[
https://issues.apache.org/jira/browse/HADOOP-12831?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15159558#comment-15159558
]
Steve Loughran commented on HADOOP-12831:
-----------------------------------------
oh, it NPEs early on. Its' not fixable...I'm just looking for a way to force
raw local/bypass checksumming and its buffering on a bit of ATS that I can't
get at directly: it's not my code creating the FS. It also means I can't get at
the rawlocalfs. I'll have to think of another way.
regarding this issue, a precondition saying "checksum block size too small" is
the only thing that makes sense. It will fail fast but with a meaningful
message.
> LocalFS/FSOutputSummer NPEs in constructor if bytes per checksum set to 0
> --------------------------------------------------------------------------
>
> Key: HADOOP-12831
> URL: https://issues.apache.org/jira/browse/HADOOP-12831
> Project: Hadoop Common
> Issue Type: Bug
> Components: fs
> Affects Versions: 2.8.0
> Reporter: Steve Loughran
> Priority: Minor
>
> If you set the number of bytes per checksum to zero,
> {code}
> conf.setInt(LocalFileSystemConfigKeys.LOCAL_FS_BYTES_PER_CHECKSUM_KEY, 0)
> {code}
> then create a "file://" instance, you get to see a stack trace
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
