I feel kind of stupid, but I've solved my problem and discovered something interesting.

App_rpt accepts connections directly from Zap and SIP channels (without the use of Local) just fine when it's invoked with the R option, as

Rpt(${NODE},R<announce-string>)

I discovered this by accidentally dialing the test extension I'd set up for checking direct IAX from my Zap phone. Wow, it worked! I got to thinking about what I had done differently, and it occurred to me that I'd probably gotten the warning I mentioned when I was still invoking app_rpt as

Rpt(${NODE},X)

So, I tried doing that again, and sure enough - no Zap or SIP connections. Anyway, I wanted to clear the air on here and not leave it appearing as though Zap and SIP connections wouldn't work for reverse-autopatch. They do, if you do it right.

Josh

Josh Freeman wrote:
OK. Well, darn. =)

Originally, I had tried to connect directly from the Zap channel, and it failed with the message
Warning: We only accept links via IAX2, Echolink, TheLinkBox or Local!!
I guess that's what got me thinking Local channels would be the way to bridge over from Zap or SIP.

Thanks again for your help!

Josh

On 10/10/2011 8:45 PM, Jim Duuuude wrote:
dont use it at all... app_rpt cant deal with it.

Date: Mon, 10 Oct 2011 19:53:14 -0500
From: cpe.jfreeman@gmail.com
To: telesistant@hotmail.com
Subject: Re: [App_rpt-users] Reverse autopatch immediately hanging up

OK, I tried that... it looks like the reverse autopatch does work correctly when dialed into via pure IAX. Unfortunately, only being able to use native IAX clients doesn't meet the requirements for what I'm trying to do. I'll study up some more on how Local channels work, maybe I'm not using that mechanism correctly.

Thanks,
Josh

On 10/10/2011 6:21 PM, Jim Duuuude wrote:
okay, try calling it directly from your IAX client, rather then through a Local channel

Date: Mon, 10 Oct 2011 18:05:16 -0500
From: cpe.jfreeman@gmail.com
To: APP_RPT-USERS@OHNOSEC.ORG
Subject: Re: [App_rpt-users] Reverse autopatch immediately hanging up

Jim,

That would be correct, if I had left the [radio] context as it was in the bundled Allstar dialplan, where this line

exten=_07XX,1,Goto(parkedcalls|${EXTEN:1}|1)

makes it necessary to dial the leading "0." I created a different radio context, and simply added

include => parkedcalls

so that the pickup code *is* actually just '701'.

Because I don't understand everything about this system, though, I tried reverting that part of my dialplan back to the Allstar configuration, just to see if that made a difference. It didn't - the autopatch still gets hung up within a second or two of connection. Dialing through the Local channel is the thing I'm doing that's significantly different than anything in the Allstar dialplan, so I'm guessing that's where the problem is - but I'm not smart enough to find it yet.

Thanks,
Josh

On 10/10/2011 1:33 PM, Jim Duuuude wrote:
Thats because youre missing a 'digits/0' between the incall and the PARKED, such as:

Rpt(${EXTEN:1}|Rrpt/node:NODE:rpt/in-call:digits/0:PARKED|120)

The code to pick that up should be '0701' *not* 701

JIM WB6NIL

Date: Mon, 10 Oct 2011 11:06:52 -0500
From: cpe.jfreeman@gmail.com
To: APP_RPT-USERS@OHNOSEC.ORG
Subject: [App_rpt-users] Reverse autopatch immediately hanging up

Hello,

I'm trying to set up a system that has a reverse autopatch as one of its features. The problem I'm having is that when a radio user dials the code to pick up an incoming call, the call gets disconnected almost immediately upon connecting. I'm certain the problem is that I've not set things up properly, but I'm still inexperienced enough that I could really use some help troubleshooting.

My Asterisk system is running from a recent ACID image (downloaded maybe three weeks ago). The machine also has a TDM400P card with an FXS module connected to an analog telephone I'm using to originate incoming test calls to the repeater system.

The relevant parts of my dialplan are as follows:

In context phones_local:

exten => 272727,1,Dial(Local/CQCQCQ@repeater)
(Apparently, I can't directly connect to app_rpt from a Zap channel, so I dial through a Local channel instead.)

Then, in context repeater:

exten => CQCQCQ,1,Rpt(5674,Rrpt/in-call:PARKED,120)
(This is an isolated system, so the node number is simply my name on a telephone keypad.)

What happens is this: I dial 272727 from the analog telephone, and I hear the announcement over the radio just fine. I send *6701 from the radio, and hear "Connecting" from the repeater along with echoing audio from the analog phone (because it's right there next to the radio). Then, I immediately hear "Call terminated" and the autopatch drops. The analog phone doesn't get hung up, though.

Here's what I see on the Asterisk console. For this trace, I was actually dialing in from an IAX softphone rather than the Zap channel, but the behavior is exactly the same.
-- Executing [272727@phones_local:1] Dial("IAX2/josh-12630", "Local/CQCQCQ@repeater") in new stack
   -- Called CQCQCQ@repeater
   -- Executing [CQCQCQ@repeater:1] Rpt("Local/CQCQCQ@repeater-781b,2", "5674|Rrpt/in-call:PARKED|120") in new stack
   -- Return Context: (repeater,CQCQCQ,2) ID: 0
   -- Warning: Return Context Invalid, call will return to default|s
   -- Music class default requested but no musiconhold loaded.
 == Parked Local/CQCQCQ@repeater-781b,2 on 701@parkedcalls. Will timeout back to extension [repeater] CQCQCQ, 2 in 45 seconds
   -- Added extension '701' priority 1 to parkedcalls
   -- Music class default requested but no musiconhold loaded.
   -- Music class default requested but no musiconhold loaded.
   -- Local/CQCQCQ@repeater-781b,1 answered IAX2/josh-12630
   -- Call Parking Called, lot: 701, timeout: 0, context: (null)
   -- Music class default requested but no musiconhold loaded.
   -- <Zap/pseudo-1800116975> Playing 'rpt/in-call' (language 'en')
   -- <Zap/pseudo-1800116975> Playing 'digits/7' (language 'en')
   -- <Zap/pseudo-1800116975> Playing 'digits/0' (language 'en')
   -- <Zap/pseudo-1800116975> Playing 'digits/1' (language 'en')
   -- Hungup 'Zap/pseudo-1800116975'
   -- Hungup 'Zap/pseudo-1018949410'
[Oct  9 19:44:18] NOTICE[2594]: chan_usbradio.c:3010 usbradio_read: Got DTMF char * duration 340 ms
[Oct  9 19:44:19] NOTICE[2594]: chan_usbradio.c:3010 usbradio_read: Got DTMF char 6 duration 235 ms
[Oct  9 19:44:20] NOTICE[2594]: chan_usbradio.c:3010 usbradio_read: Got DTMF char 7 duration 255 ms
[Oct  9 19:44:20] NOTICE[2594]: chan_usbradio.c:3010 usbradio_read: Got DTMF char 0 duration 212 ms
   -- Hungup 'Zap/pseudo-1705936882'
[Oct  9 19:44:21] NOTICE[2594]: chan_usbradio.c:3010 usbradio_read: Got DTMF char 1 duration 235 ms
   -- Executing [701@phones_radio:1] ParkedCall("Zap/pseudo-523589909", "701") in new stack
   -- Channel Zap/pseudo-523589909 connected to parked call 701
 == Spawn extension (phones_radio, 701, 1) exited non-zero on 'Local/CQCQCQ@repeater-781b,1<ZOMBIE>'
   -- Hungup 'Zap/pseudo-1742992459'
   -- Hungup 'Zap/pseudo-1417902655'
   -- <Zap/pseudo-1427730098> Playing 'rpt/callproceeding' (language 'en')
   -- Hungup 'Zap/pseudo-1427730098'
   -- <Zap/pseudo-1938338821> Playing 'rpt/callterminated' (language 'en')
   -- Hungup 'Zap/pseudo-1938338821'
   -- Hungup 'Zap/pseudo-1525511763'

I see some things about invalid return context and zombie channels, but I'm not sure what that means. Any idea what it is I'm missing?

Thanks,
Josh


_______________________________________________ App_rpt-users mailing list App_rpt-users@ohnosec.org http://ohnosec.org/cgi-bin/mailman/listinfo/app_rpt-users


_______________________________________________ App_rpt-users mailing list App_rpt-users@ohnosec.org http://ohnosec.org/cgi-bin/mailman/listinfo/app_rpt-users