mgr/cephadm: add a test for enabling cephfs mirroring module

Add a test that checks that when cephfs mirror service is enabled
the mirroring mgr module gets enabled.

Actually-written-by: Sebastian Wagner <sewagner@redhat.com>
Signed-off-by: John Mulligan <jmulligan@redhat.com>

mgr/cephadm: auto-enable mirroring module when deploying service

Automatically enable the mgr's mirroring module when creating
cephfs-mirror services. This will trigger a mgr respawn.

Fixes: https://tracker.ceph.com/issues/50593
Based roughly on 50dc1d0decb2fbf7b9129bddba940969410be5cd

Signed-off-by: John Mulligan <jmulligan@redhat.com>

Merge pull request #33934 from cbodley/wip-40177

rgw: delete full sync index before switching to incremental sync

Signed-off-by: Shilpa Jagannath <smanjara@redhat.com>

Merge pull request #40011 from cbodley/wip-49723

rgw: allow rgw_data_notify_interval_msec=0 to disable notifications

Reviewed-by: Shilpa Jagannath <smanjara@redhat.com>
Reviewed-by: Adam C. Emerson <aemerson@redhat.com>

Merge pull request #44581 from cbodley/wip-53177

rgw/dbstore: hide dbstore_log.h from rgw_main.cc

Reviewed-by: Soumya Koduri <skoduri@redhat.com>

Merge pull request #44413 from cybozu/rgw-fix-error-code-of-remove-bucket-api

rgw: remove bucket API returns NoSuchKey than NoSuchBucket

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

Merge pull request #44078 from cbodley/wip-rgw-multisite-metadata-retry-error

rgw/multisite: metadata sync only retries on errors

Reviewed-by: Adam C. Emerson <aemerson@redhat.com>

Merge pull request #44602 from neha-ojha/wip-qct-remove

doc/foundation.rst: qct is no longer a member

Reviewed-by: Dan van der Ster <daniel.vanderster@cern.ch>

Merge pull request #44505 from guits/fix-53812

ceph-volume: fix regression introcuded via #43536

Reviewed-by: Adam King <adking@redhat.com>
Reviewed-by: Teoman Onay <tonay@redhat.com>

Merge pull request #44489 from adk3798/agent-down-alerts

mgr/cephadm: still check agent deps if it is marked down

qa/cephadm: install hwe kernel only for focal

Let's install hwe kernel only on Ubuntu focal, otherwise we only shift the
problem on Ubuntu bionic given that the hwe kernel for bionic is 5.4.

Fixes: https://tracker.ceph.com/issues/53863
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>

Merge pull request #44591 from athanatos/sjust/wip-seastore-flush

crimson/os/seastore: avoid empty Transactions by adding explicit flush() call

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

Merge pull request #44556 from cyx1231st/wip-crimson-improve-log-journal

crimson/os/seastore: consolidate seastore_journal logs with cleanup and validations

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

rgw: clean up index after full metadata sync

Fixes: https://tracker.ceph.com/issues/40177
Signed-off-by: Casey Bodley <cbodley@redhat.com>

rgw: clean up index after full data sync

Fixes: https://tracker.ceph.com/issues/40177
Signed-off-by: Casey Bodley <cbodley@redhat.com>

crimson/os/seastore: implement FuturizedStore::flush

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

qa/rgw: run multisite tests with some async notifications disabled

disable the sending of async datalog notifications on one zone per
cluster. this helps to verify that tests don't rely on notifications to
succeed

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

rgw: allow rgw_data_notify_interval_msec=0 to disable notifications

the data changes log for multisite will occasionally broadcast recent
changes to other zones, which they can use to prioritize sync of some
of the most recent changes. they'll eventually see all changes as they
replay the data changes log, though, so notifications aren't required
for successful sync. the ability to turn them off is useful for testing

Fixes: https://tracker.ceph.com/issues/49723
Signed-off-by: Casey Bodley <cbodley@redhat.com>

qa/nvme_loop: fix an issue on ubuntu 18.04

The following command:

```
echo /dev/sda | tee /sys/kernel/config/nvmet/subsystems/sda/namespaces/1/device_path
```

makes nvme_loop fail because fascinatingly, it adds an unexpected newline.

See:
```
/dev/sda
/dev/sda

1
tee: /sys/kernel/config/nvmet/subsystems/sda/namespaces/1/enable: No such file or directory
/dev/sda
1
```

Other distros don't have the same behavior:

```
CentOS 8
/dev/sda
/dev/sda
1

Ubuntu 20.04
/dev/sda
/dev/sda
1
```

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>

ceph-volume: fix regression introcuded via #43536

The recent changes from PR #43536 introduced a regeression preventing from
running ceph-volume in a containerized context on Ubuntu 18.04.

Given that the path for the binary `lvs` differs between CentOS 8 and Ubuntu 18.04.
(`/usr/sbin/lvs` and `/sbin/lvs` respictively). It means that ceph-volume running
in the container on CentOS 8 sees the `lvs` binary at `/usr/sbin/lvs` and try to
run it with `nsenter` on the host which is running Ubuntu 18.04.

Fixes: https://tracker.ceph.com/issues/53812
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 95e88cda3df76b59b548ae808df0ef7f19db1f63)
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>

Merge pull request #44566 from falcon78921/minor-messaging-nit

mgr/cephadm: fix minor grammar nit in Dry-Runs message

Merge pull request #44510 from rzarzynski/wip-cephadm-docfix

doc/cephadm: improve the development doc a bit

Reviewed-by: Sebastian Wagner <sewagner@redhat.com>

Merge pull request #44485 from adk3798/agent-permissions

cephadm: fix permissions on agent files

Reviewed-by: Sebastian Wagner <sewagner@redhat.com>

Merge pull request #44506 from sebastian-philipp/orch-suite-add-scsi

qa/suites/orch/cephadm: Also run the rbd/iscsi suite

Reviewed-by: Adam King <adking@redhat.com>
Reviewed-by: Deepika Upadhyay <dupadhya@redhat.com>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
Reviewed-by: Melissa Li <mingkli@redhat.com>

Merge pull request #44603 from cbodley/wip-cmake-parquet

rgw: disable parquet by default

Reviewed-by: Matt Benjamin <mbenjamin@redhat.com>

build: revert arrow package dependency

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

cmake: disable parquet by default

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

doc/foundation.rst: qct is no longer a member

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

Merge pull request #44523 from ljflores/wip-telemetry-dashboard

mgr/dashboard/telemetry: reduce telemetry dashboard preview size

Reviewed-by: Ernesto Puerta <epuertat@redhat.com>
Reviewed-by: Laura Flores <lflores@redhat.com>
Reviewed-by: neha-ojha <NOT@FOUND>
Reviewed-by: Pere Diaz Bou <pdiazbou@redhat.com>
Reviewed-by: Yaarit Hatuka <yaarithatuka@gmail.com>

Merge pull request #44550 from jdurgin/wip-pool-get-quota

mon/OSDMonitor: avoid null dereference if stats are not available

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

Merge pull request #42735 from amathuria/wip-amathuria-scrub-stats

osd/scrub: Add stats to PG dump for number of objects scrubbed

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

Merge pull request #43667 from ifed01/wip-ifed-fix-ram-gridy-fsck

os/bluestore: make shared blob fsck much less RAM-greedy.

Reviewed-by: Adam Kupczyk <akupczyk@redhat.com>

Merge pull request #44440 from soumyakoduri/wip-skoduri-dbstore-fixes

rgw/dbstore: Misc fixes

Merge pull request #44552 from jdurgin/wip-releases-doc

doc/releases: remove outdated info and versions; mark nautilus eol

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

rgw/dbstore: hide dbstore_log.h from rgw_main.cc

dbstore_log.h sets global dout_subsys/dout_prefix macros, and was
leaking into rgw_main.cc through the common/dbstore.h. this caused all
of rgw_main's log output to start with the wrong prefix "rgw dbstore: "

Fixes: https://tracker.ceph.com/issues/53177
Signed-off-by: Casey Bodley <cbodley@redhat.com>

Merge pull request #44370 from benhanokh/NCB_expand_device_fix

NCB code doesn't update allocation file when we expand-device

Reviewed-by: Adam Kupczyk <akupczyk@redhat.com>

Merge pull request #44251 from yaarith/telemetry-opt-in

mgr/telemetry: introduce new design for varying report data

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

Merge pull request #43849 from rzarzynski/wip-bs-lucky-buffers

blk, os/bluestore: introduce huge page-based read buffers

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

Merge pull request #42576 from AmnonHanuhov/wip-port_rgw_classes

crimson/osd: Port rgw object classes to run in crimson

Reviewed-by: Kefu Chai <tchaikov@gmail.com>
Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>

Merge pull request #44518 from gregsfortytwo/wip-fix-53824

osd: PeeringState: fix selection order in calc_replicated_acting_stretch

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

crimson/os/seastore: consolidate seastore_journal logs with structured level and format

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

crimson/os/seastore/journal: validate segments before replay

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

crimson/os/seastore/seastore_types: pretty print data structures

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

crimson/os/seastore: count consumed records in cursor with cleanups

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

crimson/os/seastore: drop duplicated record_group_t::current_dlength

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

crimson/os/seastore: classify journal related logs in seastore_types.cc

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

crimson/os/seastore: convert ExtentReader to seastore logging

Also set the logger to seastore_journal as the component works at
the journal layer.

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

mgr/dashboard/telemetry: add test for formatReport()

Tests a scenario where all keys are removed, and one
where a key is ignored.

Signed-off-by: Laura Flores <lflores@redhat.com>

crimson/os/seastore/journal: convert to seastore logging

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

doc: document new OBJECTS_SCRUBBED column in pg dump

Signed-off-by: Aishwarya Mathuria <amathuri@redhat.com>

crimson/osd: Implement missing objclass functions used by cls_rgw

Signed-off-by: Amnon Hanuhov <ahanukov@redhat.com>

crimson/osd: Add support for CEPH_OSD_OP_OMAPRMKEYS

Signed-off-by: Amnon Hanuhov <ahanukov@redhat.com>

crimson/osd: Add a getter for last_user_version

last_user_version is the last user object version applied to store

Signed-off-by: Amnon Hanuhov <ahanukov@redhat.com>

crimson/osd: drop PGBackend& from OpsExecuter ctor

OpsExecuter holds a Ref<PG> so the PGBackend can be extracted from it
using get_backend()

Signed-off-by: Amnon Hanuhov <ahanukov@redhat.com>

crimson/osd: drop pg_pool_t from OpsExecuter ctor

OpsExecuter now holds a Ref<PG> so the pool info can be extracted from it
using get_pool().info

Signed-off-by: Amnon Hanuhov <ahanukov@redhat.com>

crimson/osd: Store a reference to PG inside OpsExecuter

This is needed as some ObjClass methods make use of pg information related to the given cls_method_context_t

Signed-off-by: Amnon Hanuhov <ahanukov@redhat.com>

Merge pull request #44507 from votdev/issue_53813_nfs_page_not_found

mgr/dashboard: NFS pages shows 'Page not found'

Reviewed-by: Alfonso Martínez <almartin@redhat.com>
Reviewed-by: Laura Paduano <lpaduano@suse.com>
Reviewed-by: Nizamudeen A <nia@redhat.com>
Reviewed-by: Tatjana Dehler <tdehler@suse.com>
Reviewed-by: Volker Theile <vtheile@suse.com>

Merge pull request #43685 from p-se/fix-grafana-graphs-ceph_daemon

mgr/dashboard: fix Grafana OSD/host panels

Reviewed-by: Aashish Sharma <aasharma@redhat.com>
Reviewed-by: Alfonso Martínez <almartin@redhat.com>
Reviewed-by: Avan Thakkar <athakkar@redhat.com>
Reviewed-by: Laura Paduano <lpaduano@suse.com>
Reviewed-by: Ernesto Puerta <epuertat@redhat.com>
Reviewed-by: p-se <NOT@FOUND>
Reviewed-by: Pere Diaz Bou <pdiazbou@redhat.com>

Merge pull request #44573 from rhcs-dashboard/53858-fix-smart-data-single-daemon

mgr/dashboard: fix: get SMART data from single-daemon device

Reviewed-by: Alfonso Martínez <almartin@redhat.com>
Reviewed-by: Avan Thakkar <athakkar@redhat.com>
Reviewed-by: Nizamudeen A <nia@redhat.com>
Reviewed-by: Pere Diaz Bou <pdiazbou@redhat.com>

Merge pull request #44559 from ideepika/wip-iscsi-53830

test/rbd/iscsi: correct the hostname in gwcli_create.t to match hostname -f

Reviewed-by: Ilya Dryomov <idryomov@gmail.com>

Merge pull request #44571 from idryomov/wip-xfstests-qemu-cert

qa/run_xfstests_qemu.sh: stop reporting success without actually running any tests

Reviewed-by: Deepika Upadhyay <dupadhya@redhat.com>

crimson: add and use FuturizedStore::flush() interface

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

Merge pull request #44570 from vshankar/wip-53857

qa: adjust for MDSs to get deployed before verifying their availability

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

Merge pull request #44555 from cyx1231st/wip-fix-seastore-jounral-fast-submit

crimson/os/seastore/journal: fast submit if RecordSubmitter is IDLE and no pending

Reviewed-by: Chunmei Liu <chunmei.liu@intel.com>
Reviewed-by: Samuel Just <sjust@redhat.com>

doc/releases: remove dev and pre-nautilus releases from timeline

Improve readability of the table - all this information is still
preserved in older branches.

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

Merge pull request #44583 from mgfritch/fixup-44306-docker-count

cephadm: increase number of docker.io occurances

Reviewed-by: Adam King <adking@redhat.com>

cephadm: increase number of docker.io occurances

fixup for 0fe2e54db774271e4fc18b45aba36b66cbc71779

Signed-off-by: Michael Fritch <mfritch@suse.com>

mgr/telemetry: revise format_perf_histogram

osd_perf_histograms now include only separated stats; remove the
aggregated formatting; we can revert this in case we ever add aggregated
histograms.

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

PendingReleaseNotes: add a note about telemetry

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

mgr/telemetry: add `enable / disable channel all`

Enable or disable all telemetry channels at once with:
    ceph telemetry enable channel all
    ceph telemetry disable channel all

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

mgr/telemetry: do not restore channels default when opting-out

Other modules do not reset their configuration; keep telemetry module
consistent with this behavior.

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

mgr/telemetry: verify there are new collections when nagging due to a major
upgrade

When adding a new collection we define whether to nag the user about it.
We may add many collections and nag about none of them. However, in case
of a major upgrade, we wish to notify the user about these new
collections. This commit verifies there are indeed new collections when
nagging due to a major upgrade.

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

mgr/telemetry: improve output of `ceph telemetry collection ls`

STATUS column now indicates whether a collection is being reported, and
the reasons why it's not (either the user is not opted-in to this
collection, or its channel is off).

Also, removed the ENROLLED and DEFAULT columns due to potential
confusion they may cause.

In case a user is not opted-in to certain collections, a message will
appear above the table with the missing collections:

    New collections are available:
    ['basic_base', 'basic_mds_metadata', 'crash_base', 'device_base',
    'ident_base', 'perf_perf']
    Run `ceph telemetry on` to opt-in to these collections.

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

mgr/telemetry: use dict lookup when traversing MODULE_COLLECTION

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

mgr/telemetry: add test coverage for telemetry upgrade

Test the behavior of the module after an upgrade, as we shift from our
revision design to Collections.

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

doc/mgr/telemetry: document new commands

New commands:

  ceph telemetry enable channel <channel_name>
  ceph telemetry disable channel <channel_name>
  ceph telemetry channel ls
  ceph telemetry collection ls
  ceph telemetry collection diff
  ceph telemetry preview
  ceph telemetry preview-device
  ceph telemetry preview-all

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

mgr/telemetry: add command to list all collections

List all collections, their current enrollment state, status, default,
and description, with:

$ ceph telemetry collection ls

NAME                  ENROLLED    STATUS    DEFAULT    DESC
basic_base            TRUE        ON        ON         Basic information about the cluster (capacity, number and type of daemons, version, etc.)
basic_mds_metadata    TRUE        ON        ON         MDS metadata
crash_base            TRUE        ON        ON         Information about daemon crashes (daemon type and version, backtrace, etc.)
device_base           TRUE        ON        ON         Information about device health metrics
ident_base            TRUE        OFF       OFF        User-provided identifying information about the cluster
perf_perf             TRUE        OFF       OFF        Information about performance counters of the cluster

Please note:

NAME:
=====
Collection name; prefix indicates the channel the collection belongs to.

ENROLLED:
=========
Signifies the collections that were available in the module when the
user last opted-in to telemetry. Please note: Even if a collection is
'enrolled', its metrics will be reported only if its channel is enabled.

STATUS:
=======
Indicates whether the collection metrics are reported; this is
determined by the status (enabled / disabled) of the channel the
collection belongs to, along with the enrollment status of the
collection.

DEFAULT:
========
The default status (enabled / disabled) of the channel the collection
belongs to.

DESC:
=====
Collection description.

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

mgr/telemetry: fix missing type annotations

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

mgr/telemetry: add preview-device and preview-all commands

`ceph telemetry show` will show a sample cluster report if the user is
opted-in to telemetry. The report will be compiled of the collections
the user is opted-in to. To preview a report compiled of the most recent
collection available, use `ceph telemetry preview`.

The device channel is not included in the cluster report, since it's
being sent to a different endpoint, thus we use
`ceph telemetry show-device` in case the user is opted-in to telemetry
and the device channel is enabled. If not, it can also be previewed with
`ceph telemetry preview-device`.

If telemetry is on, and device channel is enabled, both reports can be
reviewed with `ceph telemetry show-all`, otherwise use
`ceph telemetry preview-all`.

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

mgr/telemetry: add command to list all channels

List all channels, their current state, default, and description, with:

$ ceph telemetry channel ls

NAME      ENABLED    DEFAULT    DESC
basic     ON         ON         Share basic cluster information (size, version)
ident     OFF        OFF        Share a user-provided description and/or contact email for the cluster
crash     ON         ON         Share metadata about Ceph daemon crashes (version, stack straces, etc)
device    ON         ON         Share device health metrics (e.g., SMART data, minus potentially identifying info like serial numbers)
perf      ON         OFF        Share perf counter metrics summed across the whole cluster

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

mgr/telemetry: add commands to enable/disable channels

Currently we enable/disable a telemetry channel via CLI with:
  `ceph config set mgr mgr/telemetry/channel_basic true`
  `ceph config set mgr mgr/telemetry/channel_crash false`

We can now do this with:
  `ceph telemetry enable channel basic`
  `ceph telemetry disable channel crash`

We allow enabling / disabling lists of channels:
  `ceph telemetry enable channel basic device crash perf`
  `ceph telemetry disable channel basic device crash perf`

Please note, telemetry should be on for these commands to take effect.

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

mgr/telemetry: introduce new design for adding new data

The current design requires increasing the telemetry revision each time
we add new data to the report. As a result, users need to re-opt-in to
telemetry. This new design allows for adding new data to the report,
while allowing users to keep sending only what they already opted-in to,
hence no re-opt-in is required. In case users wish to report the new
data as well, they need to re-opt-in and enable any new channels.

Also, move formatting perf histograms to a function, so we can use it
both in `show` and `preview` commands.

Fix get_report call in dashboard to use get_report_locked.

Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>

Merge pull request #44554 from jdurgin/wip-rbd-qos-docs

doc/rbd: clarify and add more detail to librbd QoS docs

Reviewed-by: Ilya Dryomov <idryomov@gmail.com>

Merge pull request #40802 from galsalomon66/wip-s3select-parquet-object-processing-2

RGW/s3select : parquet implementation:

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

test/rbd/iscsi: correct the HOST name provided.

hostname -f and hostname generated from gwcli_create being different
gave rise to error:

The first gateway defined must be the local machine

Fixes: https://tracker.ceph.com/issues/53830
Signed-off-by: Deepika Upadhyay <dupadhya@redhat.com>

Merge pull request #44577 from clementperon/master

cmake: Fix Finddpdk cmake module

Reviewed-by: Kefu Chai <tchaikov@gmail.com>

Merge pull request #44498 from phlogistonjohn/jjm-root-check-later

cephadm: check if cephadm is root after cli is parsed

Reviewed-by: Adam King <adking@redhat.com>

Merge pull request #44394 from melissa-kun-li/enable-autotune

Enable autotune for osd_memory_target on bootstrap

Reviewed-by: Alfonso Martínez <almartin@redhat.com>

Merge pull request #44306 from sebastian-philipp/normalize_image_digest-ambiguity

cephadm: deal with ambiguity within normalize_image_digest

Reviewed-by: Adam King <adking@redhat.com>
Reviewed-by: Sage Weil <sage@newdream.net>

doc/rbd/rbd-config-ref: add more detail on QoS settings

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

rgw/multisite: metadata sync only retries on errors

in 866d66b8749b28ec626a8d0adba3d14fdd8abead, metadata sync was fixed to
retry on error codes other than EAGAIN/ECANCELED. but this change caused
us to retry on success as well, which means we send 10 GET requests for
each piece of metadata, and write it to rados 10 times

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

handling arm64(arrow installation)

Signed-off-by: gal salomon <gal.salomon@gmail.com>

Merge pull request #44427 from lxbsz/client_cleanup

client: remove useless Lx cap check

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

Merge pull request #44451 from lxbsz/wip-53750

mds: directly return just after responding the link request

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

Merge pull request #44561 from cbodley/wip-51727

qa/rgw: add PG_DEGRADED cluster warnings to log-ignorelist

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

mgr/dashboard: fix: get SMART data from single-daemon device

Return SMART data even when a device is only associated with a single daemon.

Fixes: https://tracker.ceph.com/issues/53858
Signed-off-by: Alfonso Martínez <almartin@redhat.com>

Merge pull request #44538 from dang/wip-dang-zipper-perf

RGW Zipper - don't load stats for every bucket load

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

Merge pull request #44002 from JoshSalomon/wip-primary-balancer

cmake: dpdk: only append common dir if it has been found

Signed-off-by: Clément Péron <peron.clem@gmail.com>

cmake: dpdk: use STREQUAL and not EQUAL when comparing strings

Signed-off-by: Clément Péron <peron.clem@gmail.com>