[ 
https://issues.apache.org/jira/browse/HDFS-1840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13021324#comment-13021324
 ] 

Suresh Srinivas commented on HDFS-1840:
---------------------------------------

# DFSClient.java
#* hdfsTimeout is initalized after LeaseChecker construction. Hence 
LeaseChecker#renewal may not be initialized right?
#* In DFSClient#run - the first time lastRenewed could be == renewal and hence 
lease is may not be renewed the first time
#* currentID is not synchronized correctly. There is a possiblity of threads 
lingering longer, because of this. You can avoid this id, by interrupting the 
current Daemon in remove method and shutting down the thread.
#* minor: //no more being written files, terminate. to // no more files being 
written


> Terminate LeaseChecker when all writing files are closed.
> ---------------------------------------------------------
>
>                 Key: HDFS-1840
>                 URL: https://issues.apache.org/jira/browse/HDFS-1840
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: hdfs client
>            Reporter: Tsz Wo (Nicholas), SZE
>            Assignee: Tsz Wo (Nicholas), SZE
>         Attachments: h1840_20110418.patch
>
>
> In {{DFSClient}}, when there are files opened for write, a {{LeaseChecker}} 
> thread is started for updating the leases periodically.  However, it never 
> terminates when when all writing files are closed.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to