I was hoping that question would catch your eye :). I have been struggling with this for a while. I will research the link you gave me, and if I still have questions I will post them. Thanks, I appreciate it.
-----Original Message----- From: Priscilla Oppenheimer To: [EMAIL PROTECTED] Sent: 1/2/2002 10:35 PM Subject: Re: Protocol type codes and SAP's [7:30744] It sounds like you already realize that you should understand SAPs from a general point of view, not just from the viewpoint of a single CCIE DLSw lab task that involves them. SAPs are a fundamental aspect of networking. They have a generic meaning as well as a specific meaning in the IEEE 802.2 Logical Link Control (LLC) layer. I strongly recommend perusing the IEEE specification which is available here: http://standards.ieee.org/getieee802/ At each protocol layer, there is typically a numeric identifier to tell the recipient protocol handler what is coming next in a frame. The recipient uses the identifier to decide which process should receive the frame. IP has a field called the Protocol Type that differentiates between UDP and TCP and other protocols carried in an IP header. TCP has a port number that differentiates between Telnet, Mail, Web access, and other applications. Ethernet II uses the EtherType for protocol identification. IEEE LANs use an 802.2 Service Access Point (SAP), specified by the 802.2 Logical Link Control (LLC) specification available from IEEE. The LLC header has three fields: 1 Destination Service Access Point (DSAP). The DSAP identifies the service to which the frame is destined. It is one byte. An example is 0xF0 for NetBEUI. 2 Source Service Access Point (SSAP). The SSAP identifies the sending service, which is almost always the same as the destination service, although not necessarily (especially not with SNA). It is one byte. 3 Control. The Control field identifies the type of frame as information transfer, supervisory, or an unnumbered command or response. The format of the Control field depends on the mode of communication service. With connection-oriented communication, the control field is two bytes to accommodate an acknowledgment number. With connectionless service, the control field is just one byte. In most cases, a well-known protocol has just one SAP. For example, IP is 0x06. NetWare is 0x0E. SNA has a few SAPs, which makes it kind of tricky. One SAP gets used during session establishment. Others get used once the session is established. SNA SAPs include: 0x04, 0x05, 0x08, 0x0C. (There is an issue with bit-ordering also, so I don't know if the ones I list are in the same order as the list that you are using and I have run out of steam to look into that, but hopefully this has been helpful. Let us know if you still have questions. Bottom line: search on LLC or 802.2 on Google ;-) Priscilla At 09:26 PM 1/2/02, Lupi, Guy wrote: >Can anyone explain the ethernet and token ring protocol type codes for me? >I know you can filter DLSw and SRB using them but I don't really understand >what they are, I found a table containing all the popular ones. IBM SNA is >80D5, does that mean that all SNA traffic has this type code, and that if >you denied it in an access list all SNA traffic would be denied? Also, >SAP's follow the format 0x0000, each of the 0s representing one byte of >data, the first 2 digits after the x are the DSAP, the last 2 are the SSAP, >is this correct? I would really appreciate it if someone could point me to >where I could find some good information on this type of stuff. I would >hate to just memorize the common values without knowing why and how they >work. Thanks. ________________________ Priscilla Oppenheimer http://www.priscilla.com Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=30757&t=30744 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
