[rescue] Static IP! Now what?

[Andrew Jones]

On 04/05/2017 02:11 PM, John Hudak wrote:
> a FAX signal being carried by VOIP  is a very bad combination.  Don't do
> it.  FAX protocol dictates that the timing relationships between and among
> tone signals and handshaking are tight and deterministic.  Attempting to
> run the FAX protocol over a temporal
>  non-deterministic transport is asking for trouble.  There are times when
> the communication will 'work'.  Most of the time, it will not. By this is
> mean, yea, you *may* get one page to FAX successfully, you won't get a doc
> of >1 to work.

Getting into the weeds a bit, VOIP audio codecs are designed around a
psycho-acoustic model, so they remove noise you can't hear very well.

A fax machine counts on using every iota of available audio bandwidth in
a conventional telephone line.  That is to say, nearly all of its signal
falls under the category "noise you can't hear very well."

If you want to do fax-over-IP, you need a fax-over-IP gateway.  The
relevant ITU standards are T.37 and T.38.  Lots of VOIP gear supports
T.38.  I've never seen a T.37 implementation.