[Olsr-users] MPRs in 0.5.x branch
Fri Mar 5 14:36:46 CET 2010
Op 5 mrt 2010, om 13:44 heeft Henning Rogge het volgende geschreven:
> On Fri March 5 2010 12:25:27 Dmitry Mikhin wrote:
>> By "not well behaving routes" do you mean routes with loops? Or there
>> are other problems?
> It means that the routing will ignore all routes that are not announced by
> both sides of the link.
Isn"t this the TcRedundancy parameter?
# TC redundancy
# Specifies how much neighbor info should
# be sent in TC messages
# Possible values are:
# 0 - only send MPR selectors
# 1 - send MPR selectors and MPRs
# 2 - send all neighbors
# defaults to 2
# TcRedundancy 2
I thought the discussion is on MPR coverage.
# MPR coverage
# Specifies how many MPRs a node should
# try select to reach every 2 hop neighbor
# Can be set to any integer >0
# MPR selection is broken at the moment, just keep 7
# MprCoverage 7
Now I wonder what is broken.
>> So with MprCoverage 1 I'll probably get buggy routes. But with simple 3
>> nodes test it looks like all works fine. Does this bug appear only on
>> complex topologies?
> 3 should be enough for most meshs.
> The bug does only appear if you have much less MPRs than your 1-hop networks,
> so I think it will mostly happen in very dense networks.
I test with dense 30 node network at the moment.
I didn't see this problem.
I did not check it in detail either.
More information about the Olsr-users