Your first monitoring configuration

Your first monitoring configuration

“Where to begin?!”

Monitoring screen

The first time you begin to set up monitoring, it can look like extremely complex and convoluted task. If you haven’t been into this before, you might get confused by all the “devices”, “hosts”, “sensors”, “monitors”, “alerts” and the rest of terms used in variety of monitoring services and software.

In the following few blog posts, we will try to make this process as simple as possible. We have prepared detailed “Getting started” reference, but even in its short form it can look overwhelming. Thus let’s begin with the basics of what monitoring is and how IPHost Network Monitor makes performing it a simpler task.

If you feel ready for setup challenge, please proceed to the above mentioned guide on getting started. Otherwise, continue reading.

The basics of monitoring

The purpose of monitoring is to get aware of possible (and existing) problems with some resource (such as website) or a device (such as a computer or network printer) accessible on the net. Be it in a local (such as home or business) network, or outside it (such as website), it’s all accessible by TCP/IP family of protocols. This is what IPHost deals with: protocols.

Apart from the basic monitors (TCP, UDP, PING), suitable for general connectivity tests, the rest of monitors use protocols (examples: HTTP/HTTPS, used by Web sites; SMTP, used to transmit and receive email).

While we have monitors covering most popular monitoring needs, there are cases when none of specialized monitors do. In such a case, IPHost offers “universal” monitor: “Script or Program” (to run a script or application on Windows host), “Python Script” (Python 3 environment is installed along with IPHost) or “SSH (Script or Program)”. Those monitors allow running almost arbitrary scripts (such as VBScript, PowerShell etc) and/or applications and interpret their output as performance value.

What is performance value? Simply put, it’s either a numerical value or text (string) returned by monitor process. It allows estimating whether the device (or service) we are monitoring is in good state. For example, we could measure how fast does our website respond to PING echo request – and warn us, if it’s longer than, say, 2 seconds. Biggest acceptable performance value in this case would be 2000 (since PING response time is measured in milliseconds). Thus, if that value is above 2000, we may wish to assume the site is in abnormal state (responds too slowly). To define what’s “abnormal”, we use so called state conditions.

State conditions

Monitors, when their state is known (defined), can be in one of 3 states: normal (“OK”), Warning (should be checked, but not yet critical) and Down (out of order; should be checked as soon as possible).

State conditions allow defining when monitor should be switched between states. For example, response time over 2500 milliseconds might be marked as Warning; or Down, if over 5000. The actual values for the mentioned conditions are entirely up to you: they depend greatly on actual network and devices setup.

When connectivity between device being monitored and the system running IPHost is broken, it also is treated as Down state.


Monitoring is useless without alerting. Simply put, when monitor enters problem state (or stays in one for long; or returns from one), resources administrator would like to get notified. For example, one can set up IPHost to send email notifications to one set of recipient in daylight time, and to another set otherwise.

It is possible to mark monitors depending on other monitors: for example, if connectivity is broken between your local network and your corporate site outside, you can receive quite a lot of notifications (since every service, every metric on the site you are monitoring becomes inaccessible). However, if you set all those metrics depending on simple connectivity test (say, PING monitor), you will only get notified of connectivity broken: all the depending monitors will simply be paused until connectivity (PING) is restored.

IPHost also offers variety of notification means. Sending mail message, posting a message to a messenger service, calling user-defined script or external service via HTTPS request – these notification means allow setting up arbitrarily flexible alerting, optimized for your needs.

The summary

Monitoring is a necessity, not a luxury. Performed either manually or by automated means, it is a must when smooth network services operation is required. IPHost makes it easy to automate this process as much as possible, while retaining full control over all credentials and other sensitive monitoring data.

In our next post, we will show how to set up a typical small network monitoring in approximately 10 minutes.