[HBASE-9216] [0.92] TestReplication.queueFailover fails often

• Type: Bug
• Status: Closed
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 0.92.3
• Fix Version/s: 0.92.3
• Assignee: Andrew Kyle Purtell
• Reporter: Andrew Kyle Purtell
• Created: 14/Aug/13 16:41
• Updated: 16/Jun/22 17:54
• Resolved: 20/Aug/13 21:03

Description

Failure trace:

java.lang.AssertionError: Waited too much time for queueFailover replication. Waited 46463ms.
	at org.junit.Assert.fail(Assert.java:93)
	at org.apache.hadoop.hbase.replication.TestReplication.queueFailover(TestReplication.java:666)

Logs show the test fails to make progress as the replication source log, after an initial few opens/replications/reports/advances, stops increasing in size, this sequence until timeout:

EventThread.replicationSource,2] regionserver.ReplicationSource(544): Opening log for replication ip-10-198-0-134.us-west-1.compute.internal%2C54449%2C1376480042448.1376480134578 at 18587
2013-08-14 11:35:39,568 DEBUG [RegionServer:0;ip-10-198-0-134.us-west-1.compute.internal,54449,1376480042448-EventThread.replicationSource,2] regionserver.ReplicationSource(489): currentNbOperations:16416 and seenEntries:0 and size: 0
2013-08-14 11:35:39,568 DEBUG [RegionServer:0;ip-10-198-0-134.us-west-1.compute.internal,54449,1376480042448-EventThread.replicationSource,2] regionserver.ReplicationSource(618): Nothing to replicate, sleeping 100 times 10
[...]
2013-08-14 11:35:45,205 INFO  [Thread-1125] replication.TestReplication(678): Only got 16607 rows instead of 17576 current i=-1
[...]