[
https://issues.apache.org/jira/browse/HDFS-15472?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17158200#comment-17158200
]
Ayush Saxena commented on HDFS-15472:
-------------------------------------
Thanx [~dzcxzl] for the report, Can you add a test as well for the issue?
Secondly, we shouldn't make changes in {{Common}} for Erasure Coding, May be
handling this at {{DfsStripedInputStream}} shall be a better. We should refrain
from making HDFS only changes in {{FSDataInputStream}}
> Erasure Coding: Support fallback read when zero copy is not supported
> ---------------------------------------------------------------------
>
> Key: HDFS-15472
> URL: https://issues.apache.org/jira/browse/HDFS-15472
> Project: Hadoop HDFS
> Issue Type: Bug
> Reporter: dzcxzl
> Priority: Trivial
> Attachments: HDFS-15472.000.patch
>
>
> [HDFS-8203|https://issues.apache.org/jira/browse/HDFS-8203]
> ec does not support zeor copy read, but currently does not support fallback
> read, it will throw an exception.
> {code:java}
> Caused by: java.lang.UnsupportedOperationException: Not support enhanced byte
> buffer access.
> at
> org.apache.hadoop.hdfs.DFSStripedInputStream.read(DFSStripedInputStream.java:524)
> at org.apache.hadoop.fs.FSDataInputStream.read(FSDataInputStream.java:188)
> at
> org.apache.hadoop.hive.shims.ZeroCopyShims$ZeroCopyAdapter.readBuffer(ZeroCopyShims.java:79)
> {code}
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
