[ 
https://issues.apache.org/jira/browse/HADOOP-16458?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16892077#comment-16892077
 ] 

Steve Loughran commented on HADOOP-16458:
-----------------------------------------

happening when the LocatedFileStatusFetcher gets null back from glob; the fact 
that the stack from the raised fault isn't propagated is something I plan to 
fix too. 
{code}
@Override
    public Result call() throws Exception {
      Result result = new Result();
      FileSystem fs = path.getFileSystem(conf);
      result.fs = fs;
      FileStatus[] matches = fs.globStatus(path, inputFilter);
      if (matches == null) {           // no matches
        result.addError(new IOException("Input path does not exist: " + path)); 
   // so error raised.
      } else if (matches.length == 0) {
        result.addError(new IOException("Input Pattern " + path
            + " matches 0 files"));
      } else {
        result.matchedFileStatuses = matches;
      }
      return result;
    }
{code}
FWIW, I'd actually tighten down the exceptions raised to an FNFE and PathIOE if 
I wasn't worried about breaking things.

> LocatedFileStatusFetcher.getFileStatuses failing intermittently with s3
> -----------------------------------------------------------------------
>
>                 Key: HADOOP-16458
>                 URL: https://issues.apache.org/jira/browse/HADOOP-16458
>             Project: Hadoop Common
>          Issue Type: Sub-task
>          Components: fs/s3
>    Affects Versions: 3.3.0
>         Environment: S3 + S3Guard
>            Reporter: Steve Loughran
>            Assignee: Steve Loughran
>            Priority: Major
>
> Intermittent failure of LocatedFileStatusFetcher.getFileStatuses(), which is 
> using globStatus to find files.
> I'd say "turn s3guard on" except this appears to be the case, and the dataset 
> being read is
> over 1h old.
> Which means it is harder than I'd like to blame S3 for what would sound like 
> an inconsistency
> We're hampered by the number of debug level statements in the globber code 
> being approximately none; there's no debugging to turn on. All we know is 
> that globFiles returns null without any explanation.



--
This message was sent by Atlassian JIRA
(v7.6.14#76016)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to