mgr/cephadm: add keep-alive requests to ssh connections

Fixes: https://tracker.ceph.com/issues/51733
Signed-off-by: Adam King <adking@redhat.com>

Merge pull request #45484 from cfsnyder/wip-53835-pacific

pacific: rgw: document S3 bucket replication support

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

Merge pull request #45483 from cfsnyder/wip-53818-pacific

pacific: rgw: RadosBucket::get_bucket_info() updates RGWBucketEnt

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

Merge pull request #45480 from cfsnyder/wip-53654-pacific

pacific: rgw: init bucket index only if putting bucket instance info succeeds

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

Merge pull request #45478 from cfsnyder/wip-53639-pacific

pacific: rgw/amqp: add default case to silence compiler warning

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

Merge pull request #45463 from cfsnyder/wip-53254-pacific

pacific: rgw: fix rgw.none statistics

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

Merge pull request #45596 from neha-ojha/wip-45512-pacific

pacific: ceph/admin: s/master/main

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

Merge pull request #45532 from idryomov/wip-fix-schedule-status-pacific

pacific: mgr/rbd_support: cast pool_id from int to str when collecting LevelSpec

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

rgw: fix rgw.none statistics

Only unaccount for objects being removed if they exist.

Also, clean up some formatting.

Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
(cherry picked from commit 95b662fa176cd841e26b76341f91bf2e6c266003)

ceph/admin: s/master/main

This PR changes the name "master" to "main" so
that builds (and, I assume, a great many other
things) will not fail.

Signed-off-by: Zac Dover <zac.dover@gmail.com>
(cherry picked from commit 6a1dd3a8a2f3dc9fe8615d402c9041273516ff89)

Merge pull request #45426 from cfsnyder/wip-51348-pacific

pacific: rgw/notification: send correct size in COPY events

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

Merge pull request #45236 from k0ste/wip-54451-pacific

pacific: mgr/prometheus: Added `avail_raw` field for Pools DF Prometheus mgr module

Reviewed-by: Neha Ojha <nojha@redhat.com>

Merge pull request #45397 from kamoltat/wip-ksirivad-pacific-backport-45078

pacific: mon/MonCommands.h: fix target_size_ratio range

Reviewed-by: Neha Ojha <nojha@redhat.com>

Merge pull request #45376 from ronen-fr/wip-rf-43303-pacific

pacific: osd: fix the truncation of an int by int division

Reviewed-by: Neha Ojha <nojha@redhat.com>

Merge pull request #45373 from smithfarm/wip-54276-pacific

pacific: common: fix FTBFS due to dout & need_dynamic on GCC-12

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

Merge pull request #45323 from ljflores/wip-54466-pacific

pacific: osd: require osd_pg_max_concurrent_snap_trims > 0

Reviewed-by: Neha Ojha <nojha@redhat.com>
Reviewed-by: Dan van der Ster <daniel.vanderster@cern.ch>

Merge pull request #45319 from ljflores/wip-52078-pacific

pacific: test: fix wrong alarm (HitSetWrite)

Reviewed-by: Myoungwon Oh <omwmw@sk.com>

Merge pull request #45152 from chrisphoffman/wip-53644-pacific

pacific: mon: Omit MANY_OBJECTS_PER_PG warning when autoscaler is on

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

Merge pull request #43698 from cfsnyder/wip-52636-pacific

pacific: mon: do not quickly mark mds laggy when MON_DOWN

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

Merge pull request #45459 from cfsnyder/wip-53079-pacific

pacific: src/rgw: Fix for malformed url

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

Merge pull request #45457 from cfsnyder/wip-53036-pacific

pacific: rgw: cls_bucket_list_unordered() might return one redundent entry every time is_truncated is true

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

Merge pull request #45442 from cfsnyder/wip-52107-pacific

pacific: radosgw-admin: 'sync status' is not behind if there are no mdlog entries

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

Merge pull request #45440 from cfsnyder/wip-52071-pacific

pacific: rgw: add the condition of lock mode conversion to PutObjRentention

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

Merge pull request #45427 from cfsnyder/wip-51351-pacific

pacific: rgw/amqp: remove the explicit "disconnect()" interface

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

Merge pull request #45445 from cfsnyder/wip-52115-pacific

pacific: qa/rgw: update apache-maven mirror for rgw/hadoop-s3a

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

Merge pull request #45439 from cfsnyder/wip-52050-pacific

pacific: rgw: default ms_mon_client_mode = secure

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

Merge pull request #45434 from cfsnyder/wip-51777-pacific

pacific: rgw: forward request in multisite for RGWDeleteBucketPolicy and RGWDeleteBucketPublicAccessBlock

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

Merge pull request #45430 from cfsnyder/wip-51701-pacific

pacific: rgw: url_decode before parsing copysource in copyobject

Merge pull request #43956 from cfsnyder/wip-52782-pacific

pacific: rgw/sts: adding role name and role session to ops log.

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

Merge pull request #45455 from cfsnyder/wip-52990-pacific

pacific: rgw: have "bucket check --fix" fix pool ids correctly

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

Merge pull request #45461 from cfsnyder/wip-53158-pacific

pacific: rgw:When KMS encryption is used and the key does not exist, we should…

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

Merge pull request #45453 from cfsnyder/wip-52988-pacific

pacific: rgw: document rgw_lc_debug_interval configuration option

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

Merge pull request #45450 from cfsnyder/wip-52611-pacific

pacific: rgw: clean-up logging of function entering to make thorough and consistent

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

Merge pull request #45448 from cfsnyder/wip-52348-pacific

pacific: rgw: change order of xml elements in ListRoles response

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

Merge pull request #45444 from cfsnyder/wip-52113-pacific

pacific: qa/rgw: barbican and pykmip tasks upgrade pip before installing pytz

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

Merge pull request #45438 from cfsnyder/wip-51810-pacific

pacific: RGW - Zipper - Make default args match in get_obj_state

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

Merge pull request #45432 from cfsnyder/wip-51703-pacific

pacific: rgw: fix md5 not match for RGWBulkUploadOp upload when enable rgw com…

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

Merge pull request #45428 from cfsnyder/wip-51352-pacific

pacific: rgw: write meta of a MP part to a correct pool

Reviewed-by: Daniel Gryniewicz <dang@redhat.com>

Merge pull request #45421 from cfsnyder/wip-54426-pacific

pacific: rgw: add OPT_BUCKET_SYNC_RUN to gc_ops_list, so that

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

Merge pull request #45413 from cfsnyder/wip-54035-pacific

pacific: qa: move certificates for kmip task into /etc/ceph

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

Merge pull request #45411 from cfsnyder/wip-54494-pacific

pacific: rgw: fix segfault in UserAsyncRefreshHandler::init_fetch

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

Merge pull request #44729 from cfsnyder/wip-53849-pacific

pacific: rgwlc:  remove lc entry on bucket delete

Reviewed-by: Daniel Gryniewicz <dang@redhat.com>

Merge pull request #44885 from BenoitKnecht/wip-54118-pacific

pacific: mgr: Fix ceph_daemon label in ceph_rgw_* metrics

Reviewed-by: BenoitKnecht <NOT@FOUND>
Reviewed-by: Avan Thakkar <athakkar@redhat.com>
Reviewed-by: Pere Diaz Bou <pdiazbou@redhat.com>

Merge pull request #45351 from nmshelke/wip-54478-pacific

pacific: ceph-fuse: perform cleanup if test_dentry_handling failed

Reviewed-by: Venky Shankar vshankar@redhat.com
Reviewed-by: Kotresh HR khiremat@redhat.com

Merge pull request #45227 from rkachach/backport_of_pr_44928

pacific: mgr/cephadm: validating service_id for MDS

Reviewed-by: Michael Fritch <mfritch@suse.com>
Reviewed-by: Adam King adking@redhat.com

Merge pull request #45205 from kotreshhr/pacific-mgr-volumes-backports-0

pacific: mgr/volumes: A few mgr volumes pacific backports

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

qa/workunits/rbd/cli_generic.sh: relax trash purge schedule status assert

Commit 08df6e0fd006 ("qa/workunits/rbd: expand LevelSpec parsing
coverage") didn't account for images with a separate data pool.  This
was missed because of small-cache-pool.yaml breakage.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 90a90ad47dd3140c796ef4da7263c9633d34e841)

rbd: fix a typo in "rbd trash purge schedule ls" formatted output

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit beffb7f9c6a4a19b6c4cfb65d04386575af50961)

qa/workunits/rbd: expand LevelSpec parsing coverage

Invoke "rbd mirror snapshot schedule ls -R" and "rbd mirror snapshot
schedule status" commands on all levels, consistently.  In particular,
make sure that an image level schedule is listed for a recursive query
at the pool level both before and after the schedule kicks in:

  $ rbd create --size 1G --mirror-image-mode snapshot -p foo bar
  $ rbd mirror snapshot schedule add -p foo --image bar 1m
  $ rbd mirror snapshot schedule ls -p foo -R
  POOL  NAMESPACE  IMAGE  SCHEDULE
  foo              bar    every 1m
  <wait for schedule to become visible in status>
  $ rbd mirror snapshot schedule ls -p foo -R
  POOL  NAMESPACE  IMAGE  SCHEDULE
  foo              bar    every 1m

Also, make sure that pool and image level status queries work:

  $ rbd mirror snapshot schedule status -p foo
  SCHEDULE TIME        IMAGE
  2022-03-04 07:14:00  foo/bar
  $ rbd mirror snapshot schedule status -p foo --image bar
  SCHEDULE TIME        IMAGE
  2022-03-04 07:14:00  foo/bar

Both of these issues are fixed by the previous commit.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 08df6e0fd00689649b802e975df320efe3b07c83)

doc: rgw: document S3 bucket replication support

Support was added at Octopus.

Fixes: https://tracker.ceph.com/issues/48755
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
(cherry picked from commit 774a247b2b854538b679490581e6950372142797)

rgw: RadosBucket::get_bucket_info() updates RGWBucketEnt

if we construct a RadosBucket without knowing its instance id,
get_bucket_info() will look it up for us and store it in
info.bucket.bucket_id

but RadosBucket::update_container_stats() gets this instance id from
'RGWBucketEnt ent', which wasn't updated by get_bucket_info(). this
causes update_container_stats() to try to read a bucket instance with
empty instance id, which fails with -ENOENT, and the RGWListBucket op
returns uninitialized X-Container-* stats

Fixes: https://tracker.ceph.com/issues/47781
Fixes: https://tracker.ceph.com/issues/47861
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit e391e56c7dda24b4a9d6981a5be31fe8b9defd9d)

rgw: init bucket index only if putting bucket instance info succeeds

Signed-off-by: Huber-ming <zhangsm01@inspur.com>
(cherry picked from commit 6e97f2a32df80f00d44ed3daceac381c46c17026)

rgw/amqp: add default case to silence compiler warning

Fixes: https://tracker.ceph.com/issues/53252
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit a32ae4b61ac4af244533ce53388411ca50fc1148)

rgw:When KMS encryption is used and the key does not exist, we should not throw ERR_ INVALID_ ACCESS_ Key error code.

When kms encryption is used, the key_id is null or the actual_key size is wrong, we should not throw "ERR_INVALID_ACCESS_KEY " error code, instead of "EINVAL"error code, is used to indicate parameter error.

Signed-off-by: wangyingbin <wangyingbin@inspur.com>
(cherry picked from commit 40dbc29984d67a3f4946a0b30d53f3db19952bf0)

src/rgw: Fix for malformed url

This PR solves: https://tracker.ceph.com/issues/52738
It is solved by making changes to rgw_url.cc
A test is also added to check it's working.

Signed-off-by: Kalpesh Pandya <kapandya@redhat.com>
(cherry picked from commit 2916f2439eb2f62bc08c3e283b13391302b3e497)

rgw: update last_added_entry when count == num_entries

RGWRados::cls_bucket_list_unordered() will produce one redundent entry
every time is_truncated is true.The issue could be easily reproduced
when a bucket is filled with amounts of incomplete multipart upload.
To be more specific, the number of incomplete multipart upload objects
should be greater than 1100.

Signed-off-by: Peng Zhang <zhangpeng@vclusters.com>
(cherry picked from commit 7511f9f675ea4e43992605dc03109bc5f356a5e1)

rgw: during "bucket check --fix" set index entry pool id correctly

The current code sets the pool id of bucket index entries to the
bucket index pool rather than the data pool. This fixes that.

Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
(cherry picked from commit 4c2ac79621c7d110d9b3664ef5ce2027d817560c)

Conflicts:
src/rgw/rgw_rados.cc

Cherry-pick notes:
- if condition under changes did not use braces in Pacific but does in later releases

rgw: document rgw_lc_debug_interval configuration option

Updates the yaml describing this config option with a "desc" and a
"long_desc".

Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
(cherry picked from commit 9171d3626b5a0181456a68555d5742109abaabbc)

Conflicts:
src/common/options/rgw.yaml.in

Cherry-pick notes:
- options not defined in yaml in Pacific

rgw: clean-up logging of function entering to make thorough and consistent

This provides more thorough and consistent function tracing in CLS/RGW
when logging is set to 10 or higher.

Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
(cherry picked from commit 9982e152095b6e7f94ed59afb5fd976f04dccd34)

Conflicts:
src/cls/rgw/cls_rgw.cc

Cherry-pick notes:
- new lines added to logs

rgw: change order of xml elements in ListRoles response

one or more AWS sdks fail to parse our response because they expect
ListRolesResult to come first. not really an rgw bug, but it's easy
enough to fix

Fixes: https://tracker.ceph.com/issues/52027
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 5f1f4212a81a7b124d657d4653de39ee85051964)

qa/rgw: update apache-maven mirror for rgw/hadoop-s3a

Fixes: https://tracker.ceph.com/issues/52069
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 9253733d0883d01988b163ee22cfc3481c01a52d)

qa/rgw: barbican and pykmip tasks upgrade pip before installing pytz

Downloading https://files.pythonhosted.org/packages/9b/77/461087a514d2e8ece1c975d8216bc03f7048e6090c5166bc34115afdaa53/cryptography-3.4.7.tar.gz (546kB)
  Complete output from command python setup.py egg_info:

          =============================DEBUG ASSISTANCE==========================
          If you are seeing an error here please try the following to
          successfully install cryptography:

          Upgrade to the latest pip and try again. This will fix errors for most
          users. See: https://pip.pypa.io/en/stable/installing/#upgrading-pip
          =============================DEBUG ASSISTANCE==========================

  Traceback (most recent call last):
    File "<string>", line 1, in <module>
    File "/tmp/pip-build-7fhnk5us/cryptography/setup.py", line 14, in <module>
      from setuptools_rust import RustExtension
  ModuleNotFoundError: No module named 'setuptools_rust'

Fixes: https://tracker.ceph.com/issues/52070
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit e5a5b4e3794baf3f6bc86ed0c02ecfc8cc0f99b4)

radosgw-admin: 'sync status' is not behind if there are no mdlog entries

if remote mdlogs are trimmed prematurely, sync status will report
that it's behind the remote's max-marker even if there are no mdlog
entries to sync

for each behind shard, we fetch the next mdlog entry from the remote. if
we get an empty listing, remove that shard from behind_shards. this
logic now has to run before we print "behind shards:" so that empty
shards aren't listed

Fixes: https://tracker.ceph.com/issues/52091
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 14d43f740d101c8d41a2ced4525bf8efd8c9d943)

rgw: add the condition of lock mode conversion to PutObjRentention

Signed-off-by: wangzhong <wangzhong@cmss.chinamobile.com>
(cherry picked from commit 585eae46a2bb7ed39ca2f3213801e0f77642c9d4)

rgw: default auth_client_required=cephx

This makes this warning go away:

2021-08-09T15:51:52.882+0000 7f2373837400 -1 warn_if_insecure(): WARNING: rgw is configured to optionally allow insecure connections to the monitors (auth_supported, ms_mon_client_mode), ssl certificates stored at the monitor configuration could leak

7e22d2a31d277ab3eecff47b0864b206a32e2332 only fixed half of the problem.

Signed-off-by: Sage Weil <sage@newdream.net>
(cherry picked from commit 6010eae66cd12358880fff64e2229a4265a5cf0a)

rgw: default ms_mon_client_mode = secure

Require a secure connection to the monitor.  This is important because
we may store SSL certs on the mon.

Note that secure mode was introduce before Nautilus, so this can be
backported at least to Pacific.

Fixes: https://tracker.ceph.com/issues/52000
Signed-off-by: Sage Weil <sage@newdream.net>
(cherry picked from commit 7e22d2a31d277ab3eecff47b0864b206a32e2332)

RGW - Zipper - Make default args match in get_obj_state

The default arg for follow_olh was different between the base class and
the Rados class.  Fix it so they're the same.

Fixes: https://tracker.ceph.com/issues/51677
Signed-off-by: Daniel Gryniewicz <dang@redhat.com>
(cherry picked from commit 664fe75c2a88cbcc27575649a6900640f4fd90b5)

Conflicts:
src/rgw/rgw_sal.h

Cherry-pick notes:
- changes to other parts of get_obj_state method signature

rgw: forward request in multisite for RGWDeleteBucketPolicy
 and RGWDeleteBucketPublicAccessBlock

Fix: https://tracker.ceph.com/issues/45551

Signed-off-by: yuliyang_yewu <yuliyang_yewu@cmss.chinamobile.com>
(cherry picked from commit 1f9267edd81fd1bf05bd6440c5e112a6cc042145)

rgw: fix md5 not match for RGWBulkUploadOp upload when enable rgw compression

Fixes: https://tracker.ceph.com/issues/46625
Signed-off-by: yuliyang_yewu <yuliyang_yewu@cmss.chinamobile.com>
(cherry picked from commit 3fc487d25e81ca2135729b120b9c832b42a26266)

Amend b7621625ed69f21a5bf701b3385ddee281ff3715 to not call url_decode excessively

Fixes: #43259
Signed-off-by: Paul Reece <paul@servercloud.com>
(cherry picked from commit c83afb4359b9f8b6d8b6942e74a52f303a474d54)

rgw: url_decode before parsing copysource in copyobject

If the copysource on copyobject call was URL-encoded, it would fail as it would not parse the '/' seperating bucket and key name

URL encoding may be necessary for certain characters in a copysource, and several public examples show URL encoding the copysource

Fixes: #43259
Signed-off-by: Paul Reece <paul@servercloud.com>
(cherry picked from commit b7621625ed69f21a5bf701b3385ddee281ff3715)

rgw: write meta of a MP part to a correct pool

When uploading a multipart part to a non-default storage class,
original logic writes the object data to the correct pool
for its storage class.
But when completing the upload, there's another write op
(with a bunch of setxattrs) to the head object targeting
the standard pool (because non-multipart uploads
write their head objects to the default pool).
As a result of this setxattrs op, a redundant zero-sized object is
created in a wrong pool. Multipart complete, multipart abort and
GC would never try to it clean up.

The fix is to make RGWRados::Object::Write::_do_write_meta() learn the
difference between multipart and non-multipart head objects and pick the
correct placement rule for RGWRados::get_obj_head_ref().

Fixes: http://tracker.ceph.com/issues/49128
Signed-off-by: Jeegn Chen <jeegnchen@tencent.com>
(cherry picked from commit 7647823bd42ce8228c3622a91e87680ee22df710)

Conflicts:
src/rgw/rgw_sal.h

Cherry-pick notes:
- RGWAttrs renamed to Attrs after pacific

rgw/amqp: remove the explicit "disconnect()" interface

this is to resolve a possible race condistion.
note that the removed interface is not used in RGW code.
and that connection cleanup should be done based on idleness and
not on explicit disconnect. see: https://tracker.ceph.com/issues/49033

Fixes: https://tracker.ceph.com/issues/51114
Signed-off-by: Yuval Lifshitz <ylifshit@redhat.com>
(cherry picked from commit a3849aaa9f5a3e95b1e6ce0a80b1df33b3a77cf9)

rgw/amqp: fix valgrind issue of uninitialized info.ssl

Signed-off-by: Yuval Lifshitz <ylifshit@redhat.com>
(cherry picked from commit 5b31052be66ff8debee1e233406e6490f8206fde)

rgw/notification: send correct size in COPY events

Fixes: https://tracker.ceph.com/issues/51305
Signed-off-by: Yuval Lifshitz <ylifshit@redhat.com>
(cherry picked from commit 17cc2a4afcd44c601901216f83f7404f4850fb9a)

Conflicts:
src/rgw/rgw_op.cc
src/test/rgw/bucket_notification/test_bn.py

Cherry-pick notes:
- src/rgw/rgw_op.cc: conflicts due to prior backport of https://github.com/ceph/ceph/pull/42155
- src/test/rgw/bucket_notification/test_bn.py: changed tests were in a different test file in Pacific

rgw: adding OPT_DATA_SYNC_RUN to gc_ops_list so that gc gets
initialized for this command.

Fixes: https://tracker.ceph.com/issues/54433
Signed-off-by: Pritha Srivastava <prsrivas@redhat.com>
(cherry picked from commit 364f997e63030c28229757cf6221f8d3bf8b1686)

Conflicts:
src/rgw/rgw_admin.cc

Cherry-pick notes:
- Conflicts due to moved code

rgw: add OPT_BUCKET_SYNC_RUN to gc_ops_list, so that
gc is initialised and send_chain does not crash.

Also deleting objects inline in case gc is uninitialised.

Fixes: https://tracker.ceph.com/issues/54417
Signed-off-by: Pritha Srivastava <prsrivas@redhat.com>
(cherry picked from commit aa3006ea34e301148779f6055ee3fa045dabbf7e)

Conflicts:
src/rgw/rgw_admin.cc

Cherry-pick notes:
- Conflicts due to differences in ops lists

rgwlc:  remove lc entry on bucket delete

Buckets with lifecycle policies installed have a state entry that
must also be deleted when the bucket is removed.

Fixes: https://tracker.ceph.com/issues/46728
N.b., should really be generic, not specific to the RADOS store, but
there doesn't seem to be a clean model for implementing generic side
effects in Zipper, currently.

Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
(cherry picked from commit cc1e812a003e2af74fe0c69ccae08dd7aa68bbe0)

Conflicts:
src/rgw/rgw_sal_rados.cc

Cherry-pick notes:
- must pass this->info as first param to remove_bucket_config
- conflict with includes

qa: move certificates for kmip task into /etc/ceph

On rhel/centos the ceph user does not have permission
to access these certs which leads to s3-test failures
in teuthology.

Signed-off-by: Ali Maredia <amaredia@redhat.com>
(cherry picked from commit 99f0e82a9590ea20804651e0a8422fd895800ae3)

rgw: fix segfault in UserAsyncRefreshHandler::init_fetch

Fixes a segfault that was occuring in error handling code of UserAsyncRefreshHandler::init_fetch.
When ruser->read_stats_async returned an error code, the instance of UserAsyncRefreshHandler had
already been deallocated in RGWSI_User_RADOS::read_stats_async and a segmentation fault occurs
when attempting to print a member variable in error logs. This commit removes the extra ref count
drop since the ref is properly dropped upstream in RGWQuotaCache::async_refresh error handling
logic.

Fixes: https://tracker.ceph.com/issues/54112
Signed-off-by: Cory Snyder <csnyder@iland.com>
(cherry picked from commit 71ef3af870e5789e71480682f11a883ff3a673e7)

Merge pull request #45365 from ronen-fr/wip-rf-42714-pacific

pacific: osd/scrub: only telling the scrubber of awaited-for 'updates' events
A backport of PR#42714.
Reviewed-by: Laura Flores <lflores@redhat.com>

Merge pull request #45194 from ronen-fr/wip-rf-45068-pacific

pacific: osd/scrub: stop sending bogus digest-update events
A backport of PR #45068, commit e1b5347.

Reviewed-by: Neha Ojha <nojha@redhat.com>

mgr/volumes: Fix clone uid/gid mismatch

This is the regression caused by commit 18b85c53a.
The 'set_attrs' function sets the uid/gid of the
group to the subvolume if uid/gid is not passed.
The attrs of the clone should match the source
snapshot. Hence, don't use the 'set_attrs'
function to set only the quota attrs for the
clone.

Fixes: https://tracker.ceph.com/issues/54066
Signed-off-by: Kotresh HR <khiremat@redhat.com>
(cherry picked from commit b3c9e6b50cf4264538e4c41d19e7ebb8b2900c3a)

mgr/volumes: Fix subvolumegroup ls

The subvolumegroup ls listed '_deleting' directory which is
internal to 'mgr/volumes' and should not be listed as a
subvolumegroup. This patch fixes the same by filtering it.

Fixes: https://tracker.ceph.com/issues/54099
Signed-off-by: Kotresh HR <khiremat@redhat.com>
(cherry picked from commit a125b0e8a22ba6c8d14f25fe85ce8d0771299c6c)

qa: Add test for subvolumegroup ls filter

Fixes: https://tracker.ceph.com/issues/54099
Signed-off-by: Kotresh HR <khiremat@redhat.com>
(cherry picked from commit 95a83efcbe7b7faf2a292889e7c7ea7fcc629749)

mgr/volumes: Inherit file quota attr to clone

The file quota attribute 'ceph.quota.max_files'
is not inherited to the cloned subvolume. This
patch fixes the same.

Fixes: https://tracker.ceph.com/issues/54121
Signed-off-by: Kotresh HR <khiremat@redhat.com>
(cherry picked from commit 72671c8ead126fdbcb39a2f179c001fb1fe43fe5)

qa: Validate file quota attrs on clone subvolume

Fixes: https://tracker.ceph.com/issues/54121
Signed-off-by: Kotresh HR <khiremat@redhat.com>
(cherry picked from commit 96c79634043b06ab8a2a1fc445998c8f64929aed)

qa/workunits/mon/pg_autoscaler.sh: clean up white space

remove white space and weird indentations

Signed-off-by: Kamoltat <ksirivad@redhat.com>
(cherry picked from commit 4add5feff657aad2afbec56a21b544bedf7f9b95)

qa/workunits/cephtool/test.sh: added test cases for target_size_ratio

Test the commands:

`osd pool create` <pool> --target_size_ratio <float>

`osd pool set` <pool> target_size_ratio <float>

`osd pool get` <pool> target_size_ratio

Signed-off-by: Kamoltat <ksirivad@redhat.com>
(cherry picked from commit 09785475f1af5050a7cae679566ac17629dfc584)

mon/OSDMonitor.cc: cannot set target_size_ratio to negative

Throw an error when user set `target_size_ratio`
to negative using the command:

`osd pool set <pool> target_size_ratio <float>`

Signed-off-by: Kamoltat <ksirivad@redhat.com>
(cherry picked from commit 1b882054ba386d026a27fcd4f3b9f38e75a531cb)

mon/MonCommands.h: fix target_size_ratio range
The `target_size_ratio` should be 0.0 -> nolimit
not limited to 0.0 -> 1.0

Fixes: https://tracker.ceph.com/issues/54316
Signed-off-by: Kamoltat <ksirivad@redhat.com>
(cherry picked from commit e5a5b81cf05e25b1e4f35ad498c7a5a1c29a7e45)

Conflicts:
src/mon/MonCommands.h
- don't add "name=yes_i_really_mean_it,type=CephBool,req=false"

Merge pull request #45381 from adk3798/pacific-crash-fixes

pacific: src/ceph-crash.in: various enhancements and fixes

Reviewed-by: Sébastien Han <seb@redhat.com>

Merge pull request #45316 from zdover23/wip-doc-pacific-2022-03-10-removing-osd-stub

pacific: doc/start: remove osd stub from hardware recs

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

src/ceph-crash.in: print stderr if available

This is not perfect, but we have seen cases where the CLI returns 0 even
on failure.
For instance:

```sh
[root@rook-ceph-crashcollector-compute-1-66bdfbd886-d2zcd /]# ceph -n client.crash crash post -i /var/lib/ceph/crash/2021-06-28T07\:47\:37.859766Z_6ffb119c-930e-4047-9cfa-a92af783cdd0/meta
malformed crash metadata: time data '2021-06-28T07:47:37.859766' does not match format '%Y-%m-%d %H:%M:%S.%f'
[root@rook-ceph-crashcollector-compute-1-66bdfbd886-d2zcd /]# echo $?
0
```

So until we find the root cause, let's mitigate and perhaps accomodate
for futur similar issues.

Signed-off-by: Sébastien Han <seb@redhat.com>
(cherry picked from commit a77b47eeeb5770eeefcf4619ab2105ee7a6a003e)

src/ceph-crash.in: adad log-level support

Now, the script can run with --log-level DEBUG to enable DEBUG level
logging.

Signed-off-by: Sébastien Han <seb@redhat.com>
(cherry picked from commit cb5d7667102a8fc8ae23dc23e0f424ebb5b85b9b)

src/ceph-crash.in: fix --name usage

Previously when the --name was used, the code will still go through the
initial list. Now if --name is passed, only this user will be used to
upload crashes.

Signed-off-by: Sébastien Han <seb@redhat.com>
(cherry picked from commit 66037d0c0ba80f2c4009558bb13ad073ca59f954)

src/ceph-crash.in: remove unused frame in handler()

frame was unused so let's remove it.

Signed-off-by: Sébastien Han <seb@redhat.com>
(cherry picked from commit 432c766a99f70884049bf7a1f268287a5a809777)

src/ceph-crash.in: remove unused variable

stdout was never used so let's remove it.

Signed-off-by: Sébastien Han <seb@redhat.com>
(cherry picked from commit 3089f0079a7fe5d2cd297d6eedd60a7b3249b42f)

osd: fix the truncation of an int by int division

The 'r' coeff calculated in OSD::tick() was always 0.

Signed-off-by: Ronen Friedman <rfriedma@redhat.com>
(cherry picked from commit 6f2b4d3dd26b0ee072949888560b40bbeb3dc27e)

common: fix FTBFS due to dout & need_dynamic on GCC-12

For details see:
https://gist.github.com/rzarzynski/d6d2df6888923bef6a3e764f4856853f.

Special thanks to Kaleb Keithley who reported the issue
and tested the fix!

Fixes: https://tracker.ceph.com/issues/53896
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
(cherry picked from commit 963d756ded40f5adf2efef53893c917bec1845c1)