[Olsr-users] Route convergence time: a function of scale?

[Breno Jacinto]

Hello,

     This is my first post, nice to be here sharing knowledge with you
guys. I have a question that has been in my head for a while.

     Consider that there's an OLSR network, and some nodes (20, for
exemple) are part of it. Now, given that OLSR is a pro-active
protocol, it's assumed that, for those 20 nodes, each has a route for
each other, whether single-hop(they are neighbors) or multi-hop (some
nodes make a node see another, more distant node).

     OK, now, consider that I wish to test if a given node address,
192.168.10.15, is on the network. So:

# ping 192.168.10.15

     Which, if my routing table is up-to-date *and* connectivity
(meaning, layer 1 and 2 conditions) is fine up to that node, then I'll
get a reply. My point is: what if I get a Host Unreachable reply? Is
it because:

1) The node really isnt in the network - that IP is not used.

2) The node is in the network, but since it may have gotten too dense
(which is OLSR's supposed especialty), there's a route convergence
time for that route to get up-to-date.

3) The node is in the network, but due to some intermittent condition
(bad wheather, interference, whatever), I cant reach it right now.

     So, how can OLSR solve those ambiguities? I'm asking this because
I'm working on a test-bed, but since it's a small-scale network, of
course I still havent got much of the issues above. But, just
wondering that scalability would bring those issues to light. I'd
appreciate if anyone has had some of those issues in a large-scale
mesh and could share some insights.

best regards,

-- 
-- 
:: Breno Jacinto ::
:: breno - at - gprt.ufpe.br ::
:: FingerPrint ::
   2F15 8A61 F566 E442 8581
   E3C0 EFF4 E202 74B7 7484
:: Persistir no difícil é a única maneira de torná-lo fácil algum dia.  ::