Re: [hylafax-users] V.21 signal reception timeout +FCERROR

[Lee Howard <faxguy@xxxxxxxxxxxxxxxx>]

dev.malst wrote:
> Some clarifications to be more possible self-employed in the log 
> analysis:
>
> 1) Nov 29 09:37:54.85: [ 3155]: --> [*10*:NO CARRIER]
>
> - You say 10 seconds. In the log, the numbers before colon are seconds ?

Date, time (HH:MM:SS.XX)... yes.

> - Arrow *--> *means modem (DCE) to hylafax (DTE) communications ? and 
> *<--* means hylafax (DTE) to modem (DCE) communication ?

Yes.
> 2) Can a VoIP T.38 line allow reception well at some time and wrong at 
> other time ? congestion or something like that ?

Yes, certainly. I assume you speak of T.38 on SIP/UDP. Understand that 
data on UDP can sometimes go wrong because that's the nature of UDP. 
Beyond that, T.38 adds another layer of T.30 protocol operations 
in-between the sender and the receiver. So in a T.38 scenario the 
receiver and sender could be 100% perfect fax devices, but if they do 
something that the T.38 relay doesn't understand, then the fax can fail. 
So as far as reliability is concerned, you can only ever be as reliable 
as the least-perfect T.30 operator.
> 3) Can a VoIP T.38 line allow reception well all times from specific 
> remote fax and produce wrong reception for some other remote fax 
> machine ? this is our situation. Perhaps best remote fax machine can 
> resolve this issue ?

Yes, same as above.

> 4) Can we set hylafax to invalidate all fax communication if REJECT by 
> Page Quality arises ? so remote fax return an error to operator and 
> operator know to resend fax ?

Certainly. For sending faxes that is "RTNHandlingMethod: Giveup" in the 
modem config file and for receiving faxes it is "BadPageHandlingMethod: 
DCN" in the modem config file.
> 5) What’s this “MODEM <Empty line>”
>
> Nov 29 14:30:06.14: [ 3153]: ANSWER: FAX CONNECTION DEVICE 
> '/dev/apsfax429'
>
> Nov 29 14:30:06.14: [ 3153]: RECV FAX: begin
>
> Nov 29 14:30:06.14: [ 3153]: <-- data [32]
>
> Nov 29 14:30:06.14: [ 3153]: <-- data [2]
>
> Nov 29 14:30:06.17: [ 3153]: --> [7:CONNECT]
>
> Nov 29 14:30:06.17: [ 3153]: <-- data [23]
>
> Nov 29 14:30:06.17: [ 3153]: <-- data [2]
>
> Nov 29 14:30:06.19: [ 3153]: --> [7:CONNECT]
>
> Nov 29 14:30:06.19: [ 3153]: <-- data [13]
>
> Nov 29 14:30:06.19: [ 3153]: <-- data [2]
>
> Nov 29 14:30:08.40: [ 3153]: --> [2:OK]
>
> Nov 29 14:30:08.40: [ 3153]: <-- [9:AT+FRH=3\r]
>
> Nov 29 14:30:15.40: [ 3153]: --> [0:]
>
> *Nov 29 14:30:15.40: [ 3153]: MODEM <Empty line>*
>
> Nov 29 14:30:15.40: [ 3153]: <-- data [1]
>
> Nov 29 14:30:15.41: [ 3153]: --> [2:

It means that a timeout occurred waiting for data from the modem. In 
this case the timeout occurred because the modem had not yet reported 
any V.21 HDLC detection.

> 6) What’s this strange character instead of “2:<modem response command 
> like OK or CONNECT or other>” ? I hope will be readable, if not it’s 
> seams a Unicode 2 byte character.
>
> Nov 29 14:30:31.91: [ 3153]: <-- data [2]
>
> Nov 29 14:30:32.24: [ 3153]: --> [2:OK]
>
> Nov 29 14:30:32.24: [ 3153]: <-- [10:AT+FRM=96\r]
>
> Nov 29 14:30:42.14: [ 3153]: <-- data [1]
>
> *Nov 29 14:30:42.15: [ 3153]: --> [2:
> ------------------------------------------------------------------------
> ]*
>
> Nov 29 14:30:42.15: [ 3153]: --> [2:OK]
>
> Nov 29 14:30:42.15: [ 3153]: <-- [9:AT+FRH=3\r]

The two bytes there are DLE+ETX. Those are required characters from the 
modem to indicate an end-of-data stream following a received +FRM 
command. (The +FRM command was aborted due to no V.29 9600 bps detection 
after 10 seconds as expected.
> 7) What’s this error ? local modem - hylafax communication error ?
>
> Nov 29 09:47:14.65: [ 2967]: <-- [9:AT+FRH=3\r]
>
> Nov 29 09:47:15.96: [ 2967]: --> [7:CONNECT]
>
> Nov 29 09:47:17.52: [ 2967]: --> [5:ERROR]
>
> *Nov 29 09:47:17.52: [ 2967]: MODEM Command error*
>
> *Nov 29 09:47:17.52: [ 2967]: FCS error*
>
> Nov 29 09:47:17.52: [ 2967]: <-- [10:AT+FRS=21\r]

A response of "ERROR" here from the modem indicates that although V.21 
HDLC was detected, the data that was demodulated was invalid (likely 
because it did not pass its FCS/CRC checksum). This happens because of 
problems with the audio distorting the modulated data.

Thanks,

Lee.


____________________ HylaFAX(tm) Users Mailing List _______________________
 To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.cgi
On UNIX: mail -s unsubscribe hylafax-users-request@xxxxxxxxxxx < /dev/null
 *To learn about commercial HylaFAX(tm) support, mail sales@xxxxxxxxx*