Suggestion for improved identD
Phil Howard
phil at charon.ipal.net
Wed May 20 04:52:46 UTC 1998
> >There isn't necessarily just a single user on the other end of a PPP
> >connection.
>
> Perhaps I should have phrased it as "single user network
> connection" and not "PPP". I'm less concerned with the
> PPP as a protocol than as its modern usage to connect the
> dialup user.
And how do you tell the difference between a single user connection
and multi-user connection? They both use PPP. Are you going to make
all the Linux users out there have to start negotiating with their
ISPs just to allow them to be on?
> >Many things will break if the actual user and the user
> >that PPP intercepted identd asserts do not match.
>
> Oh?
Yup. IRC bots, for instance. They expect certain specific information
to grant authority, and if the PPP server substitutes it, it can't be
correct all the time on systems with two or more users since the PPP
server won't know which user is on which port (without actually going
to that machine to ask ... but then what's the point).
> >Providing such information may be a violation of confidentiality if
>
> Login string. e.g. username.
Dialup account id? Unfortunately this is usually also the e-mail
address by just appending @isp-domain.net and thus giving out tons
of addresses to spammers.
I won't subject my customers to this.
> >Because the PPP access device cannot know, unless it also tracks all the
> >traffic involved, what ports are in fact in use, it would have to give
>
> If l2 is up, it's up. That's fairly basic...
So if I request an ident on port 15421, is the PPP server going to answer
it even though, there is in fact no active port 15421 on that machine?
You want PPP servers to track all those SYN and RST?
> >I believe you misunderstand the purpose of identd. It was intended to
> ...
> Nope...
So you do understand that the data wasn't intended to be trusted if
you have no trust of the machine (and certainly most of them out there
cannot be trusted).
> >Why do you want this data?
>
> My personal crusade against packet monkeys, spammers, and
> irresponsible admins who support them by pretending that
> the net is free for all to abuse.
I applaud the goals. I don't think this is a viable mechanism to
achieve them.
BTW, I blocked access to SMTP other than to my own servers for all my
dialup non-LAN customers. I don't like abuse, and won't put up with
it, either from my customers, or to them. But this identd idea is not
something I will do to my customers. The cure is worse than the disease.
The answer is simple. Don't trust identd responses. Just don't ask
for that data and then you don't have to worry about it being forged.
--
Phil Howard | no7way44 at noplace5.org ads2suck at dumbads8.com a4b3c4d5 at no0place.org
phil | no6spam2 at spam3mer.com end3it15 at no16ads4.edu stop6it8 at dumb9ads.net
at | end6it02 at s0p9a9m3.edu stop2ads at nowhere8.org end8ads7 at spammer1.com
ipal | no8way90 at no86ads8.net blow3me2 at dumb9ads.net w5x2y2z9 at spam8mer.com
dot | stop1020 at spammer5.edu stop7317 at no5place.net no4spam7 at anyplace.com
net | eat55me9 at spammer5.org no71ads8 at lame9ads.net suck4it6 at dumb7ads.org
More information about the NANOG
mailing list