headers
Bernard Aboba | 3 Sep 2002 18:49

Interim meeting on EAP state machine

Unfortunately, it doesn't look like we will be able to pull off an interim
meeting. Due to schedule conflicts, it has proven difficult to get all of
the people involved in state machine drafts (EAP and 802.1aa D3)
together on the same date.

Instead, we'll try to manage discussion of the EAP State machine issues on
the mailing list and at IETF 55.
Permalink | Reply | 0 comments |
headers
Bernard Aboba | 3 Sep 2002 18:50

EAP WG Agenda for IETF 55

IETF 55 scheduling has begun. If you have an agenda item for the IETF 55
meeting of the EAP WG, please post to Jari or myself or the list.
Permalink | Reply | 0 comments |
headers
CONGDON,PAUL (HP-Roseville,ex1 | 6 Sep 2002 21:58
Picon
Favicon

RE: Creating a design team for EAP state machine

I am interested in participating on this team.  I will be traveling next
week, but should be within reach of email now and then

Paul

+------------------------------------------+
Paul Congdon
HP ProCurve Networking
Hewlett Packard Company
8000 Foothills Blvd - M/S 5662
Roseville, CA   95747
phone: 916-785-5753
email: paul_congdon <at> hp.com
+------------------------------------------+

> -----Original Message-----
> From: Jari Arkko [mailto:jari.arkko <at> piuha.net] 
> Sent: Friday, September 06, 2002 12:57 PM
> To: eap <at> frascone.com; bdpayne <at> cs.umd.edu; 
> npetroni <at> cs.umd.edu; Glen Zorn; yohba <at> tari.toshiba.com; 
> jrv <at> interlink.com; Robert Moskowitz; CONGDON,PAUL 
> (HP-Roseville,ex1); William Arbaugh
> Cc: Bernard Aboba
> Subject: Creating a design team for EAP state machine
> 
> 
> 
> RFC 2284 does not include a state machine, and as a result, 
> questions have arisen about the behavior of the protocol, 
> which are not addressed in that document. Given the revision
(Continue reading)

Permalink | Reply | 0 comments |
headers
Jari Arkko | 6 Sep 2002 21:56

Creating a design team for EAP state machine

RFC 2284 does not include a state machine, and as a result, questions
have arisen about the behavior of the protocol, which are not addressed
in that document. Given the revision to the specification RFC 2284bis,
the desire is to develop a separate document describing in detail the
state machine behavior of EAP.

We have considered various ways to organize the production of the
state machine. A draft about the state machine exists, but some work
still remains on e.g. harmonizing the state machine with IEEE work.
We considered an interim meeting, but it proved to be hard to get
a sufficient number of people in the same place at the same time.

Bernard and I have decided that we would like to create a short-term
design team to work on the state machine. We are soliciting members
for the design team. If you are willing to help, let us know. We
are hoping the design team could be in place within a week or so,
and could include those who have worked on the current state machine
as well as folks who have knowledge of other aspects of the problem.
Note that as participation in IETF is open, folks working on
EAP-related issues in other standards bodies are also welcome.

The output of the design team is an IETF draft on the EAP state machine,
to be presented to the EAP WG at IETF 55. Desirable aspects of the state
machine include:

a. Rationalized with the IEEE 802.1aa D3 state machine, so that either
    the two state machines are compatible, or the differences are well
    understood and it can be explained why the IEEE 802.1aa D3 state
    machine needs to change.
(Continue reading)

Permalink | Reply | 3 comments |
headers
Bernard Aboba | 6 Sep 2002 22:33

Re: Creating a design team for EAP state machine

> I have one question.  I remember that the URL for the 802.1aa D3 draft
> was posted on to this mailing list at the end of August with username
> and password.  Can I legitimately use that credentials to access the
> URL though I'm not a member of the IEEE 802.1 WG?  Or all the design
> team members are assumed to be a member of the IEEE 802.1 WG?
>
> Regards,
> Yoshihiro Ohba

The IETF is an open organization, and it is not required that working
group members attend meetings (either of the IETF or any other group) in
order to participate. Our specs are openly available, in order to
encourage the maximum amount of review.

Given this, the ideal solution is for all EAP WG members to be granted
access to the 802.1aa archive, so that they can have access to the 802.1aa drafts in
progress. This will allow the two groups to collaborate on the state
machine work in the most seamless way possible. Formal liason between the
can be handled by discussions between the EAP WG chairs and the
officers of 802.1.

I have brought this issue up with Tony Jefree of 802.1. I'd note that
collaboration between IEEE and IETF groups is not new, so these kind of
issues have been handled before in the Bridge WG, and IPORPR WG. Another
example is the collaboration between IEEE 802.11 and 802.1; between 802.11
and ETSI BRAN, etc.
Permalink | Reply | 0 comments |
headers
Bryan D. Payne | 7 Sep 2002 00:44

Re: Creating a design team for EAP state machine

Please count me in on the design team.  My time is rather limited these
days, but I'd like to help out where I can.  Email is the best medium
for me right now...

-bryan
Permalink | Reply | 0 comments |
headers
Yoshihiro Ohba | 6 Sep 2002 22:54

Re: Creating a design team for EAP state machine

Yes, I am interested in working on EAP state machine and I'm 
happy if I can contribute to it.

I have one question.  I remember that the URL for the 802.1aa D3 draft
was posted on to this mailing list at the end of August with username
and password.  Can I legitimately use that credentials to access the
URL though I'm not a member of the IEEE 802.1 WG?  Or all the design
team members are assumed to be a member of the IEEE 802.1 WG?

Regards,
Yoshihiro Ohba

On Fri, Sep 06, 2002 at 10:56:37PM +0300, Jari Arkko wrote:
> 
> RFC 2284 does not include a state machine, and as a result, questions
> have arisen about the behavior of the protocol, which are not addressed
> in that document. Given the revision to the specification RFC 2284bis,
> the desire is to develop a separate document describing in detail the
> state machine behavior of EAP.
> 
> We have considered various ways to organize the production of the
> state machine. A draft about the state machine exists, but some work
> still remains on e.g. harmonizing the state machine with IEEE work.
> We considered an interim meeting, but it proved to be hard to get
> a sufficient number of people in the same place at the same time.
> 
> Bernard and I have decided that we would like to create a short-term
> design team to work on the state machine. We are soliciting members
> for the design team. If you are willing to help, let us know. We
> are hoping the design team could be in place within a week or so,
(Continue reading)

Permalink | Reply | 1 comment |
headers
Bernard Aboba | 7 Sep 2002 17:06

802.1aa Access Rights granted to EAP WG members


---------- Forwarded message ----------
Date: Sat, 07 Sep 2002 16:57:43 +0100
From: Tony Jeffree <tony <at> jeffree.co.uk>
To: Bernard Aboba <aboba <at> internaut.com>
Subject: Re: Creating a design team for EAP state machine (fwd)

Bernard -

There is no problem - they are welcome to make use of our username/password
to access the drafts directly from the 802.1 website:

Username: p8021
Password: go_wildcats

Regards,
Tony

At 13:40 06/09/2002 -0700, you wrote:
>Question for you: In order to ensure compatibility between the EAP state
>machine and the 802.1aa state machine, we'd like to allow EAP WG members
>to have access to the 802.aa drafts. Is there a way to ensure that WG
>members can obtain this access?
Permalink | Reply | 0 comments |
headers
Bernard Aboba | 9 Sep 2002 14:42

RE: Creating a design team for EAP state machine

The IEEE 802.1aa document can be retrieved from:

http://www.ieee802.org/1/pages/802.1aa.html

userid: p8021
password: go_wildcats
Permalink | Reply | 1 comment |
headers
Bernard Aboba | 10 Sep 2002 12:44

Re: Question on EAP

> Sorry for those who get this twice. I am a starter of PPP and related
> extentions. I have a question based on the case below,

Note: We now have an EAP WG -- so that this is the appropriate forum for
discussion on EAP. The AAA WG is focussed on development of the Diameter
protocol. However, it appears that your question centers on 802.11,
so that the best place to discuss this topic is one of the
802.11 task groups, such as IEEE 802.11i.

> The authentication server is not only providing
> authentication service but registration services.

EAP is an ACK/NAK protocol, so it is not particularly well suited to
supporting a complex registration process requiring substantial data
transfer.

> And the registration service of the authentication server needs to know the
> user's status in the network (registered or not). Therefore, it relys on
> some sort of information reported by the authenticator which is close to the
> terminals

Is the "Authenticator" an AP or NAS device? Does it speak 802.11/EAP to
the terminal?

> When the terminal powers off, we want the authenticator to detect this

In EAP there is now way for a peer to send a "disconnect" to an
authenticator.

Before the STA goes into "802.11 power save mode" it can communicate this
(Continue reading)

Permalink | Reply | 2 comments |

Gmane