[Olsr-dev] Test, Test, 1, 2 ...

[Henning Rogge]

Am Mittwoch 21 Mai 2008 09:28:31 schrieb Sven-Ola Tücke:
> - Does the "packet-loss" to "hello message loss" change really
> give a more accurate picture of wifi conditions?
Depends on what we consider an "accurate" picture.

The new code is independant of the amount of broadcast traffic (the old one 
allows to change the ETX value very fast if lot's of TCs are transmitted by a 
node), but it use less packages to consider the links quality.

> - Is the timer-refactoring stable enough or are the inhaerent bugs?
It seems to be more stable than the timer code in the 0.5.5 version. The 0.5.5 
version does create Hellos which are up to 60% too late (as an example).

> - Anything OK with long time values (e.g. > hour for nameservice
> announcements)?
We use 32 Bit unsigned integers for relative time values (measured in 
milliseconds), so the maximum timer would be more than 4 million seconds (or 
> 3 years).

> - Does standard timing params in default olsrd.conf fit the changes?
The hellos/tc intervals shouldn't matter for our changes I think. 
The "LQWindowSize" value has been replaced by the aging factor (at the moment 
with a default of 0.1) and we have to test if this default is okay for most 
networks (combined with the quickstart algorithm).

> May be a good idea to build a regressiontest-candiate version of the
> freifunk firmware too.
That's a good idea.

Henning


*************************************************
Diplom Informatiker Henning Rogge
Forschungsgesellschaft für
Angewandte Naturwissenschaften e. V. (FGAN) 
Neuenahrer Str. 20, 53343 Wachtberg, Germany
Tel.: 0049 (0)228 9435-961
Fax: 0049 (0)228 9435-685
E-Mail: (spam-protected)
Web: www.fgan.de
************************************************
Sitz der Gesellschaft: Bonn
Registergericht: Amtsgericht Bonn VR 2530
Vorstand: Dr. rer. nat. Ralf Dornhaus (Vors.), Prof. Dr. Joachim Ender 
(Stellv.)