[time] Monitor spike?
Ask Bjørn Hansen
Fri Apr 4 01:18:22 UTC 2008
On Apr 3, 2008, at 17:40, Chris Kuethe wrote:
> And a few more over the last couple of days. My link is not so
> saturated nor is my server so loaded that I couldn't tolerate the
> poller doing a burst of polls. Take the median of 5 or 7 probes - that
> should deal with network noise quite nicely.
That's effectively what it does in the scoring mechanism; it just
gives more weight to measurements that didn't go well and it shows the
nitty gritty of each measurement.
As always: Remember that the goal of the monitoring system is to kick
out systems that give bad time. Nothing else. It's not to monitor
the "perfect performance" of any individual system.
There are really only two important scores in the system, >= 5 and <
5. Nothing else matters for the objective of the monitoring system.
More information about the pool