IETF to make DIGEST-MD5 mandatory-to-implement

So sorry, everybody who uses TLS+PLAIN authentiation.

Because some mobile phone makers have trouble implementing SSL/TLS, the
IETF plans to replace TLS+PLAIN as the mandatory-to-implement
authentication mechanism in SMTP in favor of DIGEST-MD5 (which requires
the server to store plaintext equivalents of your passwords - a hacker's
dream!). Work is in progess to do the same for POP3, and eventually IMAP
will have to join the crowd.

See draft-siemborski-rfc2554bis-06.txt

Enjoy your reduced security.

-- Mark --

http://panda.com/mrc
Democracy is two wolves and a sheep deciding what to eat for lunch.
Liberty is a well-armed sheep contesting the vote.

Re: IETF to make DIGEST-MD5 mandatory-to-implement

Mark Crispin wrote in
news:alpine.OSX.0.81.0701151338290.5567@pangtzu.panda.com:

> So sorry, everybody who uses TLS+PLAIN authentiation.
>
> Because some mobile phone makers have trouble implementing SSL/TLS, the
> IETF plans to replace ...

Mark, lol, somehow methinks the IETF can not dictate the technologies that
independent developers choose to use. I'm sure we will keep doing what we
are doing :)

--
Jem Berkes
www.sysdesign.ca

Re: IETF to make DIGEST-MD5 mandatory-to-implement

In Mark Crispin writes:

>So sorry, everybody who uses TLS+PLAIN authentiation.

>Because some mobile phone makers have trouble implementing SSL/TLS, the
>IETF plans to replace TLS+PLAIN as the mandatory-to-implement
>authentication mechanism in SMTP in favor of DIGEST-MD5 (which requires
>the server to store plaintext equivalents of your passwords - a hacker's
>dream!). Work is in progess to do the same for POP3, and eventually IMAP
>will have to join the crowd.

>See draft-siemborski-rfc2554bis-06.txt

So, that means that somebody will have to devise a way to store
plaintext passwords on a server in a secure manner. That's a good
idea.

--
-Gary Mills- -Unix Support- -U of M Academic Computing and Networking-

Re: IETF to make DIGEST-MD5 mandatory-to-implement

On Mon, 15 Jan 2007, Gary Mills wrote:
> So, that means that somebody will have to devise a way to store
> plaintext passwords on a server in a secure manner. That's a good
> idea.

If a server can get at it, so can a bad guy. The bad guy just has to
figure out how to convince the system that he is a server. That is the
fundamental weakness of any form of plaintext-equivalent password storage.

-- Mark --

http://staff.washington.edu/mrc
Science does not emerge from voting, party politics, or public debate.
Si vis pacem, para bellum.

Re: IETF to make DIGEST-MD5 mandatory-to-implement

On Mon, 15 Jan 2007, Jem Berkes wrote:
> Mark, lol, somehow methinks the IETF can not dictate the technologies that
> independent developers choose to use. I'm sure we will keep doing what we
> are doing :)

The problem is that then you can't advertise your software as being
compliant with the standards.

What's more, you can not simultaneously implement DIGEST-MD5 and TLS+PLAIN
in the server unless you either have separate passwords (depending upon
the authentication method) or store all passwords as plaintext-equivalent.

This is part of the reason why CRAM-MD5 didn't get anywhere. DIGEST-MD5
is better (a lot better) but it doesn't address the weakness.

DIGEST-MD5 should be optional, with TLS+PLAIN being mandatory. The
proferred excuse, that TLS is too hard for some people to implement,
doesn't wash. Perhaps we need more widespread distribution of hunt and
other session hijacking tools to convince people that TLS is no longer
optional.

-- Mark --

http://staff.washington.edu/mrc
Science does not emerge from voting, party politics, or public debate.
Si vis pacem, para bellum.

Re: IETF to make DIGEST-MD5 mandatory-to-implement

Mark Crispin wrote in
news:alpine.WNT.0.81.0701151443290.1820@Shimo-Tomobiki.panda .com:

> DIGEST-MD5 should be optional, with TLS+PLAIN being mandatory. The
> proferred excuse, that TLS is too hard for some people to implement,
> doesn't wash. Perhaps we need more widespread distribution of hunt and
> other session hijacking tools to convince people that TLS is no longer
> optional.

I am actually amazed by all this, in this day and age... I thought that
enough "convincing" examples were out there. Very real, simple
hijacking/spoofing attacks on wireless LAN connection for example.

--
Jem Berkes
www.sysdesign.ca