Merge branch 'reef' into wip-doc-2025-02-08-backport-61660-to-reef

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>

Merge pull request #61731 from zdover23/wip-doc-2025-02-08-backport-61708-to-reef

reef: doc: Clarify that there are no tertiary OSDs

doc: Clarify that there are no tertiary OSDs

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit e261155a8a0234897803d38e899d10ec89b2b08f)

doc/rados/operations: Improve pools.rst

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit b1e5da594a662e0037459489fbbd449685ad9854)

Merge pull request #61602 from idryomov/wip-69679-reef

reef: mon/OSDMonitor: relax cap enforcement for unmanaged snapshots

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

Merge pull request #61367 from cbodley/wip-67268-reef

reef: rgw/rgw_rados: fix server side-copy orphans tail-objects

Merge pull request #61329 from adamemerson/wip-66340-reef

reef: rgw/multisite: Fix use-after-move in retry logic in logbacking

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

Merge pull request #61630 from idryomov/zlib-fix-reef

reef: win32_deps_build.sh: pin zlib tag

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

Merge pull request #59923 from mchangir/wip-68076-reef

reef: qa: relocate subvol creation overrides and test

Merge pull request #60689 from vshankar/wip-68110-reef

reef: mds: batch backtrace updates by pool-id when expiring a log segment

Merge pull request #60390 from rishabh-d-dave/wip-68616-reef

reef: qa/cephfs: ignore when specific OSD is reported down during upgrade

Merge pull request #60373 from anoopcs9/wip-68574-reef

reef: client: Fix opening and reading of symlinks

Merge pull request #60207 from kotreshhr/wip-68455-reef

reef: mds: misc fixes for MDSAuthCaps code

Merge pull request #60188 from kotreshhr/wip-68413-reef

reef: mgr/status: Fix 'fs status' json output

Merge pull request #59898 from joscollin/wip-68167-reef

reef: cephfs-top: fix exceptions on small/large sized windows

Merge pull request #59618 from batrick/wip-67837-reef

reef: qa: ignore warnings variations

qa: ignore warnings variations

Fixes: https://tracker.ceph.com/issues/67601
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit 782c88aa96cd4535e3eccd8bd9c25cc5d1dbe8c5)

Conflicts:
qa/cephfs/overrides/ignorelist_health.yaml: trivial

win32_deps_build.sh: pin zlib tag

The zlib Windows build started to fail, probably because of this:
https://github.com/madler/zlib/issues/1038

  Cloning into 'zlib'...
  make: *** No rule to make target 'zconf.h', needed by 'adler32.o'.

We'll pin the zlib version for now to unblock the Windows build.

Signed-off-by: Lucian Petrut <lpetrut@cloudbasesolutions.com>
(cherry picked from commit ba9270dc9fe923bb9639a29328035e78ad4346c7)

Merge pull request #61612 from adk3798/reef-ctr-label-ceph

reef: container: add label ceph=True back

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

qa: relocate subvol creation overrides and test

Fixes: https://tracker.ceph.com/issues/65829
Signed-off-by: Milind Changire <mchangir@redhat.com>
(cherry picked from commit ef68253a87b764abedee71046a0e301d408ad11e)

Merge pull request #61481 from NitzanMordhai/wip-69622-reef

reef: test: ceph daemon command with asok path

Merge pull request #61395 from NitzanMordhai/wip-nitzan-set-allow-ec-overwrites-wait

reef: (reintroduce) test/librados: fix LibRadosIoECPP.CrcZeroWrite

container: add label ceph=True back

Add a label used by cephadm internally that was always set by
ceph-container [1] back to the new containerfile. This should
prevent issues with cephadm shell command thinking official ceph images
are not official ceph images.

[1] https://github.com/ceph/ceph-container/blob/30dc8b9a55f70a40983fd2da6ac31e1b9e977143/src/__DOCKERFILE_TRACEABILITY_LABELS__#L5

Signed-off-by: John Mulligan <jmulligan@redhat.com>
(cherry picked from commit b4f4323e6ccacbff8899ef9b5c336b238101fe0e)

Merge pull request #61526 from idryomov/wip-69619-reef

reef: librbd: clear ctx before initiating close in Image::{aio_,}close()

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

Merge pull request #61169 from idryomov/wip-69319-reef

reef: rbd: open images in read-only mode for "rbd mirror pool status --verbose"

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

Merge pull request #61604 from idryomov/wip-rbd-migration-https-doc-reef

reef: doc/rbd: use https links in live import examples

Reviewed-by: Zac Dover <zac.dover@proton.me>

doc/rbd: use https links in live import examples

Even though it's explicitly said that "http" stream can be used to
import via both HTTP and HTTPS, it can still be confusing that "type":
"http" is expected to go with "url": "https://...".  Switch example
URLs from HTTP to HTTPS to make it more obvious.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 8666c5694eb61deb2b46b6b76274e58f637c068a)

osd/OSDCap: fix misleading grammar comments

The restrictions on pool name and namespace have been independent of
each other for ages.  Specifying namespace[=]<namespace> doesn't require
specifying pool[=]<pool> like is currently suggested -- neither for
regular "allow" grants nor for "profile" grants.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit d99367e1e03cbeefd33d7dd9405245e089c2237c)

mon/OSDMonitor: relax cap enforcement for unmanaged snapshots

Since commit 4972e054b32c ("mon/OSDMonitor: enforce caps when
creating/deleting unmanaged snapshots"), a) write access to the MON
service, b) write access to the OSD service for a pool or c) permission
for "osd pool op unmanaged-snap" command for a pool is required.  For
"profile rbd" we configure read-only access to the MON service and rely
on write access to the OSD service, however the corresponding check in
is_osd_writable() is too strict.

A OSD cap like "profile rbd namespace=myns" or "allow w namespace=myns"
allows write access to myns namespace of any pool, but is_osd_writable()
disallows operations with unmanaged snapshots with such a cap because
its match.pool_namespace.pool_name.empty() is true.  This condition
appears to serve as the "doesn't include support for the application
tag" guard, but it should actually be match.pool_tag.is_match_all()
(or match.pool_tag.application.empty() if open-coded) -- no restriction
on the pool name doesn't automatically mean that there is a restriction
on the application tag.

Fixes: https://tracker.ceph.com/issues/69679
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 5f3815e80028066fb3d6ae9b68d6b69045ab2bdf)

Merge pull request #61507 from ivancich/wip-69652-reef

reef: msg: insert PriorityDispatchers in sorted position

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

Merge pull request #59127 from k0ste/wip-61191-reef

reef: rgw/lua: add more info on package install errors

Reviewed-by: J. Eric Ivancich <ivancich@redhat.com>
Reviewed-by: Yuval Lifshitz <ylifshit@redhat.com>

Merge pull request #59126 from k0ste/wip-63969-reef

reef: rgw/iam: fix role deletion replication

Reviewed-by: Daniel Gryniewicz <dang@redhat.com>
Reviewed-by: J. Eric Ivancich <ivancich@redhat.com>

Merge pull request #59124 from k0ste/wip-65544-reef

reef: rgw: increase log level on abort_early

Reviewed-by: J. Eric Ivancich <ivancich@redhat.com>
Reviewed-by: Daniel Gryniewicz <dang@redhat.com>

Merge pull request #58070 from yuvalif/wip-66496-reef

reef: rgw/admin/notification: add command to dump notifications

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

Merge pull request #57229 from galsalomon66/wip-65245-reef

reef: rgw/s3select: s3select response handler refactor

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

Merge pull request #56158 from yuvalif/wip-64885-reef

reef: rgw/kafka: set message timeout to 5 seconds

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

Merge pull request #61061 from k0ste/wip-69217-reef

reef: rgw: Delete stale entries in bucket indexes while deleting obj

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

Merge pull request #61093 from shashalu/fix-data-corruption-ETIMEDOUT-reef

reef: rgw: fix data corruption when rados op return ETIMEDOUT

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

Merge pull request #61052 from cbodley/wip-69212-reef

reef: rgw: fix bucket link operation

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

Merge pull request #61557 from zdover23/wip-doc-2025-01-29-backport-61549-to-reef

reef: doc/radosgw: s/zonegroup/pools/

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

Merge pull request #59705 from vshankar/wip-67375-reef

reef: mon: fix `fs set down` to adjust max_mds only when cluster is not down

Merge pull request #59920 from mchangir/wip-67827-reef

reef: qa: failfast mount for better performance and unblock `fs volume ls`

https://tracker.ceph.com/projects/cephfs/wiki/Reef#wip-vshankar-testing-20250113141608-reef-debug

Merge pull request #57560 from ceph/wip-lusov-asok-async-reef

reef: mds: don't stall the asok thread for flush commands

Merge pull request #57177 from batrick/wip-65715-reef

reef: mds: inode_t flags may not be protected by the policylock during set_vxattr

Merge pull request #60692 from vshankar/wip-67691-reef

reef: mds: getattr just waits the xlock to be released by the previous client

Merge pull request #59686 from vshankar/wip-67743-reef

reef: qa: wait for file creation before changing mode

Merge pull request #58982 from kotreshhr/wip-67294-reef

reef: client: Fix leading / issue with mds_check_access

doc/radosgw: s/zonegroup/pools/

s/zonegroup/pools/, where this change makes the text clearer.

This change was made in reponse to an upstream comment on
https://pad.ceph.com/p/Report_Documentation_Bugs.

Fixes: https://tracker.ceph.com/issues/69689
Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit ea576ccd037ee0336f01342c86ea9b1468f67cc3)

Merge pull request #61112 from k0ste/wip-68735-reef

reef: os/bluestore: Fix BlueRocksEnv attempts to use POSIX

Merge pull request #61237 from aaSharma14/wip-69291-reef

reef: mgr/dashboard: Fix Latency chart data units in rgw overview page

Reviewed-by: Afreen Misbah <afreen@ibm.com>

Merge pull request #60850 from smanjara/wip-reef-boto-s3-client

reef: qa/multisite: add boto3.client to the library

Merge pull request #60793 from smanjara/wip-69004-reef

reef: rgw/multisite: metadata polling event based on unmodified mdlog_marker

Merge pull request #60733 from ifed01/wip-ifed-tcmalloc-rgw-reef

reef: rgw: link only radosgw with ALLOC_LIBS

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

Merge pull request #60745 from satoru-takeuchi/wip-62757-reef

reef: rgw: when there are a large number of multiparts, the unorder list result may miss objects

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

Merge pull request #59354 from k0ste/wip-63585-reef

reef: rgw/notifications: cleanup all coroutines after sending the notification

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

Merge pull request #59343 from k0ste/wip-61189-reef

reef: rgw/lua: add lib64 to the package search path

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

Merge pull request #59304 from yuvalif/wip-67608-reef

reef: test/rgw/notification: use real ip address instead of localhost

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

Merge pull request #59258 from k0ste/wip-64551-reef

reef: rgw/http/client-side: disable curl path normalization

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

Merge pull request #59134 from k0ste/wip-63250-reef

reef: rgw: fix user.rgw.user-policy attr remove by modify user

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

Merge pull request #61160 from ivancich/wip-69296-reef

reef: rgw: append query string to redirect URL if present

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

Merge pull request #61133 from cbodley/wip-69298-reef

reef: rgw/abortmp: Race condition on AbortMultipartUpload

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

Merge pull request #61105 from cbodley/wip-69259-reef

reef: <common> fix formatter buffer out-of-bounds

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

Merge pull request #61022 from cbodley/wip-69180-reef

reef: qa/rgw: bump keystone/barbican from 2023.1 to 2024.1

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

Merge pull request #60977 from cbodley/wip-69147-reef

reef: test/rgw/multisite: add meta checkpoint after bucket creation

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

Merge pull request #61458 from aaSharma14/wip-69597-reef

reef: mgr/dashboard: fix total objects/Avg object size in RGW Overview Page

Reviewed-by: Afreen Misbah <afreen@ibm.com>

Merge pull request #61038 from rhcs-dashboard/wip-69194-reef

reef: mgr/dashboard: add a custom warning message when enabling feature

Reviewed-by: Afreen Misbah <afreen@ibm.com>

Merge pull request #61236 from aaSharma14/wip-69222-reef

reef: mgr/dashboard: Update and correct zonegroup delete notification

Reviewed-by: Afreen Misbah <afreen@ibm.com>

Merge pull request #61529 from zdover23/wip-doc-2025-01-27-backport-61525-to-reef

reef: doc/cephadm: correct note

doc/cephadm: correct note

Correct the English in a note in doc/cephadm/host-management.rst.

This commit was made in response to
https://github.com/ceph/ceph/pull/50340, which corrected some of this
note in Quincy.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit d15f1f6869465a8cc3191169f038217baecf2c4f)

librbd: clear ctx before initiating close in Image::{aio_,}close()

Image::aio_close() must clear ctx before initiating close.  Otherwise
the provided callback may see a non-NULL ctx and attempt to close the
image again from Image destructor, leading to an invalid memory access
as ImageCtx and ImageState are both freed immediately after the image
is closed (i.e. before AioCompletion is completed and the callback is
executed).

The same adjustment is made to Image::close() just for consistency.

Fixes: https://tracker.ceph.com/issues/69619
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 61baa8793402d416289ff25535d6f9124da3cae9)

Merge pull request #61522 from zdover23/wip-doc-2025-01-26-backport-61503-to-reef

reef: doc/cephfs: edit disaster-recovery-experts (6 of x)

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

Merge pull request #61524 from zdover23/wip-doc-2025-01-26-backport-61520-to-reef

reef: doc/rados: pool and namespace are independent osdcap restrictions

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>

doc/rados: pool and namespace are independent osdcap restrictions

For the "profile {name}" syntax, pool and namespace restrictions are
independent of each other (i.e. specifying namespace doesn't also
require specifying pool like is currently suggested).  A cap can look
like "profile rbd namespace=myns", signifying that the RBD profile is
to be allowed in myns namespace of any pool.

For the "allow {access-spec}" syntax, pool restriction is optional.
A cap can look like "allow r namespace=myns", "allow w object_prefix
myprefix" or "allow rw namespace=myns object_prefix myprefix", for
example.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 67f5769ce6e110b89362763cfb41a0e00e595cdf)

doc/cephfs: edit disaster-recovery-experts (6 of x)

In doc/cephfs/disaster-recovery-experts.rst, incorporate Anthony's
suggestions in

https://github.com/ceph/ceph/pull/61462#discussion_r1923917812
and
https://github.com/ceph/ceph/pull/61462#discussion_r1923920724

and reword the sentences in the section "Using an alternate metadata
pool for recovery" to be in the imperative mood, which better suits the
ordered list format that was introduced in
https://github.com/ceph/ceph/pull/61493.

Follows https://github.com/ceph/ceph/pull/61493.

https://tracker.ceph.com/issues/69557

Co-authored-by: Anthony D'Atri <anthony.datri@gmail.com>
Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 5670054bd0533c8f2507d0596797214da8ba489a)

msg: insert PriorityDispatchers in sorted position

avoid calling stable_sort() after every insertion by inserting directly
into the sorted position. use lower_bound() to insert at the head and
upper_bound() to insert at the tail

this generally only happens during startup so isn't a performance
problem, but std::stable_sort() was triggering strange valgrind warnings
for "Mismatched free() / delete / delete []" when it allocates a
temporary buffer

Fixes: https://tracker.ceph.com/issues/66336
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 063d100526e9e5ece31d575568e0910806f8e74d)

Merge pull request #61102 from ivancich/wip-69255-reef

reef: qa/rgw: pull Apache artifacts from mirror instead of archive.apache.org

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

Merge pull request #61054 from smanjara/wip-69211-reef

reef: qa/rgw: fix s3 java tests by forcing gradle to run on Java 8

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

Merge pull request #61121 from ivancich/wip-69270-reef

reef: qa/rgw: force Hadoop to run under Java 1.8

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

Merge pull request #61500 from zdover23/wip-doc-2025-01-24-backport-61493-to-reef

reef: doc/cephfs: edit disaster-recovery-experts (5 of x)

doc/cephfs: edit disaster-recovery-experts (5 of x)

Put the procedure in the section called "Using an alternate metadata
pool for recovery" into an ordered list, so that it is in a proper
procedure format.

This commit is meant only to break the procedure into steps. The English
language in each of these steps could be improved, but that improvement
will be done after this formatting has been merged and backported.

Follows https://github.com/ceph/ceph/pull/61462.

https://tracker.ceph.com/issues/69557

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 9af7d9786a11e2b07a095c95ae911dfb13e4a61b)

Merge pull request #60237 from aclamk/wip-aclamk-bluefs-truncate-allocations-reef

reef: os/bluestore: Make truncate() drop unused allocations

Merge pull request #61485 from ivancich/wip-69564-reef

reef: rgw/lc: make lc worker thread name shorter

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

Merge pull request #61318 from aclamk/wip-aclamk-ifed-allocation-info-reef

[reef] os/bluestore: introduce allocator state histogram

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

rgw/lc: make lc worker thread name shorter

Fixes: https://tracker.ceph.com/issues/69459
Signed-off-by: lightmelodies <lightmelodies@outlook.com>
(cherry picked from commit 05e241245744d105e285373bb9aa7861c62dcc18)

Merge pull request #61480 from zdover23/wip-doc-2025-01-22-backport-61462-to-reef

reef: doc/cephfs: edit disaster-recovery-experts (4 of x)

test: ceph daemon command with asok path

pg-split-merge using ceph daemon command to check merge.
but it doesn't use asok path, which causes the check not to
return the correct output. change the command to use asok path.

Fixes: https://tracker.ceph.com/issues/65737
Signed-off-by: Nitzan Mordechai <nmordech@redhat.com>
(cherry picked from commit a76bf0de58547961d5fbc6d644005a3a5506d443)

Merge pull request #61360 from zdover23/wip-doc-2025-01-14-backport-61352-to-reef

reef: doc/releases: add actual_eol for quincy

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

Merge pull request #61059 from zdover23/wip-doc-2024-12-12-backport-61049-to-reef

reef: doc/cephfs: edit 2nd 3rd of mount-using-kernel-driver

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

doc/cephfs: edit disaster-recovery-experts (4 of x)

Edit the seventh and final section of
doc/cephfs/disaster-recovery-experts.rst in preparation for adding
deeper explanations of the contexts in which one should use the various
commands listed on that page.

The section edited in this commit is

 * Using an alternate metadata pool for recovery

A future commit might beneficially put this section into the format of
an ordered list. If so, such a commit should only reformat the
content and should not make any changes to the English. It's enough to
verify content or format. Let's not overload our editorial faculties by
forcing ourselves to walk and chew gum at the same time.

Follows https://github.com/ceph/ceph/pull/61442

https://tracker.ceph.com/issues/69557

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit f2529d11745d99e5b5c404193d3dd5e47e48afda)

Merge pull request #61460 from zdover23/wip-doc-2025-01-21-backport-61249-to-reef

reef: doc/cephfs: edit grammar in snapshots.rst

Merge pull request #59786 from zdover23/wip-doc-2024-09-14-backport-59732-to-reef

reef: doc/README.md: improve formatting

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

doc/cephfs: edit grammar in snapshots.rst

This commit improves the grammar in doc/cephfs/snapshots.rst. The PR
associated with this commit follows from
https://github.com/ceph/ceph/pull/61240, the PR raised by Neeraj Pratap
Singh to introduce information about snapshots into the CephFS
documentation.

See also https://tracker.ceph.com/issues/68974.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 9a34ae55ede6dea5bb784a3e6cd45b9e60d4475d)

Merge pull request #61247 from zdover23/wip-doc-2025-01-07-backport-61240-to-reef

reef: doc: add snapshots in docs under Cephfs concepts

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

Merge pull request #61309 from zdover23/wip-doc-2025-01-10-backport-61243-to-reef

reef: doc/radosgw/s3: correct eTag op match tables

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

mgr/dashboard: fix total objects/Avg object size in RGW Overview Page

Till now we are calculating the total number of objects and the average
object size in the RGW Overview Page using `ceph df` command's output.
As per the discussion with RGW team, this data is not correct as S3
objects in rgw can occupy more than one rados object. This PR tends to
make the overview page's info in sync with the RGW bucket page's info.

Fixes: https://tracker.ceph.com/issues/68733
Signed-off-by: Aashish Sharma <aasharma@redhat.com>
(cherry picked from commit 74b0749b7e63f2690a70d0226a1d730b23ddaea1)

Conflicts:
src/pybind/mgr/dashboard/frontend/src/app/ceph/rgw/rgw-overview-dashboard/rgw-overview-dashboard.component.spec.ts
src/pybind/mgr/dashboard/frontend/src/app/ceph/rgw/rgw-overview-dashboard/rgw-overview-dashboard.component.ts

Merge pull request #61332 from zdover23/wip-doc-2025-01-11-backport-60081-to-reef

reef: src/exporter: improve usage message

Merge pull request #61454 from zdover23/wip-doc-2025-01-20-backport-61442-to-reef

reef: doc/cephfs: edit disaster-recovery-experts (3 of x)

doc/cephfs: edit disaster-recovery-experts (3 of x)

Edit the fifth and sixth sections of
doc/cephfs/disaster-recovery-experts.rst in preparation for adding
deeper explanations of the contexts in which one should use the various
commands listed on that page.

The sections edited in this commit are

 - MDS Map Reset
 - Recovery From Mission Metadata Objects

Follows https://github.com/ceph/ceph/pull/61427

https://tracker.ceph.com/issues/69557

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit a69c4142b7f99146af690933b3cfe897a64c7c59)

Merge pull request #61097 from afreen23/wip-69200-reef

reef: mgr/dashboard: handle infinite values for pools

Reviewed-by: Afreen Misbah <afreen@ibm.com>

src/exporter: improve usage message

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit 725b4e184798dc38ec60ab81766577b39fd6e488)