httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Marc Slemko <>
Subject Re: [BUG] 1.3 not closing socket on NT
Date Fri, 05 Sep 1997 03:35:53 GMT
On Fri, 5 Sep 1997, Giovanni Maruzzelli wrote:

> I've made some tracing with tcpdump -vv.
> They seems to show that the problem is originated by the fact that windoz
> clients (netscape 3 on win95) reset the socket after acknoledging a FIN,
> versus Unix clients (Lynx 2.6) that after acknoledging a FIN send another
> FIN..

So are you saying that it isn't a problem when they send a RST but 
it is when they properly send the FIN?

The version of Navigator you looked at has the stupid behavior of
terminating connections with a RST.  That ain't TCPC.  I think we
convinced both MS and NS to stop that stupid behavior for non-abortive
closes; newer versions will change their behavior to properly send a FIN. 
Not sure if any released versions change it yet. 

However, I repeat: this is not a problem in Apache.  CLOSE_WAIT 
is normal.  It is necessary.  It is needed.  If NT can't 
handle it, then it is broken.  Why do you think that having sockets in
CLOSE_WAIT is a bad thing?  If NT can't handle a couple of hundred of
those, get mad a MS.

View raw message