Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

I need to set up a monitoring and alerting system for one of our business teams-- they're smart people, but not engineers or sysadmins. I basically need to do the same thing that happens for system monitoring, but with all application-specific metrics. I'll be writing all of the monitoring scripts myself.

I'm familiar with Nagios, and it's what we use in-house for our system-level stuff, but it's not the right fit for this problem. My needs:

  • A clean and simple dashboard
  • Alerting capabilities
  • Graphs
  • Obviously, the ability to write my own monitors

In other words, what's like Nagios but dumbed down?

share|improve this question
1  
What are you alerting and monitoring on? Not sure what you're after here. –  gravyface Jul 13 '10 at 18:32
    
I'm missing what you need as well...like, why not set up Nagios to alert them when things get out of a spec range? I'm not sure how "dumbed down" you want to get it; how are they going to understand what's working and what's not if it's just giving status reports as deep as a puddle? Are you just looking to alert them to stuff like low disk space (which again can be handled via email/sms alerts)? –  Bart Silverstrim Jul 13 '10 at 18:43
    
Please define "all application-specific metrics". As for dumbed down, if those users can't read a Nagios screen there's probably very little we can do for them. Green is good, red is bad and anything else is somewhere in between. How hard is that? –  John Gardeniers Jul 13 '10 at 19:28
    
We've had bad luck with this kind of thing. The more we monitor, the more likely it is that at least one thing is going wrong at any given time, and it is HARD to write dashboard logic that can handle special cases like, "2 of 3 is good, 1 of 3 is bad", and "2 of 3 is bad between 1100-1400 but good between 1800-0300". –  sysadmin1138 Jul 13 '10 at 19:33
    
My company is in the flibberwigget business. I'm alerting on things like "the systems production of flibberwiggets dropped 20% in the last hour" or "the average latency for flibberwigget production is greater than ten seconds." I can glean the flibberwigget data from scripts that monitor logs, and I'm doing that on an ad hoc basis right now. I want to pull it under a monitoring system of some sort, and Nagios doesn't seem like the right one. Plus, there's some internal politics that makes Nagios not the right solution. –  user10501 Jul 13 '10 at 19:41
show 3 more comments

5 Answers

up vote 1 down vote accepted

I would have a look at Zenoss. I've used Nagios for years to monitor and alert on issues but we recently switched to Zenoss b/c we wanted easier management and integrated RRD graphing capabilities. Zenoss has a nice web based interface which isn't just the dashboard for dealing with events but is also used for almost all configuration, device management, and alerting rules. It's much easier to setup compared to Nagios and has integrated RRD graphing capabilities.

As a bonus Zenoss supports using Nagios plugins and scripts to monitor and alert on things so if your team already has experience with writing Nagios monitoring scripts you can continue to leverage that knowledge.

share|improve this answer
add comment

+1 for ZenOSS, but it depends on what version you use. There is the free community version, and their "enterprisey" offering which I assume has shinier bells and louder whistles.

The former does not do a very good job of talking to Winboxen if you have a more recently configured Windows domain authentication policy or security guys breathing down your neck (hello NTLMv2). So, for the community version, you get to have fun configuring Samba for yourself to force that kind of authentication if you want native polling of Windows servers/clients without installing anything on the target (since it basically does remote WMI calls with an admin account of your choice). This is why it appealed to my co-workers, since it was supposed to be shiny, fun, and no-hassle, but this confused them since they never played that much with Linux. I lost time to tinker, so I never got that part to work with the community version. Just my two cents . . .

share|improve this answer
add comment

we wrapped nagios in a "cleaner" interface for the suits. nagios calls a custom script, which executes the relevant probe and returns the results. If a status changes, another script is called, which takes the appropriate actions (emails, sms etc.).

Everything the two scripts is configurable via a web interface we rigged up in PHP, operating on a DB. It doesn't do viz but the alerts work fine.

share|improve this answer
add comment

Try Zabbix. Easy to configure, not as powerfull as ZenOSS or Nagios. Has graphs, different alerting, etc.

share|improve this answer
add comment

Try PRTG Network Monitor, it should be very easy point and click install. It evens have a Windows GUI, which non-sysadmin people will love it.

It can monitor as much as Nagios can. There is 30 days free trial and also free version with limited monitoring nodes.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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