Merge branch 'jjm-qa-workunits-smb' of https://github.com/phlogistonjohn/ceph into wip-phlogistonjohn-testing-2025-12-01-1454

ARCHS: x86_64

qa/orch/cephadm/smb: add new smb.workunit task (non-AD)

Add some instances of the new smb workunit wrapper to a couple of
non-AD tests that require the user name and password to be explicitly
provided.

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

qa/orch/cephadm/smb: add new smb.workunit tasks to some AD tests

These tests will automatically use a user account from the known
AD environment.

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

qa/tasks: add a workunit wrapper function to smb.py

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

qa/workunits/smb: add test_simple_smb.py very simple smb tests

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

qa/workunits/smb: add conftest.py for pytest fixtures, etc.

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

qa/workunits/smb: add smbutil.py a testing utility module

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

qa/workunits/smb: add a stub pytest config

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

qa/workunits/smb: add a stub workunit script for smb testing

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

Merge pull request #66431 from cbodley/wip-doc-release-os-recommendations

doc: add Tentacle to os recommendations

Reviewed-by: Ville Ojamo <14869000+bluikko@users.noreply.github.com>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>

doc: remove redundant note about tested container hosts

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

doc: remove old distros from os recommendations

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

Merge pull request #63794 from clwluvw/enc-copy

rgw: implement CopyObject for encrypted objects

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

Merge pull request #66410 from aclamk/aclamk-encode-fix-debug-macro

encode: Fix bad use of DENC_DUMP_PRE

Merge pull request #66407 from rhcs-dashboard/redirect-other-propery-name

mgr/dashboard: support custom prop for table item redirection

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

Merge pull request #66437 from rhcs-dashboard/fix-74008-main

mgr/dashboard: fix multi-cluster route reload logic

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

Merge pull request #66346 from rhcs-dashboard/fix-service-form

mgr/dashboard: service creation fails if service name is same as sevice type

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

Merge pull request #66436 from rhcs-dashboard/add-sagar--to-mailmap-githubmap-organizationmap

add Sagar Gopale to githubmap mailmap organizationmap

Reviewed-by: Afreen Misbah <afreen@ibm.com>
Reviewed-by: Abhishek Desai <abhishek.desai1@ibm.com>

Merge pull request #65433 from mohit84/repeer_on_acting

test: repeer_on_down_acting_member_coming_back is continuously failing

Merge pull request #66417 from rhcs-dashboard/fix-server-sort

mgr/dashboard: fix server side table sort

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

Merge pull request #66375 from ronen-fr/wip-rf-eofread

osd/scrub: do not attempt to read past the end of an object

Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Alex Ainscow <aainscow@uk.ibm.com>

mgr/dashboard: support custom prop for table item redirection

use an extra customTemplateConfig called `customRowProperty` where
you can provide the key of the property you wish to route, instead of
relying on the cell's prop itself

Fixes: https://tracker.ceph.com/issues/73989
Signed-off-by: Nizamudeen A <nia@redhat.com>

Merge pull request #66330 from rhcs-dashboard/edit-realm

mgr/dashboard: edit realm modal not working

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

Merge pull request #66321 from rhcs-dashboard/73832-navigation-e2e-failure

mgr/dasboard : Fixes navigation e2e test

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

Merge pull request #66425 from baum/rbd_with_crc32c_nvmeof_service_spec

mgr/cephadm: add rbd_with_crc32c parameter to nvmeof service spec

qa/standalone: osd-scrub-dump.sh: additional tests

Also - adding stride override parameter to the existing
large-object scrub test. And: improving the output format
of the test results.

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

osd/scrub: extract the 'read object data' functionality

from be_deep_scrub() into a separate function,
ReplicatedBackend::be_deep_scrub_read().

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

osd/scrub: do not attempt to read past the end of an object

When performing deep scrubs, the scrubber reads object data
in strides. Existing code uses a short read to detect the end
of the object (and if the object size is a multiple of the
stride - an extra read is performed, which returns 0 bytes).

The proposed change is to avoid such extra read attempts,
by using our knowledge of the object size.

Also - some minor code cleanups in the relevant function.

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

osd: provide fmtlib formatter for ScrubMapBuilder

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

Merge pull request #66230 from ronen-fr/wip-rf-larger-stride

osd/scrub: increasing the default data-read stride

Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>
Reviewed-by: Alex Ainscow <aainscow@uk.ibm.com>
Reviewed-by: Jon Bailey <jonathan.bailey1@ibm.com>

mgr/cephadm: add rbd_with_crc32c parameter to nvmeof service spec

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

Merge pull request #64248 from zhuwei127/fix-doublefree

examples/librados: fix memory pointed to by 'rs' is freed twice.

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

Merge pull request #65771 from aainscow/ec_direct_reads_pr_1

EC Direct Reads: First PR, background work

Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Bill Scales <bill_scales@uk.ibm.com>

Merge pull request #66377 from baum/rbd_aio_write_with_crc32c_initial_fix

librbd: rbd_aio_write_with_crc32c store CRC32C with initial value -1 to match msgr2 validation

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

mgr/dashboard: fix multi-cluster route reload logic

Issue: Route was being force-reloaded using a two-step navigation hack causing unnecessary redirects and side effects.
Fix: Replaced the hack with Angular’s native same-URL reload using onSameUrlNavigation: 'reload' for a clean, stable route refresh.

Fixes: https://tracker.ceph.com/issues/74008
Signed-off-by: Aashish Sharma <aasharma@redhat.com>

Merge pull request #65739 from tchaikov/rgw-gap-list-manpage

debian: include rgw-gap-list manpage and rgw-policy-check in ceph-common

Reviewed-by: J. Eric Ivancich <ivancich@redhat.com>
Reviewed-by: Matan Breizman <mbreizma@ibm.com>

Merge pull request #66166 from cbodley/wip-cmake-breakpad-arch

cmake: disable WITH_BREAKPAD on power arch

Reviewed-by: Kefu Chai <k.chai@proxmox.com>

doc: remove os recommendations for eol releases

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

doc/dev: add os-recommendations.rst to release checklist

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

doc: add Tentacle to os recommendations

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

doc: remove Octopus and Centos7 from os recommendations

cleanup to prepare for tentacle

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

Merge pull request #66416 from bluikko/doc-fscrypt-improvements-cephfs

doc/cephfs: Small improvements in fscrypt.rst

Merge pull request #66420 from bluikko/doc-sphinx-warnings-202511

doc: Fix Sphinx warnings

Merge pull request #66421 from bluikko/doc-sphinx-warning-tentacle-202511

doc/releases: Fix Sphinx warning in tentacle.rst

Merge pull request #66423 from bluikko/doc-sphinx-warning-theme-202511

doc: Fix Sphinx warning about theme option

common/options: document osd_deep_scrub_stride

default value change (from 0.5MB to 4MB)

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

osdc: Add SplitOp capability to Objecter

This will provide the ability for Objecter to split up
certain ops and distribute them to the OSDs directly if
that provides a preformance advantage.

This is experimental code and is switched off unless the
magic pool flags are enabled. These magic pool flags were
pushed in an earlier commit in the same PR.

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osdc: Fix minor typo

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osdc: Interface to allow split reads to copy op from client op to split op

When spliting ops, certain addition sub ops (e.g. get xattr) can be simply passed
through to the child op.

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osdc: Add stub for ability to force an op to always go to a particular shard

This will eventually be used by SplitIo to direct ops to the correct OSD.

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osdc: Split handle_osd_op_reply into two functions

The functionality is not altered by this commit.

In the future we want to post-process split-ios after
recombining the read data.

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osdc: Remove unused con parameter from Objecter::_calc_target()

This parameter is not used by the _calc_target code.  It is being
removed just to clean up the code, as we are making some changes
to _calc_target in later stages of the split io PR.

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osdc: Interface to submit IO with ASIO Post.

For direct read failures, the locking is such that we cannot
immediately send a new IO without deadlocking. This new interface
allows an op to be sent as an asio post.

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osd: Call clear_repop_obc for EC as well as Replica.

This function is necessary for balanced reads and as such is required for EC too.

Rename the function to make sense, given this change of purpose, but the
functionality does not change.

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osd: Implement sync reads and sparse reads for EC for direct reads

Sparse reads for EC are simple to implement, as the code is essentially
identical to that of replica, with some address translation.

When doing a direct read in EC, only a single OSD is involved and
that OSD, by definition is the only OSD involved. As such we can
do the more performant sync read, rather than async read.

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osd: Add extent_to_shard_extent interface to PGBackend.

This allows a backend to expose how an object offset/length translates to
an offset/length on a particular shard.

For Replica, this is trivial.

For EC, this means looking up the start and end offsets, then translating
this to shard address space.

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osd: Set the from shard in the EC read reply.

This was not necessary prior to direct reads, but is essential when the
client needs to know which shard the read came from.

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osd: Generalise can_serve_replica_read for consumption by EC.

The can_serve_replica_read() function is called by replica to determine whether there are
any uncommitted writes.  If such writes exist, then the system will reject the IO to avoid
the risk of reading data from a write which may yet be rolled back.

The same code is going to be useful for EC direct reads.

The string_view code is not expensive.

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osd: Create EC Direct Read flag and pass through to EC.

This is in preperation for supporting sparse and sync reads in EC.
Such ops will only be supported for "balance reads".

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osd: Replace unused EC offset translation function with useful one.

The old chunk_aligned_shard_offset_to_ro_offset was not only unused, it
didn't actually have the correct logic. We replace it here with similar,
but more useful function that will be used in sparse reads for EC

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

osd: Introduce pool flag for "split IO" and Plugin flag for "direct read"

These flags will currently behave as follows:

1. The pool flag is never set, unless by a user with the osd_pool_default_flags
   config option.
2. The pool flag will be removed for EC pools where the plugin does not support
   direct reads.
3. Replica pools will never remove the flag.

The intention is to eventually invert this logic and allow split IOs upon
upgrade to Umberella in this same function.

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

add Sagar Gopale to githubmap mailmap organizationmap
Signed-off-by: Sagar Gopale <sagar.gopale@ibm.com>

Merge pull request #66340 from imran-imtiaz/dashboard

mgr/dashboard: add GET API endpoint for consistency groups

encode: Fix bad use of DENC_DUMP_PRE

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

doc: Fix Sphinx warning about theme option

The Sphinx theme "sphinx_rtd_theme" dropped support for "display_version"
theme option in version 3 (currently used: 3.0.2).

Because the "ceph" theme inherits that theme, remove all references to
"display_version" from it.

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

librbd: store CRC32C with initial value -1 to match msgr2 validation

Fix runtime error, using test command:
   sudo dd if=/dev/zero bs=32k of=/dev/nvme0n1 count=1

The error log:
   2025-11-23T11:24:10.512+0000 7f30f4ec0640  1 --2- [v2:192.168.13.2:6802/3444906816,v1:192.168.13.2:6803/3444906816] >> 192.168.13.3:0/3916714748 conn(0x527d400 0x728f700 crc :-1 s=THROTTLE_DONE pgs=2038703 gs=2038723 cs=0 l=1 c_cookie=0 s_cookie=0 reconnecting=0 rev1=1 crypto rx=0 tx=0 comp rx=0 tx=0)._handle_read_frame_epilogue_main bad segment crc calculated=1136411986 expected=4294967295

Ceph msgr2 validation (ceph/src/msg/async/frames_v2.cc:47):
   uint32_t crc = segment_bl.crc32c(-1);  // Uses initial value -1

Co-authored-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Alexander Indenbaum <aindenba@redhat.com>

doc/release: Fix Sphinx warning in tentacle.rst

Add an empty line between blocks to fix a warning:

/home/docs/checkouts/readthedocs.org/user_builds/ceph/checkouts/66416/doc/releases/tentacle.rst:97: ERROR: Unexpected indentation.

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

doc: Fix Sphinx warnings

Fix section title underline lengths in dev/cephfs-fscrypt.rst
radosgw/adminops.rst.

Use "figure" keyword instead of "image" and use the caption feature in
dev/cephfs-fscrypt.rst.

Remove circular toc reference in dev/crimson/index.rst.

Add an empty line after block in
rados/troubleshooting/troubleshooting-pg.rst.

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

doc/cephfs: Small improvements in fscrypt.rst

Fix Sphinx warnings about section title underline lengths.
Use title case in section titles.

Change Unicode quotation marks to ASCII.

Use ordered list for lines that were supposedly intended to be a list.

Use double backticks for literals.
Use image caption formatting.

Remove unnecessary comma and other small language improvements.

Capitalize MDS, OSD, etc.

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

mgr/dashboard: fix server side table sort

show a loading screen when the sort is being performed through
server-side since the sort will happen a little slow

It will be more visible in bigger environments, and with test env if you
try to sort too many time in a short interval and you start to see some
inconsistencies. This is only there for tables like OSDs or hosts where
we have the server side rendering enabled

Fixes: https://tracker.ceph.com/issues/73994
Signed-off-by: Nizamudeen A <nia@redhat.com>

rgw: fix offset calculation in copy_obj_data

Set ofs to total bytes read by adding 1 to end offset.
Since 'end' represents the last byte offset (zero-indexed),
we need to add 1 to get the actual number of bytes copied.

Signed-off-by: Seena Fallah <seenafallah@gmail.com>

Merge pull request #66333 from shraddhaag/wip-shraddhaag-increase-reactors

qa/clusters/crimson: increase reactors count

cmake: disable WITH_BREAKPAD on power arch

Reported-by: T K Chandra Hasan <t.k.chandra.hasan@ibm.com>
Signed-off-by: Casey Bodley <cbodley@redhat.com>

cmake: use cmake_dependent_option for WITH_BREAKPAD

a bit simpler without the WITH_BREAKPAD_DEFAULT part, and causes the
WITH_BREAKPAD option to be hidden from cmake-gui on WIN32

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

Merge pull request #66336 from Matan-B/wip-matanb-crimson-snapmapper-osddriver

osd/SnapMapper: fix Crimson logs

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

Merge pull request #66332 from rhcs-dashboard/73854-CephFS-Authorize-modal-Update-issues

mgr/dashboard : fix - CephFS Authorize Modal Update issue

Reviewed-by: Dnyaneshwari Talwekar dtalweka@redhat.com

Merge pull request #66382 from bluikko/doc-mgmt-gateway-improvements-cephadm

doc/cephadm: Fix command plus improvements in services/mgmt-gateway.rst

Merge pull request #66006 from afreen23/carbonize-chnage-password

mgr/dashboard: Carbonize the Change Password Form

Reviewed-by: Nizamudeen A <nia@redhat.com>
Reviewed-by: Aashish Sharma <aasharma@redhat.com>
Reviewed-by: Dnyaneshwari Talwekar dtalweka@redhat.com

Merge pull request #66326 from afreen23/fixes-mixins

monitoring: Fixes for development

Reviewed-by: Nizamudeen A <nia@redhat.com>
Reviewed-by: Anthony D Atri <anthony.datri@gmail.com>

mgr/dashboard : fix - CephFS Authorize Modal Update issue
fixes : https://tracker.ceph.com/issues/73854
Signed-off-by: Devika Babrekar <devika.babrekar@ibm.com>

doc/cephadm: Fix command plus improvements in service/mgmt-gateway.rst

Remove double backticks from a CLI command.

Use bash prompt consistently for CLI command blocks.

Don't capitalize word in middle of sentence.

Talk about "admin" instead of "user", similarly to the last text
paragraph in the doc.

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

qa/clusters/crimson: increase reactors count

This commit increases the number of reactors in fixed-1 and
fixed-2 crimson clusters.

Signed-off-by: Shraddha Agrawal <shraddhaag@ibm.com>

mgr/dashboard: Carbonize the Change Password Form

Fixes https://tracker.ceph.com/issues/73193

-  using carbon based stylings, typography and components
-  used grid layout for form arrangement
-  breadcrumb is slightly off, which needs to be fixed by applying grid layout to the app shell

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

Merge pull request #66372 from tchaikov/wip-qa-encoder-exclude

qa/suites/rados/encoder: exclude ceph-osd-classic when installing LTS…

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

qa/suites/rados/encoder: exclude ceph-osd-* when installing LTS releases

In a37b5b5, the ceph-osd-classic and ceph-osd-crimson packages were
added to qa/packages/packages.yaml. The "install" task uses this file as
the default package list for all branches, including LTS releases like
Reef.

However, a37b5b5 only exists in the main branch and won't be backported
to LTS branches. This causes installation failures in the rados/encoder
test suite, which verifies forward compatibility by installing LTS
releases and testing whether they can decode the latest corpus.

Exclude ceph-osd-classic and ceph-osd-crimson from LTS installations to
ensure the test suite can successfully install ceph-dencoder, which is
required for the interoperability tests.

Fixes: https://tracker.ceph.com/issues/73957
Signed-off-by: Kefu Chai <k.chai@proxmox.com>

Merge pull request #66293 from anthonyeleven/instore.dbnoonecanhearyouscream

doc: Improve start/hardware-recommendations.rst

Merge pull request #65995 from pcuzner/rocksdb_compaction_metric

rados/osd: enable compact_running perfcounter at PRIO=5

Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Neha Ojha <nojha@ibm.com>

Merge pull request #65694 from mohit84/mclock_scheduler_monc

osd: Remove monc reference from scheduler

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

Merge pull request #65687 from pdvian/wip-73272-autoscaler

pybind/mgr/pg_autoscaler: Introduce dynamic threshold to improve scal…

Reviewed-by: Kamoltat (Junior) Sirivadhna <ksirivad@redhat.com>

Merge pull request #65698 from cbodley/wip-72771

osdc: Objecter::linger_by_cookie() for safe cast from uint64

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

Merge pull request #65615 from chungfengz-syno/fix-get_addr_from_invalid_rank

mon/Elector.cc: prevent assertion failure when receiving pings from r…

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

doc: Improve start/hardware-recommendations.rst

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

Merge pull request #66311 from rhcs-dashboard/fix-73910-main

monitoring: remove cephfs.libsonnet mention from dashboards.libsonnet

Reviewed-by: Afreen Misbah <afreen@ibm.com>
Reviewed-by: Ankush Behl <cloudbehl@gmail.com>

mgr/dashboard: edit realm modal not working

Fixes: https://tracker.ceph.com/issues/73937
Signed-off-by: Naman Munet <naman.munet@ibm.com>

Merge pull request #66319 from rhcs-dashboard/fix-account-group-mode

mgr/dashboard: rgw accounts form group mode disable option is not working

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

mgr/dashboard: service creation fails if service name is same as service type

Fixes: https://tracker.ceph.com/issues/73948
Signed-off-by: Naman Munet <naman.munet@ibm.com>

rgw: sanitize copy source attrs

Co-authored-by: Marcus Watts <mwatts@redhat.com>
Signed-off-by: Seena Fallah <seenafallah@gmail.com>

Merge pull request #66323 from aainscow/pg_repeer

mon: ceph pg repeer should propose a correctly sized pg temp.

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

Merge pull request #66229 from Matan-B/wip-matanb-crimson-on

ceph.spec.in: Include Crimson by default

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

mgr/dashboard: add GET API endpoint for consistency groups

Signed-off-by: Imran Imtiaz <imran.imtiaz@uk.ibm.com>
Fixes: https://tracker.ceph.com/issues/73942
Add a consistency group dashboard API endpoint to get the list of images
in the consistency groups that match the namespace of the group.