Network monitoring systems and network testing technologies have evolved greatly over time, offering more automation, more ways to test servers and systems, and more ways to alert network administrators, sysadmins and IT managers of problems with the network and internal systems.
What is network monitoring?
Network monitoring uses an IT system or systems to constantly monitor a computer network for slow or failing components. It can check components including like application, email or web servers, peripherals and the health of the network itself.
The systems can be set to notify the network administrator in case of outages via email, pager or other means.
What sorts of problems cause network or IT failures?
Network monitoring systems monitor the network for problems caused by overloaded and crashed application servers, web servers or other systems, or issues with network connections and other devices.
These problems might be caused by faulty equipment or even human error.
How do network monitoring systems work?
They tend to check the activity and health of internal systems via the network by sending a signal, called a ping, to various system ports.
The testing system uses a huge variety of check intervals, which is basically the time between pings. This could vary from every four hours to every single minute.
Typically, most network monitoring services test the server anywhere between once every hour to once a minute.
Which network protocols do network monitoring systems check?
Network monitoring systems can check all manner of network protocols, particularly internet protocols.
For example, web site monitoring services can check HTTP pages, HTTPS, SNMP, FTP, SMTP, POP3, IMAP, DNS, SSH, TELNET, SSL, and TCP.
So, when it comes to web servers, a network monitoring program may send an HTTP request to a web server to determine its status.
But when it comes to an email server, the monitoring software may periodically send a test message through SMTP, but which is retrieved by IMAP or POP3. By doing this it can emulate the path of a typical message and check the health of the networks and servers it passes through.
What do network monitoring systems measure?
The most frequently measured metrics are to check the response time and availability, or uptime, of a server.
Network monitoring also checks for the consistency and reliability of application servers and IT equipment.
What happens when a system fails?
If a network monitoring tool detects a problem with a system, via a status request failure, for example when a connection cannot be established, this results in a time out.
Most users will see a time out when they try to access a web site service or application which is struggling to respond, or send an email to a recipient with email server issues that delay or stop the message going through.
In cases where there is a status request failure, the network monitoring system will produce an action.
These actions can vary, and might involve actively trying to get the network manager’s attention via an alarm – an SMS text, a pager message or an email.
Alternatively, automatic failover systems may be activated to remove the troubled server from duty until it can be repaired.
Commercial and Open Source Network Monitoring Tools:
The Multi Router Traffic Grapher (MRTG) monitors and graphs traffic on network links. It also creates daily, weekly, and monthly graphs.
Nagios is an open source system that offers network monitoring and notification and problem management. Nagios runs on Linux and has been around for over 10 years, with a large user base and support community.
OpenNMS is an open source network monitoring platform aimed at enterprises. It has a strong user and open source community support group.
SolarWinds Orion Network Performance Monitor (Orion NPM) is designed with simplicity and ease in mind. Solarwinds itself is known for many strong free network tools.
- Spiceworks IT Management
Spiceworks is a free, (ad sponsored) network monitoring and issue tracking platform. It has a large and active user community and a highly customisable interface.