# v12.2.1



* Bug #20594: mds: cache limits should be expressed in memory usage, not inode count
* Backport #20964: luminous: [config] switch to new config option getter methods
* Backport #20968: luminous: rgw: bytes_send and bytes_recv in the msg of usage show returning is 0 in master branch
* Backport #21045: luminous: TestMirroringWatcher.ModeUpdated: periodic failure due to injected message failures
* Backport #21046: luminous: "ceph daemon osd.0 get_command_descriptions" needs to have a line break at the end of result
* Backport #21052: luminous: RHEL 7.3 Selinux denials at OSD start
* Backport #21097: luminous: multisite: FAILED assert(prev_iter != pos_to_prev.end()) in RGWMetaSyncShardCR::collect_children()
* Backport #21103: luminous: client: missing space in some client debug log messages
* Backport #21104: luminous: ceph-fuse RPM should require fusermount
* Backport #21107: luminous: fs: client/mds has wrong check to clear S_ISGID on chown
* Backport #21108: luminous: mon/OSDMonitor: "osd pool application get" support 
* Backport #21110: luminous: rgw: send data-log list infinitely 
* Backport #21112: luminous: get_quota_root sends lookupname op for every buffered write
* Backport #21114: luminous: qa: FS_DEGRADED spurious health warnings in some sub-suites
* Backport #21115: luminous: rgw multisite: objects encrypted with SSE-KMS are stored unencrypted in target zone
* Backport #21116: luminous: rgw multisite: cannot sync objects encrypted with SSE-C
* Backport #21118: luminous: rgw: need to stream metadata full sync init
* Backport #21133: luminous: osd/PrimaryLogPG: sparse read won't trigger repair correctly
* Backport #21135: luminous: rgw: bucket index sporadically reshards to 65521 shards
* Backport #21137: luminous: mgr: 500 error when attempting to view filesystem data
* Backport #21138: luminous: RGW: object copied from remote src acl permission become full-control issue
* Backport #21139: luminous: rgw: put lifecycle configuration fails if Prefix is not set
* Backport #21182: luminous: 'osd crush rule rename' not idempotent
* Backport #21183: luminous: Crash in MonCommandCompletion
* Backport #21184: luminous: NameError: global name 'name' is not defined
* Backport #21185: luminous: rgw_file:  incorrect lane lock behavior in evict_block()
* Backport #21187: luminous: fix performance regression
* Backport #21188: luminous: dashboard: usage graph is getting more and mor big
* Bug #21222: MDS: standby-replay mds should avoid initiating subtree export
* Bug #21230: the standbys are not updated via "ceph tell mds.* command"
* Backport #21231: luminous: core: interval_set: optimize intersect_of insert operations
* Backport #21233: luminous: memory leak in MetadataHandlers
* Backport #21234: luminous: bluestore: asyn cdeferred_try_submit deadlock
* Backport #21235: luminous: thrashosds read error injection doesn't take live_osds into account
* Backport #21236: luminous: build_initial_pg_history doesn't update up/acting/etc 
* Backport #21237: luminous: osd crash when change option "bluestore_csum_type" from "none" to "CRC32" 
* Backport #21238: luminous: test_health_warnings.sh can fail
* Backport #21240: luminous: "Health check update" log spam
* Backport #21241: luminous: usage of --inconsistent-index should require user confirmation and print a warning
* Backport #21242: luminous: OSD crash: PrimaryLogPG.cc: 8396: FAILED assert(repop_queue.front() == repop)
* Bug #21243: incorrect erasure-code space in command ceph df
* Bug #21252: mds: asok command error merged with partial Formatter output
* Backport #21265: luminous: [cli] rename of non-existent image results in seg fault
* Backport #21267: luminous: Incorrect grammar in FS message "1 filesystem is have a failed mds daemon"
* Backport #21269: luminous: some generic options can not be passed by rbd-nbd
* Backport #21270: luminous: rgw: shadow objects are sometimes not removed
* Backport #21276: luminous: os/bluestore/BlueFS.cc: 1255: FAILED assert(!log_file->fnode.extents.empty())
* Backport #21277: luminous: [cls] metadata_list API function does not honor `max_return` parameter.
* Backport #21278: luminous: the standbys are not updated via "ceph tell mds.* command"
* Backport #21283: luminous: spurious MON_DOWN, apparently slow/laggy mon
* Backport #21288: luminous: [test] various teuthology errors
* Backport #21289: luminous: [rbd] image-meta list does not return all entries
* Feature #21301: expose --sync-stats via admin api
* Bug #21303: rocksdb get a error: "Compaction error: Corruption: block checksum mismatch"
* Feature #21315: Add option to view IP addresses of clients in output of 'ceph features'
* Backport #21321: luminous: mds: asok command error merged with partial Formatter output
* Backport #21322: luminous: MDS: standby-replay mds should avoid initiating subtree export
* Backport #21323: luminous: MDCache::try_subtree_merge() may print N^2 lines of debug message
* Backport #21325: luminous: bluestore: aio submission deadlock
* Bug #21328: Performance: Slow OSD startup, heavy LevelDB activity
* Backport #21341: luminous: mon/OSDMonitor: deleting pool while pgs are being created leads to assert(p != pools.end) in update_creating_pgs()
* Backport #21342: luminous: ceph mgr versions shows active mgr as "Unknown"
* Backport #21345: luminous: "'rbd/import_export.sh'" broken in upgrade:luminous-x:parallel-master
* Backport #21350: luminous: rgw: data encryption sometimes fails to follow AWS settings 
* Backport #21357: luminous: mds: segfault during `rm -rf` of large directory
* Backport #21374: luminous: incorrect erasure-code space in command ceph df
* Bug #21378: mds: up:stopping MDS cannot export directories
* Backport #21384: luminous: mds: cache limits should be expressed in memory usage, not inode count
* Backport #21385: luminous: mds: up:stopping MDS cannot export directories
* Backport #21396: Illegal instruction in RocksDB
* Backport #21436: luminous: client: Variable "onsafe" going out of scope leaks the storage it points to
* Backport #21437: luminous: test_filtered_df: assert 0.9 < ratio < 1.1
* Backport #21449: luminous: qa: test_misc creates metadata pool with dummy object resulting in WRN: POOL_APP_NOT_ENABLED
* Backport #21464: luminous: qa: ignorable MDS_READ_ONLY warning
* Backport #21472: luminous: qa: ignorable "MDS cache too large" warning
* Backport #21473: luminous: test hang after mds evicts kclient
* Backport #21484: luminous: qa: fs.get_config on stopped MDS
* Backport #21486: luminous: qa: test_client_pin times out waiting for dentry release from kernel
* Backport #21487: luminous: MDS rank add/remove log messages say wrong number of ranks
* Backport #21488: luminous: qa: failures from pjd fstest
* Backport #21490: luminous: test_rebuild_simple_altpool triggers MDS assertion
* Backport #21513: luminous: mds: src/mds/MDLog.cc: 276: FAILED assert(!capped)
* Backport #21515: luminous: qa: kcephfs: missing whitelist for evicted client
* Backport #21516: luminous: qa: kcephfs: ignore warning on expected mds failover
* Backport #21517: luminous: qa: kcephfs: client-limits: whitelist "MDS cache too large"
* Backport #21540: luminous: whitelist additions
* Bug #21588: upgrade from jewel to luminous ubuntu need firewall restart
* Bug #21619: RGW Reshard error add failed to drop lock on <bucket>
* Bug #21734: mount client shows total capacity of cluster but not of a pool 
* Bug #21736: Cannot create bluestore OSD
* Bug #21751: OSD crashes during recovery on arm64 due to assert in Throttle::put
* Bug #21770: ceph mon core dump when use ceph osd perf cmd.
* Bug #21809: Raw Used space is 70x higher than actually used space (maybe orphaned objects from pool deletion)
* Bug #21820: Ceph OSD crash with Segfault
* Bug #21827: OSD crashed while reparing inconsistent PG
* Bug #22005: result -108 xferred 2000, blk_update_request: I/O error
* Bug #22015: Civetweb reports bad response code. 
* Bug #22055: ghost rbd snapshot
* Bug #22066: bluestore osd asserts repeatedly with ceph-12.2.1/src/include/buffer.h: 882: FAILED assert(_buffers.size() <= 1024)
* Bug #22088: Docs: describe missing steps in Luminous upgrade docs
* Bug #22094: Lots of reads on default.rgw.usage pool
* Bug #22102: BlueStore crashed on rocksdb checksum mismatch
* Bug #22122: rgw: bucket index object not deleted after radosgw-admin bucket rm --purge-objects --bypass-gc
* Bug #22125: systemctl commands use literal '*'
* Bug #22149: Can not set custom port for prometheus exporter 
* Feature #22168: The RGW Admin OPS is missing the ability to filter for e.g. buckets and users
* Bug #22218: multisite: rgw crashed during meta sync
* Support #22224: memory leak
* Support #22243: Luminous: EC pool using more space than it should
* Bug #22271: vdbench's IO drop to 0 when resize the image at the same time
* Bug #22274: Luminous's radosgw cant decide if it's 32 or 64bit
* Bug #22303: mon crash: Failed to load mgr commands
* Bug #22306: Python RBD metadata_get does not work.
* Bug #22343: Using restful module of Ceph Manager to retrieve Cluster status returns invalid XML