[jira] [Created] (HBASE-16464) archive folder grows bigger and bigger due to corrupt snapshot under tmp dir

["Heng Chen (JIRA)" <ji...@apache.org>]

Heng Chen created HBASE-16464:
---------------------------------

             Summary: archive folder grows bigger and bigger due to corrupt snapshot under tmp dir
                 Key: HBASE-16464
                 URL: https://issues.apache.org/jira/browse/HBASE-16464
             Project: HBase
          Issue Type: Bug
            Reporter: Heng Chen


We met the problem on our real production cluster,  we need to cleanup some data on hbase,  we notice the archive folder is much larger than others,  so we delete all snapshots of all tables,  but the archive folder still grows bigger and bigger. 

After check the hmaster log, we notice the exception below:
{code}
2016-08-22 15:34:33,089 ERROR [f04,16000,1471240833208_ChoreService_1] snapshot.SnapshotHFileCleaner: Exception while checking if files were valid, keeping them just in case.
org.apache.hadoop.hbase.snapshot.CorruptedSnapshotException: Couldn't read snapshot info from:hdfs://f04/hbase/.hbase-snapshot/.tmp/frog_stastic_2016-08-17/.snapshotinfo
        at org.apache.hadoop.hbase.snapshot.SnapshotDescriptionUtils.readSnapshotInfo(SnapshotDescriptionUtils.java:295)
        at org.apache.hadoop.hbase.snapshot.SnapshotReferenceUtil.getHFileNames(SnapshotReferenceUtil.java:328)
        at org.apache.hadoop.hbase.master.snapshot.SnapshotHFileCleaner$1.filesUnderSnapshot(SnapshotHFileCleaner.java:85)
        at org.apache.hadoop.hbase.master.snapshot.SnapshotFileCache.getSnapshotsInProgress(SnapshotFileCache.java:303)
        at org.apache.hadoop.hbase.master.snapshot.SnapshotFileCache.getUnreferencedFiles(SnapshotFileCache.java:194)
        at org.apache.hadoop.hbase.master.snapshot.SnapshotHFileCleaner.getDeletableFiles(SnapshotHFileCleaner.java:62)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.checkAndDeleteFiles(CleanerChore.java:233)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.checkAndDeleteEntries(CleanerChore.java:157)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.checkAndDeleteDirectory(CleanerChore.java:180)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.checkAndDeleteEntries(CleanerChore.java:149)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.checkAndDeleteDirectory(CleanerChore.java:180)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.checkAndDeleteEntries(CleanerChore.java:149)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.checkAndDeleteDirectory(CleanerChore.java:180)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.checkAndDeleteEntries(CleanerChore.java:149)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.checkAndDeleteDirectory(CleanerChore.java:180)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.checkAndDeleteEntries(CleanerChore.java:149)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.checkAndDeleteDirectory(CleanerChore.java:180)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.checkAndDeleteEntries(CleanerChore.java:149)
        at org.apache.hadoop.hbase.master.cleaner.CleanerChore.chore(CleanerChore.java:124)
        at org.apache.hadoop.hbase.ScheduledChore.run(ScheduledChore.java:185)
        at java.util.concurrent.Executors$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$Worker.run(ThreadPoolExecutor.java:617)
        at java.lang.Thread.run(Thread.java:745)
Caused by: java.io.FileNotFoundException: File does not exist: /hbase/.hbase-snapshot/.tmp/frog_stastic_2016-08-17/.snapshotinfo
        at org.apache.hadoop.hdfs.server.namenode.INodeFile.valueOf(INodeFile.java:71)
        at org.apache.hadoop.hdfs.server.namenode.INodeFile.valueOf(INodeFile.java:61)
        at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getBlockLocationsInt(FSNamesystem.java:1828)
        at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getBlockLocations(FSNamesystem.java:1799)
        at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getBlockLocations(FSNamesystem.java:1712)
        at org.apache.hadoop.hdfs.server.namenode.NameNodeRpcServer.getBlockLocations(NameNodeRpcServer.java:587)
        at org.apache.hadoop.hdfs.protocolPB.ClientNamenodeProtocolServerSideTranslatorPB.getBlockLocations(ClientNamenodeProtocolServerSideTranslatorPB.java:365)
        at org.apache.hadoop.hdfs.protocol.proto.ClientNamenodeProtocolProtos$ClientNamenodeProtocol$2.callBlockingMethod(ClientNamenodeProtocolProtos.java)
        at org.apache.hadoop.ipc.ProtobufRpcEngine$Server$ProtoBufRpcInvoker.call(ProtobufRpcEngine.java:616)
{code}

It means when SnapshotHFileCleaner begin to cleanup the archive folder, it reads the the snapshot dir to check if any links to hfiles exist, but when read the file /.hbase-snapshot/.tmp/frog_stastic_2016-08-17/.snapshotinfo, corrupt exception thrown out (not sure why the file corrupt),  and cleanup will be failed.

I think we should catch up the exception and delete the file to ensure cleanup will go on.






--
This message was sent by Atlassian JIRA
(v6.3.4#6332)