doc/cephfs: edit ceph-dokan.rst (1 of x)

Edit the first 62 lines of doc/cephfs/ceph-dokan. This is part of a
larger project whose goal is the improvement of the CephFS
documentation.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 803140e92113ee7919b8188ea3677b472d91ac48)

Merge pull request #64200 from ideepika/wip-71153-reef

reef: rgw: make keystone work without admin token(service ac requirement)

Merge pull request #64645 from zdover23/wip-doc-2025-07-23-backport-64640-to-reef

reef: doc/cephfs: edit disaster-recovery.rst

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

Merge pull request #64588 from idryomov/wip-70917-reef

reef: test/rbd: remove unit tests about cache tiering

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

Merge pull request #64648 from zdover23/wip-doc-2025-07-23-backport-64639-to-reef

reef: doc/radosgw: edit config-ref.rst

Merge pull request #64596 from idryomov/wip-cephadm-rbd-iscsi-ignore-mon-down-reef

reef: qa/rbd/iscsi: ignore MON_DOWN warning in logs

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

doc/radosgw: edit config-ref.rst

Follow up on the suggestions made by Anthony D'Atri in
https://github.com/ceph/ceph/pull/64532.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 18e9b08393cac7ed8fdb1c65021f74deba3dc44c)

doc/cephfs: edit disaster-recovery.rst

Follow up on the suggestions made by Anthony D'Atri in
https://github.com/ceph/ceph/pull/64604.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 37df8e43baa1e75587686fd661310dfb4ad30312)

Merge pull request #64609 from zdover23/wip-doc-2025-07-21-backport-64604-to-reef

reef: doc/cephfs: edit disaster-recovery.rst

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

Merge pull request #64267 from zdover23/wip-doc-2025-06-30-backport-62785-to-reef

reef: pybind/mgr: Fix missing empty lines in mgr_module.py

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

rgw: make keystone work without admin token(service ac requirement)

Ceph RGW admin credentials must not be a requirement.
Both ec2 auth and keystone token validation work without an admin token.

And the user token verification will use its own token. The only
requirement for the service admin user token is the allow_expired, but
in our case we don't use this parameter.

fixes: https://tracker.ceph.com/issues/68327

Co-authored-by: @kayrus
Signed-off-by: Deepika Upadhyay <deepika.upadhyay@clyso.com>
(cherry picked from commit 55ce1782e159190fd9202b6e270a1a2c470b0626)

Merge pull request #64195 from idryomov/wip-71838-reef

reef: librbd/migration/QCOWFormat: don't complete read_clusters() inline

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

Merge pull request #64311 from ivancich/wip-70824-reef

reef: rgw: add force option to `radosgw-admin object rm ...`

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

Merge pull request #64612 from rhcs-dashboard/sys-packages-dashboard-reef

reef: mgr/dashboard: use system packages when running tox

mgr/dashboard: sitepackages only for py3

Signed-off-by: Nizamudeen A <nia@redhat.com>
(cherry picked from commit 6985ed15ce20c24f7fa87c23ac65b28cf544c9b7)

Revert "mgr/dashboard: pin lxml to fix run-dashboard-tox-make-check failure"

This reverts commit 1f84505f1232dd8445df1a2a819fa000062d3934.

Signed-off-by: Nizamudeen A <nia@redhat.com>
(cherry picked from commit 8e6b6cdd5c302577c81aa041ef4b3f494ed717a7)

 Conflicts:
src/pybind/mgr/dashboard/requirements.txt
 - only removed the lxml pinning

mgr/dashboard: disable saml2 tests when the dep is not there

Signed-off-by: Nizamudeen A <nia@redhat.com>
(cherry picked from commit 8acf8f7e4fe6d283212154857abd97b6e641de4d)
(cherry picked from commit 740df4666545c1e358757beeceba4bfbeed76e7c)

 Conflicts:
src/pybind/mgr/dashboard/controllers/saml2.py
 - kept the config options as it is in reef
src/pybind/mgr/dashboard/tox.ini
 - kept the file as is in reef

Merge pull request #62403 from adamemerson/wip-70422-reef

reef: rgw:  Try  to handle unwatch errors sensibly

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

mgr/dashboard: use system packages when running tox

Fixes: https://tracker.ceph.com/issues/70411
Signed-off-by: Nizamudeen A <nia@redhat.com>
(cherry picked from commit ffde8d044a207d304ec8f916e45c98289f740bec)
(cherry picked from commit 604b7253cdc62a3a7fd7a60d4927b09051246f22)

doc/cephfs: edit disaster-recovery.rst

Edit the section "Data Pool Damage" in doc/cephfs/disaster-recovery.rst.
This commit is part of the project of improving the data-recovery parts
of the CephFS documentation, as requested in the Ceph Power Users
Feedback Summary in mid-2025.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 891e9cf2492ad250936171e9b510e8df4996faf9)

qa/rbd/iscsi: ignore MON_DOWN warning in logs

These seem to be popping up very briefly and don't
appear to be relevant to what is being tested

Signed-off-by: Adam King <adking@redhat.com>
(cherry picked from commit 3b2064e7066fa58e0d812e85a293f74401d295f4)

test: remove unit tests about cache tiering

Cache tiering was deprecated in the Reef release,
so we are no longer supporting any issues related
to it.

Ref: https://docs.ceph.com/en/latest/rados/operations/cache-tiering/

Fixes: https://tracker.ceph.com/issues/70917
Signed-off-by: Laura Flores <lflores@ibm.com>
(cherry picked from commit b6dd408c87b384ce88b097f706f324e50fc61ace)

Merge pull request #64497 from zdover23/wip-doc-2025-07-15-backport-63877-to-reef

reef: doc/rados/ops: edit cache-tiering.rst

doc/rados/ops: edit cache-tiering.rst

Add material to doc/rados/operations/cache-tiering.rst, as suggested by
Anthony D'Atri in
https://github.com/ceph/ceph/pull/63745#discussion_r2127887785.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit d3c46820a5fc72391ef46ab4b03bbe867e0e51d2)

rgw: clean-up of some code

Comments added and formatting adjusted.

Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
(cherry picked from commit a84f984d0324e9abc060f166dd626063361a7ab0)

rgw: add doc & doc testing for force option on `radosgw-admin object rm ...`

Subject says it all.

Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
(cherry picked from commit 333b5cd1e36c53b69462961046199e69e906f1cc)

rgw: add force option to `radosgw-admin object rm ...`

The `radosgw-admin object rm ...` sub-command will give up if it
determines that there's an issue with the head object. This can make
it difficult for an admin to clean up a bucket index when there's a
damaged or missing head object.

When the user adds the "--yes-i-really-mean-it" command-line option,
it enables the "force mode". The bucket index entry(ies) will be
removed. If the object being removed is the current version in a
versioned bucket, the appropriate changes to the OLH will take place.

Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
(cherry picked from commit 7e1ac21069e14a4d6c4d4748971943b7c8b8137c)

Merge pull request #64548 from zdover23/wip-doc-2025-07-17-backport-64532-to-reef

reef: doc/radosgw: edit "Lifecycle Settings"

doc/radosgw: edit "Lifecycle Settings"

Edit the section "Lifecycle Settings" in the file
doc/radosgw/config-ref.rst. Remove solecisms and pleonasms and plain old
infelicitious formulations.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit ac2e5f502523d1bf326303e904ccb47236c81fcb)

Merge pull request #64531 from zdover23/wip-doc-2025-07-16-backport-64433-to-reef

reef: doc: update mgr modules notify_types

doc: update mgr modules notify_types

Signed-off-by: Nitzan Mordechai <nmordec@redhat.com>
(cherry picked from commit fc4396d6280fcbf0a95567cff144052d81dcd964)

Merge pull request #64340 from joscollin/wip-71949-reef

reef: qa: fix multi-fs tests in test_mds_metrics.py

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

Merge pull request #64493 from zdover23/wip-doc-2025-07-15-backport-64483-to-reef

reef: doc: add note admonitions in two files

doc: add note admonitions in two files

Add note admonitions when discussing client package support in the
context of OS Recommendations in the following two files:

- doc/cephfs/ceph-dokan.rst
- doc/rbd/rbd-windows.rst

This addresses a change requested by Ilya Dryomov in
https://github.com/ceph/ceph/pull/64374#discussion_r2199756581.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 69d641f0207d803cd9a3c3e102d5b2073e6b0f77)

Merge pull request #63343 from amathuria/wip-69773-reef

reef: mon/test_mon_osdmap_prune: Use first_pinned instead of first_committed

Merge pull request #64482 from zdover23/wip-doc-2025-07-15-backport-64374-to-reef

reef: doc: Clarify the status of MS Windows client support

doc: Clarify the status of MS Windows client support

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit 89eabfc3f0c8db3501b3ea3097e2983617c5234a)

Merge pull request #64473 from zdover23/wip-doc-2025-07-14-backport-64462-to-reef

reef: doc/cephfs: Improve mount-using-fuse.rst

Merge pull request #64476 from zdover23/wip-doc-2025-07-14-backport-63080-to-reef

reef: doc/radosgw: Improve rgw-cache.rst

doc/radosgw: Improve rgw-cache.rst

Try to improve the language by completely rewriting some sentences.
Attempt to format the document more like the rest of the docs.
Fix several errors in punctuation, capitalization, spaces etc.
Use blocks with bash prompts for CLI commands instead of hardcoded
prompts.
Fix section hierarchy and section title underline lengths.
Use admonition.

Signed-off-by: Ville Ojamo <14869000+bluikko@users.noreply.github.com>
(cherry picked from commit 6e836f8f1e1e53bc7f8d8b497960b100e6b625d6)

doc/cephfs: Improve mount-using-fuse.rst

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit 329ee7b3038e49cf0def2f2628444e3e90796c05)

Merge pull request #64270 from zdover23/wip-doc-2025-06-30-backport-64164-to-reef

reef: mgr/dashboard: Fix inline markup warning in API documentation

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

Merge pull request #64312 from ivancich/wip-69409-reef

reef: rgw: fix empty storage class on display of multipart uploads

Merge pull request #64281 from mkogan1/wip-71904-reef

reef: qa/rgw: fix perl tests missing Amazon::S3 module

Merge pull request #64397 from zdover23/wip-doc-2025-07-09-backport-64161-to-reef

reef: doc: Document ceph-mgr module configuration options

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

Merge pull request #63805 from badone/wip-tracker-70645-mon_memory_target-fix-reef

Reef: OSDMonitor: Make sure pcm is initialised

Merge pull request #56520 from k0ste/wip-64412-reef

reef: osd: don't send stale hb msgr's addresses in MOSDBoot

Merge pull request #62114 from joscollin/wip-70286-reef

reef: cephfs-journal-tool: fix segfault during 'journal import' from invalid dump file

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

Merge pull request #62116 from joscollin/wip-70280-reef

reef: qa: fix test_cephfs_mirror_stats failure

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

Merge pull request #64167 from zdover23/wip-doc-2025-06-25-backport-64160-to-reef

reef: doc/dev/config: Document how to use :confval: directive for config op…

Merge pull request #64400 from zdover23/wip-doc-2025-07-09-backport-64162-to-reef

reef: doc: do not depend on typed-ast

doc: do not depend on typed-ast

the typed-ast project was marked end of life since July 2023, and
not maintained anymore. since we build the document using readthedocs'
service, and in .readtherdocs.yml we use python 3.9, which comes with
ast module included by its standard library.

the typed-ast dependency was originally added in 30d41597, but now that
we are using python 3.9, there is no need to use this module anymore.

Signed-off-by: Kefu Chai <tchaikov@gmail.com>
(cherry picked from commit 85b6d55b1f4587121d92a061950cd7ad7e13b082)

doc: Document ceph-mgr module configuration options

Add comprehensive documentation for defining configuration options in
ceph-mgr modules, including all supported properties and their usage.

Previously, the documentation did not explain how to define ceph-mgr
module configuration options, despite subtle differences from other Ceph
components. This change documents all supported Option properties, their
types, and provides clear examples to help module developers properly
configure their options.

Signed-off-by: Kefu Chai <tchaikov@gmail.com>
(cherry picked from commit 121192f4c95df0ee282cecc7237c8ca788e9cfba)

Merge pull request #61773 from joscollin/wip-69926-reef

reef: cephfs-top: exception when terminal size greater than PAD_WIDTH

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

Merge pull request #64394 from zdover23/wip-doc-2025-07-09-backport-64334-to-reef

reef:  doc/rados: remove clonedata command

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

Merge pull request #64218 from zdover23/wip-doc-2025-06-27-backport-62944-to-reef

reef: common/options: fix typo in description

Merge pull request #63772 from zdover23/wip-doc-2025-06-06-backport-63739-to-reef

reef: doc/mgr: edit telemetry (2 of x)

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

Merge pull request #64179 from zdover23/wip-doc-2025-06-25-pr64140-reversion-reef-only

doc: Revert "doc/radosgw: add "persistent_topic_size""

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

doc/rados: remove clonedata command

Remove the "clonedata" command from the rados manpage.

The "clonedata" command was removed in
https://github.com/ceph/ceph/pull/13008.

Fixes: https://tracker.ceph.com/issues/71943
Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 1eff1c1147bdb683701a55fbe3ea846fde6937f0)

Merge pull request #64344 from zdover23/wip-doc-2025-07-04-backport-63779-to-reef

reef: doc/mgr: edit telemetry.rst

Merge pull request #63979 from cbodley/wip-71689-reef

reef: test/common: unittest_fault_injector omits unit-main target

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

Merge pull request #64339 from zdover23/wip-doc-2025-07-03-backport-64319-to-reef

reef: doc/radosgw: add rgw_enable_lc_threads & rgw_enable_gc_threads

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

doc/mgr: edit telemetry.rst

Edit doc/mgr/telemetry.rst.

Incorporate the suggestions made by Anthony D'Atri in
https://github.com/ceph/ceph/pull/63735.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 97ef126e4b6a9dcdf9872cf37cdebec289e84acd)

Merge pull request #64322 from zdover23/wip-doc-2025-07-03-backport-64306-to-reef

reef: doc/rados/ops: add caps restore command

Merge pull request #64148 from zdover23/wip-doc-2025-06-25-backport-64088-to-reef

reef: doc: Fixed a spelling error.

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

Merge pull request #64316 from zdover23/wip-doc-2025-07-03-backport-63359-to-reef

reef: doc/radosgw/s3: Document delete-if-unmodified-since

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

qa: fix multi-fs tests in test_mds_metrics.py

* Avoids the unnecessary setup, when writing a multi-fs test.
  Avoids creating the default filesystem, deleting it and creating the required filesystems, mounting them.
  This change uses the filesystems created using 'REQUIRE_BACKUP_FILESYSTEM' for conducting tests.
* This change consequently fixes the old/deleted filesystems appearing in the `perf stats` output,
  making it stale output.
* Drops unused function parameters.

Fixes: https://tracker.ceph.com/issues/68001
Fixes: https://tracker.ceph.com/issues/68446
Signed-off-by: Jos Collin <jcollin@redhat.com>
(cherry picked from commit eaf2a8360d0d70b20d5ea61022fdde4f6a9b6464)

qa: reconfigure only the appropriate client auth caps

* Fix _setup_fs to reconfigure only the appropriate client auth caps,
  by avoiding iterating the self.mounts.

* Delete unused filesystems in test_perf_stats_stale_metrics_with_multiple_filesystem.

Signed-off-by: Jos Collin <jcollin@redhat.com>
(cherry picked from commit 7ac14cac91c73779d480d42c38f4f99524ceedd7)

doc/radosgw: add rgw_enable_lc_threads & rgw_enable_gc_threads

Add rgw_enable_lc_threads and rgw_enable_gc_threads to
doc/radosgw/config-ref.rst.

Fixes: https://tracker.ceph.com/issues/58092
Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 45e5468ea7fc57b1cb60f1936558dc2678adc6fd)

doc/rados/ops: add caps restore command

Add a command that restores caps to client.admin after they have been
accidentally destroyed or removed.

Fixes: https://tracker.ceph.com/issues/23594
Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit de8c148a2f99db3a1dc4eb70a19aca19a0e9bacf)

Merge pull request #64022 from zdover23/wip-doc-2025-06-19-backport-63983-to-reef

reef: doc/radosgw/admin.rst: explain bucket and uid flags for bucket quota

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

doc: Revert "doc/radosgw: add "persistent_topic_size""

This reverts commit c25d52a28fc4984d7e42df599026bc7b2064f139.

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

doc/radosgw/s3: Document delete-if-unmodified-since

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit 3381041d190fd48cf9f6d53f679f1737028cccf4)
(cherry picked from commit 7e2bd1595d64c12e1959cc29acfaf906a9027fc6)

doc/mgr: edit telemetry (2 of x)

Improve the English and the formatting in doc/mgr/telemetry.rst. This
follows up on https://github.com/ceph/ceph/pull/63476.

This commit edits the second hundred lines in doc/mgr/telemetry.rst.

(cherry picked from commit f8975b05a6cff674ad5fb6ce93c839ca1856464c)
Signed-off-by: Zac Dover <zac.dover@proton.me>

rgw: fix empty storage class on display of multipart uploads

Some multipart uploads do not have a stored storage class, however the
code is written such that an empty storage class is treated as the
"STANDARD" storage class. So when encoding the storage class in JSON,
use the canonical storage.

Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
(cherry picked from commit eab096c188d2430ccc3e865b614b845f1a92af08)

Merge pull request #63813 from zdover23/wip-doc-2025-06-09-backport-63782-to-reef

reef: doc/radosgw: edit metrics.rst

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

Merge pull request #63810 from zdover23/wip-doc-2025-06-09-backport-63781-to-reef

reef: doc/mgr: edit telemetry.rst

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

Merge pull request #64285 from zdover23/wip-doc-2025-07-01-backport-64239-to-reef

reef: doc/mgr/crash.rst: remove outdated module enabling instructions

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

Merge pull request #64288 from zdover23/wip-doc-2025-07-01-backport-64255-to-reef

reef: doc/rados/configuration: Small improvements in ceph-conf.rst

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

Merge pull request #64171 from zdover23/wip-doc-2025-06-25-backport-64163-to-reef

reef: doc: Fix unterminated inline literal in ceph-conf.rst

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

doc: Fixed a spelling error.

Corrected communicte to communicate

Signed-off-by: Instelligence.io <github@instelligence.io>
(cherry picked from commit c328902e3929d9841f952948b0336ec5b20d36fa)

doc/rados/configuration: Small improvements in ceph-conf.rst

Fix a missing space typo.

Use title case in section titles.

Use ref for hyperlink to a section that already has a label instead of
linking to title text of the section.

Use privileged command prompt for all ceph commands except ceph commands
that invoke help.

Use INI preformatted block for last remaining block that was just using
plain preformatted.

Signed-off-by: Ville Ojamo <14869000+bluikko@users.noreply.github.com>
(cherry picked from commit 6687c9767d9764f34d7099ba6980992dd8ae8737)

doc/mgr/crash.rst: remove outdated module enabling instructions

The crash module has been enabled by default since commit 18f253aa in
Nautilus and is now in the always_on_modules list. However, the
documentation still contained instructions for manually enabling it.

When users followed these outdated instructions, they encountered:
```
  module 'crash' is already enabled (always-on)
```

The module cannot be disabled either. Running:
```
  ceph mgr module disable crash
```

Returns the error:
```
  Error EINVAL: module 'crash' cannot be disabled (always-on)
```
In this change, we remove the obsolete enabling instructions and clarify
that this module is always active and cannot be disabled.

Signed-off-by: Kefu Chai <tchaikov@gmail.com>
(cherry picked from commit 0fcd7677586f7edf3eb091420a8ace0c4f887167)

qa/rgw: fix perl tests missing Amazon::S3 module

and a second case where perl tests can fail without error output

1. fix errors like: `Can't locate Amazon/S3.pm in @INC (you may need to
   install the Amazon::S3 module)`
by priming the perl tests with installing the Amazon::S3 module from cpan

ex:
```
2025-06-23T19:18:40.162 INFO:tasks.workunit.client.0.smithi090.stderr:Can't locate Amazon/S3.pm in @INC (you may need to install the Amazon::S3 module) (@INC contains: /usr/local/lib64/perl5/5.32 ...
```

2. log an error when RGW process is not detected

Fixes: https://tracker.ceph.com/issues/71577
Signed-off-by: Mark Kogan <mkogan@redhat.com>
(cherry picked from commit 7faa23f160c9f4b40d25fe27f2345dbf999b0c84)

osd: don't send stale hb msgr's addresses in MOSDBoot

See comments in the ticket for the RCA.

NOTE: we can't just hold a reference to what `get_myaddrs()`
returns as the `safe_item_history` is involved:

```cpp
template<class T>
class safe_item_history {
  //...
  T *current = nullptr;

  // ...
  const T& operator=(const T& other) {
    std::lock_guard l(lock);
    history.push_back(other);
    current = &history.back();
    return *current;
  }
```

Fixes: https://tracker.ceph.com/issues/58915
Signed-off-by: Radosław Zarzyński <rzarzyns@redhat.com>
(cherry picked from commit 81944f60478abe2d4253caaeb4165da5ccffaca1)

mgr/dashboard: Fix inline markup warning in API documentation

Remove trailing space from summary field that was causing Sphinx build
warning.

Sphinx was generating a warning due to malformed inline markup:

```
/home/kefu/dev/ceph/doc/mgr/ceph_api/index.rst:3349: WARNING: Inline strong start-string without end-string.`
```

The openapi directive appears to convert trailing spaces into asterisk
markers, creating unterminated strong markup. This change removes the
trailing space to eliminate the warning and maintain consistency with
other entries in the file.

Signed-off-by: Kefu Chai <tchaikov@gmail.com>
(cherry picked from commit 5f1dc9af57b0398c33226410f801d16feed823f2)

pybind/mgr: Fix missing empty lines in mgr_module.py

The inline function documentation needs an empty
line between description and metadata.
Otherwise automethod from docs renders the metadata
as text together with the description instead of
rendering it as function metadata, e.g. parameters
and return value.

Signed-off-by: Ville Ojamo <14869000+bluikko@users.noreply.github.com>
(cherry picked from commit b521e969b94b93b41a37ef5f9726ed63cb77a97c)

Merge pull request #64114 from zdover23/wip-doc-2025-06-23-backport-64103-to-reef

reef: doc/radosgw: improve "pubsub_push_pending" info

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

common/options: fix typo in description

Signed-off-by: Lorenz Bausch <info@lorenzbausch.de>
(cherry picked from commit aa54c08eb01efa4f048cc8fba9f909e0226ac905)

Merge pull request #64156 from zdover23/wip-doc-2025-06-25-backport-64107-to-reef

reef: doc/radosgw: remove "pubsub_event_triggered"

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

Merge pull request #64093 from idryomov/wip-71335-reef

reef: librbd/cache/pwl: fix memory leak in SyncPoint persist context cleanup

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

Merge pull request #64091 from idryomov/wip-70951-reef

reef: rbd-mirror: release lock before calling m_async_op_tracker.finish_op()

Reviewed-by: Vinay Bhaskar Varada <vvarada@redhat.com>

Merge pull request #63927 from cbodley/wip-71670-reef

reef: qa/rgw: bump maven version in hadoop task to resolve 404 Not Found

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

Merge pull request #64052 from cbodley/wip-71750-reef

reef: fix: the RGW crash caused by special characters

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

librbd/migration/QCOWFormat: don't complete read_clusters() inline

When the cluster needs to be read, the completion is posted to ASIO.
However, in the two special cases (cluster DNE and zero cluster), the
completion is completed inline at the moment.  This violates invariants
and can eventually lead to a lockup.  For example, in a scenario of
a read from a clone image whose parent is under migration:

  io::ObjectReadRequest::read_parent()
    io::util::read_parent()
      < image_lock is taken for read >
      io::ImageDispatchSpec::send()
        migration::ImageDispatch::read()
          migration::QCOWFormat::ReadRequest::send()
            ...
            migration::QCOWFormat::ReadRequest::read_clusters()
              < cluster DNE >
              migration::QCOWFormat::ReadRequest::handle_read_clusters()
                io::AioCompletion::complete()
                  io::ObjectReadRequest::copyup()
                    is_copy_on_read()
                      < image_lock is taken for read >

copyup() expects to be called with no locks held, but going through
QCOWFormat in the "cluster DNE" case essentially maintains image_lock
taken in read_parent() and then it's taken again by the same thread in
is_copy_on_read().  Under pthreads, it's not a problem:

  A thread may hold multiple concurrent read locks on rwlock (that is,
  successfully call the pthread_rwlock_rdlock() function n times). If
  so, the thread must perform matching unlocks (that is, it must call
  the pthread_rwlock_unlock() function n times).

But according to C++ standard it's undefined behavior:

  If lock_shared is called by a thread that already owns the mutex in
  any mode (exclusive or shared), the behavior is undefined.

Other, longer and more elaborate, call chains are possible too and
there it may end up being a write lock, a tripped assertion, etc.  To
avoid this, make the special cases in read_clusters() behave the same
as the main path.

Fixes: https://tracker.ceph.com/issues/71838
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit a02cc13191d5e695620791c883ff1460d2e42f57)

librbd/migration/QCOWFormat: rename handle_read_cluster()

... to handle_read_clusters() as it's a callback for read_clusters().

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 9d9c73435341f98cecf0a450357408f8d49624fa)

Merge pull request #63616 from ceph/djg-reef-rtd

reef: .github: Fix RTD build retrigger

Merge pull request #64130 from tchaikov/wip-reef-ceph-spec.in

reef: ceph.spec.in: Remove rgw-restore-bucket-index.8* from packaging

Reviewed-by: Matan Breizman <mbreizma@redhat.com>
Reviewed-by: Yuri Weinstein <yuriw@redhat.com>
Reviewed-by: John Mulligan <jmulligan@redhat.com>
Reviewed-by: Casey Bodley <cbodley@redhat.com>

doc/radosgw: remove "pubsub_event_triggered"

Remove "pubsub_event_triggered" from the list of "Notification
Performance Statistics". It is obsolete.

Fixes: https://tracker.ceph.com/issues/71789
Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 44dc57fc15749583fc13560c9409b7687df7c223)

doc: Fix unterminated inline literal in ceph-conf.rst

Add missing space after closing backticks to properly terminate inline literal.

Sphinx was generating a warning due to an unterminated inline literal
string:

```
/home/kefu/dev/ceph/doc/rados/configuration/ceph-conf.rst:320: WARNING: Inline literal start-string without end-string.
```

This fix ensures the inline literal is properly closed and eliminates the
build warning.

Signed-off-by: Kefu Chai <tchaikov@gmail.com>
(cherry picked from commit 7f79f3d6aa7f71c9b2434b843036a3f1da660231)