Re: TPC-OPER: Re: flexfax: Large number of errors with Hylafax and

[David Woolley <david@djwhome.demon.co.uk>]

> ] We have someone stating in no uncertain terms that HylaFAX's response to
> ] 'RTN negative' is wrong. Most respondents on-list refuse to address this
> ] problem, and those who do say anything about it do not feel they have
> ] the authority to know.

I haven't got the definitive documents, but the Rockwell chipset documentation
(from their Japanese site, not the corporate one, and a superset of the
capabilities of any one modem) says that there are 5 possible statuses
at this point (Phase D):

MCF - good
RTN - bad, please retrain
RTP - good, please retrain
PIN - bad, please interrupt
PIP - good, please interrupt

>From which I would feel fairly confident to say that a sender should retransmit
the page on RTN, if possible, as the receiver would have sent RTP if it wanted
a retrain, but not a retransmission.

The only remaining question is whether there are machines that send RTN in
response to data errors, rather than analogue signal quality, and therefore
whether the RTN could be due to a fault at the sender, rather than poor line
quality (in which case it is possible that retransmissions, although correct,
could fail).

The Rockwell documents don't say what to do with a bad page, and it may well
be that the ITU-T ones don't either, as they may only specify the protocol, 
not the actions of higher level functions.