[ATLAS-2129] Atlas shutdown during progress of bulk import throws ConcurrentModificationException

• Type: Bug
• Status: Resolved
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 1.0.0
• Fix Version/s: 0.8.2, 1.0.0
• Component/s: atlas-core
• Assignee: Ashutosh Mestry
• Reporter: Sharmadha S
• Created: 11/Sep/17 11:13
• Updated: 15/Sep/17 18:26
• Resolved: 15/Sep/17 15:50

Description

1. Exported an hive_db containing 300 hive_table entities from cluster1 into a zip file .
2. Tried to import into cluster2.
3. When the import was in progress ( at 34%) , stopped Atlas.
4. Following exception was seen in application logs of cluster2 :

2017-09-11 10:17:09,192 INFO  - [pool-2-thread-9 - 83fe24a2-ff2b-4add-a94e-a54b09090912:] ~ bulkImport(): progress: 34% (of 301) - entity:last-imported:hive_table:[102]:(2d629a8e-5c94-40e8-b83f-8a9c91c6de8d) (AtlasEntityStoreV1:238)
2017-09-11 10:17:09,340 ERROR - [SIGTERM handler:] ~ Could not commit transaction [1] due to exception (StandardTitanGraph:731)
java.util.ConcurrentModificationException
	at java.util.ArrayList$Itr.checkForComodification(ArrayList.java:901)
	at java.util.ArrayList$Itr.next(ArrayList.java:851)
	at atlas.shaded.titan.guava.common.collect.Iterators$7.computeNext(Iterators.java:701)
	at atlas.shaded.titan.guava.common.collect.AbstractIterator.tryToComputeNext(AbstractIterator.java:143)
	at atlas.shaded.titan.guava.common.collect.AbstractIterator.hasNext(AbstractIterator.java:138)
	at com.thinkaurelius.titan.graphdb.database.StandardTitanGraph.prepareCommit(StandardTitanGraph.java:473)
	at com.thinkaurelius.titan.graphdb.database.StandardTitanGraph.commit(StandardTitanGraph.java:654)
	at com.thinkaurelius.titan.graphdb.transaction.StandardTitanTx.commit(StandardTitanTx.java:1337)
	at com.thinkaurelius.titan.graphdb.blueprints.TitanBlueprintsGraph.shutdown(TitanBlueprintsGraph.java:120)
	at com.thinkaurelius.titan.graphdb.database.StandardTitanGraph.shutdownInternal(StandardTitanGraph.java:171)
	at com.thinkaurelius.titan.graphdb.database.StandardTitanGraph.access$700(StandardTitanGraph.java:75)
	at com.thinkaurelius.titan.graphdb.database.StandardTitanGraph$ShutdownThread.start(StandardTitanGraph.java:756)
	at java.lang.ApplicationShutdownHooks.runHooks(ApplicationShutdownHooks.java:102)
	at java.lang.ApplicationShutdownHooks$1.run(ApplicationShutdownHooks.java:46)
	at java.lang.Shutdown.runHooks(Shutdown.java:123)
	at java.lang.Shutdown.sequence(Shutdown.java:167)
	at java.lang.Shutdown.exit(Shutdown.java:212)
	at java.lang.Terminator$1.handle(Terminator.java:52)
	at sun.misc.Signal$1.run(Signal.java:212)
	at java.lang.Thread.run(Thread.java:748)

Other operations that are called during Atlas shut down such as Shutdown hook , Stopping KafkaNotification service , NotificationHookConsumer , HBaseBasedAuditRepository are not called.

5.After that , restarted Atlas. Atlas functioned properly.
6.Resumed import with import option , startGuid= <GUID present in logs at 33 %. Atlas was stopped when Import was going on at 34% >
7. Import completed successfully.
8.Post import, only entities from 33% to 100% were imported . Initial 32% of the entities were not imported.
9.Fired an import command again without any interruption and without specifying the startGUID. All 300 hive_table entities and 1 hive_db were imported successfully.

Issue Links

• links to: Code Review