# v0.39

11/27/2011



* Bug #213: non-idempotent transactions (clone) under ext3 may not replay correct result
* Feature #1007: qa: osd failure and cluster recovery test(s)
* Bug #1351: rados bench should report errors
* Feature #1387: teuthology-nuke: don't fail on down nodes
* Bug #1432: libvirt: fix definition for rbd params/sources/etc
* Feature #1487: config: {cluster,public}_subnets
* Feature #1516: openstack: single node dev environment
* Bug #1577: rados.py: Snap.get_timestamp does not work
* Bug #1588: blogbench on kclient possibly made machine die
* Bug #1632: osd: crash in dequeue_op
* Bug #1633: osd crash in CryptoKey::decrypt
* Feature #1646: mon: catch up on committed items before attempting to join quorum
* Feature #1647: mon: robust bootstrap
* Feature #1653: librados: python binding nose tests
* Bug #1657: teuthology: testrados failed to find conf
* Bug #1665: osd: last_peering_reset incorrect on stray?
* Bug #1670: osd: crash in update_heartbeat_peers
* Bug #1676: stats mismatch during snaps workunit
* Bug #1684: mon: crash in CryptoKey::encrypt
* Bug #1694: monitor crash: FAILED assert(get_max_osd() >= crush.get_max_devices())
* Bug #1695: wrong path to ceph's libs / bash scripts in /etc/init.d/ceph
* Bug #1700: osd: invalid perfcounter usage
* Bug #1708: mon/PGMonitor.cc: 218: FAILED assert(paxos->get_version() + 1 == pending_inc.version)
* Bug #1723: timeouts during ffsb
* Bug #1724: timeout during tiobench test
* Bug #1727: osd: failed assert(pending_ops > 0) in dequeue_op
* Bug #1730: mysterious compilation error
* Bug #1731: PAXOS assert(begin->last_committed == last_committed)
* Bug #1732: osdmap assert fail during rados bench
* Bug #1733: rados bench duration can be ignored
* Bug #1742: qa: s3-tests failed 100-continue test on sepia
* Bug #1743: teuthology: not exiting with error when ceph-fuse shutdown fails
* Bug #1751: Copy in CEPH too slow