Hi Paolo,

We are running pmacct 0.14 on a Debian machine with 8GB RAM and running
Postgres as database backend.
Pmacct is used to log trafic on eth1 interface which reaches up to 1Gb.

We are offten receiving following errors in logfile:
Mar 10 22:29:52 ERROR ( default/pgsql ): We are missing data.
Mar 10 22:29:52 If you see this message once in a while, discard it.
Otherwise some solutions follow:
Mar 10 22:29:52 - increase shared memory size, 'plugin_pipe_size'; now:
'325760000'.
Mar 10 22:29:52 - increase buffer size, 'plugin_buffer_size'; now: '325760'.
Mar 10 22:29:52 - increase system maximum socket size.

I have tried different values of plugin_pipe_size and plugin_buffer_size
(from 81920000 - 81920 to current 325760000 - 325760) and also tried to
change sql_cache_entries from numbers close to 40 000 to 100 000 and now to
256 000.
Could you please advise if the configuration seems ok and what we could
change to improve performance?
I would be very grateful for any help.

pmacct.conf:

daemonize: true
pidfile: /var/run/pmacctd.pid
logfile: /var/log/pmacct.log
aggregate: src_mac,src_host,src_port,dst_host,dst_port
pcap_filter: not net 224.0.0.0 mask 240.0.0.0
interface: eth1
plugins: pgsql
sql_host: localhost
sql_passwd:
sql_table: acct_%w
sql_refresh_time: 60
sql_optimize_clauses: true
plugin_pipe_size: 325760000
plugin_buffer_size: 325760
sql_history: 1m
sql_history_roundoff: m
sql_dont_try_update: true
sql_cache_entries: 255989

Thank you in advance!

Best regards,

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

Reply via email to