[Olsr-users] Routing-Problem | olsr.org - 0.6.0 | weight | etx_ff | mode ether
Fri Mar 30 09:17:14 CEST 2012
On Fri, Mar 30, 2012 at 7:22 AM, Michel Blais <(spam-protected)>wrote:
> Use a rules to set a mark on udp port 698 and set priority for this
> mark with tc.
> I don't think it would really give result without a queue with max
> bandwith lower than the link can support to be sure the're no paquet
> lost on the link.
yes, this is the "simplest" general purpose approach to get some results.
(but as you have to know and limit your link speed to the lowest thinkable
thriupghput of the link, is a problem)
infact it depends on your interface/wifi driver,..
what approach for qos works "best".
e.g. for ethernet and some/many wireless drivers it is possible to write
qos rules, that priorize olsrd traffic without having to know (or limit)
the link speed.
but there are various approaches, all with some pros and cons
maybe we can/should create a wiki page about this
> You have to make sure than no paquet is lost on the
> link, only unpriorize paquet droped by the router to not overload the
> Le 29 mars 2012 21:35, Érico Porto <(spam-protected)> a écrit :
> > Hey Markus,
> > Do you mean using TC? I got curious, how does the OLSR traffic stands
> > the regular traffic? (so I can apply a rule to give it bigger priority)
> > Érico V. Porto
> > On Thu, Mar 29, 2012 at 9:04 PM, Markus Kittenberger
> > <(spam-protected)> wrote:
> >> On Thu, Mar 29, 2012 at 9:40 PM, Michel Blais <(spam-protected)
> >> wrote:
> >>> This often happen when the're no traffic on A link and b link have some
> >>> paquet lost.
> >>> If paquet lost for B link is too much and cost more then 3 then it will
> >>> try link A. If the're no traffic on link B then cost come back to 2 so
> >>> goes back to B.
> >> imho first to do do solve/minimize this "flapping" problem, is
> >> olsrd traffic so it gets as least as possible affected by packetloss
> due to
> >> congestion on your links,..
> >> Markus
> >> --
> >> Olsr-users mailing list
> >> (spam-protected)
> >> https://lists.olsr.org/mailman/listinfo/olsr-users
> Olsr-users mailing list
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Olsr-users