jumbograms (& em) & nfs a no go

[Barney Wolff]

On Sat, Nov 01, 2003 at 01:09:07PM -0800, Terry Lambert wrote:
> Barney Wolff wrote:
> > > Implies the sending host is not honoring the MTU restriction when
> > > deciding whether or not to frag packets.
> > 
> > 67582 looks awfully bogus even as a pre-frag length.  How could that come
> > over the wire?
> 
> The sending host is not honoring the MTU restriction?
> 
> 8-) 8-).
> 
> Most likely, a direct call to ether_output, or a code path that
> results in fragmentation not being implemented; see my other post:
> it could be that he's using NFS over UDP, and that's doing it.

Er, how is it possible to send a UDP packet > 65535?  Last time I looked
it was a 16-bit field.

-- 
Barney Wolff         http://www.databus.com/bwresume.pdf
I'm available by contract or FT, in the NYC metro area or via the 'Net.