Best OpenSource Monitoring Tool?

Question

I'm trying to decide between Zenoss and GroundWork. I need an enterprise level monitoring tool and I had heard really good comments of these tools.

Do any of you have implemented them? Which are their strong points? Shortcomings?

score 3 · Answer 1 · answered Apr 09 '10 at 17:47

3

Specifically in regards to availability monitoring, I'm a huge Nagios fan. If you haven't considered it, I'd strongly recommend it. Others like monit.

answered Apr 09 '10 at 17:47

Warner

23,440
2
57
69

score 3 · Answer 2 · answered Apr 09 '10 at 18:37

3

Nagios, or if you are looking for something with a GUI then Opsview (which is essentially written on top of Nagios and has an OSS/Community edition and a commercially supported one).

In answer to your question - I have tried a few others (groundwork included) and from an enterprise point of view Nagios/Opsview came out tops.

answered Apr 09 '10 at 18:37

Mark Regensberg

1,421
12
14

We use opsview as well. It's growing nicely but I wish they'd pick an iconography and stick with it. – jldugger Apr 09 '10 at 19:33
I am not familiar with Opsview, but I agree 100% that Nagios is a winner. – Skyhawk Apr 09 '10 at 20:15
Nagios seems to be the prefered Monitoring solution for everybody. I'll try it for sure. – alfredocambera Apr 12 '10 at 15:09

score 2 · Answer 3 · answered Apr 10 '10 at 21:25

@anders: if you have 400 servers you'd better use templates, in any environment. i'm usually faster at the console to integrate a new server/service than any interface.

and for nagios you have nagiosql which works quite nice if you know how to use templates ;-)

i prefer nagios for it's wide range of plugins and stability

score 1 · Accepted Answer · answered Apr 10 '10 at 20:54

Zabbix, it's really good. Previously we used Nagios, MRTG, Munin but Zabbix kind of mix the functionality of the previous ones, letting you know when the server is down but also:

Web Administration (PHP + MySQL): One of the worst points of Nagios is its configuration, being almost all in config files... Its OK if you have to monitor 5 machines, but what if you have 400?
Several types of monitoring: Have a native agents for almos every OS, but also lets you monitor through SNMP, IPMI, scripts, etc.
Graphs and Maps: I love them! Like MRTG, but MUCH easier to configure.

I had heard of Zabbix and yeap, It looks really good. Thanks for answering! — alfredocambera, Apr 12 '10 at 15:32

score 0 · Answer 5 · answered Apr 09 '10 at 19:50

I much prefer Nagios as well, even though I know it's not in your list of preferred tools.

I spent a fair amount of time trying to implement Zenoss in a previous company, but never got it to work as well as Nagios; the monitoring agent on one system would work while the agent on another, apparently identical system would not. I got the impression that you need to be very familiar with SNMP (which I'm not) to get the best out of Zenoss; by contrast, Nagios just uses basic TCP/IP sockets and SSH for communication, and has always worked right out of the box for me.

score 0 · Answer 6 · answered Apr 10 '10 at 22:04

0

We use a combination of ganglia, nagios, monit, and nmon. Ganglia for long term performance trends, nagios for availability alerts, and monit to restart services when they stop running. We use nmon for granular scalability testing. Nagios is great for monitoring multiple OS's, but it can take a little while to setup.

answered Apr 10 '10 at 22:04

Luke

64
3

Interesting. I friend of mine gave me a similar point of view. Probably I'll need more than one tool to solve the problem. Thankx – alfredocambera Apr 12 '10 at 15:10

Best OpenSource Monitoring Tool?

6 Answers6