[
https://issues.apache.org/jira/browse/HADOOP-3485?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12604293#action_12604293
]
Hong Tang commented on HADOOP-3485:
-----------------------------------
Why would you consider it a hack? Well, maybe true in the sense that writing
FUSE for HDFS is a hack because you will never be able to perform random writes.
In any case, my take is that by not depending on appending, the FUSE-based
mounter can be used in current and older versions of HDFS, which benefits users
who prefer hold-and-wait when some new things come out.
BTW, I do not believe this should be a major effort, and I am reasonably
comfortable to say this because my old company has an internal product very
similar to HDFS and we got it working with FUSE flawlessly.
-Hong
> implement writes
> ----------------
>
> Key: HADOOP-3485
> URL: https://issues.apache.org/jira/browse/HADOOP-3485
> Project: Hadoop Core
> Issue Type: New Feature
> Components: contrib/fuse-dfs
> Reporter: Pete Wyckoff
> Priority: Minor
> Original Estimate: 48h
> Remaining Estimate: 48h
>
> Doesn't support writes because fuse protocol first creates the file then
> closes it and re-opens it to start writing to it. So, (until hadoop-1700), we
> need a work around.
> One way to do this is to open the file with overwrite flag on the second
> open. For security, would only want to do this for zero length files (could
> even check the creation ts too, but because of clock skew, that may be
> harder).
> Doug, Craig, Nicholas - Comments?
> -- pete
> ps since mostly already implemented, this should be a very quick patch
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
