Re: [Hipsec] Re: The exact method used to generate a HIT from a HI

On Tue, Mar 18, 2003 at 04:05:35PM -0800, Tom Henderson wrote:
> > My personal opinion would be to use the RFC2536 DSA format

Well I don't see using the DNS format as any advantage at all, as the code to parse this packet (in the implementation I'm thining of) would have to be completely seperate form the resolver code anyway, so there is no point in adding a complex packet format.

> But I do think that either an octet or four is needed

Agreed, that's why I've been suggesting the format in -01 version of the optimised packet formats, since it had a one byte field anyway.

I don't see alignment as being an issue for those fields anyway (the actual key data) as it has to be unpacked form what could be any arbitrary layout, hence allignement becomes somewhat moot.

DF