[Olsr-dev] ETX mambo jambo

[Markus Kittenberger]

On Tue, May 11, 2010 at 8:17 AM, Sven-Ola Tuecke <(spam-protected)> wrote:

> Hi Mitar,
>
> what a long mail.

ACK, *G


> Besides that: you are right. Add LinkQualityAlgorithm "etx_ffeth" to your
> config.
>
but as this etx_ff_eth metric is imho only a very simple improvement to
actual etx,.. (but at least imho a big step in usability in typical freifunk
meshes)

so please take your "motivation" and invest it into a new metric plugin (or
parts of),

for example reliable measuring/finding bandwidth of various types of links,
can get some work,.
thx to henning the mac80211 has received some interface for reading such
information,.. but any other drivers have not, or different interfaces, ...
so we might need a fallback method for (at least roughly) measuring it,..

btw there are already some ideas around for measuring bandwidth with
existing olsrd messages + some additional probe packets,..

Markus

p.s. the metric plugin system currently only accepts additive metrics afaik

but imho when using very different metrics, we will might need many changes
in olsrd core aswell,. as imho the old "hated" minimal cost of 1.00 of ETX
is a big factor for routing stability of exisitng olsrd networks,..

so adding a new metric, could need much time & motivation until its working
as reliable as etx, (but of course with better routing decisions)


> // Sven-Ola
>
> Am Montag 10 Mai 2010 21:24:55 schrieb Mitar:
> > Hi!
> >
> > Probably a lot on this topic has already been written but ETX is
> > really badly conceptualized and it really does not properly influence
> > routing decisions.
> [snip]
>
> --
> Olsr-dev mailing list
> (spam-protected)
> http://lists.olsr.org/mailman/listinfo/olsr-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.olsr.org/pipermail/olsr-dev/attachments/20100511/1331187f/attachment.html>