What would be good software for monitoring multiple small and disparate networks?

My company sells a product that requires deployment of a number of Linux servers on a customer's network (typically three servers but may be more). We currently have such deployments in 25+ networks and this number is growing. We would like to monitor the general health of the machines as well as the status of various custom bits of processing on each of these networks. We connect to them in a variety of ways (direct SSH, VPN, WebEx, etc) and so being able to see everything from both the command line and a web interface would be useful.

This information would be aggregated (somehow) at our office for an overview, but that mechanism is beyond the scope of this question.

I am aware of "What tool do you use to monitor your servers?" but that seems to cover monitoring the entire network as well as large networks. Some of these seem too complex for monitoring just a handful of machines per network.

I have been experimenting with Nagios but would like to know if there were more suitable tools for our particular use case.

Similar question:


If you're looking for system-level resources (CPU/memory/disk usage, server load, etc.) then installing any number of monitoring tools with a local agent should be able to provide this data. They generally (including the one that my company, Panopta, provides) gather data locally and push the results to a central monitoring location where they are stored for reporting and alerting. Make sure that the tool you use can scale to handle what sounds like a reasonably sizable number of servers and lets you easily segment them by customer so that you have clear visibility into each customer's configuration and performance.

Checking on the availability of network services in the environment you're describing is tougher, unless you can put together a central location where you terminate each of your VPNs or SSH tunnels that would provide access to the services running in each environment. If you do have such an endpoint available, you can either run monitoring software (such as Nagios, Zabbix or Zenoss) locally and have it check services through the tunnels you have setup. Or you could use something like our monitoring appliance which can be installed at that location to perform checks and then merge those results with the server-level metrics gathered by a local agent.


You might want to have a look at Shinken.

It's still a young project but seems to fit such a "stretched" network.

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.