# v16.2.0



* Feature #47375: mgr/dashboard: support multiple RGW zone configuration
* Backport #48948: pacific: valgrind: invalid read in perm_state constructor
* Backport #48949: pacific: wait_for_recovery: failed before timeout expired with tests that override osd_async_recovery_min_cost
* Backport #48950: pacific: ObjectStore/StoreTest hangs
* Backport #48962: pacific: mgr/dashboard: incorrect validation in rgw user form for tenanted users
* Backport #48970: pacific: ocf:ceph:rbd resource agent does not support namespaces
* Backport #48971: pacific: multisite: metadata sync of bucket creation gets empty layout.logs
* Backport #48986: pacific: ceph osd df tree reporting incorrect SIZE value for rack having an empty host node
* Backport #49006: pacific: mgr should update mon metadata when mon map is updated
* Backport #49008: pacific: osd crash in OSD::heartbeat when dereferencing null session
* Backport #49011: pacific: krbd: add support for msgr2 (kernel 5.11)
* Backport #49021: pacific: mgr/dashboard: trigger alert if some nodes have a MTU different than the majority of them
* Backport #49025: pacific: build failure on fedora-34/rawhide with boost 1.75
* Backport #49027: pacific: mgr/volumes: evict clients based on auth ID and subvolume mounted
* Backport #49030: pacific: asio compilation failures with boost >= 1.74
* Backport #49038: pacific: Cannot allocate memory appears when using io_uring osd
* Backport #49042: pacific: teuthology vault unzip fails on centos-8
* Backport #49043: pacific: teuthology keystone won't install on centos-8
* Backport #49044: pacific: teuthology rgw qa/tasks/barbican.py
* Backport #49054: pacific: pick_a_shard() always select shard 0
* Backport #49058: pacific: thrash_cache_writeback_proxy_none: FAILED ceph_assert(version == old_value.version) in src/test/osd/RadosModel.h
* Backport #49060: pacific: valgrind errors when protected_fixedsize_stack is used for coroutines
* Backport #49071: pacific: RGW Bucket Lifecycle configuration on non metadata master zones returns HTTP 503
* Backport #49085: pacific: mgr/dashboard: missing root path of each session in Cephfs dashboard
* Backport #49095: pacific: Ceph-volume lvm batch fails with AttributeError: module 'ceph_volume.api.lvm' has no attribute 'is_lv'
* Backport #49097: pacific: FAILED ceph_assert(o->pinned) in BlueStore::Collection::split_cache(BlueStore::Collection*)
* Backport #49100: pacific: crash in BlueStore::Onode::put()
* Backport #49108: pacific: Ceph Zabbix module sending too long numeric values
* Backport #49129: pacific: multipart object names may have null characters
* Backport #49134: pacific: test_envlibrados_for_rocksdb.sh: EnvLibradosMutipoolTest.DBBulkLoadKeysInRandomOrder makes osds crash
* Backport #49142: pacific: fail to create OSDs because the requested extent is too large
* Backport #49144: pacific: rados/perf: cosbench workloads hang forever
* Backport #49145: pacific: out of order op
* Backport #49155: pacific: Windows RBD service issues
* Backport #49156: pacific: Segmentation fault in PrimaryLogPG::cancel_manifest_ops
* Backport #49162: pacific: mgr/dashboard: Add a badge to the Label column of the Host List
* Backport #49172: pacific: multisite: RunBucketSourcesSync ignores errors from RunBucketSync
* Backport #49176: pacific: Listing a new, empty omap datalog errors
* Backport #49183: pacific: [test] ceph_test_rbd_mirror_random_write is non-functional
* Backport #49185: pacific: rgw: tooling to locate rgw objects with missing rados components
* Backport #49194: pacific: rgw: allow rgw-orphan-list to handle intermediate files w/ binary data
* Backport #49198: pacific: C++ write API might not properly handle oversized bufferlist
* Backport #49250: pacific: Windows Event log support
* Backport #49251: pacific: ETagVerifier_MPU compilation warning
* Backport #49253: pacific: [rbd-mirror] reported snapshot corruption
* Backport #49254: pacific: mgr/dashboard: crush map viewer tables values not reloading automatically
* Backport #49264: pacific: rbd: harmless error messages when running `mirror pool status --verbose` for pool with non-mirror images
* Backport #49265: pacific: qa::ERROR: test_recover_auth_metadata_during_authorize
* Backport #49270: pacific: mgr/dashboard: delete EOF when reading passwords from file
* Backport #49288: pacific: fix typo in batch log message
* Backport #49295: pacific: pubsub: FAIL: test radosgw-admin commands, AssertionError: 11 != 10
* Backport #49310: pacific: add security to amqp bucket notification endpoint
* Backport #49315: pacific: monitoring: add some leeway for package drops and errors (1%)
* Backport #49325: pacific: mgr/dashboard: fix MTU Mismatch alert
* Backport #49326: pacific: mgr/dashboard: avoid using document.write()
* Backport #49333: pacific: drive-group subcommand potentially passes root disk to batch
* Backport #49334: pacific: [rbd_support] mirror snapshot scheduler attempts to snapshot non-primary images
* Backport #49346: pacific: vstart: volumes/nfs interface complaints cluster does not exists
* Backport #49352: pacific: device life expectancy resets on mgr restart
* Backport #49358: pacific: s3select valgrind issues
* Backport #49362: pacific: rgw: sts get_session_token always failed
* Backport #49381: pacific: multisite: etag verifier misidentifies multipart uploads with only one part
* Backport #49384: pacific: BlueFS reads might improperly rebuild internal buffer under an shared lock
* Backport #49390: pacific: mgr/dashboard: the tooltips for Provisioned/Total Provisioned fields of an RBD image are invisible
* Backport #49398: pacific: rados/dashboard: Health check failed: Telemetry requires re-opt-in (TELEMETRY_CHANGED)" in cluster log
* Backport #49401: pacific: rados: Health check failed: 1/3 mons down, quorum a,c (MON_DOWN)" in cluster log'
* Backport #49404: pacific: lazy_omap_stats_test: "ceph osd deep-scrub all" hangs
* Backport #49414: pacific: mgr/nfs: Update about user config 
* Backport #49416: pacific: [rbd-mirror] interrupted image creation might result in unlinked images
* Backport #49422: pacific: mgr/dashboard: set XFrame options and Content Security Policy headers
* Backport #49423: pacific: doc: broken links multimds and kcephfs
* Backport #49425: pacific: cephfs-mirror: test mirror daemon with valgrind
* Backport #49429: pacific: mgr/volumes: Bump up the AuthMetadataManager's version to 6
* Backport #49432: pacific: cephfs-mirror: dangling pointer in PeerReplayer 
* Backport #49442: pacific: radosgw-admin user create error message is confusing if user with supplied email address already exists
* Backport #49453: pacific: [rbd-mirror] ImageReplayer.cc: 699: FAILED ceph_assert(m_update_status_task != nullptr)
* Backport #49455: pacific: [rbd-mirror] image rename might not be propagated under snapshot-based mode
* Backport #49470: pacific: qa: ffsb workload: PG_AVAILABILITY|PG_DEGRADED warnings
* Backport #49474: pacific: nautilus: qa: "Assertion `cb_done' failed."
* Backport #49479: pacific: Bluefs improperly handles huge (>4GB) writes which causes data corruption
* Backport #49482: pacific: api_tier_pp: LibRadosTwoPoolsPP.ManifestSnapRefcount/ManifestSnapRefcount2 failed
* Backport #49496: pacific: pacific: qa: :ERROR: test_idempotency
* Backport #49502: pacific: rbd_support: should bail out if snapshot mirroring is not enabled
* Backport #49512: pacific: client: allow looking up snapped inodes by inode number+snapid tuple
* Backport #49517: pacific: pybind/cephfs: DT_REG and DT_LNK values are wrong
* Backport #49520: pacific: client: wake up the front pos waiter
* Backport #49526: pacific: mon/crush_ops.sh fails: Error EBUSY: osd.1 has already bound to class 'ssd', can not reset class to 'hdd'
* Backport #49528: pacific: "ceph osd crush set|reweight-subtree" commands do not set weight on device class subtree
* Backport #49532: pacific: osd ok-to-stop too conservative
* Backport #49546: pacific: mgr/dashboard: split tenant$user when creating S3 users
* Backport #49561: pacific: qa: file system deletion not complete because starter fs already destroyed
* Backport #49563: pacific: qa: run fs:verify with tcmalloc
* Backport #49564: pacific: mon/MonCap: `fs authorize` generates unparseable cap for file system name containing '-'
* Backport #49565: pacific: api_watch_notify: LibRadosWatchNotify.AioWatchDelete2 fails
* Backport #49569: pacific: qa: rank_freeze prevents failover on some tests
* Backport #49585: pacific: valgrind MismatchedFree warnings
* Backport #49587: pacific: missing documentation on "pubsub" in rgw_enable_apis
* Backport #49593: pacific: Master RPM build tries to git clone external project
* Backport #49595: pacific: mgr/dashboard: ERROR: test_a_set_login_credentials (tasks.mgr.dashboard.test_auth.AuthTest)
* Backport #49598: pacific: mgr/dashboard: report fsid in cluster configuration
* Backport #49601: pacific: rgw/sts: After refreshing the temporary credentials, buckets created earlier for OIDC federated user are not listed.
* Backport #49608: pacific: mds: define CephFS errors that replace standard errno values
* Backport #49609: pacific: qa: ERROR: test_damaged_dentry, KeyError: 'passed_validation'
* Backport #49610: pacific: qa: mds removed because trimming for too long with valgrind
* Backport #49612: pacific: qa: racy session evicted check
* Backport #49630: pacific: qa: slow metadata ops during scrubbing
* Backport #49631: pacific: mds: don't start purging inodes in the middle of recovery
* Backport #49634: pacific: Windows CephFS support - ceph-dokan
* Backport #49635: pacific: mgr/telemetry: check if 'ident' channel is active before compiling reports
* Backport #49642: pacific: Disable and re-enable clog_to_monitors could trigger assertion
* Backport #49652: pacific: mgr/dashboard: fix PUT - /api/host/{hostname} while adding labels
* Backport #49658: pacific: mgr/dashboard: test prometheus/alertmanager rules through promtool
* Backport #49665: pacific: mgr/dashboard: Host maintenance Feature
* Backport #49670: pacific: make check: src/test/smoke.sh: TEST_multimon: timeout 8 rados -p foo bench 4 write -b 4096 --no-cleanup
* Backport #49683: pacific: OSD: shutdown of a OSD Host causes slow requests
* Backport #49685: pacific: ls -l in cephfs-shell tries to chase symlinks when stat'ing and errors out inappropriately when stat fails
* Backport #49686: pacific: cephfs-mirror: display cephfs mirror instances in `ceph status` command
* Backport #49687: pacific: client: add metric for number of pinned capabilities
* Backport #49699: pacific: osd: Add refinements to the mclock built-in profiles implementation.
* Backport #49702: pacific: mgr/dashboard: Documented dashboard instance ssl certificate functionality not implemented
* Backport #49707: pacific: mgr now includes mon metadata as part of osd metadata
* Backport #49708: pacific: mgr/dashboard: cluster > hosts: "devices" and "inventory > devices" tabs/tables are misleading
* Backport #49713: pacific: mgr/nfs: Add interface to update export
* Backport #49722: pacific: run librgw tests in teuthology
* Backport #49728: pacific: debian ceph-common package post-inst clobbers ownership of cephadm log dirs
* Backport #49743: pacific: Segmentation fault on GC with big value of rgw_gc_max_objs
* Backport #49751: pacific: snap-schedule doc
* Backport #49753: pacific: cephfs-mirror: add mirror peers via bootstrapping
* Backport #49760: pacific: mgr/balancer: KeyError messages in balancer module
* Backport #49765: pacific: cephfs-mirror: symbolic links do not get synchronized at times
* Backport #49767: pacific: [rbd] the "trash mv" operation should support an optional "--image-id"
* Backport #49773: pacific: Compression hints are not sent to OSD when object-map is enabled and object may exist
* Backport #49774: pacific: Get more parallel scrubs within osd_max_scrubs limits
* Backport #49796: pacific: pool application metadata not propagated to the cache tier
* Backport #49814: pacific: buckets are created with ctime=0
* Backport #49817: pacific: mon: promote_standby does not update available_modules
* Backport #49852: pacific: mds: race of fetching large dirfrag
* Backport #49854: pacific: client: crashed in cct->_conf.get_val() in Client::start_tick_thread()
* Backport #49863: pacific: rgwlc: bucket_lc_prepare non-terminating loop (backport)
* Backport #49877: pacific: doc: Document mds cap acquisition readdir throttle
* Backport #49878: pacific: mgr/dashboard: support multiple RGW zone configuration
* Backport #49886: pacific: qa: Add additional wait_for_clean() check in lost_unfound tasks.
* Backport #49895: pacific: osd/scrub - a ceph_assert() in a legitimate path
* Backport #49905: pacific: mgr/volumes: setuid and setgid file bits are not retained after a subvolume snapshot restore
* Backport #49907: pacific: pybind/mgr/dashboard: bump flake8 to 3.9.0
* Backport #49908: pacific: mgr/DaemonServer.cc: FAILED ceph_assert(pending_service_map.epoch > service_map.epoch)
* Backport #49920: pacific: Disable bluestore_fsck_quick_fix_on_mount by default
* Backport #49924: pacific: Ceph-volume reports a device as available and the device cannot be used to create OSDs because it has <GPT headers>.
* Backport #49926: pacific: libcephsqlite: library for sqlite interface to ceph
* Backport #49929: pacific: test: test_mirroring_command_idempotency (tasks.cephfs.test_admin.TestMirroringCommands) failure
* Backport #49930: pacific: mon/MDSMonitor: standby-replay daemons should be removed when the flag is turned off
* Backport #49932: pacific: MDS should return -ENODATA when asked to remove xattr that doesn't exist
* Backport #49935: pacific: libcephfs: test termination "what():  Too many open files"
* Backport #49941: pacific: mgr/dashboard: error notification shown when no rgw daemons running.
* Backport #49944: pacific: mgr/dashboard: Display users current quota usage
* Backport #49951: pacific: mgr/nfs: Update about cephadm single nfs-ganesha daemon per host limitation
* Backport #49956: pacific: cephsqlite: xCurrentTimeInt64 binding uses incorrect julian day offset units
* Backport #49965: pacific: BlueStore::_collection_list causes huge latency growth pg deletion
* Backport #49973: pacific: cephfs-top: missing keys in the client_metadata
* Backport #49978: pacific: "make check" jenkins job fails
* Backport #49986: pacific: cephfs-top : allow configurable stats refresh interval
* Backport #49989: pacific: _txc_add_transaction error (39) Directory not empty not handled on operation 21 (op 1, counting from 0)
* Backport #49994: pacific: cephfs-top: fails with exception "OPENED_FILES"
* Backport #49997: pacific: daemons bind to loopback iface
* Backport #50001: pacific: ceph-libboost version conflicts
* Backport #50011: pacific: cephfs-top: flake8 E501 line too long error
* Backport #50029: pacific: Nothing provides sqlite-libs needed by libcephsqlite
* Bug #50295: cephadm bootstrap mon container fails to start with podman 3.1 in CentOS 8 Stream
* Bug #50306: /etc/hosts is not passed to ceph containers. clusters that were relying on /etc/hosts for name resolution will have strange behavior
* Bug #50359: Configure the IP address for the monitoring stack components