Merge PR #32777 into master

* refs/pull/32777/head:
common,log,msg: Fix win32 compiler warnings
win32,common: provide a default ceph.conf location
common: add newline before parsing config on Windows
common: verify unix sockets support on Windows
common: fix md_config_t::get_cluster_name
global, common: Print config parse errors
common: accept config files containing whitespaces
librbd: avoid broken mingw rng
common: define DEV_NULL
tools,common,msg: Use O_BINARY for win32 compatibility
log: fix timestamps on Windows
common,librbd: add portable ceph_memzero_s
common,msg: Fix socket handling
common: Extend win32 compatbility layer

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

Merge pull request #37984 from tchaikov/wip-fio

cmake: drop WITH_SYSTEM_FIO option

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

Merge pull request #37965 from rhcs-dashboard/about-modal-cleanup

mgr/dashboard: Moving copyright variable into the app.constants

Reviewed-by: Alfonso Martínez <almartin@redhat.com>
Reviewed-by: Ernesto Puerta <epuertat@redhat.com>
Reviewed-by: Kiefer Chang <kiefer.chang@suse.com>
Reviewed-by: Tatjana Dehler <tdehler@suse.com>

Merge pull request #37958 from ronen-fr/wip-ronenf-librbd-move1

librbd: removing a "pessimizing move"

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

Merge pull request #37990 from tchaikov/wip-liburing-fPIC

cmake: build liburing with -fPIC

Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>

Merge pull request #37915 from gregsfortytwo/wip-stretch-fixes

Fix several stretch mode issues

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

Merge PR #37741 into master

* refs/pull/37741/head:
ceph-volume: remove mention of dmcache from docs and help text

Reviewed-by: Jan Fajerski <jfajerski@suse.com>
Reviewed-by: Rishabh Dave <ridave@redhat.com>

Merge PR #37896 into master

* refs/pull/37896/head:
ceph-volume: implement the --log-level flag

Reviewed-by: Rishabh Dave <ridave@redhat.com>
Reviewed-by: Jan Fajerski <jfajerski@suse.com>

Merge pull request #37506 from rhcs-dashboard/fix-47478-master

mgr/dashboard: Improvements in cluster > OSDs

Reviewed-by: Alfonso Martínez <almartin@redhat.com>
Reviewed-by: Avan Thakkar <athakkar@redhat.com>
Reviewed-by: Nizamudeen A <nia@redhat.com>
Reviewed-by: Volker Theile <vtheile@suse.com>

Merge pull request #37649 from rhcs-dashboard/add-header-tooltip

mgr/dashboard: add description for CRUSH Ruleset

Reviewed-by: Ernesto Puerta <epuertat@redhat.com>
Reviewed-by: Laura Paduano <lpaduano@suse.com>
Reviewed-by: Nizamudeen A <nia@redhat.com>

cmake: build liburing with -fPIC

in liburing,
https://github.com/axboe/liburing/commit/75cad68b953fe0fabbc4b1ced1ab4ba0ed5da18e
partially reverts
https://github.com/axboe/liburing/commit/4e360f71131918c36774f51688e5c65dea8d43f2,
which builds liburing.a with -fPIC.

so we need to pass -fPIC by ourselves. otherwise we'd have

/usr/bin/ld: ../../liburing/src/liburing.a(setup.ol): relocation R_X86_64_PC32 against symbol `io_uring_queue_mmap' can not be used when making a shared object; recompile with -fPIC
/usr/bin/ld: final link failed: Bad value
collect2: error: ld returned 1 exit status
src/test/fio/CMakeFiles/fio_ceph_objectstore.dir/build.make:154: recipe for target 'lib/libfio_ceph_objectstore.so' failed

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

cmake: expose FIO_INCLUDE_DIR via libfio

always use the fio built from source for better consistency.

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

cmake: drop WITH_SYSTEM_FIO option

there is no well-known distro packaging fio.h, so drop this option.

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

Merge pull request #37981 from tchaikov/wip-make-dist

make-dist: add liburing to dist tarball

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

cmake: use make explicitly to build fio

we cannot assume that user uses "make" as the generator of cmake, if,
for instance, ninja is used, `$(MAKE)` is not a valid variable in the
generated `build.ninja`. so we should use "make" explicitly.

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

cmake: use src/liburing for building liburing if it exists

so we can build with liburing enabled if the dist tarball contains
liburing.

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

make-dist: use a loop to collect the ingredients of the dist tarball

for better readability

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

make-dist: add liburing to dist tarball

since liburing is enabled by default, let's included it in the dist
tarball.

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

make-dist: extract download_from() out

so we can reuse it

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

Merge PR #36998 into master

* refs/pull/36998/head:
doc/mgr/volumes: Document authorize/deauthorize cli commands
qa/tasks/cephfs: test `fs subvolume authorize/deauthorize`
mgr/volumes: Allow/deny auth IDs access to FS subvolumes
mon/MonCap: allow 'profile mgr' to create/update/del auth IDs

Reviewed-by: Ramana Raja <rraja@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>

Merge PR #37618 into master

* refs/pull/37618/head:
mds: throttle cap acquisition via readdir

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>

Merge PR #37662 into master

* refs/pull/37662/head:
mon/MDSMonitor: deal with subscription after being expired

Reviewed-by: Jos Collin <jcollin@redhat.com>
Reviewed-by: João Eduardo Luís <joao@suse.de>
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>

Merge pull request #37600 from mgfritch/cephadm-nfs-rgw

mgr/cephadm: add RGW support for NFS ganesha

Reviewed-by: Daniel-Pivonka <dpivonka@redhat.com>
Reviewed-by: Juan Miguel Olmo <jolmomar@redhat.com>
Reviewed-by: Kiefer Chang <kiefer.chang@suse.com>
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>

Merge pull request #37792 from p-se/monitoring-prom-many-to-many

mgr/cephadm: many-to-many Prometheus error

Reviewed-by: Laura Paduano <lpaduano@suse.com>
Reviewed-by: Lenz Grimmer <lgrimmer@suse.com>
Reviewed-by: Michael Fritch <mfritch@suse.com>

Merge pull request #37967 from linuxbox2/wip-rgwfile-zwrite

rgw_file: fix some zipper flow for RGWLibContinuedReq

Merge pull request #37968 from changchengx/paxos

mon/PaxosService: remove unnecessary code

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

Merge pull request #37970 from toabctl/fix-orch-cli-doc-link

mgr/orchestratori: Fix doc link in README.rst

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

mgr/orchestrator_cli: Fix doc link in README.rst

Signed-off-by: Thomas Bechtold <tbechtold@suse.com>

doc: update stretch mode doc to not be a lie

Signed-off-by: Greg Farnum <gfarnum@redhat.com>

mon: include stretch mode status in more outputs

It was missing from the monmap dump, and included in OSDMap::dump()
but not the OSDMap::print() which is used for text-based CLI queries.

Signed-off-by: Greg Farnum <gfarnum@redhat.com>

mon: retain disallowed leader list on restart

We were only setting this when new monmaps were read from paxos -- whoops!
Pull apart that mechanism a little bit and make sure to set them before
doing elections, as part of bootstrap.

Signed-off-by: Greg Farnum <gfarnum@redhat.com>

mon: Output the real leader in ::_quorum_status() and get_leader_name()

These functions previously assumed the first mon in the quorum
was the leader. That isn't accurate if the first monitor is
disallowed or it lost a connectivity-mode election, though.

Signed-off-by: Greg Farnum <gfarnum@redhat.com>

mon: monmap: print out disallowed_leaders in text form, when non-empty

This is already dumped in the programmatic output, but it got left out of
the direct-user-facing calls (as in "ceph mon dump").

Signed-off-by: Greg Farnum <gfarnum@redhat.com>

mon: Do not increase compatv when using monitor location or stretch mode

For mon_info_t, I first wrote things so that when monitors get a location
added in MonMap::mon_info_t, I bumped the struct_v to 5 and also bumped
the min_compat to 5. This made sure that nobody could decode the
struct and lose the location info, which if it were a monitor
would be very bad.

And for the MonMap, when stretch mode is enabled I bumped up the
comptav (in addition to the always-increased struct_v), for the same reason.

But clients also have to decode these structures, and we can't
disallow older clients from connecting to a stretched cluster.

Happily, usage of any stretch modes already requires a feature
bit and sets it as required in the monmap, so these are already
gated. Therefore, just don't set new compat values in these cases.

While at it, also gate setting the location on the monmap indicating
all monitors are updated.

Reported-by: Travis Nielsen <tnielsen@redhat.com>
Signed-off-by: Greg Farnum <gfarnum@redhat.com>

mon: fix erroneous commit of monmap stretch mode when failing to enable

MonmapMonitor::try_enable_stretch_mode() was accidentally writing into the
pending_map even when commit was false. Whoops!

Signed-off-by: Greg Farnum <gfarnum@redhat.com>

Merge pull request #37966 from tchaikov/wip-cmake-uring

cmake: set GIT_SHALLOW and UPDATE_DISCONNECTED

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

cmake: set GIT_SHALLOW and UPDATE_DISCONNECTED

* GIT_SHALLOW=TRUE, so we don't pull the full git history,
  as we don't care about it.
* UPDATE_DISCONNECTED=TRUE, to skip the UPDATE step, this change
  somehow works around
  https://gitlab.kitware.com/cmake/cmake/-/issues/19703. otherwise
  cmake keeps building liburing.

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

blk/kernel/io_uring: bump liburing to v0.7

* use functions exposed by liburing instead of using syscalls
* v0.7 is the latest release at the time of writing, as liburing is under
  active development. it'd be better to use a newer release.
* also use https://git.kernel.dk/liburing instead of
  http://git.kernel.dk/liburing.

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

rgw_file: fix some zipper flow for RGWLibContinuedReq

Some bits of the standard Zipper conversions were missed for
the RGWLibContinuedReq case, where the setup is encapsulated in
the request, but execution is broken up in to steps.   This
currently affects only RGWWriteRequest.

Fixes: https://tracker.ceph.com/issues/48136
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>

librbd: removing a "pessimizing move"

Silencing a compiler warning ("warning: moving a temporary object
prevents copy elision")

Signed-off-by: Ronen Friedman <rfriedma@redhat.com>

mgr/dashboard: Moving copyright into the app.constants

Move the variable copyright from about.component.ts to the app.constants.ts

https://tracker.ceph.com/issues/48135
Signed-off-by: Nizamudeen A <nia@redhat.com>

Merge pull request #37919 from tchaikov/wip-cmake-liburing

cmake: enable LIBURING bluestore backend by default

Reviewed-by: Mark Nelson <mnelson@redhat.com>

Merge pull request #37929 from BryceCao/wip-modify-error-code-of-bucket-info-API

rgw : modify error message to NoSuchBucket when bucket doesn't exist in bucket info API

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

Merge pull request #37913 from rhcs-dashboard/button-disable

Reviewed-by: Alfonso Martínez <almartin@redhat.com>
Reviewed-by: Avan Thakkar <athakkar@redhat.com>
Reviewed-by: Volker Theile <vtheile@suse.com>

Merge pull request #37832 from tchaikov/wip-doc-slow-ops

doc: reformat health-messages.rst and revise SLOW_OPS section

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Zac Dover <zac.dover@gmail.com>
Reviewed-by: Neha Ojha <nojha@redhat.com>

Merge pull request #37951 from agayev/misc-zbd-changes

doc, os/bluestore: Zoned storage related cleanup.

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

Merge pull request #37637 from p-se/dashboard-smart-logging

Reviewed-by: Kiefer Chang <kiefer.chang@suse.com>
Reviewed-by: Volker Theile <vtheile@suse.com>

doc/mgr/volumes: Document authorize/deauthorize cli commands

Fixes: https://tracker.ceph.com/issues/40401
Signed-off-by: Kotresh HR <khiremat@redhat.com>

qa/tasks/cephfs: test `fs subvolume authorize/deauthorize`

Fixes: https://tracker.ceph.com/issues/40401
Signed-off-by: Ramana Raja <rraja@redhat.com>
Signed-off-by: Kotresh HR <khiremat@redhat.com>

mgr/volumes: Allow/deny auth IDs access to FS subvolumes

... via the `ceph fs subvolume authorize/deauthorize` command.

Fixes: https://tracker.ceph.com/issues/40401
Signed-off-by: Ramana Raja <rraja@redhat.com>
Signed-off-by: Kotresh HR <khiremat@redhat.com>

Merge pull request #37454 from votdev/custom_container_docs

doc/mgr/orchestrator: Add hints related to custom containers to the docs

Reviewed-by: Kiefer Chang <kiefer.chang@suse.com>

Merge pull request #37946 from bk201/wip-47373

mgr/dashboard: disable cluster selection in NFS export editing form

Reviewed-by: Laura Paduano <lpaduano@suse.com>
Reviewed-by: Tatjana Dehler <tdehler@suse.com>

mon/MonCap: allow 'profile mgr' to create/update/del auth IDs

Signed-off-by: Ramana Raja <rraja@redhat.com>

Merge pull request #37955 from tchaikov/wip-cmake-crimson-cleanup

cmake: rename crimson tests named like foo_bar to foo-bar

Reviewed-by: Yingxin Cheng <yingxin.cheng@intel.com>

Merge pull request #37953 from tchaikov/wip-crimson-recovery-cleanup

test/crimson: do not support variadic future

Reviewed-by: Yingxin Cheng <yingxin.cheng@intel.com>

cmake: do not name demo-like tests with unittest-*

these tests cannot be exercised without human intervention.

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

cmake/test: remove duplicated target_link_libraries()

it is a copy-n-paste error.

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

cmake: rename crimson tests named like foo_bar to foo-bar

for two reasons:

* less typing: no need to press "shift" for inputting "_"
* more consistent with executable names like "ceph-conf"
* simpler to grep when compiling the tests. there is chance
  we need to kill the dead jobs on a jenkins worker node
  where it happens to be compiling the tests.

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

Merge pull request #37940 from yuvalif/remove_lua_submodule

rpm,deb,cmake: remove the lua submodule

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

Merge PR #37869 into master

* refs/pull/37869/head:
qa: for the latest kclient it will also return EIO

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>

Merge PR #37713 into master

* refs/pull/37713/head:
mds: fix wrong log text and allignment

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>

Merge PR #37664 into master

* refs/pull/37664/head:
cephfs: release client dentry_lease before send caps release to mds

Reviewed-by: Zheng Yan <zyan@redhat.com>

Merge PR #37654 into master

* refs/pull/37654/head:
mds: print the METRIC_TYPE for debug info
mds: add XXPayload:print support
mds: only update the requesting metrics

Reviewed-by: Venky Shankar <vshankar@redhat.com>

Merge PR #32479 into master

* refs/pull/32479/head:
mds: add dentry lease metric payload support
client: add _dentry_valid() helper support

Reviewed-by: Venky Shankar <vshankar@redhat.com>

crimson/common: refactor crimson::do_until()

* define `errorator_t` instead of `futurator` to simplify the code a
  little bit.
* use seastar helper function / types when appropriate

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

test/crimson: do not support variadic future

also add "unittest-seastar-errorator" to unit test.

this should address the build failure introduced by
688b95798bfd88b409abce9b70bb6d2933819bef

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

Merge pull request #37950 from tchaikov/wip-crimson-recovery-cleanup

crimson: cleanups

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

mon/PaxosService: remove unnecessary code

The class declaration has been included in the header file.

Signed-off-by: Changcheng Liu <changcheng.liu@aliyun.com>

mon/MDSMonitor: deal with subscription after being expired

check expiration condition before taking further action

Signed-off-by: Changcheng Liu <changcheng.liu@aliyun.com>

Merge PR #37899 into master

* refs/pull/37899/head:
qa: avoid using ceph.dir.entries for empty check
qa: do not test dir size equals rbytes
qa: disable check for value kclient fails to get
qa: leave one standby available to avoid warning
qa: do not check access timestamps
qa: ignore skip errors for kclient
qa: add more clients for test_volume_client
qa: accept client returning pool id for layout
qa: ignore expected warning about client missing features
qa: do not fail on skip
qa: fix mount directory
qa: rename distro link to distros
qa: accept eio for mount failure
qa: unmount volumes before removal
qa: some sessionmap tests require fuse
qa: remove obsolete test
qa: do not fail on test skip
qa: use more OSDs for workload tests
qa: use 4 clients for volumes testing
qa: relocate tasks
qa: rename workload suite
qa: relocate new tasks from fs:basic_functional
qa: remove kcephfs suite
qa: move kcephfs:mixed-clients to fs
qa: remove dead kcephfs:cephfs suite
qa: merge snap tests for kcephfs/fs
qa: add new fs:shell sub-suite
qa: remove task-less kcephfs:recovery suite
qa: add new fs:volumes sub-suite
qa: merge fs:basic_functional and kcephfs:recovery
qa: move kcephfs:thrash to fs:thrash
qa: reorganize thrash tasks
qa: move multifs/failover tests
qa: run norstats workunit for klient
qa: merge workload tests from kcephfs with fs
qa: use consistent syntax
qa: remove inline testing
qa: move log-rotate config to common yaml
qa: remove fuse default permissions config
qa: fix whitespace

Reviewed-by: Kotresh Hiremath Ravishankar <khiremat@redhat.com>
Reviewed-by: Ramana Raja <rraja@redhat.com>

os/bluestore: Cleanup.  Move ZBD related functions into one place and add zoned
prefix to all of them.

Signed-off-by: Abutalib Aghayev <agayev@psu.edu>

doc: Update contact email address and documentation for zoned-storage.rst.

Signed-off-by: Abutalib Aghayev <agayev@psu.edu>

mgr/dashboard: add description for CRUSH Ruleset

Fixes: https://tracker.ceph.com/issues/48067
Signed-off-by: Avan Thakkar <athakkar@redhat.com>
Adds a description for CRUSH Ruleset column in pools table in the pool creation dialogue

Merge pull request #32394 from zhangsw/fix-appendobj-invalidposition

rgw: using strict_strtoll for append position. If not, any invalid value will be decoded as zero which is not correct.

Reviewed-by: Yuval Lifshitz <ylifshit@redhat.com>

Merge pull request #37703 from theanalyst/rgw/sal/init-quota-fixes

rgw: op: check for bucket existance as well in init_quota

Reviewed-by: Daniel Gryniewicz <dang@redhat.com>

Merge pull request #37160 from linuxbox2/wip-rgwlc-tagfilter

rgwlc: fix clause counting in LCFilter_S3::decode_xml()

Reviewed-by: Casey Bodley <cbodley@redhat.com>
Reviewed-By: Abhishek Lekshmanan <abhishek@suse.com>

rpm/deb/cmake: remove the lua submodule

instead use the official lua-dev/el packages

Signed-off-by: Yuval Lifshitz <ylifshit@redhat.com>

Merge pull request #37066 from smanjara/wip-metadata-reshard

rgw: don't allow overwrites to BucketInfo's layout field when sync fetches bucket metadata

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

Merge pull request #37892 from IlsooByun/fix_bucket_sync_err

rgw: keep syncstopped flag when copying bucket shard headers

Reviewed-by: Casey Bodley <cbodley@redhat.com>
Reviewed-by: J. Eric Ivancich <ivancich@redhat.com>

osd/PGLog: do not use unique_ptr explicitly

* do not use unique_ptr<> explicitly, use `seastar::do_with()` for
  better readability
* use seastar::do_until() instead of seastar::repeat() for better
  readability. plain boolean is simpler than
  `seastar::stop_iteration::yes`
* do not capture variables using FuturizedStoreLogReader if we could
  pass them by instant parameters.
* rename "start()" to "read()". as "read" is more specific in this
  context.

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

osd/PGLog: move PGLog::read_log_and_missing_crimson() into .cc

to speed up the compilation. and for better readability.

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

osd/PGLog: drop template parameter of FuturizedStoreLogReader

there is no need for the generic type at this moment.

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

crimson/osd: use seastar:do_until() when appropriate

a plain boolean is easier to understand than
`seastar::stop_iteration::yes`.

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

crimson/osd: mark CyanStore::CyanOmapIterator methods final

this class is not intended to be inherited by classes, and these methods
are "final", so mark them "final".

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

crimson/osd: mark methods pure virtual

if the default implementation does not make sense in any case, just mark
it pure virtual.

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

crimson/osd: let OmapIterator mutators return future<>

instead of returning future<int> where the int represents an errno which
is not checked by any callers, we should just return a future<>, we
could erroratorize it later on, but returning an int is not the right
way to handle this, and it is not consistent with how we handle other
errors elsewhere in crimson.

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

doc/mgr/orchestrator: Add hints related to custom containers to the docs

Fixes: https://tracker.ceph.com/issues/48113
Signed-off-by: Volker Theile <vtheile@suse.com>

Merge pull request #37867 from zdover23/wip-doc-cephadm-install-oct-2020

doc/cephadm: improve install guide

Reviewed-by: Neha Ojha <nojha@redhat.com>

Merge pull request #37944 from tchaikov/wip-crimson-recovery-cleanup

crimson/osd: extract read_omap_for_push_op()

Reviewed-by: Xuehan Xu <xxhdx1985126@gmail.com>
Reviewed-by: Samuel Just <sjust@redhat.com>

mgr/dashboard: disable cluster selection in NFS export editing form

We should not allow changing an export's cluster because an export ID
might live in one cluster but not in another one. Editing a non-existing
export in a cluster causes an error.

Fixes: https://tracker.ceph.com/issues/47373
Signed-off-by: Kiefer Chang <kiefer.chang@suse.com>

mds: add dentry lease metric payload support

Fixes: https://tracker.ceph.com/issues/43423
Signed-off-by: Xiubo Li <xiubli@redhat.com>

client: add _dentry_valid() helper support

Fixes: https://tracker.ceph.com/issues/43423
Signed-off-by: Xiubo Li <xiubli@redhat.com>

crimson/osd: extract read_omap_for_push_op()

extract read_omap_for_push_op() out of build_push_op() for better
readability

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

crimson/osd: correct the function names in logging messages

it was a leftover of fbfc53c7081ea0b99fbc519d7fac9f5637cf5425

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

Merge PR #37881 into master

* refs/pull/37881/head:
doc/mgr/orchestrator: Update about nfs add in rook
mgr/rook: Refactor nfs add/apply methods
mgr/rook: Udpate to mypy py3 type style in _create(), _update() and apply()
mgr/rook: Remove unused 'current' argument from update_func method

Reviewed-by: Michael Fritch <mfritch@suse.com>
Reviewed-by: Travis Nielsen <tnielsen@redhat.com>

qa: avoid using ceph.dir.entries for empty check

This avoids a bug [1] in the kernel client.

[1] https://tracker.ceph.com/issues/48104

Fixes: https://tracker.ceph.com/issues/23718
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>

Merge PR #37825 into master

* refs/pull/37825/head:
doc: document MDS cache configuration
test: add DecayCounter test

Reviewed-by: Sidharth Anupkrishnan <sanupkri@redhat.com>

qa: do not test dir size equals rbytes

Avoiding this failure:

2020-11-03T03:11:59.980 INFO:tasks.cephfs_test_runner:test_subvolume_resize_fail_quota_lt_used_size_no_shrink (tasks.cephfs.test_volumes.TestSubvolumes) ... FAIL
2020-11-03T03:11:59.981 INFO:tasks.cephfs_test_runner:
2020-11-03T03:11:59.981 INFO:tasks.cephfs_test_runner:======================================================================
2020-11-03T03:11:59.981 INFO:tasks.cephfs_test_runner:FAIL: test_subvolume_resize_fail_quota_lt_used_size_no_shrink (tasks.cephfs.test_volumes.TestSubvolumes)
2020-11-03T03:11:59.982 INFO:tasks.cephfs_test_runner:----------------------------------------------------------------------
2020-11-03T03:11:59.982 INFO:tasks.cephfs_test_runner:Traceback (most recent call last):
2020-11-03T03:11:59.982 INFO:tasks.cephfs_test_runner:  File "/home/teuthworker/src/git.ceph.com_ceph-c_wip-pdonnell-testing-20201102.231317/qa/tasks/cephfs/test_volumes.py", line 1310, in test_subvolume_resize_fail_quota_lt_used_size_no_shrink
2020-11-03T03:11:59.982 INFO:tasks.cephfs_test_runner:    self.assertEqual(usedsize, susedsize)
2020-11-03T03:11:59.982 INFO:tasks.cephfs_test_runner:AssertionError: 8912896 != 1

From: /ceph/teuthology-archive/pdonnell-2020-11-03_02:22:55-fs:volumes-wip-pdonnell-testing-20201102.231317-distro-basic-smithi/5585383/teuthology.log

Fixes: https://tracker.ceph.com/issues/23718
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>

qa: disable check for value kclient fails to get

See-also: https://tracker.ceph.com/issues/48075
Fixes: https://tracker.ceph.com/issues/23718
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>

qa: leave one standby available to avoid warning

Four file systems will use all MDS and generate this warning:

2020-11-02T03:48:33.407 INFO:teuthology.orchestra.run.smithi003.stdout:2020-11-02T03:24:21.817337+0000 mon.a (mon.0) 481 : cluster [WRN] Health check failed: insufficient standby MDS daemons available (MDS_INSUFFICIENT_STANDBY)

Fixes: https://tracker.ceph.com/issues/23718
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>