Re: [hylafax-users] some faxes don't go through asterisk+IAXmodem

[Laurent Lesage <laurent@xxxxxxxxxxxx>]

(sorry, I sent with the wrong address)


Lee Howard a écrit :
> Laurent Lesage wrote:
> > Lee Howard a écrit :
> > > Laurent Lesage wrote:
> > > > > Mar 24 05:44:30.67: [ 3650]: RECV FAX: No sender protocol (T.30 T1 
> > > > > timeout)
> > >
> > > This means that no fax signals were heard from the caller.  It's 
> > > doubtful the caller was a fax machine.
> > Though I do not have your "fax-experience", I had the same impression 
> > (because that's the trace I get for a voice call). But as I wrote, 
> > the same machine some times works. I think it must be an old fax 
> > machine : is it possible that this fax "waits" for sthg without 
> > sending any fax tone?? Before, when there was just a hardware fax 
> > machine on our side, the guy had no problem to send faxes.
>
> Watch the Asterisk CLI on these problem calls.  See if fax detection 
> is robbing the call to somewhere else.  Disable fax detection everywhere.
There is no fax detection. The line is only for faxes. And there does
not seem to be anything wrong in the CLI :

> ^MESC[Kasterisk*CLI> ^M    -- Starting simple switch on 'DAHDI/2-1'
> ^MESC[Kasterisk*CLI> ^M[Mar 25 08:47:26] NOTICE[18841]: 
> chan_dahdi.c:6833 ss_thread: Got event 18 (Ring Begin)...
> ^MESC[Kasterisk*CLI> ^M[Mar 25 08:47:27] NOTICE[18841]: 
> chan_dahdi.c:6833 ss_thread: Got event 2 (Ring/Answered)...
>     -- Executing [s@fax-roton:1] Wait("DAHDI/2-1", "1") in new stack
> ^MESC[Kasterisk*CLI> ^M    -- Executing [s@fax-roton:2] 
> Answer("DAHDI/2-1", "") in new stack
>     -- Executing [s@fax-roton:3] Wait("DAHDI/2-1", "1") in new stack
> ^MESC[Kasterisk*CLI> ^M    -- Executing [s@fax-roton:4] 
> Goto("DAHDI/2-1", "fax-in|s|1") in new stack
>     -- Goto (fax-in,s,1)
>     -- Executing [s@fax-in:1] Dial("DAHDI/2-1", "IAX2/iaxmodem") in 
> new stack
> ^MESC[Kasterisk*CLI> ^M    -- Called iaxmodem
> ^MESC[Kasterisk*CLI> ^M    -- Call accepted by 127.0.0.1 (format alaw)
>     -- Format for call is alaw
> ^MESC[Kasterisk*CLI> ^M    -- IAX2/iaxmodem-11705 is ringing
> ^MESC[Kasterisk*CLI> ^M    -- IAX2/iaxmodem-11705 answered DAHDI/2-1
> ^MESC[Kasterisk*CLI> ^M    -- Registered IAX2 'iaxmodem' 
> (AUTHENTICATED) at 127.0.0.1:4570
> [Mar 25 08:47:41] NOTICE[18589]: chan_iax2.c:7477 update_registry: 
> Restricting registration for peer 'iaxmodem' to 60 seconds (requested 600)
> ^MESC[Kasterisk*CLI> ^M    -- Registered IAX2 'iaxmodem' 
> (AUTHENTICATED) at 127.0.0.1:57229
> [Mar 25 08:47:41] NOTICE[18591]: chan_iax2.c:7477 update_registry: 
> Restricting registration for peer 'iaxmodem' to 60 seconds (requested 600)
> ^MESC[Kasterisk*CLI> ^M    -- Hungup 'IAX2/iaxmodem-11705'
> ^MESC[Kasterisk*CLI> ^M  == Spawn extension (fax-in, s, 1) exited 
> non-zero on 'DAHDI/2-1'
> ^MESC[Kasterisk*CLI> ^M    -- Hungup 'DAHDI/2-1'
But I see shtg strange : why are there 2 "iaxmodem" registering? Is ther
a "leg" for "faxgetty" and one for sending faxes? But if my iaxmodem
config was wrong, I would not receive faxes? Right? And I can receive
faxes from most of hte machines, except very few ones.
And for the trace, same story (same call as above in CLI):
> Mar 25 08:48:05.80: [ 3650]: <-- data [2]
> Mar 25 08:48:06.38: [ 3650]: --> [2:OK]
> Mar 25 08:48:06.38: [ 3650]: <-- [9:AT+FRH=3\r]
> Mar 25 08:48:13.22: [ 3650]: --> [10:NO CARRIER]
> Mar 25 08:48:13.22: [ 3650]: MODEM No carrier
> Mar 25 08:48:13.22: [ 3650]: RECV FAX: No sender protocol (T.30 T1 
> timeout)
> Mar 25 08:48:13.22: [ 3650]: RECV FAX: end
> Mar 25 08:48:13.22: [ 3650]: No sender protocol (T.30 T1 timeout)
> Mar 25 08:48:13.22: [ 3650]: SESSION END

My iax.conf :
> [iaxmodem]
> type=friend
> requirecalltoken=no
> secret=password
> port=4570
> host=dynamic
> context=roton
> disallow=all
> allow=alaw
/etc/inittab :
> IA00:23:respawn:/usr/bin/iaxmodem ttyIAX0
> mo00:23:respawn:/usr/sbin/faxgetty ttyIAX0
and /etc/iaxmodem/ttyIAX0
> device          /dev/ttyIAX0
> owner           uucp:uucp
> mode            660
> port            4570
> refresh         600
> server          127.0.0.1
> peername        iaxmodem
> secret          password
> codec           alaw
Any suggestion?




> > > > > Mar 22 12:10:10.17: [ 3650]: RECV send RTN (retrain negative)
> > > > > Mar 22 12:10:10.17: [ 3650]: RECV FAX (000000014): from 0xxxxxxx, 
> > > > > page 2 in 0:18, INF, 3.85 line/mm, 2-D MR, 9600 bit/s
> > > > > Mar 22 12:10:10.17: [ 3650]: <-- data [3]
> > > > > Mar 22 12:10:10.17: [ 3650]: <-- data [2]
> > > > > Mar 22 12:10:11.25: [ 3650]: --> [2:OK]
> > > > > Mar 22 12:10:11.25: [ 3650]: <-- [9:AT+FRH=3\r]
> > > > > Mar 22 12:10:11.59: [ 3650]: --> [7:CONNECT]
> > > > > Mar 22 12:10:12.75: [ 3650]: --> [2:OK]
> > > > > Mar 22 12:10:12.75: [ 3650]: RECV recv DCN (disconnect) 
> > >
> > > This is actually a normal error.  RTN means that HylaFAX rejected 
> > > the page image quality, but it doesn't require the sender to 
> > > retransmit.  The sender chose to disconnect instead of retransmitting.
> > Yes, the fax was nearly OK (just the 3 last lines with some garbage 
> > in the pixels). But the remote machine "told" the sender that ther 
> > was an error and he sent the fax again, this time without any error. 
>
> Good, that's as it should be.  The end-result is what was desired: the 
> sender retransmitted.
>
> 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*



____________________ 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*