[Babel-users] hmac merge

[Dave Taht]

turning keys off and unicast on, I am interoperating with babel head
and I'm going to dogfood it for a wh

my attempt at unifying these is in the hmac-merge branch on
https://github.com/dtaht/babeld-hacking

there were a few puzzling things in the merge - num_ll? send with a "to"?

but I do seem to be getting rtt correctly with keys disabled from the
non-hmac version interoperating with babel head... (I was concerned
I'd broke that)

somewhere around here is toke's patch for bird??
On Mon, Nov 12, 2018 at 5:59 AM Dave Taht <dave.taht at gmail.com> wrote:
>
> So I got around to testing this merge on two boxes this morning. It's
> rather noisy when run on a mixed network, but the two boxes I got
> running seem to be exchanging routes on
>
> key id key1 type sha1 value deadbeefdeadbeefdeadbeefdeadbeefdeadbeef
> default enable-timestamps true hmac key1
>
> I would probably ratelimit stuff like this:
>
> static unsigned ratelimit = 0;
> if(ratelimit++ % 1024) fprintf(stderr,"Received wrong hmac from box %x", box);
>
> 'cause
>
> ./rtod -r 4096
>
> from an non-hmacing box is rather noisy.
>
> ...
>
> Received wrong hmac.
> Received wrong hmac.
> Received wrong hmac.
> Received incorrect RTT sub-TLV on IHU.



-- 

Dave Täht
CTO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-831-205-9740