman/wip-librbd-mirror-peer
librbd: update of mirror pool mode and mirror peer handling
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Josh Durgin [Sat, 20 Feb 2016 00:50:19 +0000 (16:50 -0800)]
Merge pull request #7698 from dillaman/wip-librbd-replay-flush
librbd: differentiate journal replay flush vs shut down
Reviewed-by: Josh Durgin <jdurgin@redhat.com>