I’ve sometimes seen people use non-standardized ports for applications such as HTTPs, 8080 is pretty common.
Why would we want to use non-standard ports in our network and not just stick to the ones that were standardized? So for example, if our server is running HTTPs, it would have port 443 open on it.
That’s a great question! There are a few reasons why network administrators might choose to use non-standard ports, and here are the most common:
First of all, it is a quick and easy way to establish a low level of security. By using non-standard ports, you can reduce the visibility of the service to automated scanning tools that scan default port numbers. However, not that this should not be the only security you rely on, but simply one more layer of security in addition to more thorough methods.
Another reason is to run multiple services on a single server. For example, in a network that I am network administrator for, we have a single server that has a web interface for an NMS running on port 8000, and a telephone directory for internal users running on port 80. We can run two separate web services on the same server by simply using different port numbers.
It can also be a quick and dirty way of bypassing restrictions that may be in place. Some networks may block standard ports, either for security reasons or to enforce usage policies. In such cases, you might choose to run your service on a non-standard port to bypass these restrictions.
Typically the reasons are many, but they should be used very carefully.