On 17-08-11 23:42, Schattenfell wrote:
>
> Hi Paul,
>
>>> The timeout handler is only set on the reader. So any client that hasn't
>>> written anything for longer than 'timeout' triggers a timeout event.
>>> Perhaps we could reset the read-timeout after successful writes.
>
> I think you mean it the other way round: after successful "reads"
> (downloading email) the write-timeout should also be resettet to avoid the
> server side connection close.
No. I meant what I said, talking from the perspective of the server of
course.
The read-timeout is currently disabled whenever the socket is marked as
readable (i.e. there's available data written by the client), and reset
after all data has been read.
A similar approach could be used for writing to clients. But we want to
avoid the situation where the buffer holds a lot of data for a client,
but the client fails to read any data off the wire for a long time.
Tricky stuff I'm not getting into before 3.0 final.
--
________________________________________________________________
Paul J Stevens pjstevns @ gmail, twitter, skype, linkedin
* Premium Hosting Services and Web Application Consultancy *
www.nfg.nl/[email protected]/+31.85.877.99.97
________________________________________________________________
_______________________________________________
DBmail mailing list
[email protected]
http://mailman.fastxs.nl/cgi-bin/mailman/listinfo/dbmail
