[hylafax-users] possible receive error

[shouldbe q931 <shouldbeq931@xxxxxxxxxxxxxx>]

Hi All,

 

After a 'fun' weekend working my way through the documentation, and doing several rebuilds as I documented the process and discovered things that could be done in better ways, I have Hylafax 5.2.9 running on Ubuntu Serve 9.0.4 with a Dialogic Diva Server BRI card, forwarding to specific email addresses based on the DDI, and I'm VERY happy :-)

 

I tried both the 6.0.2 and the 5.2.9, but it was on the 5.2.9 one that I fixed the DDI routing, and so it has stayed on, however the below happened on both.

 

Doing all of this remotely, I was limited in the devices that I could use to send, but I had remote access to a Windows 2003 server running fax services with an external USR Sportster modem, and a Windows 2008 server running fax services with a Diva Server BRI ISDN T/A. The USR Sportster on the 2003 server is connected to an analogue port on the same phone system as the HylaFAX install, the phone system is connected to teh PSTN by 4 BRI trunks, and the Windows 2008 server is connected to a BRI port on a different phone system which is connected to teh PSTN by 3 ISDN30 trunks.

 

With both the 2003 and the 2008 server would frequently, but not always, error with "The Remote fax machine did not respond in time, and the call was terminated".

 

I appreciate that this might well be a Microsoft issue, but as it appears from two operating systems, each with different types of devices for sending, it woudl appear at least on the surface to be a Hylafax issue.

 

I have also confirmed that the two servers can send faxes to each other in both directions without any errors

 

The Hylafax log for a session with the error from the 2003 server is as below.

 

-------------------------------------------------------------------------------------------------

Jul 06 07:42:44.12: [ 2379]: SESSION BEGIN 000000011 +442077982020
Jul 06 07:42:44.12: [ 2379]: HylaFAX (tm) Version 5.2.9
Jul 06 07:42:44.12: [ 2379]: CallID: '<DDI>' '<CLI>'
Jul 06 07:42:44.12: [ 2379]: <-- [4:ATA\r]
Jul 06 07:42:57.12: [ 2379]: --> [5:+FCON]
Jul 06 07:42:57.12: [ 2379]: ANSWER: FAX CONNECTION  DEVICE '/dev/ttyds01'
Jul 06 07:42:57.12: [ 2379]: STATE CHANGE: ANSWERING -> RECEIVING
Jul 06 07:42:57.12: [ 2379]: RECV FAX: begin
Jul 06 07:42:57.13: [ 2379]: --> [26:+FTSI:Fax                 ]
Jul 06 07:42:57.13: [ 2379]: REMOTE TSI "Fax"
Jul 06 07:42:57.13: [ 2379]: --> [22:+FDCS: 1,5,0,0,0,0,0,0]
Jul 06 07:42:57.13: [ 2379]: REMOTE wants 14400 bit/s
Jul 06 07:42:57.13: [ 2379]: REMOTE wants A4 page width (215 mm)
Jul 06 07:42:57.13: [ 2379]: REMOTE wants A4 page length (297 mm)
Jul 06 07:42:57.13: [ 2379]: REMOTE wants 7.7 line/mm
Jul 06 07:42:57.13: [ 2379]: REMOTE wants 1-D MH
Jul 06 07:42:57.13: [ 2379]: --> [2:OK]
Jul 06 07:42:57.13: [ 2379]: <-- [7:AT+FDR\r]
Jul 06 07:42:57.13: [ 2379]: --> [5:+FCFR]
Jul 06 07:42:57.13: [ 2379]: --> [22:+FDCS: 1,5,0,0,0,0,0,0]
Jul 06 07:42:57.13: [ 2379]: REMOTE wants 14400 bit/s
Jul 06 07:42:57.13: [ 2379]: REMOTE wants A4 page width (215 mm)
Jul 06 07:42:57.13: [ 2379]: REMOTE wants A4 page length (297 mm)
Jul 06 07:42:57.13: [ 2379]: REMOTE wants 7.7 line/mm
Jul 06 07:42:57.13: [ 2379]: REMOTE wants 1-D MH
Jul 06 07:42:57.13: [ 2379]: --> [7:CONNECT]
Jul 06 07:42:57.13: [ 2379]: RECV: begin page
Jul 06 07:42:57.13: [ 2379]: RECV: send trigger 021
Jul 06 07:42:57.13: [ 2379]: <-- data [1]
Jul 06 07:42:57.13: [ 2379]: Copy quality checking performed by host
Jul 06 07:43:05.30: [ 2379]: RECV/CQ: Bad 1D pixel count, row 2024, got 0, expected 1728
Jul 06 07:43:05.30: [ 2379]: RECV/CQ: Bad 1D pixel count, row 2025, got 0, expected 1728
Jul 06 07:43:05.30: [ 2379]: RECV/CQ: Bad 1D pixel count, row 2026, got 0, expected 1728
Jul 06 07:43:05.30: [ 2379]: RECV/CQ: Bad 1D pixel count, row 2027, got 0, expected 1728
Jul 06 07:43:05.30: [ 2379]: RECV/CQ: Adjusting for RTC found at row 2024
Jul 06 07:43:05.30: [ 2379]: RECV: 2024 total lines, 0 bad lines, 0 consecutive bad lines
Jul 06 07:43:05.30: [ 2379]: --> [13:+FPTS: 2,2024]
Jul 06 07:43:05.30: [ 2379]: --> [7:+FET: 2]
Jul 06 07:43:05.30: [ 2379]: RECV recv EOP (no more pages or documents)
Jul 06 07:43:05.30: [ 2379]: --> [2:OK]
Jul 06 07:43:05.30: [ 2379]: RECV send MCF (message confirmation)
Jul 06 07:43:05.30: [ 2379]: RECV FAX (000000011): from Fax, page 1 in 0:00:08, A4, 7.7 line/mm, 1-D MH, 14400 bit/s
Jul 06 07:43:06.30: [ 2379]: RECV FAX (000000011): recvq/fax000000011.tif from Fax, subaddress <unspecified>, 1 pages in 0:00:08
Jul 06 07:43:06.30: [ 2379]: <-- [7:AT+FDR\r]
Jul 06 07:43:36.30: [ 2379]: MODEM TIMEOUT: reading line from modem
Jul 06 07:43:36.30: [ 2379]: MODEM <Timeout>
Jul 06 07:43:36.30: [ 2379]: RECV FAX: bin/faxrcvd 'recvq/fax000000011.tif' 'ttyds01' '000000011' '' '<DDI>' '<CLI>'
Jul 06 07:43:36.31: [ 2379]: RECV FAX: end
Jul 06 07:43:36.31: [ 2379]: SESSION END

-------------------------------------------------------------------------------------------------

 

I have substitued <CLI> & <DDI> as a basic level of santisization

 

the config.ttyds01 is as below

 

-------------------------------------------------------------------------------------------------

# $Id: eicon-2,v 1.6 2008/07/28 16:39:16 faxguy Exp $
#
# prototype config file for the Eicon Diva ISDN Modem
#

#
#
CountryCode:  44
AreaCode:  20
FAXNumber:  +44.20.<local>
LongDistancePrefix: 0
InternationalPrefix: 00
DialStringRules: etc/dialrules
ServerTracing:  1
SessionTracing:  0xFFF
RecvFileMode:  0600
LogFileMode:  0644
DeviceMode:  0600
RingsBeforeAnswer: 2
SpeakerVolume:  off
GettyArgs:  "-b -r -s %s %l"
LocalIdentifier: "Fax Server"
TagLineFont:  etc/lutRS18.pcf
TagLineFormat:  "From %%l|%c|Page %%P of %%T"
MaxRecvPages:  25
#
#

ModemType:  Class2
ModemRate:  57600
    # this may need to be 57600
ModemFlowControl: rtscts
ModemNoFlowCmd:  AT&K0
ModemSoftFlowCmd: AT&K4
ModemHardFlowCmd: AT&K3

#
# Enable this to set Caller*ID Number
#
#ModemSetOriginCmd:     AT+iO%d

#
# Make sure the modem gives us Caller*ID in a legible way.
#
ModemResetCmds:         AT#CID=14+iQ=a1

#
# Eicon recommends initializing with AT+iQ=a1 where "1" is
# the controller number.  This command binds the tty to that
# controller and bypasses the "channel pool" approach.
#
# There are 4 controllers on the 4BRI card.
# For each controller You have to configure 2 ttyds<xx> devices.
#
# 1st controller - ttyds01 (AT#CID=14+iQ=a1), ttyds02 (AT#CID=14+iQ=a1)
# 2nd controller - ttyds03 (AT#CID=14+iQ=a2), ttyds04 (AT#CID=14+iQ=a2)
# 3rd controller - ttyds05 (AT#CID=14+iQ=a3), ttyds06 (AT#CID=14+iQ=a3)
# 4th controller - ttyds07 (AT#CID=14+iQ=a4), ttyds08 (AT#CID=14+iQ=a4)
#
#ModemResetCmds:         AT#CID=14+iQ=a1

#
# These should match the modem responses for DID and Caller*ID.
#
CallIDPattern:  "DAD: "
CallIDPattern:  "CID: "

Class2APQueryCmd: none
Class2SPLCmd:  none
Class2TBCCmd:  none
Class2PHCTOCmd:         none

#
# As Eicon's "Class 2" is a cross-breed and abomination between
# Class 2.0, 2.1, and old-style 2 it may require some testing
# to see if ECM type 2.0 or 2 is used and whether or not hexadecimal
# is presented in the +FDCC response.
#
#Class2ECMType:  2.0 # follows Class 2.0 spec, not Class 2
Class2ECMType:          2
#
Class2UseHex:           true
#Class2UseHex:           false

-------------------------------------------------------------------------------------------------

 

I have substituted <local> for the fax number as a basic level of sanitisation.

 

I'd appreciate any asistance in resolving this.

 

Cheers