I am using ganglia, but ganglia calls rrdtool to create the graph, so I'm not sure if its a parameter issue from ganglia or the way rrdtool behaves. Any pointers are appreciated.
When I select to display a metric in ganglia, I get the last hour, last 2 hours, 4 hours, day, week, month and year. Each graph has a "now" value display as part of that graph and I'm wondering from where and how that value is calculated. Our metrics are all defined as gauges (because they com through various tooling like sflow or jmxtrans or gmetric. some support type, other do not). The metric I'm collecting is an integer counter. But I'm sometimes seeing the max (and last) value as NN.nn where nn is not 00. Why would this (or the max value) never always be a whole number? When I look at the metrics being collected, I do see Slope=BOTH - does that have impact here? when I dump the rrds, I also noticed that there are only <cf> entries for AVERAGE but this is true for the standard ganglia metrics like cpu_system trying to figure out how this works so that I can explain it to the OPS people or is we need to change something. thanks, jon -- View this message in context: http://rrd-mailinglists.937164.n2.nabble.com/How-is-the-now-value-is-calculated-tp7582091.html Sent from the RRDtool Users Mailinglist mailing list archive at Nabble.com. _______________________________________________ rrd-users mailing list [email protected] https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users
