Le 21 nov. 2012 à 19:46, Paolo Lucente <[email protected]> a écrit :
> Hi Frederic,

Hi Paolo,

> Yes, to collect both NetFlow/IPFIX data and sFlow ou will have to
> start nfacctd and sfaccd separatedly - each with the own BGP daemon.
> Typical issue here is that routers do not support BGP peering over a
> port different than 179/TCP: if both nfacctd and sfacctd run on the
> same server then i recommend a simple solution: bind the BGP daemons
> to two serparate sub-interfaces. The interface the BGP daemon has to
> bind to can be defined by means of its IP configured address, using
> the bgp_daemon_ip directive. 

Ok perfect. I can listen on two ports and I will configure the routers to use 
an other port.

> Btw, do i understand correct also the Foundry boxes can export full
> BGP table? 

Sure, that's not a problem !

However I do not understand how I have to configure the nfacctd and the sfacctd 
in order to have a single aggregate plugin. For example, I'd like to have 
statistics aggregated from both netflow and sflow when I run 

pmacct -s -T bytes -p /tmp/as_stats.pipe

Should I set in both nfacctd.conf and sfacctd.conf ?

aggregate: XXX,XXX,XXX
imt_path: /tmp/as_stats.pipe

Or is it impossible, and I will have to use two different imt_path, and then run
 pmacct -s -T bytes -p /tmp/as_stats_netflow.pipe
 pmacct -s -T bytes -p /tmp/as_stats_sflow.pipe

and finally write a script to sum the statistics between netflow and sflow ?

Thanks in advance,

Best regards,

-- 
Frédéric Gabut-Deloraine | Network Architect
Dailymotion | 140 boulevard Malesherbes, 75017 Paris, France



_______________________________________________
pmacct-discussion mailing list
http://www.pmacct.net/#mailinglists

Reply via email to