[e2e] ICMP & TCP segments with IP ID = 0?

[Alex Cannara]

Just to point out again that the IP4 ID has been extremely useful for
years in quickly detecting loops in hardware, whether just due to hub
wiring, switching/bridging misconfig, or routing issues.  IP6 makes a
fundamental mistake by not requiring it.

Alex


Craig Partridge wrote:
> 
> In message <20010516111231.A16714 at host-186.dyn.dev.zks.net>, Jerome Etienne wri
> tes:
> 
> >what is so evil in setting the IPv4 id to 0, assuming the DF bit is set ?
> 
> I said it was "evil" because it is not in compliance with the specifications
> (namely RFC791 and 1122).
> 
> Interestingly enough, while RFC 1122 certainly assumes that a new identifier is
> normally required for each datagram, it doesn't say so.
> 
> But RFC 791 is quite clear:
> 
>     The originating protocol module of
>     an internet datagram sets the identification field to a value that
>     must be unique for that source-destination pair and protocol for the
>     time the datagram will be active in the internet system.
> 
> So while 0 is a valid value, you can't repeat it for the lifetime of
> a datagram or so (RFC 1122 permits reuse in the narrow case of retransmitting
> exactly the same data).
> 
> Craig