# v0.61 - Cuttlefish

03/31/2013



* Fix #1857: osd: reimplement shutdown()
* Bug #2545: init-ceph: stops if one instance fails to start
* Feature #4107: Usage quota for rados pools
* Feature #4267: osd: mark self immediately down on orderly (SIGINT, SIGTERM) shutdown
* Feature #4299: crush: command to set additional parents for an interior node or leaf
* Feature #4313: mon: generic key/value service
* Fix #4435: backfill on full osd is not being prevented
* Feature #4490: rados per-pool quota
* Feature #4510: OSD scrub/repair should handle read error and truncate reasonably
* Bug #4527: journal Unable to read past sequence 0 but header indicates the journal has committed up through 1437, journal is corrupt
* Bug #4579: kclient + ffsb workload makes osds mark themselves down
* Bug #4591: test/test_stress_watch.cc:84: Failure
* Bug #4602: osd/ReplicatedPG.cc: 6487: FAILED assert(latest->is_update())
* Bug #4653: mon: creation of cluster with many pgs fails within a reasonble time period
* Bug #4675: mon: pg creations don't get queued on mon startup
* Bug #4686: corrupt or missing osdmap on load_pgs
* Bug #4687: ceph_test_filejournal failing
* Bug #4699: osd: crash when looking at a map changing pool size from 0 to 2
* Bug #4701: osd logs are flooded during the upgrade process
* Bug #4747: Upgrade monitors from argonaut->bobtail->next fails w/"Existing store has not been converted to 0.52 format"
* Bug #4758: monitor: going through all incrementals on startup
* Bug #4765: monitor: sets global version feature but upgrades might not actually have any gv data to look at
* Bug #4778: scrub clears inconsistent flag set by deep scrub
* Bug #4784: Two Monitors Concurrently Reporting as Leaders
* Bug #4793: mon/Monitor.cc: 1126: FAILED assert(!(sync_role & SYNC_ROLE_REQUESTER))
* Bug #4815: mon: leveldb grows quickly and without bound
* Bug #4837: mon: FAILED assert(!(sync_role & SYNC_ROLE_REQUESTER))
* Bug #4851: leveldb: hang on leveldb::DBImpl::MakeRoomForWrite(bool) 
* Bug #4858: mon: doesn't necessarily call reset() during an election cycle
* Bug #4898: ceph-deploy osd create hostname fails with traceback (should catch error)
* Bug #4951: Ceph init script makes assumptions about osd data location
* Bug #4952: ceph-create-keys talked to mon, got back non-JSON
* Bug #5078: Debian missing sudo results in unclear error