If the router at the other end is non-Cisco then you'll have to specify IETF
encap for the Frame encapsulation.  The frame relay encapsulation type
should be independent of the LMI type.  If the telco is using non cisco
switches then you just set your lmi type to whatever they are using.  The
DLCI being passed is a function of LMI it has nothing to do with Cisco or
any cisco tampering with their LMI code it should work with cisco, Ansi, and
Q933a LMI types.  This should all work in any live environment unless there
is a non-standard configuration ie the telco is not advertising LMI at all
so you have to put "no keep" under the interface.

Steve

-----Original Message-----
From: Joseph Ezerski [mailto:[EMAIL PROTECTED]]
Sent: Monday, January 15, 2001 1:26 PM
To: '[EMAIL PROTECTED]'; [EMAIL PROTECTED]; [EMAIL PROTECTED]
Subject: RE: Frame Relay...Inverse-Arp..?


I think you got it right, but just to clarify....

The frame switch looks like this:

        Interface Serial0
        encapsulation frame-relay
        frame-relay intf-type dce
        frame-relay route 20 interface serial1 30

        Interface Serial1
        encapsulation frame-relay
        frame-relay intf-type dce
        frame-relay route 30 interface serial0 20

Remote CPE #1 looks like this:

        Interface Serial0
        ip address 192.168.1.1 255.255.255.252
        encapsulation frame-relay
        (notice no DLCI is assigned)

Remote CPE #2 looks like this:

        Interface Serial0
        ip address 192.168.1.2 255.255.255.252
        encapsulation frame-relay
        (notice no DLCI is assigned)

My ping tests work between the two CPE's.  My best guess is that the LMI is
passing the important information across the point-to-point links.

As for doing this with the Telco, I am not sure.  Many times they use non
Cisco switches, so IETF encapsulation is needed and/or use a different LMI
type.  But why don't you try it and report back?

Joseph

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Monday, January 15, 2001 10:57 AM
To: [EMAIL PROTECTED]; [EMAIL PROTECTED]
Subject: Re: Frame Relay...Inverse-Arp..?


Ok, now you all got me interested.

#1) From what I'm reading, you are saying that if you create a FR lab and
just place the DLCI map commands on the frame switch for the remote ends,
and
you have the LMI set right, then the remotes will automatically learn their
DLCIs through LMI?

#2) Would this also work in a real environment concerning telco assigning
you
DLCI numbers? What I'm saying is, after the telco gives you the DLCIs for
your DTE interfaces (remote ends) usually you would go in and configure
them.
If you didn't bother to go and configure them, would your remote ends
automatically learn them from the FR switch somewhere in the telco's network

through LMI???

In a message dated 1/15/01 1:49:41 PM Eastern Standard Time,
[EMAIL PROTECTED] writes:


> that should be true no matter what.  You are learning the DLCI's through
LMI
> a standard function of Frame Relay.  You will learn it regardless as long
as
> your router and the switch agree on what type of LMI they are using.
>
> Steve
>
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of
> Joseph Ezerski
> Sent: Monday, January 15, 2001 9:59 AM
> To: 'Nigel Taylor'; Cisco Group Study; Chuck Larrieu; CCIE_Lab Group
> Study
> Subject: RE: Frame Relay...Inverse-Arp..?
>
>
> I have an interesting tidbit that I hope can add to this thread.  I was
> doing a frame relay lab and set up a 2501 as a frame switch.  I had two
> other 2501s as the remote ends.  I set up the major interface for frame
> relay, but did NOT specify a DLCI number for both remote ends.  In the
frame
> switch, I added my frame-relay route commands for the respective major
> interfaces and DID specify DLCI numbers.  What I found out is that the
> remote ends "learned" their DLCI's from the frame switch.  I remember in
my
> studies that using the default Cisco LMI type buys you some auto discovery
> of sorts.  I am not sure if this is playing a part in Nigel's experiment,
> but why not give it a go by changing the default Cisco LMI type to some
> other type and see if the results stays the same.
>
> Joseph
>



_________________________________
FAQ, list archives, and subscription info:
http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]

_________________________________
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]

Reply via email to