Re: [hylafax-users] hdlc errors

[Lee Howard <faxguy@xxxxxxxxxxxxxxxx>]

Robert Haas wrote:

> I am running Hylafax 4.2.1 on a RedHat 7.3 system with a modem that 
> identifies itself as:
>
> MODEM HAYES V.90 SERIAL S052099GV
>
> The modem is connected to an analog card on a Mitel SX200 and receives 
> calls via direct inward dial.  We're using this only for purposes of 
> receiving faxes, not sending them.  Faxes from some senders come 
> through with no problem at all, but other senders experience repeated 
> failures.  The two common error messages are:
>
>  Failed to properly detect high-speed data carrier.


This error is a strong indicator that the sender did not get MCF or CFR 
and has gone back to retransmit prologue frames or PPS/PPM.

Current HylaFAX CVS HEAD (to become 4.2.2 most likely) is much better at 
handling these situations than previously.  When you see this error 
after updating, then it generally means that the sender disconnected 
unexpectedly.  Furthermore, if your modem supports AT+FAR=1, then you'll 
do well to set Class1AdaptiveRecvCmd to it (if your modem doesn't 
support it, then don't).  You'll probably want to review 
Class1RMPersistence after updating, also.  (These are documented in 'man 
hylafax-config'.)

> and
>  T.30 T2 timeout, expected signal not received


This generally happens at Phase D (after receiving image data).  
Normally we just needed to wait longer than T2 (the spec) to recover.  
This also is improved in CVS HEAD.

> May 03 20:11:40.42: [  828]: RECV recv PPS (partial page signal)
> May 03 20:11:40.42: [  828]: RECV recv MPS (more pages, same document)
> May 03 20:11:40.42: [  828]: RECV received 8 frames of block 1 of page 8
> May 03 20:11:40.43: [  828]: RECV: 1859 bytes of data, 1112 total lines
> May 03 20:11:40.43: [  828]: <-- [9:AT+FRS=7\r]
> May 03 20:11:40.57: [  828]: --> [2:OK]
> May 03 20:11:40.57: [  828]: <-- [9:AT+FTH=3\r]
> May 03 20:11:41.44: [  828]: --> [7:CONNECT]
> May 03 20:11:41.44: [  828]: RECV send MCF (message confirmation)
> May 03 20:11:41.44: [  828]: RECV FAX (000000323): from <UNSPECIFIED>, 
> page 8 in 0:06, INF, 3.85 line/mm, 2-D MMR, 14400 bit/s
> May 03 20:11:41.76: [  828]: --> [2:OK]
> May 03 20:11:41.76: [  828]: <-- [11:AT+FRM=146\r]
> May 03 20:11:48.87: [  828]: --> [2:]
> May 03 20:11:48.87: [  828]: --> [2:OK]


Yeah.  Either the sender didn't get the MCF signal properly or they 
decided to abort (in which case you may see DCN with current CVS HEAD).

> May 03 18:40:16.88: [32206]: RECV received frame number 63
> May 03 18:40:17.02: [32206]: RECV received frame number 64
> May 03 18:40:17.16: [32206]: RECV received frame number 65
> May 03 18:40:17.23: [32206]: RECV received frame number 66
> May 03 18:40:17.23: [32206]: RECV frame FCS check failed
> May 03 18:40:17.30: [32206]: Bad HDLC terminating flag received.
> May 03 18:40:17.45: [32206]: HDLC frame not byte-oriented.  Trailing 
> byte: 0xfc
> May 03 18:40:17.45: [32206]: HDLC frame not byte-oriented.  Trailing 
> byte: 0xc0
> May 03 18:40:17.45: [32206]: Bad HDLC terminating flag received.
> May 03 18:40:17.52: [32206]: Bad HDLC terminating flag received.
> May 03 18:40:17.60: [32206]: HDLC frame not byte-oriented.  Trailing 
> byte: 0xfc
> May 03 18:40:17.60: [32206]: Bad HDLC terminating flag received.
> May 03 18:40:17.60: [32206]: HDLC frame not byte-oriented.  Trailing 
> byte: 0xf8
> May 03 18:40:17.60: [32206]: HDLC frame not byte-oriented.  Trailing 
> byte: 0x80
> May 03 18:40:17.68: [32206]: Bad HDLC terminating flag received.
> May 03 18:40:17.68: [32206]: HDLC frame not byte-oriented.  Trailing 
> byte: 0x80
> May 03 18:40:17.68: [32206]: HDLC frame not byte-oriented.  Trailing 
> byte: 0xfc
> May 03 18:40:17.68: [32206]: Bad HDLC terminating flag received.
> May 03 18:40:17.82: [32206]: HDLC frame not byte-oriented.  Trailing 
> byte: 0x80
> May 03 18:40:17.90: [32206]: Bad HDLC terminating flag received.
> May 03 18:40:17.90: [32206]: Bad HDLC terminating flag received.
> May 03 18:40:17.98: [32206]: Bad HDLC terminating flag received.
> [...more crap...]


If this doesn't recover with CVS HEAD, then it's probably a 
disconnection or serious line problems.  But sometimes it's just due to 
a broken modulator (sender or receiver).

> I guess what I'm wondering is - is this a problem with my phone line, 
> my modem, my firmware, or Hylafax?  Any recommendations on how to fix?


Update to current CVS HEAD first, then re-evaluate the situation.

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*