数据仓库平台 Apache Hive 2.1.1 发布了。本次部分更新如下： Sub-task [HIVE-13409] - Fix JDK8 test failures related to COLUMN_STATS_ACCURATE [HIVE-13549] - Remove jdk version specific out files from Hive2 [HIVE-13587] - Set Hive pom to use Hadoop 2.6.1 [HIVE-13593] - HiveServer2: Performance instrumentation for HIVE-12049 (serializing thrift ResultSets in tasks) [HIVE-13723] - Executing join query on type Float using Thrift Serde will result in Float cast to Double error [HIVE-13798] - Fix the unit test failure org.apache.hadoop.hive.cli.TestCliDriver.testCliDriver_ivyDownload [HIVE-13803] - More aggressive inference of transitive predicates for inner joins [HIVE-13860] - Fix more json related JDK8 test failures [HIVE-13901] - Hivemetastore add partitions can be slow depending on filesystems [HIVE-13965] - Empty resultset run into Exception when using Thrift Binary Serde [HIVE-14028] - stats is not updated [HIVE-14039] - HiveServer2: Make the usage of server with JDBC thirft serde enabled, backward compatible for older clients [HIVE-14191] - bump a new api version for ThriftJDBCBinarySerde changes [HIVE-14276] - Update protocol version in TOpenSessionReq and TOpenSessionResp [HIVE-14277] - Disable StatsOptimizer for all ACID tables Bug [HIVE-9423] - HiveServer2: Provide the user with different error messages depending on the Thrift client exception code [HIVE-10022] - Authorization checks for non existent file/directory should not be recursive [HIVE-11402] - HS2 - add an option to disallow parallel query execution within a single Session [HIVE-13191] - DummyTable map joins mix up columns between tables [HIVE-13264] - JDBC driver makes 2 Open Session Calls for every open session [HIVE-13369] - AcidUtils.getAcidState() is not paying attention toValidTxnList when choosing the "best" base file [HIVE-13392] - disable speculative execution for ACID Compactor [HIVE-13539] - HiveHFileOutputFormat searching the wrong directory for HFiles [HIVE-13590] - Kerberized HS2 with LDAP auth enabled fails in multi-domain LDAP case [HIVE-13610] - Hive exec module won't compile with IBM JDK [HIVE-13648] - ORC Schema Evolution doesn't support same type conversion for VARCHAR, CHAR, or DECIMAL when maxLength or precision/scale is different [HIVE-13704] - Don't call DistCp.execute() instead of DistCp.run() [HIVE-13725] - ACID: Streaming API should synchronize calls when multiple threads use the same endpoint [HIVE-13744] - LLAP IO - add complex types support [HIVE-13749] - Memory leak in Hive Metastore [HIVE-13809] - Hybrid Grace Hash Join memory usage estimation didn't take into account the bloom filter size [HIVE-13836] - DbNotifications giving an error = Invalid state. Transaction has already started [HIVE-13874] - Tighten up EOF checking in Fast DeserializeRead classes; display better exception information; add new Unit Tests [HIVE-13932] - Hive SMB Map Join with small set of LIMIT failed with NPE [HIVE-13934] - Configure Tez to make nocondiional task size memory available for the Processor [HIVE-13985] - ORC improvements for reducing the file system calls in task side [HIVE-13991] - Union All on view fail with no valid permission on underneath table [HIVE-13997] - Insert overwrite directory doesn't overwrite existing files [HIVE-14000] - (ORC) Changing a numeric type column of a partitioned table to lower type set values to something other than 'NULL' [HIVE-14003] - queries running against llap hang at times - preemption issues [HIVE-14004] - Minor compaction produces ArrayIndexOutOfBoundsException: 7 in SchemaEvolution.getFileType [HIVE-14012] - some ColumnVector-s are missing ensureSize [HIVE-14022] - left semi join should throw SemanticException if where clause contains columnname from right table [HIVE-14024] - setAllColumns is called incorrectly after some changes [HIVE-14027] - NULL values produced by left outer join do not behave as NULL [HIVE-14031] - cleanup metadataReader in OrcEncodedDataReader [HIVE-14034] - Vectorization may fail with compex OR conditions [HIVE-14038] - miscellaneous acid improvements [HIVE-14041] - llap scripts add hadoop and other libraries from the machine local install to the daemon classpath [HIVE-14045] - (Vectorization) Add missing case for BINARY in VectorizationContext.getNormalizedName method [HIVE-14055] - directSql - getting the number of partitions is broken [HIVE-14059] - Missing license headers for two files [HIVE-14062] - Changes from HIVE-13502 overwritten by HIVE-13566 [HIVE-14071] - HIVE-14014 breaks non-file outputs [HIVE-14072] - QueryIds reused across different queries [HIVE-14073] - update config whiltelist for sql std authorization [HIVE-14076] - Vectorization is not supported for datatype:VOID error while inserting data into specific columns [HIVE-14079] - Remove file, method and line number from pattern layout [HIVE-14083] - ALTER INDEX in Tez causes NullPointerException [HIVE-14090] - JDOExceptions thrown by the Metastore have their full stack trace returned to clients [HIVE-14092] - Kryo exception when deserializing VectorFileSinkOperator [HIVE-14098] - Logging task properties, and environment variables might contain passwords [HIVE-14111] - better concurrency handling for TezSessionState - part I [HIVE-14114] - Ensure RecordWriter in streaming API is using the same UserGroupInformation as StreamingConnection [HIVE-14122] - VectorMapOperator: Missing update to AbstractMapOperator::numRows [HIVE-14132] - Don't fail config validation for removed configs [HIVE-14136] - LLAP ZK SecretManager should resolve _HOST in principal [HIVE-14137] - Hive on Spark throws FileAlreadyExistsException for jobs with multiple empty tables [HIVE-14140] - LLAP: package codec jars [HIVE-14144] - Permanent functions are showing up in show functions, but describe says it doesn't exist [HIVE-14147] - Hive PPD might remove predicates when they are defined as a simple expr e.g. WHERE 'a' [HIVE-14152] - datanucleus.autoStartMechanismMode should set to 'Ignored' to allow rolling downgrade [HIVE-14153] - Beeline: beeline history doesn't work on Hive2 [HIVE-14163] - LLAP: use different kerberized/unkerberized zk paths for registry [HIVE-14172] - LLAP: force evict blocks by size to handle memory fragmentation [HIVE-14173] - NPE was thrown after enabling directsql in the middle of session [HIVE-14175] - Fix creating buckets without scheme information [HIVE-14176] - CBO nesting windowing function within each other when merging Project operators [HIVE-14178] - Hive::needsToCopy should reuse FileUtils::equalsFileSystem [HIVE-14184] - Adding test for limit pushdown in presence of grouping sets [HIVE-14187] - JDOPersistenceManager objects remain cached if MetaStoreClient#close is not called [HIVE-14188] - LLAPIF: wrong user field is used from the token [HIVE-14192] - False positive error due to thrift [HIVE-14195] - HiveMetaStoreClient getFunction() does not throw NoSuchObjectException [HIVE-14197] - LLAP service driver precondition failure should include the values [HIVE-14205] - Hive doesn't support union type with AVRO file format [HIVE-14207] - Strip HiveConf hidden params in webui conf [HIVE-14210] - ExecDriver should call jobclient.close() to trigger cleanup [HIVE-14214] - ORC Schema Evolution and Predicate Push Down do not work together (no rows returned) [HIVE-14215] - Displaying inconsistent CPU usage data with MR execution engine [HIVE-14218] - LLAP: ACL validation fails if the user name is different from principal user name [HIVE-14222] - PTF: Operator initialization does not clean state [HIVE-14223] - beeline should look for jdbc standalone jar in dist/jdbc dir instead of dist/lib [HIVE-14236] - CTAS with UNION ALL puts the wrong stats in Tez [HIVE-14241] - Acid clashes with ConfVars.HIVEFETCHTASKCONVERSION <> "none" [HIVE-14242] - Backport ORC-53 to Hive [HIVE-14244] - bucketmap right outer join query throws ArrayIndexOutOfBoundsException [HIVE-14245] - NoClassDefFoundError when starting LLAP daemon [HIVE-14253] - Fix MinimrCliDriver test failure [HIVE-14262] - Inherit writetype from partition WriteEntity for table WriteEntity [HIVE-14263] - Log message when HS2 query is waiting on compile lock [HIVE-14265] - Partial stats in Join operator may lead to data size estimate of 0 [HIVE-14267] - HS2 open_operations metrics not decremented when an operation gets timed out [HIVE-14268] - INSERT-OVERWRITE is not generating an INSERT event during hive replication 改进 [HIVE-10815] - Let HiveMetaStoreClient Choose MetaStore Randomly [HIVE-13631] - Support index in HBase Metastore [HIVE-13982] - Extensions to RS dedup: execute with different column order and sorting direction if possible [HIVE-14002] - Extend limit propagation to subsequent RS operators [HIVE-14018] - Make IN clause row selectivity estimation customizable [HIVE-14021] - When converting to CNF, fail if the expression exceeds a threshold [HIVE-14057] - Add an option in llapstatus to generate output to a file [HIVE-14080] - hive.metastore.schema.verification should check for schema compatiblity [HIVE-14167] - Use work directories provided by Tez instead of directly using YARN local dirs [HIVE-14213] - Add timeouts for various components in llap status check [HIVE-14228] - Better row count estimates for outer join during physical planning [HIVE-14383] - SparkClientImpl should pass principal and keytab to spark-submit instead of calling kinit explicitely [HIVE-14545] - HiveServer2 with http transport mode spends too much time just creating configs 任务 [HIVE-14202] - Change tez version used to 0.8.4 [HIVE-15248] - Add Apache header license to TestCustomPartitionVertex 测试 [HIVE-14212] - hbase_queries result out of date on branch-2.1 [HIVE-14316] - TestLlapTokenChecker.testCheckPermissions, testGetToken fail [HIVE-14713] - LDAP Authentication Provider should be covered with unit tests 发布主页和更新日志 下载地址 https://hive.apache.org/downloads.html Apache Hive 2.1.1 发布，数据仓库平台下载地址
