Wed Jun 1 11:16:50 CEST 2005
let me make sure that I fully understand your scenario. You have notebooks
A, B, C, and D. A and B see each other, B and C see each other, C and D
see each other.
You start out with an instance of olsrd on A. Then you start olsrd on B
and wait for the route to appear on node A, which takes 8 - 9 seconds.
Then you do the same for the other nodes. First your start olsrd on C and
wait for the route to C to appear on A. Then you start olsrd on D and wait
for the route to D to appear on A. Each new route takes 8 - 9 seconds to
appear on A.
That's quite interesting, as the route to B is a result of B being A's
one-hop neighbour. The route to C is a result of C being a two-hop
neighbour of A. And the route to D is a result of a TC message issued by
C, which is an MPR for D. Three different mechanisms and always 8 - 9
Could you run olsrd on A at a debug level of 9 and also add the "-dispin"
and "-dispout" command line options when starting it? The command line
options will let us see the OLSR packets that olsrd receives and sends.
The debug output will help in tracking down this issue.
More information about the Olsr-dev