We are pleased to announce the release of Cloudera Enterprise 5.4.10 (CDH 5.4.10, Cloudera Manager 5.4.10, and Cloudera Navigator 2.3.10).
This release fixes key bugs and includes the following:
- CDH fixes for the following issues:
- FLUME-2712 - Optional channel errors slows down the Source to Main channel event rate
- FLUME-2886 - Optional Channels can cause OOMs
- HDFS-7553 - Fix the TestDFSUpgradeWithHA due to BindException
- HDFS-9083 - Replication violates block placement policy
- HDFS-9092 - Nfs silently drops overlapping write requests and causes data copying to fail
- HDFS-9406 - FSImage may get corrupted after deleting snapshot
- HDFS-9445 - Datanode may deadlock while handling a bad volume
- MAPREDUCE-6302 - Incorrect headroom can lead to a deadlock between map and reduce allocations
- MAPREDUCE-6460 - TestRMContainerAllocator.testAttemptNotFoundCausesRMCommunicatorException fails
- YARN-2902 - Killing a container that is localizing can orphan resources in the DOWNLOADING state
- YARN-4204 - ConcurrentModificationException in FairSchedulerQueueInfo
- YARN-4347 - Resource manager fails with Null pointer exception
- HBASE-14621 - ReplicationLogCleaner stuck on RS crash
- HBASE-14923 - VerifyReplication should not mask the exception during result comparison
- HBASE-14926 - Hung ThriftServer; no timeout on read from client; if client crashes, worker thread gets stuck reading
- HBASE-15019 - Replication stuck when HDFS is restarted
- HBASE-15031 - Fix merge of MVCC and SequenceID performance regression in branch-1.0
- HBASE-15213 - Fix increment performance regression caused by HBASE-8763 on branch-1.0
- HIVE-11826 - 'hadoop.proxyuser.hive.groups' configuration doesn't prevent unauthorized user to access metastore
- HIVE-12008 - Hive queries failing when using count(*) on column in view
- HIVE-12505 - Insert overwrite in same encrypted zone silently fails to remove some existing files
- HIVE-12566 - Incorrect result returns when using COALESCE in WHERE condition with LEFT JOIN
- HIVE-12784 - Group by SemanticException: Invalid column reference
- HIVE-12795 - Vectorized execution causes ClassCastException
- HIVE-13065 - Hive throws NPE when writing map type data to a HBase backed table
- IMPALA-1702 - Check for duplicate table IDs at the end of analysis (issue not entirely fixed, but now fails gracefully)
- IMPALA-2264 - Implicit casts to integers from decimals with higher precision sometimes allowed
- IMPALA-2473 - Excessive memory usage by scan nodes
- IMPALA-2643 - Nested inline view produces incorrect result when referencing the same column implicitly
- IMPALA-2765 - AnalysisException: operands of type BOOLEAN and TIMESTAMP are not comparable when OUTER JOIN with CASE statement
- IMPALA-2798 - After adding a column to avro table, Impala returns weird result if codegen is enabled.
- IMPALA-3034 - MemTracker leak on PHJ failure to spill
- IMPALA-3093 - ReopenClient() could NULL out 'client_key' causing a crash
- KITE-1114 - Kite CLI json-import HDFS temp file path not multiuser safe
- OOZIE-2413 - Kerberos credentials can expire if the KDC is slow to respond
- SENTRY-835 - Drop table leaves a connection open when using MetastoreListener
- SENTRY-953 - External Partitions which are referenced by more than one table can cause some unexpected behavior with Sentry HDFS sync
- SENTRY-957 - Exceptions in MetastoreCacheInitializer should probably not prevent HMS from starting up
- SENTRY-1002 - PathsUpdate.parsePath(path) will throw an NPE when parsing relative paths
- SENTRY-1044 - Tables with non-HDFS locations breaks HMS startup
- SOLR-8372 - backportCanceled recovery can lead to data loss
- SQOOP-2847 - Sqoop --incremental + missing parent --target-dir reports success with no data
For a full list of upstream JIRAs fixed in CDH 5.4.10, see the issues fixed section of the Release Notes.
- Cloudera Manager fixes for the following issues:
- By default, hive.compute.query.using.stats was enabled. This produced incorrect results for some queries that used stats only. This setting is now disabled by default.
- After security was enabled on a cluster, YARN jobs failed. This occurred because the contents of any previously existing YARN User Cache directory could not be overridden after security was enabled. YARN jobs now complete as expected after security is enabled.
- Diagnostic bundles collected manually included all expected logs, but bundles collected on a schedule did not include role logs. Scheduled diagnostic bundles now include all expected logs.
For full list of issues fixed in Cloudera Manager 5.4.10, see the issues fixed section of the Release Notes.
We look forward to you trying it, using the information below:
As always, we are happy to hear your feedback. Please send your comments and suggestions to the user group or through our community forums. You can also file bugs through our external JIRA projects on issues.cloudera.org.
