Re: [pmacct-discussion] Multiple updates per interval (print plugin)

2014-02-05 Thread Ruben Laban
Cause for the memory usage has been found: print_cache_entries: 91 This was needed when I was testing with random destination IP addresses. As our production environment will only be accounting for say 1 IPs, this value might not need tweaking. Regards, Ruben On 2014-02-05 08:28,

Re: [pmacct-discussion] Multiple updates per interval (print plugin)

2014-02-05 Thread Ruben Laban
Another thing I noticed was the lack of markers in the output files. Had forgotten this is now controlled by the config. So after enabling it, I get files like these: --START (1391596920+60)-- DST_IP,PACKETS,BYTES 192.168.0.1,1059620,48742520 --END-- 192.168.0.1,62895128,2893175888 --END--

[pmacct-discussion] Strange results on nfdump when using networks_file

2014-02-05 Thread Joan
(only google's 8.8.8.x and 8.8.4.x) ! generated by quagga_gen_as_network.pl at 20140205-11:25.51 193.149.55.94,15169,8.8.4.0/24 193.149.55.94,15169,8.8.8.0/24 Now I'm getting the srcas and dstas setted for all the traffic as if it was originated and destinated to google. I'm using the current

Re: [pmacct-discussion] Strange results on nfdump when using networks_file

2014-02-05 Thread Paolo Lucente
creating a full networks_file with the script at ( https://github.com/paololucente/pmacct-contrib/tree/master/st1) failed leaving all the AS fields as 0, I simplified the file to a minimal case (only google's 8.8.8.x and 8.8.4.x) ! generated by quagga_gen_as_network.pl at 20140205-11:25.51