On Fri, Nov 17, 2006 at 11:45:41AM +0000, Paolo Lucente wrote:
> Hi Alan,
>
> On Thu, Nov 16, 2006 at 11:02:01AM -0800, Alan wrote:
> > Hey all. I have 2 systems running pmacctd with virtually identical
> > versions and setups, but on one system the stamp_inserted is inserted at
> > 23:00:00 and the other is at 00:00:00. The two systems are supposed to
> > be able to back each other up and are giving slightly different reports,
> > I think due to when the system starts recording traffic each day.
> >
> > Is this a configuration option somewhere, or could it be based on a
> > timezone or something (select NOW() from the postgres DB shows the same
> > date time and timezone on both systems, and the system time it the same
> > for each as well).
>
> Yes, it looks like either a different timezone setting on the two boxes
> or a line like "sql_history_roundoff: h" is missing in your pmacct config.
> Can you please check these two things? If both of them are unsuccessful,
> having a look to your configurations can help.
On both systems /etc/timezone has Canada/Pacific in it, and 'date' on
the command line displays PST and the same time (within a couple of
minutes anyway). Both daily.conf files have sql_history_roundoff:h in
them. Here's the config from box 1 (23:00 roundoff):
debug:false
interface:eth2
aggregate[in]:dst_host
aggregate[out]:src_host
aggregate_filter[in]: dst net x.y.84.0/24 and not src net x.y.84.0/24
aggregate_filter[out]: src net x.y.84.0/24 and not dst net x.y.84.0/24
plugins:pgsql[in], pgsql[out]
sql_data:frontend
sql_db:pmacct
sql_table:acct_daily
sql_user:pmacct
sql_passwd:[xxx]
sql_refresh_time:600
sql_optimize_clauses:true
sql_history:1d
sql_host:newipa
sql_recovery_logfile:/home/path/recover-daily.log
pidfile:/home/path/pmacctd-daily.pid
sql_history_roundoff:h
plugin_pipe_size: 512000
plugin_buffer_size: 4096
! next must be prime....
sql_cache_entries: 99991
And box 2 (00:00):
debug:false
interface:eth3
aggregate[in]:dst_host,dst_mac
aggregate[out]:src_host,src_mac
aggregate_filter[in]: dst net (192.168.2.0/23 or x.y.84.0/24) and not src net
(192.168.2.0/23 or x.y.84.0/24)
aggregate_filter[out]: src net (192.168.2.0/23 or x.y.84.0/24) and not dst net
(192.168.2.0/23 or x.y.84.0/24)
plugins:pgsql[in], pgsql[out]
sql_data:frontend
sql_db:gatekeeper
sql_table:acct_daily
sql_user:[yyy]
sql_passwd:[xxx]
sql_refresh_time:600
sql_optimize_clauses:true
sql_history:1d
sql_host:localhost
sql_recovery_logfile:/opt/path/pmacct/recover-daily.log
pidfile:/opt/path/pmacct/daily.pid
sql_history_roundoff:h
plugin_pipe_size: 512000
plugin_buffer_size: 4096
! next must be prime....
sql_cache_entries: 99991
On box 1 postgres db:
pmacct=# select now();
now
-------------------------------
2006-11-17 10:16:53.113981-08
(1 row)
On box 2 postgres db:
gatekeeper=# select now();
now
-------------------------------
2006-11-17 10:21:47.027123-08
(1 row)
Any other thoughts or information that can help?
--
Alan <[EMAIL PROTECTED]> - http://arcterex.net
--------------------------------------------------------------------
"Backups are for people who don't pray." -- big Mike
_______________________________________________
pmacct-discussion mailing list
http://www.pmacct.net/#mailinglists
