You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@carbondata.apache.org by "Manish Gupta (JIRA)" <ji...@apache.org> on 2017/07/10 13:42:00 UTC

[jira] [Updated] (CARBONDATA-1277) Dictionary generation failure if there is failure in closing output stream in HDFS

     [ https://issues.apache.org/jira/browse/CARBONDATA-1277?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Manish Gupta updated CARBONDATA-1277:
-------------------------------------
    Summary: Dictionary generation failure if there is failure in closing output stream in HDFS  (was: Dictionary generation failure if there is failure in closing output steam in HDFS)

> Dictionary generation failure if there is failure in closing output stream in HDFS
> ----------------------------------------------------------------------------------
>
>                 Key: CARBONDATA-1277
>                 URL: https://issues.apache.org/jira/browse/CARBONDATA-1277
>             Project: CarbonData
>          Issue Type: Bug
>            Reporter: Manish Gupta
>            Assignee: Manish Gupta
>             Fix For: 1.2.0
>
>          Time Spent: 1h 40m
>  Remaining Estimate: 0h
>
> If there is any failure while closing the output stream of dictionary file in HDFS then on next data load, update or insert into operation dictionary generation fails. This is because we open the dictionary file in append mode and when we try to get the output stream for that file HDFS throws an exception that Lease is already acquired by some other client. Exception trace as below.
> java.io.IOException: Failed to APPEND_FILE /user/hive/warehouse/carbon.store/test/t12/Metadata/010497af-1833-4804-a7ec-849ab7b9bf10.dictmeta for DFSClient_NONMAPREDUCE_-1904161438_159 on 172.168.100.212 because lease recovery is in progress. Try again later.
> at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.recoverLeaseInternal(FSNamesystem.java:2901)
> at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.appendFileInternal(FSNamesystem.java:2655)
> at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.appendFileInt(FSNamesystem.java:2968)
> at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.appendFile(FSNamesystem.java:2937)
> at org.apache.hadoop.hdfs.server.namenode.NameNodeRpcServer.append(NameNodeRpcServer.java:776)
> at org.apache.hadoop.hdfs.protocolPB.ClientNamenodeProtocolServerSideTranslatorPB.append(ClientNamenodeProtocolServerSideTranslatorPB.java:458)
> 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)



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)