Hiveserver2 does not start after installing HDP 2.6.4.0-91 using cloudbreak on AWS

Question

Hiveserver2 does not start after installing HDP 2.6.4.0-91 using cloudbreak on AWS.

Start the hiveserver2 in the Ambari UI and check the contents of /var/log/hive/hiveserver2.log.

Below is the error log.

Any help would be appreciated.

---

Contents of hiveserver2.log

2018-03-08 04:41:53,345 WARN [main-EventThread]: server.HiveServer2 (HiveServer2.java:process(343)) - This instance of HiveServer2 has been removed from the list of server instances available for dynamic service discovery. The last client session has ended - will shutdown now.
2018-03-08 04:41:53,347 INFO [main]: zookeeper.ZooKeeper (ZooKeeper.java:close(684)) - Session: 0x16203aad5af0040 closed
2018-03-08 04:41:53,347 INFO [main]: server.HiveServer2 (HiveServer2.java:removeServerInstanceFromZooKeeper(361)) - Server instance removed from ZooKeeper.
2018-03-08 04:41:53,348 INFO [main-EventThread]: server.HiveServer2 (HiveServer2.java:stop(405)) - Shutting down HiveServer2
2018-03-08 04:41:53,348 INFO [main-EventThread]: server.HiveServer2 (HiveServer2.java:removeServerInstanceFromZooKeeper(361)) - Server instance removed from ZooKeeper.
2018-03-08 04:41:53,348 INFO [main-EventThread]: zookeeper.ClientCnxn (ClientCnxn.java:run(524)) - EventThread shut down
2018-03-08 04:41:53,348 WARN [main]: server.HiveServer2 (HiveServer2.java:startHiveServer2(508)) - Error starting HiveServer2 on attempt 1, will retry in 60 seconds
org.apache.tez.dag.api.SessionNotRunning: TezSession has already shutdown. Application application_1520480101488_0046 failed 2 times due to AM Container for appattempt_1520480101488_0046_000002 exited with exitCode: -1000
For more detailed output, check the application tracking page: http://ip-10-0-91-7.ap-northeast-2.compute.internal:8088/cluster/app/application_1520480101488_0046 Then click on links to logs of each attempt.
Diagnostics: ExitCodeException exitCode=2: tar: Removing leading `/' from member names
tar: Skipping to next header

gzip: /hadoopfs/fs1/yarn/nodemanager/filecache/60_tmp/tmp_tez.tar.gz: invalid compressed data--format violated
tar: Exiting with failure status due to previous errors

Failing this attempt. Failing the application.
at org.apache.tez.client.TezClient.waitTillReady(TezClient.java:699)
at org.apache.hadoop.hive.ql.exec.tez.TezSessionState.open(TezSessionState.java:218)
at org.apache.hadoop.hive.ql.exec.tez.TezSessionState.open(TezSessionState.java:116)
at org.apache.hadoop.hive.ql.exec.tez.TezSessionPoolManager.startPool(TezSessionPoolManager.java:76)
at org.apache.hive.service.server.HiveServer2.startHiveServer2(HiveServer2.java:488)
at org.apache.hive.service.server.HiveServer2.access$700(HiveServer2.java:87)
at org.apache.hive.service.server.HiveServer2$StartOptionExecutor.execute(HiveServer2.java:720)
at org.apache.hive.service.server.HiveServer2.main(HiveServer2.java:593)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)
at org.apache.hadoop.util.RunJar.run(RunJar.java:233)
at org.apache.hadoop.util.RunJar.main(RunJar.java:148)

Answer 1

I had exactly the same issue with HDP on AWS. FYI, In my case the issue was with HDP version 2.6.4.5-2. I'm going to show how I fixed using this version because it is the latest at this time.

As the error log shows the problem is with tez.tar.gz that is corrupted then YARN is unable to decompress it in the YARN container.

This tez.tar.gz file is copied from the hdfs:///hdp/apps/<hdp_version>/tez/tez.tar.gz.

To reproduce the error and confirm that this file is corrupted, you can run the following command:

 sudo su 
 su hdfs
 hdfs dfs -get /hdp/apps/2.6.4.5-2/tez.tar.gz
 tar -xvzf tez.tar.gz

You will get the following error:

gzip: stdin: invalid compressed data--format violated
tar: Unexpected EOF in archive
tar: Unexpected EOF in archive
tar: Error is not recoverable: exiting now

The fix is pretty simple, you must just replace the HDFS file with the one that you have on your local file-system running the following command:

hdfs dfs -rm /hdp/apps/2.6.4.5-2/tez/tez.tar.gz
hdfs dfs -put /usr/hdp/current/tez-client/lib/tez.tar.gz /hdp/apps/2.6.4.5-2/tez/tez.tar.gz    

Now restart Hive Server 2 service and done!

NOTE: If something similar happens with other services you can do the same thing. Please check the following link that has more details: https://community.hortonworks.com/articles/30096/foxing-broken-targz-and-jar-files-in-hdp-24.html

Hope this helps!