Greg A. Woods writes:
> >    not a NetBSD issue. We're doing what we need to.
> 
> Wrong.
> 
> There are many places where NetBSD (and/or 4.4bsd in general) slips away
> from full compliance with RFCs when it can do something better (or
> sometimes when it's lazy).  Leaving broken PMTUD in place even when we
> know about it, and when we can safely do something about it, is a
> seriously bad cop-out.

Path MTU discover is not broken. Our implementation is perfectly fine.

The firewalls at the remote end systems are broken -- they swallow
ICMP. This is a major violation of the RFCs and of the internet architecture.

I'm personally not interested in doing anything further on this
subject, including discussing it.

Perry