Re: gac experiment

[Jochen Kaiser <Jochen.Kaiser@xxxxxxxxxxxxxxxxxxxx>]

It's working perfectly for up to 1600 users here in Erlangen and on
netsurf it's working, too. 

Please stop polemic accusses. Give us concrete numbers.

If it's just about carrying the discussion from #opers to this
mailinglist - just stop it. There are more important things 
right now to discuss then one branch of the ircd code. 

If there will be the holy ircnet patch someday which solves the
colliding problem, we will happily use it instead of gac.

greetings,
jochen "kind" kaiser


On Tue, Feb 01, 2000 at 08:19:01PM +0100, Thomas Kuiper wrote:
> Some might find this experience interesting, and I thought
> I'd share it. It's about all these patches there are now to fix
> network issues only cause of the #channel.
> 
> Not a long time ago, irc.webbernet.net decided to run
> Guenthi's anti-collide patch (also known as gac). Only as a
> fix way for getting rid of abusers.
> 
> It only last for about 10 minutes.  Why? because that's how
> long it took the admin to figure that it was just not
> working.  The amount of traffic generated for a server with
> 4000 clients was just phenomenal, and caused more harm than
> anything else.  And this was not even the latest version of
> gac which generates even more traffic (for deops).  My
> feeling is that it's enough traffic to cause another small
> server to split.
> 
> just the facts...
> 
> Engerim
> 

-- 
Jochen Kaiser          		                   phone +49 9131 85-28134
Network Administration           mailto:jochen.kaiser@xxxxxxxxxxxxxxxxxxxx
Regionales Rechenzentrum Universitaet Erlangen-Nuernberg, Germany