# v16.2.8



* Backport #50851: pacific: mds: 32bit compilation fixes for PurgeQueue
* Backport #51484: pacific: pybind: balancer show: AttributeError: 'Plan' object has no attribute 'show'
* Backport #51979: pacific: mgr/prometheus: Prometheus manager module metrics having the wrong type
* Backport #52073: pacific: rgw: user stats showing 0 value for "size_utilized" and "size_kb_utilized" fields
* Backport #52534: pacific: rbd children:logging crashes after open or close fails.
* Backport #52571: pacific: npm fails on alpine linux trying to install fsevents
* Backport #52584: pacific: [pwl] possible data corruption in TEST_F(TestLibRBD, TestFUA)
* Backport #52631: pacific: mds: add max_mds_entries_per_dir config option
* Backport #52636: pacific: MDSMonitor: removes MDS coming out of quorum election
* Backport #52680: pacific: Add option in `fs new` command to start rank 0 in failed state
* Backport #52734: pacific: full cluster handling regressions in pacific
* Backport #52735: pacific: bogus "unable to find a keyring" warning on "rbd mirroring bootstrap import"
* Backport #53013: pacific: CEPH_OSD_FLAG_FULL_TRY isn't propagated to data_ctx for separate data pool images
* Backport #53014: pacific: write-around cache doesn't honor LIBRADOS_OP_FLAG_FADVISE_FUA flag for write_same()
* Backport #53019: pacific: [test] TestLibRBD.CreateThickRemoveFullTry fails because pool quota enforcement is delayed
* Backport #53066: pacific: rbd-nbd: generate and send device cookie with netlink connect request
* Backport #53098: pacific: rgw/crypt s3tests with vault: Failed to retrieve the actual key, kms-keyid: my-key-1
* Backport #53101: pacific: os/bluestore/AvlAllocator: introduce bluestore_avl_alloc_ff_max_* options
* Backport #53103: pacific:  os/bluestore/AvlAllocator: specialize _block_picker() and cleanups
* Backport #53105: pacific: os/bluestore: Improve _block_picker function
* Feature #53111: RFE: Improve ceph alert structure and validation
* Backport #53218: pacific: qa: Test failure: test_perf_counters (tasks.cephfs.test_openfiletable.OpenFileTable)
* Backport #53225: pacific: tempest dependency issue with 'PrettyTable'
* Backport #53256: pacific: rgw nfs export at user-level crash on readdir
* Backport #53271: pacific: beast frontend performance regressions
* Backport #53280: pacific: ceph-volume should run lvm commands on host
* Backport #53289: pacific: rgw: fix bi put not using right bucket index shard
* Backport #53296: pacific: mgr/dashboard: Add Grafana unit testing
* Backport #53332: pacific: ceph-fuse seems to need root permissions to mount (ceph-fuse-15.2.14-1.fc33.x86_64) (documentation issue?)
* Backport #53344: pacific: mgr/dashboard: ModuleNotFoundError: No module named 'tasks.mgr.dashboard.test_ganesha'
* Backport #53371: pacific: util/disk.py can't parse PB size suffix
* Backport #53380: pacific: mgr/dashboard: upgrade Cypress to the latest stable version
* Backport #53383: pacific: mgr/dashboard: dashboard does not show degraded objects if they are less than 0.5% under "Dashboard->Capacity->Objects block
* Backport #53386: pacific: rbd-mirror: TestMockMirrorStatusUpdater.RemoveImmediateUpdate may get stuck
* Backport #53421: pacific: librbd/crypto: fix various memory leaks
* Backport #53437: pacific: mgr/dashboard: avoid tooltip if disk_usage=null and fast-diff enable in RBD image table
* Backport #53445: pacific: mds: opening connection to up:replay/up:creating daemon causes message drop
* Backport #53510: pacific: mgr/dashboard: fix frontend deps' vulnerabilities
* Backport #53576: pacific: mgr/dashboard: introduce HAProxy metrics for RGW
* Backport #53602: pacific: run-tox-grafana-query-test fails on arm64
* Backport #53608: pacific: crash BlueStore::Onode::put from BlueStore::TransContext::~TransContext
* Backport #53616: pacific: mgr/prometheus: Add all healthchecks to prometheus output and provide a way of viewing history
* Backport #53617: pacific: allow skip is_root check
* Backport #53679: pacific: jenkins api test failure: No module named 'setuptools._distutils'
* Backport #53687: pacific: mgr/dashboard: implement host drain feature in dashboard
* Backport #53689: pacific: mgr/dashboard: Cluster Expansion - Review Section: fixes and improvements
* Backport #53690: pacific: mgr/dashboard: cephadm e2e job: improvements
* Backport #53714: pacific: mds: fails to reintegrate strays if destdn's directory is full (ENOSPC)
* Backport #53716: pacific: mgr/dashboard: timeout error in dashboard cephadm e2e jobs
* Backport #53717: pacific: mgr/dashboard: introduce seperate frontend component for API docs
* Backport #53733: pacific: [pwl] TestDeepCopy.NoSnaps faild w/ enable rbd pwl.
* Backport #53736: pacific: mds: recursive scrub does not trigger stray reintegration
* Backport #53744: pacific: mgr/dashboard: cephadm dashboard e2e cleanups
* Backport #53777: pacific: fs perf stats command crashes
* Backport #53780: pacific: test_dashboard_e2e.sh Failure: orchestrator/02-hosts-inventory.e2e failed.
* Backport #53825: pacific: mgr/dashboard: add test coverage for API docs (SwaggerUI)
* Backport #53828: pacific: mgr/dashboard: monitoring:Implement BlueStore onode hit/miss counters into the dashboard
* Backport #53834: pacific: mgr/dashboard: evaluate upgrade to Angular 12
* Backport #53838: pacific: diff-iterate include_parent functionality is broken in fast-diff mode
* Backport #53839: pacific: diff-iterate reports incorrect offsets in fast-diff mode
* Backport #53860: pacific: mds: crash when `ceph tell mds.0 dump tree ''`
* Backport #53861: pacific: qa: tasks.cephfs.fuse_mount:mount command failed
* Backport #53864: pacific: mds: FAILED ceph_assert(mut->is_wrlocked(&pin->filelock))
* Backport #53880: pacific: [test] run_xfstests_qemu.sh reports success without actually running any tests
* Backport #53881: pacific: mgr/dashboard: failed to load smart data when a device has only 1 daemon associated
* Backport #53908: pacific: mds: remove the duplicated or incorrect respond when the pool is full
* Backport #53916: pacific: fix lvm calls when running on a different OS family
* Backport #53921: pacific: mgr/dashboard: Improve cephadm dashboard e2e stability
* Backport #53922: pacific: mgr/dashboard: Error - yargs parser supports a minimum Node.js version of 12.
* Backport #53933: pacific: Stretch mode: peering can livelock with acting set changes swapping primary back and forth
* Backport #53938: pacific: journal: client_register does not check disconnected state when looking for min_commit_position
* Backport #53944: pacific: [RFE] Limit slow request details to mgr log
* Backport #53955: pacific: ceph-volume generates a lot of unnecessary messages
* Backport #53962: pacific: ceph-volume should ignore /dev/rbd* devices
* Backport #53981: pacific: [cli] fix mixing switches and positional arguments
* Backport #53989: pacific: [rbd-mirror] fix races in snapshot-based mirroring deletion propagation
* Backport #54168: pacific: [rbd_support] passing invalid interval removes entire schedule
* Backport #54170: pacific: [krbd] expose rxbounce option
* Backport #54187: pacific: [krbd] misleading error when no initial monitor address
* Backport #54231: pacific: [cli] mark optional positional arguments as such in help output
* Backport #54241: pacific: mds: clients can send a "new" op (file operation) and crash the MDS
* Backport #54268: pacific: DiffIterateTest/0.DiffIterate failed w/ librbd pwl cache.
* Backport #54298: pacific: librbd: bug in GroupSnapshotNamespace operator< 
* Backport #54301: pacific: [test] qemu task on centos 8.stream is missing qemu-kvm-block-rbd package
* Backport #54324: pacific: allow rbd_default_map_options to be overridden at pool and image level
* Backport #54378: pacific: [rbd-mirror] disabling and shortly after re-enabling mirroring on the image can lead to split-brain
* Backport #54379: pacific: c++20: compilation failures around BitVector::NoInitAllocator
* Backport #54381: pacific: qa/workunits/rbd/rbd_mirror_stress.sh: rbd: mirroring not enabled on the image
* Backport #54484: pacific: rbd snapshot schedule status output is missing schedule
* Bug #54615: virtual int KernelDevice::read(uint64_t, uint64_t, ceph::bufferlist*, IOContext*, bool): assert((uint64_t)r == len)
* Bug #54983: mgr/dashboard: RGW users and buckets tables are empty if the selected gateway is down
* Backport #54987: pacific: mgr/nfs:  don't always restart the NFS ganesha service for updates to CephFS NFS exports
* Backport #54989: pacific: Create test case to verify values of snaps in rbd du
* Backport #55013: pacific: librados: check latest osdmap on ENOENT in pool_reverse_lookup()
* Backport #55058: pacific: diff-iterate in fast-diff mode reports extents out of order for fancy striped images
* Backport #55073: pacific: osd: osd_fast_shutdown_notify_mon not quite right
* Backport #55139: pacific: osd: pgs went back into snaptrim state after osd restart
* Bug #55194: RGW is not starting in public_network when MON are in cluster_network even when endpoint is set
* Backport #55292: pacific: [pwl] enhance "rbd status" output and periodically update it
* Backport #55442: pacific: rocksdb omap iterators become extremely slow in the presence of large delete range tombstones
* Feature #55544: Make it possible to use custom haproxy config when using ingress