[Spread-users] Spread - sp_error (8)

Balog, Julian JBalog at activevoice.com
Wed Feb 1 11:37:11 EST 2006


Prady,

Because of the slower client side processing, the messages are piling up in
the server and sooner or later we got the sp_error(8) problem. As Cristina
mentioned, you can
either tune-up your client side processing for better performance or, and
this is what I particularly did while experimenting with the performance of
the Spread platform, you can change the MAX_SESSION_MESSAGES parameter in
the spread_params.h header file.

For my particular application (and client side Spread proxy service, written
in C#) a value of 10000 for the above mentioned parameter aleviated the
problem. I wasn't able to get the error anymore, regardless of the load my
particular application is designed to handle (as I was able to break it
before the change in less than 30 seconds).


Julian




-----Original Message-----
From: Pradyumna Sampath [mailto:pradysam at yahoo.co.in] 
Sent: Tuesday, January 31, 2006 9:27 PM
To: Cristina Nita-Rotaru; Balog, Julian
Cc: Iovescu, Adrian; spread-users at lists.spread.org
Subject: Re: [Spread-users] Spread - sp_error (8)



--- Cristina Nita-Rotaru <crisn at cerias.purdue.edu>
wrote:

> Your receiver is not processing fast enough, that's why it gets 
> disconnected. You have to slow down your senders.

What is the definition of fast enough ? I seem to have a similar problem
here when I send about 200 bytes every 160ms. Also would it have an effect
if the reciever is a java client ?

My config is like this:

Sender (C client,100Mhz proc -> Reciever (Java client,
1Ghz)

regards
Prady

Visit me at http://rtns.org/prady/
Mail : pradysam (at) yahoo (dot) co.in

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.spread.org/pipermail/spread-users/attachments/20060201/c69532aa/attachment.html 


More information about the Spread-users mailing list