* build/ops: rpm: bump epoch ahead of ceph-common in RHEL base (`issue#20508 <http://tracker.ceph.com/issues/20508>`_, `pr#21190 <https://github.com/ceph/ceph/pull/21190>`_, Ken Dreyer)
* build/ops: upstart: radosgw-all does not start on boot if ceph-base is not installed (`issue#18313 <http://tracker.ceph.com/issues/18313>`_, `pr#16294 <https://github.com/ceph/ceph/pull/16294>`_, Ken Dreyer)
* ceph-disk: factor out the retry logic into a decorator (`issue#21728 <http://tracker.ceph.com/issues/21728>`_, `pr#18169 <https://github.com/ceph/ceph/pull/18169>`_, Kefu Chai)
* ceph-disk: fix --runtime omission when enabling ceph-osd@$ID.service units for device-backed OSDs (`issue#21498 <http://tracker.ceph.com/issues/21498>`_, `pr#17942 <https://github.com/ceph/ceph/pull/17942>`_, Carl Xiong)
* ceph-disk flake8 test fails on very old, and very new, versions of flake8 (`issue#22207 <http://tracker.ceph.com/issues/22207>`_, `pr#19153 <https://github.com/ceph/ceph/pull/19153>`_, Nathan Cutler)
* cephfs: ceph.in: pass RADOS inst to LibCephFS (`issue#21406 <http://tracker.ceph.com/issues/21406>`_, `issue#21967 <http://tracker.ceph.com/issues/21967>`_, `pr#19907 <https://github.com/ceph/ceph/pull/19907>`_, Patrick Donnelly)
* cephfs: client::mkdirs not handle well when two clients send mkdir request for a same dir (`issue#20592 <http://tracker.ceph.com/issues/20592>`_, `pr#20271 <https://github.com/ceph/ceph/pull/20271>`_, dongdong tao)
* cephfs: client: prevent fallback to remount when dentry_invalidate_cb is true but root->dir is NULL (`issue#23211 <http://tracker.ceph.com/issues/23211>`_, `pr#21189 <https://github.com/ceph/ceph/pull/21189>`_, Zhi Zhang)
* cephfs-journal-tool: move shutdown to the deconstructor of MDSUtility (`issue#22734 <http://tracker.ceph.com/issues/22734>`_, `pr#20333 <https://github.com/ceph/ceph/pull/20333>`_, dongdong tao)
* cephfs: osdc/Journaler: make sure flush() writes enough data (`issue#22824 <http://tracker.ceph.com/issues/22824>`_, `pr#20435 <https://github.com/ceph/ceph/pull/20435>`_, "Yan, Zheng")
* cephfs: Processes stuck waiting for write with ceph-fuse (`issue#22008 <http://tracker.ceph.com/issues/22008>`_, `issue#22207 <http://tracker.ceph.com/issues/22207>`_, `pr#19141 <https://github.com/ceph/ceph/pull/19141>`_, "Yan, Zheng")
* ceph-fuse: failure to remount in startup test does not handle client_die_on_failed_remount properly (`issue#22269 <http://tracker.ceph.com/issues/22269>`_, `pr#21162 <https://github.com/ceph/ceph/pull/21162>`_, Patrick Donnelly)
* ceph.in: bypass codec when writing raw binary data (`issue#23185 <http://tracker.ceph.com/issues/23185>`_, `pr#20763 <https://github.com/ceph/ceph/pull/20763>`_, Oleh Prypin)
* ceph-objectstore-tool command to trim the pg log (`issue#23242 <http://tracker.ceph.com/issues/23242>`_, `pr#20882 <https://github.com/ceph/ceph/pull/20882>`_, Josh Durgin, David Zafman)
* ceph-objectstore-tool: "$OBJ get-omaphdr" and "$OBJ list-omap" scan all pgs instead of using specific pg (`issue#21327 <http://tracker.ceph.com/issues/21327>`_, `pr#20284 <https://github.com/ceph/ceph/pull/20284>`_, David Zafman)
* class rbd.Image discard----OSError: [errno 2147483648] error discarding region (`issue#16465 <http://tracker.ceph.com/issues/16465>`_, `issue#21966 <http://tracker.ceph.com/issues/21966>`_, `pr#20287 <https://github.com/ceph/ceph/pull/20287>`_, Nathan Cutler, Huan Zhang, Jason Dillaman)
* cli/crushtools/build.t sometimes fails in jenkins' make check run (`issue#21758 <http://tracker.ceph.com/issues/21758>`_, `pr#21158 <https://github.com/ceph/ceph/pull/21158>`_, Kefu Chai)
* client: release revoking Fc after invalidate cache (`issue#22652 <http://tracker.ceph.com/issues/22652>`_, `pr#19975 <https://github.com/ceph/ceph/pull/19975>`_, "Yan, Zheng")
* client: set client_try_dentry_invalidate to false by default (`issue#21423 <http://tracker.ceph.com/issues/21423>`_, `pr#17925 <https://github.com/ceph/ceph/pull/17925>`_, "Yan, Zheng")
* [cli] rename of non-existent image results in seg fault (`issue#21248 <http://tracker.ceph.com/issues/21248>`_, `pr#20280 <https://github.com/ceph/ceph/pull/20280>`_, Jason Dillaman)
* CLI unit formatting tests are broken (`issue#24733 <http://tracker.ceph.com/issues/24733>`_, `pr#22913 <https://github.com/ceph/ceph/pull/22913>`_, Jason Dillaman)
* common: compute SimpleLRU's size with contents.size() instead of lru.… (`issue#22613 <http://tracker.ceph.com/issues/22613>`_, `pr#19978 <https://github.com/ceph/ceph/pull/19978>`_, Xuehan Xu)
* common/config: set rocksdb_cache_size to OPT_U64 (`issue#22104 <http://tracker.ceph.com/issues/22104>`_, `pr#18850 <https://github.com/ceph/ceph/pull/18850>`_, Vikhyat Umrao, liuhongtong)
* core: Objecter::C_ObjectOperation_sparse_read throws/catches exceptions on -ENOENT (`issue#21844 <http://tracker.ceph.com/issues/21844>`_, `pr#18743 <https://github.com/ceph/ceph/pull/18743>`_, Jason Dillaman)
* Deleting a pool with active notify linger ops can result in seg fault (`issue#23966 <http://tracker.ceph.com/issues/23966>`_, `pr#22188 <https://github.com/ceph/ceph/pull/22188>`_, Kefu Chai, Jason Dillaman)
* doc: remove region from INSTALL CEPH OBJECT GATEWAY (`issue#21610 <http://tracker.ceph.com/issues/21610>`_, `pr#18303 <https://github.com/ceph/ceph/pull/18303>`_, Orit Wasserman)
* Filestore rocksdb compaction readahead option not set by default (`issue#21505 <http://tracker.ceph.com/issues/21505>`_, `pr#20446 <https://github.com/ceph/ceph/pull/20446>`_, Mark Nelson)
* follow-on: osd: be_select_auth_object() sanity check oi soid (`issue#20471 <http://tracker.ceph.com/issues/20471>`_, `pr#20622 <https://github.com/ceph/ceph/pull/20622>`_, David Zafman)
* HashIndex: randomize split threshold by a configurable amount (`issue#15835 <http://tracker.ceph.com/issues/15835>`_, `pr#19906 <https://github.com/ceph/ceph/pull/19906>`_, Josh Durgin)
* install-deps.sh: point gcc to the one shipped by distro (`issue#22220 <http://tracker.ceph.com/issues/22220>`_, `pr#19461 <https://github.com/ceph/ceph/pull/19461>`_, Kefu Chai)
* journal: Message too long error when appending journal (`issue#23526 <http://tracker.ceph.com/issues/23526>`_, `pr#21215 <https://github.com/ceph/ceph/pull/21215>`_, Mykola Golub)
* [journal] tags are not being expired if no other clients are registered (`issue#21960 <http://tracker.ceph.com/issues/21960>`_, `pr#20282 <https://github.com/ceph/ceph/pull/20282>`_, Jason Dillaman)
* librados: copy out data to users' buffer for xio (`issue#20616 <http://tracker.ceph.com/issues/20616>`_, `pr#17594 <https://github.com/ceph/ceph/pull/17594>`_, Vu Pham)
* librbd: cannot clone all image-metas if we have more than 64 key/value pairs (`issue#21814 <http://tracker.ceph.com/issues/21814>`_, `pr#21228 <https://github.com/ceph/ceph/pull/21228>`_, PCzhangPC)
* librbd: cannot copy all image-metas if we have more than 64 key/value pairs (`issue#21815 <http://tracker.ceph.com/issues/21815>`_, `pr#21203 <https://github.com/ceph/ceph/pull/21203>`_, PCzhangPC)
* librbd: create+truncate for whole-object layered discards (`issue#23285 <http://tracker.ceph.com/issues/23285>`_, `pr#21219 <https://github.com/ceph/ceph/pull/21219>`_, Jason Dillaman)
* librbd: list_children should not attempt to refresh image (`issue#21670 <http://tracker.ceph.com/issues/21670>`_, `pr#21224 <https://github.com/ceph/ceph/pull/21224>`_, Jason Dillaman)
* librbd: object map batch update might cause OSD suicide timeout (`issue#22716 <http://tracker.ceph.com/issues/22716>`_, `issue#21797 <http://tracker.ceph.com/issues/21797>`_, `pr#21220 <https://github.com/ceph/ceph/pull/21220>`_, Song Shun, Jason Dillaman)
* librbd: set deleted parent pointer to null (`issue#22158 <http://tracker.ceph.com/issues/22158>`_, `pr#19098 <https://github.com/ceph/ceph/pull/19098>`_, Jason Dillaman)
* mon/OSDMonitor: fix dividing by zero in OSDUtilizationDumper (`issue#22662 <http://tracker.ceph.com/issues/22662>`_, `pr#20344 <https://github.com/ceph/ceph/pull/20344>`_, Mingxin Liu)
* ObjectStore/StoreTest.FiemapHoles/3 fails with kstore (`issue#21716 <http://tracker.ceph.com/issues/21716>`_, `pr#20143 <https://github.com/ceph/ceph/pull/20143>`_, Kefu Chai, Ning Yao)
* osd: also check the exsistence of clone obc for "CEPH_SNAPDIR" requests (`issue#17445 <http://tracker.ceph.com/issues/17445>`_, `pr#17707 <https://github.com/ceph/ceph/pull/17707>`_, Xuehan Xu)
* osdc/Objecter: prevent double-invocation of linger op callback (`issue#23872 <http://tracker.ceph.com/issues/23872>`_, `pr#21754 <https://github.com/ceph/ceph/pull/21754>`_, Jason Dillaman)
* osd: objecter sends out of sync with pg epochs for proxied ops (`issue#22123 <http://tracker.ceph.com/issues/22123>`_, `pr#20518 <https://github.com/ceph/ceph/pull/20518>`_, Sage Weil)
* osd ops (sent and?) arrive at osd out of order (`issue#19133 <http://tracker.ceph.com/issues/19133>`_, `issue#19139 <http://tracker.ceph.com/issues/19139>`_, `pr#17893 <https://github.com/ceph/ceph/pull/17893>`_, Jianpeng Ma, Sage Weil)
* osd: osd_scrub_during_recovery only considers primary, not replicas (`issue#18206 <http://tracker.ceph.com/issues/18206>`_, `pr#17815 <https://github.com/ceph/ceph/pull/17815>`_, David Zafman)
* osd: recover_replicas: object added to missing set for backfill, but is not in recovering, error! (`issue#18162 <http://tracker.ceph.com/issues/18162>`_, `issue#14513 <http://tracker.ceph.com/issues/14513>`_, `pr#18690 <https://github.com/ceph/ceph/pull/18690>`_, huangjun, Adam C. Emerson, David Zafman)
* osd: update heartbeat peers when a new OSD is added (`issue#18004 <http://tracker.ceph.com/issues/18004>`_, `pr#20108 <https://github.com/ceph/ceph/pull/20108>`_, Pan Liu)
* performance: Only scan for omap corruption once (`issue#21328 <http://tracker.ceph.com/issues/21328>`_, `pr#18951 <https://github.com/ceph/ceph/pull/18951>`_, David Zafman)
* qa: src/test/libcephfs/test.cc:376: Expected: (len) > (0), actual: -34 vs 0 (`issue#22221 <http://tracker.ceph.com/issues/22221>`_, `pr#21172 <https://github.com/ceph/ceph/pull/21172>`_, Patrick Donnelly)
* qa: use xfs instead of btrfs w/ filestore (`issue#20169 <http://tracker.ceph.com/issues/20169>`_, `issue#20911 <http://tracker.ceph.com/issues/20911>`_, `pr#18165 <https://github.com/ceph/ceph/pull/18165>`_, Sage Weil)
* qa: use xfs instead of btrfs w/ filestore (`issue#21481 <http://tracker.ceph.com/issues/21481>`_, `pr#17847 <https://github.com/ceph/ceph/pull/17847>`_, Patrick Donnelly)
* radosgw: fix awsv4 header line sort order (`issue#21607 <http://tracker.ceph.com/issues/21607>`_, `pr#18080 <https://github.com/ceph/ceph/pull/18080>`_, Marcus Watts)
* rbd: clean up warnings when mirror commands used on non-setup pool (`issue#21319 <http://tracker.ceph.com/issues/21319>`_, `pr#21227 <https://github.com/ceph/ceph/pull/21227>`_, Jason Dillaman)
* rbd: disk usage on empty pool no longer returns an error message (`issue#22200 <http://tracker.ceph.com/issues/22200>`_, `pr#19186 <https://github.com/ceph/ceph/pull/19186>`_, Jason Dillaman)
* [rbd] image-meta list does not return all entries (`issue#21179 <http://tracker.ceph.com/issues/21179>`_, `pr#20281 <https://github.com/ceph/ceph/pull/20281>`_, Jason Dillaman)
* rbd: is_qemu_running in qemu_rebuild_object_map.sh and qemu_dynamic_features.sh may return false positive (`issue#23502 <http://tracker.ceph.com/issues/23502>`_, `pr#21207 <https://github.com/ceph/ceph/pull/21207>`_, Mykola Golub)
* rbd: [journal] allocating a new tag after acquiring the lock should use on-disk committed position (`issue#22945 <http://tracker.ceph.com/issues/22945>`_, `pr#21206 <https://github.com/ceph/ceph/pull/21206>`_, Jason Dillaman)
* rbd: librbd: filter out potential race with image rename (`issue#18435 <http://tracker.ceph.com/issues/18435>`_, `pr#19855 <https://github.com/ceph/ceph/pull/19855>`_, Jason Dillaman)
* rbd ls -l crashes with SIGABRT (`issue#21558 <http://tracker.ceph.com/issues/21558>`_, `pr#19801 <https://github.com/ceph/ceph/pull/19801>`_, Jason Dillaman)
* rbd-mirror: cluster watcher should ensure it has latest OSD map (`issue#22461 <http://tracker.ceph.com/issues/22461>`_, `pr#19644 <https://github.com/ceph/ceph/pull/19644>`_, Jason Dillaman)
* rbd-mirror: fix potential infinite loop when formatting status message (`issue#22932 <http://tracker.ceph.com/issues/22932>`_, `pr#20418 <https://github.com/ceph/ceph/pull/20418>`_, Mykola Golub)
* rbd-mirror: ignore permission errors on rbd_mirroring object (`issue#20571 <http://tracker.ceph.com/issues/20571>`_, `pr#21225 <https://github.com/ceph/ceph/pull/21225>`_, Jason Dillaman)
* rbd-mirror: strip environment/CLI overrides for remote cluster (`issue#21894 <http://tracker.ceph.com/issues/21894>`_, `pr#21223 <https://github.com/ceph/ceph/pull/21223>`_, Jason Dillaman)
* [rbd-nbd] Fedora does not register resize events (`issue#22131 <http://tracker.ceph.com/issues/22131>`_, `pr#19115 <https://github.com/ceph/ceph/pull/19115>`_, Jason Dillaman)
* rbd-nbd: fix ebusy when do map (`issue#23528 <http://tracker.ceph.com/issues/23528>`_, `pr#21232 <https://github.com/ceph/ceph/pull/21232>`_, Li Wang)
* rbd: possible deadlock in various maintenance operations (`issue#22120 <http://tracker.ceph.com/issues/22120>`_, `pr#20285 <https://github.com/ceph/ceph/pull/20285>`_, Jason Dillaman)
* rbd: rbd crashes during map (`issue#21808 <http://tracker.ceph.com/issues/21808>`_, `pr#18843 <https://github.com/ceph/ceph/pull/18843>`_, Peter Keresztes Schmidt)
* rbd: rbd-mirror split brain test case can have a false-positive failure until teuthology (`issue#22485 <http://tracker.ceph.com/issues/22485>`_, `pr#21205 <https://github.com/ceph/ceph/pull/21205>`_, Jason Dillaman)
* rbd: TestLibRBD.RenameViaLockOwner may still fail with -ENOENT (`issue#23068 <http://tracker.ceph.com/issues/23068>`_, `pr#20627 <https://github.com/ceph/ceph/pull/20627>`_, Mykola Golub)
* repair_test fails due to race with osd start (`issue#20705 <http://tracker.ceph.com/issues/20705>`_, `pr#20146 <https://github.com/ceph/ceph/pull/20146>`_, Sage Weil)
* rgw: 15912 15673 (Fix duplicate tag removal during GC, cls/refcount: store and use list of retired tags) (`issue#20107 <http://tracker.ceph.com/issues/20107>`_, `pr#16708 <https://github.com/ceph/ceph/pull/16708>`_, Jens Rosenboom)
* rgw: abort in listing mapped nbd devices when running in a container (`issue#22012 <http://tracker.ceph.com/issues/22012>`_, `issue#22011 <http://tracker.ceph.com/issues/22011>`_, `pr#20286 <https://github.com/ceph/ceph/pull/20286>`_, Li Wang, Pan Liu)
* rgw: add ability to sync user stats from admin api (`issue#21301 <http://tracker.ceph.com/issues/21301>`_, `pr#20179 <https://github.com/ceph/ceph/pull/20179>`_, Nathan Johnson)
* rgw: add cors header rule check in cors option request (`issue#22002 <http://tracker.ceph.com/issues/22002>`_, `pr#19057 <https://github.com/ceph/ceph/pull/19057>`_, yuliyang)
* rgw: add radosgw-admin sync error trim to trim sync error log (`issue#23287 <http://tracker.ceph.com/issues/23287>`_, `pr#21210 <https://github.com/ceph/ceph/pull/21210>`_, fang yuxiang)
* rgw: add xml output header in RGWCopyObj_ObjStore_S3 response msg (`issue#22416 <http://tracker.ceph.com/issues/22416>`_, `pr#19887 <https://github.com/ceph/ceph/pull/19887>`_, Enming Zhang)
* rgw: automated trimming of datalog and mdlog (`issue#18227 <http://tracker.ceph.com/issues/18227>`_, `pr#20061 <https://github.com/ceph/ceph/pull/20061>`_, Casey Bodley)
* rgw: bi list entry count incremented on error, distorting error code (`issue#21205 <http://tracker.ceph.com/issues/21205>`_, `pr#18207 <https://github.com/ceph/ceph/pull/18207>`_, Nathan Cutler)
* rgw: boto3 v4 SignatureDoesNotMatch failure due to sorting of sse-kms headers (`issue#21832 <http://tracker.ceph.com/issues/21832>`_, `pr#18772 <https://github.com/ceph/ceph/pull/18772>`_, Nathan Cutler)
* rgw: bucket resharding should not update bucket ACL or user stats (`issue#22124 <http://tracker.ceph.com/issues/22124>`_, `pr#20421 <https://github.com/ceph/ceph/pull/20421>`_, Orit Wasserman)
* rgw: copying part without http header x-amz-copy-source-range will be mistaken for copying object (`issue#22729 <http://tracker.ceph.com/issues/22729>`_, `pr#21294 <https://github.com/ceph/ceph/pull/21294>`_, Malcolm Lee)
* rgw: core dump, recursive lock of RGWKeystoneTokenCache (`issue#23171 <http://tracker.ceph.com/issues/23171>`_, `pr#20639 <https://github.com/ceph/ceph/pull/20639>`_, Mark Kogan, Adam Kupczyk)
* rgw: data sync of versioned objects, note updating bi marker (`issue#18885 <http://tracker.ceph.com/issues/18885>`_, `pr#21213 <https://github.com/ceph/ceph/pull/21213>`_, Yehuda Sadeh)
* rgw: dont log EBUSY errors in 'sync error list' (`issue#22473 <http://tracker.ceph.com/issues/22473>`_, `pr#19908 <https://github.com/ceph/ceph/pull/19908>`_, Casey Bodley)
* rgw: ECANCELED in rgw_get_system_obj() leads to infinite loop (`issue#17996 <http://tracker.ceph.com/issues/17996>`_, `pr#20561 <https://github.com/ceph/ceph/pull/20561>`_, Yehuda Sadeh)
* rgw: file deadlock on lru evicting (`issue#22736 <http://tracker.ceph.com/issues/22736>`_, `pr#20076 <https://github.com/ceph/ceph/pull/20076>`_, Matt Benjamin)
* rgw: fix index update in dir_suggest_changes (`issue#24280 <http://tracker.ceph.com/issues/24280>`_, `pr#22677 <https://github.com/ceph/ceph/pull/22677>`_, Tianshan Qu)
* rgw: fix marker encoding problem (`issue#20463 <http://tracker.ceph.com/issues/20463>`_, `pr#17731 <https://github.com/ceph/ceph/pull/17731>`_, Orit Wasserman, Marcus Watts)
* rgw: fix swift anonymous access (`issue#22259 <http://tracker.ceph.com/issues/22259>`_, `pr#19194 <https://github.com/ceph/ceph/pull/19194>`_, Marcus Watts)
* rgw: Fix swift object expiry not deleting objects (`issue#22084 <http://tracker.ceph.com/issues/22084>`_, `pr#18925 <https://github.com/ceph/ceph/pull/18925>`_, Pavan Rallabhandi)
* rgw: fix the bug that part's index can't be removed after completing (`issue#19604 <http://tracker.ceph.com/issues/19604>`_, `pr#16763 <https://github.com/ceph/ceph/pull/16763>`_, Zhang Shaowen, Matt Benjamin)
* rgw: fix the max-uploads parameter not work (`issue#22825 <http://tracker.ceph.com/issues/22825>`_, `pr#20479 <https://github.com/ceph/ceph/pull/20479>`_, Xin Liao)
* rgw: inefficient buffer usage for PUTs (`issue#23207 <http://tracker.ceph.com/issues/23207>`_, `pr#21098 <https://github.com/ceph/ceph/pull/21098>`_, Marcus Watts)
* rgw: list bucket which enable versioning get wrong result when user marker (`issue#21500 <http://tracker.ceph.com/issues/21500>`_, `pr#20291 <https://github.com/ceph/ceph/pull/20291>`_, yuliyang)
* rgw: log includes zero byte sometimes (`issue#20037 <http://tracker.ceph.com/issues/20037>`_, `pr#17151 <https://github.com/ceph/ceph/pull/17151>`_, Abhishek Lekshmanan)
* rgw: make init env methods return an error (`issue#23039 <http://tracker.ceph.com/issues/23039>`_, `pr#20800 <https://github.com/ceph/ceph/pull/20800>`_, Abhishek Lekshmanan)
* RGW: Multipart upload may double the quota (`issue#21586 <http://tracker.ceph.com/issues/21586>`_, `pr#18121 <https://github.com/ceph/ceph/pull/18121>`_, Sibei Gao, Matt Benjamin)
* rgw: multisite: data sync status advances despite failure in RGWListBucketIndexesCR (`issue#21735 <http://tracker.ceph.com/issues/21735>`_, `pr#20269 <https://github.com/ceph/ceph/pull/20269>`_, Casey Bodley)
* rgw: multisite: Get bucket location which is located in another zonegroup, will return 301 Moved Permanently (`issue#21125 <http://tracker.ceph.com/issues/21125>`_, `pr#18305 <https://github.com/ceph/ceph/pull/18305>`_, Shasha Lu, lvshuhua, Jiaying Ren)
* rgw: radosgw-admin: add an option to reset user stats (`issue#23335 <http://tracker.ceph.com/issues/23335>`_, `issue#23322 <http://tracker.ceph.com/issues/23322>`_, `pr#20877 <https://github.com/ceph/ceph/pull/20877>`_, Abhishek Lekshmanan)
* rgw: release cls lock if taken in RGWCompleteMultipart (`issue#21596 <http://tracker.ceph.com/issues/21596>`_, `issue#22368 <http://tracker.ceph.com/issues/22368>`_, `pr#18116 <https://github.com/ceph/ceph/pull/18116>`_, Casey Bodley, Matt Benjamin)
* rgw: resharding needs to set back the bucket ACL after link (`issue#22742 <http://tracker.ceph.com/issues/22742>`_, `pr#20039 <https://github.com/ceph/ceph/pull/20039>`_, Orit Wasserman)
* rgw: resolve Random 500 errors in Swift PutObject (22517) (`issue#22517 <http://tracker.ceph.com/issues/22517>`_, `issue#21560 <http://tracker.ceph.com/issues/21560>`_, `pr#19769 <https://github.com/ceph/ceph/pull/19769>`_, Adam C. Emerson, Matt Benjamin)
* rgw: rgw_file: recursive lane lock can occur in LRU drain (`issue#20374 <http://tracker.ceph.com/issues/20374>`_, `pr#17149 <https://github.com/ceph/ceph/pull/17149>`_, Matt Benjamin)
* rgw: S3 POST policy should not require Content-Type (`issue#20201 <http://tracker.ceph.com/issues/20201>`_, `pr#19635 <https://github.com/ceph/ceph/pull/19635>`_, Matt Benjamin)
* rgw: s3website error handler uses original object name (`issue#23201 <http://tracker.ceph.com/issues/23201>`_, `issue#20307 <http://tracker.ceph.com/issues/20307>`_, `pr#21100 <https://github.com/ceph/ceph/pull/21100>`_, liuhong, Casey Bodley)
* rgw: segfaults after running radosgw-admin data sync init (`issue#22083 <http://tracker.ceph.com/issues/22083>`_, `pr#19783 <https://github.com/ceph/ceph/pull/19783>`_, Casey Bodley, Abhishek Lekshmanan)
* rgw: segmentation fault when starting radosgw after reverting .rgw.root (`issue#21996 <http://tracker.ceph.com/issues/21996>`_, `pr#20292 <https://github.com/ceph/ceph/pull/20292>`_, Orit Wasserman, Casey Bodley)
* rgw: stale bucket index entry remains after object deletion (`issue#22555 <http://tracker.ceph.com/issues/22555>`_, `pr#20293 <https://github.com/ceph/ceph/pull/20293>`_, J. Eric Ivancich)
* rgw: system user can't delete bucket completely (`issue#22248 <http://tracker.ceph.com/issues/22248>`_, `pr#21212 <https://github.com/ceph/ceph/pull/21212>`_, Casey Bodley)
* rgw: tcmalloc (`issue#23469 <http://tracker.ceph.com/issues/23469>`_, `pr#21073 <https://github.com/ceph/ceph/pull/21073>`_, Matt Benjamin)
* rgw: upldate the max-buckets when the quota is uploaded (`issue#22745 <http://tracker.ceph.com/issues/22745>`_, `pr#20496 <https://github.com/ceph/ceph/pull/20496>`_, zhaokun)
* rgw: user creation can overwrite existing user even if different uid is given (`issue#21685 <http://tracker.ceph.com/issues/21685>`_, `pr#20074 <https://github.com/ceph/ceph/pull/20074>`_, Casey Bodley)
* RHEL 7.3 Selinux denials at OSD start (`issue#19200 <http://tracker.ceph.com/issues/19200>`_, `pr#18780 <https://github.com/ceph/ceph/pull/18780>`_, Boris Ranto)
* scrub errors not cleared on replicas can cause inconsistent pg state when replica takes over primary (`issue#23267 <http://tracker.ceph.com/issues/23267>`_, `pr#21194 <https://github.com/ceph/ceph/pull/21194>`_, David Zafman)
* snapset xattr corruption propagated from primary to other shards (`issue#20186 <http://tracker.ceph.com/issues/20186>`_, `issue#18409 <http://tracker.ceph.com/issues/18409>`_, `issue#21907 <http://tracker.ceph.com/issues/21907>`_, `pr#20331 <https://github.com/ceph/ceph/pull/20331>`_, David Zafman)
* table of contents doesn't render for luminous/jewel docs (`issue#23780 <http://tracker.ceph.com/issues/23780>`_, `pr#21503 <https://github.com/ceph/ceph/pull/21503>`_, Alfredo Deza)
* test: Adjust for Jewel quirk caused of differences with master (`issue#23006 <http://tracker.ceph.com/issues/23006>`_, `pr#20463 <https://github.com/ceph/ceph/pull/20463>`_, David Zafman)
* test_health_warnings.sh can fail (`issue#21121 <http://tracker.ceph.com/issues/21121>`_, `pr#20289 <https://github.com/ceph/ceph/pull/20289>`_, Sage Weil)
* test/librbd: fixed metadata tests under upgrade scenarios (`issue#21911 <http://tracker.ceph.com/issues/21911>`_, `pr#18548 <https://github.com/ceph/ceph/pull/18548>`_, Jason Dillaman)
* test/librbd: utilize unique pool for cache tier testing (`issue#11502 <http://tracker.ceph.com/issues/11502>`_, `pr#20524 <https://github.com/ceph/ceph/pull/20524>`_, Jason Dillaman)
* tests: rbd_mirror_helpers.sh request_resync_image function saves image id to wrong variable (`issue#21663 <http://tracker.ceph.com/issues/21663>`_, `pr#19804 <https://github.com/ceph/ceph/pull/19804>`_, Jason Dillaman)
* tests: test_admin_socket.sh may fail on wait_for_clean (`issue#23499 <http://tracker.ceph.com/issues/23499>`_, `pr#21125 <https://github.com/ceph/ceph/pull/21125>`_, Mykola Golub)
* tests: tests/librbd: updated test_notify to handle new release lock semantics (`issue#21912 <http://tracker.ceph.com/issues/21912>`_, `pr#18560 <https://github.com/ceph/ceph/pull/18560>`_, Jason Dillaman)
This point release brings a number of important bugfixes in all major
components of Ceph, we recommend all Jewel 10.2.x users to upgrade.
For a detailed list of changes refer to :download: `the complete changelog <../changelog/v10.2.10txt>`
Notable Changes
-----------------
* build/ops: Add fix subcommand to ceph-disk, fix SELinux denials, and speed up upgrade from non-SELinux enabled ceph to an SELinux enabled one (`issue#20077 <http://tracker.ceph.com/issues/20077>`_, `issue#20184 <http://tracker.ceph.com/issues/20184>`_, `issue#19545 <http://tracker.ceph.com/issues/19545>`_, `pr#14346 <https://github.com/ceph/ceph/pull/14346>`_, Boris Ranto)
* build/ops: rpm: fix python-Sphinx package name for SUSE (`issue#19924 <http://tracker.ceph.com/issues/19924>`_, `pr#15196 <https://github.com/ceph/ceph/pull/15196>`_, Nathan Cutler, Jan Matejek)
* build/ops: rpm: set subman cron attributes in spec file (`issue#20074 <http://tracker.ceph.com/issues/20074>`_, `pr#15473 <https://github.com/ceph/ceph/pull/15473>`_, Thomas Serlin)
* cephfs: ceph-fuse segfaults at mount time, assert in ceph::log::Log::stop (`issue#18157 <http://tracker.ceph.com/issues/18157>`_, `pr#16963 <https://github.com/ceph/ceph/pull/16963>`_, Greg Farnum)
* cephfs: df reports negative disk "used" value when quota exceed (`issue#20178 <http://tracker.ceph.com/issues/20178>`_, `pr#16151 <https://github.com/ceph/ceph/pull/16151>`_, John Spray)
* cephfs: get_quota_root sends lookupname op for every buffered write (`issue#20945 <http://tracker.ceph.com/issues/20945>`_, `pr#17396 <https://github.com/ceph/ceph/pull/17396>`_, Dan van der Ster)
* cephfs: osdc/Filer: truncate large file party by party (`issue#19755 <http://tracker.ceph.com/issues/19755>`_, `pr#15442 <https://github.com/ceph/ceph/pull/15442>`_, "Yan, Zheng")
* core: an OSD was seen getting ENOSPC even with osd_failsafe_full_ratio passed (`issue#20544 <http://tracker.ceph.com/issues/20544>`_, `issue#16878 <http://tracker.ceph.com/issues/16878>`_, `issue#19733 <http://tracker.ceph.com/issues/19733>`_, `issue#15912 <http://tracker.ceph.com/issues/15912>`_, `pr#15050 <https://github.com/ceph/ceph/pull/15050>`_, Sage Weil, David Zafman)
* core: disable skewed utilization warning by default (`issue#20730 <http://tracker.ceph.com/issues/20730>`_, `pr#17210 <https://github.com/ceph/ceph/pull/17210>`_, David Zafman)
* core: kv: let ceph_logger destructed after db reset (`issue#21336 <http://tracker.ceph.com/issues/21336>`_, `pr#17626 <https://github.com/ceph/ceph/pull/17626>`_, wumingqiao)
* libradosstriper silently fails to delete empty objects in jewel (`issue#20325 <http://tracker.ceph.com/issues/20325>`_, `pr#15760 <https://github.com/ceph/ceph/pull/15760>`_, Stan K)
* librbd: fail IO request when exclusive lock cannot be obtained (`issue#20168 <http://tracker.ceph.com/issues/20168>`_, `issue#21251 <http://tracker.ceph.com/issues/21251>`_, `pr#17402 <https://github.com/ceph/ceph/pull/17402>`_, Jason Dillaman)
* librbd: prevent self-blacklisting during break lock (`issue#18666 <http://tracker.ceph.com/issues/18666>`_, `pr#17412 <https://github.com/ceph/ceph/pull/17412>`_, Jason Dillaman)
* librbd: reacquire lock should update lock owner client id (`issue#19929 <http://tracker.ceph.com/issues/19929>`_, `pr#17385 <https://github.com/ceph/ceph/pull/17385>`_, Jason Dillaman)
* mds: damage reporting by ino number is useless (`issue#18509 <http://tracker.ceph.com/issues/18509>`_, `issue#16016 <http://tracker.ceph.com/issues/16016>`_, `pr#14699 <https://github.com/ceph/ceph/pull/14699>`_, John Spray, Michal Jarzabek)
* mds: log rotation doesn't work if mds has respawned (`issue#19291 <http://tracker.ceph.com/issues/19291>`_, `pr#14673 <https://github.com/ceph/ceph/pull/14673>`_, Patrick Donnelly)
* mds: save projected path into inode_t::stray_prior_path (`issue#20340 <http://tracker.ceph.com/issues/20340>`_, `pr#16150 <https://github.com/ceph/ceph/pull/16150>`_, "Yan, Zheng")
* mon: crash on shutdown, lease_ack_timeout event (`issue#19825 <http://tracker.ceph.com/issues/19825>`_, `pr#15083 <https://github.com/ceph/ceph/pull/15083>`_, Kefu Chai, Michal Jarzabek, Alexey Sheplyakov)
* mon: Disallow enabling 'hashpspool' option to a pool without some kind of --i-understand-this-will-remap-all-pgs flag (`issue#18468 <http://tracker.ceph.com/issues/18468>`_, `pr#13507 <https://github.com/ceph/ceph/pull/13507>`_, Vikhyat Umrao)
* mon: factor mon_osd_full_ratio into MAX AVAIL calc (`issue#18522 <http://tracker.ceph.com/issues/18522>`_, `pr#15236 <https://github.com/ceph/ceph/pull/15236>`_, Sage Weil)
* mon: fail to form large quorum; msg/async busy loop (`issue#20230 <http://tracker.ceph.com/issues/20230>`_, `pr#15726 <https://github.com/ceph/ceph/pull/15726>`_, Haomai Wang, Michal Jarzabek)
* osd: objecter full_try behavior not consistent with osd (`issue#19430 <http://tracker.ceph.com/issues/19430>`_, `pr#15474 <https://github.com/ceph/ceph/pull/15474>`_, Sage Weil)
* osd: omap threadpool heartbeat is only reset every 100 values (`issue#20375 <http://tracker.ceph.com/issues/20375>`_, `pr#16167 <https://github.com/ceph/ceph/pull/16167>`_, Josh Durgin)
* osd: osd_internal_types: wake snaptrimmer on put_read lock, too (`issue#19131 <http://tracker.ceph.com/issues/19131>`_, `pr#16015 <https://github.com/ceph/ceph/pull/16015>`_, Sage Weil)
* osd: PrimaryLogPG: do not call on_shutdown() if (pg.deleting) (`issue#19902 <http://tracker.ceph.com/issues/19902>`_, `pr#15065 <https://github.com/ceph/ceph/pull/15065>`_, Kefu Chai)
* osd: rados ls on pool with no access returns no error (`issue#20043 <http://tracker.ceph.com/issues/20043>`_, `issue#19790 <http://tracker.ceph.com/issues/19790>`_, `pr#16473 <https://github.com/ceph/ceph/pull/16473>`_, Nathan Cutler, Kefu Chai, John Spray, Sage Weil, Brad Hubbard)
* osd: Reset() snaptrimmer on shutdown and do not default-abort on leaked pg refs (`issue#19931 <http://tracker.ceph.com/issues/19931>`_, `pr#15322 <https://github.com/ceph/ceph/pull/15322>`_, Greg Farnum)
* osd: scrub_to specifies clone ver, but transaction include head write ver (`issue#20041 <http://tracker.ceph.com/issues/20041>`_, `pr#16405 <https://github.com/ceph/ceph/pull/16405>`_, David Zafman)
* osd: unlock sdata_op_ordering_lock with sdata_lock hold to avoid missing wakeup signal (`issue#20427 <http://tracker.ceph.com/issues/20427>`_, `pr#15947 <https://github.com/ceph/ceph/pull/15947>`_, Alexey Sheplyakov)
* qa: add a sleep after restarting osd before "tell"ing it (`issue#16239 <http://tracker.ceph.com/issues/16239>`_, `pr#15475 <https://github.com/ceph/ceph/pull/15475>`_, Kefu Chai)
* rbd: cli: ensure positional arguments exist before casting (`issue#20185 <http://tracker.ceph.com/issues/20185>`_, `pr#16295 <https://github.com/ceph/ceph/pull/16295>`_, Jason Dillaman)
* rbd: cli: map with cephx disabled results in error message (`issue#19035 <http://tracker.ceph.com/issues/19035>`_, `pr#16297 <https://github.com/ceph/ceph/pull/16297>`_, Jason Dillaman)
* rbd: default features should be negotiated with the OSD (`issue#17010 <http://tracker.ceph.com/issues/17010>`_, `pr#14874 <https://github.com/ceph/ceph/pull/14874>`_, Mykola Golub, Jason Dillaman)
* rbd: Enabling mirroring for a pool with clones may fail (`issue#19798 <http://tracker.ceph.com/issues/19798>`_, `issue#19130 <http://tracker.ceph.com/issues/19130>`_, `pr#14663 <https://github.com/ceph/ceph/pull/14663>`_, Mykola Golub, Jason Dillaman)
* rbd-mirror: image sync should send NOCACHE advise flag (`issue#17127 <http://tracker.ceph.com/issues/17127>`_, `pr#16285 <https://github.com/ceph/ceph/pull/16285>`_, Mykola Golub)
* rbd: object-map: batch updates during trim operation (`issue#17356 <http://tracker.ceph.com/issues/17356>`_, `pr#15460 <https://github.com/ceph/ceph/pull/15460>`_, Mykola Golub, Venky Shankar, Nathan Cutler)
* rbd: Potential IO hang if image is flattened while read request is in-flight (`issue#19832 <http://tracker.ceph.com/issues/19832>`_, `pr#15464 <https://github.com/ceph/ceph/pull/15464>`_, Jason Dillaman)
* rbd: rbd_clone_copy_on_read ineffective with exclusive-lock (`issue#18888 <http://tracker.ceph.com/issues/18888>`_, `pr#16124 <https://github.com/ceph/ceph/pull/16124>`_, Nathan Cutler, Venky Shankar, Jason Dillaman)
* rbd: rbd-mirror: ensure missing images are re-synced when detected (`issue#19811 <http://tracker.ceph.com/issues/19811>`_, `pr#15488 <https://github.com/ceph/ceph/pull/15488>`_, Jason Dillaman)
* rbd: rbd-mirror: failover and failback of unmodified image results in split-brain (`issue#19858 <http://tracker.ceph.com/issues/19858>`_, `pr#14977 <https://github.com/ceph/ceph/pull/14977>`_, Jason Dillaman)
* rgw: add the remove-x-delete feature to cancel swift object expiration (`issue#19074 <http://tracker.ceph.com/issues/19074>`_, `pr#14659 <https://github.com/ceph/ceph/pull/14659>`_, Jing Wenjun)
* rgw: aws4: add rgw_s3_auth_aws4_force_boto2_compat conf option (`issue#16463 <http://tracker.ceph.com/issues/16463>`_, `pr#17009 <https://github.com/ceph/ceph/pull/17009>`_, Javier M. Mellid)
* rgw: bucket index check in radosgw-admin removes valid index (`issue#18470 <http://tracker.ceph.com/issues/18470>`_, `pr#16856 <https://github.com/ceph/ceph/pull/16856>`_, Zhang Shaowen, Pavan Rallabhandi)
* rgw: cls: ceph::timespan tag_timeout wrong units (`issue#20380 <http://tracker.ceph.com/issues/20380>`_, `pr#16289 <https://github.com/ceph/ceph/pull/16289>`_, Matt Benjamin)
* rgw: Custom data header support (`issue#19644 <http://tracker.ceph.com/issues/19644>`_, `pr#15966 <https://github.com/ceph/ceph/pull/15966>`_, Pavan Rallabhandi)
* rgw: datalog trim can't work as expected (`issue#20190 <http://tracker.ceph.com/issues/20190>`_, `pr#16299 <https://github.com/ceph/ceph/pull/16299>`_, Zhang Shaowen)
* rgw: Do not decrement stats cache when the cache values are zero (`issue#20661 <http://tracker.ceph.com/issues/20661>`_, `issue#20934 <http://tracker.ceph.com/issues/20934>`_, `pr#16720 <https://github.com/ceph/ceph/pull/16720>`_, Aleksei Gutikov, Pavan Rallabhandi)
* rgw: fix crash caused by shard id out of range when listing data log (`issue#19732 <http://tracker.ceph.com/issues/19732>`_, `pr#15465 <https://github.com/ceph/ceph/pull/15465>`_, redickwang)
* rgw: fix hangs in RGWRealmReloader::reload on SIGHUP (`issue#20686 <http://tracker.ceph.com/issues/20686>`_, `pr#17281 <https://github.com/ceph/ceph/pull/17281>`_, fang.yuxiang)
* rgw: fix infinite loop in rest api for log list (`issue#20386 <http://tracker.ceph.com/issues/20386>`_, `pr#15988 <https://github.com/ceph/ceph/pull/15988>`_, xierui, Casey Bodley)
* rgw: fix race in RGWCompleteMultipart (`issue#20861 <http://tracker.ceph.com/issues/20861>`_, `pr#16767 <https://github.com/ceph/ceph/pull/16767>`_, Abhishek Varshney, Matt Benjamin)
* rgw: Fix up to 1000 entries at a time in check_bad_index_multipart (`issue#20772 <http://tracker.ceph.com/issues/20772>`_, `pr#16880 <https://github.com/ceph/ceph/pull/16880>`_, Orit Wasserman, Matt Benjamin)
* rgw: folders starting with _ underscore are not in bucket index (`issue#19432 <http://tracker.ceph.com/issues/19432>`_, `pr#16276 <https://github.com/ceph/ceph/pull/16276>`_, Giovani Rinaldi, Orit Wasserman)
* rgw: 'gc list --include-all' command infinite loop the first 1000 items (`issue#19978 <http://tracker.ceph.com/issues/19978>`_, `pr#15719 <https://github.com/ceph/ceph/pull/15719>`_, Shasha Lu, fang yuxiang)
* rgw: meta sync thread crash at RGWMetaSyncShardCR (`issue#20251 <http://tracker.ceph.com/issues/20251>`_, `pr#16711 <https://github.com/ceph/ceph/pull/16711>`_, fang yuxiang, Nathan Cutler)
* rgw: multipart parts on versioned bucket create versioned bucket index entries (`issue#19604 <http://tracker.ceph.com/issues/19604>`_, `issue#17964 <http://tracker.ceph.com/issues/17964>`_, `pr#17278 <https://github.com/ceph/ceph/pull/17278>`_, Zhang Shaowen)
* rgw: multisite: after CreateBucket is forwarded to master, local bucket may use different value for bucket index shards (`issue#19745 <http://tracker.ceph.com/issues/19745>`_, `pr#15450 <https://github.com/ceph/ceph/pull/15450>`_, Shasha Lu)
* rgw: multisite: bucket zonegroup redirect not working (`issue#19488 <http://tracker.ceph.com/issues/19488>`_, `pr#15448 <https://github.com/ceph/ceph/pull/15448>`_, Casey Bodley)
* rgw: multisite: fixes for meta sync across periods (`issue#18639 <http://tracker.ceph.com/issues/18639>`_, `pr#15556 <https://github.com/ceph/ceph/pull/15556>`_, Casey Bodley)
* rgw: multisite: lock is not released when RGWMetaSyncShardCR::full_sync() fails to write marker (`issue#18077 <http://tracker.ceph.com/issues/18077>`_, `pr#17155 <https://github.com/ceph/ceph/pull/17155>`_, Zhang Shaowen)
* rgw: multisite: log_meta on secondary zone causes continuous loop of metadata sync (`issue#20357 <http://tracker.ceph.com/issues/20357>`_, `issue#20244 <http://tracker.ceph.com/issues/20244>`_, `pr#17148 <https://github.com/ceph/ceph/pull/17148>`_, Orit Wasserman, Casey Bodley)
* rgw: multisite: operating bucket's acl&cors is not restricted on slave zone (`issue#16888 <http://tracker.ceph.com/issues/16888>`_, `pr#15453 <https://github.com/ceph/ceph/pull/15453>`_, Casey Bodley, Shasha Lu, Guo Zhandong)
* rgw: multisite: realm rename does not propagate to other clusters (`issue#19746 <http://tracker.ceph.com/issues/19746>`_, `pr#15454 <https://github.com/ceph/ceph/pull/15454>`_, Casey Bodley)
* rgw: multisite: rest api fails to decode large period on "period commit" (`issue#19505 <http://tracker.ceph.com/issues/19505>`_, `pr#15447 <https://github.com/ceph/ceph/pull/15447>`_, Casey Bodley)
* rgw: multisite: RGWPeriodPuller does not call RGWPeriod::reflect() on new period (`issue#19816 <http://tracker.ceph.com/issues/19816>`_, `issue#19817 <http://tracker.ceph.com/issues/19817>`_, `pr#17167 <https://github.com/ceph/ceph/pull/17167>`_, Casey Bodley)
* rgw: multisite: RGWRadosRemoveOmapKeysCR::request_complete return val is wrong (`issue#20539 <http://tracker.ceph.com/issues/20539>`_, `pr#17156 <https://github.com/ceph/ceph/pull/17156>`_, Shasha Lu)
* rgw: not initialized pointer cause rgw crash with ec data pool (`issue#20542 <http://tracker.ceph.com/issues/20542>`_, `pr#17164 <https://github.com/ceph/ceph/pull/17164>`_, Aleksei Gutikov, fang yuxiang)
* rgw: radosgw-admin: bucket rm with --bypass-gc and without --purge-data doesn't throw error message (`issue#20688 <http://tracker.ceph.com/issues/20688>`_, `pr#17159 <https://github.com/ceph/ceph/pull/17159>`_, Abhishek Varshney)
* rgw: radosgw-admin data sync run crash (`issue#20423 <http://tracker.ceph.com/issues/20423>`_, `pr#17165 <https://github.com/ceph/ceph/pull/17165>`_, Shasha Lu)
* rgw: reduce log level of 'storing entry at' in cls_log (`issue#19835 <http://tracker.ceph.com/issues/19835>`_, `pr#15455 <https://github.com/ceph/ceph/pull/15455>`_, Willem Jan Withagen)
* rgw: remove unnecessary 'error in read_id for object name: default' (`issue#19922 <http://tracker.ceph.com/issues/19922>`_, `pr#15197 <https://github.com/ceph/ceph/pull/15197>`_, weiqiaomiao)
* rgw: replace '+' with "%20" in canonical query string for s3 v4 auth (`issue#20501 <http://tracker.ceph.com/issues/20501>`_, `pr#16951 <https://github.com/ceph/ceph/pull/16951>`_, Zhang Shaowen, Matt Benjamin)
* rgw: rgw_common.cc: modify the end check in RGWHTTPArgs::sys_get (`issue#16072 <http://tracker.ceph.com/issues/16072>`_, `pr#16268 <https://github.com/ceph/ceph/pull/16268>`_, zhao kun)
* rgw: rgw_file: fix size and (c|m)time unix attrs in write_finish (`issue#19653 <http://tracker.ceph.com/issues/19653>`_, `pr#15449 <https://github.com/ceph/ceph/pull/15449>`_, Matt Benjamin)
* rgw: rgw_file: incorrect lane lock behavior in evict_block() (`issue#21141 <http://tracker.ceph.com/issues/21141>`_, `pr#17597 <https://github.com/ceph/ceph/pull/17597>`_, Matt Benjamin)
* rgw: rgw_file: prevent conflict of mkdir between restarts (`issue#20275 <http://tracker.ceph.com/issues/20275>`_, `pr#17147 <https://github.com/ceph/ceph/pull/17147>`_, Gui Hecheng)
* rgw: rgw_file: v3 write timer does not close open handles (`issue#19932 <http://tracker.ceph.com/issues/19932>`_, `pr#15456 <https://github.com/ceph/ceph/pull/15456>`_, Matt Benjamin)
* rgw: Segmentation fault when exporting rgw bucket in nfs-ganesha (`issue#20663 <http://tracker.ceph.com/issues/20663>`_, `pr#17285 <https://github.com/ceph/ceph/pull/17285>`_, Matt Benjamin)
* rgw: user quota did not work well on multipart upload (`issue#19285 <http://tracker.ceph.com/issues/19285>`_, `issue#19602 <http://tracker.ceph.com/issues/19602>`_, `pr#17277 <https://github.com/ceph/ceph/pull/17277>`_, Zhang Shaowen)
* rgw: VersionIdMarker and NextVersionIdMarker are not returned when listing object versions (`issue#19886 <http://tracker.ceph.com/issues/19886>`_, `pr#16316 <https://github.com/ceph/ceph/pull/16316>`_, Zhang Shaowen)
* rgw: when uploading objects continuously into a versioned bucket, some objects will not sync (`issue#18208 <http://tracker.ceph.com/issues/18208>`_, `pr#15452 <https://github.com/ceph/ceph/pull/15452>`_, lvshuhua)
* tools: ceph cli: Rados object in state configuring race (`issue#16477 <http://tracker.ceph.com/issues/16477>`_, `pr#15762 <https://github.com/ceph/ceph/pull/15762>`_, Loic Dachary)
* tools: ceph-disk: dmcrypt cluster must default to ceph (`issue#20893 <http://tracker.ceph.com/issues/20893>`_, `pr#16870 <https://github.com/ceph/ceph/pull/16870>`_, Loic Dachary)
This point release brought a number of important bugfixes in all major
components of Ceph. However, it also introduced a regression that could cause
MDS damage, and a new release, v10.2.9, was published to address this.
Therefore, Jewel users should *not* upgrade to this version - instead, we
recommend upgrading directly to v10.2.9.
For more detailed information, see :download:`the complete changelog <../changelog/v10.2.8.txt>`.
OSD Removal Caveat
------------------
There was a bug introduced in Jewel (#19119) that broke the mapping behavior
when an "out" OSD that still existed in the CRUSH map was removed with 'osd rm'.
This could result in 'misdirected op' and other errors. The bug is now fixed,
but the fix itself introduces the same risk because the behavior may vary between
clients and OSDs. To avoid problems, please ensure that all OSDs are removed
from the CRUSH map before deleting them. That is, be sure to do::
ceph osd crush rm osd.123
before::
ceph osd rm osd.123
Snap Trimmer Improvements
-------------------------
This release greatly improves control and throttling of the snap trimmer. It
introduces the "osd max trimming pgs" option (defaulting to 2), which limits
how many PGs on an OSD can be trimming snapshots at a time. And it restores
the safe use of the "osd snap trim sleep" option, wihch defaults to 0 but
otherwise adds the given number of seconds in delay between every dispatch
of trim operations to the underlying system.
Other Notable Changes
---------------------
* build/ops: "osd marked itself down" will not recognised if host runs mon + osd on shutdown/reboot (`issue#18516 <http://tracker.ceph.com/issues/18516>`_, `pr#13492 <https://github.com/ceph/ceph/pull/13492>`_, Boris Ranto)
* build/ops: enable build of ceph-resource-agents package on rpm-based os (`issue#17613 <http://tracker.ceph.com/issues/17613>`_, `issue#19546 <http://tracker.ceph.com/issues/19546>`_, `pr#13606 <https://github.com/ceph/ceph/pull/13606>`_, Nathan Cutler)
* build/ops: rbdmap.service not included in debian packaging (jewel-only) (`issue#19547 <http://tracker.ceph.com/issues/19547>`_, `pr#14383 <https://github.com/ceph/ceph/pull/14383>`_, Ken Dreyer)
* cephfs: MDS assert failed when shutting down (`issue#19204 <http://tracker.ceph.com/issues/19204>`_, `pr#14683 <https://github.com/ceph/ceph/pull/14683>`_, John Spray)
* cephfs: MDS goes readonly writing backtrace for a file whose data pool has been removed (`issue#19401 <http://tracker.ceph.com/issues/19401>`_, `pr#14682 <https://github.com/ceph/ceph/pull/14682>`_, John Spray)
* cephfs: MDS server crashes due to inconsistent metadata (`issue#19406 <http://tracker.ceph.com/issues/19406>`_, `pr#14676 <https://github.com/ceph/ceph/pull/14676>`_, John Spray)
* cephfs: No output for ceph mds rmfailed 0 --yes-i-really-mean-it command (`issue#16709 <http://tracker.ceph.com/issues/16709>`_, `pr#14674 <https://github.com/ceph/ceph/pull/14674>`_, John Spray)
* cephfs: Test failure: test_open_inode (`issue#18661 <http://tracker.ceph.com/issues/18661>`_, `pr#14669 <https://github.com/ceph/ceph/pull/14669>`_, John Spray)
* cephfs: The mount point break off when mds switch hanppened (`issue#19437 <http://tracker.ceph.com/issues/19437>`_, `pr#14679 <https://github.com/ceph/ceph/pull/14679>`_, Guan yunfei)
* cephfs: ceph-fuse does not recover after lost connection to MDS (`issue#16743 <http://tracker.ceph.com/issues/16743>`_, `issue#18757 <http://tracker.ceph.com/issues/18757>`_, `pr#14698 <https://github.com/ceph/ceph/pull/14698>`_, Kefu Chai, Henrik Korkuc, Patrick Donnelly)
* cephfs: mds is crushed, after I set about 400 64KB xattr kv pairs to a file (`issue#19033 <http://tracker.ceph.com/issues/19033>`_, `pr#14684 <https://github.com/ceph/ceph/pull/14684>`_, Yang Honggang)
* cephfs: non-local quota changes not visible until some IO is done (`issue#17939 <http://tracker.ceph.com/issues/17939>`_, `pr#15466 <https://github.com/ceph/ceph/pull/15466>`_, John Spray, Nathan Cutler)
* cephfs: normalize file open flags internally used by cephfs (`issue#18872 <http://tracker.ceph.com/issues/18872>`_, `issue#19890 <http://tracker.ceph.com/issues/19890>`_, `pr#15000 <https://github.com/ceph/ceph/pull/15000>`_, Jan Fajerski, "Yan, Zheng")
* common: monitor creation with IPv6 public network segfaults (`issue#19371 <http://tracker.ceph.com/issues/19371>`_, `pr#14324 <https://github.com/ceph/ceph/pull/14324>`_, Fabian Grünbichler)
* common: radosstriper: protect aio_write API from calls with 0 bytes (`issue#14609 <http://tracker.ceph.com/issues/14609>`_, `pr#13254 <https://github.com/ceph/ceph/pull/13254>`_, Sebastien Ponce)
* core: clear divergent_priors set off disk (`issue#17916 <http://tracker.ceph.com/issues/17916>`_, `pr#14596 <https://github.com/ceph/ceph/pull/14596>`_, Greg Farnum)
* core: improve snap trimming, enable restriction of parallelism (`issue#19241 <http://tracker.ceph.com/issues/19241>`_, `pr#14492 <https://github.com/ceph/ceph/pull/14492>`_, Samuel Just, Greg Farnum)
* core: os/filestore/HashIndex: be loud about splits (`issue#18235 <http://tracker.ceph.com/issues/18235>`_, `pr#13788 <https://github.com/ceph/ceph/pull/13788>`_, Dan van der Ster)
* core: transient jerasure unit test failures (`issue#18070 <http://tracker.ceph.com/issues/18070>`_, `issue#17762 <http://tracker.ceph.com/issues/17762>`_, `issue#18128 <http://tracker.ceph.com/issues/18128>`_, `issue#17951 <http://tracker.ceph.com/issues/17951>`_, `pr#14701 <https://github.com/ceph/ceph/pull/14701>`_, Kefu Chai, Pan Liu, Loic Dachary, Jason Dillaman)
* core: two instances of omap_digest mismatch (`issue#18533 <http://tracker.ceph.com/issues/18533>`_, `pr#14204 <https://github.com/ceph/ceph/pull/14204>`_, Samuel Just, David Zafman)
* doc: radosgw-admin: add the 'object stat' command to usage (`issue#19013 <http://tracker.ceph.com/issues/19013>`_, `pr#13872 <https://github.com/ceph/ceph/pull/13872>`_, Pavan Rallabhandi)
* doc: rgw S3 create bucket should not do response in json (`issue#18889 <http://tracker.ceph.com/issues/18889>`_, `pr#13874 <https://github.com/ceph/ceph/pull/13874>`_, Abhishek Lekshmanan)
* fs: Invalid error code returned by MDS is causing a kernel client WARNING (`issue#19205 <http://tracker.ceph.com/issues/19205>`_, `pr#13831 <https://github.com/ceph/ceph/pull/13831>`_, Jan Fajerski, xie xingguo)
* librbd: Incomplete declaration for ContextWQ in librbd/Journal.h (`issue#18862 <http://tracker.ceph.com/issues/18862>`_, `pr#14152 <https://github.com/ceph/ceph/pull/14152>`_, Boris Ranto)
* librbd: Issues with C API image metadata retrieval functions (`issue#19588 <http://tracker.ceph.com/issues/19588>`_, `pr#14666 <https://github.com/ceph/ceph/pull/14666>`_, Mykola Golub)
* librbd: Possible deadlock performing a synchronous API action while refresh in-progress (`issue#18419 <http://tracker.ceph.com/issues/18419>`_, `pr#13154 <https://github.com/ceph/ceph/pull/13154>`_, Jason Dillaman)
* librbd: is_exclusive_lock_owner API should ping OSD (`issue#19287 <http://tracker.ceph.com/issues/19287>`_, `pr#14481 <https://github.com/ceph/ceph/pull/14481>`_, Jason Dillaman)
* mds: Too many stat ops when trying to probe a large file (`issue#19955 <http://tracker.ceph.com/issues/19955>`_, `pr#15472 <https://github.com/ceph/ceph/pull/15472>`_, "Yan, Zheng")
* mds: enable start when session ino info is corrupt (`issue#19708 <http://tracker.ceph.com/issues/19708>`_, `issue#16842 <http://tracker.ceph.com/issues/16842>`_, `pr#14700 <https://github.com/ceph/ceph/pull/14700>`_, John Spray)
* mds: fragment space check can cause replayed request fail (`issue#18660 <http://tracker.ceph.com/issues/18660>`_, `pr#14668 <https://github.com/ceph/ceph/pull/14668>`_, "Yan, Zheng")
* mds: heartbeat timeout during rejoin, when working with large amount of caps/inodes (`issue#19118 <http://tracker.ceph.com/issues/19118>`_, `pr#14672 <https://github.com/ceph/ceph/pull/14672>`_, John Spray)
* mds: issue new caps when sending reply to client (`issue#19635 <http://tracker.ceph.com/issues/19635>`_, `pr#15438 <https://github.com/ceph/ceph/pull/15438>`_, "Yan, Zheng")
* mon: OSDMonitor: make 'osd crush move ...' work on osds (`issue#18587 <http://tracker.ceph.com/issues/18587>`_, `pr#13261 <https://github.com/ceph/ceph/pull/13261>`_, Sage Weil)
* msg: IPv6 Heartbeat packets are not marked with DSCP QoS - simple messenger (`issue#18887 <http://tracker.ceph.com/issues/18887>`_, `pr#13450 <https://github.com/ceph/ceph/pull/13450>`_, Yan Jun, Robin H. Johnson)
* msg: set close on exec flag (`issue#16390 <http://tracker.ceph.com/issues/16390>`_, `pr#13585 <https://github.com/ceph/ceph/pull/13585>`_, Kefu Chai)
* osd: Give requested scrubs a higher priority (`issue#15789 <http://tracker.ceph.com/issues/15789>`_, `pr#14686 <https://github.com/ceph/ceph/pull/14686>`_, David Zafman)
* osd: Object level shard errors are tracked and used if no auth available (`issue#20089 <http://tracker.ceph.com/issues/20089>`_, `pr#15416 <https://github.com/ceph/ceph/pull/15416>`_, David Zafman)
* osd: ReplicatedPG: try with pool's use-gmt setting if hitset archive not found (`issue#19185 <http://tracker.ceph.com/issues/19185>`_, `pr#13827 <https://github.com/ceph/ceph/pull/13827>`_, Kefu Chai)
* osd: allow client throttler to be adjusted on-fly, without restart (`issue#18791 <http://tracker.ceph.com/issues/18791>`_, `pr#13214 <https://github.com/ceph/ceph/pull/13214>`_, Piotr Dałek)
* osd: bypass readonly ops when osd full (`issue#19394 <http://tracker.ceph.com/issues/19394>`_, `pr#14181 <https://github.com/ceph/ceph/pull/14181>`_, Jianpeng Ma, yaoning)
* osd: degraded and misplaced status output inaccurate (`issue#18619 <http://tracker.ceph.com/issues/18619>`_, `pr#14325 <https://github.com/ceph/ceph/pull/14325>`_, David Zafman)
* osd: new added OSD always down when full flag is set (`issue#15025 <http://tracker.ceph.com/issues/15025>`_, `pr#14326 <https://github.com/ceph/ceph/pull/14326>`_, Mingxin Liu)
* osd: pg_pool_t::encode(): be compatible with Hammer <= 0.94.6 (`issue#19508 <http://tracker.ceph.com/issues/19508>`_, `pr#14392 <https://github.com/ceph/ceph/pull/14392>`_, Alexey Sheplyakov)
* rbd: [api] temporarily restrict (rbd\_)mirror_peer_add from adding multiple peers (`issue#19256 <http://tracker.ceph.com/issues/19256>`_, `pr#14664 <https://github.com/ceph/ceph/pull/14664>`_, Jason Dillaman)
* rbd: qemu crash triggered by network issues (`issue#18436 <http://tracker.ceph.com/issues/18436>`_, `pr#13244 <https://github.com/ceph/ceph/pull/13244>`_, Jason Dillaman)
* rbd: rbd --pool=x rename y z does not work (`issue#18326 <http://tracker.ceph.com/issues/18326>`_, `pr#14148 <https://github.com/ceph/ceph/pull/14148>`_, Gaurav Kumar Garg)
* rbd: systemctl stop rbdmap unmaps all rbds and not just the ones in /etc/ceph/rbdmap (`issue#18884 <http://tracker.ceph.com/issues/18884>`_, `issue#18262 <http://tracker.ceph.com/issues/18262>`_, `pr#14083 <https://github.com/ceph/ceph/pull/14083>`_, David Disseldorp, Nathan Cutler)
* rgw: "cluster [WRN] bad locator @X on object @X...." in cluster log (`issue#18980 <http://tracker.ceph.com/issues/18980>`_, `pr#14064 <https://github.com/ceph/ceph/pull/14064>`_, Casey Bodley)
* rgw: 'radosgw-admin sync status' on master zone of non-master zonegroup (`issue#18091 <http://tracker.ceph.com/issues/18091>`_, `pr#13779 <https://github.com/ceph/ceph/pull/13779>`_, Jing Wenjun)
* rgw: Change loglevel to 20 for 'System already converted' message (`issue#18919 <http://tracker.ceph.com/issues/18919>`_, `pr#13834 <https://github.com/ceph/ceph/pull/13834>`_, Vikhyat Umrao)
* rgw: Use decoded URI when verifying TempURL (`issue#18590 <http://tracker.ceph.com/issues/18590>`_, `pr#13724 <https://github.com/ceph/ceph/pull/13724>`_, Alexey Sheplyakov)
* rgw: a few cases where rgw_obj is incorrectly initialized (`issue#19096 <http://tracker.ceph.com/issues/19096>`_, `pr#13842 <https://github.com/ceph/ceph/pull/13842>`_, Yehuda Sadeh)
* rgw: add apis to support ragweed suite (`issue#19804 <http://tracker.ceph.com/issues/19804>`_, `pr#14851 <https://github.com/ceph/ceph/pull/14851>`_, Yehuda Sadeh)
* rgw: add bucket size limit check to radosgw-admin (`issue#17925 <http://tracker.ceph.com/issues/17925>`_, `pr#14787 <https://github.com/ceph/ceph/pull/14787>`_, Matt Benjamin)
* rgw: allow system users to read SLO parts (`issue#19027 <http://tracker.ceph.com/issues/19027>`_, `pr#14752 <https://github.com/ceph/ceph/pull/14752>`_, Casey Bodley)
* rgw: don't return skew time in pre-signed url (`issue#18828 <http://tracker.ceph.com/issues/18828>`_, `issue#18829 <http://tracker.ceph.com/issues/18829>`_, `pr#14605 <https://github.com/ceph/ceph/pull/14605>`_, liuchang0812)
* rgw: failure to create s3 type subuser from admin rest api (`issue#16682 <http://tracker.ceph.com/issues/16682>`_, `pr#14815 <https://github.com/ceph/ceph/pull/14815>`_, snakeAngel2015)
* rgw: fix break inside of yield in RGWFetchAllMetaCR (`issue#17655 <http://tracker.ceph.com/issues/17655>`_, `pr#14066 <https://github.com/ceph/ceph/pull/14066>`_, Casey Bodley)
* rgw: fix failed to create bucket if a non-master zonegroup has a single zone (`issue#19756 <http://tracker.ceph.com/issues/19756>`_, `pr#14766 <https://github.com/ceph/ceph/pull/14766>`_, weiqiaomiao)
* rgw: health check errors out incorrectly (`issue#19025 <http://tracker.ceph.com/issues/19025>`_, `pr#13865 <https://github.com/ceph/ceph/pull/13865>`_, Pavan Rallabhandi)
* rgw: multisite: fetch_remote_obj() gets wrong version when copying from remote (`issue#19599 <http://tracker.ceph.com/issues/19599>`_, `pr#14607 <https://github.com/ceph/ceph/pull/14607>`_, Zhang Shaowen, Casey Bodley)
* rgw: multisite: some yields in RGWMetaSyncShardCR::full_sync() resume in incremental_sync() (`issue#18076 <http://tracker.ceph.com/issues/18076>`_, `pr#13837 <https://github.com/ceph/ceph/pull/13837>`_, Casey Bodley, Abhishek Lekshmanan)
* rgw: only append zonegroups to rest params if not empty (`issue#20078 <http://tracker.ceph.com/issues/20078>`_, `pr#15312 <https://github.com/ceph/ceph/pull/15312>`_, Yehuda Sadeh, Karol Mroz)
* rgw: swift: disable revocation thread under certain circumstances (`issue#19499 <http://tracker.ceph.com/issues/19499>`_, `issue#9493 <http://tracker.ceph.com/issues/9493>`_, `pr#14789 <https://github.com/ceph/ceph/pull/14789>`_, Marcus Watts)
* rgw: the swift container acl does not support field .ref (`issue#18484 <http://tracker.ceph.com/issues/18484>`_, `pr#13833 <https://github.com/ceph/ceph/pull/13833>`_, Jing Wenjun)
* rgw: upgrade to multisite v2 fails if there is a zone without zone info (`issue#19231 <http://tracker.ceph.com/issues/19231>`_, `pr#14136 <https://github.com/ceph/ceph/pull/14136>`_, Danny Al-Gaaf, Orit Wasserman)
* rgw: use separate http_manager for read_sync_status (`issue#19236 <http://tracker.ceph.com/issues/19236>`_, `pr#14195 <https://github.com/ceph/ceph/pull/14195>`_, Casey Bodley, Shasha Lu)
* rgw: when converting region_map we need to use rgw_zone_root_pool (`issue#19195 <http://tracker.ceph.com/issues/19195>`_, `pr#14143 <https://github.com/ceph/ceph/pull/14143>`_, Orit Wasserman)
* rgw: zonegroupmap set does not work (`issue#19498 <http://tracker.ceph.com/issues/19498>`_, `issue#18725 <http://tracker.ceph.com/issues/18725>`_, `pr#14660 <https://github.com/ceph/ceph/pull/14660>`_, Orit Wasserman, Casey Bodley)
* tests: 'ceph auth import -i' overwrites caps, should alert user before overwrite (`issue#18932 <http://tracker.ceph.com/issues/18932>`_, `pr#13544 <https://github.com/ceph/ceph/pull/13544>`_, Vikhyat Umrao)
* tests: New upgrade test for #19508 (`issue#19829 <http://tracker.ceph.com/issues/19829>`_, `issue#19508 <http://tracker.ceph.com/issues/19508>`_, `pr#14930 <https://github.com/ceph/ceph/pull/14930>`_, Nathan Cutler)
* tests: [ FAILED ] TestLibRBD.ImagePollIO in upgrade:client-upgrade-kraken-distro-basic-smithi (`issue#18617 <http://tracker.ceph.com/issues/18617>`_, `pr#13107 <https://github.com/ceph/ceph/pull/13107>`_, Jason Dillaman)
* tests: [librados_test_stub] cls_cxx_map_get_XYZ methods don't return correct value (`issue#19597 <http://tracker.ceph.com/issues/19597>`_, `pr#14665 <https://github.com/ceph/ceph/pull/14665>`_, Jason Dillaman)
* tests: additional rbd-mirror test stability improvements (`issue#18935 <http://tracker.ceph.com/issues/18935>`_, `pr#14154 <https://github.com/ceph/ceph/pull/14154>`_, Jason Dillaman)
* tests: buffer overflow in test LibCephFS.DirLs (`issue#18941 <http://tracker.ceph.com/issues/18941>`_, `pr#14671 <https://github.com/ceph/ceph/pull/14671>`_, "Yan, Zheng")
* tests: clone workunit using the branch specified by task (`issue#19429 <http://tracker.ceph.com/issues/19429>`_, `pr#14371 <https://github.com/ceph/ceph/pull/14371>`_, Kefu Chai, Dan Mick)
* tests: drop upgrade/hammer-jewel-x (`issue#20574 <http://tracker.ceph.com/issues/20574>`_, `pr#15933 <https://github.com/ceph/ceph/pull/15933>`_, Nathan Cutler)
* tests: dummy suite fails in OpenStack (`issue#18259 <http://tracker.ceph.com/issues/18259>`_, `pr#14070 <https://github.com/ceph/ceph/pull/14070>`_, Nathan Cutler)
* tests: qa/workunits/rados/test-upgrade-\*: whitelist tests the right way (`issue#20575 <http://tracker.ceph.com/issues/20575>`_, `pr#15824 <https://github.com/ceph/ceph/pull/15824>`_, Kefu Chai)
* tests: test_notify.py: rbd.InvalidArgument: error updating features for image test_notify_clone2 (`issue#19692 <http://tracker.ceph.com/issues/19692>`_, `pr#14680 <https://github.com/ceph/ceph/pull/14680>`_, Jason Dillaman)
* tests: upgrade/hammer-x failing with OSD has the store locked when Thrasher runs ceph-objectstore-tool on down PG (`issue#19556 <http://tracker.ceph.com/issues/19556>`_, `pr#14416 <https://github.com/ceph/ceph/pull/14416>`_, Nathan Cutler)
* tools: brag fails to count "in" mds (`issue#19192 <http://tracker.ceph.com/issues/19192>`_, `pr#14112 <https://github.com/ceph/ceph/pull/14112>`_, Oleh Prypin, Peng Zhang)
* tools: ceph-disk does not support cluster names different than 'ceph' (`issue#17821 <http://tracker.ceph.com/issues/17821>`_, `pr#14765 <https://github.com/ceph/ceph/pull/14765>`_, Loic Dachary)
* tools: ceph-disk: Racing between partition creation and device node creation (`issue#19428 <http://tracker.ceph.com/issues/19428>`_, `pr#14329 <https://github.com/ceph/ceph/pull/14329>`_, Erwan Velu)
* tools: ceph-disk: bluestore --setgroup incorrectly set with user (`issue#18955 <http://tracker.ceph.com/issues/18955>`_, `pr#13489 <https://github.com/ceph/ceph/pull/13489>`_, craigchi)
* tools: ceph-disk: ceph-disk list reports mount error for OSD having mount options with SELinux context (`issue#17331 <http://tracker.ceph.com/issues/17331>`_, `pr#14402 <https://github.com/ceph/ceph/pull/14402>`_, Brad Hubbard)
* tools: ceph-disk: do not setup_statedir on trigger (`issue#19941 <http://tracker.ceph.com/issues/19941>`_, `pr#15504 <https://github.com/ceph/ceph/pull/15504>`_, Loic Dachary)
* tools: ceph-disk: enable directory backed OSD at boot time (`issue#19628 <http://tracker.ceph.com/issues/19628>`_, `pr#14602 <https://github.com/ceph/ceph/pull/14602>`_, Loic Dachary)
* tools: rados: RadosImport::import should return an error if Rados::connect fails (`issue#19319 <http://tracker.ceph.com/issues/19319>`_, `pr#14113 <https://github.com/ceph/ceph/pull/14113>`_, Brad Hubbard)
v10.2.7 Jewel
=============
This point release fixes several important bugs in RBD mirroring, librbd & RGW.
We recommend that all v10.2.x users upgrade.
For more detailed information, see :download:`the complete changelog <../changelog/v10.2.7.txt>`.
Notable Changes
---------------
* librbd: possible race in ExclusiveLock handle_peer_notification (`issue#19368 <http://tracker.ceph.com/issues/19368>`_, `pr#14233 <https://github.com/ceph/ceph/pull/14233>`_, Mykola Golub)
* rbd-mirror: deleting a snapshot during sync can result in read errors (`issue#18990 <http://tracker.ceph.com/issues/18990>`_, `pr#13596 <https://github.com/ceph/ceph/pull/13596>`_, Jason Dillaman)
* rgw: 'period update' does not remove short_zone_ids of deleted zones (`issue#15618 <http://tracker.ceph.com/issues/15618>`_, `pr#14140 <https://github.com/ceph/ceph/pull/14140>`_, Casey Bodley)
* rgw: DUMPABLE flag is cleared by setuid preventing coredumps (`issue#19089 <http://tracker.ceph.com/issues/19089>`_, `pr#13844 <https://github.com/ceph/ceph/pull/13844>`_, Brad Hubbard)
* rgw: fix swift cannot disable object versioning with empty X-Versions-Location (`issue#18852 <http://tracker.ceph.com/issues/18852>`_, `pr#13823 <https://github.com/ceph/ceph/pull/13823>`_, Jing Wenjun)
* rgw: librgw: RGWLibFS::setattr fails on directories (`issue#18808 <http://tracker.ceph.com/issues/18808>`_, `pr#13778 <https://github.com/ceph/ceph/pull/13778>`_, Matt Benjamin)
* rgw: make sending Content-Length in 204 and 304 controllable (`issue#16602 <http://tracker.ceph.com/issues/16602>`_, `pr#13503 <https://github.com/ceph/ceph/pull/13503>`_, Radoslaw Zarzynski, Matt Benjamin)
* rgw: multipart uploads copy part support (`issue#12790 <http://tracker.ceph.com/issues/12790>`_, `pr#13219 <https://github.com/ceph/ceph/pull/13219>`_, Yehuda Sadeh, Javier M. Mellid, Matt Benjamin)
* rgw: multisite: RGWMetaSyncShardControlCR gives up on EIO (`issue#19019 <http://tracker.ceph.com/issues/19019>`_, `pr#13867 <https://github.com/ceph/ceph/pull/13867>`_, Casey Bodley)
* rgw: radosgw/swift: clean up flush / newline behavior (`issue#18473 <http://tracker.ceph.com/issues/18473>`_, `pr#14100 <https://github.com/ceph/ceph/pull/14100>`_, Nathan Cutler, Marcus Watts, Matt Benjamin)
* rgw: radosgw/swift: clean up flush / newline behavior. (`issue#18473 <http://tracker.ceph.com/issues/18473>`_, `pr#13143 <https://github.com/ceph/ceph/pull/13143>`_, Marcus Watts, Matt Benjamin)
* rgw: rgw_fh: RGWFileHandle dtor must also cond-unlink from FHCache (`issue#19112 <http://tracker.ceph.com/issues/19112>`_, `pr#14231 <https://github.com/ceph/ceph/pull/14231>`_, Matt Benjamin)
* rgw: rgw_file: avoid interning .. in FHCache table and don't ref for them (`issue#19036 <http://tracker.ceph.com/issues/19036>`_, `pr#13848 <https://github.com/ceph/ceph/pull/13848>`_, Matt Benjamin)
* rgw: rgw_file: interned RGWFileHandle objects need parent refs (`issue#18650 <http://tracker.ceph.com/issues/18650>`_, `pr#13583 <https://github.com/ceph/ceph/pull/13583>`_, Matt Benjamin)
* rgw: rgw_file: restore (corrected) fix for dir partial match (return of FLAG_EXACT_MATCH) (`issue#19060 <http://tracker.ceph.com/issues/19060>`_, `issue#18992 <http://tracker.ceph.com/issues/18992>`_, `issue#19059 <http://tracker.ceph.com/issues/19059>`_, `pr#13858 <https://github.com/ceph/ceph/pull/13858>`_, Matt Benjamin)
* rgw: rgw_file: FHCache residence check should be exhaustive (`issue#19111 <http://tracker.ceph.com/issues/19111>`_, `pr#14169 <https://github.com/ceph/ceph/pull/14169>`_, Matt Benjamin)
* rgw: rgw_file: ensure valid_s3_object_name for directories, too (`issue#19066 <http://tracker.ceph.com/issues/19066>`_, `pr#13717 <https://github.com/ceph/ceph/pull/13717>`_, Matt Benjamin)
* build/ops: ceph daemons DUMPABLE flag is cleared by setuid preventing coredumps (`issue#17650 <http://tracker.ceph.com/issues/17650>`_, `pr#11736 <http://github.com/ceph/ceph/pull/11736>`_, Patrick Donnelly)
* build/ops: fixed compilation error when --with-radowsgw=no (`issue#18512 <http://tracker.ceph.com/issues/18512>`_, `pr#12729 <http://github.com/ceph/ceph/pull/12729>`_, Pan Liu)
* build/ops: fixed the issue when --disable-server, compilation fails. (`issue#18120 <http://tracker.ceph.com/issues/18120>`_, `pr#12239 <http://github.com/ceph/ceph/pull/12239>`_, Pan Liu)
* build/ops: install-deps.sh based on /etc/os-release (`issue#18466 <http://tracker.ceph.com/issues/18466>`_, `issue#18198 <http://tracker.ceph.com/issues/18198>`_, `pr#12405 <http://github.com/ceph/ceph/pull/12405>`_, Jan Fajerski, Nitin A Kamble, Nathan Cutler)
* build/ops: Remove the runtime dependency on lsb_release (`issue#17425 <http://tracker.ceph.com/issues/17425>`_, `pr#11875 <http://github.com/ceph/ceph/pull/11875>`_, John Coyle, Brad Hubbard)
* build/ops: rpm: /etc/ceph/rbdmap is packaged with executable access rights (`issue#17395 <http://tracker.ceph.com/issues/17395>`_, `pr#11855 <http://github.com/ceph/ceph/pull/11855>`_, Ken Dreyer)
* build/ops: selinux: Allow ceph to manage tmp files (`issue#17436 <http://tracker.ceph.com/issues/17436>`_, `pr#13048 <http://github.com/ceph/ceph/pull/13048>`_, Boris Ranto)
* build/ops: systemd: Restart Mon after 10s in case of failure (`issue#18635 <http://tracker.ceph.com/issues/18635>`_, `pr#13058 <http://github.com/ceph/ceph/pull/13058>`_, Wido den Hollander)
* build/ops: systemd restarts Ceph Mon to quickly after failing to start (`issue#18635 <http://tracker.ceph.com/issues/18635>`_, `pr#13184 <http://github.com/ceph/ceph/pull/13184>`_, Wido den Hollander)
* cephfs: fuse client crash when adding a new osd (`issue#17270 <http://tracker.ceph.com/issues/17270>`_, `pr#11860 <http://github.com/ceph/ceph/pull/11860>`_, John Spray)
* cli: ceph-disk: convert none str to str before printing it (`issue#18371 <http://tracker.ceph.com/issues/18371>`_, `pr#13187 <http://github.com/ceph/ceph/pull/13187>`_, Kefu Chai)
* client: Fix lookup of "/.." in jewel (`issue#18408 <http://tracker.ceph.com/issues/18408>`_, `pr#12766 <http://github.com/ceph/ceph/pull/12766>`_, Jeff Layton)
* client: fix stale entries in command table (`issue#17974 <http://tracker.ceph.com/issues/17974>`_, `pr#12137 <http://github.com/ceph/ceph/pull/12137>`_, John Spray)
* client: populate metadata during mount (`issue#18361 <http://tracker.ceph.com/issues/18361>`_, `pr#13085 <http://github.com/ceph/ceph/pull/13085>`_, John Spray)
* cli: implement functionality for adding, editing and removing omap values with binary keys (`issue#18123 <http://tracker.ceph.com/issues/18123>`_, `pr#12755 <http://github.com/ceph/ceph/pull/12755>`_, Jason Dillaman)
* doc: Python Swift client commands in Quick Developer Guide don't match configuration in vstart.sh (`issue#17746 <http://tracker.ceph.com/issues/17746>`_, `pr#13043 <http://github.com/ceph/ceph/pull/13043>`_, Ronak Jain)
* librbd: allow to open an image without opening parent image (`issue#18325 <http://tracker.ceph.com/issues/18325>`_, `pr#13130 <http://github.com/ceph/ceph/pull/13130>`_, Ricardo Dias)
* librbd: metadata_set API operation should not change global config setting (`issue#18465 <http://tracker.ceph.com/issues/18465>`_, `pr#13168 <http://github.com/ceph/ceph/pull/13168>`_, Mykola Golub)
* librbd: new API method to force break a peer's exclusive lock (`issue#15632 <http://tracker.ceph.com/issues/15632>`_, `issue#16773 <http://tracker.ceph.com/issues/16773>`_, `issue#17188 <http://tracker.ceph.com/issues/17188>`_, `issue#16988 <http://tracker.ceph.com/issues/16988>`_, `issue#17210 <http://tracker.ceph.com/issues/17210>`_, `issue#17251 <http://tracker.ceph.com/issues/17251>`_, `issue#18429 <http://tracker.ceph.com/issues/18429>`_, `issue#17227 <http://tracker.ceph.com/issues/17227>`_, `issue#18327 <http://tracker.ceph.com/issues/18327>`_, `issue#17015 <http://tracker.ceph.com/issues/17015>`_, `pr#12890 <http://github.com/ceph/ceph/pull/12890>`_, Danny Al-Gaaf, Mykola Golub, Jason Dillaman)
* librbd: properly order concurrent updates to the object map (`issue#16176 <http://tracker.ceph.com/issues/16176>`_, `pr#12909 <http://github.com/ceph/ceph/pull/12909>`_, Jason Dillaman)
* librbd: restore journal access when force disabling mirroring (`issue#17588 <http://tracker.ceph.com/issues/17588>`_, `pr#11916 <http://github.com/ceph/ceph/pull/11916>`_, Mykola Golub)
* mds: Cannot create deep directories when caps contain path=/somepath (`issue#17858 <http://tracker.ceph.com/issues/17858>`_, `pr#12154 <http://github.com/ceph/ceph/pull/12154>`_, Patrick Donnelly)
* mds: cephfs test failures (ceph.com/qa is broken, should be download.ceph.com/qa) (`issue#18574 <http://tracker.ceph.com/issues/18574>`_, `pr#13023 <http://github.com/ceph/ceph/pull/13023>`_, John Spray)
* mds: Cleanly reject session evict command when in replay (`issue#17801 <http://tracker.ceph.com/issues/17801>`_, `pr#12153 <http://github.com/ceph/ceph/pull/12153>`_, Yan, Zheng)
* mds: client segfault on ceph_rmdir path / (`issue#9935 <http://tracker.ceph.com/issues/9935>`_, `pr#13029 <http://github.com/ceph/ceph/pull/13029>`_, Michal Jarzabek)
* mds: Clients without pool-changing caps shouldn't be allowed to change pool_namespace (`issue#17798 <http://tracker.ceph.com/issues/17798>`_, `pr#12155 <http://github.com/ceph/ceph/pull/12155>`_, John Spray)
* mds: Decode errors on backtrace will crash MDS (`issue#18311 <http://tracker.ceph.com/issues/18311>`_, `pr#12836 <http://github.com/ceph/ceph/pull/12836>`_, Nathan Cutler, John Spray)
* mds: false failing to respond to cache pressure warning (`issue#17611 <http://tracker.ceph.com/issues/17611>`_, `pr#11861 <http://github.com/ceph/ceph/pull/11861>`_, Yan, Zheng)
* mds: finish clientreplay requests before requesting active state (`issue#18461 <http://tracker.ceph.com/issues/18461>`_, `pr#13113 <http://github.com/ceph/ceph/pull/13113>`_, Yan, Zheng)
* mds: Fix mount root for ceph_mount users and change tarball format (`issue#18312 <http://tracker.ceph.com/issues/18312>`_, `issue#18254 <http://tracker.ceph.com/issues/18254>`_, `pr#12592 <http://github.com/ceph/ceph/pull/12592>`_, Jeff Layton)
* mds: lookup of /.. in returns -ENOENT (`issue#18408 <http://tracker.ceph.com/issues/18408>`_, `pr#12783 <http://github.com/ceph/ceph/pull/12783>`_, Jeff Layton)
* mds: mds fails to respawn if executable has changed (`issue#17531 <http://tracker.ceph.com/issues/17531>`_, `pr#11873 <http://github.com/ceph/ceph/pull/11873>`_, Patrick Donnelly)
* mds: MDS: false failing to respond to cache pressure warning (`issue#17716 <http://tracker.ceph.com/issues/17716>`_, `pr#11856 <http://github.com/ceph/ceph/pull/11856>`_, Yan, Zheng)
* mds: MDS goes damaged on blacklist (failed to read JournalPointer: -108 ((108) Cannot send after transport endpoint shutdown) (`issue#17236 <http://tracker.ceph.com/issues/17236>`_, `pr#11413 <http://github.com/ceph/ceph/pull/11413>`_, John Spray)
* mds: MDS long-time blocked ops. ceph-fuse locks up with getattr of file (`issue#17275 <http://tracker.ceph.com/issues/17275>`_, `pr#11858 <http://github.com/ceph/ceph/pull/11858>`_, Yan, Zheng)
* mds: speed up readdir by skipping unwanted dn (`issue#18519 <http://tracker.ceph.com/issues/18519>`_, `pr#12921 <http://github.com/ceph/ceph/pull/12921>`_, Xiaoxi Chen)
* mds: standby-replay daemons can sometimes miss events (`issue#17954 <http://tracker.ceph.com/issues/17954>`_, `pr#13126 <http://github.com/ceph/ceph/pull/13126>`_, John Spray)
* mon: cache tiering: base pool last_force_resend not respected (racing read got wrong version) (`issue#18366 <http://tracker.ceph.com/issues/18366>`_, `pr#13115 <http://github.com/ceph/ceph/pull/13115>`_, Sage Weil)
* mon: Error EINVAL: removing mon.a at 172.21.15.16:6789/0, there will be 1 monitors (`issue#17725 <http://tracker.ceph.com/issues/17725>`_, `pr#11999 <http://github.com/ceph/ceph/pull/11999>`_, Joao Eduardo Luis)
* mon: health does not report pgs stuck in more than one state (`issue#17515 <http://tracker.ceph.com/issues/17515>`_, `pr#11660 <http://github.com/ceph/ceph/pull/11660>`_, Sage Weil)
* mon: monitor assertion failure when deactivating mds in (invalid) fscid 0 (`issue#17518 <http://tracker.ceph.com/issues/17518>`_, `pr#11862 <http://github.com/ceph/ceph/pull/11862>`_, Patrick Donnelly)
* mon: monitor cannot start because of FAILED assert(info.state == MDSMap::STATE_STANDBY) (`issue#18166 <http://tracker.ceph.com/issues/18166>`_, `pr#13123 <http://github.com/ceph/ceph/pull/13123>`_, John Spray, Patrick Donnelly)
* mon: osd flag health message is misleading (`issue#18175 <http://tracker.ceph.com/issues/18175>`_, `pr#13117 <http://github.com/ceph/ceph/pull/13117>`_, Sage Weil)
* mon: OSDMonitor: clear jewel+ feature bits when talking to Hammer OSD (`issue#18582 <http://tracker.ceph.com/issues/18582>`_, `pr#13131 <http://github.com/ceph/ceph/pull/13131>`_, Piotr Dałek)
* mon: OSDs marked OUT wrongly after monitor failover (`issue#17719 <http://tracker.ceph.com/issues/17719>`_, `pr#11947 <http://github.com/ceph/ceph/pull/11947>`_, Dong Wu)
* mon: peon wrongly delete routed pg stats op before receive pg stats ack (`issue#18458 <http://tracker.ceph.com/issues/18458>`_, `pr#13045 <http://github.com/ceph/ceph/pull/13045>`_, Mingxin Liu)
* mon: send updated monmap to its subscribers (`issue#17558 <http://tracker.ceph.com/issues/17558>`_, `pr#11743 <http://github.com/ceph/ceph/pull/11743>`_, Kefu Chai)
* osd: Add config option to disable new scrubs during recovery (`issue#17866 <http://tracker.ceph.com/issues/17866>`_, `pr#11944 <http://github.com/ceph/ceph/pull/11944>`_, Wido den Hollander)
* osd: do not send ENXIO on misdirected op by default (`issue#18751 <http://tracker.ceph.com/issues/18751>`_, `pr#13255 <http://github.com/ceph/ceph/pull/13255>`_, Sage Weil)
* osd: FileStore: fiemap cannot be totally retrieved in xfs when the number of extents > 1364 (`issue#17610 <http://tracker.ceph.com/issues/17610>`_, `pr#11998 <http://github.com/ceph/ceph/pull/11998>`_, Kefu Chai, Ning Yao)
* osd: leveldb corruption leads to Operation not permitted not handled and assert (`issue#18037 <http://tracker.ceph.com/issues/18037>`_, `pr#12789 <http://github.com/ceph/ceph/pull/12789>`_, Nathan Cutler)
* osd: limit omap data in push op (`issue#16128 <http://tracker.ceph.com/issues/16128>`_, `pr#11991 <http://github.com/ceph/ceph/pull/11991>`_, Wanlong Gao)
* osd: osd crashes when radosgw-admin bi list --max-entries=1 command runing (`issue#17745 <http://tracker.ceph.com/issues/17745>`_, `pr#11758 <http://github.com/ceph/ceph/pull/11758>`_, weiqiaomiao)
* osd: osd_max_backfills default has changed, documentation should reflect that. (`issue#17701 <http://tracker.ceph.com/issues/17701>`_, `pr#11735 <http://github.com/ceph/ceph/pull/11735>`_, huangjun)
* osd: OSDMonitor: only reject MOSDBoot based on up_from if inst matches (`issue#17899 <http://tracker.ceph.com/issues/17899>`_, `pr#12868 <http://github.com/ceph/ceph/pull/12868>`_, Samuel Just)
* osd: osd/PG: publish PG stats when backfill-related states change (`issue#18369 <http://tracker.ceph.com/issues/18369>`_, `pr#12875 <http://github.com/ceph/ceph/pull/12875>`_, Alexey Sheplyakov, Sage Weil)
* osd: Remove extra call to reg_next_scrub() during splits (`issue#16474 <http://tracker.ceph.com/issues/16474>`_, `pr#11606 <http://github.com/ceph/ceph/pull/11606>`_, David Zafman)
* osd: Revert "Merge pull request #12978 from asheplyakov/jewel-18581" (`issue#18809 <http://tracker.ceph.com/issues/18809>`_, `pr#13280 <http://github.com/ceph/ceph/pull/13280>`_, Samuel Just)
* osd: update_log_missing does not order correctly with osd_ops (`issue#17789 <http://tracker.ceph.com/issues/17789>`_, `pr#11997 <http://github.com/ceph/ceph/pull/11997>`_, Samuel Just)
* rados: crash adding snap to purged_snaps in ReplicatedPG::WaitingOnReplicas (part 2) (`issue#15943 <http://tracker.ceph.com/issues/15943>`_, `issue#18504 <http://tracker.ceph.com/issues/18504>`_, `pr#12791 <http://github.com/ceph/ceph/pull/12791>`_, Samuel Just)
* rados: Memory leaks in object_list_begin and object_list_end (`issue#18252 <http://tracker.ceph.com/issues/18252>`_, `pr#13118 <http://github.com/ceph/ceph/pull/13118>`_, Brad Hubbard)
* rados: The request lock RPC message might be incorrectly ignored (`issue#17030 <http://tracker.ceph.com/issues/17030>`_, `pr#10865 <http://github.com/ceph/ceph/pull/10865>`_, Jason Dillaman)
* rbd: add image id block name prefix APIs (`issue#18270 <http://tracker.ceph.com/issues/18270>`_, `pr#12529 <http://github.com/ceph/ceph/pull/12529>`_, Jason Dillaman)
* rbd: add max_part and nbds_max options in rbd nbd map, in order to keep consistent with (`issue#18186 <http://tracker.ceph.com/issues/18186>`_, `pr#12426 <http://github.com/ceph/ceph/pull/12426>`_, Pan Liu)
* rbd: Attempting to remove an image w/ incompatible features results in partial removal (`issue#18315 <http://tracker.ceph.com/issues/18315>`_, `pr#13156 <http://github.com/ceph/ceph/pull/13156>`_, Dongsheng Yang)
* rbd: bench-write will crash if --io-size is 4G (`issue#18422 <http://tracker.ceph.com/issues/18422>`_, `pr#13129 <http://github.com/ceph/ceph/pull/13129>`_, Gaurav Kumar Garg)
* rbd: diff calculate can hide parent extents when examining first snapshot in clone (`issue#18068 <http://tracker.ceph.com/issues/18068>`_, `pr#12322 <http://github.com/ceph/ceph/pull/12322>`_, Jason Dillaman)
* rbd: Exclusive lock improperly initialized on read-only image when using snap_set API (`issue#17618 <http://tracker.ceph.com/issues/17618>`_, `pr#11852 <http://github.com/ceph/ceph/pull/11852>`_, Jason Dillaman)
* rbd: JournalMetadata flooding with errors when being blacklisted (`issue#18243 <http://tracker.ceph.com/issues/18243>`_, `pr#12739 <http://github.com/ceph/ceph/pull/12739>`_, Jason Dillaman)
* rbd: librbd: use proper snapshot when computing diff parent overlap (`issue#18200 <http://tracker.ceph.com/issues/18200>`_, `pr#12649 <http://github.com/ceph/ceph/pull/12649>`_, Xiaoxi Chen)
* rbd: partition func should be enabled When load nbd.ko for rbd-nbd (`issue#18115 <http://tracker.ceph.com/issues/18115>`_, `pr#12754 <http://github.com/ceph/ceph/pull/12754>`_, Pan Liu)
* rbd: [qa] crash in journal-enabled fsx run (`issue#18618 <http://tracker.ceph.com/issues/18618>`_, `pr#13128 <http://github.com/ceph/ceph/pull/13128>`_, Jason Dillaman)
* rbd: 'rbd du' of missing image does not return error (`issue#16987 <http://tracker.ceph.com/issues/16987>`_, `pr#11854 <http://github.com/ceph/ceph/pull/11854>`_, Dongsheng Yang)
* rbd: rbd-nbd: invalid error code for "failed to read nbd request" messages (`issue#18242 <http://tracker.ceph.com/issues/18242>`_, `pr#12756 <http://github.com/ceph/ceph/pull/12756>`_, Mykola Golub)
* rbd: status json format has duplicated/overwritten key (`issue#18261 <http://tracker.ceph.com/issues/18261>`_, `pr#12741 <http://github.com/ceph/ceph/pull/12741>`_, Mykola Golub)
* rbd: truncate can cause unflushed snapshot data lose (`issue#17193 <http://tracker.ceph.com/issues/17193>`_, `pr#12324 <http://github.com/ceph/ceph/pull/12324>`_, Yan, Zheng)
* : ReplicatedBackend: take read locks for clone sources during recovery (`issue#17831 <http://tracker.ceph.com/issues/17831>`_, `issue#18583 <http://tracker.ceph.com/issues/18583>`_, `pr#12978 <http://github.com/ceph/ceph/pull/12978>`_, Samuel Just)
* rgw: add option to log custom HTTP headers (rgw_log_http_headers) (`issue#18891 <http://tracker.ceph.com/issues/18891>`_, `pr#12490 <http://github.com/ceph/ceph/pull/12490>`_, Matt Benjamin)
* rgw: add suport for Swift-at-root dependent features of Swift API (`issue#18526 <http://tracker.ceph.com/issues/18526>`_, `issue#16673 <http://tracker.ceph.com/issues/16673>`_, `pr#11497 <http://github.com/ceph/ceph/pull/11497>`_, Pritha Srivastava, Radoslaw Zarzynski, Pete Zaitcev, Abhishek Lekshmanan)
* rgw: add support for the prefix parameter in account listing of Swift API (`issue#17931 <http://tracker.ceph.com/issues/17931>`_, `pr#12258 <http://github.com/ceph/ceph/pull/12258>`_, Radoslaw Zarzynski)
* rgw: Add workaround for upgrade issues for older jewel versions (`issue#17820 <http://tracker.ceph.com/issues/17820>`_, `pr#12316 <http://github.com/ceph/ceph/pull/12316>`_, Orit Wasserman)
* rgw: be aware abount tenants on cls_user_bucket -> rgw_bucket conversion (`issue#18364 <http://tracker.ceph.com/issues/18364>`_, `issue#16355 <http://tracker.ceph.com/issues/16355>`_, `pr#13276 <http://github.com/ceph/ceph/pull/13276>`_, Radoslaw Zarzynski)
* rgw: bucket resharding (`issue#17549 <http://tracker.ceph.com/issues/17549>`_, `issue#17550 <http://tracker.ceph.com/issues/17550>`_, `pr#13341 <http://github.com/ceph/ceph/pull/13341>`_, Yehuda Sadeh, Robin H. Johnson)
* rgw: disable virtual hosting of buckets when no hostnames are configured (`issue#17440 <http://tracker.ceph.com/issues/17440>`_, `issue#15975 <http://tracker.ceph.com/issues/15975>`_, `issue#17136 <http://tracker.ceph.com/issues/17136>`_, `pr#11760 <http://github.com/ceph/ceph/pull/11760>`_, Casey Bodley, Robin H. Johnson)
* rgw: do not abort when accept a CORS request with short origin (`issue#18187 <http://tracker.ceph.com/issues/18187>`_, `pr#12397 <http://github.com/ceph/ceph/pull/12397>`_, LiuYang)
* rgw: don't store empty chains in gc (`issue#17897 <http://tracker.ceph.com/issues/17897>`_, `pr#12174 <http://github.com/ceph/ceph/pull/12174>`_, Yehuda Sadeh)
* rgw:fix for deleting objects name beginning and ending with underscores of one bucket using POST method of js sdk. (`issue#17888 <http://tracker.ceph.com/issues/17888>`_, `pr#12320 <http://github.com/ceph/ceph/pull/12320>`_, Casey Bodley)
* rgw: fix put_acls for objects starting and ending with underscore (`issue#17625 <http://tracker.ceph.com/issues/17625>`_, `pr#11675 <http://github.com/ceph/ceph/pull/11675>`_, Orit Wasserman)
* rgw: fix use of marker in List::list_objects() (`issue#18331 <http://tracker.ceph.com/issues/18331>`_, `pr#13358 <http://github.com/ceph/ceph/pull/13358>`_, Yehuda Sadeh)
* rgw: for the create_bucket api, if the input creation_time is zero, we … (`issue#16597 <http://tracker.ceph.com/issues/16597>`_, `pr#11990 <http://github.com/ceph/ceph/pull/11990>`_, weiqiaomiao)
* rgw: Have a flavor of bucket deletion in radosgw-admin to bypass garbage collection (`issue#15557 <http://tracker.ceph.com/issues/15557>`_, `pr#10661 <http://github.com/ceph/ceph/pull/10661>`_, Pavan Rallabhandi)
* rgw: json encode/decode of RGWBucketInfo missing index_type field (`issue#17755 <http://tracker.ceph.com/issues/17755>`_, `pr#11759 <http://github.com/ceph/ceph/pull/11759>`_, Yehuda Sadeh)
* rgw: librgw: objects created from s3 apis are not visible from nfs mount point (`issue#18651 <http://tracker.ceph.com/issues/18651>`_, `pr#13177 <http://github.com/ceph/ceph/pull/13177>`_, Matt Benjamin)
* rgw: log name instead of id for SystemMetaObj on failure (`issue#15776 <http://tracker.ceph.com/issues/15776>`_, `pr#12622 <http://github.com/ceph/ceph/pull/12622>`_, Wido den Hollander, Abhishek Lekshmanan)
* rgw: multimds: mds entering up:replay and processing down mds aborts (`issue#17670 <http://tracker.ceph.com/issues/17670>`_, `pr#11857 <http://github.com/ceph/ceph/pull/11857>`_, Patrick Donnelly)
* rgw: multipart upload copy (`issue#12790 <http://tracker.ceph.com/issues/12790>`_, `pr#13068 <http://github.com/ceph/ceph/pull/13068>`_, Yehuda Sadeh, Javier M. Mellid, Matt Benjamin)
* rgw: multisite: after finishing full sync on a bucket, incremental sync starts over from the beginning (`issue#17661 <http://tracker.ceph.com/issues/17661>`_, `issue#17624 <http://tracker.ceph.com/issues/17624>`_, `pr#11864 <http://github.com/ceph/ceph/pull/11864>`_, Zengran Zhang, Casey Bodley)
* rgw: multisite: metadata master can get the wrong value for 'oldest_log_period' (`issue#16894 <http://tracker.ceph.com/issues/16894>`_, `pr#11868 <http://github.com/ceph/ceph/pull/11868>`_, Casey Bodley)
* rgw: multisite: race between ReadSyncStatus and InitSyncStatus leads to EIO errors (`issue#17568 <http://tracker.ceph.com/issues/17568>`_, `pr#11865 <http://github.com/ceph/ceph/pull/11865>`_, Casey Bodley)
* rgw: multisite requests failing with '400 Bad Request' with civetweb 1.8 (`issue#17822 <http://tracker.ceph.com/issues/17822>`_, `pr#12313 <http://github.com/ceph/ceph/pull/12313>`_, Casey Bodley)
* rgw: multisite: segfault after changing value of rgw_data_log_num_shards (`issue#18488 <http://tracker.ceph.com/issues/18488>`_, `pr#13180 <http://github.com/ceph/ceph/pull/13180>`_, Casey Bodley)
* rgw: multisite: sync status reports master is on a different period (`issue#18064 <http://tracker.ceph.com/issues/18064>`_, `pr#13175 <http://github.com/ceph/ceph/pull/13175>`_, Abhishek Lekshmanan)
* rgw: Realm set does not create a new period (`issue#18333 <http://tracker.ceph.com/issues/18333>`_, `pr#13182 <http://github.com/ceph/ceph/pull/13182>`_, Orit Wasserman)
* rgw: remove spurious mount entries for RGW buckets (`issue#17850 <http://tracker.ceph.com/issues/17850>`_, `pr#12045 <http://github.com/ceph/ceph/pull/12045>`_, Matt Benjamin)
* rgw: Replacing '+' with "%20" in canonical uri for s3 v4 auth. (`issue#17076 <http://tracker.ceph.com/issues/17076>`_, `pr#12542 <http://github.com/ceph/ceph/pull/12542>`_, Pritha Srivastava)
* rgw: RGW will not list Argonaut-era bucket via HTTP (but radosgw-admin works) (`issue#17372 <http://tracker.ceph.com/issues/17372>`_, `pr#11863 <http://github.com/ceph/ceph/pull/11863>`_, Yehuda Sadeh)
* rgw: TempURL properly handles accounts created with the implicit tenant (`issue#17961 <http://tracker.ceph.com/issues/17961>`_, `pr#12079 <http://github.com/ceph/ceph/pull/12079>`_, Radoslaw Zarzynski)
* rgw: the value of total_time is wrong in the result of 'radosgw-admin log show' opt (`issue#17598 <http://tracker.ceph.com/issues/17598>`_, `pr#11876 <http://github.com/ceph/ceph/pull/11876>`_, weiqiaomiao)
* rgw: Unable to commit period zonegroup change (`issue#17364 <http://tracker.ceph.com/issues/17364>`_, `pr#12315 <http://github.com/ceph/ceph/pull/12315>`_, Orit Wasserman)
* tests: assertion failure in a radosgw-admin related task (`issue#17167 <http://tracker.ceph.com/issues/17167>`_, `pr#12764 <http://github.com/ceph/ceph/pull/12764>`_, Orit Wasserman)
* tests: qa/suites/upgrade/hammer-x/stress-split: finish thrashing before final upgrade (`issue#19004 <http://tracker.ceph.com/issues/19004>`_, `pr#13222 <http://github.com/ceph/ceph/pull/13222>`_, Sage Weil)
* tests: qa/tasks/ceph_deploy.py: use dev option (`issue#18736 <http://tracker.ceph.com/issues/18736>`_, `pr#13106 <http://github.com/ceph/ceph/pull/13106>`_, Vasu Kulkarni)
* tests: qa/workunits/rbd: use more recent qemu-iotests that support Xenial (`issue#18149 <http://tracker.ceph.com/issues/18149>`_, `issue#10773 <http://tracker.ceph.com/issues/10773>`_, `pr#13103 <http://github.com/ceph/ceph/pull/13103>`_, Jason Dillaman)
* tests: SUSE yaml facets in qa/distros/all are out of date (`issue#18856 <http://tracker.ceph.com/issues/18856>`_, `issue#18846 <http://tracker.ceph.com/issues/18846>`_, `pr#13331 <http://github.com/ceph/ceph/pull/13331>`_, Nathan Cutler)
* tests: update rbd/singleton/all/formatted-output.yaml to support ceph-ci (`issue#18440 <http://tracker.ceph.com/issues/18440>`_, `pr#12822 <http://github.com/ceph/ceph/pull/12822>`_, Nathan Cutler, Venky Shankar)
* tests: update Ubuntu image url after ceph.com refactor (`issue#18542 <http://tracker.ceph.com/issues/18542>`_, `pr#12959 <http://github.com/ceph/ceph/pull/12959>`_, Jason Dillaman)
* tests: upgrade:hammer-x: install firefly only on Ubuntu 14.04 (`issue#18089 <http://tracker.ceph.com/issues/18089>`_, `pr#13153 <http://github.com/ceph/ceph/pull/13153>`_, Nathan Cutler)
* tests: use ceph-jewel branch for s3tests (`issue#18384 <http://tracker.ceph.com/issues/18384>`_, `pr#12745 <http://github.com/ceph/ceph/pull/12745>`_, Nathan Cutler)
* test: test fails due to The UNIX domain socket path (`issue#16014 <http://tracker.ceph.com/issues/16014>`_, `pr#12151 <http://github.com/ceph/ceph/pull/12151>`_, Loic Dachary)
This point release fixes several important bugs in RBD mirroring, RGW multi-site, CephFS, and RADOS.
We recommend that all v10.2.x users upgrade. Also note the following when upgrading from hammer
Upgrading from hammer
---------------------
When the last hammer OSD in a cluster containing jewel MONs is
upgraded to jewel, as of 10.2.4 the jewel MONs will issue this
warning: "all OSDs are running jewel or later but the
'require_jewel_osds' osdmap flag is not set" and change the
cluster health status to HEALTH_WARN.
This is a signal for the admin to do "ceph osd set require_jewel_osds" - by
doing this, the upgrade path is complete and no more pre-Jewel OSDs may be added
to the cluster.
Notable Changes
---------------
For more detailed information, see :download:`the complete changelog <../changelog/v10.2.4.txt>`.
* build/ops: aarch64: Compiler-based detection of crc32 extended CPU type is broken (`issue#17516 <http://tracker.ceph.com/issues/17516>`_, `pr#11492 <http://github.com/ceph/ceph/pull/11492>`_, Alexander Graf)
* build/ops: allow building RGW with LDAP disabled (`issue#17312 <http://tracker.ceph.com/issues/17312>`_, `pr#11478 <http://github.com/ceph/ceph/pull/11478>`_, Daniel Gryniewicz)
* build/ops: backport 'logrotate: Run as root/ceph' (`issue#17381 <http://tracker.ceph.com/issues/17381>`_, `pr#11201 <http://github.com/ceph/ceph/pull/11201>`_, Boris Ranto)
* build/ops: ceph installs stuff in %_udevrulesdir but does not own that directory (`issue#16949 <http://tracker.ceph.com/issues/16949>`_, `pr#10862 <http://github.com/ceph/ceph/pull/10862>`_, Nathan Cutler)
* build/ops: ceph-osd-prestart.sh fails confusingly when data directory does not exist (`issue#17091 <http://tracker.ceph.com/issues/17091>`_, `pr#10812 <http://github.com/ceph/ceph/pull/10812>`_, Nathan Cutler)
* build/ops: disable LTTng-UST in openSUSE builds (`issue#16937 <http://tracker.ceph.com/issues/16937>`_, `pr#10794 <http://github.com/ceph/ceph/pull/10794>`_, Michel Normand)
* build/ops: include more files in "make dist" tarball (`issue#17560 <http://tracker.ceph.com/issues/17560>`_, `pr#11431 <http://github.com/ceph/ceph/pull/11431>`_, Ken Dreyer)
* build/ops: incorrect value of CINIT_FLAG_DEFER_DROP_PRIVILEGES (`issue#16663 <http://tracker.ceph.com/issues/16663>`_, `pr#10278 <http://github.com/ceph/ceph/pull/10278>`_, Casey Bodley)
* common: Enable/Disable of features is allowed even the features are already enabled/disabled (`issue#16079 <http://tracker.ceph.com/issues/16079>`_, `pr#11460 <http://github.com/ceph/ceph/pull/11460>`_, Lu Shi)
* common: Proxied operations shouldn't result in error messages if replayed (`issue#16130 <http://tracker.ceph.com/issues/16130>`_, `pr#11461 <http://github.com/ceph/ceph/pull/11461>`_, Vikhyat Umrao)
* common: Request exclusive lock if owner sends -ENOTSUPP for proxied maintenance op (`issue#16171 <http://tracker.ceph.com/issues/16171>`_, `pr#10784 <http://github.com/ceph/ceph/pull/10784>`_, Jason Dillaman)
* common: msgr/async: Messenger thread long time lock hold risk (`issue#15758 <http://tracker.ceph.com/issues/15758>`_, `pr#10761 <http://github.com/ceph/ceph/pull/10761>`_, Wei Jin)
* doc: fix description for rsize and rasize (`issue#17357 <http://tracker.ceph.com/issues/17357>`_, `pr#11171 <http://github.com/ceph/ceph/pull/11171>`_, Andreas Gerstmayr)
* filestore: can get stuck in an unbounded loop during scrub (`issue#17859 <http://tracker.ceph.com/issues/17859>`_, `pr#12001 <http://github.com/ceph/ceph/pull/12001>`_, Sage Weil)
* fs: Log path as well as ino when detecting metadata damage (`issue#16973 <http://tracker.ceph.com/issues/16973>`_, `pr#11418 <http://github.com/ceph/ceph/pull/11418>`_, John Spray)
* fs: client: add missing client_lock for get_root (`issue#17197 <http://tracker.ceph.com/issues/17197>`_, `pr#10921 <http://github.com/ceph/ceph/pull/10921>`_, Patrick Donnelly)
* fs: client: fix shutdown with open inodes (`issue#16764 <http://tracker.ceph.com/issues/16764>`_, `pr#10958 <http://github.com/ceph/ceph/pull/10958>`_, John Spray)
* fs: client: nlink count is not maintained correctly (`issue#16668 <http://tracker.ceph.com/issues/16668>`_, `pr#10877 <http://github.com/ceph/ceph/pull/10877>`_, Jeff Layton)
* fs: multimds: allow_multimds not required when max_mds is set in ceph.conf at startup (`issue#17105 <http://tracker.ceph.com/issues/17105>`_, `pr#10997 <http://github.com/ceph/ceph/pull/10997>`_, Patrick Donnelly)
* librados: modify Pipe::connect() to return the error code (`issue#15308 <http://tracker.ceph.com/issues/15308>`_, `pr#11193 <http://github.com/ceph/ceph/pull/11193>`_, Vikhyat Umrao)
* librados: remove new setxattr overload to avoid breaking the C++ ABI (`issue#18058 <http://tracker.ceph.com/issues/18058>`_, `pr#12207 <http://github.com/ceph/ceph/pull/12207>`_, Josh Durgin)
* librbd: cannot disable journaling or remove non-mirrored, non-primary image (`issue#16740 <http://tracker.ceph.com/issues/16740>`_, `pr#11337 <http://github.com/ceph/ceph/pull/11337>`_, Jason Dillaman)
* librbd: discard after write can result in assertion failure (`issue#17695 <http://tracker.ceph.com/issues/17695>`_, `pr#11644 <http://github.com/ceph/ceph/pull/11644>`_, Jason Dillaman)
* librbd::Operations: update notification failed: (2) No such file or directory (`issue#17549 <http://tracker.ceph.com/issues/17549>`_, `pr#11420 <http://github.com/ceph/ceph/pull/11420>`_, Jason Dillaman)
* mds: Crash in Client::_invalidate_kernel_dcache when reconnecting during unmount (`issue#17253 <http://tracker.ceph.com/issues/17253>`_, `pr#11414 <http://github.com/ceph/ceph/pull/11414>`_, Yan, Zheng)
* mon: Display full flag in ceph status if full flag is set (`issue#15809 <http://tracker.ceph.com/issues/15809>`_, `pr#9388 <http://github.com/ceph/ceph/pull/9388>`_, Vikhyat Umrao)
* mon: Error EINVAL: removing mon.a at 172.21.15.16:6789/0, there will be 1 monitors (`issue#17725 <http://tracker.ceph.com/issues/17725>`_, `pr#12267 <http://github.com/ceph/ceph/pull/12267>`_, Joao Eduardo Luis)
* mon: OSDMonitor: only reject MOSDBoot based on up_from if inst matches (`issue#17899 <http://tracker.ceph.com/issues/17899>`_, `pr#12067 <http://github.com/ceph/ceph/pull/12067>`_, Samuel Just)
* mon: OSDMonitor: Missing nearfull flag set (`issue#17390 <http://tracker.ceph.com/issues/17390>`_, `pr#11272 <http://github.com/ceph/ceph/pull/11272>`_, Igor Podoski)
* mon: the %USED of ceph df is wrong (`issue#16933 <http://tracker.ceph.com/issues/16933>`_, `pr#10860 <http://github.com/ceph/ceph/pull/10860>`_, Kefu Chai)
* osd: condition OSDMap encoding on features (`issue#18015 <http://tracker.ceph.com/issues/18015>`_, `pr#12167 <http://github.com/ceph/ceph/pull/12167>`_, Sage Weil)
* osd: PG::_update_calc_stats wrong for CRUSH_ITEM_NONE up set items (`issue#16998 <http://tracker.ceph.com/issues/16998>`_, `pr#10883 <http://github.com/ceph/ceph/pull/10883>`_, Samuel Just)
* osd: PG::choose_acting valgrind error or ./common/hobject.h: 182: FAILED assert(!max || (\*this == hobject_t(hobject_t::get_max()))) (`issue#13967 <http://tracker.ceph.com/issues/13967>`_, `pr#10885 <http://github.com/ceph/ceph/pull/10885>`_, Tao Chang)
* osd: Potential crash during journal::Replay shut down (`issue#16433 <http://tracker.ceph.com/issues/16433>`_, `pr#10645 <http://github.com/ceph/ceph/pull/10645>`_, Jason Dillaman)
* osd: adjust scrub boundary to object without SnapSet (`issue#17470 <http://tracker.ceph.com/issues/17470>`_, `pr#11311 <http://github.com/ceph/ceph/pull/11311>`_, Samuel Just)
* osd: ceph osd df does not show summarized info correctly if one or more OSDs are out (`issue#16706 <http://tracker.ceph.com/issues/16706>`_, `pr#10759 <http://github.com/ceph/ceph/pull/10759>`_, xie xingguo)
* osd: journal: do not prematurely flag object recorder as closed (`issue#17590 <http://tracker.ceph.com/issues/17590>`_, `pr#11634 <http://github.com/ceph/ceph/pull/11634>`_, Jason Dillaman)
* rbd: Disabling pool mirror mode with registered peers results orphaned mirrored images (`issue#16984 <http://tracker.ceph.com/issues/16984>`_, `pr#10857 <http://github.com/ceph/ceph/pull/10857>`_, Jason Dillaman)
* rbd: ImageWatcher: use after free within C_UnwatchAndFlush (`issue#17289 <http://tracker.ceph.com/issues/17289>`_, `issue#17254 <http://tracker.ceph.com/issues/17254>`_, `pr#11466 <http://github.com/ceph/ceph/pull/11466>`_, Jason Dillaman)
* rbd: Prevent the creation of a clone from a non-primary mirrored image (`issue#16449 <http://tracker.ceph.com/issues/16449>`_, `pr#10650 <http://github.com/ceph/ceph/pull/10650>`_, Mykola Golub)
* rbd: RBD should restrict mirror enable/disable actions on parents/clones (`issue#16056 <http://tracker.ceph.com/issues/16056>`_, `pr#11459 <http://github.com/ceph/ceph/pull/11459>`_, zhuangzeqiang)
* rbd: bench io-size should not be larger than image size (`issue#16967 <http://tracker.ceph.com/issues/16967>`_, `pr#10796 <http://github.com/ceph/ceph/pull/10796>`_, Jason Dillaman)
* rbd: ceph 10.2.2 rbd status on image format 2 returns (2) No such file or directory (`issue#16887 <http://tracker.ceph.com/issues/16887>`_, `pr#10652 <http://github.com/ceph/ceph/pull/10652>`_, Jason Dillaman)
* rbd: helgrind: TestLibRBD.TestIOPP potential deadlock closing an image with read-ahead enabled (`issue#17198 <http://tracker.ceph.com/issues/17198>`_, `pr#11463 <http://github.com/ceph/ceph/pull/11463>`_, Jason Dillaman)
* rbd: image.stat() call in librbdpy fails sometimes (`issue#17310 <http://tracker.ceph.com/issues/17310>`_, `pr#11464 <http://github.com/ceph/ceph/pull/11464>`_, Jason Dillaman)
* rbd: krbd qa scripts and concurrent.sh test fix (`issue#17223 <http://tracker.ceph.com/issues/17223>`_, `pr#11018 <http://github.com/ceph/ceph/pull/11018>`_, Ilya Dryomov)
* rbd: rbd-nbd IO hang (`issue#16921 <http://tracker.ceph.com/issues/16921>`_, `pr#11467 <http://github.com/ceph/ceph/pull/11467>`_, Jason Dillaman)
* rbd: update_features API needs to support backwards/forward compatibility (`issue#17330 <http://tracker.ceph.com/issues/17330>`_, `pr#11462 <http://github.com/ceph/ceph/pull/11462>`_, Jason Dillaman)
* rgw: Config parameter rgw keystone make new tenants in radosgw multitenancy does not work (`issue#17293 <http://tracker.ceph.com/issues/17293>`_, `pr#11473 <http://github.com/ceph/ceph/pull/11473>`_, SirishaGuduru)
* rgw: Do not archive metadata by default (`issue#17256 <http://tracker.ceph.com/issues/17256>`_, `pr#11321 <http://github.com/ceph/ceph/pull/11321>`_, Pavan Rallabhandi, Matt Benjamin)
* rgw: ERROR: got unexpected error when trying to read object: -2 (`issue#17111 <http://tracker.ceph.com/issues/17111>`_, `pr#11472 <http://github.com/ceph/ceph/pull/11472>`_, Yang Honggang)
* rgw: Modification for TEST S3 ACCESS section in INSTALL CEPH OBJECT GATEWAY page (`issue#15603 <http://tracker.ceph.com/issues/15603>`_, `pr#11475 <http://github.com/ceph/ceph/pull/11475>`_, la-sguduru)
* rgw: RGW loses realm/period/zonegroup/zone data: period overwritten if somewhere in the cluster is still running Hammer (`issue#17371 <http://tracker.ceph.com/issues/17371>`_, `pr#11519 <http://github.com/ceph/ceph/pull/11519>`_, Orit Wasserman)
* rgw: RGWDataSyncCR fails on errors from RGWListBucketIndexesCR (`issue#17073 <http://tracker.ceph.com/issues/17073>`_, `pr#11330 <http://github.com/ceph/ceph/pull/11330>`_, Casey Bodley)
* rgw: S3 object versioning fails when applied on a non-master zone (`issue#16494 <http://tracker.ceph.com/issues/16494>`_, `pr#11367 <http://github.com/ceph/ceph/pull/11367>`_, Yehuda Sadeh)
* rgw: add orphan options to radosgw-admin --help and man page (`issue#17281 <http://tracker.ceph.com/issues/17281>`_, `issue#17280 <http://tracker.ceph.com/issues/17280>`_, `pr#11139 <http://github.com/ceph/ceph/pull/11139>`_, Ken Dreyer, Thomas Serlin)
* rgw: back off bucket sync on failures, don't store marker (`issue#16742 <http://tracker.ceph.com/issues/16742>`_, `pr#11021 <http://github.com/ceph/ceph/pull/11021>`_, Yehuda Sadeh)
* rgw: combined LDAP backports (`issue#17544 <http://tracker.ceph.com/issues/17544>`_, `issue#17185 <http://tracker.ceph.com/issues/17185>`_, `pr#11332 <http://github.com/ceph/ceph/pull/11332>`_, Harald Klein, Matt Benjamin)
* rgw: cors auto memleak (`issue#16564 <http://tracker.ceph.com/issues/16564>`_, `pr#10656 <http://github.com/ceph/ceph/pull/10656>`_, Yan Jun)
* rgw: doc: description of multipart part entity is wrong (`issue#17504 <http://tracker.ceph.com/issues/17504>`_, `pr#11342 <http://github.com/ceph/ceph/pull/11342>`_, weiqiaomiao)
* rgw: don't loop forever when reading data from 0 sized segment. (`issue#17692 <http://tracker.ceph.com/issues/17692>`_, `pr#11626 <http://github.com/ceph/ceph/pull/11626>`_, Marcus Watts)
* rgw: fix put_acls for objects starting and ending with underscore (`issue#17625 <http://tracker.ceph.com/issues/17625>`_, `pr#11669 <http://github.com/ceph/ceph/pull/11669>`_, Orit Wasserman)
* rgw: period commit return error when the current period has a zonegroup which doesn't have a master zone (`issue#17110 <http://tracker.ceph.com/issues/17110>`_, `pr#10867 <http://github.com/ceph/ceph/pull/10867>`_, weiqiaomiao)
* rgw: rgw file uses too much CPU in gc/idle thread (`issue#16976 <http://tracker.ceph.com/issues/16976>`_, `pr#10889 <http://github.com/ceph/ceph/pull/10889>`_, Matt Benjamin)
* rgw: upgrade from old multisite to new multisite fails (`issue#16751 <http://tracker.ceph.com/issues/16751>`_, `pr#10891 <http://github.com/ceph/ceph/pull/10891>`_, Orit Wasserman)
* rgw:response information is error when geting token of swift account (`issue#15195 <http://tracker.ceph.com/issues/15195>`_, `pr#11474 <http://github.com/ceph/ceph/pull/11474>`_, Qiankun Zheng)
* rgw:user email can modify to empty when it has values (`issue#13286 <http://tracker.ceph.com/issues/13286>`_, `pr#11469 <http://github.com/ceph/ceph/pull/11469>`_, Yehuda Sadeh, Weijun Duan)
* tests: test/ceph_test_msgr: do not use Message::middle for holding transient… (`issue#17365 <http://tracker.ceph.com/issues/17365>`_, `issue#17728 <http://tracker.ceph.com/issues/17728>`_, `issue#16955 <http://tracker.ceph.com/issues/16955>`_, `pr#11742 <http://github.com/ceph/ceph/pull/11742>`_, Haomai Wang, Kefu Chai, Michal Jarzabek, Sage Weil)
* tools: Missing comma in ceph-create-keys causes concatenation of arguments (`issue#17815 <http://tracker.ceph.com/issues/17815>`_, `pr#11822 <http://github.com/ceph/ceph/pull/11822>`_, Patrick Donnelly)
* tools: add a tool to rebuild mon store from OSD (`issue#17179 <http://tracker.ceph.com/issues/17179>`_, `issue#17400 <http://tracker.ceph.com/issues/17400>`_, `pr#11126 <http://github.com/ceph/ceph/pull/11126>`_, Kefu Chai, xie xingguo)
* tools: ceph-create-keys: sometimes blocks forever if mds allow is set (`issue#16255 <http://tracker.ceph.com/issues/16255>`_, `pr#11417 <http://github.com/ceph/ceph/pull/11417>`_, John Spray)
* tools: ceph-disk should timeout when a lock cannot be acquired (`issue#16580 <http://tracker.ceph.com/issues/16580>`_, `pr#10758 <http://github.com/ceph/ceph/pull/10758>`_, Loic Dachary)
* tools: ceph-disk: expected systemd unit failures are confusing (`issue#15990 <http://tracker.ceph.com/issues/15990>`_, `pr#10884 <http://github.com/ceph/ceph/pull/10884>`_, Boris Ranto)
* tools: ceph-disk: using a regular file as a journal fails (`issue#16280 <http://tracker.ceph.com/issues/16280>`_, `issue#17662 <http://tracker.ceph.com/issues/17662>`_, `pr#11657 <http://github.com/ceph/ceph/pull/11657>`_, Jayashree Candadai, Anirudha Bose, Loic Dachary, Shylesh Kumar)
* tools: ceph-objectstore-tool: add a way to split filestore directories offline (`issue#17220 <http://tracker.ceph.com/issues/17220>`_, `pr#11252 <http://github.com/ceph/ceph/pull/11252>`_, Josh Durgin)
* tools: ceph-post-file: use new ssh key (`issue#14267 <http://tracker.ceph.com/issues/14267>`_, `pr#11746 <http://github.com/ceph/ceph/pull/11746>`_, David Galloway)
v10.2.3 Jewel
=============
This point release fixes several important bugs in RBD mirroring, RGW multi-site, CephFS, and RADOS.
We recommend that all v10.2.x users upgrade.
For more detailed information, see :download:`the complete changelog <../changelog/v10.2.3.txt>`.
Notable Changes
---------------
* build/ops: 60-ceph-partuuid-workaround-rules still needed by debian jessie (udev 215-17) (`issue#16351 <http://tracker.ceph.com/issues/16351>`_, `pr#10653 <http://github.com/ceph/ceph/pull/10653>`_, runsisi, Loic Dachary)
* build/ops: ceph Resource Agent does not work with systemd (`issue#14828 <http://tracker.ceph.com/issues/14828>`_, `pr#9917 <http://github.com/ceph/ceph/pull/9917>`_, Nathan Cutler)
* build/ops: mount.ceph: move from ceph-base to ceph-common and add symlink in /sbin for SUSE (`issue#16598 <http://tracker.ceph.com/issues/16598>`_, `issue#16645 <http://tracker.ceph.com/issues/16645>`_, `pr#10357 <http://github.com/ceph/ceph/pull/10357>`_, Nathan Cutler, Dan Horák, Ricardo Dias, Kefu Chai)
* build/ops: need rocksdb commit 7ca731b12ce for ppc64le build (`issue#17092 <http://tracker.ceph.com/issues/17092>`_, `pr#10816 <http://github.com/ceph/ceph/pull/10816>`_, Nathan Cutler)
* cli: ceph command line tool chokes on ceph –w (the dash is unicode 'en dash' &ndash, copy-paste to reproduce) (`issue#12287 <http://tracker.ceph.com/issues/12287>`_, `pr#10420 <http://github.com/ceph/ceph/pull/10420>`_, Oleh Prypin, Kefu Chai)
* common: global-init: fixup chown of the run directory along with log and asok files (`issue#15607 <http://tracker.ceph.com/issues/15607>`_, `pr#8754 <http://github.com/ceph/ceph/pull/8754>`_, Karol Mroz)
* fs: ceph-fuse: link to libtcmalloc or jemalloc (`issue#16655 <http://tracker.ceph.com/issues/16655>`_, `pr#10303 <http://github.com/ceph/ceph/pull/10303>`_, Yan, Zheng)
* fs: client: crash in unmount when fuse_use_invalidate_cb is enabled (`issue#16137 <http://tracker.ceph.com/issues/16137>`_, `pr#10106 <http://github.com/ceph/ceph/pull/10106>`_, Yan, Zheng)
* fs: fix mdsmap print_summary with standby replays (`issue#15705 <http://tracker.ceph.com/issues/15705>`_, `pr#9547 <http://github.com/ceph/ceph/pull/9547>`_, John Spray)
* fs: fuse mounted file systems fails SAMBA CTDB ping_pong rw test with v9.0.2 (`issue#12653 <http://tracker.ceph.com/issues/12653>`_, `issue#15634 <http://tracker.ceph.com/issues/15634>`_, `pr#10108 <http://github.com/ceph/ceph/pull/10108>`_, Yan, Zheng)
* librados: Missing export for rados_aio_get_version in src/include/rados/librados.h (`issue#15535 <http://tracker.ceph.com/issues/15535>`_, `pr#9574 <http://github.com/ceph/ceph/pull/9574>`_, Jim Wright)
* librados: osd: bad flags can crash the osd (`issue#16012 <http://tracker.ceph.com/issues/16012>`_, `pr#9997 <http://github.com/ceph/ceph/pull/9997>`_, Sage Weil)
* librbd: Close journal and object map before flagging exclusive lock as released (`issue#16450 <http://tracker.ceph.com/issues/16450>`_, `pr#10053 <http://github.com/ceph/ceph/pull/10053>`_, Jason Dillaman)
* librbd: Crash when utilizing advisory locking API functions (`issue#16364 <http://tracker.ceph.com/issues/16364>`_, `pr#10051 <http://github.com/ceph/ceph/pull/10051>`_, Jason Dillaman)
* librbd: ExclusiveLock object leaked when switching to snapshot (`issue#16446 <http://tracker.ceph.com/issues/16446>`_, `pr#10054 <http://github.com/ceph/ceph/pull/10054>`_, Jason Dillaman)
* librbd: Image removal doesn't necessarily clean up all rbd_mirroring entries (`issue#16471 <http://tracker.ceph.com/issues/16471>`_, `pr#10009 <http://github.com/ceph/ceph/pull/10009>`_, Jason Dillaman)
* librbd: Object map/fast-diff invalidated if journal replays the same snap remove event (`issue#16350 <http://tracker.ceph.com/issues/16350>`_, `pr#10010 <http://github.com/ceph/ceph/pull/10010>`_, Jason Dillaman)
* librbd: Timeout sending mirroring notification shouldn't result in failure (`issue#16470 <http://tracker.ceph.com/issues/16470>`_, `pr#10052 <http://github.com/ceph/ceph/pull/10052>`_, Jason Dillaman)
* librbd: Whitelist EBUSY error from snap unprotect for journal replay (`issue#16445 <http://tracker.ceph.com/issues/16445>`_, `pr#10055 <http://github.com/ceph/ceph/pull/10055>`_, Jason Dillaman)
* librbd: cancel all tasks should wait until finisher is done (`issue#16517 <http://tracker.ceph.com/issues/16517>`_, `pr#9752 <http://github.com/ceph/ceph/pull/9752>`_, Haomai Wang)
* librbd: delay acquiring lock if image watch has failed (`issue#16923 <http://tracker.ceph.com/issues/16923>`_, `pr#10827 <http://github.com/ceph/ceph/pull/10827>`_, Jason Dillaman)
* librbd: fix missing return statement if failed to get mirror image state (`issue#16600 <http://tracker.ceph.com/issues/16600>`_, `pr#10144 <http://github.com/ceph/ceph/pull/10144>`_, runsisi)
* librbd: flag image as updated after proxying maintenance op (`issue#16404 <http://tracker.ceph.com/issues/16404>`_, `pr#9883 <http://github.com/ceph/ceph/pull/9883>`_, Jason Dillaman)
* librbd: mkfs.xfs slow performance with discards and object map (`issue#16707 <http://tracker.ceph.com/issues/16707>`_, `issue#16689 <http://tracker.ceph.com/issues/16689>`_, `pr#10649 <http://github.com/ceph/ceph/pull/10649>`_, Jason Dillaman)
* librbd: potential use after free on refresh error (`issue#16519 <http://tracker.ceph.com/issues/16519>`_, `pr#9952 <http://github.com/ceph/ceph/pull/9952>`_, Mykola Golub)
* librbd: rbd-nbd does not properly handle resize notifications (`issue#15715 <http://tracker.ceph.com/issues/15715>`_, `pr#10679 <http://github.com/ceph/ceph/pull/10679>`_, Mykola Golub)
* librbd: the option 'rbd_cache_writethrough_until_flush=true' dosn't work (`issue#16740 <http://tracker.ceph.com/issues/16740>`_, `issue#16386 <http://tracker.ceph.com/issues/16386>`_, `issue#16708 <http://tracker.ceph.com/issues/16708>`_, `issue#16654 <http://tracker.ceph.com/issues/16654>`_, `issue#16478 <http://tracker.ceph.com/issues/16478>`_, `pr#10797 <http://github.com/ceph/ceph/pull/10797>`_, Mykola Golub, xinxin shu, Xiaowei Chen, Jason Dillaman)
* mds: tell command blocks forever with async messenger (TestVolumeClient.test_evict_client failure) (`issue#16288 <http://tracker.ceph.com/issues/16288>`_, `pr#10501 <http://github.com/ceph/ceph/pull/10501>`_, Douglas Fuller)
* mds: Confusing MDS log message when shut down with stalled journaler reads (`issue#15689 <http://tracker.ceph.com/issues/15689>`_, `pr#9557 <http://github.com/ceph/ceph/pull/9557>`_, John Spray)
* mds: Deadlock on shutdown active rank while busy with metadata IO (`issue#16042 <http://tracker.ceph.com/issues/16042>`_, `pr#10502 <http://github.com/ceph/ceph/pull/10502>`_, Patrick Donnelly)
* mds: Failing file operations on kernel based cephfs mount point leaves unaccessible file behind on hammer 0.94.7 (`issue#16013 <http://tracker.ceph.com/issues/16013>`_, `pr#10199 <http://github.com/ceph/ceph/pull/10199>`_, Yan, Zheng)
* mds: Fix shutting down mds timed-out due to deadlock (`issue#16396 <http://tracker.ceph.com/issues/16396>`_, `pr#10500 <http://github.com/ceph/ceph/pull/10500>`_, Zhi Zhang)
* mds: enforce a dirfrag limit on entries (`issue#16164 <http://tracker.ceph.com/issues/16164>`_, `pr#10104 <http://github.com/ceph/ceph/pull/10104>`_, Patrick Donnelly)
* mds: wrongly treat symlink inode as normal file/dir when symlink inode is stale on kcephfs (`issue#15702 <http://tracker.ceph.com/issues/15702>`_, `pr#9405 <http://github.com/ceph/ceph/pull/9405>`_, Zhi Zhang)
* mon: "mon metadata" fails when only one monitor exists (`issue#15866 <http://tracker.ceph.com/issues/15866>`_, `pr#10654 <http://github.com/ceph/ceph/pull/10654>`_, John Spray, Kefu Chai)
* mon: Monitor: validate prefix on handle_command() (`issue#16297 <http://tracker.ceph.com/issues/16297>`_, `pr#10036 <http://github.com/ceph/ceph/pull/10036>`_, You Ji)
* mon: OSDMonitor: drop pg temps from not the current primary (`issue#16127 <http://tracker.ceph.com/issues/16127>`_, `pr#9998 <http://github.com/ceph/ceph/pull/9998>`_, Samuel Just)
* mon: prepare_pgtemp needs to only update up_thru if newer than the existing one (`issue#16185 <http://tracker.ceph.com/issues/16185>`_, `pr#10001 <http://github.com/ceph/ceph/pull/10001>`_, Samuel Just)
* msgr: backports of all asyncmsgr fixes to jewel (`issue#15503 <http://tracker.ceph.com/issues/15503>`_, `issue#15372 <http://tracker.ceph.com/issues/15372>`_, `pr#9633 <http://github.com/ceph/ceph/pull/9633>`_, Yan Jun, Haomai Wang, Piotr Dałek)
* osd: FileStore: umount hang because sync thread doesn't exit (`issue#15695 <http://tracker.ceph.com/issues/15695>`_, `pr#9105 <http://github.com/ceph/ceph/pull/9105>`_, Kefu Chai)
* osd: Fixes for list-inconsistent-* (`issue#15766 <http://tracker.ceph.com/issues/15766>`_, `issue#16192 <http://tracker.ceph.com/issues/16192>`_, `issue#15719 <http://tracker.ceph.com/issues/15719>`_, `pr#9565 <http://github.com/ceph/ceph/pull/9565>`_, David Zafman)
* osd: New pools have bogus stuck inactive/unclean HEALTH_ERR messages until they are first active and clean (`issue#14952 <http://tracker.ceph.com/issues/14952>`_, `pr#10007 <http://github.com/ceph/ceph/pull/10007>`_, Sage Weil)
* osd: OSD crash with Hammer to Jewel Upgrade: void FileStore::init_temp_collections() (`issue#16672 <http://tracker.ceph.com/issues/16672>`_, `pr#10561 <http://github.com/ceph/ceph/pull/10561>`_, David Zafman)
* osd: OSD failed to subscribe skipped osdmaps after ceph osd pause (`issue#17023 <http://tracker.ceph.com/issues/17023>`_, `pr#10804 <http://github.com/ceph/ceph/pull/10804>`_, Kefu Chai)
* osd: crash adding snap to purged_snaps in ReplicatedPG::WaitingOnReplicas (`issue#15943 <http://tracker.ceph.com/issues/15943>`_, `pr#9575 <http://github.com/ceph/ceph/pull/9575>`_, Samuel Just)
* osd: partprobe intermittent issues during ceph-disk prepare (`issue#15176 <http://tracker.ceph.com/issues/15176>`_, `pr#10497 <http://github.com/ceph/ceph/pull/10497>`_, Marius Vollmer, Loic Dachary)
* osd: sparse_read on ec pool should return extends with correct offset (`issue#16138 <http://tracker.ceph.com/issues/16138>`_, `pr#10006 <http://github.com/ceph/ceph/pull/10006>`_, kofiliu)
* rbd: ImageReplayer::is_replaying does not include flush state (`issue#16970 <http://tracker.ceph.com/issues/16970>`_, `pr#10790 <http://github.com/ceph/ceph/pull/10790>`_, Jason Dillaman)
* rbd: Journal duplicate op detection can cause lockdep error (`issue#16363 <http://tracker.ceph.com/issues/16363>`_, `pr#10044 <http://github.com/ceph/ceph/pull/10044>`_, Jason Dillaman)
* rbd: Journal needs to handle duplicate maintenance op tids (`issue#16362 <http://tracker.ceph.com/issues/16362>`_, `pr#10045 <http://github.com/ceph/ceph/pull/10045>`_, Jason Dillaman)
* rbd: Unable to disable journaling feature if in unexpected mirror state (`issue#16348 <http://tracker.ceph.com/issues/16348>`_, `pr#10042 <http://github.com/ceph/ceph/pull/10042>`_, Jason Dillaman)
* rbd: bashism in src/rbdmap (`issue#16608 <http://tracker.ceph.com/issues/16608>`_, `pr#10786 <http://github.com/ceph/ceph/pull/10786>`_, Jason Dillaman)
* rbd: doc: format 2 now is the default image format (`issue#17026 <http://tracker.ceph.com/issues/17026>`_, `pr#10732 <http://github.com/ceph/ceph/pull/10732>`_, Chengwei Yang)
* rbd: hen journaling is enabled, a flush request shouldn't flush the cache (`issue#15761 <http://tracker.ceph.com/issues/15761>`_, `pr#10041 <http://github.com/ceph/ceph/pull/10041>`_, Yuan Zhou)
* rbd: possible race condition during journal transition from replay to ready (`issue#16198 <http://tracker.ceph.com/issues/16198>`_, `pr#10047 <http://github.com/ceph/ceph/pull/10047>`_, Jason Dillaman)
* rbd: rbd-mirror: ensure replay status formatter has completed before stopping replay (`issue#16352 <http://tracker.ceph.com/issues/16352>`_, `pr#10043 <http://github.com/ceph/ceph/pull/10043>`_, Jason Dillaman)
* rbd: rbd-mirror: include local pool id in resync throttle unique key (`issue#16536 <http://tracker.ceph.com/issues/16536>`_, `issue#15239 <http://tracker.ceph.com/issues/15239>`_, `issue#16488 <http://tracker.ceph.com/issues/16488>`_, `issue#16491 <http://tracker.ceph.com/issues/16491>`_, `issue#16329 <http://tracker.ceph.com/issues/16329>`_, `issue#15108 <http://tracker.ceph.com/issues/15108>`_, `issue#15670 <http://tracker.ceph.com/issues/15670>`_, `pr#10678 <http://github.com/ceph/ceph/pull/10678>`_, Ricardo Dias, Jason Dillaman)
* rbd: rbd-mirror: potential race condition accessing local image journal (`issue#16230 <http://tracker.ceph.com/issues/16230>`_, `pr#10046 <http://github.com/ceph/ceph/pull/10046>`_, Jason Dillaman)
* rgw: A query on a static large object fails with 404 error (`issue#16015 <http://tracker.ceph.com/issues/16015>`_, `pr#9544 <http://github.com/ceph/ceph/pull/9544>`_, Radoslaw Zarzynski)
* rgw: Add zone rename to radosgw_admin (`issue#16934 <http://tracker.ceph.com/issues/16934>`_, `pr#10663 <http://github.com/ceph/ceph/pull/10663>`_, Shilpa Jagannath)
* rgw: Bucket index shards orphaned after bucket delete (`issue#16412 <http://tracker.ceph.com/issues/16412>`_, `pr#10525 <http://github.com/ceph/ceph/pull/10525>`_, Orit Wasserman)
* rgw: Bug when using port 443s in rgw. (`issue#16548 <http://tracker.ceph.com/issues/16548>`_, `pr#10664 <http://github.com/ceph/ceph/pull/10664>`_, Pritha Srivastava)
* rgw: Fallback to Host header for bucket name. (`issue#15975 <http://tracker.ceph.com/issues/15975>`_, `pr#10693 <http://github.com/ceph/ceph/pull/10693>`_, Robin H. Johnson)
* rgw: Fix civetweb IPv6 (`issue#16928 <http://tracker.ceph.com/issues/16928>`_, `pr#10580 <http://github.com/ceph/ceph/pull/10580>`_, Robin H. Johnson)
* rgw: Increase log level for messages occuring while running rgw admin command (`issue#16935 <http://tracker.ceph.com/issues/16935>`_, `pr#10765 <http://github.com/ceph/ceph/pull/10765>`_, Shilpa Jagannath)
* rgw: No Last-Modified, Content-Size and X-Object-Manifest headers if no segments in DLO manifest (`issue#15812 <http://tracker.ceph.com/issues/15812>`_, `pr#9265 <http://github.com/ceph/ceph/pull/9265>`_, Radoslaw Zarzynski)
* rgw: RGWPeriodPuller tries to pull from itself (`issue#16939 <http://tracker.ceph.com/issues/16939>`_, `pr#10764 <http://github.com/ceph/ceph/pull/10764>`_, Casey Bodley)
* rgw: Set Access-Control-Allow-Origin to a Asterisk if allowed in a rule (`issue#15348 <http://tracker.ceph.com/issues/15348>`_, `pr#9453 <http://github.com/ceph/ceph/pull/9453>`_, Wido den Hollander)
* rgw: Swift API returns double space usage and objects of account metadata (`issue#16188 <http://tracker.ceph.com/issues/16188>`_, `pr#10148 <http://github.com/ceph/ceph/pull/10148>`_, Albert Tu)
* rgw: account/container metadata not actually present in a request are deleted during POST through Swift API (`issue#15977 <http://tracker.ceph.com/issues/15977>`_, `issue#15779 <http://tracker.ceph.com/issues/15779>`_, `pr#9542 <http://github.com/ceph/ceph/pull/9542>`_, Radoslaw Zarzynski)
* rgw: add socket backlog setting for via ceph.conf (`issue#16406 <http://tracker.ceph.com/issues/16406>`_, `pr#10216 <http://github.com/ceph/ceph/pull/10216>`_, Feng Guo)
* rgw: aws4: add STREAMING-AWS4-HMAC-SHA256-PAYLOAD support (`issue#16146 <http://tracker.ceph.com/issues/16146>`_, `pr#10167 <http://github.com/ceph/ceph/pull/10167>`_, Radoslaw Zarzynski, Javier M. Mellid)
* rgw: backport merge of static sites fixes (`issue#15555 <http://tracker.ceph.com/issues/15555>`_, `issue#15532 <http://tracker.ceph.com/issues/15532>`_, `issue#15531 <http://tracker.ceph.com/issues/15531>`_, `pr#9568 <http://github.com/ceph/ceph/pull/9568>`_, Robin H. Johnson)
* rgw: can set negative max_buckets on RGWUserInfo (`issue#14534 <http://tracker.ceph.com/issues/14534>`_, `pr#10655 <http://github.com/ceph/ceph/pull/10655>`_, Yehuda Sadeh)
* rgw: cleanup radosgw-admin temp command as it was deprecated (`issue#16023 <http://tracker.ceph.com/issues/16023>`_, `pr#9390 <http://github.com/ceph/ceph/pull/9390>`_, Vikhyat Umrao)
* rgw: comparing return code to ERR_NOT_MODIFIED in rgw_rest_s3.cc (needs minus sign) (`issue#16327 <http://tracker.ceph.com/issues/16327>`_, `pr#9790 <http://github.com/ceph/ceph/pull/9790>`_, Nathan Cutler)
* rgw: data sync stops after getting error in all data log sync shards (`issue#16530 <http://tracker.ceph.com/issues/16530>`_, `pr#10073 <http://github.com/ceph/ceph/pull/10073>`_, Yehuda Sadeh)
* rgw: default zone and zonegroup cannot be added to a realm (`issue#16839 <http://tracker.ceph.com/issues/16839>`_, `pr#10658 <http://github.com/ceph/ceph/pull/10658>`_, Casey Bodley)
* rgw: document multi tenancy (`issue#16635 <http://tracker.ceph.com/issues/16635>`_, `pr#10217 <http://github.com/ceph/ceph/pull/10217>`_, Pete Zaitcev)
* rgw: don't unregister request if request is not connected to manager (`issue#15911 <http://tracker.ceph.com/issues/15911>`_, `pr#9242 <http://github.com/ceph/ceph/pull/9242>`_, Yehuda Sadeh)
* rgw: failed to create bucket after upgrade from hammer to jewel (`issue#16627 <http://tracker.ceph.com/issues/16627>`_, `pr#10524 <http://github.com/ceph/ceph/pull/10524>`_, Orit Wasserman)
* rgw: multisite: preserve zone's extra pool (`issue#16712 <http://tracker.ceph.com/issues/16712>`_, `pr#10537 <http://github.com/ceph/ceph/pull/10537>`_, Abhishek Lekshmanan)
* rgw: object expirer's hints might be trimmed without processing in some circumstances (`issue#16705 <http://tracker.ceph.com/issues/16705>`_, `issue#16684 <http://tracker.ceph.com/issues/16684>`_, `pr#10763 <http://github.com/ceph/ceph/pull/10763>`_, Radoslaw Zarzynski)
* rgw: radosgw-admin failure for user create after upgrade from hammer to jewel (`issue#15937 <http://tracker.ceph.com/issues/15937>`_, `pr#9294 <http://github.com/ceph/ceph/pull/9294>`_, Orit Wasserman, Abhishek Lekshmanan)
* rgw: segmentation fault on error_repo in data sync (`issue#16603 <http://tracker.ceph.com/issues/16603>`_, `pr#10523 <http://github.com/ceph/ceph/pull/10523>`_, Casey Bodley)
* rgw: selinux denials in RGW (`issue#16126 <http://tracker.ceph.com/issues/16126>`_, `pr#10519 <http://github.com/ceph/ceph/pull/10519>`_, Boris Ranto)
* rgw: support size suffixes for --max-size in radosgw-admin command (`issue#16004 <http://tracker.ceph.com/issues/16004>`_, `pr#9743 <http://github.com/ceph/ceph/pull/9743>`_, Vikhyat Umrao)
* rgw: updating CORS/ACLs might not work in some circumstances (`issue#15976 <http://tracker.ceph.com/issues/15976>`_, `pr#9543 <http://github.com/ceph/ceph/pull/9543>`_, Radoslaw Zarzynski)
* rgw: use zone endpoints instead of zonegroup endpoints (`issue#16834 <http://tracker.ceph.com/issues/16834>`_, `pr#10659 <http://github.com/ceph/ceph/pull/10659>`_, Casey Bodley)
* tests: improve rbd-mirror test case coverage (`issue#16197 <http://tracker.ceph.com/issues/16197>`_, `pr#9631 <http://github.com/ceph/ceph/pull/9631>`_, Mykola Golub, Jason Dillaman)
* tools: ceph-disk: Accept bcache devices as data disks (`issue#13278 <http://tracker.ceph.com/issues/13278>`_, `pr#8497 <http://github.com/ceph/ceph/pull/8497>`_, Peter Sabaini)
* tools: rados: Add cleanup message with time to rados bench output (`issue#15704 <http://tracker.ceph.com/issues/15704>`_, `pr#9740 <http://github.com/ceph/ceph/pull/9740>`_, Vikhyat Umrao)
* tools: src/script/subman fails with KeyError: 'nband' (`issue#16961 <http://tracker.ceph.com/issues/16961>`_, `pr#10625 <http://github.com/ceph/ceph/pull/10625>`_, Loic Dachary, Ali Maredia)
v10.2.2 Jewel
=============
This point release fixes several important bugs in RBD mirroring, RGW multi-site, CephFS, and RADOS.
We recommend that all v10.2.x users upgrade.
For more detailed information, see :download:`the complete changelog <../changelog/v10.2.2.txt>`.
Notable Changes
---------------
* ceph: cli: exception when pool name has non-ascii characters (`issue#15913 <http://tracker.ceph.com/issues/15913>`_, `pr#9320 <http://github.com/ceph/ceph/pull/9320>`_, Ricardo Dias)
* cephfs: backports needed for Manila (`issue#15599 <http://tracker.ceph.com/issues/15599>`_, `issue#15417 <http://tracker.ceph.com/issues/15417>`_, `issue#15045 <http://tracker.ceph.com/issues/15045>`_, `pr#9430 <http://github.com/ceph/ceph/pull/9430>`_, John Spray, Ramana Raja, Xiaoxi Chen)
* ceph.spec.in: drop support for RHEL<7 and SUSE<1210 in jewel and above (`issue#15725 <http://tracker.ceph.com/issues/15725>`_, `issue#15627 <http://tracker.ceph.com/issues/15627>`_, `issue#13445 <http://tracker.ceph.com/issues/13445>`_, `issue#15822 <http://tracker.ceph.com/issues/15822>`_, `issue#15472 <http://tracker.ceph.com/issues/15472>`_, `issue#15987 <http://tracker.ceph.com/issues/15987>`_, `issue#15516 <http://tracker.ceph.com/issues/15516>`_, `issue#15549 <http://tracker.ceph.com/issues/15549>`_, `pr#8938 <http://github.com/ceph/ceph/pull/8938>`_, Boris Ranto, Sage Weil, Nathan Cutler, Lars Marowsky-Bree)
* ceph_test_librbd_fsx crashes during journal replay shut down (`issue#16123 <http://tracker.ceph.com/issues/16123>`_, `pr#9556 <http://github.com/ceph/ceph/pull/9556>`_, Jason Dillaman)
* cls_journal: initialize empty commit position upon client register (`issue#15757 <http://tracker.ceph.com/issues/15757>`_, `pr#9376 <http://github.com/ceph/ceph/pull/9376>`_, runsisi, Venky Shankar)
* cls::rbd: mirror_image_status_list returned max 64 items (`pr#9069 <http://github.com/ceph/ceph/pull/9069>`_, Mykola Golub)
* cls_rbd: mirror image status summary should read full directory (`issue#16178 <http://tracker.ceph.com/issues/16178>`_, `pr#9608 <http://github.com/ceph/ceph/pull/9608>`_, Jason Dillaman)
* common: BackoffThrottle spins unnecessarily with very small backoff while the throttle is full (`issue#15953 <http://tracker.ceph.com/issues/15953>`_, `pr#9579 <http://github.com/ceph/ceph/pull/9579>`_, Samuel Just)
* common: Do not link lttng into libglobal (`pr#9194 <http://github.com/ceph/ceph/pull/9194>`_, Karol Mroz)
* doc: update mirroring guide to include pool/image status commands (`issue#15746 <http://tracker.ceph.com/issues/15746>`_, `pr#9180 <http://github.com/ceph/ceph/pull/9180>`_, Mykola Golub)
* librbd: Disabling journaling feature results in "Transport endpoint is not connected" error (`issue#15863 <http://tracker.ceph.com/issues/15863>`_, `pr#9548 <http://github.com/ceph/ceph/pull/9548>`_, Yuan Zhou)
* librbd: do not shut down exclusive lock while acquiring' (`issue#16291 <http://tracker.ceph.com/issues/16291>`_, `issue#16260 <http://tracker.ceph.com/issues/16260>`_, `pr#9691 <http://github.com/ceph/ceph/pull/9691>`_, Jason Dillaman)
* librbd: Initial python APIs to support mirroring (`issue#15656 <http://tracker.ceph.com/issues/15656>`_, `pr#9550 <http://github.com/ceph/ceph/pull/9550>`_, Mykola Golub)
* librbd: journal IO error results in failed assertion in AioCompletion (`issue#16077 <http://tracker.ceph.com/issues/16077>`_, `issue#15034 <http://tracker.ceph.com/issues/15034>`_, `issue#15791 <http://tracker.ceph.com/issues/15791>`_, `pr#9611 <http://github.com/ceph/ceph/pull/9611>`_, Hector Martin, Jason Dillaman)
* librbd: journal: live replay might skip entries from previous object set (`issue#15864 <http://tracker.ceph.com/issues/15864>`_, `issue#15665 <http://tracker.ceph.com/issues/15665>`_, `pr#9217 <http://github.com/ceph/ceph/pull/9217>`_, Jason Dillaman)
* librbd: journal: support asynchronous shutdown (`issue#15949 <http://tracker.ceph.com/issues/15949>`_, `issue#14530 <http://tracker.ceph.com/issues/14530>`_, `issue#15993 <http://tracker.ceph.com/issues/15993>`_, `pr#9373 <http://github.com/ceph/ceph/pull/9373>`_, Jason Dillaman)
* librbd: Metadata config overrides are applied synchronously (`issue#15928 <http://tracker.ceph.com/issues/15928>`_, `pr#9318 <http://github.com/ceph/ceph/pull/9318>`_, Jason Dillaman)
* librbd: Object Map is showing as invalid, even when Object Map is disabled for that Image. (`issue#16076 <http://tracker.ceph.com/issues/16076>`_, `pr#9555 <http://github.com/ceph/ceph/pull/9555>`_, xinxin shu)
* librbd: prevent error messages when journal externally disabled (`issue#16114 <http://tracker.ceph.com/issues/16114>`_, `pr#9610 <http://github.com/ceph/ceph/pull/9610>`_, Zhiqiang Wang, Jason Dillaman)
* librbd: recursive lock possible when disabling journaling (`issue#16235 <http://tracker.ceph.com/issues/16235>`_, `pr#9654 <http://github.com/ceph/ceph/pull/9654>`_, Jason Dillaman)
* librbd: refresh image if needed in mirror functions (`issue#16096 <http://tracker.ceph.com/issues/16096>`_, `pr#9609 <http://github.com/ceph/ceph/pull/9609>`_, Jon Bernard)
* librbd: remove should ignore mirror errors from older OSDs (`issue#16268 <http://tracker.ceph.com/issues/16268>`_, `pr#9692 <http://github.com/ceph/ceph/pull/9692>`_, Jason Dillaman)
* librbd: reuse ImageCtx::finisher and SafeTimer for lots of images case (`issue#13938 <http://tracker.ceph.com/issues/13938>`_, `pr#9580 <http://github.com/ceph/ceph/pull/9580>`_, Haomai Wang)
* mds: order directories by hash and fix simultaneous readdir races (`issue#15508 <http://tracker.ceph.com/issues/15508>`_, `pr#9655 <http://github.com/ceph/ceph/pull/9655>`_, Yan, Zheng, Greg Farnum)
* mon: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?) (`issue#13990 <http://tracker.ceph.com/issues/13990>`_, `pr#9100 <http://github.com/ceph/ceph/pull/9100>`_, Kefu Chai)
* osd: acting_primary not updated on split (`issue#15523 <http://tracker.ceph.com/issues/15523>`_, `pr#8968 <http://github.com/ceph/ceph/pull/8968>`_, Sage Weil)
* osd: boot race with noup being set (`issue#15678 <http://tracker.ceph.com/issues/15678>`_, `pr#9101 <http://github.com/ceph/ceph/pull/9101>`_, Sage Weil)
* pybind: rbd API should default features parameter to None (`issue#15982 <http://tracker.ceph.com/issues/15982>`_, `pr#9553 <http://github.com/ceph/ceph/pull/9553>`_, Mykola Golub)
* qa/workunits: ensure replay has started before checking position (`issue#16248 <http://tracker.ceph.com/issues/16248>`_, `pr#9674 <http://github.com/ceph/ceph/pull/9674>`_, Jason Dillaman)
* rbd-mirror: Delete local image mirror when remote image mirroring is disabled (`issue#15916 <http://tracker.ceph.com/issues/15916>`_, `issue#14421 <http://tracker.ceph.com/issues/14421>`_, `pr#9372 <http://github.com/ceph/ceph/pull/9372>`_, runsisi, Mykola Golub, Ricardo Dias)
* rbd-mirror: do not propagate deletions when pool unavailable (`issue#16229 <http://tracker.ceph.com/issues/16229>`_, `pr#9630 <http://github.com/ceph/ceph/pull/9630>`_, Jason Dillaman)
* rbd-mirror: do not re-use image id from mirror directory if creating image (`issue#16253 <http://tracker.ceph.com/issues/16253>`_, `pr#9673 <http://github.com/ceph/ceph/pull/9673>`_, Jason Dillaman)
* rbd-mirror: fix journal shut down ordering (`issue#16165 <http://tracker.ceph.com/issues/16165>`_, `pr#9628 <http://github.com/ceph/ceph/pull/9628>`_, Jason Dillaman)
* rbd-mirror: potential crash during image status update (`issue#15909 <http://tracker.ceph.com/issues/15909>`_, `pr#9226 <http://github.com/ceph/ceph/pull/9226>`_, Mykola Golub, Jason Dillaman)
* rbd-mirror: refresh image after creating sync point (`issue#16196 <http://tracker.ceph.com/issues/16196>`_, `pr#9627 <http://github.com/ceph/ceph/pull/9627>`_, Jason Dillaman)
* rbd-mirror should disable the rbd cache for local images (`issue#15930 <http://tracker.ceph.com/issues/15930>`_, `pr#9317 <http://github.com/ceph/ceph/pull/9317>`_, Jason Dillaman)
* rpm: ceph gid mismatch on upgrade from hammer with pre-existing ceph user (SUSE) (`issue#15869 <http://tracker.ceph.com/issues/15869>`_, `pr#9424 <http://github.com/ceph/ceph/pull/9424>`_, Nathan Cutler)
* systemd: ceph-{mds,mon,osd,radosgw} systemd unit files need wants=time-sync.target (`issue#15419 <http://tracker.ceph.com/issues/15419>`_, `pr#8802 <http://github.com/ceph/ceph/pull/8802>`_, Nathan Cutler)
* test: failure in journal.sh workunit test (`issue#16011 <http://tracker.ceph.com/issues/16011>`_, `pr#9377 <http://github.com/ceph/ceph/pull/9377>`_, Mykola Golub)
* journal: incorrectly computed object offset within set (`issue#15765 <http://tracker.ceph.com/issues/15765>`_, `pr#9038 <http://github.com/ceph/ceph/pull/9038>`_, Jason Dillaman)
* librbd: client-side handling for incompatible object map sizes (`issue#15642 <http://tracker.ceph.com/issues/15642>`_, `pr#9039 <http://github.com/ceph/ceph/pull/9039>`_, Jason Dillaman)
* librbd: constrain size of AioWriteEvent journal entries (`issue#15750 <http://tracker.ceph.com/issues/15750>`_, `pr#9048 <http://github.com/ceph/ceph/pull/9048>`_, Jason Dillaman)
* librbd: does not crash if image header is too short (`pr#9044 <http://github.com/ceph/ceph/pull/9044>`_, Kefu Chai)
* librbd: Errors encountered disabling object-map while flatten is in-progress (`issue#15572 <http://tracker.ceph.com/issues/15572>`_, `pr#8869 <http://github.com/ceph/ceph/pull/8869>`_, Jason Dillaman)
* librbd: fix get/list mirror image status API (`issue#15771 <http://tracker.ceph.com/issues/15771>`_, `pr#9036 <http://github.com/ceph/ceph/pull/9036>`_, Mykola Golub)
* librbd: Parent image is closed twice if error encountered while opening (`issue#15574 <http://tracker.ceph.com/issues/15574>`_, `pr#8867 <http://github.com/ceph/ceph/pull/8867>`_, Jason Dillaman)
* librbd: possible double-free of object map invalidation request upon error (`issue#15643 <http://tracker.ceph.com/issues/15643>`_, `pr#8865 <http://github.com/ceph/ceph/pull/8865>`_, runsisi)
* librbd: possible race condition leads to use-after-free (`issue#15690 <http://tracker.ceph.com/issues/15690>`_, `pr#9009 <http://github.com/ceph/ceph/pull/9009>`_, Jason Dillaman)
* librbd: potential concurrent event processing during journal replay (`issue#15755 <http://tracker.ceph.com/issues/15755>`_, `pr#9040 <http://github.com/ceph/ceph/pull/9040>`_, Jason Dillaman)
* librbd: put the validation of image snap context earlier (`pr#9046 <http://github.com/ceph/ceph/pull/9046>`_, runsisi)
* librbd: reduce log level for image format 1 warning (`issue#15577 <http://tracker.ceph.com/issues/15577>`_, `pr#9003 <http://github.com/ceph/ceph/pull/9003>`_, Jason Dillaman)
* mds/MDSAuthCap parse no longer fails on paths with hyphens (`issue#15465 <http://tracker.ceph.com/issues/15465>`_, `pr#8969 <http://github.com/ceph/ceph/pull/8969>`_, John Spray)
* mds: MDS incarnation no longer gets lost after remove filesystem (`issue#15399 <http://tracker.ceph.com/issues/15399>`_, `pr#8970 <http://github.com/ceph/ceph/pull/8970>`_, John Spray)
* mon/OSDMonitor: avoid underflow in reweight-by-utilization if max_change=1 (`issue#15655 <http://tracker.ceph.com/issues/15655>`_, `pr#9006 <http://github.com/ceph/ceph/pull/9006>`_, Samuel Just)
* python: clone operation will fail if config overridden with "rbd default format = 1" (`issue#15685 <http://tracker.ceph.com/issues/15685>`_, `pr#8972 <http://github.com/ceph/ceph/pull/8972>`_, Jason Dillaman)
* rbd CLI to retrieve rbd mirror state for a pool / specific image (`issue#15144 <http://tracker.ceph.com/issues/15144>`_, `issue#14420 <http://tracker.ceph.com/issues/14420>`_, `pr#8868 <http://github.com/ceph/ceph/pull/8868>`_, Mykola Golub)
* rbd disk-usage CLI command should support calculating full image usage (`issue#14540 <http://tracker.ceph.com/issues/14540>`_, `pr#8870 <http://github.com/ceph/ceph/pull/8870>`_, Jason Dillaman)
* rbd: help message distinction between commands and aliases (`issue#15521 <http://tracker.ceph.com/issues/15521>`_, `pr#9004 <http://github.com/ceph/ceph/pull/9004>`_, Yongqiang He)
* rbd-mirror can crash if start up is interrupted (`issue#15630 <http://tracker.ceph.com/issues/15630>`_, `pr#8866 <http://github.com/ceph/ceph/pull/8866>`_, Jason Dillaman)
* rbd-mirror: image sync needs to handle snapshot size and protection status (`issue#15110 <http://tracker.ceph.com/issues/15110>`_, `pr#9050 <http://github.com/ceph/ceph/pull/9050>`_, Jason Dillaman)
* rbd-mirror: lockdep error during bootstrap (`issue#15664 <http://tracker.ceph.com/issues/15664>`_, `pr#9008 <http://github.com/ceph/ceph/pull/9008>`_, Jason Dillaman)
* rgw: add AWS4 completion support for RGW_OP_SET_BUCKET_WEBSITE (`issue#15626 <http://tracker.ceph.com/issues/15626>`_, `pr#9018 <http://github.com/ceph/ceph/pull/9018>`_, Javier M. Mellid)
* rgw: radosgw-admin zone set cuts pool names short if name starts with a period (`issue#15598 <http://tracker.ceph.com/issues/15598>`_, `pr#9029 <http://github.com/ceph/ceph/pull/9029>`_, Yehuda Sadeh)
* autotools: fix out of tree build (Krxysztof Kosinski)
* autotools: improve make check output (Loic Dachary)
* Be more careful about directory fragmentation and scrubbing (`issue#15167 <http://tracker.ceph.com/issues/15167>`_, `pr#8180 <http://github.com/ceph/ceph/pull/8180>`_, Yan, Zheng)
* buffer: increment history alloc as well in raw_combined (`issue#14955 <http://tracker.ceph.com/issues/14955>`_, `pr#7910 <http://github.com/ceph/ceph/pull/7910>`_, Samuel Just)
* buffer: make usable outside of ceph source again (`pr#6863 <http://github.com/ceph/ceph/pull/6863>`_, Josh Durgin)
* buffer: raw_combined allocations buffer and ref count together (`pr#7612 <http://github.com/ceph/ceph/pull/7612>`_, Sage Weil)
* build/ops: rbd-replay moved from ceph-test-dbg to ceph-common-dbg (`issue#13785 <http://tracker.ceph.com/issues/13785>`_, `pr#6578 <http://github.com/ceph/ceph/pull/6578>`_, Loic Dachary)
* build/ops: systemd ceph-disk unit must not assume /bin/flock (`issue#13975 <http://tracker.ceph.com/issues/13975>`_, `pr#6803 <http://github.com/ceph/ceph/pull/6803>`_, Loic Dachary)
* ceph-disk: add -f flag for btrfs mkfs (`pr#7222 <http://github.com/ceph/ceph/pull/7222>`_, Darrell Enns)
* ceph-disk: Add --setuser and --setgroup options for ceph-disk (`pr#7351 <http://github.com/ceph/ceph/pull/7351>`_, Mike Shuey)
* ceph-disk: ceph-disk list fails on /dev/cciss!c0d0 (`issue#13970 <http://tracker.ceph.com/issues/13970>`_, `issue#14233 <http://tracker.ceph.com/issues/14233>`_, `issue#14230 <http://tracker.ceph.com/issues/14230>`_, `pr#6879 <http://github.com/ceph/ceph/pull/6879>`_, Loic Dachary)
* ceph-disk: compare parted output with the dereferenced path (`issue#13438 <http://tracker.ceph.com/issues/13438>`_, `pr#6219 <http://github.com/ceph/ceph/pull/6219>`_, Joe Julian)
* ceph-disk: get Nonetype when ceph-disk list with --format plain on single device. (`pr#6410 <http://github.com/ceph/ceph/pull/6410>`_, Vicente Cheng)
* ceph-disk: warn for prepare partitions with bad GUIDs (`issue#13943 <http://tracker.ceph.com/issues/13943>`_, `pr#6760 <http://github.com/ceph/ceph/pull/6760>`_, David Disseldorp)
* ceph-fuse,libcephfs: Fix client handling of "lost" open directories on shutdown (`issue#14996 <http://tracker.ceph.com/issues/14996>`_, `pr#7994 <http://github.com/ceph/ceph/pull/7994>`_, Yan, Zheng)
* ceph-fuse,libcephfs: fix free fds being exhausted eventually because freed fds are never put back (`issue#14798 <http://tracker.ceph.com/issues/14798>`_, `pr#7685 <http://github.com/ceph/ceph/pull/7685>`_, Zhi Zhang)
* ceph.in: avoid a broken pipe error when use ceph command (`issue#14354 <http://tracker.ceph.com/issues/14354>`_, `pr#7212 <http://github.com/ceph/ceph/pull/7212>`_, Bo Cai)
* ceph.in: correct dev python path for automake builds (`pr#8360 <http://github.com/ceph/ceph/pull/8360>`_, Josh Durgin)
* ceph.in: fix python libpath for automake as well (`pr#8362 <http://github.com/ceph/ceph/pull/8362>`_, Josh Durgin)
* ceph.in: Minor python3 specific changes (`pr#7947 <http://github.com/ceph/ceph/pull/7947>`_, Sarthak Munshi)
* ceph-kvstore-tool: handle bad out file on command line (`pr#6093 <http://github.com/ceph/ceph/pull/6093>`_, Kefu Chai)
* ceph.spec.in: do not install Ceph RA on systemd platforms (`issue#14828 <http://tracker.ceph.com/issues/14828>`_, `pr#7894 <http://github.com/ceph/ceph/pull/7894>`_, Nathan Cutler)
* ceph.spec.in: fix openldap and openssl build dependencies for SUSE (`issue#15138 <http://tracker.ceph.com/issues/15138>`_, `pr#8120 <http://github.com/ceph/ceph/pull/8120>`_, Nathan Cutler)
* ceph.spec.in: limit _smp_mflags when lowmem_builder is set in SUSE's OBS (`issue#13858 <http://tracker.ceph.com/issues/13858>`_, `pr#6691 <http://github.com/ceph/ceph/pull/6691>`_, Nathan Cutler)
* client: check if Fh is readable when processing a read (`issue#11517 <http://tracker.ceph.com/issues/11517>`_, `pr#7209 <http://github.com/ceph/ceph/pull/7209>`_, Yan, Zheng)
* client: close mds sessions in shutdown() (`pr#6269 <http://github.com/ceph/ceph/pull/6269>`_, John Spray)
* client: don't invalidate page cache when inode is no longer used (`pr#6380 <http://github.com/ceph/ceph/pull/6380>`_, Yan, Zheng)
* client: don't mark_down on command reply (`pr#6204 <http://github.com/ceph/ceph/pull/6204>`_, John Spray)
* client: drop prefix from ints (`pr#6275 <http://github.com/ceph/ceph/pull/6275>`_, John Coyle)
* client: sys/file.h includes for flock operations (`pr#6282 <http://github.com/ceph/ceph/pull/6282>`_, John Coyle)
* client: use null snapc to check pool permission (`issue#13714 <http://tracker.ceph.com/issues/13714>`_, `pr#6497 <http://github.com/ceph/ceph/pull/6497>`_, Yan, Zheng)
* cls_rbd: pass WILLNEED fadvise flags during object map update (`issue#15332 <http://tracker.ceph.com/issues/15332>`_, `pr#8380 <http://github.com/ceph/ceph/pull/8380>`_, Jason Dillaman)
* cls_rbd: protect against excessively large object maps (`issue#15121 <http://tracker.ceph.com/issues/15121>`_, `pr#8099 <http://github.com/ceph/ceph/pull/8099>`_, Jason Dillaman)
* cls_rbd: read_peers: update last_read on next cls_cxx_map_get_vals (`pr#8374 <http://github.com/ceph/ceph/pull/8374>`_, Mykola Golub)
* common: assert: abort() rather than throw (`pr#6804 <http://github.com/ceph/ceph/pull/6804>`_, Adam C. Emerson)
* common: assert: __STRING macro is not defined by musl libc. (`pr#6210 <http://github.com/ceph/ceph/pull/6210>`_, John Coyle)
* common/bit_vector: use hard-coded value for block size (`issue#14747 <http://tracker.ceph.com/issues/14747>`_, `pr#7610 <http://github.com/ceph/ceph/pull/7610>`_, Jason Dillaman)
* common: buffer: add cached_crc and cached_crc_adjust counts to perf dump (`pr#6535 <http://github.com/ceph/ceph/pull/6535>`_, Ning Yao)
* common: buffer/assert minor fixes (`pr#6990 <http://github.com/ceph/ceph/pull/6990>`_, Matt Benjamin)
* common: fix time_t cast in decode (`issue#15330 <http://tracker.ceph.com/issues/15330>`_, `pr#8419 <http://github.com/ceph/ceph/pull/8419>`_, Adam C. Emerson)
* common/Formatter: avoid newline if there is no output (`pr#5351 <http://github.com/ceph/ceph/pull/5351>`_, Aran85)
* common: improve shared_cache and simple_cache efficiency with hash table (`pr#6909 <http://github.com/ceph/ceph/pull/6909>`_, Ning Yao)
* common/lockdep: increase max lock names (`pr#6961 <http://github.com/ceph/ceph/pull/6961>`_, Sage Weil)
* common: reduce CPU usage by making stringstream in stringify function thread local (`pr#6543 <http://github.com/ceph/ceph/pull/6543>`_, Evgeniy Firsov)
* common: re-enable backtrace support (`pr#6771 <http://github.com/ceph/ceph/pull/6771>`_, Jason Dillaman)
* common: set thread name from correct thread (`pr#7845 <http://github.com/ceph/ceph/pull/7845>`_, Igor Podoski)
* common: signal_handler: added support for using reentrant strsignal() implementations vs. sys_siglist[] (`pr#6796 <http://github.com/ceph/ceph/pull/6796>`_, John Coyle)
* common: snappy decompressor may assert when handling segmented input bufferlist (`issue#14400 <http://tracker.ceph.com/issues/14400>`_, `pr#7268 <http://github.com/ceph/ceph/pull/7268>`_, Igor Fedotov)
* common/str_map: cleanup: replaced get_str_map() function overloading by using default parameters for delimiters (`pr#7266 <http://github.com/ceph/ceph/pull/7266>`_, Sahithi R V)
* common/strtol.cc: fix the coverity warnings (`pr#7967 <http://github.com/ceph/ceph/pull/7967>`_, Kefu Chai)
* common: unit test for interval_set implementations (`pr#6 <http://github.com/ceph/ceph/pull/6>`_, Igor Fedotov)
* common: use namespace instead of subclasses for buffer (`pr#6686 <http://github.com/ceph/ceph/pull/6686>`_, Michal Jarzabek)
* common: various fixes from SCA runs (`pr#7680 <http://github.com/ceph/ceph/pull/7680>`_, Danny Al-Gaaf)
* common: WorkQueue: new PointerWQ base class for ContextWQ (`issue#13636 <http://tracker.ceph.com/issues/13636>`_, `pr#6525 <http://github.com/ceph/ceph/pull/6525>`_, Jason Dillaman)
* compat: use prefixed typeof extension (`pr#6216 <http://github.com/ceph/ceph/pull/6216>`_, John Coyle)
* config: complains when a setting is not tracked (`issue#11692 <http://tracker.ceph.com/issues/11692>`_, `pr#7085 <http://github.com/ceph/ceph/pull/7085>`_, Kefu Chai)
* config: fix osd_crush_initial_weight (`pr#7975 <http://github.com/ceph/ceph/pull/7975>`_, You Ji)
* config: increase default async op threads (`pr#7802 <http://github.com/ceph/ceph/pull/7802>`_, Piotr Dałek)
* config_opts: disable filestore throttle soft backoff by default (`pr#8265 <http://github.com/ceph/ceph/pull/8265>`_, Samuel Just)
* configure.ac: boost_iostreams is required, not optional (`pr#7816 <http://github.com/ceph/ceph/pull/7816>`_, Hector Martin)
* configure.ac: macro fix (`pr#6769 <http://github.com/ceph/ceph/pull/6769>`_, Igor Podoski)
* configure.ac: make "--with-librocksdb-static" default to 'check' (`issue#14463 <http://tracker.ceph.com/issues/14463>`_, `pr#7317 <http://github.com/ceph/ceph/pull/7317>`_, Dan Mick)
* configure.ac: update help strings for cython (`pr#7856 <http://github.com/ceph/ceph/pull/7856>`_, Josh Durgin)
* configure: Add -D_LARGEFILE64_SOURCE to Linux build. (`pr#8402 <http://github.com/ceph/ceph/pull/8402>`_, Ira Cooper)
* crushtool: Don't crash when called on a file that isn't a crushmap (`issue#8286 <http://tracker.ceph.com/issues/8286>`_, `pr#8038 <http://github.com/ceph/ceph/pull/8038>`_, Brad Hubbard)
* debian: package librgw_file* tests (`pr#7930 <http://github.com/ceph/ceph/pull/7930>`_, Ken Dreyer)
* debian: packaging fixes for jewel (`pr#7807 <http://github.com/ceph/ceph/pull/7807>`_, Ken Dreyer, Ali Maredia)
* debian/rpm split servers (`issue#10587 <http://tracker.ceph.com/issues/10587>`_, `pr#7746 <http://github.com/ceph/ceph/pull/7746>`_, Ken Dreyer)
* debian/rules: put init-ceph in /etc/init.d/ceph, not ceph-base (`issue#15329 <http://tracker.ceph.com/issues/15329>`_, `pr#8406 <http://github.com/ceph/ceph/pull/8406>`_, Dan Mick)
* doc: adding "--allow-shrink" in decreasing the size of the rbd block to distinguish from the increasing option (`pr#7020 <http://github.com/ceph/ceph/pull/7020>`_, Yehua)
* doc: add orphans commands to radosgw-admin(8) (`issue#14637 <http://tracker.ceph.com/issues/14637>`_, `pr#7518 <http://github.com/ceph/ceph/pull/7518>`_, Ken Dreyer)
* doc: add v0.80.11 to the release timeline (`pr#6658 <http://github.com/ceph/ceph/pull/6658>`_, Loic Dachary)
* doc: admin/build-doc: add lxml dependencies on debian (`pr#6610 <http://github.com/ceph/ceph/pull/6610>`_, Ken Dreyer)
* doc: admin/build-doc: make paths absolute (`pr#7119 <http://github.com/ceph/ceph/pull/7119>`_, Dan Mick)
* doc: misc updates (Fracois Lafont, Ken Dreyer, Kefu Chai, Owen Synge, Gael Fenet-Garde, Loic Dachary, Yannick Atchy-Dalama, Jiaying Ren, Kevin Caradant, Robert Maxime, Nicolas Yong, Germain Chipaux, Arthur Gorjux, Gabriel Sentucq, Clement Lebrun, Jean-Remi Deveaux, Clair Massot, Robin Tang, Thomas Laumondais, Jordan Dorne, Yuan Zhou, Valentin Thomas, Pierre Chaumont, Benjamin Troquereau, Benjamin Sesia, Vikhyat Umrao, Nilamdyuti Goswami, Vartika Rai, Florian Haas, Loic Dachary, Simon Guinot, Andy Allan, Alistair Israel, Ken Dreyer, Robin Rehu, Lee Revell, Florian Marsylle, Thomas Johnson, Bosse Klykken, Travis Rhoden, Ian Kelling)
* doc: rgw: port changes from downstream to upstream (`pr#7264 <http://github.com/ceph/ceph/pull/7264>`_, Bara Ancincova)
* doc: rgw_region_root_pool option should be in [global] (`issue#15244 <http://tracker.ceph.com/issues/15244>`_, `pr#8271 <http://github.com/ceph/ceph/pull/8271>`_, Vikhyat Umrao)
* doc: update doc for with new pool settings (`pr#5951 <http://github.com/ceph/ceph/pull/5951>`_, Guang Yang)
* doc: Updated the rados command man page to include the --run-name opt… (`issue#12899 <http://tracker.ceph.com/issues/12899>`_, `pr#5900 <http://github.com/ceph/ceph/pull/5900>`_, ritz303)
* fix FTBFS introduced by d0af316 (`pr#7792 <http://github.com/ceph/ceph/pull/7792>`_, Kefu Chai)
* fix: use right init_flags to finish CephContext (`pr#6549 <http://github.com/ceph/ceph/pull/6549>`_, Yunchuan Wen)
* fs: be more careful about the "mds setmap" command to prevent breakage (`issue#14380 <http://tracker.ceph.com/issues/14380>`_, `pr#7262 <http://github.com/ceph/ceph/pull/7262>`_, Yan, Zheng)
* ghobject_t: use # instead of ! as a separator (`pr#8055 <http://github.com/ceph/ceph/pull/8055>`_, Sage Weil)
* global: do not start two daemons with a single pid-file (`issue#13422 <http://tracker.ceph.com/issues/13422>`_, `pr#7075 <http://github.com/ceph/ceph/pull/7075>`_, shun song)
* global: do not start two daemons with a single pid-file (part 2) (`issue#13422 <http://tracker.ceph.com/issues/13422>`_, `pr#7463 <http://github.com/ceph/ceph/pull/7463>`_, Loic Dachary)
* helgrind: additional race conditionslibrbd: journal replay should honor inter-event dependencies (`pr#7274 <http://github.com/ceph/ceph/pull/7274>`_, Jason Dillaman)
* helgrind: fix real (and imaginary) race conditions (`issue#14163 <http://tracker.ceph.com/issues/14163>`_, `pr#7208 <http://github.com/ceph/ceph/pull/7208>`_, Jason Dillaman)
* include/encoding: do not try to be clever with list encoding (`pr#7913 <http://github.com/ceph/ceph/pull/7913>`_, Sage Weil)
* init-ceph: do umount when the path exists. (`pr#6866 <http://github.com/ceph/ceph/pull/6866>`_, Xiaoxi Chen)
* init-ceph.in: allow case-insensitive true in `osd crush update on start' (`pr#7943 <http://github.com/ceph/ceph/pull/7943>`_, Eric Cook)
* init-ceph.in: skip ceph-disk if it is not present (`issue#10587 <http://tracker.ceph.com/issues/10587>`_, `pr#7286 <http://github.com/ceph/ceph/pull/7286>`_, Ken Dreyer)
* init-ceph: use getopt to make option processing more flexible (`issue#3015 <http://tracker.ceph.com/issues/3015>`_, `pr#6089 <http://github.com/ceph/ceph/pull/6089>`_, Nathan Cutler)
* journal: async methods to (un)register and update client (`pr#7832 <http://github.com/ceph/ceph/pull/7832>`_, Mykola Golub)
* journal: disconnect watch after watch error (`issue#14168 <http://tracker.ceph.com/issues/14168>`_, `pr#7113 <http://github.com/ceph/ceph/pull/7113>`_, Jason Dillaman)
* journal: fire replay complete event after reading last object (`issue#13924 <http://tracker.ceph.com/issues/13924>`_, `pr#6762 <http://github.com/ceph/ceph/pull/6762>`_, Jason Dillaman)
* journal: fix final result for JournalTrimmer::C_RemoveSet (`pr#8516 <http://github.com/ceph/ceph/pull/8516>`_, runsisi)
* journal: fix race condition between Future and journal shutdown (`issue#15364 <http://tracker.ceph.com/issues/15364>`_, `pr#8477 <http://github.com/ceph/ceph/pull/8477>`_, Jason Dillaman)
* journal: flush commit position on metadata shutdown (`pr#7385 <http://github.com/ceph/ceph/pull/7385>`_, Mykola Golub)
* journal: improve commit position tracking (`pr#7776 <http://github.com/ceph/ceph/pull/7776>`_, Jason Dillaman)
* journal: incremental improvements and fixes (`pr#6552 <http://github.com/ceph/ceph/pull/6552>`_, Mykola Golub)
* journal: prevent race injecting new records into overflowed object (`issue#15202 <http://tracker.ceph.com/issues/15202>`_, `pr#8220 <http://github.com/ceph/ceph/pull/8220>`_, Jason Dillaman)
* librados: aix gcc librados port (`pr#6675 <http://github.com/ceph/ceph/pull/6675>`_, Rohan Mars)
* librados: avoid malloc(0) (which can return NULL on some platforms) (`issue#13944 <http://tracker.ceph.com/issues/13944>`_, `pr#6779 <http://github.com/ceph/ceph/pull/6779>`_, Dan Mick)
* librados: cancel aio notification linger op upon completion (`pr#8102 <http://github.com/ceph/ceph/pull/8102>`_, Jason Dillaman)
* librados: check connection state in rados_monitor_log (`issue#14499 <http://tracker.ceph.com/issues/14499>`_, `pr#7350 <http://github.com/ceph/ceph/pull/7350>`_, David Disseldorp)
* librados: clean up Objecter.h (`pr#6731 <http://github.com/ceph/ceph/pull/6731>`_, Jie Wang)
* librados: fix pool alignment API overflow issue (`issue#13715 <http://tracker.ceph.com/issues/13715>`_, `pr#6489 <http://github.com/ceph/ceph/pull/6489>`_, xie xingguo)
* librados: fix potential null pointer access when do pool_snap_list (`issue#13639 <http://tracker.ceph.com/issues/13639>`_, `pr#6422 <http://github.com/ceph/ceph/pull/6422>`_, xie xingguo)
* librados: fix PromoteOn2ndRead test for EC (`pr#6373 <http://github.com/ceph/ceph/pull/6373>`_, Sage Weil)
* librados: fix rare race where pool op callback may hang forever (`issue#13642 <http://tracker.ceph.com/issues/13642>`_, `pr#6426 <http://github.com/ceph/ceph/pull/6426>`_, xie xingguo)
* librados: fix several flaws introduced by the enumeration_objects API (`issue#14299 <http://tracker.ceph.com/issues/14299>`_, `issue#14301 <http://tracker.ceph.com/issues/14301>`_, `issue#14300 <http://tracker.ceph.com/issues/14300>`_, `pr#7156 <http://github.com/ceph/ceph/pull/7156>`_, xie xingguo)
* librados_test_stub: protect against notify/unwatch race (`pr#7540 <http://github.com/ceph/ceph/pull/7540>`_, Jason Dillaman)
* librados: wrongly passed in argument for stat command (`issue#13703 <http://tracker.ceph.com/issues/13703>`_, `pr#6476 <http://github.com/ceph/ceph/pull/6476>`_, xie xingguo)
* librbd: API: async open and close (`issue#14264 <http://tracker.ceph.com/issues/14264>`_, `pr#7259 <http://github.com/ceph/ceph/pull/7259>`_, Mykola Golub)
* librbd: automatically flush IO after blocking write operations (`issue#13913 <http://tracker.ceph.com/issues/13913>`_, `pr#6742 <http://github.com/ceph/ceph/pull/6742>`_, Jason Dillaman)
* librbd: better handling of exclusive lock transition period (`pr#7204 <http://github.com/ceph/ceph/pull/7204>`_, Jason Dillaman)
* librbd: block maintenance ops until after journal is ready (`issue#14510 <http://tracker.ceph.com/issues/14510>`_, `pr#7382 <http://github.com/ceph/ceph/pull/7382>`_, Jason Dillaman)
* librbd: block read requests until journal replayed (`pr#7627 <http://github.com/ceph/ceph/pull/7627>`_, Jason Dillaman)
* librbd: check for presence of journal before attempting to remove (`issue#13912 <http://tracker.ceph.com/issues/13912>`_, `pr#6737 <http://github.com/ceph/ceph/pull/6737>`_, Jason Dillaman)
* librbd: clear error when older OSD doesn't support image flags (`issue#14122 <http://tracker.ceph.com/issues/14122>`_, `pr#7035 <http://github.com/ceph/ceph/pull/7035>`_, Jason Dillaman)
* librbd: correct include guard in RenameRequest.h (`pr#7143 <http://github.com/ceph/ceph/pull/7143>`_, Jason Dillaman)
* librbd: correct issues discovered during teuthology testing (`issue#14108 <http://tracker.ceph.com/issues/14108>`_, `issue#14107 <http://tracker.ceph.com/issues/14107>`_, `pr#6974 <http://github.com/ceph/ceph/pull/6974>`_, Jason Dillaman)
* librbd: correct issues discovered via valgrind memcheck (`pr#8132 <http://github.com/ceph/ceph/pull/8132>`_, Jason Dillaman)
* librbd: correct issues discovered when cache is disabled (`issue#14123 <http://tracker.ceph.com/issues/14123>`_, `pr#6979 <http://github.com/ceph/ceph/pull/6979>`_, Jason Dillaman)
* librbd: correct race conditions discovered during unit testing (`issue#14060 <http://tracker.ceph.com/issues/14060>`_, `pr#6923 <http://github.com/ceph/ceph/pull/6923>`_, Jason Dillaman)
* librbd: deadlock while attempting to flush AIO requests (`issue#13726 <http://tracker.ceph.com/issues/13726>`_, `pr#6508 <http://github.com/ceph/ceph/pull/6508>`_, Jason Dillaman)
* librbd: differentiate journal replay flush vs shut down (`pr#7698 <http://github.com/ceph/ceph/pull/7698>`_, Jason Dillaman)
* librbd: disable copy-on-read when not exclusive lock owner (`issue#14167 <http://tracker.ceph.com/issues/14167>`_, `pr#7129 <http://github.com/ceph/ceph/pull/7129>`_, Jason Dillaman)
* librbd: disable image mirroring when image is removed (`issue#15265 <http://tracker.ceph.com/issues/15265>`_, `pr#8375 <http://github.com/ceph/ceph/pull/8375>`_, Ricardo Dias)
* librbd: do not ignore self-managed snapshot release result (`issue#14170 <http://tracker.ceph.com/issues/14170>`_, `pr#7043 <http://github.com/ceph/ceph/pull/7043>`_, Jason Dillaman)
* librbd: enable/disable image mirroring automatically for pool mode (`issue#15143 <http://tracker.ceph.com/issues/15143>`_, `pr#8204 <http://github.com/ceph/ceph/pull/8204>`_, Ricardo Dias)
* librbd: ensure copy-on-read requests are complete prior to closing parent image (`pr#6740 <http://github.com/ceph/ceph/pull/6740>`_, Jason Dillaman)
* librbd: ensure librados callbacks are flushed prior to destroying (`issue#14092 <http://tracker.ceph.com/issues/14092>`_, `pr#7040 <http://github.com/ceph/ceph/pull/7040>`_, Jason Dillaman)
* librbd: exit if parent's snap is gone during clone (`issue#14118 <http://tracker.ceph.com/issues/14118>`_, `pr#6968 <http://github.com/ceph/ceph/pull/6968>`_, xie xingguo)
* librbd: fix known test case race condition failures (`issue#13969 <http://tracker.ceph.com/issues/13969>`_, `pr#6800 <http://github.com/ceph/ceph/pull/6800>`_, Jason Dillaman)
* librbd: not necessary to hold owner_lock while releasing snap id (`issue#13914 <http://tracker.ceph.com/issues/13914>`_, `pr#6736 <http://github.com/ceph/ceph/pull/6736>`_, Jason Dillaman)
* librbd: optionally validate new RBD pools for snapshot support (`issue#13633 <http://tracker.ceph.com/issues/13633>`_, `pr#6925 <http://github.com/ceph/ceph/pull/6925>`_, Jason Dillaman)
* librbd: partial revert of commit 9b0e359 (`issue#13969 <http://tracker.ceph.com/issues/13969>`_, `pr#6789 <http://github.com/ceph/ceph/pull/6789>`_, Jason Dillaman)
* librbd: perf counters might not be initialized on error (`issue#13740 <http://tracker.ceph.com/issues/13740>`_, `pr#6523 <http://github.com/ceph/ceph/pull/6523>`_, Jason Dillaman)
* librbd: perf section name: use hyphen to separate components (`issue#13719 <http://tracker.ceph.com/issues/13719>`_, `pr#6516 <http://github.com/ceph/ceph/pull/6516>`_, Mykola Golub)
* librbd: reduce mem copies to user-buffer during read (`pr#7548 <http://github.com/ceph/ceph/pull/7548>`_, Jianpeng Ma)
* librbd: reduce verbosity of common error condition logging (`issue#14234 <http://tracker.ceph.com/issues/14234>`_, `pr#7114 <http://github.com/ceph/ceph/pull/7114>`_, Jason Dillaman)
* librbd: refresh image if required before replaying journal ops (`issue#14908 <http://tracker.ceph.com/issues/14908>`_, `pr#7978 <http://github.com/ceph/ceph/pull/7978>`_, Jason Dillaman)
* librbd: remove canceled tasks from timer thread (`issue#14476 <http://tracker.ceph.com/issues/14476>`_, `pr#7329 <http://github.com/ceph/ceph/pull/7329>`_, Douglas Fuller)
* librbd: remove duplicate read_only test in librbd::async_flatten (`pr#5856 <http://github.com/ceph/ceph/pull/5856>`_, runsisi)
* librbd: remove last synchronous librados calls from open/close state machine (`pr#7839 <http://github.com/ceph/ceph/pull/7839>`_, Jason Dillaman)
* librbd: replaying a journal op post-refresh requires locking (`pr#8028 <http://github.com/ceph/ceph/pull/8028>`_, Jason Dillaman)
* librbd: resize should only update image size within header (`issue#13674 <http://tracker.ceph.com/issues/13674>`_, `pr#6447 <http://github.com/ceph/ceph/pull/6447>`_, Jason Dillaman)
* librbd: retrieve image name when opening by id (`pr#7736 <http://github.com/ceph/ceph/pull/7736>`_, Mykola Golub)
* librbd: return error if we fail to delete object_map head object (`issue#14098 <http://tracker.ceph.com/issues/14098>`_, `pr#6958 <http://github.com/ceph/ceph/pull/6958>`_, xie xingguo)
* librbd: Revert "librbd: use task finisher per CephContext" (`issue#14780 <http://tracker.ceph.com/issues/14780>`_, `pr#7667 <http://github.com/ceph/ceph/pull/7667>`_, Josh Durgin)
* librbd: send notifications for mirroring status updates (`pr#8355 <http://github.com/ceph/ceph/pull/8355>`_, Jason Dillaman)
* librbd: several race conditions discovered under single CPU environment (`pr#7653 <http://github.com/ceph/ceph/pull/7653>`_, Jason Dillaman)
* librbd: simplify IO method signatures for 32bit environments (`pr#6700 <http://github.com/ceph/ceph/pull/6700>`_, Jason Dillaman)
* librbd: small fixes for error messages and readahead counter (`issue#14127 <http://tracker.ceph.com/issues/14127>`_, `pr#6983 <http://github.com/ceph/ceph/pull/6983>`_, xie xingguo)
* librbd: start perf counters after id is initialized (`issue#13720 <http://tracker.ceph.com/issues/13720>`_, `pr#6494 <http://github.com/ceph/ceph/pull/6494>`_, Mykola Golub)
* librbd: truncate does not need to mark the object as existing in the object map (`issue#14789 <http://tracker.ceph.com/issues/14789>`_, `pr#7772 <http://github.com/ceph/ceph/pull/7772>`_, xinxin shu)
* librbd: uninitialized state in snap remove state machine (`pr#6982 <http://github.com/ceph/ceph/pull/6982>`_, Jason Dillaman)
* librbd: update of mirror pool mode and mirror peer handling (`pr#7718 <http://github.com/ceph/ceph/pull/7718>`_, Jason Dillaman)
* librbd: use async librados notifications (`pr#7668 <http://github.com/ceph/ceph/pull/7668>`_, Jason Dillaman)
* log: do not repeat errors to stderr (`issue#14616 <http://tracker.ceph.com/issues/14616>`_, `pr#7983 <http://github.com/ceph/ceph/pull/7983>`_, Sage Weil)
* mds: expose state of recovery to status ASOK command (`issue#14146 <http://tracker.ceph.com/issues/14146>`_, `pr#7068 <http://github.com/ceph/ceph/pull/7068>`_, Yan, Zheng)
* mds: Extend the existing pool access checking to include specific RADOS namespacse. (`pr#8444 <https://github.com/ceph/ceph/pull/8444>`_, Yan, Zheng)
* mds: messages/MOSDOp: cast in assert to eliminate warnings (`issue#13625 <http://tracker.ceph.com/issues/13625>`_, `pr#6414 <http://github.com/ceph/ceph/pull/6414>`_, David Zafman)
* mds: tear down connections from `tell` commands (`issue#14048 <http://tracker.ceph.com/issues/14048>`_, `pr#6933 <http://github.com/ceph/ceph/pull/6933>`_, John Spray)
* mon: compact full epochs also (`issue#14537 <http://tracker.ceph.com/issues/14537>`_, `pr#7396 <http://github.com/ceph/ceph/pull/7396>`_, Kefu Chai)
* mon: consider pool size when creating pool (`issue#14509 <http://tracker.ceph.com/issues/14509>`_, `pr#7359 <http://github.com/ceph/ceph/pull/7359>`_, songbaisen)
* mon: consider the pool size when setting pool crush rule (`issue#14495 <http://tracker.ceph.com/issues/14495>`_, `pr#7341 <http://github.com/ceph/ceph/pull/7341>`_, song baisen)
* mon: degrade a log message to level 2 (`pr#6929 <http://github.com/ceph/ceph/pull/6929>`_, Kongming Wu)
* mon: do not send useless pg_create messages for split pgs (`pr#8247 <http://github.com/ceph/ceph/pull/8247>`_, Sage Weil)
* mon: don't require OSD W for MRemoveSnaps (`issue#13777 <http://tracker.ceph.com/issues/13777>`_, `pr#6601 <http://github.com/ceph/ceph/pull/6601>`_, John Spray)
* mon: go into ERR state if multiple PGs are stuck inactive (`issue#13923 <http://tracker.ceph.com/issues/13923>`_, `pr#7253 <http://github.com/ceph/ceph/pull/7253>`_, Wido den Hollander)
* mon: initialize last_* timestamps on new pgs to creation time (`issue#14952 <http://tracker.ceph.com/issues/14952>`_, `pr#7980 <http://github.com/ceph/ceph/pull/7980>`_, Sage Weil)
* mon: initialize recorded election epoch properly even when standalone (`issue#13627 <http://tracker.ceph.com/issues/13627>`_, `pr#6407 <http://github.com/ceph/ceph/pull/6407>`_, huanwen ren)
* mon: should not set isvalid = true when cephx_verify_authorizer return false (`issue#13525 <http://tracker.ceph.com/issues/13525>`_, `pr#6306 <http://github.com/ceph/ceph/pull/6306>`_, Ruifeng Yang)
* mon: show the pool quota info on ceph df detail command (`issue#14216 <http://tracker.ceph.com/issues/14216>`_, `pr#7094 <http://github.com/ceph/ceph/pull/7094>`_, song baisen)
* mon: some cleanup in MonmapMonitor.cc (`pr#7418 <http://github.com/ceph/ceph/pull/7418>`_, huanwen ren)
* mon: warn if pg(s) not scrubbed (`issue#13142 <http://tracker.ceph.com/issues/13142>`_, `pr#6440 <http://github.com/ceph/ceph/pull/6440>`_, Michal Jarzabek)
* mount.fuse.ceph: better parsing of arguments passed to mount.fuse.ceph by mount command (`issue#14735 <http://tracker.ceph.com/issues/14735>`_, `pr#7607 <http://github.com/ceph/ceph/pull/7607>`_, Florent Bautista)
* msg: async: small cleanups (`pr#7871 <http://github.com/ceph/ceph/pull/7871>`_, xie xingguo)
* msg/async: smarter MSG_MORE (`pr#7625 <http://github.com/ceph/ceph/pull/7625>`_, Piotr Dałek)
* msg: async: start over after failing to bind a port in specified range (`issue#14928 <http://tracker.ceph.com/issues/14928>`_, `issue#13002 <http://tracker.ceph.com/issues/13002>`_, `pr#7852 <http://github.com/ceph/ceph/pull/7852>`_, xie xingguo)
* msg/async: support of non-block connect in async messenger (`issue#12802 <http://tracker.ceph.com/issues/12802>`_, `pr#5848 <http://github.com/ceph/ceph/pull/5848>`_, Jianhui Yuan)
* msg/async: _try_send trim already sent for outcoming_bl more efficient (`pr#7970 <http://github.com/ceph/ceph/pull/7970>`_, Yan Jun)
* msg/async: will crash if enabling async msg because of an assertion (`pr#6640 <http://github.com/ceph/ceph/pull/6640>`_, Zhi Zhang)
* msg: filter out lo addr when bind osd addr (`pr#7012 <http://github.com/ceph/ceph/pull/7012>`_, Ji Chen)
* osd: avoid double-check for replaying and can_checkpoint() in FileStore::_check_replay_guard (`pr#6471 <http://github.com/ceph/ceph/pull/6471>`_, Ning Yao)
* osd: avoid duplicate op->mark_started in ReplicatedBackend (`pr#6689 <http://github.com/ceph/ceph/pull/6689>`_, Jacek J. Łakis)
* osd: avoid FORCE updating digest been overwritten by MAYBE when comparing scrub map (`pr#7051 <http://github.com/ceph/ceph/pull/7051>`_, Zhiqiang Wang)
* osd: bluestore/blockdevice: use std::mutex et al (`pr#7568 <http://github.com/ceph/ceph/pull/7568>`_, Sage Weil)
* osd: bluestore: bluefs: fix several small bugs (`issue#14344 <http://tracker.ceph.com/issues/14344>`_, `issue#14343 <http://tracker.ceph.com/issues/14343>`_, `pr#7200 <http://github.com/ceph/ceph/pull/7200>`_, xie xingguo)
* osd: bluestore/BlueFS: initialize super block_size earlier in mkfs (`pr#7535 <http://github.com/ceph/ceph/pull/7535>`_, Sage Weil)
* osd: bluestore: don't include when building without libaio (`issue#14207 <http://tracker.ceph.com/issues/14207>`_, `pr#7169 <http://github.com/ceph/ceph/pull/7169>`_, Mykola Golub)
* osd: bluestore: fix check for write falling within the same extent (`issue#14954 <http://tracker.ceph.com/issues/14954>`_, `pr#7892 <http://github.com/ceph/ceph/pull/7892>`_, Jianpeng Ma)
* osd: bluestore: for overwrite a extent, allocate new extent on min_alloc_size write (`pr#7996 <http://github.com/ceph/ceph/pull/7996>`_, Jianpeng Ma)
* osd: bluestore: improve fs-type verification and tidy up (`pr#7651 <http://github.com/ceph/ceph/pull/7651>`_, xie xingguo)
* osd: cancel failure reports if we fail to rebind network (`pr#6278 <http://github.com/ceph/ceph/pull/6278>`_, Xinze Chi)
* osdc: Fix race condition with tick_event and shutdown (`issue#14256 <http://tracker.ceph.com/issues/14256>`_, `pr#7151 <http://github.com/ceph/ceph/pull/7151>`_, Adam C. Emerson)
* osd: change mutex to spinlock to optimize thread context switch. (`pr#6492 <http://github.com/ceph/ceph/pull/6492>`_, Xiaowei Chen)
* osd: check do_shutdown before do_restart (`pr#6547 <http://github.com/ceph/ceph/pull/6547>`_, Xiaoxi Chen)
* osd: check health state before pre_booting (`issue#14181 <http://tracker.ceph.com/issues/14181>`_, `pr#7053 <http://github.com/ceph/ceph/pull/7053>`_, Xiaoxi Chen)
* osd: disable filestore_xfs_extsize by default (`issue#14397 <http://tracker.ceph.com/issues/14397>`_, `pr#7265 <http://github.com/ceph/ceph/pull/7265>`_, Ken Dreyer)
* osd: do not keep ref of old osdmap in pg (`issue#13990 <http://tracker.ceph.com/issues/13990>`_, `pr#7007 <http://github.com/ceph/ceph/pull/7007>`_, Kefu Chai)
* osd: don't do random deep scrubs for user initiated scrubs (`pr#6673 <http://github.com/ceph/ceph/pull/6673>`_, David Zafman)
* osd: enable perfcounters on sharded work queue mutexes (`pr#6455 <http://github.com/ceph/ceph/pull/6455>`_, Jacek J. Łakis)
* osd: ensure new osdmaps commit before publishing them to pgs (`issue#15073 <http://tracker.ceph.com/issues/15073>`_, `pr#8096 <http://github.com/ceph/ceph/pull/8096>`_, Sage Weil)
* osd: filestore: FALLOC_FL_PUNCH_HOLE must be used with FALLOC_FL_KEEP_SIZE (`pr#7768 <http://github.com/ceph/ceph/pull/7768>`_, xinxin shu)
* osd: filestore: fast abort if statfs encounters ENOENT (`pr#7703 <http://github.com/ceph/ceph/pull/7703>`_, xie xingguo)
* osd: FileStore: fix initialization order for m_disable_wbthrottle (`pr#8067 <http://github.com/ceph/ceph/pull/8067>`_, Samuel Just)
* osd: filestore: fix race condition with split vs collection_move_rename and long object names (`issue#14766 <http://tracker.ceph.com/issues/14766>`_, `pr#8136 <http://github.com/ceph/ceph/pull/8136>`_, Samuel Just)
* osd: filestore: fix result code overwritten for clone (`issue#14817 <http://tracker.ceph.com/issues/14817>`_, `issue#14827 <http://tracker.ceph.com/issues/14827>`_, `pr#7711 <http://github.com/ceph/ceph/pull/7711>`_, xie xingguo)
* osd: filestore: fix wrong scope of result code for error cases during mkfs (`issue#14814 <http://tracker.ceph.com/issues/14814>`_, `pr#7704 <http://github.com/ceph/ceph/pull/7704>`_, xie xingguo)
* osd: filestore: fix wrong scope of result code for error cases during mount (`issue#14815 <http://tracker.ceph.com/issues/14815>`_, `pr#7707 <http://github.com/ceph/ceph/pull/7707>`_, xie xingguo)
* osd: fix epoch check in handle_pg_create (`pr#8382 <http://github.com/ceph/ceph/pull/8382>`_, Samuel Just)
* osd: fixes for several cases where op result code was not checked or set (`issue#13566 <http://tracker.ceph.com/issues/13566>`_, `pr#6347 <http://github.com/ceph/ceph/pull/6347>`_, xie xingguo)
* osd: fix reactivate (check OSDSuperblock in mkfs() when we already have the superblock) (`issue#13586 <http://tracker.ceph.com/issues/13586>`_, `pr#6385 <http://github.com/ceph/ceph/pull/6385>`_, Vicente Cheng)
* osd: fix unnecessary object promotion when deleting from cache pool (`issue#13894 <http://tracker.ceph.com/issues/13894>`_, `pr#7537 <http://github.com/ceph/ceph/pull/7537>`_, Igor Fedotov)
* osd: os: skip checking pg_meta object existance in FileStore (`pr#6870 <http://github.com/ceph/ceph/pull/6870>`_, Ning Yao)
* osd: partial revert of "ReplicatedPG: result code not correctly set in some cases." (`issue#13796 <http://tracker.ceph.com/issues/13796>`_, `pr#6622 <http://github.com/ceph/ceph/pull/6622>`_, Sage Weil)
* osd/PG: set epoch_created and parent_split_bits for child pg (`issue#15426 <http://tracker.ceph.com/issues/15426>`_, `pr#8552 <http://github.com/ceph/ceph/pull/8552>`_, Kefu Chai)
* osd: pool size change triggers new interval (#11771 Samuel Just)
* osd: prepopulate needs_recovery_map when only one peer has missing (#9558 Guang Yang)
* osd: release related sources when scrub is interrupted (`pr#6744 <http://github.com/ceph/ceph/pull/6744>`_, Jianpeng Ma)
* osd: release the message throttle when OpRequest unregistered (`issue#14248 <http://tracker.ceph.com/issues/14248>`_, `pr#7148 <http://github.com/ceph/ceph/pull/7148>`_, Samuel Just)
* osd: remove __SWORD_TYPE dependency (`pr#6262 <http://github.com/ceph/ceph/pull/6262>`_, John Coyle)
* osd: remove up_thru_pending field, which is never used (`pr#7991 <http://github.com/ceph/ceph/pull/7991>`_, xie xingguo)
* osd: reorder bool fields in PGLog struct (`pr#6279 <http://github.com/ceph/ceph/pull/6279>`_, Piotr Dałek)
* osd: Replace snprintf with faster implementation in eversion_t::get_key_name (`pr#7121 <http://github.com/ceph/ceph/pull/7121>`_, Evgeniy Firsov)
* osd/ReplicatedPG: be more careful about calling publish_stats_to_osd() (`issue#14962 <http://tracker.ceph.com/issues/14962>`_, `pr#8039 <http://github.com/ceph/ceph/pull/8039>`_, Greg Farnum)
* osd: replicatedpg: break out loop if we encounter fatal error during do_pg_op() (`issue#14922 <http://tracker.ceph.com/issues/14922>`_, `pr#7844 <http://github.com/ceph/ceph/pull/7844>`_, xie xingguo)
* osd: ReplicatedPG: clean up unused function (`pr#7211 <http://github.com/ceph/ceph/pull/7211>`_, Xiaowei Chen)
* osd/ReplicatedPG: clear watches on change after applying repops (`issue#15151 <http://tracker.ceph.com/issues/15151>`_, `pr#8163 <http://github.com/ceph/ceph/pull/8163>`_, Sage Weil)
* osd/ReplicatedPG::_rollback_to: update the OMAP flag (`issue#14777 <http://tracker.ceph.com/issues/14777>`_, `pr#8495 <http://github.com/ceph/ceph/pull/8495>`_, Samuel Just)
* osd: repop and lost-unfound overhaul (`pr#7765 <http://github.com/ceph/ceph/pull/7765>`_, Samuel Just)
* osd: reset primary and up_primary when building a new past_interval. (`issue#13471 <http://tracker.ceph.com/issues/13471>`_, `pr#6240 <http://github.com/ceph/ceph/pull/6240>`_, xiexingguo)
* osd: resolve boot vs NOUP set + clear race (`pr#7483 <http://github.com/ceph/ceph/pull/7483>`_, Sage Weil)
* osd: scrub: do not assign value if read error (`pr#6568 <http://github.com/ceph/ceph/pull/6568>`_, Li Wang)
* osd/ScrubStore: remove unused function (`pr#8045 <http://github.com/ceph/ceph/pull/8045>`_, Kefu Chai)
* os/filestore: fix result handling logic of destroy_collection (`pr#7721 <http://github.com/ceph/ceph/pull/7721>`_, xie xingguo)
* os/filestore: force lfn attrs to be written atomically, restructure name length limits (`pr#8496 <http://github.com/ceph/ceph/pull/8496>`_, Samuel Just)
* os/filestore: require offset == length == 0 for full object read; add test (`pr#7957 <http://github.com/ceph/ceph/pull/7957>`_, Jianpeng Ma)
* os/ObjectStore: try_move_rename in transaction append and add coverage to store_test (`issue#15205 <http://tracker.ceph.com/issues/15205>`_, `pr#8359 <http://github.com/ceph/ceph/pull/8359>`_, Samuel Just)
* PG: pg down state blocked by osd.x, lost osd.x cannot solve peering stuck (`issue#13531 <http://tracker.ceph.com/issues/13531>`_, `pr#6317 <http://github.com/ceph/ceph/pull/6317>`_, Xiaowei Chen)
* pybind: port the rbd bindings to Cython (`issue#13115 <http://tracker.ceph.com/issues/13115>`_, `pr#6768 <http://github.com/ceph/ceph/pull/6768>`_, Hector Martin)
* pybind/rados: fix object lifetime issues and other bugs in aio (`pr#7778 <http://github.com/ceph/ceph/pull/7778>`_, Hector Martin)
* pybind/rados: python3 fix (`pr#8331 <http://github.com/ceph/ceph/pull/8331>`_, Mehdi Abaakouk)
* pybind/rados: use __dealloc__ since __del__ is ignored by cython (`pr#7692 <http://github.com/ceph/ceph/pull/7692>`_, Mehdi Abaakouk)
* pybind: remove next() on iterators (`pr#7706 <http://github.com/ceph/ceph/pull/7706>`_, Mehdi Abaakouk)
* pybind: replace __del__ with __dealloc__ for rbd (`pr#7708 <http://github.com/ceph/ceph/pull/7708>`_, Josh Durgin)
* pybind: support ioctx:exec (`pr#6795 <http://github.com/ceph/ceph/pull/6795>`_, Noah Watkins)
* python binding of librados with cython (`pr#7621 <http://github.com/ceph/ceph/pull/7621>`_, Mehdi Abaakouk)
* python: use pip instead of python setup.py (`pr#7605 <http://github.com/ceph/ceph/pull/7605>`_, Loic Dachary)
* qa: add workunit to run ceph_test_rbd_mirror (`pr#8221 <http://github.com/ceph/ceph/pull/8221>`_, Josh Durgin)
* qa: disable rbd/qemu-iotests test case 055 on RHEL/CentOSlibrbd: journal replay should honor inter-event dependencies (`issue#14385 <http://tracker.ceph.com/issues/14385>`_, `pr#7272 <http://github.com/ceph/ceph/pull/7272>`_, Jason Dillaman)
* qa/workunits: merge_diff shouldn't attempt to use striping (`issue#14165 <http://tracker.ceph.com/issues/14165>`_, `pr#7041 <http://github.com/ceph/ceph/pull/7041>`_, Jason Dillaman)
* qa/workunits/rados/test.sh: capture stderr too (`pr#8004 <http://github.com/ceph/ceph/pull/8004>`_, Sage Weil)
* qa/workunits/rados/test.sh: test tmap_migrate (`pr#8114 <http://github.com/ceph/ceph/pull/8114>`_, Sage Weil)
* qa/workunits/rbd: do not use object map during read flag testing (`pr#8104 <http://github.com/ceph/ceph/pull/8104>`_, Jason Dillaman)
* qa/workunits/rbd: new online maintenance op tests (`pr#8216 <http://github.com/ceph/ceph/pull/8216>`_, Jason Dillaman)
* qa/workunits/rbd: rbd-nbd test should use sudo for map/unmap ops (`issue#14221 <http://tracker.ceph.com/issues/14221>`_, `pr#7101 <http://github.com/ceph/ceph/pull/7101>`_, Jason Dillaman)
* qa/workunits/rbd: use POSIX function definition (`issue#15104 <http://tracker.ceph.com/issues/15104>`_, `pr#8068 <http://github.com/ceph/ceph/pull/8068>`_, Nathan Cutler)
* radosgw-admin: Checking the legality of the parameters (`issue#13018 <http://tracker.ceph.com/issues/13018>`_, `pr#5879 <http://github.com/ceph/ceph/pull/5879>`_, Qiankun Zheng)
* radosgw-admin: Create --secret-key alias for --secret (`issue#5821 <http://tracker.ceph.com/issues/5821>`_, `pr#5335 <http://github.com/ceph/ceph/pull/5335>`_, Yuan Zhou)
* radosgw-admin: fix for 'realm pull' (`pr#8404 <http://github.com/ceph/ceph/pull/8404>`_, Casey Bodley)
* radosgw-admin: metadata list user should return an empty list when user pool is empty (`issue#13596 <http://tracker.ceph.com/issues/13596>`_, `pr#6465 <http://github.com/ceph/ceph/pull/6465>`_, Orit Wasserman)
* rbd: add pool name to disambiguate rbd admin socket commands (`pr#6904 <http://github.com/ceph/ceph/pull/6904>`_, wuxiangwei)
* rbd: add RBD pool mirroring configuration API + CLI (`pr#6129 <http://github.com/ceph/ceph/pull/6129>`_, Jason Dillaman)
* rbd: add support for mirror image promotion/demotion/resync (`pr#8138 <http://github.com/ceph/ceph/pull/8138>`_, Jason Dillaman)
* rbd: allow librados to prune the command-line for config overrides (`issue#15250 <http://tracker.ceph.com/issues/15250>`_, `pr#8282 <http://github.com/ceph/ceph/pull/8282>`_, Jason Dillaman)
* rbdmap: systemd support (`issue#13374 <http://tracker.ceph.com/issues/13374>`_, `pr#6479 <http://github.com/ceph/ceph/pull/6479>`_, Boris Ranto)
* rbd: merge_diff test should use new --object-size parameter instead of --order (`issue#14106 <http://tracker.ceph.com/issues/14106>`_, `pr#6972 <http://github.com/ceph/ceph/pull/6972>`_, Na Xie, Jason Dillaman)
* rbd-mirror: asok commands to get status and flush on Mirror and Replayer level (`pr#8235 <http://github.com/ceph/ceph/pull/8235>`_, Mykola Golub)
* rbd-mirror: enabling/disabling pool mirroring should update the mirroring directory (`issue#15217 <http://tracker.ceph.com/issues/15217>`_, `pr#8261 <http://github.com/ceph/ceph/pull/8261>`_, Ricardo Dias)
* rbd-mirror: fix image replay test failures (`pr#8158 <http://github.com/ceph/ceph/pull/8158>`_, Jason Dillaman)
* rbd-mirror: fix long termination due to 30sec wait in main loop (`pr#8185 <http://github.com/ceph/ceph/pull/8185>`_, Mykola Golub)
* rbd-mirror: fix missing increment for iterators (`pr#8352 <http://github.com/ceph/ceph/pull/8352>`_, runsisi)
* rbd-mirror: initial failover / failback support (`pr#8287 <http://github.com/ceph/ceph/pull/8287>`_, Jason Dillaman)
* rbd-mirror: integrate with image sync state machine (`pr#8079 <http://github.com/ceph/ceph/pull/8079>`_, Jason Dillaman)
* rbd-mirror: make remote context respect env and argv config params (`pr#8182 <http://github.com/ceph/ceph/pull/8182>`_, Mykola Golub)
* rbd-mirror: minor fix-ups for initial skeleton implementation (`pr#7958 <http://github.com/ceph/ceph/pull/7958>`_, Mykola Golub)
* rbd-mirror: prevent enabling/disabling an image's mirroring when not in image mode (`issue#15267 <http://tracker.ceph.com/issues/15267>`_, `pr#8332 <http://github.com/ceph/ceph/pull/8332>`_, Ricardo Dias)
* rbd-mirror: remote to local cluster image sync (`pr#7979 <http://github.com/ceph/ceph/pull/7979>`_, Jason Dillaman)
* rbd-mirror: switch fsid over to mirror uuid (`issue#15238 <http://tracker.ceph.com/issues/15238>`_, `pr#8280 <http://github.com/ceph/ceph/pull/8280>`_, Ricardo Dias)
* rbd-mirror: use pool/image names in asok commands (`pr#8159 <http://github.com/ceph/ceph/pull/8159>`_, Mykola Golub)
* rbd-mirror: use the mirroring directory to detect candidate images (`issue#15142 <http://tracker.ceph.com/issues/15142>`_, `pr#8162 <http://github.com/ceph/ceph/pull/8162>`_, Ricardo Dias)
* rbd-mirror: workaround for intermingled lockdep singletons (`pr#8476 <http://github.com/ceph/ceph/pull/8476>`_, Jason Dillaman)
* rbd: must specify both of stripe-unit and stripe-count when specifying stripingv2 feature (`pr#7026 <http://github.com/ceph/ceph/pull/7026>`_, Donghai Xu)
* rbd-nbd: add copyright (`pr#7166 <http://github.com/ceph/ceph/pull/7166>`_, Li Wang)
* rbd-nbd: network block device (NBD) support for RBD (`pr#6657 <http://github.com/ceph/ceph/pull/6657>`_, Yunchuan Wen, Li Wang)
* rbd-nbd: small improvements in logging and forking (`pr#7127 <http://github.com/ceph/ceph/pull/7127>`_, Mykola Golub)
* rbd: output formatter may not be closed upon error (`issue#13711 <http://tracker.ceph.com/issues/13711>`_, `pr#6706 <http://github.com/ceph/ceph/pull/6706>`_, xie xingguo)
* rbd: rbdmap improvements (`pr#6445 <http://github.com/ceph/ceph/pull/6445>`_, Boris Ranto)
* rbd: rbd order will be place in 22, when set to 0 in the config_opt (`issue#14139 <http://tracker.ceph.com/issues/14139>`_, `issue#14047 <http://tracker.ceph.com/issues/14047>`_, `pr#6886 <http://github.com/ceph/ceph/pull/6886>`_, huanwen ren)
* rbd: stripe unit/count set incorrectly from config (`pr#6593 <http://github.com/ceph/ceph/pull/6593>`_, Mykola Golub)
* rbd: striping parameters should support 64bit integers (`pr#6942 <http://github.com/ceph/ceph/pull/6942>`_, Na Xie)
* rbd: support for enabling/disabling mirroring on specific images (`issue#13296 <http://tracker.ceph.com/issues/13296>`_, `pr#8056 <http://github.com/ceph/ceph/pull/8056>`_, Ricardo Dias)
* rgw: accept data only at the first time in response to a request (`pr#8084 <http://github.com/ceph/ceph/pull/8084>`_, sunspot)
* rgw: add a few more help options in admin interface (`pr#8410 <http://github.com/ceph/ceph/pull/8410>`_, Abhishek Lekshmanan)
* rgw: add a method to purge all associate keys when removing a subuser (`issue#12890 <http://tracker.ceph.com/issues/12890>`_, `pr#6002 <http://github.com/ceph/ceph/pull/6002>`_, Sangdi Xu)
* rgw: add a missing cap type (`pr#6774 <http://github.com/ceph/ceph/pull/6774>`_, Yehuda Sadeh)
* rgw: add an inspection to the field of type when assigning user caps (`pr#6051 <http://github.com/ceph/ceph/pull/6051>`_, Kongming Wu)
* rgw: add bucket request payment feature usage statistics integration (`issue#13834 <http://tracker.ceph.com/issues/13834>`_, `pr#6656 <http://github.com/ceph/ceph/pull/6656>`_, Javier M. Mellid)
* rgw: add compat header for TEMP_FAILURE_RETRY (`pr#6294 <http://github.com/ceph/ceph/pull/6294>`_, John Coyle)
* rgw: add default quota config (`pr#6400 <http://github.com/ceph/ceph/pull/6400>`_, Daniel Gryniewicz)
* rgw: add LifeCycle feature (`pr#6331 <http://github.com/ceph/ceph/pull/6331>`_, Ji Chen)
* rgw: adds the radosgw-admin sync status command that gives a human readable status of the sync process at a specific zone (`pr#8030 <http://github.com/ceph/ceph/pull/8030>`_, Yehuda Sadeh)
* rgw: add support for caching of Keystone admin token. (`pr#7630 <http://github.com/ceph/ceph/pull/7630>`_, Radoslaw Zarzynski)
* rgw: add support for "end_marker" parameter for GET on Swift account. (`issue#10682 <http://tracker.ceph.com/issues/10682>`_, `pr#4216 <http://github.com/ceph/ceph/pull/4216>`_, Radoslaw Zarzynski)
* rgw: add support for getting Swift's DLO without manifest handling (`pr#6206 <http://github.com/ceph/ceph/pull/6206>`_, Radoslaw Zarzynski)
* rgw: add support for metadata upload during PUT on Swift container. (`pr#8002 <http://github.com/ceph/ceph/pull/8002>`_, Radoslaw Zarzynski)
* rgw: add support for Static Large Objects of Swift API (`issue#12886 <http://tracker.ceph.com/issues/12886>`_, `issue#13452 <http://tracker.ceph.com/issues/13452>`_, `pr#6643 <http://github.com/ceph/ceph/pull/6643>`_, Yehuda Sadeh, Radoslaw Zarzynski)
* rgw: add support for system requests over Swift API (`pr#7666 <http://github.com/ceph/ceph/pull/7666>`_, Radoslaw Zarzynski)
* rgw: add zone delete to rgw-admin help (`pr#8184 <http://github.com/ceph/ceph/pull/8184>`_, Abhishek Lekshmanan)
* rgw: adjust error code when bucket does not exist in copy operation (`issue#14975 <http://tracker.ceph.com/issues/14975>`_, `pr#7916 <http://github.com/ceph/ceph/pull/7916>`_, Yehuda Sadeh)
* rgw: adjust the request_uri to support absoluteURI of http request (`issue#12917 <http://tracker.ceph.com/issues/12917>`_, `pr#7675 <http://github.com/ceph/ceph/pull/7675>`_, Wenjun Huang)
* rgw: admin api for retrieving usage info (Ji Chen) (`pr#8031 <http://github.com/ceph/ceph/pull/8031>`_, Yehuda Sadeh, Ji Chen)
* rgw_admin: orphans finish segfaults (`pr#6652 <http://github.com/ceph/ceph/pull/6652>`_, Igor Fedotov)
* rgw-admin: remove unused iterator and fix error message (`pr#8507 <http://github.com/ceph/ceph/pull/8507>`_, Karol Mroz)
* rgw:bucket link now set the bucket.instance acl (bug fix) (`issue#11076 <http://tracker.ceph.com/issues/11076>`_, `pr#8037 <http://github.com/ceph/ceph/pull/8037>`_, Zengran Zhang)
* rgw: bucket request payment support (`issue#13427 <http://tracker.ceph.com/issues/13427>`_, `pr#6214 <http://github.com/ceph/ceph/pull/6214>`_, Javier M. Mellid)
* rgw: Bug fix for mtime anomalies in RadosGW and other places (`pr#7328 <http://github.com/ceph/ceph/pull/7328>`_, Adam C. Emerson, Casey Bodley)
* rgw: build-related fixes (`pr#8076 <http://github.com/ceph/ceph/pull/8076>`_, Yehuda Sadeh, Matt Benjamin)
* rgw: calculate payload hash in RGWPutObj_ObjStore only when necessary. (`pr#7869 <http://github.com/ceph/ceph/pull/7869>`_, Radoslaw Zarzynski)
* rgw: check the return value when call fe->run() (`issue#14585 <http://tracker.ceph.com/issues/14585>`_, `pr#7457 <http://github.com/ceph/ceph/pull/7457>`_, wei qiaomiao)
* rgw: clarify the error message when trying to create an existed user (`pr#5938 <http://github.com/ceph/ceph/pull/5938>`_, Zeqiang Zhuang)
* rgw: cleanups to comments and messages (`pr#7633 <http://github.com/ceph/ceph/pull/7633>`_, Pete Zaitcev)
* rgw: do not abort radowgw server when using admin op API with bad parameters (`issue#14190 <http://tracker.ceph.com/issues/14190>`_, `issue#14191 <http://tracker.ceph.com/issues/14191>`_, `pr#7063 <http://github.com/ceph/ceph/pull/7063>`_, Dunrong Huang)
* rgw: Do not send a Content-Type on a '304 Not Modified' response (`issue#15119 <http://tracker.ceph.com/issues/15119>`_, `pr#8253 <http://github.com/ceph/ceph/pull/8253>`_, Wido den Hollander)
* rgw: don't use s->bucket for metadata api path entry (`issue#14549 <http://tracker.ceph.com/issues/14549>`_, `pr#7408 <http://github.com/ceph/ceph/pull/7408>`_, Yehuda Sadeh)
* rgw: Drop a debugging message (`pr#7280 <http://github.com/ceph/ceph/pull/7280>`_, Pete Zaitcev)
* rgw: drop permissions of rgw/civetweb after startup (`issue#13600 <http://tracker.ceph.com/issues/13600>`_, `pr#8019 <http://github.com/ceph/ceph/pull/8019>`_, Karol Mroz)
* rgw: Drop unused usage_exit from rgw_admin.cc (`pr#7632 <http://github.com/ceph/ceph/pull/7632>`_, Pete Zaitcev)
* rgw: extend rgw_extended_http_attrs to affect Swift accounts and containers as well (`pr#5969 <http://github.com/ceph/ceph/pull/5969>`_, Radoslaw Zarzynski)
* rgw: fix objects can not be displayed which object name does not cont… (`issue#12963 <http://tracker.ceph.com/issues/12963>`_, `pr#5738 <http://github.com/ceph/ceph/pull/5738>`_, Weijun Duan)
* rgw: fix wrong etag calculation during POST on S3 bucket. (`issue#11241 <http://tracker.ceph.com/issues/11241>`_, `pr#6030 <http://github.com/ceph/ceph/pull/6030>`_, Radoslaw Zarzynski)
* rgw: fix wrong handling of limit=0 during listing of Swift account. (`issue#14903 <http://tracker.ceph.com/issues/14903>`_, `pr#7821 <http://github.com/ceph/ceph/pull/7821>`_, Radoslaw Zarzynski)
* rgw: modify command stucking when operating radosgw-admin metadata list user (`pr#7032 <http://github.com/ceph/ceph/pull/7032>`_, Peiyang Liu)
* rgw: modify documents and help infos' descriptions to the usage of option date when executing command "log show" (`pr#6080 <http://github.com/ceph/ceph/pull/6080>`_, Kongming Wu)
* rgw: modify the conditional statement in parse_metadata_key method. (`pr#5875 <http://github.com/ceph/ceph/pull/5875>`_, Zengran Zhang)
* rgw: move signal.h dependency from rgw_front.h (`pr#7678 <http://github.com/ceph/ceph/pull/7678>`_, Matt Benjamin)
* rgw: Multipart ListPartsResult ETag quotes (`issue#15334 <http://tracker.ceph.com/issues/15334>`_, `pr#8387 <http://github.com/ceph/ceph/pull/8387>`_, Robin H. Johnson)
* rgw: multiple improvements regarding etag calculation for SLO/DLO of Swift API. (`pr#7764 <http://github.com/ceph/ceph/pull/7764>`_, Radoslaw Zarzynski)
* rgw: multiple Swift API compliance improvements for TempURL (Radoslaw Zarzynsk) (`issue#14806 <http://tracker.ceph.com/issues/14806>`_, `issue#11163 <http://tracker.ceph.com/issues/11163>`_, `pr#7891 <http://github.com/ceph/ceph/pull/7891>`_, Radoslaw Zarzynski)
* rgw: refuse to calculate digest when the s3 secret key is empty (`issue#13133 <http://tracker.ceph.com/issues/13133>`_, `pr#6045 <http://github.com/ceph/ceph/pull/6045>`_, Sangdi Xu)
* rgw: remove duplicated code in RGWRados::get_bucket_info() (`pr#7413 <http://github.com/ceph/ceph/pull/7413>`_, liyankun)
* rgw: remove extra check in RGWGetObj::execute (`issue#12352 <http://tracker.ceph.com/issues/12352>`_, `pr#5262 <http://github.com/ceph/ceph/pull/5262>`_, Javier M. Mellid)
* rgw: remove unused vector (`pr#7990 <http://github.com/ceph/ceph/pull/7990>`_, Na Xie)
* rgw: reset return code in when iterating over the bucket the objects (`issue#14826 <http://tracker.ceph.com/issues/14826>`_, `pr#7803 <http://github.com/ceph/ceph/pull/7803>`_, Orit Wasserman)
* rgw: retry RGWRemoteMetaLog::read_log_info() while master is down (`pr#8453 <http://github.com/ceph/ceph/pull/8453>`_, Casey Bodley)
* rgw: signature mismatch with escaped characters in url query portion (`issue#15358 <http://tracker.ceph.com/issues/15358>`_, `pr#8445 <http://github.com/ceph/ceph/pull/8445>`_, Javier M. Mellid)
* rgw: static large objects (Radoslaw Zarzynski, Yehuda Sadeh)
* rgw: store system object meta in cache when creating it (`issue#14678 <http://tracker.ceph.com/issues/14678>`_, `pr#7615 <http://github.com/ceph/ceph/pull/7615>`_, Yehuda Sadeh)
* rgw: support core file limit for radosgw daemon (`pr#6346 <http://github.com/ceph/ceph/pull/6346>`_, Guang Yang)
* rgw: support for aws authentication v4 (Javier M. Mellid) (`issue#10333 <http://tracker.ceph.com/issues/10333>`_, `pr#7720 <http://github.com/ceph/ceph/pull/7720>`_, Yehuda Sadeh, Javier M. Mellid)
* rgw: support json format for admin policy API (Dunrong Huang) (`issue#14090 <http://tracker.ceph.com/issues/14090>`_, `pr#8036 <http://github.com/ceph/ceph/pull/8036>`_, Yehuda Sadeh, Dunrong Huang)
* rgw: swift use Civetweb ssl can not get right url (`issue#13628 <http://tracker.ceph.com/issues/13628>`_, `pr#6408 <http://github.com/ceph/ceph/pull/6408>`_, Weijun Duan)
* rgw:Use count fn in RGWUserBuckets for quota check (`pr#8294 <http://github.com/ceph/ceph/pull/8294>`_, Abhishek Lekshmanan)
* rgw: use pimpl pattern for RGWPeriodHistory (`pr#7809 <http://github.com/ceph/ceph/pull/7809>`_, Casey Bodley)
* rgw: user quota may not adjust on bucket removal (`issue#14507 <http://tracker.ceph.com/issues/14507>`_, `pr#7586 <http://github.com/ceph/ceph/pull/7586>`_, root)
* selinux: Update policy to grant additional access (`issue#14870 <http://tracker.ceph.com/issues/14870>`_, `pr#7971 <http://github.com/ceph/ceph/pull/7971>`_, Boris Ranto)
* set 128MB tcmalloc cache size by bytes (`pr#8427 <http://github.com/ceph/ceph/pull/8427>`_, Star Guo)
* systemd: set up environment in rbdmap unit file (`issue#14984 <http://tracker.ceph.com/issues/14984>`_, `pr#8222 <http://github.com/ceph/ceph/pull/8222>`_, Nathan Cutler)
* systemd: start/stop/restart ceph services by daemon type (`issue#13497 <http://tracker.ceph.com/issues/13497>`_, `pr#6276 <http://github.com/ceph/ceph/pull/6276>`_, Zhi Zhang)
* sysvinit: allow custom cluster names (`pr#6732 <http://github.com/ceph/ceph/pull/6732>`_, Richard Chan)
* Test exit values on test.sh, fix tier.cc (`issue#15165 <http://tracker.ceph.com/issues/15165>`_, `pr#8266 <http://github.com/ceph/ceph/pull/8266>`_, Samuel Just)
* test: Fix test to run with btrfs which has snap_### dirs (`issue#15347 <http://tracker.ceph.com/issues/15347>`_, `pr#8420 <http://github.com/ceph/ceph/pull/8420>`_, David Zafman)
* test: fixup and improvements for rbd-mirror test (`pr#8090 <http://github.com/ceph/ceph/pull/8090>`_, Mykola Golub)
* test: fix ut test failure caused by lfn change (`issue#15464 <http://tracker.ceph.com/issues/15464>`_, `pr#8544 <http://github.com/ceph/ceph/pull/8544>`_, xie xingguo)
* test: fix valgrind memcheck issues for rbd-mirror test cases (`issue#15354 <http://tracker.ceph.com/issues/15354>`_, `pr#8493 <http://github.com/ceph/ceph/pull/8493>`_, Jason Dillaman)
* test: handle exception thrown from close during rbd lock test (`pr#8124 <http://github.com/ceph/ceph/pull/8124>`_, Jason Dillaman)
* test/libcephfs/flock: add sys/file.h include for flock operations (`pr#6310 <http://github.com/ceph/ceph/pull/6310>`_, John Coyle)
* test/librados/test.cc: clean up EC pools' crush rules too (`issue#13878 <http://tracker.ceph.com/issues/13878>`_, `pr#6788 <http://github.com/ceph/ceph/pull/6788>`_, Loic Dachary, Dan Mick)
* test/librbd/fsx: Use c++11 std::mt19937 generator instead of random_r() (`pr#6332 <http://github.com/ceph/ceph/pull/6332>`_, John Coyle)
* tests: allow docker-test.sh to run under root (`issue#13355 <http://tracker.ceph.com/issues/13355>`_, `pr#6173 <http://github.com/ceph/ceph/pull/6173>`_, Loic Dachary)
* tests: allow object corpus readable test to skip specific incompat instances (`pr#6932 <http://github.com/ceph/ceph/pull/6932>`_, Igor Podoski)
* tests: centos7 needs the Continuous Release (CR) Repository enabled for (`issue#13997 <http://tracker.ceph.com/issues/13997>`_, `pr#6842 <http://github.com/ceph/ceph/pull/6842>`_, Brad Hubbard)
* tests: ceph-disk.sh: should use "readlink -f" instead (`pr#7594 <http://github.com/ceph/ceph/pull/7594>`_, Kefu Chai)
* tests: ceph-disk.sh: use "readlink -f" instead for fullpath (`pr#7606 <http://github.com/ceph/ceph/pull/7606>`_, Kefu Chai)
* tests: fix typo in TestClsRbd.snapshots test case (`issue#13727 <http://tracker.ceph.com/issues/13727>`_, `pr#6504 <http://github.com/ceph/ceph/pull/6504>`_, Jason Dillaman)
* tests: flush op work queue prior to destroying MockImageCtx (`issue#14092 <http://tracker.ceph.com/issues/14092>`_, `pr#7002 <http://github.com/ceph/ceph/pull/7002>`_, Jason Dillaman)
* tests: notification slave needs to wait for master (`issue#13810 <http://tracker.ceph.com/issues/13810>`_, `pr#7220 <http://github.com/ceph/ceph/pull/7220>`_, Jason Dillaman)
* tests: --osd-scrub-load-threshold=2000 for more consistency (`issue#14027 <http://tracker.ceph.com/issues/14027>`_, `pr#6871 <http://github.com/ceph/ceph/pull/6871>`_, Loic Dachary)
* tests: osd-scrub-snaps.sh to display full osd logs on error (`issue#13986 <http://tracker.ceph.com/issues/13986>`_, `pr#6857 <http://github.com/ceph/ceph/pull/6857>`_, Loic Dachary)
* tests: port uniqueness reminder (`pr#6387 <http://github.com/ceph/ceph/pull/6387>`_, Loic Dachary)
* test/system/rados_list_parallel: print oid if rados_write fails (`issue#15240 <http://tracker.ceph.com/issues/15240>`_, `pr#8309 <http://github.com/ceph/ceph/pull/8309>`_, Kefu Chai)
* test/system/\*: use dynamically generated pool name (`issue#15240 <http://tracker.ceph.com/issues/15240>`_, `pr#8318 <http://github.com/ceph/ceph/pull/8318>`_, Kefu Chai)
* test/test_pool_create.sh: fix port (`pr#8361 <http://github.com/ceph/ceph/pull/8361>`_, Sage Weil)
* test/time: no need to abs(uint64_t) for comparing (`pr#7726 <http://github.com/ceph/ceph/pull/7726>`_, Kefu Chai)
* test: update rbd integration cram tests for new default features (`pr#8001 <http://github.com/ceph/ceph/pull/8001>`_, Jason Dillaman)
* test: use sequential journal_tid for object cacher test (`issue#13877 <http://tracker.ceph.com/issues/13877>`_, `pr#6710 <http://github.com/ceph/ceph/pull/6710>`_, Josh Durgin)
* tools: add cephfs-table-tool 'take_inos' (`pr#6655 <http://github.com/ceph/ceph/pull/6655>`_, John Spray)
* tools/cephfs: add tmap_upgrade (`pr#7003 <http://github.com/ceph/ceph/pull/7003>`_, John Spray)
* tools/cephfs: fix overflow writing header to fixed size buffer (#13816) (`pr#6617 <http://github.com/ceph/ceph/pull/6617>`_, John Spray)
* tools/cephfs: fix tmap_upgrade (`issue#15135 <http://tracker.ceph.com/issues/15135>`_, `pr#8128 <http://github.com/ceph/ceph/pull/8128>`_, John Spray)
* tools/rados: reduce "rados put" memory usage by op_size (`pr#7928 <http://github.com/ceph/ceph/pull/7928>`_, Piotr Dałek)
* tools:remove duplicate references (`pr#5917 <http://github.com/ceph/ceph/pull/5917>`_, Bo Cai)
* tools: support printing part cluster map in readable fashion (`issue#13079 <http://tracker.ceph.com/issues/13079>`_, `pr#5921 <http://github.com/ceph/ceph/pull/5921>`_, Bo Cai)
* unittest_compression_zlib: do not assume buffer will be null terminated (`pr#8064 <http://github.com/ceph/ceph/pull/8064>`_, Sage Weil)
* unittest_erasure_code_plugin: fix deadlock (Alpine) (`pr#8314 <http://github.com/ceph/ceph/pull/8314>`_, John Coyle)
* vstart: fix up cmake paths when VSTART_DEST is given (`pr#8363 <http://github.com/ceph/ceph/pull/8363>`_, Casey Bodley)
* vstart: grant full access to Swift testing account (`pr#6239 <http://github.com/ceph/ceph/pull/6239>`_, Yuan Zhou)
* vstart: make -k with optional mon_num. (`pr#8251 <http://github.com/ceph/ceph/pull/8251>`_, Jianpeng Ma)
* vstart: set cephfs root uid/gid to caller (`pr#6255 <http://github.com/ceph/ceph/pull/6255>`_, John Spray)
* vstart.sh: add mstart, mstop, mrun wrappers for running multiple vstart-style test clusters out of src tree (`pr#6901 <http://github.com/ceph/ceph/pull/6901>`_, Yehuda Sadeh)
* vstart.sh: avoid race condition starting rgw via vstart.sh (`issue#14829 <http://tracker.ceph.com/issues/14829>`_, `pr#7727 <http://github.com/ceph/ceph/pull/7727>`_, Javier M. Mellid)
* vstart.sh: silence a harmless msg where btrfs is not found (`pr#7640 <http://github.com/ceph/ceph/pull/7640>`_, Patrick Donnelly)
* xio: add prefix to xio msgr logs (`pr#8148 <http://github.com/ceph/ceph/pull/8148>`_, Roi Dayan)
* xio: fix compilation against latest accelio (`pr#8022 <http://github.com/ceph/ceph/pull/8022>`_, Roi Dayan)
* xio: fix incorrect ip being assigned in case of multiple RDMA ports (`pr#7747 <http://github.com/ceph/ceph/pull/7747>`_, Subramanyam Varanasi)
* xio: remove duplicate assignment of peer addr (`pr#8025 <http://github.com/ceph/ceph/pull/8025>`_, Roi Dayan)
* xio: remove redundant magic methods (`pr#7773 <http://github.com/ceph/ceph/pull/7773>`_, Roi Dayan)
* xio: remove unused variable (`pr#8023 <http://github.com/ceph/ceph/pull/8023>`_, Roi Dayan)
* xio: xio_init needs to be called before any other xio function (`pr#8227 <http://github.com/ceph/ceph/pull/8227>`_, Roi Dayan)
* xxhash: use clone of xxhash.git; add .gitignore (`pr#7986 <http://github.com/ceph/ceph/pull/7986>`_, Sage Weil)
