SAIL Community Supported PBX

Had to
unload app_voicemail_imap.so
and
load app_voicemail_imap

In the modules.conf so now phone calls get to the phone. Before the system sent a bye packet before the system rang.
But the voicemail gets the prompts but fails the password.

From what I have been reading only 1 voicemail module should be loaded. But it seems voicemail_imap.so supports comedian Mail (ie) *51* although it answered on *50* and then fails the password.
If you load app_voicemail_imap.so from the modules.conf and then try and unload from the cli (which fails) and then load app_voicemail.so we get the VoiceMailMain application back.

Is there an answer to this riddle? How to get it back to being stable?

PBX release: 20.4.0
SAIL Release: 6.2.0-35
HPE Release: 6.0.0-9

May have solved it
When we have this it works
Module                         Description                              Use Count  Status      Support Level
app_voicemail.so               Comedian Mail (Voicemail System)         0          Running              core
app_voicemail_imap.so          Comedian Mail (Voicemail System) with IM 0          Running              core
app_voicemail_odbc.so          Comedian Mail (Voicemail System) with OD 0          Not Running          core
3 modules loaded
So the default is only app_voicemail_imap.so running but that does not load VoiceMailMain

module load app_voicemail.so

unloading the odbc module also works
Module                         Description                              Use Count  Status      Support Level
app_voicemail.so               Comedian Mail (Voicemail System)         0          Running              core
app_voicemail_imap.so          Comedian Mail (Voicemail System) with IM 0          Running              core

But this should be set from modules.conf. Might try a preload in the conf file to get it automatically. It also solved another issue where trunks did not answer if linked to voicemail. Just dropped the call with a bye command.

Hi
I have tried this on Jammy and Mantic and no change to the issue.
This is standard output with the modules.conf file as installed

---

pbx-ub*CLI> module show like voicemail
Module                         Description                              Use Count  Status      Support Level
app_voicemail.so               Comedian Mail (Voicemail System)         0          Not Running          core
app_voicemail_imap.so          Comedian Mail (Voicemail System) with IM 0          Running              core                           
app_voicemail_odbc.so          Comedian Mail (Voicemail System) with OD 0          Not Running          core

---

the result
[2024-03-29 09:58:05] ERROR[572]: app_voicemail_imap.c:2816 inboxcount2: Couldn't find mailbox 501 in context default
    -- AGI Script Executing Application: (VoiceMailMain) Options: (501)
[2024-03-29 09:58:06] WARNING[1147][C-00000001]: res_agi.c:3181 handle_exec: Could not find application (VoiceMailMain)
    -- <SIP/501-00000000>AGI Script sarkhpe completed, returning 0
    -- Auto fallthrough, channel 'SIP/501-00000000' status is 'UNKNOWN'
[2024-03-29 09:58:28] ERROR[572]: app_voicemail_imap.c:2816 inboxcount2: Couldn't find mailbox 501 in context default
[2024-03-29 09:58:39] ERROR[572]: app_voicemail_imap.c:2816 inboxcount2: Couldn't find mailbox 501 in context default[/size]

If we noload app_voicemail_imap.so we do not get the error messages "Couldn't find mailbox..." But still get "Could not find application (VoiceMailMain)"

In Ubuntu there is a package (deb) for VoiceMail but it is for asterisk 16.
asterisk-voicemail_16.2.1~dfsg-2ubuntu1_amd64.deb
and if we unload off app_voicemail_imap.so and use the app_voicemail.so we loose the Mailbox errors but still have no VoiceMailMain appication. See attached list

Any ideas on how to get voicemail back?
Thanks.

Solved
The DID was the same ID as used for a trunk. We need the full number for the DID so dropped the trunk and recreated with different ID. Is a error trap possible instead of a blank screen?

PBX release: 20.1.0
SAIL Release: 6.2.0-35
HPE Release: 6.0.0-9
Endpoints defined: 4
Serial Num: 936191

Hi S
Anyone have an interpretation of this

---

Wed Feb 07 07:53:21.819550 2024] [php:warn] [pid 270] [client xxx.xxx.xxx.xxx:55830] PHP Warning:  Undefined array key "smartlink" in /opt/sark/php/sarkddi/view.php on line 370, refer
er: https://mail.torresbusiness.com.au:28443/php/sarkddi/main.php                                                                                                                     
[Wed Feb 07 07:53:21.819618 2024] [php:error] [pid 270] [client xxx.xxx.xxx.xxx:55830] PHP Fatal error:  Uncaught Error: Undefined constant "trunk" in /opt/sark/php/sarkddi/view.php:31
7\nStack trace:\n#0 /opt/sark/php/sarkddi/view.php(58): sarkddi->saveNew()\n#1 /opt/sark/php/srkmain.php(58): sarkddi->showForm()\n#2 /opt/sark/php/sarkddi/main.php(3): require_once(
'...')\n#3 {main}\n  thrown in /opt/sark/php/sarkddi/view.php on line 317, referer: https://xxx.xxx.xxx.xxx:28443/php/sarkddi/main.php

---

I have 3 DID's already in the list just need to add a 4th. Thanks

Thanks S
That solved the issue.
Are you planning on drooping Debian as a base OS?

OK Edited the /opt/sark/www/sark-common/manuf.txt file and added a line for that manufacturer. So now get the "mac already exists"
Looking in the IPphone table in the DB, both extensions that use 24:9A:D8 macs are there but do not display in extensions screen. I reset the system to factory and reapplied a backup and still the same failure.
Any clues as to why the phones with those macs would not add or display?

Hi S
PBX release: 16.28.0
SAIL Release: 6.2.0-16+deb10u1
HPE Release: 6.0.0-9

If I add and extension using 249AD82AB0FA as the mac it says it is invalid, but if I add 001565.... it works. Both are Yealink phones.
Maybe a clue, it is a system where I restored the backup from ver5 and all extensions bar the 2 that start with 249AD8... came across.
Do we need to update the DB with the OUI?

Thank you for the reply.
What lead us to this was in sngrep the packets were not getting back to the PBX from ITSP. Turn off shorewall and worked perfectly. The local.lan file read "LAN=/" and on some other systems it was 0.0.0.0/0 which works.

The PBX is a node in a Proxmox server with only 1 NIC defined.

I will run the php line and see what we get back.

May have found the answer
If you run
shorewall show zones

it returns 0.0.0.0/0 and when I modify the local.lan file from "/" to "0.0.0.0/0" it now works as expected.

Hi
Hope all is well on your side of the world.
PBX release: 16.28.0
SAIL Release: 6.2.0-16+deb10u1
HPE Release: 6.0.0-9

My trunks show as unreachable whilst I have shorewall running. Turn it off and the number of messages for options and register drop from 34 to 2.

Is there an adjustment to shorewall to reduce interference on sip traffic?


local.lan
xxx.xxx.xxx.x/24

local.if1
IF1=enp6s18

ACCEPT net:$LAN $FW tcp 5060 - -   
ACCEPT net:$LAN $FW tcp 80 - - # HTTP
ACCEPT net:$LAN $FW tcp 443 - - # HTTPS
ACCEPT net:$LAN $FW tcp 22 - - # SSH
ACCEPT net:$LAN $FW udp 123 - - # NTP
ACCEPT net:$LAN $FW tcp 389 - - # LDAP
ACCEPT net:$LAN $FW udp 389 - - # LDAP
ACCEPT net:$LAN $FW udp 4569 - - # IAX2
ACCEPT net:$LAN $FW udp 5060 - - 4/min:5 # SIP
ACCEPT net:$LAN $FW udp 10000:20000 - -  # RTP
This is the rules file
ACCEPT net:$LAN $FW tcp 5060 - -   
ACCEPT net:$LAN $FW tcp 80 - - # HTTP
ACCEPT net:$LAN $FW tcp 443 - - # HTTPS
ACCEPT net:$LAN $FW tcp 22 - - # SSH
ACCEPT net:$LAN $FW udp 123 - - # NTP
ACCEPT net:$LAN $FW tcp 389 - - # LDAP
ACCEPT net:$LAN $FW udp 389 - - # LDAP
ACCEPT net:$LAN $FW udp 4569 - - # IAX2
ACCEPT net:$LAN $FW udp 5060 - - 4/min:5 # SIP
ACCEPT net:$LAN $FW udp 10000:20000 - -  # RTP

Hi S
Thank you for the reply. The strange thing is, if the call is put onhold or transferred by the extension the MOH plays. Just not as a replacement of ring tone.

So is the solution to create a moh-default folder?
Thanks

I am a little confused to the outcome of below
musiconhold.conf
[default]
mode=files
directory=moh
.....
#include sark_moh.conf

but sark_moh.conf says
[default]
mode=files
directory=moh-default
moh-default is not in the asterisk directory set and "moh show files" does not look at it. The sark_moh.conf is overwritten by the workbench so it can not be edited.

Do we need moh-default which is in /etc/asterisk and not /usr/share/asterisk?

Thanks