| Hortonworks Bug ID | Apache JIRA | Component | Summary |
|---|---|---|---|
| BUG-28507 | Ranger | Add timeout for connection settings | |
| BUG-33182 | HIVE-11123 HIVE-11628 | Hive | Oracle 12:direct sql is failing while running test_HIVE_8295 |
| BUG-33843 | Tez | Log warnings or fail early if there are too many tasks for a job | |
| BUG-33845 | Tez | Backport TEZ-1796 to Baikal, Champlain | |
| BUG-36322 | Tez | Sometimes Tez Client keeps on showing old status even if application is finished | |
| BUG-36587 | HBASE-12366 | HBase | Inclusion of HBASE-12366 in HDP 2.2-Next |
| BUG-36601 | HBase | Backport scanner keyvalue size / partial results related fixes to 2.2 | |
| BUG-36823 | HBase | [2.2] Backports of fixes to region replicas for bloomberg | |
| BUG-38391 | Ranger | Ranger setup in Ambari does not respect custom service users | |
| BUG-39321 | Falcon | falcon regression test for tuture tests .ELExp_FutureAndLatestTest, lastestTest keeps on failing on both 2.1 and 2.2-maint lines saying endtime of feed before startime of feed | |
| BUG-40259 | Ranger |
Issue: Upgrade to HDP 2.2.6 breaks usersync. Details: Ranger in HDP 2.2.6 introduced
authentication between usersync and the admin REST API. If you upgrade to HDP 2.2.6,
there will be a previous version of
Workaround:
| |
| BUG-41828 | Hive | Hiveserver2 failing with OOM PermGen when using temporary functions due to Hadoop reflections utils caching classloaders | |
| BUG-42084 | Hive | Prevent misconfiguration when StorageBasedAuthorization is set on hive.security.authorization.manager | |
| BUG-42151 | Hive | HS2 JVM settings | |
| BUG-42203 | Hive | Hiveserver2 failing with OOM PermGen when using temporary functions due to DataNucleus caching classloaders | |
| BUG-42263 | HDFS | Allow reconfiguration without restart for parameters where applicable. | |
| BUG-42275 | HDFS | A new healthcheck and heartbeat port | |
| BUG-42498 | HDFS | Block received messages put heavy load on namenode | |
| BUG-42674 | Hive | HS2 concurr tests intermittently fail via http mode throwing org.apache.hive.service.auth.HttpAuthenticationException: Kerberos authentication failed | |
| BUG-42698 | HADOOP-12318 | Hadoop Common | BUG-42808 [2.2] Expose underlying LDAP exceptions in SaslPlainServer (HADOOP-12318) |
| BUG-42699 | HADOOP-12324 | Hadoop Common | BUG-42808 [2.2] Better exception reporting in SaslPlainServer (HADOOP-12324) |
| BUG-42709 | HDFS-7278 | HDFS | BUG-42808 [2.2] Add a command that allows sysadmins to manually trigger full block reports from a DN (HDFS-7278) |
| BUG-42710 | HDFS-7433 | HDFS | BUG-42808 [2.2] Optimize performance of DatanodeManager's node map (HDFS-7433) |
| BUG-42713 | YARN | Timeline server OOM | |
| BUG-42721 | HIVE-11407 | Hive | JDBC DatabaseMetaData.getTables with large no of tables call leads to HS2 OOM |
| BUG-42725 | HIVE-10925 | Hive | HIVE-10925: Non-static threadlocals in metastore code can potentially cause memory leak |
| BUG-42790 | Hive | BUG-42658 Gen Plan Changes to support multi col in clause | |
| BUG-42802 | HDFS-7923 HDFS-8930 | HDFS | BUG-42808 [2.2] HDFS-7923: The DataNodes should rate-limit their full block reports by asking the NN on heartbeat messages |
| BUG-42808 | HDFS | [2.2] Backport fixes to HDP 2.2-october-next | |
| BUG-42860 | Ranger | EAR-1858 - Ranger UserSync seems to be limited to pulling a up to 100 users from Active Directory. | |
| BUG-42890 | HBase | Configuration for bulkload needs to be updated | |
| BUG-42916 | HIVE-11606 | Hive | (2.2) - bucket mapjoin error "capacity must be power of two" when #rows < #buckets |
| BUG-42937 | HDFS-8864 | HDFS | BUG-42808 [2.2] Padding is needed in remaining space check (HDFS-8864) |
| BUG-42938 | HDFS-8871 | HDFS | BUG-42808 [2.2] Decommissioning of a node with a failed volume may not start (HDFS-8871) |
| BUG-42939 | HDFS-8870 | HDFS | BUG-42808 [2.2] Lease is leaked on write failure (HDFS-8870) |
| BUG-43023 | HDFS-5215 | HDFS | BUG-42808 [2.2] dfs.datanode.du.reserved is not considered while computing available space (HDFS-5215) |
| BUG-43027 | HBase | Have metrics for blockcache take into account replicas | |
| BUG-43029 | Hive | Explain plan is missing PPD | |
| BUG-43032 | Hive | (2.2) Backport HIVE-11502 (Map side aggregation is extremely slow) | |
| BUG-43069 | Hue | CLONE - Hue does not respect Support dfs.umaskmode, fs.permissions.umask-mode when creating files or folders | |
| BUG-43070 | Hue | CLONE - Select has a 1024 fd limit. If the FD number goes above 1024 for some reason (long running process), then all apps relying on thrift will fail. | |
| BUG-43071 | HIVE-9223 | Hive, Hue | CLONE - HS2/hue/tez: multiple concurrent queries not allowed in single tez session |
| BUG-43082 | YARN | BUG-42343 [ATS V1.5] backport YARN-3942: Timeline store to read events from HDFS (ATS V1.5 patch)into 2.2-maint | |
| BUG-43110 | Hive | NPE explain in vectorization with empty tables | |
| BUG-43163 | Hive, Oozie, Tez | In Tez, fix the same issue of MR that spawn task doesn't take files in distributed cache into classpath. | |
| BUG-43176 | Hive | Backport HIVE-5277 (HBase handler skips rows with null valued first cells when only row key is selected) | |
| BUG-43215 | Tez | Fix Standalone Tez UI to work with RM HA | |
| BUG-43221 | Tez | ClassNotFound in InputInitializer causes AM to crash | |
| BUG-43224 | HDFS-8155 | HDFS | BUG-42808 [2.2] Support OAuth2 in WebHDFS (HDFS-8155) |
| BUG-43226 | HDFS-8435 | HDFS | BUG-42808 [2.2] Support CreateFlag in WebHdfs (HDFS-8435) |
| BUG-43228 | HDFS-8885 | HDFS | BUG-42808 [2.2] ByteRangeInputStream used in webhdfs does not override available() (HDFS-8885) |
| BUG-43229 | HADOOP-12358 | HDFS | BUG-42808 [2.2] FSShell should prompt before deleting directories bigger than a configured size. (HADOOP-12358) |
| BUG-43238 | HIVE-10528 | Hive | (2.2) - Hiveserver2 in HTTP mode is not applying auth_to_local rules |
| BUG-43246 | Ranger | (2.2) Ranger admin should redirect back to login page when login cookies are no longer valid | |
| BUG-43266 | Tez | CLONE[2.2] - When you kill a job in the middle of running it, it Diagnostics shows 'killed/failed due to:null". | |
| BUG-43273 | HDFS | HDFS limits: Namenode goes down due to one bad app doing lots of calls | |
| BUG-43278 | Hive | Password information shows up as clear text when pulling Tez configs | |
| BUG-43281 | HDFS | NN web UI should display storage used for rollback during upgrade | |
| BUG-43297 | YARN | Jobclient prints End of File Exception if RM failover when application is running | |
| BUG-43305 | HDFS-8809 | HDFS | BUG-42808 [2.2] HDFS fsck reports under construction blocks as "CORRUPT" (HDFS-8809) |
| BUG-43307 | Hive | select query on json formatted table with map type column fails | |
| BUG-43330 | Tez, YARN | Functional standalone UI ( hosted in ATS ) for Tez | |
| BUG-43331 | YARN | BUG-43330 [YARN-4009] CORS support for RM webservices | |
| BUG-43335 | Ranger | Processing done by Audit Shutdown hooks can confuse someone looking at logs to think that shutdown of a service is held up due to Ranger plugin | |
| BUG-43341 | HDFS-8965 | HDFS | BUG-42808 [2.2] Harden edit log reading code against out of memory errors (HDFS-8965) |
| BUG-43350 | Ranger | Ranger install fails as - ojdbc.jar missing in classpath in ambari2.1.0 | |
| BUG-43403 | HIVE-11581 | Hive | (2.2) - HS2 url parameters should be stored in zookeeper |
| BUG-43411 | HDP / Stack | Need a central doc for calling out all the steps for applying a hot-fix for each and every component | |
| BUG-43416 | YARN-4107 | YARN | [YARN-4107]Both RM becomes Active if all zookeepers can not connect to active RM |
| BUG-43433 | RANGER-629 | Hive | [Ranger][hive] security issue with hive & ranger for import table command |
| BUG-43465 | Hive | (2.2) - Hive "SLF4J: Class path contains multiple SLF4J bindings." error | |
| BUG-43466 | HBase | (2.2) - Cherry-pick HBASE-14302 TableSnapshotInputFormat should not create back references when restoring snapshot | |
| BUG-43468 | HBase | (2.2) - Integrate 'HBASE-14313 After a Connection sees ConnectionClosingException it never recovers' | |
| BUG-43470 | HDFS | CLONE(2.2) - HDFS: Track DFSClient name in hdfs-audit.log | |
| BUG-43493 | YARN | Active RM can not be started if 2/3 ZKs are killed | |
| BUG-43520 | Tez, YARN | [ATS v1.5] Events for APP attemtps published to ATS 1.5 and ATS 1.0 differs for TEZ | |
| BUG-43543 | HBase | Integrate HBASE-14309 'Allow load balancer to operate when there is region in transition by adding force flag' | |
| BUG-43583 | SLIDER-931 | HBase, Slider | test_runintegrationtestzkandfspermissions failed on HBase-Slider |
| BUG-43588 | Tez | Auto clean tez history log when meeting hdfs limit of 1048576 (can't start hive cli, etc) | |
| BUG-43591 | TEZ-2781 | Tez | Unhelpful error message on Hive console from a failing Hive on Tez query |
| BUG-43592 | Hive, Tez | Tez queries stopped running - ApplicationNotFoundException | |
| BUG-43596 | Tez | [TEZ-2064] ApplicationNotFound exception should be translated to SessionNotRunning | |
| BUG-43598 | HDFS | BUG-42808 [2.2] computeReplicationWork does not record all failure paths | |
| BUG-43603 | Hive | hive does not translate ${system:java.io.tmpdir}/${system:user.name}/operation_logs properly | |
| BUG-43604 | Hive | Exception in FileSink when num_bucket > num_reducer | |
| BUG-43609 | Hive | Adjust compression buffer size to avoid creation of too many small stripes | |
| BUG-43625 | RANGER-631 | Ranger | audit log is not present for deleting user |
| BUG-43658 | HDFS-8995 | HDFS | BUG-42808 [2.2] Flaw in registration bookeeping can make DN die on reconnect (HDFS-8995) |
| BUG-43660 | Hive, Hive View, Tez, YARN | [Umbrella] ATS v 1.5 | |
| BUG-43681 | RANGER-635 | Ranger | delete an user and then add a new user with same id but with diffenent role creates user with old role |
| BUG-43698 | Hive |
There are some issues that can arise while upgrading older Hive Metastore versions, in the case that the Metastore tables have been created using datanucleus.autoCreateSchema rather than using the Hive Metastore creation scipts. Make a backup of your Metastore database before attempting any changes It can be the case that autoCreateSchema may not have created all of the Metastore tables. If there are some expected Metastore tables missing, the Metastore upgrade scripts will fail. Workaround: One way to fix this is to look at the Metastore creation script to the current version of your Hive Metastore, and determine which tables are missing in your Metastore compared to the tables that are being created in the Metastore creation script. Then create a new script containing the commands in the creation script relevant to the Metastore tables that are missing in your Hive Metastore, and run this script to ensure that your Metastore has all of the necessary tables. Some auto-created columns may have a different name - for example the COMMENT column in COLUMNS_V2 might instead have the name FCOMMENT. The upgrade scripts will also fail in this case as it was expecting the column names to match the names from the creation scripts. Workaround: In this case you can try to rename the column name of the Metastore column, to match the column names as they appear in the Metastore creation script. | |
| BUG-43699 | HDFS, Tez | CLONE(2.2) - Log a useful error message when the summary stream cannot be closed when shutting down an AM | |
| BUG-43738 | Hive | Ensure Hive does not propagate credentials file property to tasks, plugins | |
| BUG-43756 | Phoenix | Data not returned correctly for query on local index columns | |
| BUG-43766 | HDFS-8180 | HDFS | BUG-42808 [2.2] AbstractFileSystem Implementation for WebHdfs (HDFS-8180) |
| BUG-43773 | Spark |
Issue: Spark: Fails to find table Details: Spark + Kerberos + yarn-cluster mode + remote hive meta store for SparkSQL does not work. Workaround: Try this with Spark + Kerberos + yarn-client mode for SparkSQL. | |
| BUG-43789 | HDFS-6945 | HDFS | BUG-42808 [2.2] BlockManager should remove a block from excessReplicateMap and decrement ExcessBlocks metric when the block is removed (HDFS-6945) |
| BUG-43799 | HDFS-8542 | HDFS | BUG-42808 [2.2] WebHDFS getHomeDirectory behavior does not match specification (HDFS-8542) |
| BUG-43801 | HDFS-8939 | HDFS | BUG-42808 [2.2] Test(S)WebHdfsFileContextMainOperations failing on branch-2 (HDFS-8939) |
| BUG-43895 | HDFS-8797 | HDFS | BUG-42808 [2.2] WebHdfsFileSystem creates too many connections for pread (HDFS-8797) |
| BUG-43966 | YARN-4105 | YARN | (2.2) - [YARN-4105] Capacity Scheduler headroom for DRF is wrong |
| BUG-43970 | Hive | smb join query generates empty result in mr mode | |
| BUG-43973 | Ranger | Error in test_security_agent_admin.py: "E AttributeError: 'NoneType' object has no attribute 'endswith'" | |
| BUG-43979 | YARN | RM UI prints ZK RM connection state="CONNECTED" , if >50% ZKs are disconnected from active RM | |
| BUG-43980 | YARN | [UI usability] 'ResourceManager HA zookeeper connection state' should be highlighted if connection between ZK-RM is lost | |
| BUG-43986 | HBase | Integrate HBASE-12635 'Delete acl notify znode of table after the table is deleted' | |
| BUG-44026 | HDFS | [2.2-maint] Backport HDFS-9019 Adding informative message to sticky bit permission denied exception | |
| BUG-44039 | Ranger | (2.2) Data consistency across API and UI to include PUT | |
| BUG-44055 | ACCUMULO-3967 | Accumulo | (2.2) Backport "ACCUMULO-3967 bulk import loses records when loading pre-split table" |
| BUG-44108 | Flume | Flume agent fails to start after downgrade.Workaround: Downgrade, in this order: 1) clients 2) core. | |
| BUG-50531 | Kafka |
Kafka file system support Issue: Encrypted file systems such as SafenetFS are not supported for Kafka. Index file corruption can occur. For more information, see: Install Kafka. | |
| RMP-3126 | HDFS | Accurate metrics and messages from Balancer | |
| RMP-3532 | HDFS-7858 | HDFS | Improve HA Namenode Failover detection on the client |
| RMP-3847 | HDFS | Provide way to observe HDFS LeaseManager state |
