Re: Hit capacity of Classic HTTP Server and a module 170 -- WEB400

I had the same thing happen today on the original server running on an
820 under V5R2. The web server just shutdown. I was also under a heavy
workload. This is not the first time this has happened. We do not get a
heavy load very often but when we do the server shutting down has not
been uncommon. We have been trying to get moved to the Apache server but
an app we purchased will not run under Apache without an upgrade. We
just stopped using this app and can now migrate and after today that
might become a top priority. I am not an expert on web servers but each
time this has happened and we talk to IBM we end up thinking it was a
problem with the number of threads available to the server. Too few or
too many and the server crashes. That has been my experience

>>> franz400@xxxxxxxxxxxx 01/05/05 8:01 PM >>>
Bob - what are you serving?  static html or dynamic? lot's of
pictures?
How much memory in the 170? what's the cpw?  any heavy sql or rpg
processing?
I've got a customer with a S10 512 meg 73 cpw serving rpg cgi
that sometimes hits this volume. Only have a performance problem when
some sql queries have to roll thru all records in a big file (where x
like
%yy%)
but never crashes.
Are you saying crash as in restart the server? Did you get any job logs
of
some other problem or just timeouts?
jim


----- Original Message ----- 
From: "Bob Cozzi" <cozzi@xxxxxxxxx>
To: "'Web Enabling the AS400 / iSeries'" <web400@xxxxxxxxxxxx>
Sent: Wednesday, January 05, 2005 6:26 PM
Subject: [WEB400] Hit capacity of Classic HTTP Server and a module 170


> Hi,
> This morning my little old 170 running IBM's Classic HTTP Server on
OS/400
> V4.5 "crashed".
> The cause for this was due to extremely high volume of hits during a
2
hour
> period.
> When my free email newsletter went out last night and everybody on
the
east
> coast got in this morning, the server started to get very, very busy.
In
> about a 90 minute period we had nearly 5,000 hits on the website.
The
entire
> system slowed to a crawl and the webserver pages were effectively
not
being
> served to the web users.
>
> My theory is that too many server jobs were being started for the
system
> capacity and some CGI jobs where starting/ending so quickly with each
hit
> that the system just couldn't keep up.
> Now I have been lobbying to upgrade to V5R1 and go to the Apache
server
but
> haven't been able to get the hosting agent to do that yet (after all
> RPGIV.com is a free site that produces no revenue).
> So my question is, have others had this kind of thing happen on
their
> systems?
> Can the 400 handle 5,000+ to 10,000+ hits per hour on an entry-level
> machine?
> -Bob Cozzi
>
> _______________________________________________
> This is the Web Enabling the AS400 / iSeries (WEB400) mailing list
> To post a message email: WEB400@xxxxxxxxxxxx
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/web400
> or email: WEB400-request@xxxxxxxxxxxx
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/web400.
>
>


_______________________________________________
This is the Web Enabling the AS400 / iSeries (WEB400) mailing list
To post a message email: WEB400@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/web400
or email: WEB400-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/web400.