The HBase Team is pleased to announce the immediate release of HBase 0.94.15.
Download it from your favorite Apache mirror . This release has also been pushed to Apache's maven repository.
All previous 0.92.x and 0.94.x releases can upgraded to 0.94.15 via a rolling upgrade without downtime, intermediary versions can be skipped.
0.94.15 contains 30 fixes and improvements, including some performance fixes:
[HBASE-7886] - [replication] hlog zk node will not be deleted if client roll hlog
[HBASE-9485] - TableOutputCommitter should implement recovery if we don't want jobs to start from 0 on RM restart
[HBASE-9995] - Not stopping ReplicationSink when using custom implementation for the ReplicationSink
[HBASE-10014] - HRegion#doMiniBatchMutation rollbacks the memstore even if there is nothing to rollback.
[HBASE-10015] - Replace intrinsic locking with explicit locks in StoreScanner
[HBASE-10026] - HBaseAdmin#createTable could fail if region splits too fast
[HBASE-10046] - Unmonitored HBase service could accumulate Status objects and OOM
[HBASE-10057] - TestRestoreFlushSnapshotFromClient and TestRestoreSnapshotFromClient fail to finish occasionally
[HBASE-10061] - TableMapReduceUtil.findOrCreateJar calls updateMap(null, ) resulting in thrown NPE
[HBASE-10064] - AggregateClient.validateParameters can throw NPE
[HBASE-10089] - Metrics intern table names cause eventual permgen OOM in 0.94
[HBASE-10111] - Verify that a snapshot is not corrupted before restoring it
[HBASE-10112] - Hbase rest query params for maxVersions and maxValues are not parsed
[HBASE-10117] - Avoid synchronization in HRegionScannerImpl.isFilterDone
[HBASE-10120] - start-hbase.sh doesn't respect --config in non-distributed mode
[HBASE-10179] - HRegionServer underreports readRequestCounts by 1 under certain conditions
[HBASE-10181] - HBaseObjectWritable.readObject catches DoNotRetryIOException and wraps it back in a regular IOException
[HBASE-9931] - Optional setBatch for CopyTable to copy large rows in batches
[HBASE-10001] - Add a coprocessor to help testing the performances without taking into account the i/o
[HBASE-10007] - PerformanceEvaluation: Add sampling and latency collection to randomRead test
[HBASE-10010] - eliminate the put latency spike on the new log file beginning
[HBASE-10048] - Add hlog number metric in regionserver
[HBASE-10049] - Small improvments in region_mover.rb
[HBASE-10093] - Unregister ReplicationSource metric bean when the replication source thread is terminated
[HBASE-9047] - Tool to handle finishing replication when the cluster is offline
[HBASE-10119] - Allow HBase coprocessors to clean up when they fail
[HBASE-9927] - ReplicationLogCleaner#stop() calls HConnectionManager#deleteConnection() unnecessarily
[HBASE-9986] - Incorporate HTTPS support for HBase (0.94 port)
[HBASE-10058] - Test for HBASE-9915 (avoid reading index blocks)
[HBASE-10189] - Intermittent TestReplicationSyncUpTool failure
See also the full release notes .
Thanks to everybody who contributed to this release!
The HBase Team