> Small outgoing packets to the pop3 server or to smtp servers seem
> to work fine though -- it looks like the TCP stack is only able to
> do one thing at a time and it processes stuff in LIFO order (or in
> this case, FINO).
>
> Are you trying to do bulk transfers and "choppy" connections at the
> same time? This is where PPP tends to break down, and one suggested
> workaround is to explicitly set your tx queue length to some small
> value at ifconfig up time (say "4"), this makes congestion control
> kick in much faster in these kinds of connection loads. You will then
> see that load is spread out more evenly and connections make equal
> progress.
OK. I'll try different things and document them for the
linuxperf site so that others can learn from it. I'm
also downloading a new netbase package since ifconfig 1.46
doesn't seem to know about txqueuelen :)
Btw, are the net gurus working on improving the worst-case
default behaviour of the Linux networking code? In this
case workarounds are needed to get regular work done, this
just isn't the Right Thing :(
regards,
Rik -- Open Source: you deserve to be in control of your data.
+-------------------------------------------------------------------+
| Le Reseau netwerksystemen BV: http://www.reseau.nl/ |
| Linux Memory Management site: http://www.linux.eu.org/Linux-MM/ |
| Nederlandse Linux documentatie: http://www.nl.linux.org/ |
+-------------------------------------------------------------------+
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/