librbd: disallow "rbd trash mv" if image is in a group

Removing an image that is a member of a group has always been
disallowed.  However, moving an image that is a member of a group to
trash is currently allowed and this is deceptive -- the only reason for
a user to move an image to trash should be the intent to remove it.

More importantly, group APIs operate in terms of image names -- there
are no corresponding variants that would operate in terms of image IDs.
For example, even though internally GroupImageSpec struct stores an
image ID, the public rbd_group_image_info_t struct insists on an image
name.  When rbd_group_image_list() encounters a trashed member image
(i.e. one that doesn't have a name), it just fails with ENOENT and no
listing gets produced at all until the offending image is restored from
trash.  Something like this can be very hard to debug for an average
user, so let's make rbd_trash_move() fail with EMLINK the same way as
rbd_remove() does in this scenario.

The one case where moving a member image to trash makes sense is live
migration where the source image gets trashed to be almost immediately
replaced by the destination image as part of preparing migration.

Fixes: https://tracker.ceph.com/issues/71026
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 2ae4a57e160f21201b9dfb4c8adb02f2826c7106)

Conflicts:
PendingReleaseNotes [ moved to >=19.2.2 section ]

pybind/rbd: add ImageMemberOfGroup exception

EMLINK is returned by rbd_remove() if the image is a member of a group.
Add a dedicated exception similar to ImageBusy or ImageHasSnapshots and
a test for it.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 61075e344c18e57b7416072f66c2e335d4d89d75)

Conflicts:
src/test/pybind/test_rbd.py [ commits 68eea0eb814e
  ("src/tools/rbd: add group info command to output group id")
  and e5ccce14c4b0 ("rbd: add group snap info command") not in
  squid ]

rbd: don't print "image will expire at" message when trash_move() fails

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 1fd79d41a527ae0e0e4a66a18d67c4b180f66106)

Merge pull request #62337 from yuvalif/wip-67309-squid

squid: rgw/async/notifications: use common async waiter in pubsub push

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

Merge pull request #61546 from aiivashchenko/wip-69588-squid

squid: rgw: add missing last_modified field to swift API

Reviewed-by: Daniel Gryniewicz <dang@redhat.com>
Reviewed-by: Tobias Urdin <tobias.urdin@binero.com>

Merge pull request #61995 from cbodley/wip-70170-squid

squid: rgw: fix to correctly store updated attrs in backend store after erasing an attr/attrs for delete ops on a bucket.

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

Merge pull request #61532 from soumyakoduri/wip-skoduri-squid

squid: rgw: Fix a bug in LCOpAction_Transition::check()

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

Merge pull request #61782 from cbodley/wip-69933-squid

squid: rgw: allow management of `accounts` user caps

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

Merge pull request #61242 from kchheda3/wip-69410-squid

squid: rgw/notification: Forward Topic & Notification creation request to master when notification_v2 enabled

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

Merge pull request #61161 from ivancich/wip-69136-squid

squid: rgw: relax RGWPutACLs_ObjStore::get_params read data log

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

Merge pull request #61840 from vshankar/wip-69683

squid: mds: prevent duplicate wrlock acquisition for a single request

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

Merge pull request #62531 from aaSharma14/wip-70686-squid

squid: mgr/dashboard: fix image filter's query on rbd-details grafana panel

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

common: missing std include with GCC 14

In file included from src/rgw/driver/posix/bucket_cache.h:19,
                 from src/test/rgw/test_posix_bucket_cache.cc:4:
src/common/cohort_lru.h: In member function _void cohort::lru::TreeX<T, TTree, CLT, CEQ, K, LK>::lock()_:
src/common/cohort_lru.h:334:14: error: _for_each_ is not a member of _std_
  334 |         std::for_each(locks.begin(), locks.end(),
      |              ^~~~~~~~
src/common/cohort_lru.h: In member function _void cohort::lru::TreeX<T, TTree, CLT, CEQ, K, LK>::unlock()_:
/home/yuvalif/ceph5/src/common/cohort_lru.h:339:14: error: _for_each_ is not a member of _std_
  339 |         std::for_each(locks.begin(), locks.end(),
      |              ^~~~~~~~

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

rgw/async/notifications: use common async waiter in pubsub push

* use the "yield_waiter" and "waiter" from common/async insteasd of the "waiter"
  implemented inside the bucket notification code (this is so we don't
  need separate investigations for 2 implementations)
* added a unit test that simulate how a separate thread (kafka or amqp) is
resuming a coroutine which is created by either the frontend or the
notification manager.

before using "defer" the unit test is passing, however,
when executed under thread sanitizer (using the WITH_TSAN cmake flag)
the following errors are observed: https://0x0.st/Xp4P.txt
after using "defer" the unit test passes under TSAN without errors.

Fixes: https://tracker.ceph.com/issues/64184
Signed-off-by: Yuval Lifshitz <ylifshit@ibm.com>
(cherry picked from commit 2872c75f184c9e715219dfa9ad44f5b6cfe4e1fe)

common/async: add yield_waiter template

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit dd779c74e1eebaf888d95b2329c7d5ead176f0a9)

Merge pull request #62386 from pritha-srivastava/wip-70535-squid

squid: rgw/sts: replacing load_stats with list_buckets

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

Merge pull request #61062 from k0ste/wip-69218-squid

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

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

Merge pull request #62586 from cbodley/wip-70738-squid

squid: rgw: fixing tempest ObjectTestACLs and ObjectACLsNegativeTest cases

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

Merge pull request #62590 from cbodley/wip-70739-squid

squid: cls/rgw: non-versioned listings skip past version suffix

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

Merge pull request #62365 from cbodley/wip-70538-squid

squid: qa/rgw: fix user cleanup in s3tests task

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

Merge pull request #62623 from rhcs-dashboard/wip-70762-squid

squid: mgr/dashboard: enable ha by default on subsystem POST API

Reviewed-by: Naman Munet <nmunet@redhat.com>

Merge pull request #62495 from kchheda3/wip-66805-squid

squid: rgw/notification: For kafka include user-id & password as part of the key along with endpoint for connection pooling.

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

Merge pull request #62881 from cbodley/wip-doc-rgw-user-admin-system-squid

squid: doc/rgw: document Admin and System Users

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

doc/rgw: document Admin and System Users

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit dc93b888f16793b00390b32d6814f9243b6808e2)

doc/rgw: Admin Capabilities section links to Admin Ops API

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit cdc04b87842e02f81add403c8843b77e56cb3ef4)

Merge pull request #62874 from zdover23/wip-doc-2025-04-18-backport-62300-to-squid

squid: doc/cephfs: link section for pausing async threads in section for...

doc/cephfs: link section for pausing async threads in section for...

disabling volumes plugin part.

Fixes: https://tracker.ceph.com/issues/70479
Signed-off-by: Rishabh Dave <ridave@redhat.com>
(cherry picked from commit 36cfbfb0eacbfbd459558d1bdff699daf8fae575)

Merge pull request #60978 from cbodley/wip-69148-squid

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

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

Merge pull request #62856 from zdover23/wip-doc-2025-04-17-backport-62763-to-squid

squid: doc/radosgw: Promptify CLI, cosmetic fixes

doc/radosgw: Promptify CLI, cosmetic fixes

Use the more modern prompt block for CLI commands
and use right one $ vs #.
Fix indentation on JSON example outputs and
some CLI command switches.
Add some arguably missing comma in JSON example output.
Add a full stop at the end of a one-sentence paragraph.
Remove extra comma mid-sentence in another.
Fix missing backslashes or typo at end of multiline commands.
Lines under section headings as long as heading text.
Fix hyperlinks. Fix list items prefixed with - insted of *.
Format configuration syntax in the middle of text as code.
Fix typo "PI" to "API" and remove extra space.
Remove colons at the end of section headers in a few places.
Use Title Case in section titles consistently with short words lowercase.
Possibly controversial: don't add whitespace before and
after main title section header text.
Possibly controversial: don't indent line continuation
backslashes, leave only 1 space before them.

Signed-off-by: Ville Ojamo <14869000+bluikko@users.noreply.github.com>
(cherry picked from commit 7fdaa94181c6fca7d4db0b6f4ee6646528121d43)

Merge pull request #58285 from pponnuvel/wip-66701-squid

squid: mgr/nfs: Don't crash ceph-mgr if NFS clusters are unavailable.
Reviewed-by: Kamoltat Sirivadhna <ksirivad@redhat.com>

Merge pull request #62834 from zdover23/wip-doc-2025-04-16-backport-62795-to-squid

squid: doc/radosgw/cloud-transition: fix details

doc/radosgw/cloud-transition: fix details

Fix documentation on Ceph cloud storage transitions - broken link urls, returning back reStructuredText and some other details.

Signed-off-by: Laimis Juzeliunas <laimis.juzeliunas@oxylabs.io>
Co-authored-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit c47b559547326f837dde9dfed9eef526e275fa8e)

Merge pull request #62202 from ifed01/wip-ifed-fix-expand-squi

squid: os/bluestore: fix bdev expansion and more

Reviewed-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
Reviewed-by: Adam Kupczyk <akupczyk@ibm.com>

tool/ceph-bluestore-tool: Make bluefs-bdev-expand command output nicer.

Here is a sample output:

inferring bluefs devices from bluestore path
0 : device size 0x4049c000(1.0 GiB) : using 0x1801000(24 MiB)
1 : device size 0x140000000(5 GiB) : using 0x1502000(21 MiB)
2 : device size 0x8c0000000(35 GiB) : using 0x40014000(1.0 GiB)
Expanding DB/WAL...
0 : nothing to do, skipped
1 : Expanding to 0x140000000(5 GiB)
1 : size updated to 0x140000000(5 GiB)
2 : Expanding to 0x8c0000000(35 GiB)
2 : size updated to 0x8c0000000(35 GiB)

Fixes: https://tracker.ceph.com/issues/67966
Signed-off-by: Igor Fedotov <igor.fedotov@croit.io>
(cherry picked from commit ac7789139e3d4ba3bfd69ddcd4fe504c35b42bc3)

os/bluestore: do not log undecodable bdev label as an error

Signed-off-by: Igor Fedotov <igor.fedotov@croit.io>
(cherry picked from commit a70a3ced7e1f4ba0d124550d053e54d3a320f782)

doc/ceph-bluestore-tool: add description for show-label-at command.

Signed-off-by: Igor Fedotov <igor.fedotov@croit.io>
(cherry picked from commit 4f5f8483444e4fa16939d917ca5e42120e7ccbc9)

os/bluestore: fix alloc update after bdev expansion in NCB mode.

Fixes: https://tracker.ceph.com/issues/69997
Signed-off-by: Igor Fedotov <igor.fedotov@croit.io>
(cherry picked from commit bd7764bc699b4cc6b3edaf99369e40b59f335b21)

os/bluestore: be less strict in main bdev label validation.

This eliminates treating as an error the case when valid bdev label(s)
exists at location(s)  beyond the size in bdev label.
This is effectively not an error but _check_main_bdev_label() returns an
error in this case. Which is undetectable by fsck and unrecoverable by repair.

Signed-off-by: Igor Fedotov <igor.fedotov@croit.io>
(cherry picked from commit 2a9c776f6aa39e8f73ed0159f3c58ad2522a114a)

os/bluestore: use proper device when updating DB/WAL label while
expanding.

Fixes: https://tracker.ceph.com/issues/69999
Signed-off-by: Igor Fedotov <igor.fedotov@croit.io>
(cherry picked from commit ba3e4d2c9daf6b3af8340607e722f31b83e69db8)

os/bluestore: don't use bdev.size() when dealing with bdev labels in fsck.

This might cause assertions after incomplete volume expansion
(expand-device cmd hasn't been called) as allocmap bitmaps are initialized with
bdev label.size not bdev.size() and hence they are accessed
out-of-bound.

Signed-off-by: Igor Fedotov <igor.fedotov@croit.io>
(cherry picked from commit 173bbe7dba7f0b835246fab9ecdba1f13c544460)

os/bluestore: fix bdev label.size update when expanding device.

Fixes: https://tracker.ceph.com/issues/69981
Signed-off-by: Igor Fedotov <igor.fedotov@croit.io>
(cherry picked from commit c06ccf57de36d8485754a9e65ce2ce997dbe7c5c)

tools/bluestore-tool: add "show-label-at" command

Signed-off-by: Igor Fedotov <igor.fedotov@croit.io>
(cherry picked from commit 19788291c991d11313fb2bebce48dcb7354b882e)

tools/bluestore-tool: dump label locations with "show-label" cmd

Signed-off-by: Igor Fedotov <igor.fedotov@croit.io>
(cherry picked from commit 97bbbb11b3b4c5e5c69059f3c07a3d026efc8261)

Merge pull request #62151 from YiteGu/wip-70148-squid

squid: blk/kerneldevice: notify_all only required when discard_drain wait for condition

Merge pull request #62576 from YiteGu/wip-70671-squid

squid: blk/kerneldevice: fix invalid iterator usage after erase in discard_q…

Merge pull request #62810 from zdover23/wip-doc-2025-04-15-backport-62799-to-squid

squid: doc/cephadm/services: Fix formatting in osd.rst

doc/cephadm/services: Fix formatting in osd.rst

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit c56d2a5aaf79a3440607946a55b9e59ca8a6de5b)

Merge pull request #61323 from ljflores/wip-69493-squid

squid: common/pick_address: Add IPv6 support to is_addr_in_subnet

Merge pull request #62796 from zdover23/wip-doc-2025-04-14-backport-62784-to-squid

squid: doc/cephadm: Add admonition re restarting an OSD service

doc/cephadm: Add admonition re restarting an OSD service

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit 524db4609f5f0b4639828b327cfc972243e08ce7)

Merge pull request #62711 from cbodley/wip-70783-squid

squid: rgw: keep the tails when copying object to itself

Reviewed-by: Jane Zhu <jzhu116@bloomberg.net>

Merge pull request #62739 from ceph/squid-release

v19.2.2

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

Merge pull request #62311 from adamemerson/wip-63786-squid

squid: test/cls_2pc_queue: fix multi-threaded access to non-atomic variables

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

Merge pull request #62307 from adamemerson/wip-63378-squid

squid: rgw/multisite: handle errors properly in RGWDataFullSyncSingleEntryCR()

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

Merge pull request #60785 from awojno-bloomberg/wip-68995-squid

squid: rgw: implement x-amz-replication-status for PENDING & COMPLETED

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

Merge pull request #62731 from afreen23/wip-70836-squid

squid: mgr/dashboard: Fix empty ceph version in GET api/hosts

Reviewed-by: Afreen Misbah <afreen@ibm.com>
Reviewed-by: Naman Munet <nmunet@redhat.com>

Merge pull request #62395 from shraddhaag/wip-70135-squid

squid: rados/test_crash.sh: add PG_DEGRADED to ignorelist

Merge pull request #62402 from adamemerson/wip-70422-squid

squid: rgw: Try to handle unwatch errors sensibly

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

Merge pull request #61986 from zdover23/wip-doc-2025-02-25-backport-61984-to-squid

squid: doc/rgw: clarify path-style vs virtual-hosted-style access

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

Merge pull request #62540 from jbaergen-do/btree2-backport-upstream-squid

squid: Backport the hybrid_btree2 allocator and prereqs

Reviewed-by: Igor Fedotov <igor.fedotov@croit.io>

blk/kerneldevice: notify_all only required when discard_drain wait for condition

This will cause discard threads to wake up each other
if bdev_async_discard_threads >= 2.

Fixes: https://tracker.ceph.com/issues/67835
Signed-off-by: Yite Gu <yitegu0@gmail.com>
(cherry picked from commit cec2e4550fe520fa60445678934c04ba605913be)

Merge pull request #62143 from ifed01/wip-ifed-no-compression-plugin-reload-squid

squid: os/bluestore: do cache locally compressor engines ever used

Merge pull request #61854 from batrick/wip-69960-squid

squid: pybind/mgr/mgr_module: turn off all automatic transactions

Merge pull request #62221 from ifed01/wip-ifed-limit-discard-qlen-squid

squid: blk/KernelDevice: Introduce a cap on the number of pending discards

Merge pull request #60562 from rishabh-d-dave/wip-68678-squid

squid: mgr: allow disabling always-on modules

Merge pull request #62211 from NitzanMordhai/wip-70383-squid

squid: suites/rados: cache tier deprecated, no need to keep the tests for it

19.2.2

Signed-off-by: Ceph Release Team <ceph-maintainers@ceph.io>

PendingReleaseNote: add release note that mgr/volumes plugin can be...

disabled now.

Signed-off-by: Rishabh Dave <ridave@redhat.com>
(cherry picked from commit 3232f6b7af15c011774c3ff37c8a98e5cf7c4887)

qa/cephfs: test enable/disable of volumes plugin

Signed-off-by: Rishabh Dave <ridave@redhat.com>
(cherry picked from commit 0abc227131c3e16077013f25a7c0fcfe1b7285a1)

Conflicts:
qa/cephfs/overrides/ignorelist_health.yaml
- this file in main had more entries than on this (squid) branch,
  resulting in cherry-picking conflict.

qa/tasks/cephfs/test_admin.py
- this file in main had more tests and a new set of test adjacent to
  tests added by this patch-series, resulting in cherry-picking conflict.

mon/MgrMonitor: allow disabling always-on MGR modules

Add a new command ("ceph mgr module force disable <module>") that allows
forcibly disabling an always-on module. This command should ideally only
be used to for cluster recovery.

Fixes: https://tracker.ceph.com/issues/66005
Signed-off-by: Rishabh Dave <ridave@redhat.com>
(cherry picked from commit 9962772358048a98a6e871dccf1bfd0a15b4d791)

mon/MgrMontitor: add log entry to support debugging

It becomes easier to understand logs when whether command succeeded or
not are also logged.

Signed-off-by: Rishabh Dave <ridave@redhat.com>
(cherry picked from commit 4374ac74d283ce1c50616330a0127a53623d96c3)

mon/MgrMonitor: improve a log message

Following log message has 3 distinct information (enabled modules,
modules that are alwats on and total number of commands enabled) printed
on the same line which makes it hard to find one of the information and
also makes it comparatively hard to read -

2024-07-17T17:21:16.609+0530 7f534ae006c0  0 mon.a@0(leader).mgr e0 create_initial initial modules iostat,nfs,volumes, always on modules balancer,crash,devicehealth,orchestrator,pg_autoscaler,progress,status,telemetry, 50 commands

With this commit it would be easier to read and find needed info -

2024-07-17T17:21:16.609+0530 7f534ae006c0  0 mon.a@0(leader).mgr e0 create_initial initial modules: iostat,nfs,volumes
2024-07-17T17:21:16.609+0530 7f534ae006c0  0 mon.a@0(leader).mgr e0 create_initial always on modules: balancer,crash,devicehealth,orchestrator,pg_autoscaler,progress,status,telemetry
2024-07-17T17:21:16.609+0530 7f534ae006c0  0 mon.a@0(leader).mgr e0 create_initial 50 commands

Signed-off-by: Rishabh Dave <ridave@redhat.com>
(cherry picked from commit e830f89240e382fdf5da676437ff783c4ea6ae42)

Merge pull request #62270 from cybozu/wip-70445-squid

squid: debian: add ceph-exporter package

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

mgr/dashboard: Fix empty ceph version in GET api/hosts

Fixes https://tracker.ceph.com/issues/70821

Due to the pagination the host list is being fetched from orchestrator which caused a regression as via orchestrator list ceph version is always marked empty.
Caused by https://github.com/ceph/ceph/pull/52154

Also fixed tests , as the new version addition causing whole json object mock to fail in tests

Signed-off-by: Afreen Misbah <afreen@ibm.com>
(cherry picked from commit 1aa74cf3fce6394ad08e83e91a90923d983b03c5)

Merge pull request #61496 from ljflores/wip-69168-squid

squid: qa/workunits/mon: ensure election strategy is "connectivity" for stretch mode
Reviewed-by: Kamoltat Sirivadhna <ksirivad@redhat.com>

rgw: remove keep_tail from RGWObjState

Signed-off-by: Jane Zhu <jzhu116@bloomberg.net>
(cherry picked from commit fd76b6466c298121994ba65cce3c0e76f8568841)

Conflicts:
src/rgw/rgw_sal_store.h RGWObjState is in rgw_sal.h on squid

rgw: keep the tails when copying object to itself

Signed-off-by: Jane Zhu <jzhu116@bloomberg.net>
(cherry picked from commit 333e4a9b0de745cf5be40c5f6c32df7a340b007a)

rgw: remove keep_tail from RGWObjState

Signed-off-by: Jane Zhu <jzhu116@bloomberg.net>
(cherry picked from commit fd76b6466c298121994ba65cce3c0e76f8568841)

Conflicts:
src/rgw/rgw_sal_store.h RGWObjState is in rgw_sal.h on squid

rgw: keep the tails when copying object to itself

Signed-off-by: Jane Zhu <jzhu116@bloomberg.net>
(cherry picked from commit 333e4a9b0de745cf5be40c5f6c32df7a340b007a)

Merge pull request #62625 from aaSharma14/wip-70763-squid

squid: monitoring: Fix OSDs panel in host-details grafana dashboard

Reviewed-by: Pedro Gonzalez Gomez <pegonzal@redhat.com>

Merge pull request #62694 from zdover23/wip-doc-2025-04-07-backport-62687-to-squid

squid: doc/cephadm/services: improve rgw.rst and snmp-gateway.rst

doc/cephadm/services: improve rgw.rst and snmp-gateway.rst

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit 3d18222b46d65784335b4af0d651118b5e588d5e)

Merge pull request #62213 from NitzanMordhai/wip-70380-squid

squid: LogMonitor: set no_reply for forward MLog commands

Merge pull request #62123 from ifed01/wip-ifed-some-recover-stuff-squi

squid: tools/ceph-objectstore-tool: tricks to tolerate disk errors for "pg export" command

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

Merge pull request #62053 from k0ste/wip-67124-squid

squid: os/bluestore: Fix ExtentDecoderPartial::_consume_new_blob

Reviewed-by: Igor Fedotov <igor.fedtovo@croit.io>

rgw: fix stale entries in bucket indexes

If rados_osd_op_timeout is set, the primary osd is slow, the rgw_rados_operate for deleting the rgw head obj may return -ETIMEDOUT
rgw can't determine whether or not the delete succeeded, we shouldn't be calling index_op.complete_del or cancel()
Instead, we should leave that pending entry in the index so than bucket listing can recover with check_disk_state() and cls_rgw_suggest_changens()
When raced with another delete op, deleting the rgw head obj may return ENOENT, calling index_op.complete_del() instead of index_op.cancel()

Fixes: https://tracker.ceph.com/issues/58965
Signed-off-by: Shasha Lu <lu.shasha@aishu.cn>
(cherry picked from commit de9ce1e27322cdf5c1572a0b8eff4a33547757c0)

os/bluestore: Fix ExtentDecoderPartial::_consume_new_blob

In case of corrupted data invalid iterator could be dereferenced.
Fixes: https://tracker.ceph.com/issues/66361
Signed-off-by: Adam Kupczyk <akupczyk@ibm.com>
(cherry picked from commit e59495b331765f4081d5aab66c939ec10b4b8344)

Merge pull request #62666 from zdover23/wip-doc-2025-04-04-backport-62661-to-squid

squid: doc/radosgw: Improve cloud-restore and cloud-transition

doc/radosgw: Improve cloud-restore and cloud-transition

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit 6a6807470850ea972c5c3786be3735d26875b221)

Merge pull request #62644 from zdover23/wip-doc-2025-04-03-backport-62643-to-squid

squid: doc/cephadm: s/confg/config/

doc/cephadm: s/confg/config/

Correct a misspelled command.

Re: https://github.com/ceph/ceph/pull/62367#discussion_r2025832937

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 0f4d9a9ea29f12333609045928499963b37ba5f3)

Merge pull request #62504 from batrick/wip-70667-squid

squid: mgr: add status command

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

Merge pull request #62620 from zdover23/wip-doc-2025-04-02-backport-62617-to-squid

squid: doc/rados/configuration: Correct admonition in ceph-conf.rst

Merge pull request #62605 from cbodley/wip-70753-squid

squid: cephfs-top, qa: Remove unnecessary global statements in tests

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

Merge pull request #62420 from clwluvw/bucket-creation-multi-zg

squid: rgw: bucket creation fixes for multi-zonegroup scenario

Merge pull request #62310 from adamemerson/wip-69124-squid

squid: rgw/multisite: fix forwarded requests for tenanted buckets

Merge pull request #62305 from adamemerson/wip-65473-squid

squid: rgw: exclude logging of request payer for 403 requests

monitoring: Fix OSDs panel in host-details grafana dashboard

OSDs panel in host-details grafana dashboard shows total of all OSDs
across all hosts even if a particular host is selected from the
ceph_hosts filter. This PR intends to fix this issue

Fixes: https://tracker.ceph.com/issues/70226
Signed-off-by: Aashish Sharnma <aasharma@redhat.com>
(cherry picked from commit 223c1363791981a35bcb3cac841e665e080cc1f5)

mgr/dashboard: enable ha by default on subsystem POST API

Fixes: https://tracker.ceph.com/issues/70745
Signed-off-by: Nizamudeen A <nia@redhat.com>
(cherry picked from commit 4ae817f703560232e96c0d58ae76a9e194e20996)