Re: Unspecified Transmit Phase C Error

On 05/09/2018 08:37 AM, Brian R wrote:
> I have changed the modemflowcontrol to RTSCTS per other threads with=20
> similar problems.

iaxmodem does not support hardware flow control as the pty interface=20
does not emulate RTS/CTS pins.

> May 08 13:02:06.27: [ 3625]: REMOTE best rate 14400 bit/s
> May 08 13:02:06.27: [ 3625]: REMOTE max A4 page width (215 mm)
> May 08 13:02:06.27: [ 3625]: REMOTE max unlimited page length
> May 08 13:02:06.27: [ 3625]: REMOTE best vres 7.7 line/mm
> May 08 13:02:06.27: [ 3625]: REMOTE format support: MH
> May 08 13:02:06.27: [ 3625]: REMOTE best 0 ms/scanline

The receiver does not support error correction mode, unfortunately.

> May 08 13:02:06.27: [ 3625]: USE 9600 bit/s
> May 08 13:02:06.27: [ 3625]: SEND file "docq/doc.ps;30"
> May 08 13:02:06.27: [ 3625]: USE A4 page width (215 mm)
> May 08 13:02:06.27: [ 3625]: USE unlimited page length
> May 08 13:02:06.27: [ 3625]: USE 3.85 line/mm
> May 08 13:02:06.27: [ 3625]: USE 1-D MH
> May 08 13:02:06.27: [ 3625]: USE 0 ms/scanline
> May 08 13:02:06.27: [ 3625]: SEND training at v.29 9600 bit/s

... and for some reason you've chosen to not try V.17.

> May 08 13:03:00.33: [ 3625]: SEND send EOP (no more pages or documents)
> May 08 13:03:00.33: [ 3625]: <-- [9:AT+FTH=3D3\r]
> May 08 13:03:00.35: [ 3625]: --> [7:CONNECT]
> May 08 13:03:00.35: [ 3625]: <-- data [3]
> May 08 13:03:00.35: [ 3625]: <-- data [2]
> May 08 13:03:01.43: [ 3625]: --> [2:OK]
> May 08 13:03:01.43: [ 3625]: <-- [9:AT+FRH=3D3\r]
> May 08 13:03:03.07: [ 3625]: --> [7:CONNECT]
> May 08 13:03:04.11: [ 3625]: --> [2:OK]
> May 08 13:03:04.11: [ 3625]: SEND recv RTN (retrain negative)

RTN is a complicated signal to interpret.=C2=A0 It only appears when erro=
r=20
correction mode is not used.

You're sending a one-page document, and the receiver rejected the page=20
image quality of the first attempt to send the page image data.=C2=A0 For=
 any=20
further pages to be sent a retrain must occur.

The fax specification does not stipulate whether or not the receiver is=20
to expect the sender to retransmit a page following RTN.=C2=A0 The receiv=
er=20
may or may not save or print that page.=C2=A0 All we know is that the=20
receiver rejects the page image quality.=C2=A0 So, if the receiver is not=
=20
expecting the sender to retransmit pages following RTN then at this=20
point you'd have to hang up and dial again to get a receipt confirmation=20
(MCF).=C2=A0 By default HylaFAX will retransmit pages up to two times mor=
e=20
before proceeding with the next page.

> May 08 13:03:04.11: [ 3625]: SEND training at v.29 7200 bit/s
> ....
> May 08 13:03:10.97: [ 3625]: TRAINING succeeded

The receiver appears to have tolerated a retraining after RTN.

> May 08 13:03:11.47: [ 3625]: SEND end page
> May 08 13:04:11.47: [ 3625]: Unspecified Transmit Phase C error

However, after transmitting the page image data the modem does not=20
respond for 60 seconds.=C2=A0 That produces a failure.

There were 55889 bytes in that page, but because you don't have any=20
functional flow control all of those bytes got sent to the modem at=20
once.=C2=A0 At 7200 bps those 55889 bytes will take a minimum of 62 secon=
ds=20
to transmit, so the 60-second modem response "dataTimeout" occurred=20
because the modem was busy transmitting the data still after 60=20
seconds.=C2=A0 So, the failure in this case was due to the lack of flow=20
control and/or a timeout that was too short.

It's still possible that even with flow control that you would have run=20
into the same problem because iaxmodems have a large buffer. (So, flow=20
control may never have even come into play in the first place.)=C2=A0 The=
re=20
were fixes in HylaFAX+ 5.5.7 to address the needed increase in the=20
"dataTimeout".

https://sourceforge.net/p/hylafax/HylaFAX+/2370/
https://sourceforge.net/p/hylafax/HylaFAX+/2364/
(And the 5 commits in-between were all related, too.)

Thanks,

Lee.


____________________ HylaFAX(tm) Users Mailing List _____________________=
__
  To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.c=
gi
 On UNIX: mail -s unsubscribe [email protected] < /dev/nu=
ll
  *To learn about commercial HylaFAX(tm) support, mail [email protected].*