]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Abhishek Lekshmanan [Fri, 15 Sep 2017 21:20:27 +0000 (23:20 +0200)]
doc: rgw: add a note for resharding in 12.2.1 docs
Since we're now enabling dynamic resharding, mention this in release
notes
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
Sage Weil [Fri, 15 Sep 2017 20:35:03 +0000 (15:35 -0500)]
Merge pull request #17709 from liewegas/wip-bluestore-repair
ceph-bluestore-tool: better fsck/repair, bluefs-bdev-{expand,sizes}
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
Kefu Chai [Fri, 15 Sep 2017 17:23:56 +0000 (01:23 +0800)]
Merge pull request #17388 from tchaikov/wip-rocksdb
rocksdb: sync with upstream
Reviewed-by: Mark Nelson <mnelson@redhat.com>
Reviewed-by: Sage Weil <sage@redhat.com>
Kefu Chai [Fri, 15 Sep 2017 11:57:36 +0000 (19:57 +0800)]
Merge pull request #17706 from liewegas/wip-test-listsnaps
ceph_test_rados_api_tier: add ListSnap test
Reviewed-by: Sage Weil <sage@redhat.com>
Kefu Chai [Fri, 15 Sep 2017 11:55:55 +0000 (19:55 +0800)]
Merge pull request #17705 from dzafman/wip-21328
osd: Only scan for omap corruption once
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Kefu Chai [Fri, 15 Sep 2017 11:54:58 +0000 (19:54 +0800)]
Merge pull request #17703 from dzafman/wip-misc
Erasure code read test and code cleanup
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Kefu Chai [Fri, 15 Sep 2017 11:54:04 +0000 (19:54 +0800)]
Merge pull request #17699 from tchaikov/wip-lookout-for-large-elen
msg/msg_types: fix the entity_addr_t's decoder
Reviewed-by: Sage Weil <sage@redhat.com>
Kefu Chai [Fri, 15 Sep 2017 11:52:08 +0000 (19:52 +0800)]
Merge pull request #17667 from scienceluo/wip-luo-fix-cli-output-branch
pybind/ceph_argparse: fix cli output info
Reviewed-by: Chang Liu <liuchang0812@gmail.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
Kefu Chai [Fri, 15 Sep 2017 11:51:31 +0000 (19:51 +0800)]
Merge pull request #17664 from Yan-waller/wip-walle-0912moninjectargs
common: fix daemon abnormal exit at parsing invalid arguments
Reviewed-by: Kefu Chai <kchai@redhat.com>
Kefu Chai [Fri, 15 Sep 2017 11:51:01 +0000 (19:51 +0800)]
Merge pull request #17655 from optimistyzy/911_2
src/bluestore/NVMEDevice: make all read use aio_submit
Reviewed-by: Haomai Wang <haomai@xsky.com>
Reviewed-by: Pan Liu <liupan1111@gmail.com>
Kefu Chai [Fri, 15 Sep 2017 11:49:52 +0000 (19:49 +0800)]
Merge pull request #17611 from chardan/jfw-wip-atomic_t-cleanup
follow-up fixups for atomic_t spinlocks
Reviewed-by: Adam Emerson <aemerson@redhat.com>
Kefu Chai [Fri, 15 Sep 2017 11:49:02 +0000 (19:49 +0800)]
Merge pull request #17605 from jcsp/wip-21253
mgr: fix py calls for dne service perf counters
Reviewed-by: Sage Weil <sage@redhat.com>
Kefu Chai [Fri, 15 Sep 2017 11:48:40 +0000 (19:48 +0800)]
Merge pull request #17568 from amitkumar50/cov-msg-2
msg: Initializing class members in module msg
Reviewed-by: Jos Collin <jcollin@redhat.com>
Kefu Chai [Fri, 15 Sep 2017 11:48:08 +0000 (19:48 +0800)]
Merge pull request #16699 from jcsp/wip-mgr-perf-threshold
mgr: apply a threshold to perf counter prios
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
Kefu Chai [Wed, 13 Sep 2017 14:55:58 +0000 (22:55 +0800)]
msg/msg_types: fix the entity_addr_t's decoder
the daemon is vulnerable to malicious client, which is able to send
large elen, and corrupt the stack, etc.
* throw at seeing corrupted entity_addr_t where its elen exceeds
the length of sockaddr
* handle the exception thrown when decoding entity_addr_t in messenger
layer.
* if a malicious client manages to send a corrutped entity_addr_t to
daemon, daemon will crash because decode fails and the exception is
not handled. it's better than continuing working with the bogus
message.
Signed-off-by: Kefu Chai <kchai@redhat.com>
Kefu Chai [Wed, 6 Sep 2017 02:46:13 +0000 (10:46 +0800)]
cmake: build portable rocksdb
* so rocksdb is not built with -march=native, because
- this is not portable
- -march=native is not supported by older version of GCC on aarch64.
* and drop the sse42 specific settings for rocksdb. since rocksdb uses
"target" attribute to build sse42 optimized crc32, as long as the
compiler on building host is able to emit sse42 instructions.
see https://github.com/facebook/rocksdb/pull/2807
Fixes: http://tracker.ceph.com/issues/20529
Signed-off-by: Kefu Chai <kchai@redhat.com>
Kefu Chai [Thu, 31 Aug 2017 05:02:08 +0000 (13:02 +0800)]
kv/RocksDBStore: update to accomodate the change in rocksdb 5.6.0
per rocksdb's HISTORY.md, it
> replace global variable `PerfContext perf_context` with `PerfContext*
> get_perf_context()`.
hence the change.
Signed-off-by: Kefu Chai <kchai@redhat.com>
Kefu Chai [Thu, 31 Aug 2017 04:54:16 +0000 (12:54 +0800)]
rocksdb: sync with upstream
to pickup the fix of https://github.com/facebook/rocksdb/pull/2692.
Signed-off-by: Kefu Chai <kchai@redhat.com>
Dan Mick [Fri, 15 Sep 2017 03:38:36 +0000 (20:38 -0700)]
Merge pull request #17361 from kungf/ulimit_setup
run-make-check.sh: run ulimit without sudo
Reviewed-by: Dan Mick <dmick@redhat.com>
yang.wang [Wed, 30 Aug 2017 08:32:49 +0000 (16:32 +0800)]
run-make-check.sh: run ulimit without sudo
ulimit is a shell builtin like cd, not a separate program.
sudo looks for a binary to run, but there is no ulimit binary,
then you will get an error message like:
"sudo: ulimit: command not found"
ulimit can only be run with root if you want raise the ulimit -n
Change-Id: I59ab09240cb15cabe5b0a7342c3f562cd50c75e9
Signed-off-by: yang.wang <yang.wang@easystack.cn>
Dan Mick [Fri, 15 Sep 2017 02:07:11 +0000 (19:07 -0700)]
Merge pull request #17713 from tchaikov/wip-ceph.in-python
ceph.in: execv using the same python
Josh Durgin [Thu, 14 Sep 2017 23:36:26 +0000 (16:36 -0700)]
Merge pull request #17583 from neha-ojha/wip-cbt-teuthology-integration
qa: add cbt task for performance testing
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
vasukulkarni [Thu, 14 Sep 2017 23:13:37 +0000 (16:13 -0700)]
Merge pull request #17680 from ceph/wip-smoke-whitelist
qa/tests: Various whitelists for smoke suite
Vasu Kulkarni [Tue, 12 Sep 2017 20:49:39 +0000 (13:49 -0700)]
Various whitelists for tests to pass
See tracker http://tracker.ceph.com/issues/21376 for details
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
Sage Weil [Wed, 13 Sep 2017 22:33:21 +0000 (18:33 -0400)]
ceph-bluestore-tool: better default logging; --log-file and --log-level options
- keep derr going to stderr (so we see 'fsck error' messages)
- hide the rest
- add friendly --log-file option
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Sun, 10 Sep 2017 19:37:10 +0000 (15:37 -0400)]
os/bluestore: fsck: fix error prefixes
__func__ is now _fsck instead of fsck
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Thu, 7 Sep 2017 22:27:20 +0000 (18:27 -0400)]
ceph-bluestore-tool: add 'bluefs-bdev-expand' to expand wal or db usage
If you are using the wal or db devices, this will expand bluefs's usage to
include the entire block device.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Thu, 7 Sep 2017 22:20:27 +0000 (18:20 -0400)]
ceph-bluestore-tool: add 'bluefs-bdev-sizes' command
Show bdev sizes vs owned extents.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Thu, 7 Sep 2017 22:12:21 +0000 (18:12 -0400)]
ceph-bluestore-tool: factor out bluefs mount
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Thu, 7 Sep 2017 16:29:32 +0000 (12:29 -0400)]
os/bluestore: allow repair invocation
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil [Thu, 7 Sep 2017 16:27:35 +0000 (12:27 -0400)]
os/ObjectStore: add repair interface
Signed-off-by: Sage Weil <sage@redhat.com>
Kefu Chai [Thu, 14 Sep 2017 14:53:47 +0000 (22:53 +0800)]
Merge pull request #17733 from yaozongyou/fix-doc-boolean-type-zero-to-false
doc: change boolean option default value from zero to false.
Reviewed-by: Kefu Chai <kchai@redhat.com>
Sage Weil [Thu, 14 Sep 2017 14:18:41 +0000 (09:18 -0500)]
Merge pull request #17610 from liewegas/wip-bluestore-fm-mas
os/bluestore: set bitmap freelist resolution to min_alloc_size
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
Reviewed-by: Varada Kari <varada.kari@gmail.com>
Kefu Chai [Thu, 14 Sep 2017 13:42:56 +0000 (21:42 +0800)]
Merge pull request #17617 from liewegas/wip-drop-disk-log
mon/PGMap: drop DISK LOG column
Reviewed-by: Kefu Chai <kchai@redhat.com>
Yao Zongyou [Thu, 14 Sep 2017 13:37:04 +0000 (21:37 +0800)]
doc: change boolean option default value from zero to false.
Signed-off-by: Yao Zongyou yaozongyou@vip.qq.com
Sage Weil [Thu, 14 Sep 2017 13:30:17 +0000 (08:30 -0500)]
Merge pull request #17628 from majianpeng/bluestore-narrow-deferred-lock
os/bluestore/BlueStore: narrow deferred_lock in _deferred_submit_unlock.
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
Xie Xingguo [Thu, 14 Sep 2017 10:27:02 +0000 (18:27 +0800)]
Merge pull request #17627 from majianpeng/bluefs-exce-debug-code
os/bluestore/BlueFS: Don't call debug related code under any condition.
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
Xie Xingguo [Thu, 14 Sep 2017 10:26:12 +0000 (18:26 +0800)]
Merge pull request #17371 from xiexingguo/wip-per-pool-full-control
mon, osd: per pool space-full flag support
Reviewed-by: Sage Weil <sage@redhat.com>
Xie Xingguo [Thu, 14 Sep 2017 10:25:24 +0000 (18:25 +0800)]
Merge pull request #17368 from xiexingguo/wip-nice-data
mon/PGMap: nice numbers for 'data' section of 'ceph df' command
Reviewed-by: Kefu Chai <kchai@redhat.com>
Kefu Chai [Wed, 13 Sep 2017 06:20:48 +0000 (14:20 +0800)]
ceph.in: execv using the same python
otherwise we could be using another python when respawing the ceph cli.
Signed-off-by: Kefu Chai <kchai@redhat.com>
Patrick Donnelly [Thu, 14 Sep 2017 03:08:51 +0000 (20:08 -0700)]
Merge PR #17657 into master
* refs/remotes/upstream/pull/17657/head:
mds: optimize MDCache::rejoin_scour_survivor_replicas()
mds: fix MDSCacheObject::clear_replica_map
mds: support limiting cache by memory
common: refactor of lru
mds: resolve unsigned coercion compiler warning
common: use safer uint64_t for list size
common: add bytes2str pretty print function
mds: check if waiting is allocated before use
mds: go back to compact_map for replicas
mds: use mempool for cache objects
mds: cleanup replica_map access
common: add alloc_ptr smart pointer
common: add warning on base class use of mempool
common: use atomic uin64_t for counter
Reviewed-by: Zheng Yan <zyan@redhat.com>
Patrick Donnelly [Thu, 14 Sep 2017 01:29:20 +0000 (18:29 -0700)]
Merge PR #17689 into master
* refs/remotes/upstream/pull/17689/head:
ceph: fix "stopping mds can't export subtree"
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Thu, 14 Sep 2017 01:29:18 +0000 (18:29 -0700)]
Merge PR #17545 into master
* refs/remotes/upstream/pull/17545/head:
client: fix request reference leak
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Jeff Layton <jlayton@redhat.com>
vasukulkarni [Thu, 14 Sep 2017 00:44:24 +0000 (17:44 -0700)]
Merge pull request #17700 from cbodley/wip-qa-smoke-s3tests
qa/smoke: add rgw crypto config for s3tests
Xuehan Xu [Wed, 13 Sep 2017 21:46:59 +0000 (17:46 -0400)]
ceph_test_rados_api_tier: add ListSnap test
Signed-off-by: Xuehan Xu <xuxuehan@360.cn>
David Zafman [Wed, 13 Sep 2017 00:17:13 +0000 (17:17 -0700)]
osd: Only scan for omap corruption once
Before
state 2: Can have complete tables (some may be bad)
state 3: Never had complete tables
After
state 2: Can have complete tables (some may be bad)
state 3 with legacy: Can have complete tables (bad ones are cleared)
state 3: Never had complete tables
Once OSDs boot with this change you can't downgrade to a previous release.
If someone does downgrade they could have unstable OSDs that hit assert(state.v < 3).
The following command run after shutting down the cluster but before downgrading
ceph packages would be a way to fix this.
ceph-osdomap-tool --omap-path ... --command resetv2
Fixes: http://tracker.ceph.com/issues/21328
Signed-off-by: David Zafman <dzafman@redhat.com>
David Zafman [Wed, 13 Sep 2017 06:14:15 +0000 (23:14 -0700)]
tools: Add the ability to reset state to v2
Available for testing and user downgrade.
Signed-off-by: David Zafman <dzafman@redhat.com>
David Zafman [Wed, 13 Sep 2017 05:12:52 +0000 (22:12 -0700)]
tools: Show DB state information
Signed-off-by: David Zafman <dzafman@redhat.com>
David Zafman [Wed, 13 Sep 2017 01:06:10 +0000 (18:06 -0700)]
tools: Add --backend option to ceph-osdomap-tool default to rocksdb
Fix hard-coded "leveldb" backend. The command is broken in Luminous
now that "rocksdb" is the default.
Signed-off-by: David Zafman <dzafman@redhat.com>
Patrick Donnelly [Wed, 13 Sep 2017 20:37:57 +0000 (13:37 -0700)]
Merge PR #17608 into master
* refs/remotes/upstream/pull/17608/head:
doc/cephfs/posix: put posix notes in perspective
Reviewed-by: John Spray <john.spray@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
David Zafman [Tue, 12 Sep 2017 22:09:14 +0000 (15:09 -0700)]
osd: CLEANUP: Change getattr_maybe_cache() to reflect how it is being used
Signed-off-by: David Zafman <dzafman@redhat.com>
David Zafman [Tue, 12 Sep 2017 22:03:33 +0000 (15:03 -0700)]
test: Add a removal test for erasure code read
Test feature: http://tracker.ceph.com/issues/14513
Signed-off-by: David Zafman <dzafman@redhat.com>
Yuri Weinstein [Wed, 13 Sep 2017 19:57:45 +0000 (12:57 -0700)]
Merge pull request #17641 from amitkumar50/cov-messages-12
messages: Initialization of variable beat
Reviewed-by: Jos Collin <jcollin@redhat.com>
Yuri Weinstein [Wed, 13 Sep 2017 15:30:03 +0000 (08:30 -0700)]
Merge pull request #10460 from weiqiaomiao/wqm-wip-rgw-dellc
rgw: add tenant to shard_id in RGWDeleteLC::execute()
Reviewed-by: Daniel Gryniewicz <dang@redhat.com>
Yuri Weinstein [Wed, 13 Sep 2017 15:28:24 +0000 (08:28 -0700)]
Merge pull request #17061 from shashalu/fix-rgw-lc
rgw: fix lc process only schdule the first item of lc objects
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Reviewed-by: fang.yuxiang <fang.yuxiang@eisoo.com>
Reviewed-by: Amit Kumar <amitkuma@redhat.com>
Reviewed-by: Daniel Gryniewicz <dang@redhat.com>
Reviewed-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
Casey Bodley [Wed, 13 Sep 2017 15:20:51 +0000 (11:20 -0400)]
qa/smoke: add rgw crypto config for s3tests
Signed-off-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Wed, 13 Sep 2017 15:20:38 +0000 (08:20 -0700)]
Merge pull request #17631 from joke-lee/RGW-CORS-Access-Control-Allow-Origin-HEADER
rgw: return 'Access-Control-Allow-Origin' header when the set bucket versioning through XMLHttpRequest
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Wed, 13 Sep 2017 15:20:17 +0000 (08:20 -0700)]
Merge pull request #17632 from joke-lee/RGW-CORS-Access-Control-Allow-Origin-HEADER-WEBSITE
rgw: return 'Access-Control-Allow-Origin' header when the set and delete bucket website through XMLHttpRequest
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Wed, 13 Sep 2017 15:19:33 +0000 (08:19 -0700)]
Merge pull request #17663 from shashalu/fix-rgw-LCFilter
rgw: Fix closing tag for Prefix
Reviewed-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
Reviewed-by: Jos Collin <jcollin@redhat.com>
Sage Weil [Wed, 13 Sep 2017 14:38:19 +0000 (09:38 -0500)]
Merge pull request #17543 from iliul/fix-dup-global-section
vstart.sh: Remove duplicate global section
Reviewed-by: Sage Weil <sage@redhat.com>
iliul [Thu, 7 Sep 2017 07:31:17 +0000 (07:31 +0000)]
vstart.sh: Remove duplicate global section
Signed-off-by: iliul <liul124@chinaunicom.cn>
Sage Weil [Wed, 13 Sep 2017 13:19:26 +0000 (08:19 -0500)]
Merge pull request #17544 from joelin/wip-delete-forgot-mark
msg: remove the ),it's redundant.
Reviewed-by: Adam Emerson <aemerson@redhat.com>
Xie Xingguo [Wed, 13 Sep 2017 10:31:10 +0000 (18:31 +0800)]
Merge pull request #17515 from xiexingguo/wip-data-digest
osd/PrimaryLogPG: do not set data digest for bluestore
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Sage Weil <sage@redhat.com>
Yan Jun [Wed, 13 Sep 2017 10:15:33 +0000 (18:15 +0800)]
test/config: regulate parameter order of ASSERT_EQ
Signed-off-by: Yan Jun <yan.jun8@zte.com.cn>
Yan Jun [Tue, 12 Sep 2017 03:20:34 +0000 (11:20 +0800)]
common: fix daemon abnormal exit at parsing invalid arguments
Signed-off-by: Yan Jun <yan.jun8@zte.com.cn>
Yan, Zheng [Wed, 13 Sep 2017 07:55:46 +0000 (15:55 +0800)]
mds: optimize MDCache::rejoin_scour_survivor_replicas()
avoid iterating dentries if dirfrag is non-auth
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
Yan, Zheng [Wed, 13 Sep 2017 07:54:45 +0000 (15:54 +0800)]
mds: fix MDSCacheObject::clear_replica_map
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
Nathan Cutler [Wed, 13 Sep 2017 07:05:01 +0000 (09:05 +0200)]
Merge pull request #17366 from smithfarm/wip-21176
build/ops: python-numpy-devel build dependency for SUSE
Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
Reviewed-by: Ewaznazari <Nazari@keyinternational.academy>
Jos Collin [Wed, 13 Sep 2017 05:29:35 +0000 (05:29 +0000)]
Merge pull request #17685 from ygtzf/HashIndex-clerical-error
os/filestore/HashIndex.h: fixed a typo in comment
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Jos Collin <jcollin@redhat.com>
Haomai Wang [Wed, 13 Sep 2017 05:25:26 +0000 (22:25 -0700)]
Merge pull request #17688 from TsaiJin/wip-fix-assert-bug
msg/asyc/rmda: fix the bug of assert when Infiniband::recv_msg receives disconnect message
Reviewed-by: Haomai Wang <haomai@xsky.com>
xie xingguo [Wed, 6 Sep 2017 02:25:02 +0000 (10:25 +0800)]
osd/osd_types.h: default to no data/omap digest for new object
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
xie xingguo [Tue, 5 Sep 2017 12:56:32 +0000 (20:56 +0800)]
osd/PrimaryLogPG: do not generate data digest for BlueStore by default
BlueStore enables CRC by default, so this is a dup and gains
no more benefits.
Turn this off by default, which is good for performance.
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
Yan, Zheng [Wed, 13 Sep 2017 03:27:18 +0000 (11:27 +0800)]
ceph: fix "stopping mds can't export subtree"
caused by commit
790c34138f "Make standby-replay mds avoid initiating subtree export"
Fixes: http://tracker.ceph.com/issues/21378
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
Jos Collin [Wed, 13 Sep 2017 03:39:23 +0000 (03:39 +0000)]
Merge pull request #17687 from yonghengdexin735/wip-zzz-error
common: fix typo deamon in comments
Reviewed-by: Jos Collin <jcollin@redhat.com>
Patrick Donnelly [Mon, 11 Sep 2017 22:21:52 +0000 (15:21 -0700)]
mds: support limiting cache by memory
This introduces two config parameters:
mds_cache_memory_limit: Sets the soft maximum of the cache to the given
byte count. (Like mds_cache_size, this doesn't actually limit the maximum
size of the cache. It just dictates the steady-state size.)
mds_cache_reservation: This replaces mds_health_cache_threshold everywhere
except the Beacon heartbeat sent to the mons. The idea here is to specify a
reservation of memory (5% by default) for operations and the MDS tries to
always maintain that reservation. So, the MDS will recall caps from clients
when it begins dipping into its reservation of memory.
mds_cache_size still limits the cache by Inode count but is now by-default 0
(i.e. unlimited). The new preferred way of specifying cache limits is by memory
size. The default is 1GB.
Fixes: http://tracker.ceph.com/issues/20594
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1464976
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Jin Cai [Wed, 13 Sep 2017 02:32:08 +0000 (10:32 +0800)]
msg/asyc/rmda: fix the bug of assert when Infiniband::recv_msg receives
disconnect message
Signed-off-by: Jin Cai <caijin.caij@alibaba-inc.com>
yonghengdexin735 [Wed, 13 Sep 2017 02:22:08 +0000 (10:22 +0800)]
common:fix error word
Signed-off-by: yonghengdexin735 <zhang.zezhu@zte.com.cn>
Sage Weil [Wed, 13 Sep 2017 02:12:50 +0000 (21:12 -0500)]
Merge pull request #17613 from smithfarm/wip-12056
doc: document include/ipaddr.h
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil [Wed, 13 Sep 2017 02:03:42 +0000 (21:03 -0500)]
Merge pull request #17644 from Vogtinator/readme-vstart
README: Improve vstart.sh usage
yaoguotao [Wed, 13 Sep 2017 01:59:53 +0000 (21:59 -0400)]
os/filestore/HashIndex.h: fixed a clerical error
Signed-off-by: yaoguotao <yaoguot@gmail.com>
Luo Kexue [Tue, 12 Sep 2017 07:20:04 +0000 (15:20 +0800)]
pybind/ceph_argparse: fix cli output info
Signed-off-by: Luo Kexue <luo.kexue@zte.com.cn>
Sage Weil [Tue, 12 Sep 2017 23:56:25 +0000 (18:56 -0500)]
Merge pull request #17654 from yuyuyu101/wip-spdk-doc
doc/bluestore: add SPDK usage for bluestore
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 12 Sep 2017 23:54:29 +0000 (18:54 -0500)]
Merge pull request #17658 from ukernel/wip-cephfs-seekdir
doc/cephfs/posix: remove stale information for seekdir
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 12 Sep 2017 23:39:14 +0000 (18:39 -0500)]
Merge pull request #17669 from ygtzf/parameter-default-value-error
doc: update default value of parameter mon_subscribe_interval
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil [Tue, 12 Sep 2017 23:36:59 +0000 (18:36 -0500)]
Merge pull request #17672 from FrankYu/master
doc: the client inputs the pool name instead of pool ID
Reviewed-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
Patrick Donnelly [Tue, 12 Sep 2017 22:56:32 +0000 (15:56 -0700)]
Merge PR #17679 into master
* refs/remotes/upstream/pull/17679/head:
qa: get asok path from ceph.conf
qa: use config_path property instead of literal
Reviewed-by: John Spray <john.spray@redhat.com>
Patrick Donnelly [Tue, 12 Sep 2017 22:51:30 +0000 (15:51 -0700)]
qa: get asok path from ceph.conf
The asok path was recently moved to /tmp to resolve unix path too long errors.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 12 Sep 2017 22:50:33 +0000 (15:50 -0700)]
qa: use config_path property instead of literal
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Fri, 8 Sep 2017 04:01:31 +0000 (21:01 -0700)]
common: refactor of lru
Avoids an unnecessary "max" size of the LRU which was used to calculate the
midpoint. Instead, just dynamically move the LRUObjects between top and bottom
on-the-fly.
This change is necessary for a cache which which does not limit by the number
of objects but by some other metric. (In this case, memory.)
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 12 Sep 2017 20:40:17 +0000 (13:40 -0700)]
mds: resolve unsigned coercion compiler warning
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Fri, 8 Sep 2017 03:51:56 +0000 (20:51 -0700)]
common: use safer uint64_t for list size
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Sun, 30 Jul 2017 23:09:32 +0000 (16:09 -0700)]
common: add bytes2str pretty print function
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 12 Sep 2017 21:31:18 +0000 (14:31 -0700)]
mds: check if waiting is allocated before use
This prevents accidental allocation of the map.
Also, privatize the variable to protect from this in child classes.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Tue, 12 Sep 2017 21:29:49 +0000 (14:29 -0700)]
mds: go back to compact_map for replicas
Zheng observed that an alloc_ptr doesn't really work in this case since any
call to get_replicas() will cause the map to be allocated, nullifying the
benefit. Use a compact_map until a better solution can be written. (This means
that the map will be allocated outside the mempool.)
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Fri, 28 Jul 2017 00:21:54 +0000 (17:21 -0700)]
mds: use mempool for cache objects
The purpose of this is to allow us to track memory usage by cached objects so
we can limit cache size based on memory available/allocated to the MDS.
This commit is a first step: it adds CInode, CDir, and CDentry to the mempool
but not all of the containers in these classes (e.g. std::map). However,
MDSCacheObject has been changed to allocate its containers through the mempool
by converting compact_* containers to the std versions offered through mempool
via the new alloc_ptr.
(A compact_* class simply wraps a pointer to the std:: version to reduce memory
usage of an object when the container is only occasionally used. The alloc_ptr
allows us to achieve the same thing explicitly with only a little handholding:
when all entries in the wrapped container are deleted, the caller must call
alloc_ptr.release().)
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Thu, 27 Jul 2017 19:06:23 +0000 (12:06 -0700)]
mds: cleanup replica_map access
The gymnastics protecting the map failed as the code evolved. Just expose it
normally with a getter.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Patrick Donnelly [Thu, 27 Jul 2017 19:10:14 +0000 (12:10 -0700)]
common: add alloc_ptr smart pointer
This ptr is like a unique_ptr except it allocates the underlying object on
access. The idea being that we can save memory if the object is only needed
sometimes.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Tamilarasi Muthamizhan [Tue, 12 Sep 2017 22:07:24 +0000 (15:07 -0700)]
Merge pull request #17650 from ceph/wip-ceph-deploy-fix
qa/tests: Fix get_system_type failure due to invalid remote name
Mykola Golub [Tue, 12 Sep 2017 18:17:19 +0000 (21:17 +0300)]
Merge pull request #17636 from dillaman/wip-21319
rbd: clean up warnings when mirror commands used on non-setup pool
Reviewed-by: Mykola Golub <mgolub@mirantis.com>
Neha Ojha [Tue, 12 Sep 2017 15:26:27 +0000 (08:26 -0700)]
qa: avoid using make install for fio
Signed-off-by: Neha Ojha <nojha@redhat.com>