Re: [hylafax-users] Cannot recieve from certain senders

[Lee Howard <faxguy@xxxxxxxxxxxxxxxx>]

johnh@xxxxxxxxxxxxxxxxx wrote:

> Greetings,
>
> My Setup: Hylafax 4.3.3 on Fedora Core 6 x64, 4 Modems, all MODEM LT V.90 
> 1.0 MT5634ZBA SERIAL DATA/FAX MODEM VERSION 8.19J, Digi Xp serial card.
>
> May 01 09:21:03.70: [13564]: <-- [9:AT+FRH=3\r]
> May 01 09:21:03.72: [13564]: --> [7:CONNECT]
> May 01 09:21:04.19: [13564]: --> [2:OK]
> May 01 09:21:04.19: [13564]: REMOTE wants 14400 bit/s
> May 01 09:21:04.19: [13564]: REMOTE wants A4 page width (215 mm)
> May 01 09:21:04.19: [13564]: REMOTE wants unlimited page length
> May 01 09:21:04.19: [13564]: REMOTE wants 3.85 line/mm
> May 01 09:21:04.19: [13564]: REMOTE wants 2-D MMR
> May 01 09:21:04.19: [13564]: REMOTE wants T.30 Annex A, 256-byte ECM
> May 01 09:21:04.19: [13564]: RECV training at v.17 14400 bit/s
> May 01 09:21:04.19: [13564]: <-- [11:AT+FRM=145\r]
> May 01 09:21:04.27: [13564]: --> [8:+FCERROR]
> May 01 09:21:04.27: [13564]: <-- [11:AT+FRM=145\r]
> May 01 09:21:04.34: [13564]: --> [8:+FCERROR]
> May 01 09:21:04.34: [13564]: <-- [9:AT+FRH=3\r]
> May 01 09:21:04.80: [13564]: --> [7:CONNECT]
> May 01 09:21:05.11: [13564]: --> [10:NO CARRIER]
> May 01 09:21:05.11: [13564]: MODEM No carrier

It would seem that the modem said "OK" up there at 09:21:04.19 before 
the sender's V.21 carrier actually dropped.  And it would seem that the 
sender is rather slow to drop the carrier.  Consequently, when the modem 
is told to go look for the V.17 carrier it finds the V.21 carrier still 
present, and it's present long enough to exhaust Class1RMPersistence.

The easy thing to do would be to increase Class1RMPersistence to 3 or 
maybe 4.  That should make this problem go away (although it may 
complicate other situations).

One of the down-sides here is that the AT+FRH=3 command reports CONNECT 
when it's hearing the V.17 carrier - which it should not do... and that 
makes the session fumble a bit more.  There is a workaround to this in 
HylaFAX+ 5.1.3 called Class1HasRHConnectBug, which, when enabled (and 
you should enable it on this modem) would help the situation a bit, 
however in this case the real culprit is the premature OK coupled with 
the slow-to-drop V.21 carrier at the sender's end.

What you *really* need here is for Class1TCFRecv hack to be done 
differently, with an AT+FRS=1 command.  If you'd like to test that I can 
send you a patch.

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*