[HBASE-15933] NullPointerException may be thrown from SimpleRegionNormalizer#getRegionSize()

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 1.4.0, 2.0.0
• Hadoop Flags: Reviewed
• Assignee: Ted Yu
• Reporter: Ted Yu
• Created: 01/Jun/16 16:01
• Updated: 02/Jun/16 21:55
• Resolved: 02/Jun/16 16:11

Description

The following was observed in master log:

2016-05-26 00:09:54,716 INFO  [AM.ZK.Worker-pool2-t6917] master.RegionStates: Transition {7669ce6bec40dd44bb2388ff4d610b7b state=CLOSED, ts=1464221394716, server=x.net,60020,1463615066136} to {7669ce6bec40dd44bb2388ff4d610b7b state=OFFLINE, ts=1464221394716, server=x.net,60020,1463615066136}
2016-05-26 00:09:54,717 ERROR [y.net,60000,1463614725717_ChoreService_1] normalizer.RegionNormalizerChore: Caught error
java.lang.NullPointerException
  at org.apache.hadoop.hbase.master.normalizer.SimpleRegionNormalizer.getRegionSize(SimpleRegionNormalizer.java:174)
  at org.apache.hadoop.hbase.master.normalizer.SimpleRegionNormalizer.computePlanForTable(SimpleRegionNormalizer.java:99)
  at org.apache.hadoop.hbase.master.HMaster.normalizeRegions(HMaster.java:1406)
  at org.apache.hadoop.hbase.master.normalizer.RegionNormalizerChore.chore(RegionNormalizerChore.java:49)
  at org.apache.hadoop.hbase.ScheduledChore.run(ScheduledChore.java:185)
  at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
  at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:308)
  at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:180)
  at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:294)
  at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
  at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)

DEBUG was not turned on.
Looks like race condition where the Map returned by getRegionsLoad() no longer contained certain region.