crimson/os/seastore/transaction_manager: minor refinement to remap_pin() asserts

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

test/crimson: drop cout from seastore tests

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

crimson/os/seastore/../fixed_kv_node: add comment to copy-from-source

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

crimson/os/seastore: reuse link_phy_tree_root_node()

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

crimson/os/cache: get_extent_viewable_by_trans, the extent should be valid

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

crimson/os/cache: cleanup get_extent(_by_type)

Distinguish caching vs absent get_extent interfaces, and misc related
cleanups.

After the lba parent-child pointer optimization, the absent path should
be used.

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

crimson/os/seastore: remove unecessary public interfaces

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

crimson/os/seastore/cache: drop the unnecessary friend fixed_kv_btree

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

Merge pull request #55967 from mkogan1/wip-rgw-ssl-ses-id-reuse

rgw/beast: Enable SSL session-id reuse speedup mechanism

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

Merge pull request #55895 from linuxbox2/wip-sigv4-boundary

rgw_sigv4: handle '0;' as trailer-chunk boundary

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

Merge pull request #55885 from ivancich/wip-enhance-restore-bi

rgw: make rgw-restore-bucket-index more robust

Reviewed-by: Michael Kidd <linuxkidd@gmail.com>

Merge pull request #55977 from myoungwon/wip-allow-dirty-remap

crimson/os/seastore: allow to remap the dirty extent

Reviewed-by: Xuehan Xu <xuxuehan@qianxin.com>
Reviewed-by: Yingxin Cheng <yingxin.cheng@intel.com>

Merge PR #55909 into main

* refs/pull/55909/head:
qa/cephfs: add `MON_DOWN' and `deprecated feature inline_data' to ignorelist

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

Merge PR #55908 into main

* refs/pull/55908/head:
qa: ignore container checkpoint/restore related selinux denials for centos9
qa: use allowlist instead of whitelist for selinux denials

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

Merge PR #55639 into main

* refs/pull/55639/head:
mds: relax certain asserts in mdlog replay thread

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

Merge pull request #56000 from cbodley/wip-64678

rpm: disable system_qat for non-x86_64 arch

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

rpm: disable system_qat for non-x86_64 arch

Fixes: https://tracker.ceph.com/issues/64678
Signed-off-by: Casey Bodley <cbodley@redhat.com>

Merge PR #55820 into main

* refs/pull/55820/head:
qa/crontab: bring back nightlies of release branches
qa/crontab: update upgrade runs for squid/main
qa/crontab: update reef upgrade tests
qa/crontab: update quincy upgrade tests
qa/crontab: remove EOL branches
qa/crontab: add spacing
qa/crontab: update smoke tests for releases

Reviewed-by: Adam King <adking@redhat.com>
Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Ilya Dryomov <idryomov@redhat.com>
Reviewed-by: Neha Ojha <nojha@redhat.com>
Reviewed-by: Laura Flores <lflores@redhat.com>

Merge pull request #55955 from guits/node-proxy-fixes-7.1-blockers-1

mgr/node-proxy: handle 'None' statuses returned by RedFish

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

Merge pull request #55536 from Matan-B/wip-crimson-snapmapper-get_next

osd/SnapMapper: Crimson - fix OSDriver::get_next

Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: chunmei-liu <chunmei.liu@intel.com>

Merge pull request #55982 from zdover23/wip-doc-2024-03-06-dev-developer-guide-testing-integration-tests-workflow

doc/dev: edit teuthology workflow

qa/cephfs: add `MON_DOWN' and `deprecated feature inline_data' to ignorelist

Fixes: http://tracker.ceph.com/issues/64746
Signed-off-by: Venky Shankar <vshankar@redhat.com>

qa: ignore container checkpoint/restore related selinux denials for centos9

Fixes: http://tracker.ceph.com/issues/64616
Signed-off-by: Venky Shankar <vshankar@redhat.com>

qa: use allowlist instead of whitelist for selinux denials

https://github.com/ceph/teuthology/pull/1757 introduced this change,
however, qa/distros continued to use `whitelist' keyword.

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

mgr/node-proxy: handle 'None' statuses returned by RedFish

Looks like RedFish might return 'None' values for some attributes.

for instance:

```
[root@ceph-node-01 ~]# curl -s -k -X GET https://169.254.1.1/redfish/v1/Systems/System.Embedded.1/Storage/AHCI.SL.6-1/Drives/Disk.Direct.0-0:AHCI.SL.6-1 -H "X-Auth-Token: 3264251c28191fa5e7c9ebec49ef90fc"  | jq .Status
{
  "Health": "OK",
  "HealthRollup": "OK",
  "State": "Enabled"
}
[root@ceph-node-01 ~]# curl -s -k -X GET https://169.254.1.1/redfish/v1/Systems/System.Embedded.1/Storage/NonRAID.Slot.2-1/Drives/Disk.Bay.0:Enclosure.Internal.0-1:NonRAID.Slot.2-1 -H "X-Auth-Token: 3264251c28191fa5e7c9ebec49ef90fc" | jq .Status
{
  "Health": null,
  "HealthRollup": null,
  "State": "Enabled"
}
[root@ceph-node-01 ~]#
```

Although this seems to be a bug from RedFish, we need to handle
the case when it happens otherwise it makes the mgr orchestrator module
throw an error.

The idea here is to create a new status "unknown" when we can't fetch the
real status of a component.

Fixes: https://tracker.ceph.com/issues/64712
Signed-off-by: Guillaume Abrioux <gabrioux@ibm.com>

Merge pull request #52267 from xxhdx1985126/wip-crimson-pg-snapmapper-obj

crimson/osd: put snapmapper's key-value pairs into dedicated objs

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

Merge pull request #55910 from zhscn/fix-int-string

osd/SnapManager: Crimson - fix std::string constructed from 0

Reviewed-by: chunmei-liu <chunmei.liu@intel.com>
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Radosław Zarzyński <rzarzyns@redhat.com>
Reviewed-by: Matan Breizman <mbreizma@redhat.com>

Merge pull request #55801 from xxhdx1985126/wip-counting-semaphore

crimson/os/alienstore: use homebrew counting_semaphore

Reviewed-by: chunmei-liu <chunmei.liu@intel.com>
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Kefu Chai <kefu.chai@scylladb.com>

Merge PR #54202 into main

* refs/pull/54202/head:
mds: fix `dump dir` indicate dir is not cached

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

Merge PR #55457 into main

* refs/pull/55457/head:
client: check for bad file handle in low level I/O APIs
client: check for bad file handle in ll_preadv_pwritev
client: add function to check if file handle exists
src/test: test async I/O with invalid/closed file handle

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

Merge PR #55619 into main

* refs/pull/55619/head:
client: fix leak of file handles

Reviewed-by: Venky Shankar <vshankar@redhat.com>
Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Dhairya Parmar <dparmar@redhat.com>

doc/dev: edit teuthology workflow

Improve a couple of sentences in
doc/dev/developer_guide_testing_integration_tests/tests-integration-testing-teutology-workflow.rst.

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

Merge pull request #55954 from ivoalmeida/fix-cephfs-attach-command

mgr/dashboard: fixed cephfs mount command

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

Merge pull request #54964 from rhcs-dashboard/add-prometheus-federation-cli

mgr/dashboard: add prometheus federation config for mullti-cluster monitoring

Reviewed-by: Adam King <adking@redhat.com>
Reviewed-by: Ankush Behl <cloudbehl@gmail.com>
Reviewed-by: frittentheke <NOT@FOUND>
Reviewed-by: Nizamudeen A <nia@redhat.com>
Reviewed-by: Redouane Kachach <rkachach@redhat.com>

Merge pull request #55574 from ceph/feature-multi-cluster-management-monitoring

mgr/dashboard: introduce multi cluster management and monitoring in ceph dashboard

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

Merge pull request #55946 from myoungwon/wip-less-verbose-rbm-test

crimson/os/seastore: make less verbose log when testing performance on RBM

Reviewed-by: Yingxin Cheng <yingxin.cheng@intel.com>
Reviewed-by: Chunmei Liu <chunmei.liu@intel.com>

crimson/os/seastore: allow to remap the dirty extent

Signed-off-by: Myoungwon Oh <myoungwon.oh@samsung.com>

rgw/beast: enablment of SSL session-id reuse speedup mechanism

Enable the OpenSSL session-id reuse acceleration mechanism that is described in:

https://www.openssl.org/docs/man1.0.2/man3/SSL_CTX_set_session_id_context.html
SSL_CTX_set_session_id_context, SSL_set_session_id_context - set context
within which session can be reused (server side only)

Fixes: https://tracker.ceph.com/issues/64719
Signed-off-by: Mark Kogan <mkogan@redhat.com>

mgr/dashboard: introduce multi-cluster overview page

https://tracker.ceph.com/issues/64530
Signed-off-by: Nizamudeen A <nia@redhat.com>
Signed-off-by: Aashish Sharma <aasharma@redhat.com>

mgr/dashboard: fixed cephfs mount command

Fixes: https://tracker.ceph.com/issues/64716
Signed-off-by: Ivo Almeida <ialmeida@redhat.com>

Merge pull request #55905 from ronen-fr/wip-rf-23a

common: fix string creation from '0' in LogEntry

Reviewed-by: Matan Breizman <mbreizma@redhat.com>
Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Kefu Chai <tchaikov@gmail.com>

Merge pull request #55464 from rhcs-dashboard/hardware-statuses

mgr/dashboard: add hardware status summary

crimson/os/seastore: make less verbose log when testing performance on RBM

Signed-off-by: Myoungwon Oh <myoungwon.oh@samsung.com>

mgr/dashboard: add prometheus federation config for mulkti-cluster
monitoring

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

Merge pull request #55887 from xxhdx1985126/wip-seastore-omap-get-values

crimson/os/seastore: omap_get_values with range [start, "infinite") should return all results from "start"

Reviewed-by: Ronen Friedman <rfriedma@redhat.com>
Reviewed-by: Yingxin Cheng <yingxin.cheng@intel.com>
Reviewed-by: Chunmei Liu <chunmei.liu@intel.com>

Merge pull request #55889 from xxhdx1985126/wip-seastore-async-cleaner-comments

crimson/os/seastore/async_cleaner: add comments to clarify what kind of extents would be considered dead by the cleaner

Reviewed-by: Yingxin Cheng <yingxin.cheng@intel.com>
Reviewed-by: Chunmei Liu <chunmei.liu@intel.com>

Merge pull request #55840 from xxhdx1985126/wip-seastore-large-omap-nodes

crimson/os/seastore/omap_manager: expand omap tree nodes

Reviewed-by: Yingxin Cheng <yingxin.cheng@intel.com>
Reviewed-by: Chunmei Liu <chunmei.liu@intel.com>

Merge pull request #55559 from tchaikov/ceph-exporter-cleanups

ceph-exporter: cleanups

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

Merge pull request #55890 from tchaikov/wip-cmake-build-rocksdb-cc

cmake/modules/BuildRocksDB.cmake: inherit CMAKE_C_COMPILER from parent

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

Merge pull request #55918 from zdover23/wip-doc-2024-03-04-rados-ops-pools

doc/rados: link to pg setting commands

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

rgw: make rgw-restore-bucket-index more robust

This experimental tool write a series of temporary files, the combined
size of which is roughly proportional to some combination of the
number of objects in the bucket and their respective sizes. Previously
the script stored all temporary files in /tmp, which may not be
sufficiently large in many cases.

There are now two enhancements. First, using the "-t" command-line
argument, the directory for temporary files can be specified. Second,
if the script consumes all available space on the file system
containing the temporary directory, an error message to that effect is
displayed, so the user understands what corrective action to take.

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

Merge pull request #54383 from asm0deuz/bz2068026-nfs-idmap

mgr/cephadm: Allow idmap overrides in nfs-ganesha configuration

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

Merge pull request #55891 from galsalomon66/s3select_remove_assert_and_updates

rgw/S3select: remove assert from csv-parser, adding updates

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

Merge pull request #55884 from kamoltat/wip-revert-55794

Revert "src/scripts/ceph-backport.sh: Add PR checklist"
Reviewed by: Casey Bodley <cbodley@redhat.com>

Merge pull request #55587 from adk3798/label-pattern-placement

mgr/cephadm: fix placement with label and host pattern

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

Merge pull request #55694 from ceph/wip-yuriw-release-16.2.15-main

doc: 16.2.15 Release Notes

Reviewed-by: Laura Flores <lflores@redhat.com>
Reviewed-by: Ilya Dryomov <idryomov@redhat.com>
Reviewed-by: Casey Bodley <cbodley@redhat.com>

Merge pull request #55924 from zdover23/wip-doc-2024-03-04-architecture-2

doc/architecture: remove pleonasm

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

common: fix string creation from '0' in LogEntry

C++23 disallows conversion from 'int' to 'string'.
That includes returning '0' from a function that returns a string.

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

Merge pull request #55921 from zdover23/wip-doc-2024-03-04-architecture

doc/architecture: fix spelling and syntax

doc/architecture: remove pleonasm

Remove the word "detailed" in a sentence immediately following a
sentence containing the word "detail".

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

doc/architecture: fix spelling and syntax

Fix the spelling of the word "algorithm" (which was "algoritm") and make
a sentence a little more natural.

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

doc/rados: link to pg setting commands

Link to the instructions for manually setting the number of PGs per
pool, from the mention of placement groups. These instructions are
included here in response to a request from Ronen Friedman on the
occasion of the removal of links to the PGcalc (see
https://github.com/ceph/ceph/pull/55899#pullrequestreview-1912940118).

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

Merge pull request #55897 from Matan-B/wip-matanb-crimson-seastar-sub-march24

src/seastar: update seastar submodule to fix FTBFS

Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Kefu Chai <tchaikov@gmail.com>

Merge pull request #55659 from batrick/i64503

client: log debug message when requesting unmount

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

crimson/os/seastore/async_cleaner: add comments to clarify what kind of
extents would be considered dead by the cleaner

This could prevent future developers from rendering the cleaner abnormal
unexpectedly

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

osd/SnapManager: fix std::string constructed from 0

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>

mds: relax certain asserts in mdlog replay thread

The calls to journaler->is_readable() and journaler->get_error()
in MDLog::_replay_thread() will drop Journaler::lock between
invocations, so, theoretically, its possible that the initial check:

  // loop
  int r = 0;
  while (1) {
    // wait for read?
    while (!journaler->is_readable() &&
       journaler->get_read_pos() < journaler->get_write_pos() &&
       !journaler->get_error()) {
      C_SaferCond readable_waiter;
      journaler->wait_for_readable(&readable_waiter);
      r = readable_waiter.wait();
    }
    if (journaler->get_error()) {
      r = journaler->get_error();
      dout(0) << "_replay journaler got error " << r << ", aborting" << dendl;

journaler->is_readable() returned true, thereby breaking out of
the (inner) while loop and by passing the journaler->get_error()
check, and by the time this hits the next set of checks:

    if (!journaler->is_readable() &&
    journaler->get_read_pos() == journaler->get_write_pos())
      break;

    ceph_assert(journaler->is_readable() || mds->is_daemon_stopping());

It's possible that the journal is unreadable due to some error that
happened during prefetch. In short, these checks are racy.

So, remove these racy assert check along with journaler->is_readable()
check when validating the journal end and rely on the next iteration
of reading the journal for error handling.

Fixes: http://tracker.ceph.com/issues/57048
Signed-off-by: Venky Shankar <vshankar@redhat.com>

mgr/dashboard: add hardware status summary

On the landing page of the Dashboard, add the hardware status summary

Fixes:https://tracker.ceph.com/issues/64329
Signed-off-by: Pedro Gonzalez Gomez <pegonzal@redhat.com>

src/seastar: update seastar submodule to fix FTBFS

See: https://github.com/ceph/seastar/commit/d382f247629a7349afc0cea495a1cd0502dc0b22

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

using the it.length() for segment processing.
s3select submodule

Signed-off-by: Gal Salomon <gal.salomon@gmail.com>

Merge pull request #55899 from zdover23/wip-doc-2024-03-02-rados-radosgw-pgcalc

doc/rados: remove PGcalc from docs

Reviewed-by: Ronen Friedman <rfriedman@redhat.com>

doc/rados: remove PGcalc from docs

Remove mention of the "PG calc" tool from the documentation. I have
removed all mention of this in one fell swoop to help posterity restore
mention of this tool if we decide we need to do so.

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

rgw_sigv4: handle '0;' as trailer-chunk boundary

https://tracker.ceph.com/issues/64676

Signed-off-by: matt benjamin <mbenjamin@redhat.com>

Merge pull request #52540 from petrutlucian94/single_process

rbd-wnbd: use a single daemon process per host

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

cmake/modules/BuildRocksDB.cmake: inherit CMAKE_C_COMPILER from parent

if we set the CFLAGS globally, and the CFLAGS contains options
only acceptable by a certain C compiler, RocksDB could fail to
configure. for instance, if we set CXXFLAGS so it contains
`--config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg` and use clang++
as the CMAKE_CXX_COMPILER, while keep CMAKE_C_COMPILER unchanged.
RocksDB would fail to configure like:

```-- Check for working C compiler: /usr/bin/cc - broken
CMake Error at /usr/share/cmake/Modules/CMakeTestCCompiler.cmake:67 (message):
  The C compiler

    "/usr/bin/cc"

  is not able to compile a simple test program.

  It fails with the following output:

    Change Dir: /home/jenkins-build/build/workspace/ceph-dev-new-build/ARCH/x86_64/AVAILABLE_ARCH/x86_64/AVAILABLE_DIST/centos9/DIST/centos9/MACHINE_SIZE/gigantic/release/19.0.0-1717-g0f726187/rpm/el9/BUILD/ceph-19.0.0-1717-g0f726187/redhat-linux-build/src/rocksdb/CMakeFiles/CMakeScratch/TryCompile-RU5UFV

    Run Build Command(s):/usr/bin/cmake -E env VERBOSE=1 /usr/bin/gmake -f Makefile cmTC_65e36/fast && gmake[3]: Entering directory '/home/jenkins-build/build/workspace/ceph-dev-new-build/ARCH/x86_64/AVAILABLE_ARCH/x86_64/AVAILABLE_DIST/centos9/DIST/centos9/MACHINE_SIZE/gigantic/release/19.0.0-1717-g0f726187/rpm/el9/BUILD/ceph-19.0.0-1717-g0f726187/redhat-linux-build/src/rocksdb/CMakeFiles/CMakeScratch/TryCompile-RU5UFV'
    /usr/bin/gmake  -f CMakeFiles/cmTC_65e36.dir/build.make CMakeFiles/cmTC_65e36.dir/build
    gmake[4]: Entering directory '/home/jenkins-build/build/workspace/ceph-dev-new-build/ARCH/x86_64/AVAILABLE_ARCH/x86_64/AVAILABLE_DIST/centos9/DIST/centos9/MACHINE_SIZE/gigantic/release/19.0.0-1717-g0f726187/rpm/el9/BUILD/ceph-19.0.0-1717-g0f726187/redhat-linux-build/src/rocksdb/CMakeFiles/CMakeScratch/TryCompile-RU5UFV'
    Building C object CMakeFiles/cmTC_65e36.dir/testCCompiler.c.o
    /usr/bin/cc   -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security  -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64-v2 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection  -fPIE -o CMakeFiles/cmTC_65e36.dir/testCCompiler.c.o -c /home/jenkins-build/build/workspace/ceph-dev-new-build/ARCH/x86_64/AVAILABLE_ARCH/x86_64/AVAILABLE_DIST/centos9/DIST/centos9/MACHINE_SIZE/gigantic/release/19.0.0-1717-g0f726187/rpm/el9/BUILD/ceph-19.0.0-1717-g0f726187/redhat-linux-build/src/rocksdb/CMakeFiles/CMakeScratch/TryCompile-RU5UFV/testCCompiler.c
    cc: error: unrecognized command-line option ‘--config’; did you mean ‘-mpconfig’?
    gmake[4]: *** [CMakeFiles/cmTC_65e36.dir/build.make:78: CMakeFiles/cmTC_65e36.dir/testCCompiler.c.o] Error 1
```

where RocksDB tries to check C compiler -- /usr/bin/cc along with
the said CFLAGS, and fails to compile the test C program, because
GCC does not support this option.

so, in this change, let's pass the CMAKE_C_COMPILER as well.

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

Merge pull request #55787 from tchaikov/wip-cmake-liburing-2.5

cmake: bump liburing from 0.7 to 2.5

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

crimson/os/seastore: omap_get_values with range [start, "infinite")
should return all results from "start"

Currently, this range retrieval has a default max result size, which is
wrong

Signed-off-by: Xuehan Xu <xxhdx1985126@gmail.com>

cmake: bump liburing from 0.7 to 2.5

this allows us to use newer liburing features. Seastar is using
some of them which are not provided by liburing 0.7.

in this change, `--use-libc` is passed to configure. otherwise
it does not link against libc, and the symbles like memset()
won't be available when compiling liburing.so with -fPIC using
clang, which does not pull libc in that case.

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

Merge pull request #55878 from athanatos/sjust/wip-seastar-module

crimson: update seastar submodule to fix prometheus build error

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

crimson/.../interruptible_future: remove SEASTAR_CONCEPT guard

Seastar commit 8dc3398a removed this macro, no longer necessary.

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

src/seastar: update seastar submodule to fix prometheus build failure

Fixes: https://tracker.ceph.com/issues/64589
Signed-off-by: Samuel Just <sjust@redhat.com>

Revert "src/scripts/ceph-backport.sh: Add PR checklist"

This reverts commit 4aba45e2374168e4485f31fea4fa1c441c35443e

commit causes backport script to fail
reverting and recommiting is the best option

Fixes: https://tracker.ceph.com/issues/64674
Signed-off-by: Kamoltat <ksirivad@redhat.com>

Merge pull request #55856 from dmick/wip-workflow-update

.github/workflows/create-backport-trackers.yml: update actions

Merge pull request #55869 from zdover23/wip-doc-2024-03-01-install-manual-radosgw

doc/install: add manual RADOSGW install procedure

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

.github/workflows/create-backport-trackers.yml: update versions of actions

Getting warning about node16 being deprecated.  The workflow doesn't use node
directly, but through the external actions.  Moving to node20 requires
changing setup-python version; Bhacaz/checkout-files is deprecated and
recommends actions/checkout.

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

doc/install: add manual RADOSGW install procedure

Add a manual RADOSGW installation procedure to
doc/install/manual-deployment.rst. This procedure was developed by Janne
Johansson and reported to the ceph-users mailing list on 29 Jan 2024
here: https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/message/LB3YRIKAPOHXYCW7MKLVUJPYWYRQVARU/

Co-authored-by: Janne Johansson <icepic.dz@gmail.com>
Co-authored-by: Anthony D'Atri <anthony.datri@gmail.com>
Signed-off-by: Zac Dover <zac.dover@proton.me>

Merge pull request #55797 from ajarr/wip-64574

qa: add diff-continuous and compare-mirror-image tests to rbd and krbd suites respectively

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

rbd-wnbd: use the right AdminSocket instance

The rbd-wnbd daemon currently caches one rados context per cluster.
However, it's registering hooks against the global context
admin socket, which won't be available. For this reason,
the "rbd-wnbd stats" command no longer works.

To address this issue, we'll ensure that rbd-wnbd sets command hooks
against the right admin socket instance, leveraging the image
context.

Signed-off-by: Lucian Petrut <lpetrut@cloudbasesolutions.com>

rbd-wnbd: adjust admin socket hook to accept image path

For each rbd-wnbd mapping we set an admin socket hook that can
be used to retrieve IO stats.

Now that the same daemon is reused for multiple mappings, we need
to distinguish the images when receiving a "stats" request.

For this reason, we'll add the image identifier to "wnbd stats"
admin socket commands.

Signed-off-by: Lucian Petrut <lpetrut@cloudbasesolutions.com>

qa: update rbd-wnbd test, retrying image rm operations

The "rbd-wnbd unmap" command is currently telling the WNBD driver
to remove the mapping without contacting the rbd-wnbd daemon
and waiting for it to perform its cleanup.

For this reason, attempting to delete the image immediately after
unmapping it can fail due to existing watchers.

As a temporary solution, we'll retry the image remove operation.
At a later time, we'll update the "rbd-wnbd unmap" command to go
through the rbd-wnbd daemon, ensuring that all the necessary
cleanup is performed before returning.

While at it, we're dropping a redundant LOG.error call so that we
won't print expected exceptions.

Signed-off-by: Lucian Petrut <lpetrut@cloudbasesolutions.com>

rbd-wnbd: update registry settings handling

This commit will store the mapping config in the Windows registry
only after initializing the mapping. This ensures that we aren't
replacing the registry settings for already mapped images.

We'll also check if the registry setting was added by us before
cleaning it up.

Signed-off-by: Lucian Petrut <lpetrut@cloudbasesolutions.com>

rbd-wnbd: use one daemon process per host

We're currently using one rbd-wnbd process per image mapping.
Since OSD connections aren't shared across those processes,
we end up with an excessive amount of TCP sessions, potentially
exceeding Windows limits:
https://ask.cloudbase.it/question/3598/ceph-for-windows-tcp-session-count/

In order to improve rbd-wnbd's scalability, we're going to use
a single process per host (unless "-f" is passed when mapping the
image, in which case the daemon will run as part of the same
process). This allows OSD sessions to be shared across image
mappings.

Another advantage is that the "ceph-rbd" service starts faster,
especially when having a large number of image mappings.

Signed-off-by: Lucian Petrut <lpetrut@cloudbasesolutions.com>

rbd-wnbd: introduce RbdMapping class

We're moving most of the WNBD mapping handling to a separate
class called RbdMapping. This simplifies cleanup and makes it
easier to reuse.

The WnbdHandler class covers WNBD specific operations and IO
callbacks while the RbdMapping wrapper will take care of RBD
operations.

A subsequent change will make use of it while switching from
one process per mapping to a single process per host.

While at it, we're also moving the rbd-wnbd config helpers
to separate files.

Signed-off-by: Lucian Petrut <lpetrut@cloudbasesolutions.com>

Merge pull request #54767 from climb-mountain123/worm_multipart

src/rgw: fix for the multipart interface in the WORM function

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

mgr/cephadm: fix placement with label and host pattern

Previously, when both the label and host pattern were
provided, only the label was actually used for the placement

Fixes: https://tracker.ceph.com/issues/64428
Signed-off-by: Adam King <adking@redhat.com>

Merge pull request #55719 from phlogistonjohn/jjm-teuth-tasks-cephadm-jt

qa/tasks/cephadm: add generic templating where subst_vip was used

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

doc: 16.2.15 Release Notes

Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
Signed-off-by: Laura Flores <lflores@ibm.com>
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>

rgw: remove unused object lock stuff from CompleteMultipart

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

Merge pull request #55817 from rkhudov/src-test-common-test_hobject-remove-constexpr

src/test/common/test_hobject: remove constexpr

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

Merge pull request #55727 from cbodley/wip-64549

rgw/auth: do_aws4_auth_completion() catches exceptions

Reviewed-by: Matt Benjamin <mbenjamin@redhat.com>