Re: portfast

From: Gert Doering (gert@greenie.muc.de)
Date: Sat Jul 07 2001 - 17:50:11 EDT


Hi,

On Sat, Jul 07, 2001 at 12:30:49PM +0100, Ryan O'Connell wrote:
> Network Engineers should be aware of the issues and it's just one of
> the many minor irritations caused by deployment of Spanning Tree. There are
[..]

Is it just me, or is it really time to completely do away with STP?

- STP has no protection against malicious insertion of BPDU packets from
  evildoers

- STP has no concept to get decent "routing" of packets in the face of
  multiple switches connected over a mesh or a ring structure - packets
  always have to travel over the "root path", even if there might be a
  direct connection between two switches which just happens to be in STP
  block instate.

The main suggestion that has come up in the part is a kind of "L2 SPF"
protocol, where the switches would do with MAC addresses what OSPF does
with IP addresses (+networks) - flood them around, calculate Dijstra, send
packets over the shortest path to their target.

The actual switching process would not even be affected that much - just
the way the CAM table is built has to change.

No more problems with slow STP failover, "portfast" kludges, etc.

gert

-- 
USENET is *not* the non-clickable part of WWW!
                                                           //www.muc.de/~gert/
Gert Doering - Munich, Germany                             gert@greenie.muc.de
fax: +49-89-35655025                        gert.doering@physik.tu-muenchen.de



This archive was generated by hypermail 2b29 : Sun Aug 04 2002 - 04:12:44 EDT