Hello Tom,
Thanks for the reply. So in other words, the postgres clients do not use a
specific range for the source port (as a specific decision by the developers,
or as written down in a given spec), but rather relies on the operating
system's socket implementation to assign a source port?
So in practice the source port can be any non-privileged port (from postgreses
point of view)?
I only ask this clarification to be sure to pass on the correct info on to the
administrator of the firewall...
Thanks,
Matt Smith
On Mon, 6 Feb 2006 04:25 pm, Tom Lane wrote:
> Matthew Smith <mps@utas.edu.au> writes:
> > I am trying to access a postgresql server though a tightly configured
> > firewall, and I'd like to know the source port range is that postgresql
> > clients use to access the database.
>
> You seem to be under the misimpression that Postgres might have
> something to do with that. We don't --- this is entirely determined
> by the whims of your local kernel. I would even go so far as to say
> that if you write code that makes any assumption in this regard,
> you will have only yourself to blame when (not if) it fails.
>
> regards, tom lane
