Re: [MBONED] Barry Leiba's No Objection on draft-ietf-mboned-driad-amt-discovery-10: (with COMMENT)

[Barry Leiba <[e-mail]>]

Hi, Jake, and thanks for the response.

> NEW:
>    In this simple example, the sender IP is 198.51.100.15, it is sending
>    traffic to the group address 232.252.0.2, and the relay ...

Excellent; thanks.

> >    Among relay addresses that still have an equivalent preference after
> >    the above orderings, a gateway MUST make a non-deterministic choice
> >    for relay preference ordering, in order to support load balancing by
> >   DNS configurations that provide many relay options.
> >
> > What do you have in mind here?  Random selection?  Something else?  If so, what
> > else that can be assured to be non-deterministic, given the "MUST"?
>
> I was thinking "random or pseudorandom" and thought this phrasing was
> cleaner.
>
> I can change it to "random or pseudorandom", or if there's a better or more
> normal way to phrase it I'd be happy to use that.  (I was worried "random" is
> too strong, if taken literally.)

I think "non-deterministic" is fine; I was mostly wondering whether
you had other things in mind.  It wouldn't be unreasonable to make the
text say something like "a non-deterministic choice (such as a
pseudorandom selection)", but use your judgment and no need to reply
further here.

Barry