Toshihiro Suzuki created HBASE-23359:
----------------------------------------
Summary: RS going down with NPE when splitting a region with
compaction disabled in branch-1
Key: HBASE-23359
URL: https://issues.apache.org/jira/browse/HBASE-23359
Project: HBase
Issue Type: Bug
Reporter: Toshihiro Suzuki
Trying to backport HBASE-22096 to brach-1, I faced the issue where a RS goes
down with NPE when splitting a region with compaction disabled.
The steps to reproduce this issue are as follows:
{code}
compaction_switch false
create "test", "cf"
(0...2000).each{|i| put "test", "row#{i}", "cf:col", "val#{i}"}
split "test"
{code}
Looking at the regionserver log, I saw the following log:
{code}
2019-12-03 22:25:38,611 INFO [RS:0;10.0.1.11:53504-splits-0]
regionserver.SplitRequest: Running rollback/cleanup of failed split of
test,,1575379535506.50e322ec68162025e17cddffdc2fb17e.; null
java.lang.NullPointerException
at
org.apache.hadoop.hbase.regionserver.HStore.cancelRequestedCompaction(HStore.java:1834)
at
org.apache.hadoop.hbase.regionserver.CompactSplitThread$Rejection.rejectedExecution(CompactSplitThread.java:656)
at
java.util.concurrent.ThreadPoolExecutor.reject(ThreadPoolExecutor.java:830)
at
java.util.concurrent.ThreadPoolExecutor.execute(ThreadPoolExecutor.java:1379)
at
org.apache.hadoop.hbase.regionserver.CompactSplitThread.requestCompactionInternal(CompactSplitThread.java:401)
at
org.apache.hadoop.hbase.regionserver.CompactSplitThread.requestSystemCompaction(CompactSplitThread.java:348)
at
org.apache.hadoop.hbase.regionserver.HRegionServer.postOpenDeployTasks(HRegionServer.java:2111)
at
org.apache.hadoop.hbase.regionserver.HRegionServer.postOpenDeployTasks(HRegionServer.java:2097)
at
org.apache.hadoop.hbase.regionserver.SplitTransactionImpl.openDaughters(SplitTransactionImpl.java:478)
at
org.apache.hadoop.hbase.regionserver.SplitTransactionImpl.stepsAfterPONR(SplitTransactionImpl.java:549)
at
org.apache.hadoop.hbase.regionserver.SplitTransactionImpl.execute(SplitTransactionImpl.java:532)
at
org.apache.hadoop.hbase.regionserver.SplitRequest.doSplitting(SplitRequest.java:82)
at
org.apache.hadoop.hbase.regionserver.SplitRequest.run(SplitRequest.java:153)
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
2019-12-03 22:25:38,613 FATAL [RS:0;10.0.1.11:53504-splits-0]
regionserver.HRegionServer: ABORTING region server
10.0.1.11,53504,1575379011279: Abort; we got an error after point-of-no-return
{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
