Greg A. Woods writes:
> [ On Sat, November 21, 1998 at 07:24:04 (-0500), Perry E. Metzger wrote: ]
> > Subject: Re: SOLVED! The cause of puzzling TCP (eg. WHOIS) connection failu
res with some InterNIC.net hosts 
> >
> > This totally violates the required behavior of routers. It is a
> > bad idea. Among other things, it probably breaks path MTU
> > discovery.
> 
> Where in the heck did you get that idea?

By thinking.

If you silently fragment packets, you've broken the protocol. The
protocol WORKS by sending you repeated probes with DF bits set and
hoping you'll not lie when replying with ICMPs.

> My proposal fixes broken PMTUD

No, it just breaks it for people for whom it isn't broken.

> *Something* needs to be done since it's clear that there will be ongoing
> problems with broken PMTUD.

Yeah. Firewall administrators should turn on ICMP message replies. BFD.

> Either the protocol needs fixing (and I
> admit I've not yet read the RFCs to see if the problem is actually in
> the protocol, or just in NetBSD-1.3.2's implementation of it),

You haven't read the RFCs and you're proposing to "fix" the protocol?

Perry