Question

Why is there this distinction between the registrated ports in well-knows and normal registred ports?
Why if I decide to bind a socket on 23 I need the root privileges while on the port 1025 I do not need? What reason brought to this decision?
And why, let's suppose I make a software that use a socket it is better to register the port on which it bind itself?

Was it helpful?

Solution

In the time of multiuser unix systems in university campuses, where normal users never had root access, unix systems were using ports below 1024 to indicate that the process that created the port/connection was 'trusted' by the local admin. This was used to f.e. indicate what the local user name in system A was so that a connection would use that user in system B.

Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top