Finding Security & Performance in a Server Monitoring Agent

Did you know that you can download monitoring agents from Monitis that check the health of your servers, networks and applications — regardless of whether your firewall is up and running? In fact, these agents (which work for all operating systems) are such a convenience for so many sysadmins and website masters that Monitis has just published a new whitepaper — “Monitis Internal Monitoring Agent: Security and Performance” — to explain the workings behind the technology.

The whitepaper, which can be downloaded free from Monitis’s portal, discusses how these agents deliver the monitoring results in real-time to the Monitis Main Server, to be further displayed via Monitis’s web-based dashboard. The information — sometimes not so good, as in, a site is down — also triggers warnings and critical alerts through a whole menu of notifications, for example, email, mobile phone, SMS, IMs, Twitter and URL callback.

Monitis agents use the Monitis API, which is well documented and open.

Finding Security & Performance in a Server Monitoring Agent

Check out the diagram below to see how Monitis agents work:

If you’re thinking: “How will these agents affect my site?” or “Will Monitis’s tools slow down my server or network?” then you need to read this whitepaper — as it discusses how the Monitis agents perform, as well as how they operate securely (Don’t we have enough to worry about as far as security goes?).

For example, as far as performance goes, generally, the internal agent uses an Internet connection very economically: it requests the main server for current tests configuration every minute and then sends the accumulated test information every five minutes. And agents operating in both the Linux and Windows OSs use tiny bits of CPU and occupy modest amounts of virtual memory.

Now let’s talk security. Monitis agents provide standard security with the following features and practices:

  • Encrypted HTTPs protocol (used for the connection with main server)
  • Client initiated encrypted HTTP — which eliminates altering firewalls, thus reducing risk for possible attacks
  • Heartbeat checks performed to ensure that agents are alive and connected with the main server
  • The tests configuration and necessary parameters are kept on the main server and sent periodically to the client by using an encrypted channel. Current configuration is stored by the client internally in the memory (not in the file);
  • The internal client does nothing other than testing and therefore cannot cause unpredictable system damage — that’s a Monitis guarantee;
  • Monitis performs daily and weekly backups, including backups to storage outside of the main server’s data center.

Check out the new whitepaper for more detailed information on how Monitis agents work! And we’d love to get your feedback; please share.