Merge pull request #58529 from joscollin/wip-66200-reef

reef: client: set LIBMOUNT_FORCE_MOUNT2=always

Merge pull request #58232 from batrick/wip-66592-reef

reef: tools/cephfs: recover alternate_name of dentries from journal

Merge pull request #57920 from batrick/wip-66387-reef

reef: qa: disable mon_warn_on_pool_no_app in fs suite

Merge pull request #57761 from joscollin/wip-66278-reef

reef: cephfs-journal-tool: Add preventive measures to avoid fs corruption

Merge pull request #57683 from batrick/wip-66194-reef

reef: mds: use regular dispatch for processing beacons

Merge pull request #57681 from batrick/wip-66190-reef

reef: mds: use regular dispatch for processing metrics

Merge pull request #57524 from batrick/wip-66043-reef

reef: qa: unmount clients before damaging the fs

Merge pull request #58157 from vshankar/wip-66324-reef

reef: qa: enable debug logs for fs:cephadm:multivolume subsuite

Merge pull request #58172 from batrick/wip-65272-reef

reef: qa: ignore PG health warnings in CephFS QA

Merge pull request #58231 from batrick/wip-65845-reef

reef: qa: ignore variation of PG_DEGRADED health warning

Merge pull request #58251 from joscollin/wip-66673-reef

reef: qa: enhance per-client labelled perf counters test

Merge pull request #58295 from batrick/wip-66625-reef

reef: mds: do remove the cap when seqs equal or larger than last issue

Merge pull request #58328 from joscollin/wip-65926-reef

reef: mds/purgequeue: add l_pq_executed_ops counter

Merge pull request #58323 from zdover23/wip-doc-2024-06-28-backport-57685-to-reef

reef: cephadm: use importlib.metadata for querying ceph_iscsi's version

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

Merge pull request #58576 from zdover23/wip-doc-2024-07-14-backport-58451-to-reef

reef: doc/dev: edit "Principles for format change"

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

Merge pull request #58475 from rhcs-dashboard/wip-66871-reef

reef: mgr/dashboard: nfs export enhancement for CEPHFS

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

Merge pull request #58579 from zdover23/wip-doc-2024-07-15-backport-58574-to-reef

reef: doc/cephfs: edit "is mount helper present"

doc/cephfs: edit "is mount helper present"

Edit the section called "Is mount helper present?", the title of which
prior to this commit was "Is mount helper is present?". Other small
disambiguating improvements have been made to the text in the section.
An unselectable prompt has been added before a command.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 0ebfb3eda8e2fc88662530bbddc84f2efcc4e13c)

doc/dev: edit "Principles for format change"

Improve "Principles for format change" in doc/dev/encoding.rst. This
commit started as a response to Anthony D'Atri's suggestion here: https://github.com/ceph/ceph/pull/58299/files#r1656985564

Review of this section suggested to me that certain minor English usage
improvements would be of benefit. The numbered lists in this section
could still be made a bit clearer.

Co-authored-by: Anthony D'Atri <anthony.datri@gmail.com>
Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 570797e5588b67b8c72e5297b61f84d9aa48dc45)

Merge pull request #58549 from idryomov/wip-test-cookie-reef

reef: test/librbd: clean up unused TEST_COOKIE variable

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

Merge pull request #58245 from rhcs-dashboard/wip-66424-reef

reef: mgr/dashboard: fix edit bucket failing in other selected gateways

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

test/ librbd: clean up unused 'TEST_COOKIE'

Clean them up because
```
/home/jenkins-build/build/workspace/ceph-pull-requests-arm64/src/test/librbd/exclusive_lock/test_mock_PreAcquireRequest.cc:47:26: error: unused variable 'TEST_COOKIE' [-Werror,-Wunused-const-variable]
static const std::string TEST_COOKIE("auto 123");
                         ^
1 error generated.
```

Ref: https://jenkins.ceph.com/job/ceph-pull-requests-arm64/59242/consoleFull#-5120798632a811ea2-3e7b-466b-84b4-d13df7e35809

Signed-off-by: Rongqi Sun <sunrongqi@huawei.com>
(cherry picked from commit 49869222c816023f696ff2a42898e2d9f3f54215)

Merge pull request #56877 from rhcs-dashboard/wip-65170-reef

reef: mgr/dashboard: bump follow-redirects from 1.15.3 to 1.15.6 in /src/pybind/mgr/dashboard/frontend

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

Merge pull request #58338 from rhcs-dashboard/wip-66738-reef

reef: mgr/dashboard: fix clone async validators with different groups

Reviewed-by: Ankush Behl <cloudbehl@gmail.com>
Reviewed-by: nmunet <NOT@FOUND>

Merge pull request #58522 from cbodley/wip-66896-reef

reef: install-deps.sh, do_cmake.sh: almalinux is another el flavour

Reviewed-by: Kyr Shatskyy <kyrylo.shatskyy@gmail.com>
Reviewed-by: Dan van der Ster <dan.vanderster@clyso.com>

Merge pull request #58247 from rhcs-dashboard/wip-66487-reef

reef: mgr/dashboard: add dueTime to rgw bucket validator

Reviewed-by: Pedro Gonzalez Gomez <pegonzal@redhat.com>
Reviewed-by: nmunet <NOT@FOUND>

Merge pull request #57752 from rhcs-dashboard/wip-66246-reef

reef: mgr/dashboard: fix readonly landingpage

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

client: set LIBMOUNT_FORCE_MOUNT2=always

This is a workaround for libmount 2.38 or later on linux kernel 5.1 or
later.

libmount defaults to the new fsconfig interface which does not support
remounting the way the old mount(2) syscall did. Thus, dentry trimming
fails.

Fixes: https://tracker.ceph.com/issues/62664
Signed-off-by: Jakob Haufe <sur5r@sur5r.net>
(cherry picked from commit 9ca1d4be2b587ce1d3e6c10dc2a98195b2ab3ba9)

install-deps.sh, do_cmake.sh: almalinux is another el flavour

Signed-off-by: Dan van der Ster <dan.vanderster@clyso.com>
(cherry picked from commit 6e85f76385c3cffd860a8a3054bf14ce85a43b59)

Conflicts:
do_cmake.sh add rocky too

Merge pull request #58489 from zdover23/wip-doc-2024-07-10-backport-58484-to-reef

reef: doc: Correct link to "Device management"

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

Merge pull request #58469 from zdover23/wip-doc-2024-07-09-backport-58454-to-reef

reef: doc/cephfs: fix typo

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

doc: Correct link to "Device management"

The current link goes to `cephadm/services/rados/operations/devices` which is a 404, it instead needs to be `rados/operations/devices/`.

Signed-off-by: Matthew Vernon <mvernon@wikimedia.org>
(cherry picked from commit d1a7cc4978adacd4e11df2d3e0886b191b118cc1)

mgr/dashboard: nfs export enhancement for CEPHFS

Fixes: https://tracker.ceph.com/issues/66718
Signed-off-by: Avan Thakkar <athakkar@redhat.com>
(cherry picked from commit 69a45db9c3cc9ed4ac36e7f2d6db78bd7940530b)

doc/cephfs: fix typo

blanacer -> balancer

Signed-off-by: spdfnet <32593931+spdfnet@users.noreply.github.com>
(cherry picked from commit d94f90b757189e269882523178a401cd4411a946)

Merge pull request #58207 from idryomov/wip-66300-reef

reef: librbd: disallow group snap rollback if memberships don't match

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

Merge pull request #58345 from idryomov/wip-66570-reef

reef: librbd: make diff-iterate in fast-diff mode aware of encryption

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

librbd: make diff-iterate in fast-diff mode aware of encryption

diff-iterate wasn't updated when librbd was being prepared to support
encryption in commit 8d6a47933269 ("librbd: add crypto image dispatch
layer").  This is even noted in [1]:

> The two places I skipped for now are DiffIterate and TrimRequest.

CryptoImageDispatch has since been removed, but diff-iterate in
fast-diff mode is still unaware of encryption and just assumes that all
offsets are raw.  This means that the callback gets invoked with
incorrect image offsets when encryption is loaded.  For example, for
a LUKS1-formatted image with some data at offsets 0 and 20971520,
diff-iterate with encryption loaded reports

  0~4194304
  4194304~4194304
  25165824~4194304

instead of

  0~4194304
  20971520~4194304

as "exists".  For any piece of code that is using diff-iterate to
optimize block-by-block processing (e.g. copy an encrypted source image
to a differently-encrypted destination image), this is fatal: it would
skip processing block 20971520 which has data and instead process block
25165824 which doesn't have any data and was to be skipped, producing
a corrupted destination image.

[1] https://github.com/ceph/ceph/pull/37935#issue-735278403

Fixes: https://tracker.ceph.com/issues/66570
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit cdeb0efce3f9f857ad6d5b7ff3965f3292cb571a)

test/librbd: extend DiffIterateTest.DiffIterateDeterministic{,PP}

Currently we are laying data only at the beginning of an object.
Extend the skeletons to write to three different offsets in the middle
and also at the end of the object.

Separately, make C and C++ API test variants slightly different in
terms of offsets being targeted to not go through exactly the same
scenario twice.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit da69d1fdbf69cf37ba8b75ea9a230634bb25ae76)

test/librbd: refactor DiffIterateTest.DiffIterateDeterministic{,PP}

In preparation for extending coverage, extract test logic into a couple
of parametrized helpers.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 1df87998dc30be6e3b4c90f5833539c7ccc70911)

Merge pull request #58211 from idryomov/wip-66418-reef

reef: librbd: diff-iterate shouldn't crash on an empty byte range

Reviewed-by: Mykola Golub <mgolub@suse.com>

Merge pull request #57765 from idryomov/wip-66232-reef

reef: qa/suites/rbd: override extra_system_packages directly on install task

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

Merge pull request #57603 from ajarr/wip-66116-reef

reef: librbd: create rbd_trash object during pool initialization and namespace creation

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

Merge pull request #57570 from idryomov/wip-65813-reef

reef: librbd: don't crash on a zero-length read if buffer is NULL

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

test/librbd: avoid caching effects TestGroup.add_snapshot{,PP}

After rollback started being tested in commit b3977c53c930
("test/librbd: make rollback in TestGroup.add_snapshot{,PP}
meaningful"), these tests can fail on comparing post-rollback
data to expected data if run with exclusive lock disabled.
This doesn't occur with exclusive lock enabled because the RBD
cache gets invalidated implicitly before releasing the lock.

While at it, pass LIBRADOS_OP_FLAG_FADVISE_FUA to avoid relying
on any cache settings that happen to be in effect.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit a7e314c2ef899406eb3ade6c5f862622ce1e698f)

test/librbd: give groups unique names in TestGroup tests

Otherwise, a failing add_snapshot test causes add_snapshotPP test to
fail too, for example.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 4c69370f6b0e4c04e77fde67a9c199af6ea13649)

test/librbd: drop unused variable in TestGroup.group_create

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit db52fca5b1fc826ce9d7cf63d7fd1aa86addccf0)

librbd: drop unused parameter on group_snap_rollback_by_record()

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit afbb744ebe9bddc1d71c84598fee832bf7843c2b)

librbd: disallow group snap rollback if memberships don't match

Before proceeding with group rollback, ensure that the set of images
that took part in the group snapshot matches the set of images that are
currently part of the group.  Otherwise, because we preserve affected
snapshots when an image is removed from the group, data loss can ensue
where an image gets rolled back while part of another group or not part
of any group but long repurposed for something else.

Similarly, ensure that the group snapshot is complete.

Fixes: https://tracker.ceph.com/issues/66300
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit f212a9ca5b9238b80ae5c728a7bf99366ed910a1)

Conflicts:
src/test/pybind/test_rbd.py [ commit d7fd66ec9944 ("librbd: add
  rbd_clone4() API to take parent snapshot by ID") not in reef ]

Merge pull request #58421 from guits/wip-66831-reef

reef: qa: fix log errors for cephadm tests

Merge pull request #58220 from guits/wip-66024-reef

reef: ceph-volume: create LVs when using partitions

Merge pull request #58432 from zdover23/wip-doc-2024-07-05-backport-58368-to-reef

reef: doc/rados: document manually passing search domain

doc/rados: document manually passing search domain

Document how to manually pass the search domain to "mon_dns_srv_name" in
doc/rados/configuration/mon-lookup-dns.rst.

This commit is made in response to a request by Lander Duncan that was made on the [ceph-users] mailing list, and can be seen here: https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/F7V4CWLIYCAJ4JXI2JLNY6QPCFPR4SLA/

Co-authored-by: Anthony D'Atri <anthony.datri@gmail.com>
Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 98938a0312dd0c8e0b293ed9aa2e0760cc9619fa)

qa: fix log errors for cephadm tests

This fixes a lot of errors induced by the log scrapper check.

Fixes: https://tracker.ceph.com/issues/66751
Signed-off-by: Guillaume Abrioux <gabrioux@ibm.com>
(cherry picked from commit 53b462764c6fe11282b6cdb21131c176c1ed2350)

Merge pull request #57451 from joscollin/wip-65990-reef

reef: cephfs_mirror: fix crash in update_fs_mirrors()

Merge pull request #58088 from rishabh-d-dave/wip-66502-reef

reef: cephfs-shell: prints warning, hangs and aborts when launched

Merge pull request #58411 from ljflores/wip-tracker-66505

qa/suites/upgrade/reef-p2p/reef-p2p-parallel: increment upgrade to 18.2.2

qa/suites/upgrade/reef-p2p/reef-p2p-parallel: increment upgrade to 18.2.2

Instead of installing 18.2.0, which still contains the osdmap crc bug tracked
in https://tracker.ceph.com/issues/63389, we should install v18.2.2 since this contains
the fix. Then, we upgrade to reef_latest. In this scenario, we do not expect to see the
crc bug. If we test any upgrade path before that, we will hit the warning and the test will fail.

Fixes: https://tracker.ceph.com/issues/66505
Signed-off-by: Laura Flores <lflores@ibm.com>

Merge pull request #58390 from zdover23/wip-doc-2024-07-02-backport-58092-to-reef

reef: doc/start: remove mention of Centos 8 support

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

doc/start: remove mention of Centos 8 support

Remove information so that it is clear that Centos 8 will no longer be
supported by Ceph.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 6b149e4f6ab2ac716e73f42b0d6c67482a768da9)

Merge pull request #57932 from cbodley/wip-reef-drop-centos8

reef: qa/distros: remove centos 8 from supported distros

Reviewed-by: Adam King adking@redhat.com

Merge pull request #57999 from ljflores/wip-tracker-66459

reef: qa/suites/rados/thrash-old-clients: update supported releases and distro

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

tests: make sure nvmetcli and nvme-cli are up to date

Otherwise jobs end up with the following failure:

```
2024-06-25T14:22:18.659 INFO:teuthology.orchestra.run.smithi098.stderr:Failed to write to /dev/nvme-fabrics: Invalid argument
```

Also, the output of nvme list has changed so we have to update
qa/tasks/nvme_loop.py accordingly.

Fixes: https://tracker.ceph.com/issues/66707
Signed-off-by: Guillaume Abrioux <gabrioux@ibm.com>
(cherry picked from commit d707c41eacb513bdcb9fec17c823bfbfdc182700)

qa/rgw/upgrade/pacific: remove centos_8.stream.yaml and rely on ubuntu_20.04.yaml

we can't test this pacific->reef upgrade path on centos because pacific doesn't
have centos 9 builds, and reef no longer has centos 8 builds. only test
this upgrade on ubuntu focal which is still supported for both releases

this commit targets the reef branch directly because this rgw/upgrade/pacific
suite no longer exists on main and squid branches

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

qa/distros: removed rhel 8 from supported distros

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 7a1dce1ebd883741b5003b9e18d4765526cbbb3e)

qa/crimson-rados: remove centos 8 symlinks

As we're trying to drop centos 8 from the distros we
test on these symlinks are now dead and need to be
cleaned up. In main, there was no replacement for
these symlinks (it just relies on the
crimson-supposted-all-distro dir for its distro)
so I'm just removing them here.

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

qa/upgrade: make reef p2p parallel upgrade tests to use centos 9

Since we're no longer using centos 8 for tests after the removal
of the mirrors

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

qa/distros: add ubuntu 22.04 for containerized tests

Partial backport of 0fa3eb67387eaf403b5a6e716a81582949dcecf1
that adds the symlinks for the containerized tests to use
ubuntu 22.04 but leaves out the part dropping ubuntu 20.04

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

qa/distros: add supported-container-hosts for cephadm and upgrade suites

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 849a58b0873154a6713b4c8d838634f13b0b806d)

qa/workunits/cephadm: enable nodejs:18 on centos 9 stream

Signed-off-by: Laura Flores <lflores@ibm.com>
(cherry picked from commit a602e5f0e0e48c40fd728ae382502826ca354828)

qa/cephadm: fix iscsi pids limit check for centos 9

Centos 9 uses cgroups v2 which has a slightly
different file location for the pids.max. This commit
updates the test to also check the new location
so the test can pass on centos 9

Signed-off-by: Adam King <adking@redhat.com>
(cherry picked from commit 97efa97c2c72dba32774d0f105ae90462b93ac4e)

qa/cephadm: use quincy for add-repo test

There are no centos 9 build for octopus, so if we
want to start testing on cnetos 9 as a distro we need
the add-repo test to be done on a newer release
for which there are actual builds

Signed-off-by: Adam King <adking@redhat.com>
(cherry picked from commit 80dc6ce72a2c209424698ce31549362e9148022f)

qa: remove container hosts install from centos 9 distros

We needed the container-hosts version of podman on centos 8,
but there's no need to keep using it in centos 9

Additionally, changes from crun variants to runc variants since
crun is now the default container runtime on centos 9

Signed-off-by: Adam King <adking@redhat.com>
(cherry picked from commit c8873c6591d368e12907669c274fd3d6391e3f68)

qa/distos: remove rhel from podman and container-hosts

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 7193f76a1502de392af0b407db0073266f9bc9c8)

qa/suites/fs: workload suite uses centos 9 instead of rhel

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit b9ba9642fea18055e768c3ddee5ba5e372264ede)

qa/distros: bump podman from centos 8 to 9

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 00b3eef5b846dee6e9aae1f1295271c89f2801e5)

qa/distros: bump container-hosts from centos 8 to 9

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit bac9e74abf0f5cb60c74640d6c4689169acf7abe)

qa/distros: remove a-supported-distro.yaml that pointed to centos 7.2

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 3825863893a8577a306d70e171224f051d7e70ee)

qa/distros: replace supported-all-distro with supported

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 0e5746d44ef44ba1c7e7c0c3c3828fdfc8b14fc2)

qa/rgw/upgrade: fix broken links to distro files

replace the broken links to qa/distros/supported/*.yaml with links under
qa/distros/all/

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit a60978fc51fa45399fbab5f2ca408f771c607c6d)

qa/fs/upgrade: fix broken links to distro files

replace the broken links to qa/distros/supported/centos_8.stream.yaml
with links under qa/distros/all/

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 7486856ba1259ff2c35f5017d12a21f797c869d1)

qa/rgw/notifications: pin to centos_latest

the subsuite had a supported-all-distro$/ subdirectory, but that only
contained centos_8.yaml. qa/tasks/rabbitmq.py is hardcoded to use 'yum'
and rpm packages, so replace supported-all-distro$ with a link to
centos_latest.yaml

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 2d9348ad564f9199ae58062c1d12b7020cba0790)

qa/distros: removed centos 8 from supported distros

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 449db416aa82df12b349ce10603c9646c88bde83)

cephadm: use importlib.metadata for querying ceph_iscsi's version

use importlib.metadata for querying ceph_iscsi's version and fallback to
pkg_resources. as the former is only available in Python 3.8, while
the latter is deprecated.

Refs https://tracker.ceph.com/issues/66201

This commit is functionally equivalent to a Reef-targeted backport of
https://github.com/ceph/ceph/pull/57685.

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

Merge pull request #58371 from zdover23/wip-doc-2024-07-01-backport-58366-to-reef

reef: doc/cephfs: update cephfs-shell link

Merge pull request #58364 from zdover23/wip-doc-2024-07-01-backport-55577-to-reef

reef: doc/start: fix wording & syntax

doc/cephfs: update cephfs-shell link

Repair the link to cephfs-shell.rst in doc/cephfs/cephfs-shell.rst that
was broken in https://github.com/ceph/ceph/pull/41165/ when
doc/cephfs/cephfs-shell.rst was moved to doc/man/8/cephfs-shell.rst.

This commit is made in response to a request by Lander Duncan that was
made on the [ceph-users] mailing list, and can be seen here: https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/F7V4CWLIYCAJ4JXI2JLNY6QPCFPR4SLA/

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 4f8903cf80a94596a1b17b4a07affb874f21ae76)

Merge pull request #58113 from vshankar/wip-66518-reef

reef: qa: use centos9 for fs:upgrade

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Jos Collin <jcollin@redhat.com>
Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

doc/start: fix wording & syntax

Signed-off-by: Piotr Parczewski <piotr@stackhpc.com>
(cherry picked from commit a8b27af84910d324023a37e616f42d6d3faa4ecf)

Merge pull request #58360 from zdover23/wip-doc-2024-06-30-backport-58358-to-reef

reef: doc/cephfs: take Anthony's suggestion

doc/cephfs: take Anthony's suggestion

Incorporate Anthony's suggestion in
https://github.com/ceph/ceph/pull/57567#discussion_r1607016050 into
doc/cephfs/fs-volumes.rst.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 7a78dda0a92339db27359f91f0cb081703e2e7f9)

qa/cephfs: fs:upgrade:mds_upgrade_sequence uses centos9

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

qa: upgrade sub-suite upgraded_client from from n-1|n-2 releases

... and move to centos9.

Fixes: https://tracker.ceph.com/issues/65580
Signed-off-by: Dhairya Parmar <dparmar@redhat.com>
Signed-off-by: Venky Shankar <vshankar@redhat.com>
(cherry picked from commit 76589659480c6e9cd2ad09b3be21deeb6d1686d1)

 Conflicts:
qa/suites/fs/upgrade/upgraded_client/tasks/0-from/nautilus.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/0-install.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/0-install/quincy.yaml

qa: upgrade sub-suite nofs from n-1 and n-2 releases

also adjust the respective upgrade YAMLs and move to
centos9.

For reef release, start upgrade from pacific/quincy.

Fixes: https://tracker.ceph.com/issues/65580
Signed-off-by: Dhairya Parmar <dparmar@redhat.com>
Signed-off-by: Venky Shankar <vshankar@redhat.com>
(cherry picked from commit f00a69f714dd07d580731a326aff41cacca6d0b6)

qa: use supported releases for featureful_client

... and switch for centos9.

For reef release, start upgrade from pacific/quincy.

Fixes: https://tracker.ceph.com/issues/65580
Signed-off-by: Venky Shankar <vshankar@redhat.com>
(cherry picked from commit 3715ce671de97530dda30d59fe643cebc3ebc508)

qa: refactor client upgrade yamls and other minor touchups

* start testing new_ops and stress_tests with both the drivers(i.e. fuse and kclient)
therefore moved 0-clients/ from tasks/3-workload/new_ops/ to tasks/ and renamed it to
2-clients/

* since new_ops/ and stress_tests/ now share the common upgrade yaml, moved the
tests yamls(in stress_tests/1-tests) directly under 3-workload/stress_tests/

* renamed 1-client-sanity.yaml in new_ops/ to newops.yaml

Fixes: https://tracker.ceph.com/issues/62953
Signed-off-by: Dhairya Parmar <dparmar@redhat.com>
(cherry picked from commit c556c7324763332df62f4d382da02314900f1a60)

 Conflicts:
qa/suites/fs/upgrade/upgraded_client/tasks/2-clients/fuse-upgrade.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-clients/kclient.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/new_ops/0-clients/fuse-upgrade.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/new_ops/0-clients/kclient.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/new_ops/1-client-sanity.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/new_ops/newops.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/stress_tests/1-tests/blogbench.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/stress_tests/1-tests/dbench.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/stress_tests/1-tests/fsstress.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/stress_tests/1-tests/iozone.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/stress_tests/1-tests/kernel_untar_build.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/stress_tests/blogbench.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/stress_tests/dbench.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/stress_tests/fsstress.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/stress_tests/iozone.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/2-workload/stress_tests/kernel_untar_build.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/3-workload/new_ops/0-clients/kclient.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/3-workload/new_ops/1-client-sanity.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/3-workload/stress_tests/1-tests/blogbench.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/3-workload/stress_tests/1-tests/dbench.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/3-workload/stress_tests/1-tests/fsstress.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/3-workload/stress_tests/1-tests/iozone.yaml
qa/suites/fs/upgrade/upgraded_client/tasks/3-workload/stress_tests/1-tests/kernel_untar_build.yaml
qa/suites/rbd/device/workloads/.qa

mgr/dashboard: cephfs snapshot and clones e2e tests

Fixes: https://tracker.ceph.com/issues/65180
Signed-off-by: Nizamudeen A <nia@redhat.com>
(cherry picked from commit 3d1a7c56ccb8eb6d988e034e99e199cd226b1704)
(cherry picked from commit 9fbcf8658863b5a03d7d8fbf0f3e5d23baf6dc59)

mgr/dashboard: fix clone async validators with different groups

Providing a way to dynamically update the async validator based on the
selector field so that when the selected value changes, the depended
field like the clone name gets validated again against the new value

Fixes: https://tracker.ceph.com/issues/66703
Signed-off-by: Nizamudeen A <nia@redhat.com>
(cherry picked from commit 38bcce2346b3430e93d67a443a27ebdf9232fec9)

mds/purgequeue: add l_pq_executed_ops counter

We need to know cluster deleting pressure better. And only pq_executed
 is not enough, because file size is different.

Fixes: https://tracker.ceph.com/issues/63468
Signed-off-by: shimin <shimin@kuaishou.com>
(cherry picked from commit bdae273235104c8d7e1d46add3b64d78511f4d0c)

Merge pull request #58308 from zdover23/wip-doc-2024-06-27-backport-58287-to-reef

reef: doc/rados: update how to install c++ header files

Merge pull request #58305 from zdover23/wip-doc-2024-06-27-backport-58299-to-reef

reef: doc/dev: Fix typos in encoding.rst