Frame Relay Question..

Dokudorf

I keep mixing up the following and their purpose (when to use/not use em):

IETF
ANSI
q933a

Anyone got a minute to edumacate me on them? Any assistance appreciated, thanks.

Taking my CCNA exam thursday and this is one of the last things on the stuff i keep mixing up on the practice exams Cisco stuck with their book. Textbook has the stuff but 'real world' definitions have more merit imo as i usually take things too literaly when i textbooks and miss the hidden meanings/work-arounds.

-Jon

sikdogg

It appears that you getting confused between Frame Relay encapsulation types with the Frame Relay LMI types. I'll try to explain the two.

Cisco, ANSI, and q933a are Frame Relay LMI types. LMI is a signaling standard between a CPE device (your router) and the Frame relay switch at the telco side. LMI are responsible for a couple of things including: support for a keepalive mechanism, which verifies that data is flowing, and a status mechanism, which provides an on-going status report on the DLCIs known to the switch.

Here's a description of the three LMI types:
Cisco - was jointly developed by Cisco, StrataCom, Nortel, and DEC
ANSI - the annex D of the ANSI T1.617 standard
q933a - the annex A of the ITU-T Q.933 standard

The type you use is based on the type the Frame Relay switch at the telco is using. You can manually configure the LMI type on an interface using the following syntax in the interface config mode (default is Cisco, of course):

frame-relay lmi-type [ansi | cisco | q933a]

Because you are setting the LMI type in the interface, this means that you can have different LMI types on each interface.

Beginning with IOS release 11.2, routers began support of autosensing for LMI types. My experience with Frame Relay is such that I have never had to manually configure the LMI type.


Frame Relay encapsulation has to do with the framing of the Frame Relay packet prior to transmitting on the wire. The two type of encapsulations are IETF and Cisco (the default, of course). IETF is based on the RFCs 1294 and 1490. It uses a two-byte header. According to Cisco's website, you would use this method of encapsulation for backward compatibility and interoperability.

Cisco's encapsulation deviates from the standard in that instead of a two-byte header, Cisco uses a four-byte header. Two-bytes for the DLCI and two-bytes for the packet type.

Here's the syntax for configuring the Frame Relay encapsulation type (in interface config mode):

encapsulation frame-relay [cisco | ietf]

or

frame-relay map ip x.x.x.x DLCI [broadcast] [cisco | ietf]

Like the LMI types, Frame Relay encapsulations are configured at the interface. This means that all VCs that come from an interface will use the same encapsulation type. If all routers used were from Cisco, you could use Cisco encapsulation without any problems. If on the other hand, one or more routers used to terminate a Frame Relay VC is NOT Cisco, you must use IETF encapsulation.

To sum it up for you, LMI has to do with the communication between CPE (your router) and the Frame Relay switch. Frame Relay encapsulation has to do with communication between two CPE devices (end-to-end traffic).

Kinda long winded, but I hope it helps.