# v0.28

05/06/2011



* Cleanup #813: libceph interface cleanup
* Cleanup #892: osd: remove class distribution infrastructure?
* Bug #910: Multi-MDS Ceph does not pass fsstress
* Tasks #916: fsstress results in failed CDentry assert
* Tasks #921: Snaprealm issue
* Bug #935: ceph client heavy loading in Metadata server
* Feature #958: rbd: map/unmap kernel rbd device via rbd(?) tool
* Bug #963: osd: errors part way through multiple ops leave inconsistent object context
* Bug #1003: mon: health should show % degraded, unfound (and warn appropriately)
* Tasks #1005: xlock is not unpinning during rename across MDSes
* Feature #1008: filestore: handle long object names
* Cleanup #1009: osd: push pg message handling code into ReplicatedPG
* Feature #1010: mon: adjust mon-osd-[near]full-ratio on the fly
* Bug #1011: config observer structure to eliminate SIGHUP requirement
* Feature #1024: rbdtool: list current kernel mappings
* Bug #1033: osd: CephxClientHandler::handle_response
* Bug #1034: librados: Calling connect twice causes a segfault
* Bug #1038: obsync: support --xuser
* Bug #1043: Filestore is broken on object pushes
* Bug #1045: master doesn't pass testsnaps while changing replication
* Bug #1049: obsync: synchronize ACLs
* Bug #1053: rgw XML parsing exploits and flaws
* Bug #1054: RGW segfaults if the Owner field is not set in an ACL
* Tasks #1055: RGW segfaults if the Owner field is not set in an ACL (subtask)
* Bug #1057: obsync: implement --force
* Bug #1062: pg_refactor: active primary is getting MPGLog
* Bug #1066: update mkcephfs man page
* Bug #1073: osd: failed assert: info.last_complete == info.last_update
* Bug #1079: pgs stuck peering or degraded
* Bug #1088: osd: assert(is_up) failed when sending queries