[
https://issues.apache.org/jira/browse/HADOOP-16100?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16765213#comment-16765213
]
Eric Yang commented on HADOOP-16100:
------------------------------------
[~salamani] Most compression libraries are not big endian compliant. If I
recall correctly in porting Hadoop to IBM Power7 several years ago, and there
were some changes required in compression codec like lz0 and snappy to get
proper checksum. Try to switch compression codecs, and see if it makes any
difference.
> checksum errors observed while executing examples on big-endian system
> ----------------------------------------------------------------------
>
> Key: HADOOP-16100
> URL: https://issues.apache.org/jira/browse/HADOOP-16100
> Project: Hadoop Common
> Issue Type: Bug
> Components: common
> Reporter: salamani
> Priority: Major
> Attachments: hadoop_checksumerror_on_bigendian.log,
> hadoop_test_2.9.2_log.txt
>
>
> I have observed the checksum error with big-endian system while executing an
> example with hadoop 2.7.1/2.7.7.
> I have also tried by building the hadoop 2.7.7 on big-endian system. Build
> was successful, however still observed the checksum error while executing an
> example and testcases. Please find the logs for more information
> can you help us know how to go about resolving this checksum issues or are
> they known issues for big endian platform
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
