>>> Ranjit Gopi <[EMAIL PROTECTED]> 11-Aug-00 3:23:40 PM >>>
>But my question is something else.
>It is not the performance impact of the Outout Stream or
>the buffering logic. It comes before that.
Yes... but you're question is more or less answered by those posts,
perhaps by absence, but it is answered.
>Basically if the servlet does something for infinity(before it
>actually comes to write the response back) is there some
>WebServer parameter that can kill it?(could do with a timer
>thread , just wanted to check if there is any server which
>provides this, knowing well that the transaction state is
>ambigous. It doesnt matter for what I am doing)
You're talking about when an HTTP connection is timed out; ie: when
you issue a request from a browser when does the browser (or server)
consider it timed out?
Basically the answer is: the server *can* timeout your connection but
most don't (it doesn't make much sense for the server to do it).
The browser can timeout your connection but the timeout is high
(can't remember what it is exactly).
When using HTTP/1.1 persistent connections have different timeout
issues which (should be) explained in the posts Ralhan refers to.
Nic Ferrier
___________________________________________________________________________
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff SERVLET-INTEREST".
Archives: http://archives.java.sun.com/archives/servlet-interest.html
Resources: http://java.sun.com/products/servlet/external-resources.html
LISTSERV Help: http://www.lsoft.com/manuals/user/user.html
