[
https://issues.apache.org/jira/browse/HDFS-12165?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16300603#comment-16300603
]
Wei-Chiu Chuang commented on HDFS-12165:
----------------------------------------
It seems to me HDFS-12594 is a possible for fix this issue.
> getSnapshotDiffReport throws NegativeArraySizeException for very large
> snapshot diff summary
> --------------------------------------------------------------------------------------------
>
> Key: HDFS-12165
> URL: https://issues.apache.org/jira/browse/HDFS-12165
> Project: Hadoop HDFS
> Issue Type: Bug
> Reporter: Wei-Chiu Chuang
>
> For a really large snapshot diff, getSnapshotDiffReport throws
> NegativeArraySizeException
> {noformat}
> 2017-07-19 11:14:16,415 WARN org.apache.hadoop.ipc.Server: Error serializing
> call response for call
> org.apache.hadoop.hdfs.protocol.ClientProtocol.getSnapshotDiffReport
> from 10.17.211.10:58223 Call#0 Retry#0
> java.lang.NegativeArraySizeException
> at
> com.google.protobuf.CodedOutputStream.newInstance(CodedOutputStream.java:105)
> at
> com.google.protobuf.AbstractMessageLite.writeDelimitedTo(AbstractMessageLite.java:87)
> at
> org.apache.hadoop.ipc.ProtobufRpcEngine$RpcResponseWrapper.write(ProtobufRpcEngine.java:468)
> at org.apache.hadoop.ipc.Server.setupResponse(Server.java:2410)
> at org.apache.hadoop.ipc.Server.access$500(Server.java:134)
> at org.apache.hadoop.ipc.Server$Handler.run(Server.java:2182)
> {noformat}
> This particular snapshot diff contains more than 25 million different file
> system objects, and which means the serialized response can be more than 2GB,
> overflowing protobuf length calculation.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
