users@grizzly.java.net

Re: maxKeepAliveRequests and maxThreads

From: Bongjae Chang <bongjae.chang_at_gmail.com>
Date: Wed, 09 Mar 2011 17:22:10 +0900

Hi Amrit,

I think "maxKeepAliveRequests" means how many requests can be allowed per
a keep alive connection over HTTP/1.1.

A browser and server will keep the connection of first request if they use
keep-alive of HTTP/1.1 and reuse it.

In "x"th request, Grizzly doesn't send "Connection: Keep-Alive" but
"Connection: close" to the browser, then the connection will be closed and
"x+1"th request will establish new connection.

"maxThreads" means the maximum worker thread number.

I think "maxThreads" should be determined according to concurrent work
load.

I hope this helps you.

Thanks.

Regards,
Bongjae Chang


On 11. 3. 9. ¿ÀÈÄ 1:53, "amrit" <amritjyotibaruah_at_gmail.com> wrote:

>
>Hi,
>
>Could someone please explain me the use of the parameters
>maxKeepAliveRequests and maxThreads for Grizzly server?
>
>
>If we set the maxKeepAliveRequests to a value 'x' suppose, what behaviour
>an
>end user can expect considering he is submitting HTTP/1.1 requests from a
>browswer? Is it like only 'x' many number of requests can processed
>concurrently? In that case what happens to the 'x+1' th request?
>
>Similar way, what differece could be seen if we set the 'maxThreads' to
>different values?
>
>Related link -
>
>http://svn-mirror.glassfish.org/grizzly/trunk/www/apidocs/com/sun/grizzly/
>http/SelectorThread.html#setMaxKeepAliveRequests(int)
>
>Regards,
>Amrit
>--
>View this message in context:
>http://old.nabble.com/maxKeepAliveRequests-and-maxThreads-tp31103465p31103
>465.html
>Sent from the Grizzly - Users mailing list archive at Nabble.com.
>