[OLSR-users] Network stability

[Thomas Lopatic]

[...]

> HystScaling     1.00

[...]

Whoops, make that 0.10 instead of 1.00 for HystScaling. 1.00 causes the 
link to be dropped when a single packet is missed and to be 
re-established when the next packet is received. That would make the 
link pretty flaky.

[...]

>>(You're still not using link quality, right?)
> 
> No.  I'm torn over this one.  I want to, but I also don't want to deviate from 
> the RFC, and I don't want to risk stranding a router when I change over.

[...]

Quite understandable. If you cannot access routers easily, then waiting 
for a few releases until the LQ code has proved itself in larger 
networks is a good idea.

[...]

> Thanks, as always.  I'll go run these tests and let you know how it looks.  
> How far through the network should I propagate these?  Every affected router?

[...]

Let's quickly have a look at how these parameters work. Let's assume 
that we change the parameters on node X.

The hysteresis settings affect X. Change HystScaling to 0.10 on X and X 
will be pretty inert with respect to re-establishing a dropped link or 
dropping an established link.

The validity times, however, affect all nodes that see messages sent by 
X. The validity times are announced in each HELLO message 
(HelloValidityTime) and TC message (TcValidityTime) that X sends. So, 
this affects how long other nodes consider a message form X to be valid.

So, if you have a flaky link, it's advisable to change these parameters 
on both ends of the link.

-Thomas