Re: [hylafax-users] Large number of failures receiving faxes ~38% failure rate

[Lee Howard <faxguy@xxxxxxxxxxxxxxxx>]

Aug 19 11:27:19.88: [ 3993]: SESSION BEGIN 000005985 18125551234
Aug 19 11:27:19.88: [ 3993]: HylaFAX (tm) Version 6.0.4
Aug 19 11:27:19.88: [ 3993]: <-- [13:AT+FCLASS=1A\r]
Aug 19 11:27:25.14: [ 3993]: --> [7:CONNECT]
Aug 19 11:27:25.14: [ 3993]: ANSWER: FAX CONNECTION DEVICE '/dev/ttyR2'
Aug 19 11:27:25.14: [ 3993]: RECV FAX: begin

Okay, so you're using HylaFAX 6.0.4 and are using a Comtrol RocketModem 
of some kind...

Aug 19 11:27:25.14: [ 3993]: <-- data [32]
Aug 19 11:27:25.14: [ 3993]: <-- data [2]
Aug 19 11:27:27.11: [ 3993]: --> [7:CONNECT]
Aug 19 11:27:27.11: [ 3993]: <-- data [23]
Aug 19 11:27:27.11: [ 3993]: <-- data [2]
Aug 19 11:27:28.28: [ 3993]: --> [7:CONNECT]
Aug 19 11:27:28.28: [ 3993]: <-- data [13]
Aug 19 11:27:28.28: [ 3993]: <-- data [2]
Aug 19 11:27:29.17: [ 3993]: --> [2:OK]
Aug 19 11:27:29.17: [ 3993]: <-- [9:AT+FRH=3\r]
Aug 19 11:27:30.23: [ 3993]: --> [7:CONNECT]
Aug 19 11:27:30.26: [ 3993]: --> [5:ERROR]
Aug 19 11:27:30.26: [ 3993]: MODEM Command error
Aug 19 11:27:30.26: [ 3993]: FCS error

This FCS error here is indicative of audio corruption.  However, audio 
corruption at the outset of a call is not entirely unusual as echo 
cancelers and other features on certain types of phone lines may be 
initializing at this point.  The fax protocol appears to recover...

Aug 19 11:27:30.26: [ 3993]: DELAY 70 ms
Aug 19 11:27:30.33: [ 3993]: <-- [9:AT+FTH=3\r]
Aug 19 11:27:30.44: [ 3993]: --> [7:CONNECT]
Aug 19 11:27:30.44: [ 3993]: <-- data [3]
Aug 19 11:27:30.44: [ 3993]: <-- data [2]
Aug 19 11:27:31.65: [ 3993]: --> [2:OK]
Aug 19 11:27:31.65: [ 3993]: RECV send CRP (command repeat)
Aug 19 11:27:31.65: [ 3993]: <-- [9:AT+FRH=3\r]
Aug 19 11:27:37.09: [ 3993]: --> [7:CONNECT]
Aug 19 11:27:37.34: [ 3993]: --> [2:OK]
Aug 19 11:27:37.34: [ 3993]: RECV recv DCS (command signal)
Aug 19 11:27:37.34: [ 3993]: REMOTE wants 9600 bit/s
Aug 19 11:27:37.34: [ 3993]: REMOTE wants A4 page width (215 mm)
Aug 19 11:27:37.34: [ 3993]: REMOTE wants unlimited page length
Aug 19 11:27:37.34: [ 3993]: REMOTE wants 3.85 line/mm
Aug 19 11:27:37.34: [ 3993]: REMOTE wants 2-D MMR
Aug 19 11:27:37.34: [ 3993]: REMOTE wants T.30 Annex A, 256-byte ECM
Aug 19 11:27:37.34: [ 3993]: RECV training at v.29 9600 bit/s
Aug 19 11:27:37.34: [ 3993]: <-- [10:AT+FRM=96\r]
Aug 19 11:27:37.70: [ 3993]: --> [7:CONNECT]
Aug 19 11:27:39.27: [ 3993]: RECV: TCF 1896 bytes, 3% non-zero, 1830 
zero-run
Aug 19 11:27:39.30: [ 3993]: --> [10:NO CARRIER]

This tells us that TCF looked clean.  1830 bytes at 9600 bps is 1.525 
seconds, and so the 3% non-zero is probably trailing noise that occurred 
as the modem dropped the carrier.

Aug 19 11:27:39.30: [ 3993]: DELAY 70 ms
Aug 19 11:27:39.37: [ 3993]: TRAINING succeeded
Aug 19 11:27:39.37: [ 3993]: <-- [9:AT+FTH=3\r]
Aug 19 11:27:39.49: [ 3993]: --> [7:CONNECT]
Aug 19 11:27:39.49: [ 3993]: <-- data [3]
Aug 19 11:27:39.49: [ 3993]: <-- data [2]
Aug 19 11:27:40.71: [ 3993]: --> [2:OK]
Aug 19 11:27:40.71: [ 3993]: <-- [10:AT+FRM=96\r]
Aug 19 11:27:41.68: [ 3993]: --> [7:CONNECT]
Aug 19 11:27:41.78: [ 3993]: HDLC frame not byte-oriented. Trailing 
byte: 0xfc
Aug 19 11:27:41.89: [ 3993]: Bad HDLC terminating flag received.
Aug 19 11:27:41.89: [ 3993]: HDLC frame with bad address field 0xa2
Aug 19 11:27:41.89: [ 3993]: HDLC frame not byte-oriented. Trailing 
byte: 0xf0
Aug 19 11:27:42.11: [ 3993]: Bad HDLC terminating flag received.
Aug 19 11:27:42.12: [ 3993]: Bad HDLC terminating flag received.
Aug 19 11:27:42.23: [ 3993]: HDLC frame not byte-oriented. Trailing 
byte: 0xf0
Aug 19 11:27:42.23: [ 3993]: Bad HDLC terminating flag received.
Aug 19 11:27:42.34: [ 3993]: HDLC frame not byte-oriented. Trailing 
byte: 0x80

And here we see that the modem didn't train very well at all.  So either 
there is a problem with the modem firmware or there is a problem with 
the audio on the call.

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*