[SENTRY-1630] out of sequence error in HMSFollower

• Type: Sub-task
• Status: Resolved
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 2.0.0
• Fix Version/s: 2.0.0
• Component/s: Hdfs Plugin
• Assignee: Alex Kolbasov
• Reporter: Hao Hao
• Created: 15/Feb/17 20:38
• Updated: 24/Jul/17 15:36
• Resolved: 11/Jul/17 20:20

Description

When HMSFollower exits in an abnormal way, for all the following connection would encounter out of sequence and SocketTimeoutException: Read timed out. Looking at HIVE-6893, it seems to relate to leakage connection problem.

2017-02-15 19:03:42,822 ERROR org.apache.sentry.hdfs.FullUpdateInitializer: Task did not complete successfully after 0 tries. Exception got: org.apache.thrift.TApplicationException: get_database failed: out of sequence response
2017-02-15 19:03:42,827 ERROR org.apache.sentry.hdfs.FullUpdateInitializer: Task did not complete successfully after 0 tries. Exception got: MetaException(message:Got exception: org.apache.thrift.transport.TTransportException null)
2017-02-15 19:03:43,803 INFO hive.metastore: Closed a connection to metastore, current connections: 0
2017-02-15 19:03:43,803 ERROR org.apache.sentry.service.thrift.HMSFollower: Exception occurred persisting Hive full snapshot into DB
java.lang.RuntimeException: org.apache.thrift.TApplicationException: get_database failed: out of sequence response
        at org.apache.sentry.hdfs.FullUpdateInitializer.createInitialUpdate(FullUpdateInitializer.java:324)
        at org.apache.sentry.service.thrift.HMSFollower.fetchFullUpdate(HMSFollower.java:343)
        at org.apache.sentry.service.thrift.HMSFollower.run(HMSFollower.java:244)
        at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
        at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:304)
        at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:178)
        at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
        at java.lang.Thread.run(Thread.java:745)
Caused by: org.apache.thrift.TApplicationException: get_database failed: out of sequence response
        at org.apache.thrift.TServiceClient.receiveBase(TServiceClient.java:84)
        at org.apache.hadoop.hive.metastore.api.ThriftHiveMetastore$Client.recv_get_database(ThriftHiveMetastore.java:662)
        at org.apache.hadoop.hive.metastore.api.ThriftHiveMetastore$Client.get_database(ThriftHiveMetastore.java:649)
        at org.apache.hadoop.hive.metastore.HiveMetaStoreClient.getDatabase(HiveMetaStoreClient.java:1213)
        at org.apache.sentry.hdfs.FullUpdateInitializer$DbTask.doTask(FullUpdateInitializer.java:256)
        at org.apache.sentry.hdfs.FullUpdateInitializer$BaseTask$RetryStrategy.exec(FullUpdateInitializer.java:110)
        at org.apache.sentry.hdfs.FullUpdateInitializer$BaseTask.call(FullUpdateInitializer.java:150)
        at org.apache.sentry.hdfs.FullUpdateInitializer$BaseTask.call(FullUpdateInitializer.java:79)
        at java.util.concurrent.FutureTask.run(FutureTask.java:262)
        ... 3 more
2017-02-15 19:03:43,849 INFO org.apache.sentry.service.thrift.HMSFollower: Making a kerberos connection to HMS

Issue Links

• is related to: SENTRY-1676 FullUpdateInitializer#createInitialUpdate should not throw RuntimeException
• relates to: SENTRY-1628 In HMSFollower failing of catching error causes the executor to halt
• links to: Code Review