QUESTION ring group goes straight to voicemail

[bobkoure]

I've been trying to figure this one out but no joy.

I have a ring group with a single extension (removed all the others)
Ring Strategy: ringall
Ring Time: 100 (increased this to see if it would fix things)
Announcement: none (removed the one I had)

Destination if no answer: voicemail (above ext. unavail)

the extension in the ring group
Ring Time: 40


So now, without getting inbound routes involved (no need)
- if I dial the extension, it works as expected, rings 40-ish seconds, goes to voicemail unavail
- if I dial the ring group, it goes immediately to voicemail

I've restarted asterisk (no joy), deleted and recreated the ring group (no joy), restarted linux (also no joy).

I've probably missed an obvious setting. Anyone care to tell me what that might be?
Thanks!

Oh - IncrediblePBX, (most recent as of a couple of weeks ago) asterisk 11.15.1, Centos7/64 - on a dual core Atom box. Extensions are on Obihai ATAs (not sure how that would matter for this)

 

[kenn10]

The only reason calling the ring-group would send the call directly to voice mail is if: (1) The extension is busy or unavailable, or (2) not entered into the ring group correctly. If you are calling the ring group from the extension in the ring group, then the extension will be unavailable and the call will go to voice mail.

Logging in to the CLI from the Linux prompt with asterisk -rvvvvvvvv and watching the logging as the call is made should reveal more clues.

 

[bobkoure]

I'd done that, but it appears to go straight to voicemail there as well
ring group being tested: 113
contains single extension: 102
ring group destination no answer: voice mail for ext 102
calling from extension 101 (not in ring group 113)

My first guess was that the extension was not available (even though I can call it directly).
Is that what this is saying? (entire trace attached as file)

    -- Goto (macro-vm,vmx,1)
    -- Executing [vmx@macro-vm:1] Set("SIP/101-00000021", "MEXTEN=102") in new stack
    -- Executing [vmx@macro-vm:2] Set("SIP/101-00000021", "MMODE=NOANSWER") in new stack
    -- Executing [vmx@macro-vm:3] Set("SIP/101-00000021", "RETVM=") in new stack
    -- Executing [vmx@macro-vm:4] Set("SIP/101-00000021", "MODE=unavail") in new stack
    -- Executing [vmx@macro-vm:5] GotoIf("SIP/101-00000021", "1?chknomsg") in new stack
    -- Goto (macro-vm,vmx,7)
    -- Executing [vmx@macro-vm:7] GotoIf("SIP/101-00000021", "0?s-NOANSWER,1") in new stack
    -- Executing [vmx@macro-vm:8] GotoIf("SIP/101-00000021", "1?notdirect") in new stack
    -- Goto (macro-vm,vmx,10)
    -- Executing [vmx@macro-vm:10] NoOp("SIP/101-00000021", "Checking if ext 102 is enabled: ") in new stack
    -- Executing [vmx@macro-vm:11] GotoIf("SIP/101-00000021", "1?s-NOANSWER,1") in new stack
    -- Goto (macro-vm,s-NOANSWER,1)
    -- Executing [s-NOANSWER@macro-vm:1] Macro("SIP/101-00000021", "get-vmcontext,102") in new stack
    -- Executing [s@macro-get-vmcontext:1] Set("SIP/101-00000021", "VMCONTEXT=default") in new stack
    -- Executing [s@macro-get-vmcontext:2] GotoIf("SIP/101-00000021", "0?200:300") in new stack
    -- Goto (macro-get-vmcontext,s,300)
    -- Executing [s@macro-get-vmcontext:300] NoOp("SIP/101-00000021", "") in new stack
    -- Executing [s-NOANSWER@macro-vm:2] VoiceMail("SIP/101-00000021", "102@default,u") in new stack
    -- <SIP/101-00000021> Playing '/var/spool/asterisk/voicemail/default/102/unavail.slin' (language 'en')
    -- <SIP/101-00000021> Playing 'vm-intro.gsm' (language 'en')

I created a new ring group. I get the same behavior when I use different extensions, including an IAX softphone and a PSTN number as extension (all of which I can call directly).

I think this following means extensions 101 and 102 are registered (trunk details obfuscated).
But I don't know if registered means 'available'.

pbx*CLI> sip show peers
Name/username            Host                                    Dyn Forcerport Comedia    ACL Port    Status      Description
101/101                  172.16.51.7                              D  No        No          A  5060    OK (4 ms)
102/102                  172.16.51.8                              D  No        No          A  5060    OK (3 ms)
103                      (Unspecified)                            D  No        No          A  0        UNKNOWN
OBITRUNK1/OBITRUNK1      172.16.51.7                              D  No        No            5061    OK (4 ms)
localphone-outbound/[NNN] 94.75.247.45                                No        No            5060    Unmonitored
vitel-outbound/[AAAA]    64.2.142.9                                  Yes        Yes            5060    Unmonitored
voipms/[NNNN_AAAA]        107.6.67.237                                Yes        Yes            5060    OK (16 ms)
7 sip peers [Monitored: 4 online, 1 offline Unmonitored: 2 online, 0 offline]

Thanks for the help!

 

[islandtech]

What model Obihai's?

 

[bobkoure]

I get this issue with ring groups that contain only an IAX softphone or PSTN-as-extension or Yealink T28.
That said...
I've two, an OBI110 and an OBI100. Target extension is on the 100. The 110 is also connected to my single land line so it comes in as a trunk.
I had these working (AFAICT perfectly) with my previous IncrediblePi.
I recently got a dual core atom box, put CentOS7/64 on it, then IncrediblePBX. Pretty much just changed the address the OBIs were registering to from the RasPi to the CenttOS, recreated the extensions using details from IncrediblePi and both extensions worked fine. Also the trunk.
Calls to/from both extensions (and trunk) work fine.

 

[islandtech]

I can't duplicate the trouble in my lab using an Obhi110.
tried every ring strategy
the only time it would go immediately to vm is when the ext is in DND

 

[bobkoure]

Thanks for checking!
Unfortunately, I have no problem duplicating the issue.
Tried creating new ring group(s). Used non-Obihai extension. Same issue with ring strategies ringall, hunt, hunt+mem.

 

[islandtech]

went back and looked at your attached trace and noticed this

Executing [s@macro-dial:3] AGI("SIP/101-00000021", "dialparties.agi") in new stack
-- Launched AGI Script /var/lib/asterisk/agi-bin/dialparties.agi
dialparties.agi: Failed to execute '/var/lib/asterisk/agi-bin/dialparties.agi': Permission denied
-- Executing [s@macro-dial:4] NoOp("SIP/101-00000021", "Returned from dialparties with no extensions to call and DIALSTATUS: ") in new stack

check permissions from ssh or the pbx console

 ls -l /var/lib/asterisk/agi-bin/dialparties.agi

and the response should be similar to this

-rwxr-xr-- 1 asterisk asterisk 31865 Feb 18 13:11 /var/lib/asterisk/agi-bin/dialparties.agi

main part is the asterisk asterisk
if not then change it

 

[bobkoure]

Got response

-rwxr-xr--. 1 asterisk asterisk 31865 Feb 18 08:34 /var/lib/asterisk/agi-bin/dialparties.agi

Looks... pretty similar.

I looked at the file (PHP?) and noticed there's a
require_once "phpagi.php";
So I tried
ls -l phpagi.php in the same directory

-rwxrw-r--. 1 asterisk asterisk 65350 Feb  2 13:29 phpagi.php

So maybe not that, either.

If it is PHP, unless the include path is /var/lib/asterisk/agi-bin - which it probably is, that require would only work with a '.' before the filename to include.

Any ideas?
Thanks!

 

[bobkoure]

From the Linux command line, logged in as 'root', I ran
amportal chown
amportal restart

Calling a ring group no longer goes directly to voicemail, instead I get the ring group announcement and then silence.
And If I change the announcement to none, things actually work (!)
I'd put that there because, with my prev IncrediblePi, if an outside call came in, got routed to my pstn mobile phone, there were audio issues (silence).
I just tested that (took a bit to find cell signal where I am) and same issue (silence without announcement).
[edit]
and if I change the announcement to one that "comes as installed" rather then the one I uploaded, it all works.
I'll go chase permissions on the sound files.
Are uploaded files also stored in /var/lib/asterisk/sounds?
[edit]
Thanks!

 

[bobkoure]

I've got it working with the recording I was using before.
The way I know to make a recording appear in the ring group announcement selection list is to upload it via admin / system recordings. I deleted the recording, then uploaded \\pbx\var\lib\asterisk\sounds\en\pls-wait-connect-call.gsm (yes, uploaded a file from the pbx that I have access to via samba). ls -l of /var/lib/asterisk/sounds/custom and files all look the same, with the single difference of -rwxrwxr-x. vs-rwxrwxr-x - note the trailing dot, which I think, means 'alternate access' (?)

I'll try progressinband when I get a chance (I'm supposed to be working but an unsolved issue is like a loose tooth). It did not work with the RasPi IncrediblePi. Maybe it will on CentOS.

Thanks very much to everyone who helped!

 

[bobkoure]

Quick to try and test.
progressinband=yes and no ring group announcement = no audio when call completed.
So it doesn't work for my CentOS, either.
No problem - back to the "please wait" lady