<div dir="ltr">The babeld version is 1.8.0-3<div>babel-pinger 0.1-1</div></div><div class="gmail_extra"><br><div class="gmail_quote">2017-06-14 15:46 GMT+02:00 Leonardo Maccari <span dir="ltr"><<a href="mailto:mail@leonardo.ma" target="_blank">mail@leonardo.ma</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class=""><br>
<br>
On 14/06/17 13:45, Juliusz Chroboczek wrote:<br>
>> <a href="https://github.com/battlemesh/battlemeshv10-testbed/blob/master/scripts/babel" rel="noreferrer" target="_blank">https://github.com/battlemesh/<wbr>battlemeshv10-testbed/blob/<wbr>master/scripts/babel</a><br>
> Cool, thanks.  So Babel was run with the default 4s Hello interval and the<br>
> default 20s update interval, which is much higher than the other<br>
> protocols.  This might or might not explain the packet loss and the spikes<br>
> in latency, I'm not sure.<br>
><br>
> If you've got some time, I'd really like to see the time distribution of<br>
> the packet loss and the CDF graph of the latency.<br>
<br>
</span>Hi Juliusz,<br>
<br>
You can find all the data for the ping test here:<br>
<br>
<a href="https://github.com/battlemesh/battlemeshv10-testbed/tree/master/tests/failure_recovery/delay-one-run/BABEL" rel="noreferrer" target="_blank">https://github.com/battlemesh/<wbr>battlemeshv10-testbed/tree/<wbr>master/tests/failure_recovery/<wbr>delay-one-run/BABEL</a><br>
<br>
each file is one flow, in each file you have an "entries" field with all<br>
the rtt.<br>
<br>
we used this script to parse them, which you can modify to extract cdf<br>
(and the parse_all_pings.sh in the same folder to run them all):<br>
<br>
<a href="https://github.com/battlemesh/battlemeshv10-testbed/blob/master/tests/failure_recovery/parser_scripts/parse_ping.py" rel="noreferrer" target="_blank">https://github.com/battlemesh/<wbr>battlemeshv10-testbed/blob/<wbr>master/tests/failure_recovery/<wbr>parser_scripts/parse_ping.py</a><br>
<br>
if you are wondering why the tests are in the failure_recovery folder,<br>
this  is because my idea was to do the tests and then let some node in<br>
between fail, but we never got to that point :-)<br>
<span class="HOEnZb"><font color="#888888"><br>
l.<br>
</font></span><div class="HOEnZb"><div class="h5"><br>
><br>
> -- Juliusz<br>
> ______________________________<wbr>_________________<br>
> Battlemesh mailing list<br>
> <a href="mailto:Battlemesh@ml.ninux.org">Battlemesh@ml.ninux.org</a><br>
> <a href="http://ml.ninux.org/mailman/listinfo/battlemesh" rel="noreferrer" target="_blank">http://ml.ninux.org/mailman/<wbr>listinfo/battlemesh</a><br>
><br>
<br>
<br>
______________________________<wbr>_________________<br>
Battlemesh mailing list<br>
<a href="mailto:Battlemesh@ml.ninux.org">Battlemesh@ml.ninux.org</a><br>
<a href="http://ml.ninux.org/mailman/listinfo/battlemesh" rel="noreferrer" target="_blank">http://ml.ninux.org/mailman/<wbr>listinfo/battlemesh</a><br>
</div></div></blockquote></div><br></div>