# v13.2.5



* Backport #35932: mimic: mds: retry remounting in ceph-fuse on dcache invalidation
* Backport #36209: mimic: mds: runs out of file descriptors after several respawns
* Backport #36503: mimic: qa: infinite timeout on asok command causes job to die
* Backport #36578: mimic: qa: teuthology may hang on diagnostic commands for fuse mount
* Backport #36639: mimic: rename does not old ref to replacement onode at old name
* Backport #36751: mimic: cmake: do not pass -B{symbolic,symbolic-functions} to linker on FreeBSD
* Backport #36753: mimic: "log last" command skips latest entry
* Backport #37342: mimic: Prioritize user specified scrubs
* Backport #37417: mimic: mgr: various python3 fixes
* Backport #37424: mimic: qa: wrong setting for msgr failures
* Backport #37426: mimic: ceph-volume-client: cannot set mode for cephfs volumes as required by OpenShift
* Backport #37428: mimic: msg/async: crashes when authenticator provided by verify_authorizer not implemented
* Backport #37437: mimic: crushtool: add --reclassify operation to convert legacy crush maps to use device classes 
* Backport #37447: mimic: add a command to trim old bucket instances after resharding completes
* Backport #37465: mimic: rgw: master zone deletion without a zonegroup rm would break rgw rados init
* Backport #37477: mimic: src/mgr/DaemonServer.cc: 912: FAILED ceph_assert(daemon_state.exists(key))
* Backport #37480: mimic: mds: MDCache.cc: 11673: abort()
* Backport #37494: mimic: bluefs-bdev-expand aborts
* Backport #37496: mimic: OSD mkfs might assert when working agains bluestore disk that already has a superblock
* Backport #37550: mimic: multisite: sync gets stuck retrying deletes that fail with ERR_PRECONDITION_FAILED
* Backport #37552: mimic: linger op get lost during ceph osd pause and ceph osd unpause
* Backport #37554: mimic: rgw: resharding leaves old bucket info objs and index shards behind
* Backport #37556: mimic: multisite: es sync null versioned object failed because of olh info
* Backport #37558: mimic: presigned URL for PUT with metadata fails: SignatureDoesNotMatch
* Backport #37560: mimic: radosgw coredump  RGWGC::process
* Backport #37562: mimic: rgw: version bucket stats not correct
* Backport #37599: mimic: Python bindings use iterkeys method which is not Python 3 compatible
* Backport #37601: mimic: doc: broken link on troubleshooting-mon page
* Backport #37603: mimic: mds: severe internal fragment when decoding xattr_map from log event
* Backport #37605: mimic: mds: PurgeQueue write error handler does not handle EBLACKLISTED
* Backport #37607: mimic: mds: directories pinned keep being replicated back and forth between exporting mds and importing mds
* Backport #37609: mimic: MDS admin socket command `dump cache` with a very large cache will hang/kill the MDS
* Backport #37611: mimic: qa: pjd test appears to require more than 3h timeout for some configurations
* Backport #37624: mimic: fs status command broken in py3-only environments
* Backport #37626: mimic: mds: fix incorrect l_pq_executing_ops statistics when meet an invalid item in purge queue
* Backport #37628: mimic: mds: do not call Journaler::_trim twice
* Backport #37630: mimic: client: do not move f->pos untill success write
* Backport #37632: mimic: mds: remove duplicated l_mdc_num_strays perfcounter set
* Backport #37634: mimic: race of updating wanted caps
* Backport #37642: mimic: ceph-create-keys: fix octal notation for Python 3 without losing compatibility with Python 2
* Backport #37649: mimic: rgw: resharding events are not logged 
* Backport #37684: mimic: Remove capability reset command
* Backport #37686: mimic: list-inconsistent-obj output truncated, causing osd-scrub-repair.sh failure
* Backport #37688: mimic: Command failed on smithi191 with status 1: '\n sudo yum -y install ceph-radosgw\n '
* Backport #37689: mimic: ceph-objectstore-tool: Add HashInfo to object dump output
* Backport #37693: mimic: CephFS remove snapshot result in slow ops
* Backport #37695: mimic: client: fix failure in quota size limitation when using samba
* Backport #37698: mimic: osd_memory_target: failed assert when options mismatch
* Backport #37699: mimic: fuse client can't read file due to can't acquire Fr
* Backport #37738: mimic: MDSMonitor: ignores stopping MDS that was formerly laggy
* Backport #37740: mimic: extend reconnect period when mds is busy
* Backport #37742: mimic: rbd-mirror and radosgw packages should requires ceph-base
* Backport #37744: mimic: Mgr: OSDMap.cc: 4140: FAILED assert(osd_weight.count(i.first))
* Backport #37755: mimic: osd/PrimaryLogPG: fix the extent length error of the sync read
* Backport #37757: mimic: standby-replay MDS spews message to log every second
* Backport #37759: mimic: mds: mds state change race 
* Backport #37810: mimic: Empty pg_temps are added to incremental map even if there're no changes in new epoch
* Backport #37812: mimic: mon: segmentation fault during shutdown
* Backport #37814: mimic: workunits/rados/test_health_warnings.sh fails with <9 osds down
* Backport #37818: mimic: mds crashes frequently when using snapshots in CephFS on mimic
* Backport #37819: mimic: mds: create separate config for heartbeat timeout
* Backport #37821: mimic: ceph-objectstore-tool export from luminous, import to master clears same_interval_since
* Backport #37822: mimic: mds: output client IP of blacklisted/evicted clients to cluster log
* Backport #37824: mimic: BlueStore: ENODATA not fully handled
* Backport #37826: mimic: mgr crash when handle_report updating existing DaemonState for rgw
* Backport #37828: mimic: ceph-fuse: hang because it miss reconnect phase when hot standby mds switch occurs
* Backport #37832: mimic: FAILED assert(is_up(osd)) in OSDMap::get_inst(int)
* Backport #37842: mimic: mimic->nautilus: test_rbd.TestClone.test_list_children failure
* Backport #37870: mgr/dashboard: (Mimic) UnboundLocalError: local variable 'mirror_mode' referenced before assignment
* Backport #37888: mimic: osd/OSDMap: calc_pg_upmaps - potential access violation
* Backport #37890: mimic: possible "deferment time has not expired" error on "rbd trash remove" just after "rbd trash move"
* Backport #37894: mimic: doc: updated Ceph documentation links
* Backport #37896: mimic: msg/async: mark_down vs accept race leaves connection registered
* Backport #37898: mimic: mds: purge queue recovery hangs during boot if PQ journal is damaged
* Backport #37900: mimic: [journal] max journal order is incorrectly set at 64
* Backport #37902: mimic: osd: pg log hard limit can cause crash during upgrade
* Backport #37904: mimic: FAILED ceph_assert(can_write == WriteStatus::NOWRITE) in ProtocolV1::replace() 
* Backport #37907: mimic: mds: wait shorter intervals to send beacon if laggy
* Backport #37920: mimic: TestMockLeaderWatcher.InitReleaseShutdown: Mutex.cc: 79: FAILED ceph_assert(r == 0)
* Backport #37921: mimic: qa: test_damage expectations wrong for Truncate on some objects
* Backport #37923: mimic: qa: test_damage performs truncate test on same object repeatedly
* Backport #37925: mimic: Broken parameter parsing in /etc/ceph/rbdmap
* Backport #37948: mimic: debug logging for v4 auth does not sanitize encryption keys
* Backport #37952: mimic: qa: test_damage needs to silence MDS_READ_ONLY
* Backport #37984: mimic: cli: dump osd-fsid as part of osd find <id>
* Backport #37986: mimic: Throttle.cc: 194: FAILED assert(c >= 0) due to invalid ceph_osd_op union
* Backport #37988: mimic: MDSMonitor: missing osdmon writeable check
* Backport #37990: mimic: Compression not working, and when applied OSD disks are failing randomly
* Backport #37992: mimic: ec pool lost data due to snap clone
* Backport #38032: mimic: trash purge tries to remove non-user images
* Backport #38036: mimic: upmap balancer won't refill underfull osds if zero overfull found
* Backport #38039: mimic: RGW fails to start on Fedora 28 from default configuration
* Backport #38045: mimic: qa/overrides/short_pg_log.yaml: reduce osd_{min,max}_pg_log_entries
* Backport #38096: mimic: doc/rados/configuration: refresh osdmap section
* Backport #38101: mimic: mds: cache drop should trim cache before flushing journal
* Backport #38103: mimic: client: session flush does not cause cap release message flush
* Backport #38106: mimic: osd/ECBackend.cc: 1547: FAILED ceph_assert(!(*m).is_missing(hoid))
* Backport #38107: mimic: Adding back the IOPS line for client and recovery IO in cluster logs
* Backport #38109: mimic: mgr will refuse connection from the monitor who starts behind it
* Backport #38111: mimic: force-backfill sets forced_recovery instead of forced_backfill in 13.2.1
* Backport #38139: mimic: mgr: add get_latest_counter() to C++ -> Python interface.
* Backport #38141: mimic: Add hashinfo testing for dump command of ceph-objectstore-tool
* Backport #38143: mimic: os/bluestore: fixup access a destroy cond cause deadlock or undefine behaviors
* Backport #38147: mimic: rgw: `radosgw-admin bucket rm ... --purge-objects` can hang...
* Support #38152: rocksdb problem in the 108 down osds
* Backport #38179: mimic: build/ops: boost/noncopyable.hpp is missing
* Backport #38181: mimic: Not enough PG information sent to Zabbix
* Backport #38183: mimic: restful: py got exception when get osd info
* Backport #38185: mimic: krbd discard no longer guarantees zeroing
* Backport #38187: mimic: deep fsck fails on inspecting very large onodes
* Backport #38189: mimic: mds: broadcast quota message to client when disable quota
* Backport #38192: mimic: Object can still be deleted even if s3:DeleteObject policy is set
* Backport #38208: mimic: A PG repairing doesn't mean PG is damaged
* Bug #38213: Cmake complains with unresovable Boost:python.
* Bug #38222: cmake: change order of searching for StdFilesystem libs.
* Backport #38239: mimic: radosbench tests hit ENOSPC
* Backport #38243: mimic: scrub warning check incorrectly uses mon scrub interval
* Backport #38266: mimic: test: Allow multi instances of "make tests" on the same machine
* Backport #38317: mimic: filestore: fsync(2) return value not checked
* Backport #38319: mimic: mgr deadlock: _check_auth_rotating possible clock skew, rotating keys expired way too early
* Bug #38323: ceph-volume -osds-per-device : Unable to convert to integer with locale FR
* Backport #38334: mimic: MDCache::finish_snaprealm_reconnect() create and drop MClientSnap message
* Backport #38335: mimic: mds: fix potential re-evaluate stray dentry in _unlink_local_finish
* Backport #38342: mimic: luminous->(mimic,nautilus): PGMapDigest decode error on luminous end
* Bug #38366: mimic: ceph-volume: add osd_ids argument
* Backport #38451: mimic: Destruction of basic_string _GLIBCXX_USE_CXX11_ABI=0 and C++17 mode results in invalid delete
* Feature #38458: Ceph does not have command to show current osd primary-affinity
* Bug #38461: Ceph osd out is the same as ceph osd reweight 0 (result in same bucket weights)
* Bug #38480: ceph-mon does not start when using a custom cluster name
* Bug #38489: bluestore_prefer_deferred_size_hdd units are not clear
* Backport #38535: mimic: rgw: civetweb drops all requests with absoluteURI (mimic)
* Backport #38536: mimic: rados civetweb fd leak
* Bug #38659: librbd: avoid aggregate-initializing any static_visitor
* Bug #38660: librbd: avoid aggregate-initializing IsWriteOpVisitor
* Bug #38722: rgw: fix RGWDeleteMultiObj::verify_permission
* Bug #38763: Error in Mimic repo for Ubuntu 18.04
* Bug #38783: Changing mon_pg_warn_max_object_skew has no effect.
* Bug #38800: mgr/dashboard: User list does not load
* Bug #39056: localize-reads does not increment pg stats read count
* Bug #39115: ceph pg repair doesn't fix itself if osd is bluestore
* Bug #39116: Draining filestore osd, removing, and adding new bluestore osd causes OSDs to crash
* Bug #39120: rados: Segmentation fault in thread 7f0aebfff700 thread_name:fn_anonymous
* Bug #39484: mon: "FAILED assert(pending_finishers.empty())" when paxos restart