mgr/cephadm: Handle daemon colocation port for cluster qos port

Fixes: https://tracker.ceph.com/issues/69861
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/cephadm: Fixed scheduling test for nfs qos port

Fixes: https://tracker.ceph.com/issues/69861
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/cephadm: For colocating nfs daeoms, check daemon qos ports instead of fetching port from spec

Fixes: https://tracker.ceph.com/issues/69861
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: 1. Removed the option to enable and disable cluster wide qos, it will be enabled by default
2. Removed the cluster_enable_qos field from the cluster-level block as it was causing confusion for the user.
3. Instead of using cluster use global while showing cluster level qos values in export qos get

Fixes: https://tracker.ceph.com/issues/69861
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: Update cluster qos_type from int to str

Fixes: https://tracker.ceph.com/issues/69861
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/cephadm: support nfs cluster level qos
Added below CEPH_NODES_LIST block in ganesha.conf and enable_cluster_qos in cluster level QoS block
CEPH_NODES_LIST {
        Ceph_Nodes = 192.168.100.100, 192.168.100.101, 192.168.100.102;
}

Fixes: https://tracker.ceph.com/issues/69861
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>
mgr/cephadm: Changes to add NFS cluster qos inter node communication port in spec

mgr/nfs: Addressed review comments for cluster level qos support

mgr/nfs: add enable_cluster_qos = true while enabling qos

mgr/cephadm: NFSSpec changes to apply cluster level QOS while nfs service creation

Fixes: https://tracker.ceph.com/issues/69458
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: update QOS minimum BandWidth value and IOPS max value

Fixes: https://tracker.ceph.com/issues/69458
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: Changes for get QoS command to return bandwidth in GiB instead of GB

Fixes: https://tracker.ceph.com/issues/69458
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: Changed IOPS max limit to 409600 and Bandwidth max limit to 100GB

Fixes: https://tracker.ceph.com/issues/69458
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: Addressd review comments

Fixes: https://tracker.ceph.com/issues/69458
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

nfs/mgr: Updated logs and exception messages

fixes: https://tracker.ceph.com/issues/69458
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: Added skip_notify_nfs_server parameter to update export commands

fixes: https://tracker.ceph.com/issues/69458
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: When cluster level qos is disabled and export still has qos parameters, then allow nfs export apply command if file has same qos block which is already set

fixes: https://tracker.ceph.com/issues/69458
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: Add cluster level qos status to export get qos command

fixes: https://tracker.ceph.com/issues/69458
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: Updated get qos to return bandwidth in bytes as dashboard needs bw in bytes

Fixes: https://tracker.ceph.com/issues/69458
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: Updating QoS min and max limit of bandwidth and iops values

fixes: https://tracker.ceph.com/issues/69458
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: NFS cluster and export commands to enable and disable ops/s control

Fixes: https://tracker.ceph.com/issues/69861
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

mgr/nfs: NFS commands to enable, disable and get QOS config for cluster and export

fixes: https://tracker.ceph.com/issues/69458
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

Merge pull request #68407 from knrt10/crimson-rgw-lifecycle

qa: Port rgw lifecycle test to crimson-rados

Merge pull request #68618 from oduameh/wip-doc-nvmeof-subcommands

doc: document ceph nvmeof CLI subcommands for target configuration

Reviewed-by: Aviv Caro <Aviv.Caro@ibm.com>

Merge pull request #68338 from fivetime/ceph-volume-raw-osd-fsid

ceph-volume: add --osd-fsid support to raw mode prepare

Merge pull request #66586 from ShwetaBhosale1/execute_command_retry_logic_if_command_fails_with_connection_error

mgr/cephadm: Added retry logic for execute command if command fails with connection error

Reviewed-by: Adam King <adking@redhat.com>
Reviewed-by: Redouane Kachach <rkachach@ibm.com>

Merge pull request #67651 from ShwetaBhosale1/fix_issue_75189_nfs_rdma_support

mgr/cephadm: nfs rdma support

Reviewed-by: Redouane Kachach <rkachach@ibm.com>

Merge pull request #68152 from ShwetaBhosale1/fix_issue_75704_make_haproxy_peer_configuration_port_configurable

mgr/cephadm: Provide option to configure custom haproxy peer communication port

Reviewed-by: Kushal Deb <Kushal.Deb@ibm.com>

Merge pull request #67906 from ShwetaBhosale1/update_nfs

Bump NFS Version in Ceph

Reviewed-by: Shubha Jain <SHUBHA.JAIN1@ibm.com>

Merge pull request #68549 from afreen23/service-form-fix

Reviewed-by: Nizamudeen A <nia@redhat.com>
Reviewed-by: Sagar Gopale <sagar.gopale@ibm.com>

Merge pull request #66331 from ujjawal-ibm/daemon-upgrade-issue-with-space

mgr/cephadm: Handle whitespace in upgrade daemon types

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

Merge pull request #68164 from phlogistonjohn/jjm-cephadm-smb-deps-action

cephadm: add get_dependencies to smb service class

Reviewed-by: Redouane Kachach <rkachach@ibm.com>

Merge pull request #68608 from tchaikov/wip-minimal-release

doc/rados: document require_osd_release and require_min_compat_client

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>
Reviewed-by: Ville Ojamo <git2233+ceph@ojamo.eu>

qa: Port rgw lifecycle test to crimson-rados

This adds lifecycle qa tests of rgw to crimson-rados suite

Signed-off-by: Kautilya Tripathi <kautilya.tripathi@ibm.com>

Merge pull request #67372 from knrt10/wip-crimson-rgw-multifs

qa: Port rgw multifs test to crimson-rados

Merge pull request #68341 from ronen-fr/wip-rf-dev3-crimson

crimson/osd: move to a unified on-device super-block structure

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

doc/rados: document require_osd_release

Add a standalone concept page for the OSDMap require_osd_release field,
the upgrade-gate counterpart to require_min_compat_client. Cover:

 - how to set it and how to check it;
 - the full set of pre-commit guards the monitor runs, rendered as a
   table with each guard's error text and bypass status;
 - which commands and features become available as the flag is raised,
   per release;
 - the OSD boot window that refuses OSDs more than two releases ahead
   of the flag;
 - the OSD_UPGRADE_FINISHED health warning that prompts admins to set
   the flag after an upgrade;
 - the initial value on new clusters and the two mon_debug_* knobs
   that override it for testing.

Also cross-link the new page from the related-flags table on
require-min-compat-client.rst, and from the rados operations index.

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

doc/rados: document require_min_compat_client

Add a standalone concept page for the OSDMap require_min_compat_client
field, covering: how to set and check it, the non-monotonic lowering
behavior (with the features-in-use floor derived from
OSDMap::get_min_compat_client()), and the operator commands it gates.
Include tables for the floor-pinning features and the flag-gated
commands, so operators can reason about transitions without reading
OSDMonitor.cc.

Cross-reference to the CephFS per-filesystem required_client_features
mechanism, which is the MDSMap-side equivalent for client-protocol
features. Add an anchor on the existing CephFS Required Client Features
section so the cross-reference resolves.

Link the new page from the rados operations index.

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

Merge pull request #68394 from yuvalif/wip-yuval-75928

rgw/kafka: setting maximum batch size when sending messages

doc: document ceph nvmeof CLI subcommands for target configuration

Replaces verbose podman run container commands with native ceph nvmeof
CLI subcommands. The nvmeof-cli container approach is preserved as an
alternative in a note block, with a clarification that its option names
differ from the ceph nvmeof CLI.

Fixes: https://tracker.ceph.com/issues/76042
Signed-off-by: Emmanuel Ameh <oduameh@users.noreply.github.com>
Signed-off-by: oduameh <oduameh@users.noreply.github.com>

Merge pull request #68127 from Matan-B/wip-matanb-seastore-conflict-user-lba

crimson/os/seastore: Prepare no invalidation to LBA user txns

Reviewed-by: Xuehan Xu <xuxuehan@qianxin.com>

test/rgw/kafka: make batch size tests manual

currently failing in teuthology

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

rgw/kafka: setting maximum batch size when sending messages

Fixes: https://tracker.ceph.com/issues/75928
Signed-off-by: Yuval Lifshitz <ylifshit@ibm.com>

Merge PR #68582 into main

* refs/pull/68582/head:
script/ptl-tool: improve help messages during startup

Reviewed-by: John Mulligan <jmulligan@redhat.com>
Reviewed-by: Yaarit Hatuka <yaarithatuka@gmail.com>

Merge pull request #68606 from tchaikov/wip-doc-silence-warning

doc/scripts: use raw string for regex in gen_state_diagram.py

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

Merge pull request #68366 from ljflores/wip-tracker-74517

qa/suites/rados: temporarily disable ceph-post-file test

Reviewed-by: Sridhar Seshasayee <Sridhar.Seshasayee@ibm.com>

doc/scripts: use raw string for regex in gen_state_diagram.py

Python 3.12 emits SyntaxWarning for invalid escape sequences in ordinary
string literals. The re.search() call on line 162 was the only pattern
in the file passed as a non-raw string, causing:

  doc/scripts/gen_state_diagram.py:162: SyntaxWarning: invalid escape
    sequence '\s'
    i = re.search("return\s+transit<\s*(\w*)\s*>()", line)

Add the r"" prefix to match the other re.search / re.finditer / re.sub
call sites in the same file. No behavior change; \s was already being
interpreted as a regex whitespace class because Python leaves unknown
escapes untouched, but this will become a SyntaxError in a future
Python release.

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

Merge pull request #68576 from rhcs-dashboard/fix-76239-main

mr/dashboard: remove rgw_servers filter from radosgw-sync-overview grafana dashboard

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

Merge pull request #68483 from rhcs-dashboard/fix-smb-mgrdb

mgr: fix continous smb MgrDBNotReady

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

ceph-volume: add tests for raw prepare --osd-fsid

Add unit tests to cover the raw prepare help text for --osd-fsid,
assert generate_uuid is used when no osd_fsid is supplied and
assert an externally provided osd_fsid is passed through to
create_id without generating a new UUID.

Signed-off-by: Guillaume Abrioux <gabrioux@ibm.com>

ceph-volume: add --osd-fsid support to raw mode prepare

The LVM mode already supports --osd-fsid to allow external tools
(e.g., Kubernetes operators) to pre-register an OSD ID+UUID via
"ceph osd new" and then pass both to ceph-volume, ensuring the
operator retains full control of the OSD ID lifecycle and can
reliably clean up on prepare failure (no orphan OSDs).

The raw mode was missing this support: prepare() unconditionally
called system.generate_uuid(), ignoring any --osd-fsid value.
When an operator pre-registered osd.N with uuid_A and then ran
"ceph-volume raw prepare --osd-id N --dmcrypt", ceph-volume
generated uuid_B internally and called "ceph osd new uuid_B N",
which failed with EINVAL because the ID was already registered
with a different UUID.

This commit:
- Adds --osd-fsid argument to the raw mode argument parser
  (devices/raw/common.py), consistent with the LVM mode.
- Changes raw.prepare() to honor an externally provided osd_fsid,
  falling back to generate_uuid() only when none is given
  (objectstore/raw.py), consistent with the LVM mode.

Signed-off-by: Duanming Zhou <zhouduanming@gmail.com>

crimson/tools: add dump-superblock command to crimson_objectstore_tool

We now also check magic before attempting full decode
so that non-crimson or corrupt devices produce a clean
error instead of a decode exception.

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

crimson/tests: fix unitest-seastore-randomblock-manager

test following a change to the device header block.

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

crimson/osd: use a unified super-block for devices

This commit refactors the on-hardware super-block structure
used by the seastore to a unified format that
can accommodate all three device types (HDD, ZBD, RBM).

All devices now have a 60 bytes header at address 0,
similar to the existing BlueStore layout. A 23-byte magic
string ("CRIMSON_DEVICE") is placed at the beginning of
the header, followed by 37 bytes of null padding (to
match the existing 60 bytes of the super-block), and
then the DENC-encoded device_superblock_t structure starting
at offset 60.

A unified device_config_t is now used for all device types.

The per-shard data structure is also unified, now including a union
of all relevant fields for each device type.

We are also adding a check for the super-block magic value in the
RBMDevice::read_rbm_superblock() method, similar to the existing check
in SegmentManager::read_segment_manager_superblock().

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

Merge PR #68555 into main

* refs/pull/68555/head:
script/redmine-upkeep: resolve comment misformatting

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

script/ptl-tool: improve help messages during startup

For newer folks who've never run it before.

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

Merge pull request #68414 from guits/has_bs_repl

ceph-volume: has_bluestore_label checks all bluestore label replica o…

Merge pull request #66704 from ivancich/wip-document-object-manifest-cmd

rgw: add internal docs for `radosgw-admin object manifest ...` sub-command

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

mr/dashboard: remove rgw_servers filter from radosgw-sync-overview grafana dashboard

Fixes: https://tracker.ceph.com/issues/76239
Signed-off-by: Aashish Sharma <aasharma@redhat.com>

Merge pull request #66432 from stzuraski898/unittests-mgr

test/mgr: Improve unit-test coverage for ceph-mgr

Reviewed by: Nitzan Mordechai <nmordech@ibm.com>
Reviewed by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed by: Ronen Friedman <rfriedma@redhat.com>

Merge pull request #68472 from yuvalif/wip-yuval-76127

test/rgw/kafka: cleanup kafka tgz file

qa/cephadm: zap raw devices before OSD deployment

follow up on commit f3c938f5032885434dd2ab77f9b2e6bc05917f3c

```
2026-04-22T10:16:21.350 DEBUG:teuthology.orchestra.run.trial118:> sudo ceph-bluestore-tool zap-device --dev /dev/nvme3n1 --yes-i-really-really-mean-it
2026-04-22T10:16:21.358 INFO:teuthology.orchestra.run.trial118.stderr:sudo: ceph-bluestore-tool: command not found
```

ceph-bluestore-tool is not available on the host and must be run from within a container.

Instead, use `ceph-volume lvm zap` which internally calls `ceph-bluestore-tool zap-device`.

Fixes: https://tracker.ceph.com/issues/76238
Signed-off-by: Guillaume Abrioux <gabrioux@ibm.com>

Fix unit test

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

Merge pull request #68541 from rhcs-dashboard/rgw-nfs-export-path-issue

mgr/dashboard: Difference in "path" value observed when rgw user level export created via dashboard vs cli

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

mgr/dashboard: Nvmeof gateway group should account for labels

- updates gateway group sections to account for placement label

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

Merge PR #65656 into main

* refs/pull/65656/head:
client: do not allow zero‑length reads
src/test: test zero-length async-fsync read using ceph_ll_nonblocking_readv_writev
src/test: test zero-length async-fsync read using ll_preadv_pwritev

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

crimson/os/seastore: introduce stage_visibility_handoff

Generalize prepare_rewrite_publish_to_prior() into stage_visibility_handoff().

* introduce should_use_no_conflict_publish

* Replace is_rewrite_transaction() checks with should_use_no_conflict_publish(),
  so adding new no-conflict users becomes straightforward.

* Stop committing metadata (commit_state + sync_checksum) during prepare_record()
  (pre-commit). While it is correct for rewrite, doing it pre-commit doesn't buy
  us anything today because readers are still blocked until the publish finishes.

  Moving metadata commit to the after commit phase would also make future
  non-rewrite users easier to support.

This is a prep step for expanding no-conflict publish coverage.

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

Merge pull request #68562 from bluikko/wip-doc-sphinx-warnings-202604b

doc: fix Sphinx complaints

Merge pull request #68529 from rhcs-dashboard/fix-step-3

mgr/dashboard: Allow moving to next when Bidirectional mode is chosen

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

doc: fix Sphinx complaints

Fix section title underlines in ec_stretch_cluster.rst.
Fix table that was not rendered at all in lua-scripting.rst.

Signed-off-by: Ville Ojamo <git2233+ceph@ojamo.eu>

Merge pull request #68469 from VallariAg/fix-top-image-name

mgr/dashboard: show rados ns in 'ceph nvmeof top io'

Merge pull request #68535 from bluikko/wip-doc-rados-troubleshooting-mon-backticks

doc/rados: fix markup in troubleshooting-mon.rst

mgr/dashboard: Persist labels and hosts in service form

Fixes https://tracker.ceph.com/issues/75280

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

mgr/dashboard: Fix service form to take into account labels

Fixes https://tracker.ceph.com/issues/76215

- placement label for services takes only a single label
- when multiple labels or an array of labels is passed the service was not getting created

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

Merge pull request #68547 from rhcs-dashboard/encryption-fix

mgr/dashboard: Encryption key not being passed in gateway group form

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

Merge pull request #68531 from rhcs-dashboard/fix-step-2

mgr/dashboard: Validate no added hosts in second step

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

Merge pull request #68525 from rhcs-dashboard/y-axis-are

mgr/dashboard: Round off y-axis value of area chart

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

ceph-volume: has_bluestore_label checks all bluestore label replica offsets

BlueStore replicates the block device label at fixed offsets (0 and
multiples of 1Gb up to 1000gb). has_bluestore_label() only read the
first 22 bytes, so disks with a wiped primary label but intact
replicas are missed.

with this commit, has_bluestore_label() scans each known offset with
seek/read and compares the ASCII prefix as bytes.

Fixes: https://tracker.ceph.com/issues/76060
Signed-off-by: Guillaume Abrioux <gabrioux@ibm.com>

script/redmine-upkeep: resolve comment misformatting

See for example [1].

[1] https://github.com/ceph/ceph/pull/67343#issuecomment-4298533030

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

doc/rados: fix markup in troubleshooting-mon.rst

Double backticks do not stack with strong emphasis.
Use automatic emphasis instead.

Signed-off-by: Ville Ojamo <git2233+ceph@ojamo.eu>

Merge pull request #67019 from Jayaprakash-ibm/wip-object-fragmentation

os/bluestore: add static and runtime object fragmentation tracking

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

Merge pull request #68183 from rhcs-dashboard/set-motd

mgr/dashboard: Add option to set motd via api

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

Merge pull request #68467 from ujjawal-ibm/filter-virtual-ram-devices

ceph-volume: skip /dev/ram* devices in inventory

Merge pull request #68280 from xhernandez/proxy-fix-fscrypt-buffers

libcephfs_proxy: fix passed buffers for fscrypt functions

Reviewed-by: John Mulligan <jmulligan@redhat.com>
Reviewed-by: Anoop C S <anoopcs@cryptolab.net>

Merge pull request #68493 from rhcs-dashboard/76158-rgw-restart-stop-bug

mgr/dashboard : Fix services restart/stop issue

Reviewed-by: Afreen Misbah <afreen@ibm.com>
Reviewed-by: Dnyaneshwari Talwekar <dtalweka@redhat.com>

mgr/dashboard: Difference in "path" value observed when rgw user level export created via dashboard vs cli

Fixes: https://tracker.ceph.com/issues/75425
Signed-off-by: Dnyaneshwari Talwekar <dtalweka@redhat.com>

Merge pull request #68036 from ifed01/wip-ifed-print-effective-vsel-extra

os/bluestore: print effective extra in 'bluefs stats' report

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

Merge pull request #68177 from ifed01/wip-ifed-better-ebd-detect

blk,os/bluestore: do not call extblkdev::detect_device on every

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

mgr/dashboard: Encryption key not being passed in gateway group form

- encryption_key despite fo being added by user is not added in service spec
- this blocks both bi and uni directional auth

Fixes https://tracker.ceph.com/issues/76214

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

mgr/dashboard: Allow moving to next when Bidirectional mode is chosen

- there is an issue with bi-auth , where the validation is getting empty due to some state refresh
- fixed by removing re-setting of form state

Fixes https://tracker.ceph.com/issues/76194

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

Merge pull request #68353 from Jayaprakash-ibm/wip-clean-discard-warn

mon: simplify discard queue health warning

Reviewed-by: Adam Kupczyk <akupczyk@ibm.com>
Reviewed-by: Igor Fedotov <igor.fedotov@croit.io>
Reviewed-by: Kefu Chai <tchaikov@gmail.com>

Merge pull request #68441 from rhcs-dashboard/mgr/dashboard/carbonize-multisite-wizard-realm-token

mgr/dashboard: mgr/dashboard: Carbonize Realm Name and Token block in Multi-site Replication Wizard

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

mgr/dashboard: Fix review page showing auth details when all hosts selected

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

mgr/dashboard: Round off y-axis value of area chart

- by default y-axos set to 1 for all
- the value round off for area chart is seperated from y-axis ticks
- also fixes a bug where all IOPS y-ticks being repeated 1,1,0,0

The following values are set for now:

IOPS: valueDecimals=0, axisDecimals=1
Latency: valueDecimals=2, axisDecimals=1
Throughput: valueDecimals=2, axisDecimals=1
Consumption: valueDecimals=2, axisDecimals=1

Fixes https://tracker.ceph.com/issues/76191

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

mgr/dashboard: Add option to set motd via api

Signed-off-by: Aashish Sharma <aasharma@redhat.com>

Merge pull request #68537 from rhcs-dashboard/fix-76200

mgr/dashboard: [snap-visibility]Edit Client config option remains stuck in loading when nfs user is configured

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

Merge pull request #68521 from rhcs-dashboard/fix-osd

mgr/dashboard: Fix osd modal spacings and footer

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

Merge pull request #67995 from rhcs-dashboard/rename-namespacesize

mgr/dashboard: Rename namespace block size field and reorder namespace form order

Reviewed-by: Devika Babrekar <devika.babrekar@ibm.com>
Reviewed-by: pujaoshahu <pshahu@redhat.com>
Reviewed-by: Afreen Misbah <afreen@ibm.com>

client: do not allow zero‑length reads

fixing this in Client::_read which is called by both async and sync code paths.

Fixes: https://tracker.ceph.com/issues/73037
Signed-off-by: Dhairya Parmar <dparmar@redhat.com>

src/test: test zero-length async-fsync read using ceph_ll_nonblocking_readv_writev

Fixes: https://tracker.ceph.com/issues/73037
Signed-off-by: Dhairya Parmar <dparmar@redhat.com>

src/test: test zero-length async-fsync read using ll_preadv_pwritev

Fixes: https://tracker.ceph.com/issues/73037
Signed-off-by: Dhairya Parmar <dparmar@redhat.com>

mgr: fix continous smb MgrDBNotReady

Check db_ready before trying to fetch prometheus metrics and smb status (from dashboard)

Fixes: https://tracker.ceph.com/issues/76151
Signed-off-by: Pedro Gonzalez Gomez <pegonzal@ibm.com>

Merge branch 'main' into daemon-upgrade-issue-with-space