Hello Ron, Below the score graph, there is a link 'CSV log'. For your server, it shows
ts_epoch,ts,offset,step,score,leap 1453206900,"2016-01-19 12:35:00",0.00105476379394531,1,8.1,0 1453205790,"2016-01-19 12:16:30",0.00391614437103271,1,7.5,0 1453204699,"2016-01-19 11:58:19",,-5,6.8,0 1453203648,"2016-01-19 11:40:48",,-5,12.5,0 1453202569,"2016-01-19 11:22:49",0.00187373161315918,1,18.4,0 The time stamp is in UTC, so you might need to correct it for your local time zone. If there is no offset at all (i.e. at 11:40:48 and 11:58:19) it means the monitor in L.A. did not receive a reply from your system. The package may have been lost on its way to your system or on its way back to L.A. It happened 5 times in the last 36 hours with your server. If it happens more often, or if you are just curious what's going on, you can check if you receive the time requests from the monitor with the normal interval (about every 18 minutes for IPv4), or if you miss them. That gives you a clue in which direction packets are lost. You can also traceroute the monitor server to see your upstream networks to LA (I guess aorta.net/UPC and GTT ;-) ) For now, I should not worry too much ;-) Kind regards, Arnold On 19-01-16 15:14, Ron Hahn wrote: > Colleagues, > > Looking at the offset monitoring and scores for my pool server I am seeing > several dips, one below the 10 score today. > > I am first thinking that my fibre connection has gone crazy but cannot > correlate zabbix alerts with the dips. > > Can some of the probes from Los Angeles to Dublin be timing out on the > route? Or is there something else that I have to look at? > > All Zabbix dashboards are normal all day - maybe I am missing some thing? > > Thank you for the help, > > Ron > _______________________________________________ > pool mailing list > [email protected] > http://lists.ntp.org/listinfo/pool > > _______________________________________________ pool mailing list [email protected] http://lists.ntp.org/listinfo/pool
