Merge pull request #62725 from VallariAg/nvmeof-teuthology-fio

qa/suites/nvmeof: Fix thrasher and fio script

Merge pull request #60731 from joscollin/wip-B68954-check-headers-journal-recovery

cephfs-journal-tool: check the headers in dump file after journal recovery

Reviewed-by: Dhairya Parmar <dparmar@redhat.com>
Reviewed-by: Rishabh Dave <ridave@redhat.com>

Merge pull request #62914 from baum/ms_dispatch2_clean_up

src/nvmeof/NVMeofGwMonitorClient.cc: ms_dispatch2 clean up

Merge pull request #62696 from anthonyeleven/mgr-prom

doc/mgr: Improve prometheus.rst

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

Merge PR #62577 into main

* refs/pull/62577/head:
libcephfs_proxy: avoid libc buffering for logging

Reviewed-by: Venky Shankar <vshankar@redhat.com>
Reviewed-by: Anoop C S <anoopcs@cryptolab.net>

Merge branch 'main' into mgr-prom

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

qa: test 'journal import' recognizes invalid headers post journal recovery

Fixes: https://tracker.ceph.com/issues/68954
Signed-off-by: Jos Collin <jcollin@redhat.com>

cephfs-journal-tool: check the headers in dump file after journal recovery

Fixes: https://tracker.ceph.com/issues/68954
Signed-off-by: Jos Collin <jcollin@redhat.com>

doc/mgr: Improve prometheus.rst

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

Merge pull request #62911 from bluikko/doc-cleanup-radosgw

doc/radosgw: Fix indentation in admin.rst

Merge pull request #62896 from zdover23/wip-doc-2025-04-21-revert-62782-c4f0f8e

doc: Revert "doc/mgr: Promptify CLI commands and other formatting fixes"

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

Merge pull request #62416 from kamoltat/wip-ksirivad-fix-connection-score

Merge pull request #61212 from rishabh-d-dave/mgr-vol-count-clones

mgr/vol: count number of ongoing clones in CloneProgressReporter...

Reviewed-by: Milind Changire <mchangir@redhat.com>

Merge pull request #62870 from MaxKellermann/mds_includes

mds: include cleanup

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

Merge pull request #62708 from rishabh-d-dave/vols-snap-path

mgr/vol: add command to get snapshot path

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

Merge pull request #62837 from athanatos/sjust/wip-crimson-stuck-backfilling

crimson: fix several bugs causing stuck backfills

Reviewed-by: Matan Breizman <mbreizma@redhat.com>

Merge pull request #62619 from athanatos/sjust/wip-replica-read-crimson-mosdpct

crimson: add MOSDPGPCT support

Reviewed-by: Matan Breizman <mbreizma@redhat.com>

src/nvmeof/NVMeofGwMonitorClient.cc: ms_dispatch2 clean up

- return ACKNOWLEDGED/HANDLED
- remove registration for unwanted keys

Signed-off-by: Alexander Indenbaum <aindenba@redhat.com>

Merge pull request #62899 from tchaikov/cmake-build-boost

cmake: Fix b2 build with postfixed compiler versions

Reviewed-by: Matan Breizman <mbreizma@redhat.com>

doc/radosgw: Fix indentation in admin.rst

Indent the CLI command continuation lines correctly to start at the same
position as the other such commands, add one space on 2 lines.
Introduced in #62877.

Signed-off-by: Ville Ojamo <14869000+bluikko@users.noreply.github.com>

Merge pull request #61997 from Naveenaidu/wip-naveen-telemetry-show-labeled-perf-counters

telemetry: include labeled perf counters in report

Reviewed-by: Ronen Friedman <rfriedma@redhat.com>
Reviewed-by: Yaarit Hatuka <yaarithatuka@gmail.com>
Reviewed-by: Afreen Misbah <afreen@ibm.com>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>

Merge pull request #62712 from leonidc/heuristic-redeploy-fix

nvmeofgw: fix host issue during redeploy, improves previous redeploy fix

Merge PR #62578 into main

* refs/pull/62578/head:
mds: fix dump stray command

Reviewed-by: Laura Flores <lflores@redhat.com>
Reviewed-by: Venky Shankar <vshankar@redhat.com>
Reviewed-by: Dhairya Parmar <dparmar@redhat.com>

Merge PR #62674 into main

* refs/pull/62674/head:
mon: Fix cast warning

Reviewed-by: Kefu Chai <tchaikov@gmail.com>
Reviewed-by: Venky Shankar <vshankar@redhat.com>

Merge pull request #62877 from bluikko/doc-formatting-radosgw

doc/radosgw: Improve and more consistent formatting

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

doc: Revert "doc/mgr: Promptify CLI commands and other formatting fixes"

This reverts commit c4f0f8edad46c852961a622795d14b735f660d94.

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

Merge pull request #62869 from afreen23/wip-nvme

mgr/dashboard: Fix pool update on edit

Reviewed-by: Nizamudeen A <nia@redhat.com>

Merge pull request #62900 from ceph/revert-61786-ns-create-size-fix

Revert "mgr/dashboard: Ns create size fix"

cmake: Fix b2 build with postfixed compiler versions

Previously, the build process used `bootstrap.sh` to build the b2 tool,
which automatically selected the compiler based on the specified toolset.
This failed when the compiler executable had a version postfix (e.g.,
/usr/bin/clang++-19) without a symlink at the expected name, producing
errors like:

```
A C++11 capable compiler is required for building the B2 engine.
Toolset 'clang' does not appear to support C++11.

> clang++ -x c++ -std=c++11 -pthread  check_clib.cpp check_cxx11.cpp
./tools/build/src/engine/build.sh: 120: clang++: not found
> clang++ -x c++ -std=c++11  check_clib.cpp check_cxx11.cpp
./tools/build/src/engine/build.sh: 120: clang++: not found

** Note, the C++11 capable compiler is _only_ required for building the B2
** engine. The B2 build system allows for using any C++ level and any other
** supported language and resource in your projects.

You can specify the toolset as the argument, i.e.:
    ./build.sh [options] gcc
```

The issue occurred because `bootstrap.sh` hardcodes the compiler name
based on the toolset (e.g., `clang++` for Clang) without supporting
postfixed versions.

This commit replaces the `bootstrap.sh` approach with an explicit build
command using Boost's `build.sh` script. We now:
1. Directly specify the full compiler path from CMake variables
2. Manually configure the build with `--cxx=...` and `--toolset=...`
3. Avoid reliance on symlinks or `bootstrap.sh`'s internal detection

This ensures the B2 engine is always built with the user-specified
compiler, even when installed with version postfixes.

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

Merge pull request #62728 from adk3798/cephadm-core-status-update-no-cinfo

cephadm: still set keep_container_info key in CoreStatusUpdater when cinfo is None

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

Merge pull request #62660 from Kushal-deb/reset-config-for-grafana

cephadm: Add pre_remove logic to clean up configs for grafana

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

Merge pull request #61487 from adk3798/cephadm-rgw-exit-timeout-secs

mgr/cephadm: allow setting up RGW delaying shutdown to complete client connections

Reviewed-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

Merge pull request #62652 from shraddhaag/wip-shraddhaag-fix-pg-degraded-warning

qa/suites/orch/cephadm: add PG_DEGRADED to ignorelist

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

Merge pull request #62681 from ShwetaBhosale1/fix_issue_70800_moving_last_host_having_nvmeof_to_maintainance_mode_should_not_be_allowed

mgr/cephadm: Give warning for moving last nvmeof host to maintainane mode

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

Revert "mgr/dashboard: Ns create size fix"

Signed-off-by: Tomer Haskalovitch <il033030@Tomers-MBP.lan>

qa/suites/nvmeof: add OSD warnings in ignorelist

And these to ignorelist:
- OSD_DOWN
- OSD_HOST_DOWN
- OSD_ROOT_DOWN
- PG_DEGRADED

These warnings come up in "cephadm" task during OSD
setup and are unrelated to nvmeof testing.

Fixes: https://tracker.ceph.com/issues/70906
Signed-off-by: Vallari Agrawal <vallari.agrawal@ibm.com>

osd/pg_features: advertise PG_FEATURE_PCT for crimson

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

crimson: wire up crimson pct timer and message

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

crimson/osd/pg: call PGBackend::on_actingset_changed in PG::stop

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

crimson: remove unused OperationThrottler::with_throttle

It's now unused, and these combinator style wrappers are easy to
misuse and hard to read compared with RAII and coroutines.

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

crimson: use make_interruptible, coroutine, and RAII releaser for recover_object_with_throttle

791772f1c used with_throttle here in a way which caused
then_interruptible in PGRecovery::recover_object to be called outside of
an interruptible context.

Instead of using a wrapper taking a lambda, rephrase as an RAII releaser
suitable for use in a coroutine.  This avoids needing to structure
with_throttle to deal correctly with both interruptible and
non-interruptible contexts.

Fixes: https://tracker.ceph.com/issues/70939
Signed-off-by: Samuel Just <sjust@redhat.com>

crimson: update PGRecovery::recover_object_with_throttle to use logging macros

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

crimson/.../osd_operation: remove max_in_progress shortcut in with_throttle

This made sense prior to the addition of the scheduler.  Now, we need
to go through the scheduler whether there's a max or not.

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

crimson/osd/osd_operation: remove unused with_throttle[_while] variants

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

crimson/osd/recovery_backend: correctly handle missing object in scan_for_backfill

load_and_lock can return enoent if object is missing.

Fixes: https://tracker.ceph.com/issues/70936
Signed-off-by: Samuel Just <sjust@redhat.com>

crimson/osd/recovery_backend: do not resolve_clone in scan_for_backfill

We're looking for the exact hobject_t specified.

Fixes: https://tracker.ceph.com/issues/70935
Signed-off-by: Samuel Just <sjust@redhat.com>

Merge pull request #62861 from cloudbehl/nfs-ip-port-fix

mgr/dashboard: Fix virtual IP:port in NFS list

Reviewed-by: Afreen Misbah <afreen@ibm.com>
Reviewed-by: Nizamudeen A <nia@redhat.com>

Merge pull request #62733 from adk3798/cephadm-rm-cluster-zap-devices-fix

cephadm: check "ceph_device_lvm" field instead of "ceph_device" during zap

Reviewed-by: Guillaume Abrioux <gabrioux@ibm.com>
Reviewed-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

Merge pull request #62853 from ceph/wip-yuriw-18.2.6-release.notes-main

doc: reef 18.2.6 release notes

Reviewed-by: Laura Flores <lflores@redhat.com>
Reviewed-by: Adam King adking@redhat.com
Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

Merge pull request #62788 from phlogistonjohn/jjm-defence

mgr/cephadm: do not delete smb fs cephx keys

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

add qa/suites/nvmeof/thrash/gateway-initiator-setup/16-subsys-4-namespace.yaml

Signed-off-by: Vallari Agrawal <vallari.agrawal@ibm.com>

qa/suites/nvmeof/thrash: Run fio for 20mins

Instead of 30mins.

Signed-off-by: Vallari Agrawal <vallari.agrawal@ibm.com>

qa/tasks/nvmeof.py: set default daemon_max_thrash_times to 4

Signed-off-by: Vallari Agrawal <vallari.agrawal@ibm.com>

qa/tasks/nvmeof.py: force restart nvmeof daemons

We need to add "--force" to "ceph orch daemon restart"
command after merge of PR https://github.com/ceph/ceph/pull/61925

Signed-off-by: Vallari Agrawal <vallari.agrawal@ibm.com>

qa/workunits/nvmeof/fio_test.sh: improve fio command

Signed-off-by: Vallari Agrawal <vallari.agrawal@ibm.com>

qa/workunit/nvmeof/fio: log when fio fails

Log using "[nvmeof.fio]" when fio fails.
This helps to grep "[nvmeof." to see all
test events.

Signed-off-by: Vallari Agrawal <vallari.agrawal@ibm.com>

qa/workunits/nvmeof/fio_test: add "nvme list-subsys"

This would help us to see connected devices
and subsystems.

Signed-off-by: Vallari Agrawal <vallari.agrawal@ibm.com>

Merge pull request #62782 from bluikko/doc-prompts-mgr

doc/mgr: Promptify CLI commands and other formatting fixes

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

Merge pull request #62794 from afreen23/main

mgr/dashboard: refactor about.component.ts

Reviewed-by: Nizamudeen A <nia@redhat.com>
Reviewed-by: Ankush Behl <cloudbehl@gmail.com>

Merge pull request #62879 from tchaikov/pkgutil.find_loader

pybind: switch from pkgutil.find_loader() to importlib.util.find_spec()

Reviewed-by: John Mulligan <jmulligan@redhat.com>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>

pybind: switch from pkgutil.find_loader() to importlib.util.find_spec()

Replace pkgutil.find_loader() with importlib.util.find_spec() throughout
Python bindings. This addresses the deprecation warning in Python 3.10
(scheduled for removal in 3.14) that appeared when generating librbd
Python bindings.

The importlib.util.find_spec() API has been available since Python 3.4
and is compatible with our minimum required Python version (3.9, since
commit 51f71fc1).

The warning resolved:
```
/home/kefu/dev/ceph/src/pybind/rbd/setup.py:8: DeprecationWarning: 'pkgutil.find_loader' is deprecated and slated for removal in Python 3.14; use importlib.util.find_spec() instead
  if not pkgutil.find_loader('setuptools'):
```

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

Merge pull request #62736 from adamemerson/wip-unbreak-d4n

rgw: Unbreak D4N

Reviewed-by: Pritha Srivastava <prsrivas@redhat.com>
Reviewed-by: Samarah Uriarte <samarah.uriarte@ibm.com>

Merge pull request #62850 from ivancich/wip-shard-report-crash

rgw: prevent crash in `radosgw-admin bucket object shard ...`

Reviewed-by: Casey Bodley <cbodley@redhat.com>
Reviewed-by: Jesse Williamson <jfw@ibm.com>

Merge pull request #62871 from cbodley/wip-doc-rgw-user-admin-system

doc/rgw: document Admin and System Users

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

Merge pull request #61486 from cbodley/wip-qa-rgw-s3tests-logging

qa/rgw: run-s3tests.sh disables bucket_logging tests

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

Merge pull request #62497 from bill-scales/ec_rollback_changes

osd: EC optimizations: changes to rollback to support partial writes

Merge pull request #57740 from bill-scales/wip-bscales-align-write-buffers-v2

os/transaction: page align write data buffers to improve performance - version 2

doc/rgw: document Admin and System Users

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

rgw: prevent crash in `radosgw-admin bucket object shard ...`

This subcommand is used to ask radosgw-admin which bucket index shard
a given object in a given bucket would have its bucket index entry
on. The user is required to supply the number of shards (i.e., the
command doesn't look that up). If 0 is provided it would result in a
divide by zero runtime exception. Values less than or equal to zero
are now protected.

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

Merge pull request #62803 from ShwetaBhosale1/fix_issue_70870_update_log_message_for_enabling_logrotate.timer

cephadm: Updated log message for enabling logrotate.timer service while bootstrap

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

doc/radosgw: Improve and more consistent formatting

Use inline code formatting consistently for command
line switches, data, hostnames, etc.
Correctly indent text and child lists in list items.
Remove a mid-sentence double spaces.
Capitalize "RGW" and "API" in text.
Remove unordered lists that are just regular text
everywhere else.
Use correct prompt # instead of $ for privileged
commands.
Use line continuation for multi-line example commands
instead of render them incorrectly as separate
single-line commands.
Use Title Case in few section header text that
missed it.
multisite.rst: Don't repeat "(RGW)" after "RADOS
Gateway" beyond the first instance in the same
paragraph.
multisite.rst: Change one "multisite" to "multi-site"
because all other instances use this spelling (EXCEPT
the title of the document??).
multisite.rst: Fix indentation of continuation lines in
prompted example commands.
Use pre-formatted block, as seen elsewhere in docs,
instead of strange unordered list plus inline code for
syntax example.
Add space before backslash for multi-line command
continuation.

Signed-off-by: Ville Ojamo <14869000+bluikko@users.noreply.github.com>

Merge pull request #61800 from MaxKellermann/blk_includes

blk: add missing includes

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

Merge pull request #62849 from VinayBhaskar-V/WIP-instancereplayer-crash

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

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

mgr/dashboard: Fix virtual IP:port in NFS list

fixes:https://tracker.ceph.com/issues/70964

Signed-off-by: Ankush Behl <cloudbehl@gmail.com>

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

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

mds/MDSMap: un-inline methods to reduce compile times

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

Merge pull request #62819 from pritha-srivastava/wip-rgw-iam-oidc-caps

rgw/iam: correcting the caps for OIDC Provider for a user.

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

doc: reef 18.2.6 release notes

Fixes https://tracker.ceph.com/issues/70938
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>

Merge pull request #62300 from rishabh-d-dave/mgr-vol-link-docs

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

Reviewed-by: Zac Dover <zac.dover@gmail.com>
Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>
Reviewed-by: Venky Shankar <vshankar@redhat.com>

mds/*: forward-declare MDSContext

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/MD{Cache,SRank}: move classes to Retry{Message,Request}.h

Prepare for eliminating the header dependency on MDSContext.h.

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/cephfs_features.h: do not include cephfs/metrics/Types.h

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/Server: forward-declare types from CInode.h

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

Merge pull request #62766 from adamemerson/wip-70864

rgw: Don't print spurious error on `radosgw-admin` exit

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

mds/Server: add missing includes

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

Merge pull request #62793 from BBoozmen/wip-oozmen-70270

rgw: metadata and data sync notification to retry upon any failure case

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

mds/Locker: forward-declare types from CInode.h

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/Locker: add missing includes

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/{LocalLockC,SimpleLock}: un-inline methods using `MutationRef`

Commit e8bc28407117 added a forward declaration for `MutationRef` but
that doesn't work as long as the header constructs and destructs
instances, causing errors such as:

 build/debug/boost/include/boost/smart_ptr/intrusive_ptr.hpp: In instantiation of ‘boost::intrusive_ptr<T>::~intrusive_ptr() [with T = MutationImpl]’:
 src/mds/SimpleLock.h:424:5:   required from here
   424 |     ceph_assert(!get_xlock_by());
       |                  ~~~~~~~~~~~~^~
 build/debug/boost/include/boost/smart_ptr/intrusive_ptr.hpp:100:44: error: ‘intrusive_ptr_release’ was not declared in this scope; did you mean ‘ceph::common::intrusive_ptr_release’?
   100 |         if( px != 0 ) intrusive_ptr_release( px );
       |                       ~~~~~~~~~~~~~~~~~~~~~^~~~~~
       |                       ceph::common::intrusive_ptr_release

This never occurred previously because `Mutation.h` happened to be
already included by somebody else.

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/CInode: remove the default value from will_block_for_quiesce()

Defaulting a `MDRequestRef` parameter works only if the type is fully
defined and thus disallows forward declarations.

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/CInode: un-inline methods to reduce header dependencies

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/flock: un-inline methods to reduce header dependencies

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/SessionMap: un-inline methods to reduce header dependencies

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/CDentry: include cleanup

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/CInode: include cleanup

Un-inline the destructor to eliminate a few more dependencies from the
public header.

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/SimpleLock: add missing includes

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mgr/dashboard: Fix pool update on edit

Issue: Pool field was blank on editing namespace form and user needs to type out pool
Reason: `image` field is no longer supported in form yet edit function trying to fetch it and failing on that, hence no following updates for pool
Fix: Removed stale `image` field

Additional changes:
- included unit tests for edit to capture such errors
- enhanced unit tests to sue ActivatedRouteStub and `router.url`
- pre populating pool form on create with first rbd pool in the list

Fixes https://tracker.ceph.com/issues/70966

Regression by: https://tracker.ceph.com/issues/69900

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

mds/Mutation: un-inline print() to resolve circular dependency

Mutation.h and SimpleLock.h have a circular dependency and we have to
resolve it using forward-declarations.  To make that possible, move
print() to Mutation.cc.

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/Mutation: add missing includes

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>

mds/OpenFileTable: include cleanup

Signed-off-by: Max Kellermann <max.kellermann@ionos.com>