On Tue, Jul 08, 2003 at 03:10:58PM -0300, xzhao2 wrote: > I'm now working on integrating QuickPage into MRTG (or vice versa), meaning > by > reading the data from MRTG log file every 5 minutes and comparing with some > thresholds, QuickPage is automatically triggered and a message is sent to a > cellphone or pager. Anybody has any idea to do this? Thanks. Simon
Honestly? Don't. Well, that's a little too strict, but here's what I mean: MRTG does something really, really well. What it does well is collecting and managing historical data; with the addition of some tool to graph the data, MRTG becomes a powerful way for you to know both how some resource is doing right now and whether how it's doing now is normal or abnormal (/var is at 82%! Oh, it's been increasing at 1%/week for a few weeks now, OK). I *LOVE* MRTG, and I use it heavily. However, it's not IMHO a very good alerting and fault detection system. These things are built into it, but they're very rudimentary and basic. I believe one is better served by using a product that is designed to do alerting from the ground up for that functionality; here (and at various other places where I've implemented it) we use Nagios; as an example, you can see either the Nagios site (www.nagios.org) or my own instance, at nagios.inorganic.org (log in as guest/guest). Nagios, in a word, rocks. Its design is incredibly sophisticated; it allows for very fine granularity; it allows for the concept of time-based reporting (email me if it's during business hours; page me otherwise); it allows for notification escalation (wow, there've been five notifications with no response. Time to let management now). It lets you put the whole configuration (almost) in a database. Did I mention it rocks? -roy -- Unsubscribe mailto:[EMAIL PROTECTED] Archive http://www.ee.ethz.ch/~slist/mrtg FAQ http://faq.mrtg.org Homepage http://www.mrtg.org WebAdmin http://www.ee.ethz.ch/~slist/lsg2.cgi
