On 2012-06-12 0857, Paolo Lucente wrote:
> Hi Tiernan,
> 
> On Mon, Jun 11, 2012 at 02:10:10PM +1000, Tiernan Messmer wrote:
> 
>> With this in mind, and that the core process is what is using almost all
>> of the CPU usage - the only intensive task I can think of the core
>> process doing is moving around packet headers to the plugin processes -
>> which certaintly seems at this stage to be a beefier task than
>> incrementing counters - or is there another task that the core process
>> does as well?
> 
> Or tagging or filtering or correlation with BGP and/or IGP. It can help
> as a start to see your current configuration file. Feel free to follow up
> here or privately.  

It's the same as in my first message to the list (other than the two
changes noted earlier), but for completeness here it is again. The
filters have been culled for privacy reasons but contain clauses to
match direction of traffic based on IP address (either matching the
private internal ipv4 address range, public ipv4 address, or public ipv6
range for source/dest as appropriate) and then matching based on if the
traffic is metered or not (consists of 80 CIDR ranges).

If you think the exact filter may be contributing, I can send those
privately if you'd like. On an unrelated note, is it possible to filter
based on traffic direction in/out of an interface rather than relying on
matching based on network? May simplify the filters slightly if that's
what the bottleneck is.

debug: false
daemonize: true
aggregate[in]: dst_host
aggregate[out]: src_host
aggregate[infree]: dst_host
aggregate[outfree]: src_host
aggregate[intotal]: dst_host
aggregate[outtotal]: src_host
aggregate_filter[in]: <filter with about 90 network clauses>
aggregate_filter[out]: <filter with about 90 network clauses>
aggregate_filter[infree]: <filter with about 90 network clauses>
aggregate_filter[outfree]: <filter with about 90 network clauses>
aggregate_filter[intotal]: <filter with 6 network clauses>
aggregate_filter[outtotal]: <filter with 6 network clauses>
plugins: mysql[in], mysql[out], mysql[infree], mysql[outfree],
memory[intotal], memory[outtotal]
plugin_pipe_size: 10485760
plugin_buffer_size: 1048576
pmacctd_flow_buffer_buckets: 65536
interface: tun0
interface_wait: true
snaplen: 128
promisc: false
syslog: daemon
pidfile: /var/run/pmacctd.pid
sql_db: pmacct
sql_table[in]: in
sql_table[out]: out
sql_table[infree]: infree
sql_table[outfree]: outfree
sql_table_version: 1
sql_user: pmacct
sql_passwd: <password>
sql_refresh_time: 60
sql_table_schema[in]: /usr/local/etc/pmacctd/pmacct.in.db
sql_table_schema[out]: /usr/local/etc/pmacctd/pmacct.out.db
sql_table_schema[infree]: /usr/local/etc/pmacctd/pmacct.infree.db
sql_table_schema[outfree]: /usr/local/etc/pmacctd/pmacct.outfree.db
sql_history: 1h
sql_history_roundoff: m
sql_optimize_clauses: true

imt_path[intotal]: /tmp/pmacct_in.pipe
imt_path[outtotal]: /tmp/pmacct_out.pipe


> Cheers,
> Paolo

Cheers,
~Tiernan


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

Reply via email to