ipid-debuggery (was Re: [e2e] ICMP & TCP segments with IP ID = 0?

[Bob Braden]

  *> 
  *> An IP ID of 0 is legal as any other value.
  *> 
  *> Incrementing IP  ID when DF is set  is not mandated by  any document, as
  *> far as I know, and not even recommended.
  *> 

Francesco,

I am confused by your claim.  Near the beginning of this thread,
someone quoted RFC 791 as saying exactly that incrementing the IP ID is
(always) required.  It did not qualify this requirement by whether or
not DF is set.  I said that I strongly suspected  was an oversight, but
subsequent discussion has suggested to me that I might be wrong about
this, and that Jon may in fact have been strictly and deliberately
correct in RFC 791; being conservative in what you send would seem to
require always incrementing the ID.   In any case, I cannot understand
your claim that it is not mandated by any document.

  *> If the architect of an IP layer can nonetheless increment it without too
  *> much effort, that is a good think,  because it turns out to be useful in
  *> debugging, tracing, measuring, experimenting.

And because it increases the system's robustness against failures caused
by liberties (like turning off DF without resetting the ID)
that boxes in the middle of the network might take.

Bob Braden

  *> 
  *> -- 
  *> Francesco Potortì (researcher)         Voice: +39 050 315 3058 (op.2111)
  *> Area della ricerca CNR - CNUCE         Fax:   +39 050 313 8091
  *> via Alfieri 1, I-56010 Ghezzano, Pisa  Email: F.Potorti at cnuce.cnr.it
  *> Web: http://fly.cnuce.cnr.it/          Key:   fly.cnuce.cnr.it/public.key
  *>