[
https://issues.apache.org/jira/browse/FLUME-1350?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13478747#comment-13478747
]
Juhani Connolly commented on FLUME-1350:
----------------------------------------
I had a more detailed look at the source and I don't really think as it is it
will solve a problem without introducing a new one(the thrashing that Mike
referred to).
An alternative solution I would suggest is tracking last writes to each open
file, and having a watcher thread close them after a configured timeout period
where the file has received no writes. This would solve every case of unclused
idle files I can think of, and if a file becomes active again due to a
temporarily out of commission source reactivating, it would not result in
thrashing(the file would be reopened, and then closed again soon after all the
backlog has been handled)
Yongcheng/Mike: What do you think? If no-one else wants to do it, I can put it
together.
> HDFS file handle not closed properly when date bucketing
> ---------------------------------------------------------
>
> Key: FLUME-1350
> URL: https://issues.apache.org/jira/browse/FLUME-1350
> Project: Flume
> Issue Type: Bug
> Components: Sinks+Sources
> Affects Versions: v1.1.0, v1.2.0
> Reporter: Robert Mroczkowski
> Attachments: HDFSEventSink.java.patch
>
>
> With configuration:
> agent.sinks.hdfs-cafe-access.type = hdfs
> agent.sinks.hdfs-cafe-access.hdfs.path =
> hdfs://nga/nga/apache/access/%y-%m-%d/
> agent.sinks.hdfs-cafe-access.hdfs.fileType = DataStream
> agent.sinks.hdfs-cafe-access.hdfs.filePrefix = cafe_access
> agent.sinks.hdfs-cafe-access.hdfs.rollInterval = 21600
> agent.sinks.hdfs-cafe-access.hdfs.rollSize = 10485760
> agent.sinks.hdfs-cafe-access.hdfs.rollCount = 0
> agent.sinks.hdfs-cafe-access.hdfs.txnEventMax = 1000
> agent.sinks.hdfs-cafe-access.hdfs.batchSize = 1000
> #agent.sinks.hdfs-cafe-access.hdfs.codeC = snappy
> agent.sinks.hdfs-cafe-access.hdfs.hdfs.maxOpenFiles = 5000
> agent.sinks.hdfs-cafe-access.channel = memo-1
> When new directory is created previous file handle remains opened.
> rollInterval setting is used only with files in current date bucket.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
