]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Patrick Donnelly [Wed, 10 May 2017 20:56:06 +0000 (16:56 -0400)]
mds: delay export until dir is stable
This catches an assertion failure due to exports on unjournaled directories:
2017-05-10 12:45:18.193640
7eff65a04700 7 mds.0.migrator export_dir [dir
10000000003 /a/ [2,head] auth pv=3 v=1 cv=0/0 ap=1+2+2 state=
1073742082 |complete f()->f(v0 m2017-05-10 12:45:14.977151 1=0+1) n()->n(v0 rc2017-05-10 12:45:14.977151 1=0+1) hs=0+1,ss=0+0 | child=1 authpin=1 0x55ef655de000] to 1
2017-05-10 12:45:18.193654
7eff65a04700 15 mds.0.18 hit export target 10 @ 2017-05-10 12:45:18.193653
2017-05-10 12:45:18.193661
7eff65a04700 10 mds.0.cache.dir(
10000000003 ) auth_pin by 0x55ef650cc130 on [dir
10000000003 /a/ [2,head] auth pv=3 v=1 cv=0/0 ap=2+2+2 state=
1073742082 |complete f()->f(v0 m2017-05-10 12:45:14.977151 1=0+1) n()->n(v0 rc2017-05-10 12:45:14.977151 1=0+1) hs=0+1,ss=0+0 | child=1 authpin=1 0x55ef655de000] count now 2 + 2
2017-05-10 12:45:18.193701
7eff65a04700 7 mds.0.cache request_start_internal request(mds.0:4) op 5377
2017-05-10 12:45:18.193708
7eff65a04700 7 mds.0.migrator dispatch_export_dir request(mds.0:4)
2017-05-10 12:45:18.198281
7eff66205700 20 mgrc operator() 234 counters, of which 0 new
2017-05-10 12:45:18.198294
7eff66205700 20 mgrc send_report encoded 2006 bytes
2017-05-10 12:45:18.198298
7eff66205700 1 -- 127.0.0.1:6825/
693985647 --> 127.0.0.1:6826/1025 -- mgrreport(+0-0 packed 2006) v2 -- 0x55ef655abe00 con 0
2017-05-10 12:45:18.198364
7eff6a9ab700 10 _calc_signature seq 14 front_crc_ =
2388339344 middle_crc = 0 data_crc = 0 sig =
5026349537430007662
2017-05-10 12:45:18.198376
7eff6a9ab700 20 Putting signature in client message(seq # 14): sig =
5026349537430007662
2017-05-10 12:45:18.198411
7eff65a04700 -1 /home/pdonnell/ceph/src/mds/MDCache.cc: In function 'void MDCache::make_trace(std::vector<CDentry*>&, CInode*)' thread
7eff65a04700 time 2017-05-10 12:45:18.194136
/home/pdonnell/ceph/src/mds/MDCache.cc: 8225: FAILED assert(parent)
ceph version
12.0.1-2198-ge757c02 (
e757c025fa3270b12fb2fca17cf159fa1bd72747 )
1: (ceph::__ceph_assert_fail(char const*, char const*, int, char const*)+0x102) [0x55ef5b780f02]
2: (MDCache::make_trace(std::vector<CDentry*, std::allocator<CDentry*> >&, CInode*)+0x1cb) [0x55ef5b4fa12b]
3: (Migrator::get_export_lock_set(CDir*, std::set<SimpleLock*, std::less<SimpleLock*>, std::allocator<SimpleLock*> >&)+0x55) [0x55ef5b5e8215]
4: (Migrator::dispatch_export_dir(boost::intrusive_ptr<MDRequestImpl>&, int)+0xa74) [0x55ef5b5f72d4]
5: (Migrator::export_dir(CDir*, int)+0x9ca) [0x55ef5b5ea75a]
6: (MDBalancer::handle_export_pins()+0x7b4) [0x55ef5b61ab24]
7: (MDBalancer::tick()+0x1e8) [0x55ef5b61b748]
8: (MDSRankDispatcher::tick()+0x5f1) [0x55ef5b44bdb1]
9: (Context::complete(int)+0x9) [0x55ef5b43bcc9]
10: (SafeTimer::timer_thread()+0x452) [0x55ef5b77dd52]
11: (SafeTimerThread::entry()+0xd) [0x55ef5b77f15d]
12: (()+0x76ba) [0x7eff6d4bb6ba]
13: (clone()+0x6d) [0x7eff6c52782d]
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Yan, Zheng [Tue, 9 May 2017 08:28:13 +0000 (16:28 +0800)]
mds: fix dir auth calculation in CDir::merge
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
Yan, Zheng [Tue, 9 May 2017 08:26:48 +0000 (16:26 +0800)]
mds: properly cleanup export states
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
Patrick Donnelly [Tue, 9 May 2017 02:26:48 +0000 (22:26 -0400)]
mds: update export targets even when not active
Problem caught test_migration_on_shutdown. Migration could not proceed because
export_targets were not being updated.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Sat, 6 May 2017 18:06:06 +0000 (14:06 -0400)]
mds: check mdsdir against dest
Introduced by
aebc1caee86b70f51d7d2d16f25f0ee9e4119f43 . Found by Zheng.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Yan, Zheng [Fri, 5 May 2017 03:35:16 +0000 (11:35 +0800)]
mds: handle aux subtree when splitting/merging dirfrag
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
Patrick Donnelly [Fri, 5 May 2017 23:10:53 +0000 (19:10 -0400)]
mds: call maybe_export_pin on all fetched dirfrags
If the dirfrag is pinned to the current mds, it needs to be made into an aux
subtree.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Fri, 5 May 2017 23:10:27 +0000 (19:10 -0400)]
mds: handle export pin on unjournaled directory
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Fri, 5 May 2017 23:00:31 +0000 (19:00 -0400)]
qa: improve time handling for test_exports test
Also catches corner-case found by Zheng where an unjournaled directory will
cause export pinning to fail because it cannot be made a subtree until its
parent is stable.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Thu, 4 May 2017 20:59:26 +0000 (16:59 -0400)]
mds: do not try to export pin special directories
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Thu, 4 May 2017 02:42:57 +0000 (22:42 -0400)]
mds: remove unnecessary check for parent pins
With Zheng's help, now that the code has captured all the paths where an inode
should be checked for export pins, we don't need to look at parents anymore.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Wed, 3 May 2017 14:39:48 +0000 (10:39 -0400)]
mds: check export_pin on dirfrag load
Zheng's suggestion: "[maybe_export_pin is called by MDCache::add_inode() and
Migrator. For the first case, inode has no dirfrag (I think it's better to call
this function in CInode::get_or_open_dirfrag)"
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 2 May 2017 22:29:08 +0000 (18:29 -0400)]
qa: remove errant mount requirement
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 2 May 2017 20:28:17 +0000 (16:28 -0400)]
mds: cleanup aux subtree pin debug
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 2 May 2017 20:11:42 +0000 (16:11 -0400)]
mds: use clearer name for export pin cont
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 2 May 2017 04:30:49 +0000 (00:30 -0400)]
mds: handle imported directory that is pinned
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 2 May 2017 04:28:15 +0000 (00:28 -0400)]
mds: use aux subtrees for export pinned inodes
Idea here is that a pinned inode should not be exported when its parent is.
Setting the pinned inode's dirfrags to aux subtrees prevents them from being
merged with a parent subtree.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Fri, 28 Apr 2017 17:44:53 +0000 (13:44 -0400)]
Revert "mds: prevent export of pinned inodes"
This reverts commit
df340e820d0fe964a4e8a71a86aaa92cad25b91f .
Going to use Zheng's aux subtrees instead.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Yan, Zheng [Fri, 28 Apr 2017 02:53:41 +0000 (10:53 +0800)]
mds: simplify auxiliary subtrees when sending subtree resolve
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
Yan, Zheng [Thu, 27 Apr 2017 15:52:08 +0000 (23:52 +0800)]
mds: create auxiliary subtree inside auth subtree
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
Patrick Donnelly [Tue, 25 Apr 2017 15:13:44 +0000 (11:13 -0400)]
mds: break ancestor walk if node is export_pinned
i.e. don't keep walking the parents after adding the ancestor (or the node
itself) to the export_pin_queue.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Fri, 21 Apr 2017 14:31:10 +0000 (10:31 -0400)]
mds: use projected parent to avoid unsafe access
Add-on to:
26a08f31f7965afb062b027a4455620b52e7a67e
Credit to Zheng for noticing.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Wed, 19 Apr 2017 02:28:39 +0000 (22:28 -0400)]
mds: prevent export of pinned inodes
This adds a chain of linked lists to CInode which can be followed to CInodes
that are export pinned to this rank.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 11 Apr 2017 19:16:02 +0000 (15:16 -0400)]
mds: add export_pin feature
This allows the client/admin to pin a directory tree to a particular rank,
preventing its export by the dynamic balancer.
Fixes: http://tracker.ceph.com/issues/17834
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 18 Apr 2017 02:39:19 +0000 (22:39 -0400)]
mds: unify export_targets handling for all exports
This commit moves the MDSMap export_targets updates/handling to MDSRank. It is
necessary to wait for export_targets to be updated prior to doing any exports
as clients must have sessions open with targets of exports before any export
can be performed. Before this commit, this handling was only done for
migrations initiated by the balancer and not for manual migrations done by the
admin socket.
This fix and refactoring does the following:
o MDSRank now manages export_targets via a map of ranks with DecayCounters.
MDSRank::hit_export_target enables the Migrator/MDBalancer to hit a rank to
indicate migration is either desired or in progress. Importantly, updating
export_targets is now no longer tied to the previous MDBalancer cycle.
o mds_bal_target_removal_min and mds_bal_target_removal_max are removed in
favor of a DecayRate, via mds_bal_target_decay, which is independent of the
tick rate.
o Certain balancing state has been pulled out of the MDBalancer into a stack
variable type (balance_state_t). This is to avoid unnecessary persistence
of my_targets, imported, and exported maps which made the code confusing.
o try_rebalance is no longer called on MDSMap updates. This was done before
export target checking was part of the balancer, in
3e36d3202 .
o The Migrator now hits a rank in export_targets via MDSRank::hit_export_target
proportional to how much is being exported and periodically during the
course of the export. In my testing, one "default" hit (-1) will at least
keep the target in the export map for about 2 minutes.
o The Migrator will wait until a target is in the export_targets before
it actually does the export, or abort the export if the target is not
added in a timely manner.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Mon, 17 Apr 2017 20:03:58 +0000 (16:03 -0400)]
common: assoc. DecayRate with DecayCounter
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 18 Apr 2017 02:35:00 +0000 (22:35 -0400)]
mds: dispatch export request
Funnel dispatch through MDCache::dispatch_request so we have only one call site
for Migrator::dispatch_export_dir.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 18 Apr 2017 02:33:19 +0000 (22:33 -0400)]
mds: check projected parent to avoid unsafe access
An anonymous inode may not have a stable parent so immediate migration would
cause a segfault when checking for strays.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 18 Apr 2017 02:28:13 +0000 (22:28 -0400)]
common/DecayCounter: add delta to value for ::get
This resolves a bug where getting the DecayCounter value before one second has
elapsed will result in 0 always being returned. This is because ::hit will add
to the delta, not the value. The delta is added to the value only in the decay
function which only processes changes after 1 second has elapsed since the last
decay.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 18 Apr 2017 02:27:37 +0000 (22:27 -0400)]
common/DecayCounter: remove redundant qualifiers
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Mon, 17 Apr 2017 20:09:04 +0000 (16:09 -0400)]
mds: use uint64_t for sum of exported inodes
An int is too likely to overflow.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Mon, 17 Apr 2017 20:07:14 +0000 (16:07 -0400)]
mds: organize headers
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Mon, 17 Apr 2017 20:04:51 +0000 (16:04 -0400)]
messages: add missing header to MMDSLoadTargets
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 11 Apr 2017 19:07:44 +0000 (15:07 -0400)]
mds: add comment for subtrees MDCache member
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 11 Apr 2017 19:06:22 +0000 (15:06 -0400)]
mds: simplify loops to range-for
No functional change.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 11 Apr 2017 19:00:06 +0000 (15:00 -0400)]
mds: organize Migrator headers
This follows the coding style guidelines:
https://google.github.io/styleguide/cppguide.html#Declaration_Order
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 11 Apr 2017 18:59:28 +0000 (14:59 -0400)]
mds: remove unimplemented function
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Thu, 12 Jan 2017 09:36:06 +0000 (04:36 -0500)]
mds: order MDBalancer header
This follows the coding style guidelines:
https://google.github.io/styleguide/cppguide.html#Declaration_Order
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Josh Durgin [Tue, 11 Apr 2017 19:35:48 +0000 (12:35 -0700)]
Merge pull request #14461 from dmick/master
stop.sh: boilerplate error (don't stop mon when stopping mgr)
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Dan Mick [Tue, 11 Apr 2017 01:35:12 +0000 (18:35 -0700)]
stop.sh: boilerplate error (don't stop mon when stopping mgr)
Signed-off-by: Dan Mick <dan.mick@redhat.com>
Orit Wasserman [Tue, 11 Apr 2017 17:30:33 +0000 (20:30 +0300)]
Merge pull request #14376 from cbodley/wip-rgw-forward-canned-acl
rgw: fix for EINVAL errors on forwarded bucket put_acl requests
Reviewed-by: Orit Wasserman <owasserm@redhat.com>
Jason Dillaman [Tue, 11 Apr 2017 15:24:21 +0000 (11:24 -0400)]
Merge pull request #13105 from rjfd/wip-18481
librbd: deferred image deletion
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Casey Bodley [Tue, 11 Apr 2017 15:13:04 +0000 (11:13 -0400)]
Merge pull request #14355 from cbodley/wip-19505
rgw: allow larger payload for period commit
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
Casey Bodley [Tue, 11 Apr 2017 15:12:44 +0000 (11:12 -0400)]
Merge pull request #14319 from cbodley/wip-19488
rgw: multisite: fixes for zonegroup redirect
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
Casey Bodley [Tue, 11 Apr 2017 15:11:41 +0000 (11:11 -0400)]
Merge pull request #14343 from Liuchang0812/wip-rgw-read-user-buckets
rgw: update is_truncated in function rgw_read_user_buckets
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Kefu Chai [Tue, 11 Apr 2017 14:10:37 +0000 (22:10 +0800)]
Merge pull request #14438 from theanalyst/fix/build-doc
script: build-doc/serve-doc fixes
Reviewed-by: Loic Dachary <ldachary@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
Sage Weil [Tue, 11 Apr 2017 13:47:40 +0000 (08:47 -0500)]
Merge pull request #14455 from optimistyzy/411_1
os/bluestore/NVMEDevice: fix the compilation issue for collect_metadata
Reviewed-by: Haomai Wang <haomai@xsky.com>
John Spray [Tue, 11 Apr 2017 13:06:19 +0000 (14:06 +0100)]
Merge pull request #13698 from ukernel/wip-multimds-misc
mds: miscellaneous multimds fixes
Reviewed-by: John Spray <john.spray@redhat.com>
Jason Dillaman [Tue, 11 Apr 2017 12:10:20 +0000 (08:10 -0400)]
Merge pull request #14453 from trociny/wip-test_mock_InstanceWatcher
test/rbd_mirror: race in TestMockInstanceWatcher on destroy
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Ricardo Dias [Mon, 27 Mar 2017 14:34:26 +0000 (15:34 +0100)]
pybind: librbd: added trash API to python RBD binding
Signed-off-by: Ricardo Dias <rdias@suse.com>
Ricardo Dias [Thu, 2 Mar 2017 15:29:36 +0000 (15:29 +0000)]
qa/workunits/rbd: cli testing of rbd trash commands
Signed-off-by: Ricardo Dias <rdias@suse.com>
Ricardo Dias [Fri, 24 Mar 2017 13:50:52 +0000 (13:50 +0000)]
rbd: added image-id optional to Info command
Signed-off-by: Ricardo Dias <rdias@suse.com>
Ricardo Dias [Mon, 10 Apr 2017 13:32:46 +0000 (14:32 +0100)]
rbd: Added image-id optional to Snap commands
Signed-off-by: Ricardo Dias <rdias@suse.com>
Ricardo Dias [Mon, 10 Apr 2017 13:31:54 +0000 (14:31 +0100)]
rbd: added image_id param to Utils::init_and_open_image
Signed-off-by: Ricardo Dias <rdias@suse.com>
Ricardo Dias [Mon, 10 Apr 2017 13:26:52 +0000 (14:26 +0100)]
rbd: added group image remove by id to CLI
Signed-off-by: Ricardo Dias <rdias@suse.com>
Ricardo Dias [Mon, 10 Apr 2017 13:26:21 +0000 (14:26 +0100)]
librbd: added group image remove by id to API
Signed-off-by: Ricardo Dias <rdias@suse.com>
Ricardo Dias [Fri, 24 Feb 2017 18:37:43 +0000 (18:37 +0000)]
rbd: added CLI support for trash operations
Signed-off-by: Ricardo Dias <rdias@suse.com>
Ricardo Dias [Fri, 24 Feb 2017 18:36:55 +0000 (18:36 +0000)]
librbd: added trash methods to API
Signed-off-by: Ricardo Dias <rdias@suse.com>
Ricardo Dias [Fri, 24 Feb 2017 18:35:31 +0000 (18:35 +0000)]
librbd: open image gets name from directory or trash
Signed-off-by: Ricardo Dias <rdias@suse.com>
Ricardo Dias [Fri, 24 Feb 2017 18:00:00 +0000 (18:00 +0000)]
librbd: added open an image by id to API
Signed-off-by: Ricardo Dias <rdias@suse.com>
Ricardo Dias [Fri, 24 Feb 2017 17:57:04 +0000 (17:57 +0000)]
cls_rbd: methods to manipulate the rbd_trash object
Signed-off-by: Ricardo Dias <rdias@suse.com>
Ziye Yang [Tue, 11 Apr 2017 10:31:36 +0000 (18:31 +0800)]
bluestore/NVMEDevice: fix the compilation issue for collect_metadata
Signed-off-by: optimistyzy <optimistyzy@gmail.com>
Mykola Golub [Tue, 11 Apr 2017 09:55:59 +0000 (11:55 +0200)]
test/rbd_mirror: race in TestMockInstanceWatcher on destroy
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
Loic Dachary [Tue, 11 Apr 2017 09:44:29 +0000 (11:44 +0200)]
Merge pull request #14295 from theanalyst/release/10.2.7/notes
doc: release notes for v10.2.7 Jewel
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
Loic Dachary [Tue, 11 Apr 2017 09:43:15 +0000 (11:43 +0200)]
Merge pull request #14441 from theanalyst/doc/changelog/v10.2.7
doc: add changelog for v10.2.7
Reviewed-by: Loic Dachary <ldachary@redhat.com>
Kefu Chai [Tue, 11 Apr 2017 09:41:56 +0000 (17:41 +0800)]
Merge pull request #14426 from tchaikov/wip-13562-kefu
misc: SCA fixes
Reviewed-by: Kefu Chai <kchai@redhat.com>
Haomai Wang [Tue, 11 Apr 2017 01:13:03 +0000 (09:13 +0800)]
Merge pull request #13940 from Adirl/out
msg/async/rdma: Update fix broken compilation
Reviewed-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Tue, 11 Apr 2017 01:11:39 +0000 (09:11 +0800)]
Merge pull request #14409 from Adirl/conn_info
msg/async/rdma: RDMA-CM, Pass specific ConnMgr info in constructor
Reviewed-by: Haomai Wang <haomai@xsky.com>
Haomai Wang [Tue, 11 Apr 2017 01:10:26 +0000 (09:10 +0800)]
Merge pull request #14410 from Adirl/get_device
msg/async/rdma: RDMA-CM, get_device() by ibv_context
Reviewed-by: Haomai Wang <haomai@xsky.com>
Sage Weil [Mon, 10 Apr 2017 21:24:14 +0000 (16:24 -0500)]
Merge pull request #13073 from chardan/jfw-wip-15475-admin_heap_command
osd: add "heap *" admin command
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Sage Weil [Mon, 10 Apr 2017 21:23:18 +0000 (16:23 -0500)]
Merge pull request #14235 from nvedant07/bug_fix_3
rados: more info added to pool deletion error
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil [Mon, 10 Apr 2017 21:22:59 +0000 (16:22 -0500)]
Merge pull request #14022 from batrick/admin-socket-close-error
common: do not print error when asok is closed
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil [Mon, 10 Apr 2017 21:22:37 +0000 (16:22 -0500)]
Merge pull request #14292 from joscollin/wip-radosacl-warning
tools: warning, ‘%.16x’ directive output truncated writing 16 bytes into a region of size 9.
Reviewed-by: Kefu Chai <kchai@redhat.com>
Sage Weil [Mon, 10 Apr 2017 21:22:21 +0000 (16:22 -0500)]
Merge pull request #14301 from Andrey1994/master
build: Add Virtuozzo Linux support
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil [Mon, 10 Apr 2017 21:22:07 +0000 (16:22 -0500)]
Merge pull request #14336 from Adirl/fix_sockaddr
src/msg/async: Update fix broken compilation for Posix
Reviewed-by: Haomai Wang <haomai@xsky.com>
Abhishek Lekshmanan [Mon, 10 Apr 2017 20:33:23 +0000 (22:33 +0200)]
doc: add changelog for v10.2.7
Signed-off-by: Abhishek Lekshmanan <alekshmanan@suse.com>
Yuri Weinstein [Mon, 10 Apr 2017 20:30:32 +0000 (13:30 -0700)]
Merge pull request #14039 from dachary/wip-19307-set-deviceclass
mon/OSDMonitor: osd crush set-device-class
Reviewed-by: Sage Weil <sage@redhat.com>
Yuri Weinstein [Mon, 10 Apr 2017 20:28:41 +0000 (13:28 -0700)]
Merge pull request #14208 from dachary/wip-crush-uniform-weights
crush: bucket: crush_add_uniform_bucket_item should check for uniformity
Reviewed-by: Sage Weil <sage@redhat.com>
Yuri Weinstein [Mon, 10 Apr 2017 20:27:32 +0000 (13:27 -0700)]
Merge pull request #14209 from dachary/wip-crush-create-default-tunables
crush: builder: creating crush map with optimal configurations
Reviewed-by: Sage Weil <sage@redhat.com>
Mykola Golub [Mon, 10 Apr 2017 19:23:53 +0000 (22:23 +0300)]
Merge pull request #14401 from dillaman/wip-rbd-import-fixes
rbd: correct issues with image importing
Reviewed-by: Mykola Golub <mgolub@mirantis.com>
Sage Weil [Mon, 10 Apr 2017 19:20:32 +0000 (14:20 -0500)]
Merge pull request #14437 from cernceph/dvanders_mgrrotate
ceph-mgr: rotate logs on sighup
Reviewed-by: Sage Weil <sage@redhat.com>
Abhishek Lekshmanan [Mon, 10 Apr 2017 17:16:56 +0000 (19:16 +0200)]
script: serve-doc print the port we're serving
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
Abhishek Lekshmanan [Mon, 10 Apr 2017 17:15:25 +0000 (19:15 +0200)]
script: build-doc use python2 only for virtualenv
This is because the sphinx-ditaa package and ceph's rbd/rgw/rados pybind
require python2
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
Abhishek Lekshmanan [Mon, 3 Apr 2017 09:40:56 +0000 (11:40 +0200)]
doc: release notes for v10.2.7 Jewel
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
Kefu Chai [Mon, 10 Apr 2017 16:25:25 +0000 (00:25 +0800)]
Merge pull request #14413 from hansbogert/patch-3
doc: rgw: remove mention of megabytes for quotas
Reviewed-by: Daniel Gryniewicz <dang@redhat.com>
Dan van der Ster [Mon, 10 Apr 2017 15:34:48 +0000 (17:34 +0200)]
logrotate.conf: also sighup the ceph-mgr
Tell ceph-mgr to reopen its logs postrotate.
Fixes: http://tracker.ceph.com/issues/19568
Signed-off-by: Dan van der Ster <daniel.vanderster@cern.ch>
Danny Al-Gaaf [Tue, 21 Feb 2017 15:50:13 +0000 (16:50 +0100)]
ceph_fuse.cc: fix 64/32bit detection
Fix for:
[src/ceph_fuse.cc:111]: (style) Condition 'sizeof(long)==4' is always false
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Danny Al-Gaaf [Tue, 21 Feb 2017 15:39:34 +0000 (16:39 +0100)]
lua_bufferlist.cc: use static_cast instead of c-style
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Danny Al-Gaaf [Tue, 21 Feb 2017 15:25:00 +0000 (16:25 +0100)]
BitAllocator_test.cc: use static_cast instead of c-style
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Danny Al-Gaaf [Tue, 21 Feb 2017 15:24:43 +0000 (16:24 +0100)]
test_librbd.cc: use static_cast instead of c-style
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Danny Al-Gaaf [Tue, 21 Feb 2017 15:24:26 +0000 (16:24 +0100)]
rgw_rest_s3.cc: use static_cast instead of c-style
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Danny Al-Gaaf [Tue, 21 Feb 2017 15:24:06 +0000 (16:24 +0100)]
PrimaryLogPG.cc: use static_cast instead of c-style
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Danny Al-Gaaf [Tue, 21 Feb 2017 15:23:40 +0000 (16:23 +0100)]
osd/PG.cc: use static_cast instead of c-style
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Danny Al-Gaaf [Tue, 21 Feb 2017 15:23:12 +0000 (16:23 +0100)]
BitAllocator.cc: use static_cast instead of c-style
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Kefu Chai [Mon, 10 Apr 2017 15:31:06 +0000 (23:31 +0800)]
Merge pull request #14430 from jcsp/wip-19566-monassert
mon: avoid segfault in wait_auth_rotating
Reviewed-by: Kefu Chai <kchai@redhat.com>
Dan van der Ster [Mon, 10 Apr 2017 15:29:05 +0000 (17:29 +0200)]
ceph-mgr: reopen logs on SIGHUP
Wire up the sighup_handler to reopen logs, like the other daemons.
Fixes: http://tracker.ceph.com/issues/19568
Signed-off-by: Dan van der Ster <daniel.vanderster@cern.ch>
Kefu Chai [Mon, 10 Apr 2017 15:16:09 +0000 (23:16 +0800)]
Merge pull request #14428 from tchaikov/wip-ceph_objectstore_tool-kill-all
tests: ceph_objectstore_tool.py: kill all daemons
Reviewed-by: Willem Jan Withagen <wjw@digiware.nl>
John Spray [Mon, 10 Apr 2017 12:52:04 +0000 (13:52 +0100)]
Merge pull request #10636 from fullerdj/wip-djf-15069
cephfs: Permit recovering metadata into a new RADOS pool
Reviewed-by: John Spray <john.spray@redhat.com>
John Spray [Mon, 10 Apr 2017 12:00:41 +0000 (13:00 +0100)]
mon: avoid segfault in wait_auth_rotating
MonClient users should not be calling into MonClient
after calling shutdown(). However, MonClient should
assert out rather than proceeding to try and
follow a maybe-null pointer.
Related to http://tracker.ceph.com/issues/19566
Signed-off-by: John Spray <john.spray@redhat.com>
Kefu Chai [Sat, 8 Apr 2017 14:27:33 +0000 (22:27 +0800)]
tests: ceph_objectstore_tool.py: kill all daemons
init-ceph kills all daemons that can be identified by get_name_list() by
default, so no need to list them explicitly. also, without this change,
mgr is not stopped when the test finishes.
Signed-off-by: Kefu Chai <kchai@redhat.com>