Re: cancel-tcpip-forward matching

[der Mouse <mouse%Rodents.Montreal.QC.CA@localhost>]

>>> Or is all this left up to the implementations?
>> i would only cancel on exact string matches.  but this is left up to
>> the implementation.
> I don't think it's so much up to the implementation... I think this
> is the Correct Way.  The idea being - what you effect with one
> request, you can cancel with a symmetric one, without having to
> understand the implementation semantics.

Yes; I agree that if the same string is given, the preceding request
should be undone.

My question really bears more on what happens when a different string
is given.  _Must_ the server reject the cancel attempt, or is it
permitted to possibly cancel previous requests, or portions of previous
requests, that match the cancel request it in an implementation-defined
way?

> I think an implementation that did not adopt this idea would be
> broken... The whole idea of standards is to get consistent output
> when supplying consistent input.

For what it may be worth, my implementation as it stands does not quite
do this.  For numeric and wildcard forwarding, I think it does - but if
the host string is a FQDN, and the set of addresses it resolves to at
cancel time is not the same as the set of addresses it resolves to at
request time, the cancel may not cancel exactly and only the
forwardings established by the forward request.

I'm not sure whether I consider this behaviour broken.  After all,
telnet and the DNS are standardized too, and yet nobody thinks it
broken that "telnet foo.example.net" may connect you to two different
hosts at different times if the DNS has changed in between.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse%rodents.montreal.qc.ca@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B