* [B.A.T.M.A.N.] Very strange behaviour
@ 2011-11-19 19:22 Filippo Sallemi
2011-11-19 19:31 ` Antonio Quartulli
0 siblings, 1 reply; 3+ messages in thread
From: Filippo Sallemi @ 2011-11-19 19:22 UTC (permalink / raw)
To: The list for a Better Approach To Mobile Ad-hoc Networking
Hi,
I'm just working on script that try to autorecover the node in case of
failure and I noticed something very strange.
Ii have 3 node setted to channel 3. Well when I change the channel of
one of these nodes for example from 3 to 5 after some time I can see
however the entirely topology of my network.
How it is possible? My script is not ready and iwconfig output show
the changed channel!!!!
apparently do not have the same behavior if channel is far from other
nodes (es 6)
This is very very strange for me!
Rgds
--
Filippo Sallemi
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [B.A.T.M.A.N.] Very strange behaviour
2011-11-19 19:22 [B.A.T.M.A.N.] Very strange behaviour Filippo Sallemi
@ 2011-11-19 19:31 ` Antonio Quartulli
2011-11-19 19:42 ` Daniele Furlan
0 siblings, 1 reply; 3+ messages in thread
From: Antonio Quartulli @ 2011-11-19 19:31 UTC (permalink / raw)
To: The list for a Better Approach To Mobile Ad-hoc Networking
Hello Filippo,
On Sat, Nov 19, 2011 at 08:22:41 +0100, Filippo Sallemi wrote:
> Hi,
> I'm just working on script that try to autorecover the node in case of
> failure and I noticed something very strange.
> Ii have 3 node setted to channel 3. Well when I change the channel of
> one of these nodes for example from 3 to 5 after some time I can see
> however the entirely topology of my network.
>
> How it is possible? My script is not ready and iwconfig output show
> the changed channel!!!!
>
> apparently do not have the same behavior if channel is far from other
> nodes (es 6)
>
> This is very very strange for me!
>
I know you are testing batman-adv on your network, but this particular issue is
not strictly related to batman anymore..
However I'd say that, since channels overlap (yes they do), it is possible that some
devices overhear packets (not all of course) from near channels too. If you want
to be sure that overlapping is not happening, you should choose channel 1, 7 and
13.
By the way this is just my explanation. Could be not correct.
Cheers,
--
Antonio Quartulli
..each of us alone is worth nothing..
Ernesto "Che" Guevara
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [B.A.T.M.A.N.] Very strange behaviour
2011-11-19 19:31 ` Antonio Quartulli
@ 2011-11-19 19:42 ` Daniele Furlan
0 siblings, 0 replies; 3+ messages in thread
From: Daniele Furlan @ 2011-11-19 19:42 UTC (permalink / raw)
To: The list for a Better Approach To Mobile Ad-hoc Networking
2011/11/19 Antonio Quartulli <ordex@autistici.org>:
> Hello Filippo,
>
> On Sat, Nov 19, 2011 at 08:22:41 +0100, Filippo Sallemi wrote:
>> Hi,
>> I'm just working on script that try to autorecover the node in case of
>> failure and I noticed something very strange.
>> Ii have 3 node setted to channel 3. Well when I change the channel of
>> one of these nodes for example from 3 to 5 after some time I can see
>> however the entirely topology of my network.
>>
>> How it is possible? My script is not ready and iwconfig output show
>> the changed channel!!!!
>>
>> apparently do not have the same behavior if channel is far from other
>> nodes (es 6)
>>
>> This is very very strange for me!
>>
>
> I know you are testing batman-adv on your network, but this particular issue is
> not strictly related to batman anymore..
>
> However I'd say that, since channels overlap (yes they do), it is possible that some
> devices overhear packets (not all of course) from near channels too. If you want
> to be sure that overlapping is not happening, you should choose channel 1, 7 and
> 13.
>
> By the way this is just my explanation. Could be not correct.
>
> Cheers,
>
> --
> Antonio Quartulli
>
> ..each of us alone is worth nothing..
> Ernesto "Che" Guevara
>
I agree with Antonio explanation, I think the cause can be the following:
The node that changes channel start sending beacon
containing the new channel, if the channel is near the
previous one, they are partially overlapping so probably
neighbor nodes are able to receive beacon and consequently
changes their channel.
This cannot happen if you choose non overlapping channel.
Also a possible solution can be to fix the channel on every
node in the ad-hoc network, not only the essid, so every node
maintains the channel set at start-up.
Regards
--
Daniele Furlan
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2011-11-19 19:42 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-11-19 19:22 [B.A.T.M.A.N.] Very strange behaviour Filippo Sallemi
2011-11-19 19:31 ` Antonio Quartulli
2011-11-19 19:42 ` Daniele Furlan
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.