The place I work for is interested in purchasing a product that, among other things, doesn't set hard thresholds for application alerts. Rather, it detects peaks and slopes, etc., and kind of "intelligently" figures out how the resources of a device are used, and sends out alerts when "abnormal" activity is occuring.

I'm wondering if there are any monitoring tools out there that have this kind of setup, or if there are plugins or APIs between several tools that would allow my network/application monitor to have more automagically flexible alert thresholds.

  • rrdtool (oss.oetiker.ch/rrdtool/doc/rrdtool.en.html) has aberrant behaviour detection - I wish there were some applications that took advantage of it for alerting.
    – James
    Jun 18, 2010 at 17:46
  • Is is possible to hook rrdtool into Nagios or OpenNMS to set rolling thresholds on things like CPU utilization and disk usage? That would be nice. Jun 18, 2010 at 18:55

1 Answer 1


The company I work for uses Application Monitor from ManageEngine, and while it doesn't set thresholds dynamically the defaults that it uses are pretty good. Many of the values that it monitors do not have dynamic limits. For example: Response time on a web server should always be quick no matter how much traffic there is. It is better to set alerts based on desired SLA or response times than using connections or bandwidth.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.