common: get rid of "warning: ignoring return value of ‘strerror_r’"

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>

Merge pull request #12770 from yuriw/wip-fix-distro-master

qa: fixed distros links

Reviewed-by: Josh Durgin <jdurgin@redhat.com>

qa: fixed distros links

Signed-off-by: Yuri Weinstein <yweinste@redhat.com>

Merge pull request #11787 from SUSE/wip-17802

doc: clarify "ceph quorum" syntax

Reviewed-by: Joao Eduardo Luis <joao@suse.de>
Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12385 from dlan17/master

cmake: explictly call find_package(PythonInterp) first to fix build err

Reviewed-by: Ali Maredia <amaredia@redhat.com>

Merge pull request #12658 from oritwas/wip-rgw-17239

rgw: clear master_zonegroup when reseting RGWPeriodMap

Reviewed-by: Casey Bodley <cbodley@redhat.com>

Merge pull request #12723 from gaurav36/rbd-remove-unused-condition

rbd: remove unused condition within group action handler

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #12509 from SUSE/wip-18262

doc: add verbiage to rbdmap manpage

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #12667 from trociny/wip-librados_test_stub-tmap_udate

test: librados_test_stub: tmap_update: return -ENOENT when removing nonexisent key

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #12712 from liewegas/wip-18366

mon/OSDMonitor: set last_force_op_resend on overlay pool too

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12717 from hrchu/patch-1

doc: Update keystone.rst

Reviewed-by: Sage Weil <sage@redhat.com>

doc: Update keystone.rst

Make it clear that project/domain options accept name instead of id.

Signed-off-by: Chu, Hua-Rong <hrchu@cht.com.tw>

Merge pull request #12734 from liupan1111/wip-fix-nvme-driver-compile-issue

spdk: upgrade spdk to v16.12

Reviewed-by: Haomai Wang <haomai@xsky.com>

Merge pull request #12751 from tchaikov/wip-18395

ceph_test_rados_api_watch_notify: move global variables into test class

Reviewed-by: Brad Hubbard <bhubbard@redhat.com>

Merge pull request #12730 from Liuchang0812/fix-dev-doc-pglog

doc/dev: update log_based_pg.rst, fix some display problem

Reviewed-by: Kefu Chai <kchai@redhat.com>

ceph_test_rados_api_watch_notify: move global variables into test class

so the class instances using mempool are destructed before the mempools
are destroyed. this avoids the "invalid read" issue reported by
valgrind.

Fixes: http://tracker.ceph.com/issues/18395
Signed-off-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12713 from badone/wip-ceph_test_rados_api_watch_notify-valgrind-fixes

test_rados_watch_notify: Fix trivial memory leaks

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12680 from badone/wip-ceph_test_rados_api_misc-valgrind-fixes

ceph_test_rados_api_misc: Fix trivial memory leak

Reviewed-by: Kefu Chai <kchai@redhat.com>

doc/dev: update log_based_pg.rst, fix some display problem

Signed-off-by: liuchang0812 <liuchang0812@gmail.com>

os/bluestore: add more useful log information in NVMEDevice.cc

Signed-off-by: Pan Liu <pan.liu@istuary.com>

Merge pull request #12747 from SUSE/wip-18388

tests: rbd/test_lock_fence.sh: fix rbdrw.py relative path

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

tests: rbd/test_lock_fence.sh: fix rbdrw.py relative path

This commit fixes a regression introduced by
cf294777ea92f0911813a7132068584d4f73a65a

Fixes: http://tracker.ceph.com/issues/18388
Signed-off-by: Nathan Cutler <ncutler@suse.com>

os/bluestore: fixed compilation issue when use spdk v16.12

Signed-off-by: Pan Liu <pan.liu@istuary.com>

spdk: fixed the compilation error for testcases when use spdk v16.12

Signed-off-by: Pan Liu <pan.liu@istuary.com>

dpdk: make compilation successful when use dpdk v16.11

Signed-off-by: Pan Liu <pan.liu@istuary.com>

Merge pull request #12706 from Liuchang0812/wip-qa-rados-1228

test: update test_rados_tool.sh, use POOL and OBJ var

Reviewed-by: Kefu Chai <kchai@redhat.com>

spdk: upgrade spdk version to 16.12

Signed-off-by: Pan Liu <pan.liu@istuary.com>

dpdk: upgrade dpdk version to 16.11

Signed-off-by: Pan Liu <pan.liu@istuary.com>

Merge pull request #12716 from kylinstorage/wip-ctypos_h

misc: fix code typos in header files

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>

Merge pull request #12721 from chendave/wip-authx-options

auth: Let's not use the deprecated cephx option

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12676 from chendave/master

common/ceph_context: Show clear message if all features are enabled

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12633 from liewegas/wip-15784

msg/simple: call clear_pipe in wait() shutdown path

Reviewed-by: Haomai Wang <haomai@xsky.com>

Merge pull request #12692 from XinzeChi/wip-bluestore-statfs

os/bluestore: _txc_release_alloc when do wal cleaning

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12691 from runsisi/wip-dump-osdop

osd/OpRequest: dump both name and addr for the client op

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12363 from wjwithagen/wip-wjw-freebsd-strerror

common/module.c: do not use strerror_r the GNU way.

Reviewed-by: Sage Weil <sage@redhat.com>

rbd: remove unused condition

Signed-off-by: Gaurav Kumar Garg <garg.gaurav52@gmail.com>

cephx: Let's not use the deprecated option

Enable cephx by setting the option of "auth supported" has been
deprecated for a long time, it's better to use the recommended
options instead.

Signed-off-by: Dave Chen <wei.d.chen@intel.com>

Merge pull request #12718 from kylinstorage/wip-orig

organizationmap: add Xianxia Xiao to Kylin Cloud team

organizationmap: add Xianxia Xiao to Kylin Cloud team

Signed-off-by: Yunchuan Wen <yunchuan.wen@kylin-cloud.com>

src/tools/rbd_mirror/Replayer.h : fix typos

Signed-off-by: Xianxia Xiao <xianxia.xiao@kylin-cloud.com>

src/messages/MMDSSlaveRequest.h : fix typos

Signed-off-by: Xianxia Xiao <xianxia.xiao@kylin-cloud.com>

src/tools/rbd_recover_tool/epoch_h : fix typos

Signed-off-by: Xianxia Xiao <xianxia.xiao@kylin-cloud.com>

src/librados/IoCtxImpl.h : fix typos

Signed-off-by: Xianxia Xiao <xianxia.xiao@kylin-cloud.com>

src/journal/JournalTrimmer.h : fix typos

Signed-off-by: Xianxia Xiao <xianxia.xiao@kylin-cloud.com>

src/common/hobject.h : fix typos

Signed-off-by: Xianxia Xiao <xianxia.xiao@kylin-cloud.com>

Merge pull request #12652 from badone/wip-coverity-TracepointProvider

tracing: Fix handle leak in TracepointProvider

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

tracing: Fix handle leak in TracepointProvider

Silences Coverity 1397733

Signed-off-by: Brad Hubbard <bhubbard@redhat.com>

test_rados_watch_notify: Fix trivial memory leaks

Signed-off-by: Brad Hubbard <bhubbard@redhat.com>

Merge pull request #12711 from yuyuyu101/wip-fix-rdma

msg/async/rdma: fix ceph_clock_now calls

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12451 from liewegas/wip-osd-on-applied

osd: never send rados ack (only commit)

Reviewed-by: Samuel Just <sjust@redhat.com>

Merge pull request #12480 from dx9/wip-boost-python

cmake: fix hard coded boost python lib

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12406 from songbaisen/s5

tool: add some ceph relate processes to ps-ceph.pl

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12621 from fiskn/master

doc: Update disk thread section to reflect that scrubbing is no longe…

mon/OSDMonitor: set last_force_op_resend on overlay pool too

We currently set the last_force_op_resend field on the
base pool when we set or clear the overlay.  Set it on
the cache/overlay pool too.  The Objecter should resend
even with a change only to the base pool, but the OSD
needs to see the change on the overlay pool to correctly
discard the op.

Fixes: http://tracker.ceph.com/issues/18366
Signed-off-by: Sage Weil <sage@redhat.com>

test: update test_rados_tool.sh, use POOL and OBJ var

* replace hard-code pool name with $POOL
* replace hard-code object name with $OBJ
* introduce a new variable called $POOL_EC
* clean up pool
* simplify test case

Signed-off-by: liuchang0812 <liuchang0812@gmail.com>

msg/async/rdma: fix ceph_clock_now calls

Signed-off-by: Haomai Wang <haomai@xsky.com>

Merge pull request #12403 from liewegas/wip-mon-mgr-cleanups

mon: clean up some osdmon/pgmon interactions

Reviewed-by: Kefu Chai <kchai@redhat.com>

qa/workunits: remove crash_replay_interval tests

Signed-off-by: Sage Weil <sage@redhat.com>

doc: remove crash_replay_interval mentions

Signed-off-by: Sage Weil <sage@redhat.com>

mon/MDSMonitor: no need to set crash_replay_interval on data pools

Signed-off-by: Sage Weil <sage@redhat.com>

osd/PG: remove waiting_for_ack

No longer used.

Signed-off-by: Sage Weil <sage@redhat.com>

osd: simplify dup/resent logic

No need to special ack path.

Signed-off-by: Sage Weil <sage@redhat.com>

osd: remove replay_queue

Signed-off-by: Sage Weil <sage@redhat.com>

osd/PG: do not go into replay state

Signed-off-by: Sage Weil <sage@redhat.com>

osd/PG: do not queue ops for replay

We no longer send ACKs, so this can't happen anyway.

Signed-off-by: Sage Weil <sage@redhat.com>

osd: we can discard "replay" ops

Signed-off-by: Sage Weil <sage@redhat.com>

osd/ReplicatedPG: do not send ACK separate from COMMIT

This never comes before COMMIT, making it useless to clients.  And it's
just unnecessary overhead.

Signed-off-by: Sage Weil <sage@redhat.com>

osd: remove 'rlat' read latency metrics

These are measuring how long of a delay we see before ops are readable.
They are calculated in the on_applied path, which we'd like to remove,
and weren't reliably calculated anyway.

Signed-off-by: Sage Weil <sage@redhat.com>

osd/ReplicatedPG: kill hit_set_flushing

This was added before we started taking the ondisk lock in order to allow
reading a just-written hitset.  Since this is a rare path anyway, just
take the ondisk lock.

Signed-off-by: Sage Weil <sage@redhat.com>

Merge pull request #12522 from hjwsm1989/ecwrite

osd/ECBackend: Send write message to peers first, then do local write

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12524 from yuyuyu101/wip-bluestore-flush

os/bluestore: prevent lock for almost "flush" calls

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12528 from kylinstorage/wip-JournalingObjectStore-cleanup

os/filestore/JournalingObjectStore cleanup

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12670 from liewegas/wip-nits

mon: fix a few nits

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12684 from xiexingguo/xxg-wip-fix-onodesizetracking

os/bluestore: fix OnodeSizeTracking testing

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12707 from tchaikov/wip-dpdk-v16.07

builds with dpdk v16.07

Reviewed-by: Haomai Wang <haomai@xsky.com>

msg/async: bump dpdk to v16.07

as some distros include it already. so sync it up to avoid building
failures on them with dpdk-dev package(s) shipped by distro.

Signed-off-by: Kefu Chai <kchai@redhat.com>

msg/async: be compatible with dpdk v16.07

Signed-off-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12698 from yuyuyu101/wip-dpdk-fix

msg/async/dpdk: fix compile errors

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12699 from wjwithagen/patch-15

qa/test_rados_tool.sh: POSIX dd only accepts 'k' as multiplier

Reviewed-by: Kefu Chai <kchai@redhat.com>

osd/OpRequest: dump both name and addr for the client op

Signed-off-by: runsisi <runsisi@zte.com.cn>

mon/OSDMonitor: remove workaround for dumpling-era bug

We will definitely have published (and stored) a new osdmap since then.

Signed-off-by: Sage Weil <sage@redhat.com>

Merge pull request #12703 from kylinstorage/wip-remove-unused-variable

os/bluestore/BlueStore.cc: remove unused variable

Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>

Merge pull request #12594 from wjwithagen/wip-wjw-freebsd-rebind-timing

common/config_opts.h: FreeBSD timing changed due to no SO_REUSEADDR

Reviewed-by: Sage Weil <sage@redhat.com>

os/bluestore/BlueStore.cc: remove unused variable

Signed-off-by: Li Wang <li.wang@kylin-cloud.com>

qa/test_rados_tool.sh: POSIX dd only accepts 'k' as multiplier

FreeBSD does not want KB as 1k multiplier.
see also: http://www.unix.com/man-page/posix/1posix/dd/

Signed-off-by: Willem Jan Withagen <wjw@digiware.nl>

test_userspace_event: fix missing std

Signed-off-by: Haomai Wang <haomai@xsky.com>

msg/dpdk/TCP: remove unneeded CephContext argument

Signed-off-by: Haomai Wang <haomai@xsky.com>

os/bluestore: _txc_release_alloc when do wal cleaning

Signed-off-by: Xinze Chi <xinze@xsky.com>

Merge pull request #12688 from Adirl/clean_line_end

msg/async/rdma: clean line endings

Reviewed-by: Haomai Wang <haomai@xsky.com>

os/bluestore: fix OnodeSizeTracking testing

[  FAILED  ] 1 test, listed below:
[  FAILED  ] ObjectStore/StoreTest.OnodeSizeTracking/2, where GetParam() = "bluestore"

 1 FAILED TEST

The above test failure happens as the bluestore mount() process
will try to load all collections and put them into the coll_map,
which will be also tracked as mempool::bluestore_meta_other.
So total_bytes from mempool won't be equal to zero.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>

common/ceph_context: Show clear message if all features are enabled

It's saying the following dangerous and experimental featuresare enabled
while the value is "*", it's not quite clear what's "*" stands for.

The patch updates the message so that it's more clear when "*" is used
in place.

Signed-off-by: Dave Chen <wei.d.chen@intel.com>

Merge pull request #12336 from kylinstorage/wip-cleanup-use-string-ref

osd: cleanup: use string & to avoid unnecessary copy

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12569 from liewegas/wip-bluestore-raw-unused

os/bluestore/bluestore_types: drop std::bitset for blob unused

Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>
Reviewed-by: Varada Kari <varada.kari@sandisk.com>

Merge pull request #12612 from fangyuxiangGL/master

osd/ReplicatedBackend: do not set omap header if it is empty

Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Haomai Wang <haomai@xsky.com>

Merge pull request #12668 from hjwsm1989/pg-unfound

osd/PG: add new have_unfound() function in MissingLoc

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #12674 from Liuchang0812/tools-rados-offset-support

tools/rados: add a parameter "--offset" to rados put command

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #12675 from Liuchang0812/fix-test-case-in-rados

test: replace hard-code binary names with varibles

Reviewed-by: Kefu Chai <kchai@redhat.com>

mon/OSDMonitor: ceph osd pool get/set crush_rule by name

This means users don't have to manually translate a rule
they just created to a ruleset ID in order to map a pool
to it.

Signed-off-by: Sage Weil <sage@redhat.com>

mon/PGMap: remove CATEGORY column from 'df detail'

Signed-off-by: Sage Weil <sage@redhat.com>

mon/PGMap: fix alignment of 'cache io' in ceph -s output

Currently it looks like

  client io 3508 B/s wr, 0 op/s rd, 5 op/s wr
  cache io 0 op/s promote

but should be

  client io 3508 B/s wr, 0 op/s rd, 5 op/s wr
   cache io 0 op/s promote

Signed-off-by: Sage Weil <sage@redhat.com>