crimson/osd: wire the EC-specific messages up

Signed-off-by: Radosław Zarzyński <rzarzyns@redhat.com>

crimson/osd: bring ECRepRequest

Signed-off-by: Radosław Zarzyński <rzarzyns@redhat.com>

crimson/osd: always dencode object_info_t with hobject_t

The classical OSD never uses the the `encode_no_oid()`
and `decode_no_oid()` variants of object info's encoding
machinery. crimson departed from this to such an extent
on recovery it sends `OI_ATTR` with meaningless `soid`-
related bytes. The net result is the following crash in
an heterogenous cluster:

```
    -7> 2023-08-22T14:36:57.621+0200 7f25113c8700  0 osd.1 pg_epoch: 62 pg[1.0( v 42'43 lc 39'37 (0'0,42'43] local-lis/les=45/58 n=4 ec=6/6 lis/c=45/6 les/c/f=58/8/0 sis=45) [1] r=0 lpr=53 pi=[6,45)/1 crt=42'43 lcod 0'0 mlcod 0'0 active+recovering rops=2 m=2 mbc={255={(0+1)=2}}] get_object_context:11843: soid.pool=1
    -6> 2023-08-22T14:36:57.621+0200 7f25113c8700 10 osd.1 pg_epoch: 62 pg[1.0( v 42'43 lc 39'37 (0'0,42'43] local-lis/les=45/58 n=4 ec=6/6 lis/c=45/6 les/c/f=58/8/0 sis=45) [1] r=0 lpr=53 pi=[6,45)/1 crt=42'43 lcod 0'0 mlcod 0'0 active+recovering rops=2 m=2 mbc={255={(0+1)=2}}] get_object_context: obc NOT found in cache: 1:30306672:devicehealth::main.db.0000000000000000:head
    -5> 2023-08-22T14:36:57.621+0200 7f25113c8700  0 osd.1 pg_epoch: 62 pg[1.0( v 42'43 lc 39'37 (0'0,42'43] local-lis/les=45/58 n=4 ec=6/6 lis/c=45/6 les/c/f=58/8/0 sis=45) [1] r=0 lpr=53 pi=[6,45)/1 crt=42'43 lcod 0'0 mlcod 0'0 active+recovering rops=2 m=2 mbc={255={(0+1)=2}}] get_object_context:11865 bv came from OI_ATTR
    -4> 2023-08-22T14:36:57.621+0200 7f25113c8700  0 osd.1 pg_epoch: 62 pg[1.0( v 42'43 lc 39'37 (0'0,42'43] local-lis/les=45/58 n=4 ec=6/6 lis/c=45/6 les/c/f=58/8/0 sis=45) [1] r=0 lpr=53 pi=[6,45)/1 crt=42'43 lcod 0'0 mlcod 0'0 active+recovering rops=2 m=2 mbc={255={(0+1)=2}}] get_object_context:11897 just after bv decode oi.soid.pool=-9223372036854775808
    -3> 2023-08-22T14:36:57.621+0200 7f25113c8700  0 osd.1 pg_epoch: 62 pg[1.0( v 42'43 lc 39'37 (0'0,42'43] local-lis/les=45/58 n=4 ec=6/6 lis/c=45/6 les/c/f=58/8/0 sis=45) [1] r=0 lpr=53 pi=[6,45)/1 crt=42'43 lcod 0'0 mlcod 0'0 active+recovering rops=2 m=2 mbc={255={(0+1)=2}}] soid.pool=-9223372036854775808 info.pgid.pool()=1
    -2> 2023-08-22T14:36:57.641+0200 7f25113c8700 -1 ../src/osd/PrimaryLogPG.cc: In function 'ObjectContextRef PrimaryLogPG::get_object_context(const hobject_t&, bool, const std::map<std::__cxx11::basic_string<char>, ceph::buffer::v15_2_0::list, std::less<void> >*)' thread 7f25113c8700 time 2023-08-22T14:36:57.625282+0200
../src/osd/PrimaryLogPG.cc: 11906: FAILED ceph_assert(oi.soid.pool == (int64_t)info.pgid.pool())

 ceph version 18.0.0-4293-g6cd888f2315 (6cd888f2315e525412ee7d39587d11d0a19245c2) reef (dev)
 1: (ceph::__ceph_assert_fail(char const*, char const*, int, char const*)+0x127) [0x5602381712da]
 2: (ceph::register_assert_context(ceph::common::CephContext*)+0) [0x560238171505]
 3: (PrimaryLogPG::get_object_context(hobject_t const&, bool, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, ceph::buffer::v15_2_0::list, std::less<void>, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, ceph::buffer::v15_2_0::list> > > const*)+0xc9e) [0x560237b5aec6]
 4: (non-virtual thunk to PrimaryLogPG::get_obc(hobject_t const&, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, ceph::buffer::v15_2_0::list, std::less<void>, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, ceph::buffer::v15_2_0::list> > > const&)+0x30) [0x560237c2f9d5]
 5: (ReplicatedBackend::handle_pull_response(pg_shard_t, PushOp const&, PullOp*, std::__cxx11::list<ReplicatedBackend::pull_complete_info, std::allocator<ReplicatedBackend::pull_complete_info> >*, ceph::os::Transaction*)+0x5dc) [0x560237e9d116]
 6: (ReplicatedBackend::_do_pull_response(boost::intrusive_ptr<OpRequest>)+0x477) [0x560237e9e6d5]
 7: (ReplicatedBackend::_handle_message(boost::intrusive_ptr<OpRequest>)+0x1e1) [0x560237e9f2ef]
 8: (PGBackend::handle_message(boost::intrusive_ptr<OpRequest>)+0x59) [0x560237c3f3bb]
 9: (PrimaryLogPG::do_request(boost::intrusive_ptr<OpRequest>&, ThreadPool::TPHandle&)+0x93f) [0x560237bb7bdd]
 10: (OSD::dequeue_op(boost::intrusive_ptr<PG>, boost::intrusive_ptr<OpRequest>, ThreadPool::TPHandle&)+0x531) [0x5602379dcb41]
 11: (ceph::osd::scheduler::PGRecoveryMsg::run(OSD*, OSDShard*, boost::intrusive_ptr<PG>&, ThreadPool::TPHandle&)+0x22c) [0x560237d5b8f8]
 12: (OSD::ShardedOpWQ::_process(unsigned int, ceph::heartbeat_handle_d*)+0x3386) [0x5602379ff9f4]
 13: (ShardedThreadPool::shardedthreadpool_worker(unsigned int)+0x979) [0x56023815f9c3]
 14: (ShardedThreadPool::WorkThreadSharded::entry()+0x17) [0x560238163279]
 15: (Thread::entry_wrapper()+0x43) [0x56023814c0af]
 16: (Thread::_entry_func(void*)+0xd) [0x56023814c0cb]
 17: /lib/x86_64-linux-gnu/libpthread.so.0(+0x8609) [0x7f2531844609]
 18: clone()
```

Fixes: https://tracker.ceph.com/issues/62526
Signed-off-by: Radosław Zarzyński <rzarzyns@redhat.com>

crimson/osd, osd: build and link ECCommon

The modifications of the code shared with the classical OSD boil
down to just adding some `WITH_CRIMSON` guards; nothing more.

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

Merge PR #67954 into main

* refs/pull/67954/head:
orchestrator/test/test_orchestrator: fix return code to negative
mgr/mgr_module: fix tox test missing a type annotation
mgr/selftest: mypy error fix missing a type annotation

Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>
Reviewed-by: John Mulligan <jmulligan@redhat.com>

Merge pull request #67830 from dmick/container-manifest-versions

container/make-manifest-list.py: add version parameter

Merge pull request #67420 from Jayaprakash-ibm/wip-fix-bluefs_used-accounting

os/bluestore: fix incorrect bluefs_used accounting when releasing old extents

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

orchestrator/test/test_orchestrator: fix return code to negative

After changes of PR #67652

Signed-off-by: Nitzan Mordechai <nmordech@ibm.com>

mgr/mgr_module: fix tox test missing a type annotation

After changes of PR #67327

Signed-off-by: Nitzan Mordechai <nmordech@ibm.com>

Merge pull request #67853 from bluikko/wip-doc-cephfs-spelling

doc/cephfs: Fix spelling etc errors

Merge pull request #67959 from djgalloway/r10-default-fromimage

container/build.sh: FROM_IMAGE=rockylinux-10 default for >=tentacle

container/build.sh: FROM_IMAGE=rockylinux-10 default for >=tentacle

We build centos9 and rocky10 packages and containers by default now for wip, main, and tentacle branches as of https://github.com/ceph/ceph-build/pull/2557.

Starting with tentacle, we want a `podman pull quay.ceph.io/ceph-ci/ceph:tentacle` or `podman pull quay.ceph.io/ceph-ci/ceph:$SHA1` to get the container with Rocky 10 as the Base OS image, or FROM_IMAGE.

Fixes: https://tracker.ceph.com/issues/75673
Signed-off-by: David Galloway <david.galloway@ibm.com>

Merge pull request #67696 from cbodley/wip-75375

rgw: CompleteMultipartUpload renews cls_lock on multipart meta object

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

Merge PR #61981 into main

* refs/pull/61981/head:

Reviewed-by: Venky Shankar <vshankar@redhat.com>
Reviewed-by: Rishabh Dave <ridave@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>

mgr/selftest: mypy error fix missing a type annotation

After changes of PR #67327

Signed-off-by: Nitzan Mordechai <nmordech@ibm.com>

Merge pull request #67804 from VallariAg/wip-nvmeof-thrasher-fixes

qa: Fix nvmeof 'errors during thrashing'

Merge pull request #67833 from rhcs-dashboard/storage-class-zone-12403

mgr/dashboard: [storage-class]: Deleting local storage class from UI does not remove its entry from zone.

Reviewed-by: Nizamudeen A <nia@redhat.com>
Reviewed-by: Pedro Gonzalez Gomez <pegonzal@ibm.com>

Merge pull request #67809 from rhcs-dashboard/fix-add-node-issue

mgr/dashboard: fix-add-node-issue

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

Merge pull request #67781 from rhcs-dashboard/Bug-75443

mgr/dashboard: Title says "Gateways" but it displays gateway group names in NVME/TCP

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

doc/cephfs: Fix spelling etc errors

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

Merge pull request #67896 from idryomov/wip-75209

qa/workunits/rbd: drop racy assert in test_tasks_recovery()

Reviewed-by: Mykola Golub <mykola.golub@clyso.com>

Merge pull request #67919 from idryomov/wip-75378

librbd/migration/QCOWFormat: avoid use-after-free in execute_request()

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

Merge PR #67928 into main

* refs/pull/67928/head:
doc: fix release date of v18.2.8

Reviewed-by: Yuri Weinstein <yweins@redhat.com>

Merge pull request #67918 from bluikko/wip-doc-monitoring-improve

doc/monitoring: Improve index.rst

doc/monitoring: Improve index.rst

Improve consistency and formatting.
Fix typos, markup errors, missing words.
Fix wrong metric explanation.
Change a broken MD link to an RST link.

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

doc: fix release date of v18.2.8

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

Merge PR #67746 into main

* refs/pull/67746/head:
doc: add notable/security changes in v18.2.8
doc: add known issues for v18.2.8
doc: add reef release notes
script: fix argument change for github module

Reviewed-by: Redouane Kachach <rkachach@redhat.com>
Reviewed-by: Anthony D Atri <anthony.datri@gmail.com>

Merge PR #67652 into main

* refs/pull/67652/head:
pybind/orchestrator/cli: fix OrchestratorError retval sign

Merge PR #67557 into main

* refs/pull/67557/head:
qa: resolve py3.12 regression for random.sample

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

Merge PR #67327 into main

* refs/pull/67327/head:
mgr/dashboard: use __name__ for module-specific logging
selftest: Add logging self tests
pybind/mgr/mgr_module: isolate logging per mgr module

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

Merge PR #67230 into main

* refs/pull/67230/head:
qa: suppress false positive delete map mismatch errors

Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>

Merge PR #67127 into main

* refs/pull/67127/head:
qa/suites/upgrade: exclude rocky when Squid is involved
qa/suites/upgrade: update upgrade paths

Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>
Reviewed-by: Nitzan Mordechai <nmordech@redhat.com>

Merge PR #67102 into main

* refs/pull/67102/head:
qa/workunits/rados/test_envlibrados_for_rocksdb.sh: Add Rocky support
qa/workunits/ceph-helpers-root: Add Rocky support for install packages

Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>

Merge PR #66651 into main

* refs/pull/66651/head:
qa: suppress MismatchedFree operator delete RocksDB

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

Merge PR #66396 into main

* refs/pull/66396/head:
neorados: specify alignments for aligned_storage

Reviewed-by: Adam C. Emerson <aemerson@redhat.com>
Reviewed-by: Laura Flores <lflores@redhat.com>
Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Kefu Chai <k.chai@proxmox.com>
Reviewed-by: Mark Kogan <mkogan@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>

Merge PR #66244 into main

* refs/pull/66244/head:
mgr/Gil.cc: simplify Gil(), ~Gil()
mgr/Gil.cc: do not use PyGILState_Check()
mgr: add mgr_subinterpreter_modules config
python-common/.../service_spec: implement ServiceSpec.__getnewargs__ to allow unpickle to work correctly
mgr: serialize python objects sent between subinterpreters via remote

Reviewed-by: Nitzan Mordechai <nmordech@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>

Merge PR #66055 into main

* refs/pull/66055/head:
qa/distros: re-install nvme-cli package in rocky tests
qa: allowlist bpf podman denials on Rocky 10
qa/distros: bump rocky to 10.1
qa/distros: add rocky_10 as supported container host
qa/distros: bump rpm_latest.yaml to rocky_10.yaml
qa/distros: rename centos_latest.yaml to rpm_latest.yaml
qa/distros: add rocky_9 and rocky_10

Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>
Reviewed-by: Ilya Dryomov <idryomov@redhat.com>

Merge PR #63859 into main

* refs/pull/63859/head:
qa/workunits/mgr: account for nvmeof module being "always-on"
mgr, qa: clarify module checks in DaemonServer
mgr, qa: add `pending_modules` to asock command
mgr, common, qa, doc: issue health error after max expiration is exceeded
mgr: ensure that all modules have started before advertising active mgr

Reviewed-by: Nitzan Mordechai <nmordech@redhat.com>
Reviewed-by: Anthony D Atri <anthony.datri@gmail.com>
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>

Merge pull request #67568 from tchaikov/wip-rgw-fix-co_waiter-stack-use-after-scope

common/async: fix stack-use-after-scope in co_waiter

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

Merge pull request #66624 from leonidc/stretched-cluster

Stretched cluster

Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: baum <alexander.indenbaum@gmail.com>

Merge pull request #67798 from ceph/wip-oozmen-75466

rgw/pubsub: fix topic num_shards serialization and some multisite test improvements

Reviewed-by: Shilpa Jagannath <smanjara@redhat.com>
Reviewed-by: Yuval Lifshitz <ylifshit@ibm.com>

Merge pull request #67336 from BBoozmen/oozmen-wip-74917

RGW: handle plain-text object tags in RGWObjTags::decode()

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

librbd/migration/QCOWFormat: avoid use-after-free in execute_request()

Both L2TableCache and QCOWFormat can be destroyed after the completion
for the last L2 cache request is posted, particularly so in unit tests.
The strand destructor doesn't drain the handler queue in any way but
merely ensures that previously posted handlers would get dispatched in
a non-concurrent fashion.  As a result, use-after-free can ensue when
execute_request() unnecessarily dispatches itself for the last time.

Fixes: https://tracker.ceph.com/issues/75378
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

qa/distros: re-install nvme-cli package in rocky tests

Fixes: https://tracker.ceph.com/issues/73823
Signed-off-by: Adam King <adking@redhat.com>

qa: allowlist bpf podman denials on Rocky 10

Rocky Linux 10 logs SELinux AVCs for systemd BPF operations during container startup due to incomplete SELinux policy coverage. These AVCs occur in permissive mode, are reproducible without Ceph, and do not indicate functional failure. Tests should ignore this specific AVC class while continuing to fail on enforced denials.

Signed-off-by: David Galloway <david.galloway@ibm.com>

qa/distros: bump rocky to 10.1

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

qa/distros: add rocky_10 as supported container host

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

qa/distros: bump rpm_latest.yaml to rocky_10.yaml

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

qa/distros: rename centos_latest.yaml to rpm_latest.yaml

rename qa/distros/all/centos_latest.yaml to rpm_latest.yaml, along with
all symlinks thereto

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

qa/distros: add rocky_9 and rocky_10

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

Merge pull request #67745 from timqn22/db-device-not-available-osd-fail-apply

src/ceph-volume: fast device unavailable as error

mgr/dashboard: Title says "Gateways" but it displays gateway group names in NVME/TCP

Fixes: https://tracker.ceph.com/issues/75443
Signed-off-by: pujaoshahu <pshahu@redhat.com>

Merge pull request #67406 from batrick/i75013

mon/AuthMonitor: add osd w cap for superuser client

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

Merge pull request #65885 from edwinzrodriguez/ceph-wip-73431

mds: Remove unnecessary std::move on trivially-copyable LockOp

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

Merge pull request #66923 from thmour/main

rgw: ListRoles returns "Access Denied" for a regular user with valid allow policy

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

container/make-manifest-list.py: handle "no tags for this version"

Especially now that --version is required, there may be no
useful tag; catch early and error out.

Signed-off-by: Dan Mick <dmick@ibm.com>

container/make-manifest-list.py: add version support

Add mandatory -v/--version to select version to examine (to allow
multiple prerelease tags to exist).  Reorder arguments so that
usage help in the 'missing version' case shows the long option names.

Requires change to ceph-release-containers job as well to pass
the --version argument.

This commit is part of a PR that includes an update to the "promote"
invocation of make-manifest-list.py, which is done manually and must
also contain the --version argument.

Signed-off-by: Dan Mick <dmick@ibm.com>

rgw/posix: MPSerializer sets and clears locked status

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

src/ceph-volume: fast device unavailable as error

Normally when fast devices are passed to batch command but
no fast allocations could be found the batch command will
do nothing and return an empty plan. This leads to issues
however because the return essentially makes this issue silent
which makes it hard to debug in certain scenarios. I propose
to change this to raise error, and have made changes in osd.py
to better log the errors and process the exceptions. This
shouldn't affect processes that much and the change in
osd.py ensures the raised errors will not interrupt the return
output. I've also changed the unit tests to account for
change.

Signed-off-by: Timothy Q Nguyen <timqn22@gmail.com>

rgw/dbstore: MPSerializer sets and clears locked status

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

qa/workunits/mgr: account for nvmeof module being "always-on"

Post the merge of this: https://github.com/ceph/ceph/pull/67641

Fixes: https://tracker.ceph.com/issues/71631
Signed-off-by: Laura Flores <lflores@ibm.com>

qa/workunits/rbd: drop racy assert in test_tasks_recovery()

Even though "ceph rbd task list" is executed immediately after
a successful "ceph rbd task add flatten", the operation may complete
in the interim and the task listing may come back empty legitimately.
Given that we are asserting that flatten actually occurs based on
"rbd info" output, there is no real need to try to briefly observe
the flatten task in the task list.

Fixes: https://tracker.ceph.com/issues/75209
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

Merge pull request #67877 from aainscow/fix_makefile_75537

src: Move the decision to build the ISA plugin to the top level makefile

Merge pull request #65404 from MaxKellermann/msg__includes

msg: include cleanup

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

Merge pull request #67405 from cbodley/wip-75008

rgw: `account rm --purge-data` can delete users/roles/groups/oidcs too

Reviewed-by: Krunal Chheda <kchheda3@bloomberg.net>

doc: add notable/security changes in v18.2.8

AI-Assisted: scan of changeset to indicate notable changes
Signed-off-by: Patrick Donnelly <pdonnell@ibm.com>

doc: add known issues for v18.2.8

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

doc: add reef release notes

Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
Signed-off-by: Patrick Donnelly <pdonnell@ibm.com>

script: fix argument change for github module

    Traceback (most recent call last):
        File "/home/batrick/scm/ceph/./src/script/ceph-release-notes", line 400, in <module>
            gh = github.GitHub(
            access_token=args.token)
    TypeError: GitHub.__init__() got an unexpected keyword argument 'access_token'

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

Merge PR #67632 into main

* refs/pull/67632/head:
pybind/mgr: call new _ceph_exit for killpoints
mgr: add ceph binding for exit

Reviewed-by: Nitzan Mordechai <nmordech@redhat.com>

src: Move the decision to build the ISA plugin to the top level make file

Previously, the first time you build ceph, common did not see the correct
value of WITH_EC_ISA_PLUGIN.  The consequence is that the global.yaml gets
build with osd_erasure_code_plugins not including isa.  This is not great
given its our default plugin.

We considered simply removing this parameter from make entirely, but this
may require more discussion about supporting old hardware.

So the slightly ugly fix is to move this erasure-code specific declartion
to the top-level.

Fixes: https://tracker.ceph.com/issues/75537
Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>

Merge pull request #67643 from rhcs-dashboard/75288-bugfix

mgr/dashboard: Breadcrumb should allow going back to subsystem tab

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

msg/async/crypto_onwire: include cleanup

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

msg/async/frames_v2: include cleanup

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

msg/Message: do not include Connection.h in Crimson mode

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

msg/compressor_registry: include cleanup

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

msg/msg_types: un-inline methods to reduce header dependencies

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

include/Context: add missing include

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

java/native/libcephfs_jni: add missing include

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

Merge pull request #64439 from myoungwon/wip-logstoremanager

crimson/os/seastore: Introduce LogStoreManager

Reviewed-by: Josh Durgin <jdurgin@redhat.com>

qa: test case for evicting clients with id having multiple number of consecutive zeroes

Fixes: https://tracker.ceph.com/issues/70198
Signed-off-by: neeraj pratap singh <neerajpratapsingh@li-ff7f0d4c-3462-11b2-a85c-d4004c0fa1a0.ibm.com>

mds: add a proper check to evict clients with id 0

Fixes: https://tracker.ceph.com/issues/70198
Signed-off-by: Neeraj Pratap Singh <neesingh@redhat.com>

mgr/dashboard: [storage-class]: Deleting local storage class from UI does not remove its entry from zone

Fixes: https://tracker.ceph.com/issues/75541
Signed-off-by: Dnyaneshwari Talwekar <dtalwekar@redhat.com>

mds: Remove unnecessary std::move on trivially-copyable type LockOp in add_lock method

Fixes: https://tracker.ceph.com/issues/73431
Signed-off-by: Edwin Rodriguez <edwin.rodriguez1@ibm.com>

Merge pull request #66026 from lvshuo2016/featurecrc-new

src/common: add crc32c support for riscv64

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

Merge pull request #67815 from rhcs-dashboard/helper-spacing-fix

mgr/dashboard: add helper text to bucket form > policy and other spacing fixes

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

Merge pull request #67847 from rhcs-dashboard/fix-services-page

mgr/dashboard: services page breaking for test_orchestrator module

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

Merge pull request #67544 from tchaikov/wip-fix-unittest-peeringstate-leaks

test/osd: Fix unittest_peeringstate message leaks:

Reviewed-by: Bill Scales <bill_scales@uk.ibm.com>

Merge pull request #67807 from afreen23/overview-content

mgr/dashboard: Stabilizing the storage overview card

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

mgr/dashboard: add helper text to bucket form > policy and other spacing fixes

Fixes: https://tracker.ceph.com/issues/75525
Signed-off-by: Naman Munet <naman.munet@ibm.com>

Merge pull request #65509 from MaxKellermann/mon_includes

mon: reduce header dependencies

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

Merge pull request #67865 from tchaikov/doc-runtime-updatable

doc: show runtime-updatable config metadata

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

doc: show runtime-updatable config metadata

Extend the confval Sphinx extension to render whether an option
can be updated at runtime.

Use the same policy as Option::can_update_at_runtime() so the
documentation reflects both native config options and mgr module
options consistently.

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

rgw: config options for mp lock error/delay injection

rgw_mp_lock_inject_delay can inject delays in multipart complete to make
sure lock renewal requests get issued

rgw_mp_lock_inject_renewal_error can inject lock renewal errors to test
the error handling

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

rgw/rados: MPRadosSerializer spawns lock renewal coroutine

CompleteMultipartUpload depends on this lock to ensure consistency of
uploads and protect against data loss, so we should try very hard to
hold this lock as long as it takes to complete successfully

MPRadosSerializer accomplishes this by spawning a background lock
renewal coroutine. this coroutine is started during a successful call to
try_lock(), and stopped before unlock() releases the lock

Fixes: https://tracker.ceph.com/issues/75375
Signed-off-by: Casey Bodley <cbodley@redhat.com>

common/async: async_cond Executor defaults to any_io_executor

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

rgw/sal: Object::get_serializer() takes optional_yield

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

mgr/dashboard: Add capacity thresholds

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

rgw/sal: Serializer::try_lock() takes ceph::timespan

this duration ultimately gets passed down to cls_lock's set_duration()
function, which has overloads for both utime_t and ceph::timespan.
prefer ceph::timespan because it also works with boost asio timers

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

rgw/sal: StoreMPSerializer uses atomic bool locked

make it safe for other threads to call clear_locked() and is_locked()

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