[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: IPComp CPI's



At 09:33 11.10.2001 -0400, steve@tril-inc.com wrote:

>Hello,
>
>I have a question regarding the negotiation of IPComp CPI's vs. the use of 
>well known values. In looking at the IPComp RFC there seems to be some 
>ambiguity regarding the use of well-known CPI's (the CPI indicating the 
>compression algorithm) vs. negotiated CPI's.
>
>If an initiator of Phase2 sends a well-known CPI for IPComp, what are the 
>requirements of the responding node? Can the responding node send a 
>response with a CPI in the negotiated range?

yes.

>If Phase2 completes with the initiator using a well-known CPI and the 
>responder is using a CPI in the negotiated range, is the initiator 
>required to use the CPI that was negotiated during phase2?

Each host can choose a CPI, and that will be used for the traffic he receives.
If the initiator uses CPI 2 and the responder CPI 50000 in QM, then all 
traffic from i to r
will carry the CPI 50000, and all traffic from r to i the CPI 2.


>Thanks for any insight.
>
>Steven Erickson
>Senior Software Engineer
>Trilogy, Inc.
>978-371-3980 x112
>http://www.tril-inc.com

J–rn


References: