[
https://issues.apache.org/jira/browse/HDFS-1111?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12882746#action_12882746
]
Rodrigo Schmidt commented on HDFS-1111:
---------------------------------------
If I remember correctly, there is another message when the list is empty
telling that no corrupt files could have been found. That part of the code was
not changed by this patch.
The -list-corruptfiles option on fsck emerged after a number of incremental
changes, in the very organic way open source works.
The reason why getCorruptFiles() belongs to ClientProtocol is to allow external
services to query it. We are working on having the RaidNode use this call to
automatically recover corrupted blocks. I guess this shows we did think some
things through.
On a design related note, if we want to build a truly distributed namenode, we
should be thinking about taking things out of it, like fsck itself.
Last, HDFS-1265 seems to be an orthogonal issue, since it doesn't guarantee the
list is complete. Adding filters DOES NOT prevent the queried directory from
having thousands of corrupted files, out of which, only a few hundred will be
reported.
> getCorruptFiles() should give some hint that the list is not complete
> ---------------------------------------------------------------------
>
> Key: HDFS-1111
> URL: https://issues.apache.org/jira/browse/HDFS-1111
> Project: Hadoop HDFS
> Issue Type: New Feature
> Reporter: Rodrigo Schmidt
> Assignee: Rodrigo Schmidt
> Attachments: HADFS-1111.0.patch
>
>
> If the list of corruptfiles returned by the namenode doesn't say anything if
> the number of corrupted files is larger than the call output limit (which
> means the list is not complete). There should be a way to hint incompleteness
> to clients.
> A simple hack would be to add an extra entry to the array returned with the
> value null. Clients could interpret this as a sign that there are other
> corrupt files in the system.
> We should also do some rephrasing of the fsck output to make it more
> confident when the list is not complete and less confident when the list is
> known to be incomplete.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
