Fixed Issues in Apache Hive
Review the list of Hive issues that are resolved in Cloudera Runtime 7.1.4.
- Fixed issues running Hive ACID when using Cloudera Runtime with Hive Virtual Warehouses
- Hive ACID is supported for Hive Virtual Warehouses using Cloudera Runtime 7.1.4 in CDP Private Cloud Base.
- CDPD-13345: In jdbcUrl add configuration to create External + purge table by default.
- This issue is now resolved.
- CDPD-16123: Fixed the problem preventing spin off of a separate Hive on Tez session for the Hive Warehouse Connector.
- CDPD-16040: Fixed the problem requiring the
--hs2-urloption insqoop import. - CDPD-12301: Fixed Spark Hive Streaming issue that caused HWC to fail and throw a java.lang.NoSuchMethodException message.
- CDPD-9214: Fixed absence of query progress indicators when Spark-Hive queries through the Hive JDBC connector include a where clause.
- CDPD-9214: Fixed the problem identifying the Tez YARN application ID or IDs run by a Spark session.
- CDPD-9297: Fixed the issue with DESCRIBE statements from SparkSQL to fail.
- CDPD-6761: Fixed how Apache Hive ignores the database location provided by Apache Spark.
- OPSAPS-57198: Fixed the issue causing Erasure coding warnings to appear Cloudera Manager even though you were not using erasure coding.
Compaction issues fixed in this release
- CDH-84872: HIVE-22705: LLAP cache is polluted by query-based compactor
- CDPD-12497: HIVE-22971: Eliminate file rename in insert-only compactor
- CDPD-12576: HIVE-23531: Major CRUD QB compaction failing with ClassCastException when vectorization off
- CDPD-12577: HIVE-23703: Major QB compaction with multiple FileSinkOperators results in data loss and one
- CDPD-13000: HIVE-23555: Cancel compaction jobs when hive.compactor.worker.timeout is reached
- CDPD-13000: HIVE-23591: When Worker fails to connect to metastore it should wait before retrying
- CDPD-14959: HIVE-23825: Create a flag to turn off _orc_acid_version file creation
- CDPD-15116: HIVE-23853: CRUD based compaction also should update ACID file version metadata
- CDPD-15884: HIVE-23966: Minor query-based compaction always results in delta directories with minWriteId=1
- CDPD-15955: HIVE-23929: Run compaction as partition owner
- CDPD-15956: HIVE-22255: Hive doesn’t trigger Major Compaction automatically if table contains only base files
- CDPD-16031: HIVE-24001: Don’t cache MapWork in tez/ObjectCache during query-based compaction
- CDPD-16132: HIVE-24024: Improve logging around CompactionTxnHandler
- CDPD-16149: HIVE-24015: Disable query-based compaction on MR execution engine
- CDPD-16183: HIVE-24023: Hive parquet reader can’t read files with length=0
- CDPD-16298: HIVE-23763: Query based minor compaction produces wrong files when rows with different buckets Ids are processed by the same FileSinkOperator
- CDPD-16471: HIVE-24089: Run QB compaction as table directory user with impersonation
- CDPD-16578: HIVE-24096: Abort failed compaction’s txn on TException or IOException
Hive snapshot isolation issues fixed in this release
- CDPD-17113: HIVE-23032: Add batching in Lock generation
- CDPD-17114: HIVE-22750: Consolidate LockType naming
- CDPD-17108: HIVE-23349: ACID: Concurrent MERGE INSERT operations produce duplicates
- CDPD-17112: HIVE-24000: Put exclusive MERGE INSERT under the feature flag
- CDPD-17109: HIVE-23503: ValidTxnManager doesn’t consider txns opened and committed between snapshot generation and locking when evaluating ValidTxnListState
- CDPD-11683: HIVE-23311: Fix ValidTxnManager regression
- CDPD-16308: HIVE-24125: Incorrect transaction snapshot invalidation with unnecessary writeset check for exclusive operations
