Fixed Issues
Fixed issues represents selected issues that were previously logged via Cloudera Support, but are now addressed in the current release. These issues may have been reported in previous versions within the Known Issues section; meaning they were reported by customers or identified by Cloudera Quality Engineering team.
Incorrect-results
| Bug ID | Apache Jira | Apache Component | Summary |
|---|---|---|---|
| BUG-120118 | CALCITE-2929 | Calcite | Simplification of IS NULL checks are incorrectly assuming that CAST-s are possible |
| BUG-121707 | HDFS-14499 | HDFS | Misleading REM_QUOTA value with snapshot and trash feature enabled for a directory |
| BUG-122333 | HIVE-22476 | Hive | Hive datediff function provided inconsistent results when hive.fetch.task.conversion is set to none |
| BUG-122230 | HIVE-22331 | Hive | unix_timestamp without argument returns timestamp in millisecond instead of second. |
| BUG-122127 | HIVE-22360 | Hive | MultiDelimitSerDe returns wrong results in last column when the loaded file has more columns than those in table schema |
| BUG-122078 | HIVE-21924 | Hive | Split text files even if header/footer exists |
| BUG-121958 | HIVE-21915 | Hive | Hive with TEZ UNION ALL and UDTF results in data loss |
| BUG-121917 | HIVE-22170 | Hive | from_unixtime and unix_timestamp should use user session time zone |
| BUG-121761 | HIVE-22107 | Hive | Correlated subquery producing wrong schema |
| BUG-121748 | HIVE-22099 | Hive | Several date related UDFs can't handle Julian dates properly since HIVE-20007 |
| BUG-121669 | HIVE-22121 | Hive | Turning on hive.tez.bucket.pruning produce wrong results |
| BUG-121634 | HIVE-22120 | Hive | Fix wrong results/ArrayOutOfBound exception in left outer map joins on specific boundary conditions |
| BUG-121905 | N/A | HWC | HWC: Sharing a parent dataframe for more than one queries causes problems in Spark v2.3.x - No need to OR filters |
| BUG-122387 | ZEPPELIN-3701 | Zeppelin | Missing first several '0' and losing digital accuracy in result table |
Other
| Bug ID | Apache JIRA | Apache Component | Summary |
|---|---|---|---|
| BUG-122279 | HADOOP-16640 | Hadoop Common | WASB: Override getCanonicalServiceName() to return full url of WASB filesystem |
| BUG-122067 | N/A | HBase | Add hadoop-aws dependency in ams-hbase |
| BUG-122066 | N/A | HBase | Build hbase shaded jar for AWS dependencies |
| BUG-122521 | HIVE-22948 | Hive | Schematool MergeCatalog |
| BUG-122494 | N/A | Hive | Spark shouldn't write into hive bucketed tables |
| BUG-122492 | N/A | Hive | HMS should return true metadata to client without bucketing capability |
| BUG-122368 | HIVE-22429 | Hive | Migrated clustered tables using bucketing_version 1 on hive 3 uses bucketing_version 2 for inserts |
| BUG-121519 | HIVE-22009 | Hive | CTLV with user specified location is not honoured |
| BUG-122308 | N/A | Knox | Knox Proxy related changes |
| BUG-122275 | RANGER-2538 | Ranger | Ranger policy import calls via knox trusted proxy failing |
| BUG-122206 | RANGER-2531RANGER-2552 | Ranger | Removing a user from a group is not reflected properly in unix based sync. |
| BUG-122189 | RANGER-2597 | Ranger | Allow auditor role user to get details of services and policies from public API |
| BUG-121945 | RANGER-2507RANGER-2517RANGER-2526RANGER-2569RANGER-2574 | Ranger | Support Deny-All-Else Capability |
| BUG-121921 | RANGER-2528RANGER-2580 | Ranger | Import / Export API improvements |
| BUG-122493 | N/A | Spark | Spark client won't send in bucketing capability to HMS |
Performance
| Bug ID | Apache JIRA | Apache Component | Summary |
|---|---|---|---|
| BUG-122153 | HADOOP-16548 | Hadoop Common | ABFS: Config to enable/disable flush operation |
| BUG-122152 | HADOOP-16404 | Hadoop Common | ABFS default blocksize change(256MB from 512MB) |
| BUG-122030 | HDFS-14171HDFS-14366HDFS-14632HDFS-14859 | HDFS | Improvement on namenode startup |
| BUG-121849 | HIVE-19661HIVE-21171HIVE-21182HIVE-22106HIVE-22115HIVE-22161HIVE-22168HIVE-22169HIVE-22204 | Hive | Query performance fixes |
| BUG-121846 | HIVE-22221 | Hive | Llap external client - Need to reduce LlapBaseInputFormat#getSplits() footprint |
| BUG-121837 | HIVE-21177 | Hive | Optimize AcidUtils.getLogicalLength() |
| BUG-121177 | HIVE-20113 | Hive | Shuffle avoidance: Disable 1-1 edges for sorted shuffle |
| BUG-115077 | HIVE-20932 | Hive | Vectorize Druid Storage Handler Reader |
| BUG-121862 | N/A | HWC | Fast serialization in HiveWarehouseDataReaderFactory |
| BUG-122550 | RANGER-2637RANGER-2651 | Ranger | Performance improvements |
| BUG-122226 | TEZ-4075TEZ-4088TEZ-4091 | Tez | Query performance degrades over time and LLAP needs a frequent restart to recover. |
| BUG-121078 | TEZ-4075 | Tez | Tez: Reimplement tez.runtime.transfer.data-via-events.enabled |
Query-failure
| Bug ID | Apache Jira | Apache Component | Summary |
|---|---|---|---|
| BUG-121613 | HADOOP-16247 | HDFS | NPE in FsUrlConnection |
| BUG-122113 | HIVE-22208 | Hive | Column name with reserved keyword is unescaped when query including join on table with mask column is re-written |
| BUG-121967 | HIVE-22231 | Hive | Hive query with big size via knox fails with Broken pipe Write failed |
| BUG-121842 | HIVE-21799 | Hive | NullPointerException in DynamicPartitionPruningOptimization, when join key is on aggregation column |
| BUG-121822 | HIVE-15956 | Hive | StackOverflowError when drop lots of partitions |
| BUG-121682 | HIVE-21280ORC-540 | Hive | Null pointer exception on running compaction against a MM table. |
| BUG-121634 | HIVE-22120 | Hive | Fix wrong results/ArrayOutOfBound exception in left outer map joins on specific boundary conditions |
| BUG-122407 | N/A | HWC | Insert overwrite into managed table shows "does not match the file format" exception. |
| BUG-121727 | N/A | HWC | HWC: Sharing a parent dataframe for more than one queries causes problems in Spark v2.3.x (actual fix) |
| BUG-121781 | SPARK-26307 | Spark | Fix CTAS when INSERT a partitioned table using Hive serde |
| BUG-121408 | SPARK-25271 | Spark | Creating parquet table with all the column null throws exception |
| BUG-121406 | SPARK-26307 | Spark | Fix CTAS when INSERT a partitioned table using Hive serde |
Security
| Bug ID | Apache Jira | Apache Component | Summary |
|---|---|---|---|
| BUG-122199 | HADOOP-16619 | Hadoop Common | Upgrade jackson and jackson-databind to 2.9.10 |
| BUG-122198 | HADOOP-16533 | Hadoop Common | Update jackson-databind to 2.9.9.3 |
| BUG-122197 | HADOOP-16487 | Hadoop Common | Update jackson-databind to 2.9.9.2 |
| BUG-122158 | N/A | Hadoop Common | Upgrade jackson-databind to 2.10.0 |
| BUG-122103 | N/A | HDFS | Fix potential FsImage corruption - CVE-2018-11768 |
| BUG-122094 | HIVE-22273 | Hive | Access check is failed when a temporary directory is removed |
| BUG-122008 | HIVE-22243 | Hive | Align Apache Thrift version to 0.9.3-1 in standalone-metastore as well |
| BUG-121671 | N/A | Hive | Upgrade to guava 28.0-jre in ORC |
| BUG-122085 | N/A | Livy | Upgrade Netty to 4.1.42.Final |
| BUG-122084 | LIVY-695 | Livy | Upgrade JQuery to 3.4.1 and Bootstrap to 3.4.1 |
| BUG-122082 | N/A | Livy | Upgrade Jackson to 2.10.0 |
| BUG-121900 | OOZIE-3543 | Oozie | Upgrade quartz to 2.3.1 |
| BUG-121899 | OOZIE-3544 | Oozie | Upgrade commons-beanutils to 1.9.4 |
| BUG-122139 | N/A | Spark | Upgrade Jackson to 2.10.0 |
| BUG-122131 | SPARK-29445 | Spark | Bump netty-all from 4.1.39.Final to 4.1.42.Final |
| BUG-122130 | N/A | Spark | Upgrade Jackson to 2.10.0 |
| BUG-121389 | SPARK-26895 | Spark | When running spark 2.3 as a proxy user (--proxy-user), SparkSubmit fails to resolve globs owned by target user |
| BUG-121841 | SQOOP-3447 | Sqoop | Remove usage of org.codehaus.jackson and org.json packages |
| BUG-121930 | STORM-3201 | Storm | kafka spout lag on UI needs some cleanup - CVE-2018-11779 |
Stability
| Bug ID | Apache Jira | Apache Component | Summary |
|---|---|---|---|
| BUG-122155 | HADOOP-16578 | Hadoop Common | ABFS: fileSystemExists() should not call container level apis |
| BUG-121929 | HADOOP-16248HADOOP-16278 | Hadoop Common | Instability issues related to Hive Metastore |
| BUG-121515 | HDFS-14042 | Hadoop Common | Fix NPE when PROVIDED storage is missing |
| BUG-121935 | HBASE-22380 | HBase | break circle replication when doing bulkload |
| BUG-122241 | HDFS-14423YARN-9833 | HDFS | Stability fixes |
| BUG-121613 | HADOOP-16247 | HDFS | NPE in FsUrlConnection |
| BUG-122324 | HIVE-22373 | Hive | File Merge tasks fail when containers are reused |
| BUG-122201 | HIVE-22219 | Hive | Bringing a node manager down blocks restart of LLAP service |
| BUG-122106 | HIVE-22275 | Hive | OperationManager.queryIdOperation does not properly clean up multiple queryIds |
| BUG-121959 | HIVE-21866 | Hive | LLAP status service driver may get stuck with wrong Yarn app ID |
| BUG-121623 | HIVE-22113 | Hive | Prevent LLAP shutdown on AMReporter related RuntimeException |
| BUG-121609 | HIVE-22068 | Hive | Return the last event id dumped as repl status to avoid notification event missing error. |
| BUG-121421 | PHOENIX-5188PHOENIX-5289 | Phoenix | NPE reading KeyValue in WAL with Phoenix |
| BUG-122525 | YARN-9968 | YARN | Public Localizer is exiting in NodeManager due to NullPointerException |
| BUG-122252 | YARN-9921 | YARN | Issue in PlacementConstraint when YARN Service AM retries allocation on component failure. |
| BUG-122121 | YARN-9816 | YARN | EntityGroupFSTimelineStore#scanActiveLogs fails when undesired files are present under /ats/active. |
| BUG-122120 | YARN-9744 | YARN | RollingLevelDBTimelineStore.getEntityByTime fails with NPE |
| BUG-121938 | YARN-8015YARN-9209 | YARN | When nodePartition is not set in Placement Constraints, containers are allocated only in default partition |
Supportability
| Bug ID | Apache Jira | Apache Component | Summary |
|---|---|---|---|
| BUG-121578 | HIVE-22129HIVE-22134 | Hive | Hive 3.1 driver includes org.glassfish.jersey.* which can interfere with Cognos |
| BUG-121499 | HIVE-21009 | Hive | LDAP - Specify binddn for ldap-search |
| BUG-121863 | N/A | HWC | HIVE HiveWarehouseSession().HIVE_WAREHOUSE_CONNECTOR class is static and cannot be accessed outside of that class. |
| BUG-122317 | N/A | Oozie | Bigtop-Tomcat Ubuntu package is not being upgraded |
| BUG-122014 | OOZIE-3551 | Oozie | Configure working defaults for Spark action in Oozie |
| BUG-121875 | OOZIE-3542 | Oozie | Handle better old Hdfs implementations in ECPolicyDisabler |
| BUG-121836 | OOZIE-3405 | Oozie | SSH action shows empty error Message and Error code |
| BUG-121783 | N/A | Spark | Set spark.sql.parquet.writeLegacyFormat=true in spark |
Usability
| Bug ID | Apache Jira | Apache Component | Summary |
|---|---|---|---|
| BUG-121956 | HIVE-22241 | Hive | Implement UDF to interpret date/timestamp using its internal representation and Gregorian-Julian hybrid calendar |
| BUG-121535 | HIVE-22110 | Hive | Initialize ReplChangeManager before starting actual dump |
| BUG-122276 | N/A | HWC | HWC does not propagate some confs to hive |
| BUG-121835 | OOZIE-3024 | Oozie | Email action ignores value of content_type attribute when attachment attribute is set |
| BUG-122458 | PHOENIX-5552 | Phoenix | Hive against Phoenix gets 'Expecting "RPAREN", got "L"' in Tez mode |
| BUG-122051 | PHOENIX-5506 | Phoenix | Psql load fails with lower table name |
| BUG-122023 | RANGER-1912 | Ranger | Ranger setup fails with mariadb/mysql when binary logging is enabled |
| BUG-121869 | HIVE-12261HIVE-14080 | Spark | Incorrect Hive version is updated in Database while running spark-sql command. |
| BUG-121686 | YARN-9197 | YARN | NPE in service AM when failed to launch container |
| Technical Service Bulletin | Apache JIRA | Apache Component | Summary |
|---|---|---|---|
| TSB-327 | HDFS-5698 | HDFS | CVE-2018-11768: HDFS FSImage Corruption (potential DoS, file/dir
takeover) In very large clusters, the in-memory format to store the user, group, acl, and extended attributes may exceed the size of the on disk format, causing corruption of fsImage. For the latest update on this issue, see the corresponding Knowledge article: CVE-2018-11768: HDFS FSImage Corruption (potential DoS, file/dir takeover) |
| TSB-450 | HBASE-21000 | HBase |
Default limits for PressureAwareCompactionThroughputController are too low HDP and CDH releases suffer from low compaction throughput limits, which cause storefiles to back up faster than compactions can re-write them. For the latest update on this issue see the corresponding Knowledge article: Cloudera Customer Advisory: Default limits for PressureAwareCompactionThroughputController are too low |