# v0.24

12/17/2010

stable clustered recovery

* Bug #248: rbdtool import should use fiemap
* Feature #425: trigger osd scrub automatically
* Feature #453: osd: return error (instead of blocking) on lost objects
* Bug #479: ceph/mount crash badly when writing
* Subtask #485: osd: cooperative scrub scheduling
* Subtask #486: osd: make scrub not block writes
* Bug #503: osd: query osds since last_epoch_clean before concluding objects lost?
* Bug #515: osd: recovery isn't completing
* Feature #526: osd: unfound objects rework
* Bug #559: osd: dup requests can ack early
* Feature #567: osd: background scrub frequency, scheduling
* Feature #568: debian: build with --as-needed?
* Bug #574: timer: event cancellation apparently broken
* Bug #575: monmaptool terminates when input file is not a monmap
* Bug #579: OSD::sched_scrub: FAILED assert(pg_map.count(pgid)
* Bug #589: OSD: crash on startup, PG::read_state
* Bug #592: osd: rebind cluster_messenger when wrongly marked down
* Bug #595: Autogen: not a literal
* Bug #598: osd: journal reset in parallel mode acts weird
* Bug #611: OSD: OSDMap::get_cluster_inst
* Bug #612: OSD: Crash during auto scrub
* Tasks #616: radosacl needs a man page
* Tasks #617: cephfs needs a man page
* Bug #621: error building unstable branch, rbd.cc:837: error: no matching function for call to 'min(uint64_t&, off_t)'
* Bug #622: crushtool useless parse error
* Bug #625: make install should create dirs
* Bug #627: replace openssl with crypto++
* Feature #643: filestore: update btrfs ioctl interface for soon-to-be-pushed SNAP_CREATE_V2
* Bug #658: osd: scrub crash