As Johan Danielsson wrote ...
> Wilko Bulte <wilko@yedi.iaf.nl> writes:
> 
> > The trick is in using the :vm=rfc1048. If this is omitted the boot
> > will not work at all. SRM 7.0 is the latest SRM for the DEC3000
> > series BTW.
> 
> Are you really sure about this? RFC1048 defines extensions to BOOTP,

Yes, in the sense that it works for me.. And yes, I was surprised too that
it works. But I counted my blessings to be honest..

> and I can very well believe that the SRM is using them, but I have a
> hard time understanding why this would affect the fact that it's
> sending out bogus UDP packets. What usually happens is that the kernel
> throws those packets away before any application (such as
> rpc.bootparamd) even gets them.
> 
> Are you sure you're not running your bootparamd on a SunOS machine (or
> another OS that doesn't check UDP checksums)?

No, this was bootpd running on a FreeBSD machine (an Intel) serving
a DEC 3000/400 Sandpiper. I don't have a SunOS box anymore ;-)

Wilko
-- 
|   / o / /  _  	Arnhem, The Netherlands	  - Powered by FreeBSD -
|/|/ / / /( (_) Bulte 	WWW : http://www.tcja.nl  http://www.freebsd.org