Layer2 Framing questions (ethernet)

This is my understanding of ethernet layer 2 framing methods:

802.3 raw   =     Does not have a type field.

802.2 w/LLC=     Uses LLC to determine which protocol (ip, ipx etc) to hand to at layer3

802.2 SNAP =      Why was it made? Was it created after LLC?

Ethernet II  =      Includes a type field to identify upper layer

My questions are:

1.  When I capture SNAP frames, why am I seeing LLC fields in it?  I thought 802.2 SNAP and 802.2 LLC were different beasts? Or does SNAP expand upon LLC in some way?

2. Is it possible to have differnet frame formats, all in one broadcast domain, communicating with each other?
    Or can hosts framing with 802.3 only communicate with 802.3 hosts....... vice versa....
Who is Participating?
grbladesConnect With a Mentor Commented:
Hi dissolved,
The 802.2 frame is formatted as follows:-

8 bytes - preamble
6 bytes - destination mac address
6 bytes - source mac address
2 bytes - length
variable - data
4 bytes - checksum

In Ethernet II the 'length' filed is not used but instead this are is used for the 'Type' such as IP or ARP etc...

Within the 'Data' area you can have either SNAP or SAP LLC frames.
SAP has a 1 bytes source SAP, 1 byte destination SAP, 1 or 2 byte control and a variable data area.
SNAP came about later and has additional OUI and Type fields so that it works better with TCP/IP which requires the type field. In orderr to identify a SNAP frame the destination and source SAP frames are both set to AA.
I'm smiling....this was a test question on the CCIE exam years ago.  To expand a little further, the 'AA' stands for Alien Address, because you expect a SNAP (Sub Network Address Protocol) destination (and usually source) in your data, for things like, oh, Appletalk Phase II....something just out of reach of SAP.

grblades, I love elegant straightforward answers, and yours qualifies :)
All Courses

From novice to tech pro — start learning today.