[Olsr-users] Re : : Re : : TC redundancy 0

[(spam-protected)]

Ok.
Thanks.

So if I modify the code to act like "tc rendundancy = 0":
- only MPRs send TC message 
- TC contains only MPR selectors; 

Will this cause problem with the Dijkstra code of OLSRD?



--
Sarette



-----E-mail d'origine-----
De : Henning Rogge <(spam-protected)>
A: sarette25 <(spam-protected)>
Cc: teco <(spam-protected)>; olsr-users <(spam-protected)>
Envoyé le : Ma, 18 Sep 2012 17:29
Sujet : Re: Re : [Olsr-users] Re : : TC redundancy 0


At the moment every nodes sends TCs and every node includes every
ymmetric neighbor into the TC.
There is also currently no real support for MPRs, which would also
ffect the flooding patterns. Every node floods every new packet.
Thats the only supported mode at the moment.
Henning Rogge
On Tue, Sep 18, 2012 at 5:26 PM,  <(spam-protected)> wrote:
 Well, sorry I don't understand all.

 I am using OLSRD 0.6.2. Any node sends TC messages with its symetric links.
 And the option TC redundancy is set to 2.

 So for me, OLSRD uses TC redundancy level 2. And on the code it is written
 that we can't use TC redundancy levels 1 and 2, because a problem with
 Dijkstra.
 There is no flooding optimizations with TC redundancy, because any node
 sends TC message, so all the nodes floods.
 No?



 -----E-mail d'origine-----
 De : Teco Boot <(spam-protected)>
 A: sarette25 <(spam-protected)>
 Cc: hrogge <(spam-protected)>; olsr-users <(spam-protected)>
 Envoyé le : Ma, 18 Sep 2012 17:11

 Sujet : Re: [Olsr-users] Re : : TC redundancy 0


 Op 18 sep. 2012, om 17:00 heeft (spam-protected) het volgende geschreven:

 Sorry my question is about TC redundancy on OLSRD.
 The option was available but there is a bug with Dijkstra (version 0.5.6). I
 want to know if the bug is fixed or not.
 So if we can use TC redundancy = 0 with OLSRD (0.6.3) ???


 The description of the defect is not 100% correct.
 OLSR - as specified in RFC - works on minimum hops. Experiments showed that
 using some kind of link metrics works better. In current olsr.org code, ETX
 is implemented. The new OLSRv2 standard incorporates the lessons learned.
 But this needs new algorithms for reduced topology (distinct rules for
 Routing MPR and Flooding MPR). Rules for optimized flooding are still valid.
 Plans are that the next version with fixed optimizations is based on OLSRv2,
 not on something in between.

 Teco






 --
 Sarette


 -----E-mail d'origine-----
 De : Henning Rogge <(spam-protected)>
 A: sarette25 <(spam-protected)>
 Cc: henning.rogge <(spam-protected)>; olsr-users
 <(spam-protected)>
 Envoyé le : Ma, 18 Sep 2012 16:45
 Sujet : Re: [Olsr-users] Re : : TC redundancy 0

 On Tue, Sep 18, 2012 at 3:54 PM,  <(spam-protected)> wrote:
> Ok..
> Well, would you have any information about TC redundancy on OLSRD?

 http://tools.ietf.org/html/rfc3626#section-15.1

 in short, it decides how many of a nodes neighbors will be put into
 the TC... all of them, just the one who select you as a MPR or both
 your MPRs and the one who select you as an MPR.

 Henning Rogge
 --
 Steven Hawkings about cosmic inflation: "An increase of billions of
 billions of percent in a tiny fraction of a second. Of course, that
 was before the present government."

 --
 Olsr-users mailing list
 (spam-protected)
 https://lists.olsr.org/mailman/listinfo/olsr-users



-- 
teven Hawkings about cosmic inflation: "An increase of billions of
illions of percent in a tiny fraction of a second. Of course, that
as before the present government."

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.olsr.org/pipermail/olsr-users/attachments/20120918/aff74ea0/attachment.html>