flexfax: The all familiar MPS!!!!!

[Dmitri Granovski <grand@ica.net>]

So, it seems that I have run into an interesting problem:
Sending FROM Hylafax using External AOPEN 56K setup as follows:
ModemType:              Class2          # use class 2 interface
ModemRate:              38400
ModemFlowControl:       rtscts          # default
ModemDialCmd:           ATDT9%s         # Dial 9 to pass though
#ModemResetCmds:        AT&F2           # what you want for a Supra
ModemNoFlowCmd:         AT&K            # disable flow control cmd
ModemHardFlowCmd:       AT&K3           # hardware flow control cmd
ModemSoftFlowCmd:       AT&K4           # software flow control cmd
ModemSetupDTRCmd:       AT&D3           # DTR off causes modem to reset
#ModemSetupDTRCmd:      AT&D2S25=1S38=0 # Dynalink 1414VE needs this
ModemSetupDCDCmd:       AT&C1           # DCD follows carrier
ModemSetupAACmd:        AT+FAA=1+FCLASS=0 # enable in current class
#
ModemSendFillOrder:     LSB2MSB         # as expected
ModemRecvFillOrder:     MSB2LSB         # opposite of what makes sense
#
ModemAnswerFaxBeginCmd: "<19200><xon>"  # modem switches on fax receive

stty </dev/ttyS0
speed 38400 baud; line = 0;
min = 1; time = 0;
-brkint -icrnl -imaxbel
-opost -onlcr
-isig -icanon -iexten -echo -echoe -echok -echoctl -echoke

Receiving TO Hylafax using External Hayes 14.4 Accura setup as follows:
ModemType:              Class1          # use class 1 interface
ModemRate:              38400
ModemFlowControl:       rtscts          # default
#
ModemNoFlowCmd:         AT&K            # disable flow control cmd
ModemHardFlowCmd:       AT&K3           # hardware flow control cmd
ModemSoftFlowCmd:       AT&K4           # software flow control cmd
ModemSetupDTRCmd:       AT&D3           # setup so DTR drop resets modem

ModemSetupDCDCmd:       AT&C1           # setup so DCD reflects carrier
(or not)
ModemSetupAACmd:        AT+FAA=0        # in case modem supports Class 2
also
#
ModemMfrQueryCmd:       !Rockwell       # no way to query modem
ModemModelQueryCmd:     !RC32ACL        # not always right
ModemRevQueryCmd:       ATI3

stty </dev/ttyS1
speed 38400 baud; line = 0;
eof = ^A; min = 1; time = 0;
-brkint -icrnl -imaxbel
-opost -onlcr
-isig -icanon -iexten -echo -echoe -echok -echoctl -echoke

Now.
Here's the story:
The sending Hylafax has been set up and running for a few months now on
RH6.0. I almost never received the MPS error when sending.

The receiving Hylafax on RH 6.1 is a new addition, I have tried to get
it to work with:

this Hayes Accura 144+ fax 144 External
USR Sportster 33.6 Vi internal
Zoom 56K External (which whould hang up before the first page was even
beginning to send)
another AOPEN 56K External
and no-name 14.4 internal (forget it - last desperation move)
Motorolla Lifestyle 28.8 External (I used this one with MsFAX to test
sending capabilities of the hylafax under RH6.0)

Of course, I can produce all the logs (want them?). For both - send and
receive.
The situation is a bit funny though. Both sending and receiving is done
through *good* lines - so no question about line quality,
All the fax/modems are more or less new;

Can someone explain to me why am I getting MPS (or equally EOP with a 1
page fax) 100% of the time?

send:
Mar 10 01:01:32.20: [13562]: <-- data [491]
Mar 10 01:01:32.20: [13562]: SENT 38379 bytes of data
Mar 10 01:01:32.20: [13562]: <-- data [2]
Mar 10 01:01:32.20: [13562]: SEND end page
Mar 10 01:01:35.48: [13562]: --> [2:OK]
Mar 10 01:01:35.48: [13562]: SEND send MPS (more pages, same document)
Mar 10 01:01:35.48: [13562]: <-- [9:AT+FET=0\r]
Mar 10 01:01:49.22: [13562]: --> [9:+FHNG: 52]
Mar 10 01:01:49.22: [13562]: REMOTE HANGUP: No response to MPS repeated
3 times
(code 52)
Mar 10 01:01:49.22: [13562]: <-- [5:ATH0\r]
Mar 10 01:01:50.64: [13562]: --> [2:OK]
Mar 10 01:01:50.64: [13562]: SESSION END

receive:
Mar 10 00:06:21.76: [ 1860]: <-- data [3]
Mar 10 00:06:21.76: [ 1860]: <-- data [2]
Mar 10 00:06:23.16: [ 1860]: --> [2:OK]
Mar 10 00:06:23.16: [ 1860]: TRAINING succeeded
Mar 10 00:06:23.16: [ 1860]: <-- [11:AT+FRM=146\r]
Mar 10 00:06:23.95: [ 1860]: --> [7:CONNECT]
Mar 10 00:06:23.95: [ 1860]: RECV: begin page

The worst part about it is that I have to stop hylafax every time and
kill faxgetty, otherwise, the receiving fax never hangs up!

Dmitri