Show/Hide Toolbars

MonitorWare Agent

Navigation: Getting Started > MonitorWare Agent Tutorial

Monitoring External Devices via a Port Probe

Scroll Prev Top Next More

This sample is very similar to the ping probe sample directly above. Thus, we describe it briefly, only.

 

The main difference between the ping probe and the port probe is that the port probe tries to connect to a specific TCP port. As such, it can only be used with TCP based services like mail server, web servers or ftp servers. For the very same reason, the port probe does not only check the status of the machine it is connecting to but rather if a specific, service is available. Let us assume you are interested in monitoring a mail server. If you do a ping probe, the mail server itself might have died while the machine is still running. The ping probe cannot detect this. The port probe, on the other hand, directly connects to the mail server, e.g. on port 25 (the default SMTP port). If the mail server has died, it will probably not answer this connection request and thus the port probe is able to detect the failing state of the service.

 

In our sample, we probe a web server, which typically listens to port 80 (the default port for http). We will send an email alert if the port probe cannot connect successfully to the web server.

 

Because this sample is so close to the previous one, we do not create a new rule set specifically for email alerting. It is already covered in the "Ping Probe". This is a good sample of rule set re-use. It might be clever to rename the rule set in such a case. For simplicity reasons, we have not done this here.

 

Therefore, we begin by creating the new service, done by right-clicking "Services":

 

tutorials_038

Monitoring External Devices via a PortProbe - Figure 1

 

 

After doing so, select the newly created service in the tree view to look at its properties. Be sure to bind it to the "Ping Probe" rule set as seen below:

 

tutorials_039

Monitoring External Devices via a PortProbe - Figure 2

 

Save the configuration and restart the service. From now on, the following mail alert will be generated when the port cannot be connected to:

 

Event message:

Facility: 16

Priority: 6

Source: 192.168.1.1

 

Message:

PortProbe status="fail" target="192.168.1.1" port="80" netstate="10065" message="Couldn't connect to host"