[
https://issues.apache.org/jira/browse/NET-258?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12993092#comment-12993092
]
Sebb commented on NET-258:
--------------------------
Thanks for the patch.
However, as it stands, it's not guaranteed to always work. The boolean
stopKeepAlive is set by one thread and read by another, and therefore needs to
be volatile (or read in a sync. block). The code assumes that NET FTP is
thread-safe - I'm not sure that is the case.
Also, what happens if the thread sends a NOOP whilst the main code is busy
doing a transfer? Or vice versa? I suspect that would cause errors.
==
I'm not sure a separate keep-alive thread is the way to go here - too tricky to
co-ordinate.
Maybe it would be possible to use some kind of worker queue to ensure requests
are single-threaded.
Keep-alive could be supported by having the worker use wait with timeout and
send a NOOP if there was nothing else to do.
> Implement A Keepalive Mechanism
> -------------------------------
>
> Key: NET-258
> URL: https://issues.apache.org/jira/browse/NET-258
> Project: Commons Net
> Issue Type: Improvement
> Affects Versions: 2.0
> Reporter: Rory Winston
> Assignee: Rory Winston
> Attachments: ftp-keepalive.diff
>
>
> For routers/firewalls that terminate idle connections, a separate heartbeat
> mechanism may need to be implemented to keep the control connection active.
> Some potential issues:
> * Synchronization between a heartbeat write and a __getReply() on an active
> control connection command;
> * Should use the NOOP command as a heartbeat signal;
> * Make the timeout configurable;
> * Default SocketImpl::setKeepAlive() wont do here.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
