Shared line between IP Phones and FXS Port with SCCP setup

Unanswered Question
Oct 25th, 2008
User Badges:
  • Silver, 250 points or more

Hi,


We have a shared extension (2000) across 5 IP Phones and an FXS port

on a gateway for an overhead bell. The FXS port is configured as SCCP

with call manager so it can be a shared line. It works fine, and when

2000 is called the IP Phones ring as well as the bell but when a IP

Phone answers, the FXS port stays offhook until the IP Phone caller

hangs up the call. This means other calls into this extension while

they have a phone call, the overhead bell doesn't ring. We put a

regular POTS telephone on the FXS port and when 2000 is called the IP

Phones ring as well as the regular POTS phone and when IP Phone

answers, the POTS phone stops ringing but the port on the gateway is

offhook still. When we lift the handset on the POTS phone it is just

dead air/silence and no one can hear.


When the port is configured as MGCP, when the IP Phone picks up the

call the FXS port goes idle and a 2nd/3rd call into the line ring the

FXS port fine but with MGCP after awhile the port gets hung up due to

a bug but when we try newer IOS versions where that is supposedly

fixed the FAX calls stop working due to other bugs.


Just wondering if SCCP should act as same as MGCP with this shared

line scenario, and why with SCCP the port stays offhook after another

phone picks up the call to the shared line.


The gateway is a 2800 with a VIC2-FXS and we've tried IOS 12.4(11)T

through 12.4(20)T with same problem on CCM 5.1(3) and CCM 6.1(2).

Again, with FXS port configured as MGCP the port goes idle after

another phone picks up the call on the shared lines.


This is what the FXS port state is after another IP Phone answers the

call on the shared line, with a regular POTS home phone or the

overhead bell plugged into it with SCCP setup.


Gateway#show voice call summ

PORT CODEC VAD VTSP STATE VPM STATE

============== ========= === ==================== ======================

0/1/0 g711ulaw n S_SETUP_REQ_PROC FXSLS_WAIT_OFFHOOK

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 5 (3 ratings)
Loading.
amuslimcsr Tue, 10/05/2010 - 17:49
User Badges:

Hi,


Have you got any solution to this issue yet? We're having a similar issue with sharing the phone (SCCP) DN with VG224 FXS (MGCP) connected to an overhead bell. The MGCP port gets hung up after a while and need to be reset periodically. Can you tell me the bug number for this?

VG224 is running vg224-i6k9s-mz.124-25.bin, CCM 7.1.3.10000-11


thanks,

Arief

Daniel Bosch Thu, 01/22/2015 - 21:07
User Badges:

This is a very old post, but don't see an answer provided to handle the shared line across multiple IP phones and an analog port used to ring an overhead bell.  I recently encountered an identical scenario and used a hunt pilot and a broadcast line group to make it work.  Using this method, it doesn't matter if the FXS port is SCCP or an MGCP endpoint.  In my case, I was able to configure unique DNs on each IP phone, but a shared line will work too - as long as a unique DN is configured for the FXS port.

Here were my steps.

1. Configure a unique DN (or shared DN) on each IP phone - but it must NOT be the number used to reach the department.

2. Configure a unique DN on the FXS port (SCCP or MGCP)

3. Configure a Line Group containing all the DNs (IP Phones and FXS) - Distribution Algorithm = Broadcast

4. Configure a Hunt List containing the Line Group from step 3 (don't forget to check the enable box.)

5. Configure a Hunt Pilot with the number used to reach the department.  Select the Hunt List from step 4.

 

Works like a charm - every time!

 

If the department also requires voice mail, you'll also need to configure the forwarding parameters on the Hunt Pilot accordingly.  Then you configure the Unity Connection subscriber mailbox using the extension matching the Hunt Pilot number.  After that you configure the mailbox with Alternate Extensions and MWI that match the DNs associated with the IP phones.  If someone leaves a message, MWI will be sent to all the IP phones and Unity Connection will recognize that any of the DNs on the IP Phones are associated with that mailbox.

 

There are probably other viable solutions too.  For example, you could probably configure the departmental number as a shared line on the IP phones, and then use Single Number Reach (Remote Destination Profile) with the analog DN configured as the remote destination.  This would eliminate the requirement to add Alternate Extensions and MWI on the mailbox.  The con would be that the Remote Destination Profile method might consume additional license - depending on how you setup the associated user.

Udit Mehrotra Tue, 10/05/2010 - 19:16
User Badges:
  • Cisco Employee,

I believe shared lines between IP Phones and MGCP controlled FXS are not supported.

As far as SCCP controlled FXS port and IP Phones sharing DNs, this setup is supported.


What have you set Max Call/Busy trigger on CM DN page for SCCP controlled FXS port?

amuslimcsr Tue, 10/05/2010 - 19:46
User Badges:

The max call/busy trigger is set to 2/1 for the mgcp fxs port and 4/2 for the sccp phones. Can you give me the link that says the set up is not supported?

Steven Holl Wed, 10/06/2010 - 07:52
User Badges:
  • Cisco Employee,

The previous poster is correct.  MGCP developers won't support a shared line between an MGCP controlled FXS and a SCCP controlled extension (regardless whether the latter be an IP phone or SCCP FXS).


There isn't official documentation stating this, but I've had first-hand conversations with the MGCP DE Manager regarding this design.  It commonly causes 'ghost ringing' scenarios,and I've encountered it several times along my adventures.


Here is the original issue I observed:


1. Call rings shared DN.
2. IP phone picks call up.
3. While call is active on DN, another call rings the same shared DN.
4. Caller hangs up first call to answer second call with IP phone.
*Here, the analog phone will keep ringing, until you manually go
OFF/ONHOOK with the analog handset.

If I disconnect the *first* call and wait a few seconds for the
DLCX to be sent before I pickup the second line, then the POTS line will
stop ringing for the *second* call.  However, if I don't wait any time
between going onhook with the first call and picking up the second, the
POTS line will continue to ring after the IP phone is onhook with the
second call.

I've determined the phone continues to ring because it never gets a
DLCX.  Actually, sometimes, the router will send the DLCX to the POTS
line like it should, but it is maybe 1 in 4 attempts.  The issue where
the second POTS call ends ringing early (due to the IP phone
disconnecting the first call) is due to a DLCX message being sent to the
port (when it shouldn't be sent one).


The MGCP team pointed out that this was not a supported scenario due to major issues with the way it works internally.  They pointed me to two bugs which were closed for the same reason of this design not being supported.


CSCin75549
Externally found moderate defect: Closed (C) FXS port not released after transfer from a shared line


CSCeg76321
Externally found moderate defect: Closed (C) Callmanager MGCP CRCX not sent to FXS port shared line with IP Phone




Now, what is supported is a shared line between SCCP endpoints.  If you configure the FXS port as SCCP controlled, you will be good to go.  Just run code that has shared line support for SCCP (run 15.0(1)M or higher to prevent hitting CSCsw88646).


Configure:


stcapp supplementary-service
 port 
  hold-resume

And then configure the max-call on the extension to 2 or higher.  If you *don't* want hold-resume configured, max-call should be configured for 1.

Actions

This Discussion