PPP Virtual template and Frame-relay

To configure a virtual template for MLP LFI on a frame-relay subinterface, you have to remove the DLCI configured under that subinterface first before reapplying it back with the ppp keyword.

int s0/0

no frame-relay interface-dlci 30

 frame-relay interface-dlci 30 ppp virtual-template 1

Apply any LLQ, IP settings, ppp configs under the virtual template interface.

Alias huntgroups in SRST

Have phones 2001,2002,2003 in SRST mode in a router. Calls coming into the system from PSTn should hit 2001 first and then on 2003.

 alias 1 6175222001 to 2001 preference 1 cfw 2003 timeou 4 huntstop
 alias 2 6175222002 to 2001 preference 1 cfw 2003 timeout 4 huntstop
 alias 3 6175222003 to 2001 preference 1 cfw 2003 timeout 4 huntstop

Dialplan pattern doesnt have any effect in the called number when alias is configured. So you have to configure full 10 digits in the alias command. Alias

Call Park

When you park a call, Callmanager will search the available Park numbers configured on the server that is processing the calls and will also look up the partitions that are available in the CSS of the phone and will pick the first available part number that is defined in a partition that is reachable by the phone.

When you go to another phone to retrieve the parked call, Callmanager will search for the parked call by using the CSS of the device.

This can create an issue if you create same park numbers in different partitions in different servers. It is recommended to give different park number ranges for different servers.

Call Pickup group

Make sure the call pickup group numbers does not overlap with the exiting dial plan

Call Pickup features include Pickup, GPickup, and OPickup:

•Pickup allows you to answer a call that is ringing on another phone within your “group” (a collection of extensions that your system administrator defines).

•GPickup allows you to answer a call ringing on a phone in another group.

•OPickup allows you to answer a call ringing on a phone in another group that is associated with your group

CUE Database compacting

show sysdb /sw/info/filesys  — will show how much disk space is being used for voicemail. It will also show disk usage in percentage

To compact the database.

# offline

database compact

How to debug DTMF relay?

For H323

debug h245 asn1

For SIP

debug voip rtp session named-event

Sample output

H323

Aug 21 21:07:37.902: h245_decode_one_pdu: more_pdus = 0, bytesLeftToDecode = 7
Aug 21 21:07:37.902: H245 MSC INCOMING ENCODE BUFFER::= 6D810447200063
Aug 21 21:07:37.902:
Aug 21 21:07:37.902: H245 MSC INCOMING PDU ::=

value MultimediaSystemControlMessage ::= indication : userInput : signal :
{
signalType “9″
duration 100
}

SIP

Aug 21 21:07:37.902:          Pt:101    Evt:9       Pkt:04 00 00  <Snd>>>
Aug 21 21:07:37.902:          Pt:101    Evt:9       Pkt:04 00 00  <Snd>>>
Aug 21 21:07:37.902:          Pt:101    Evt:9       Pkt:04 00 00  <Snd>>>
Aug 21 21:07:37.902:          Pt:101    Evt:9       Pkt:04 01 90  <Snd>>>
Aug 21 21:07:37.902:          Pt:101    Evt:9       Pkt:84 03 20  <Snd>>>
Aug 21 21:07:37.902:          Pt:101    Evt:9       Pkt:84 03 20  <Snd>>>
Aug 21 21:07:37.902:          Pt:101    Evt:9       Pkt:84 03 20  <Snd>>>

Configuring PPP multilink and fragmentation over Frame-relay

* Remove ip address of serial subinterface interface

* Remove DLCI of subinterface.

* Its recommended to do this configuration on the remote site first.

int s0/0.9

no ip address

frame-relay interface-dlci 900 ppp virtual-template 1

int virtual-template 1

ip add 10.1.1.1 255.255.255.0

ppp multilink

ppp multilink interleave

ppp multilink fragment-delay 10 — milliseconds.

service-policy output VATS

Frame relay voice adaptive traffic shaping

** You will need to disable frame relay traffic shaping at the main interface level for this to work **

Configure LLQ for voice 

policy-map BR2-HQ
class RTP
priority percent 33
class Sig
bandwidth percent 2
class class-default
fair-queue
Configure Voice activated FRTS:

policy-map VATS  ———–  Line rate is 768kbps, CIR is 384kbps. Use 95 % of CIR for shape average and shape adaptive.
class class-default
shape average 729600 3648 0 
shape adaptive 364800
shape fr-voice-adapt deactivation 50
service-policy BR2-HQ
Configure FRTS class-map and apply VATS policy to the class-map

map-class frame-relay BR2-HQ
frame-relay fragment 480  ——— configures end-to-end FRF.12 fragmentation based on CIR of 384kbps
service-policy output VATS

interface Serial0/0
no ip address
encapsulation frame-relay
no dce-terminal-timing-enable
frame-relay fragmentation voice-adaptive deactivation 50
!
interface Serial0/0.201 point-to-point
ip address 10.201.2.2 255.255.255.0
frame-relay interface-dlci 201
class BR2-HQ

AAR template

1. Check the dial plan to find out if the calls between the sites are long distance or international.

2. If the calls are long distance, then you need to create only 1 AAR group with the prefix of 91. If the calls are international, then you would need to create a AAR group for each site.

3. Check the dial plan to see if the calls to the other site are using the local gateway only. If yes, then you can reuse the route pattern. Also you can reuse one of the existing CSS that has access to that route pattern as the AAR CSS.  If no, then create a route pattern to specifically match the AAR calls and add it to the long distance or international partition for that site.

How to prepend digits to caller id in H323?

voice translation-rule 1
rule 1 /\(^[2-9]………$\)/ /91\1/ — for national calls
rule 2 /\(.*\)/ /9011\1/ — for international calls

!
!
voice translation-profile ANI
translate calling 1

IPIPGW Gatekeeper lookup

When using IPIPGW, you need to define a zone prefix to route calls. invia zones does not use the default technology.

CCM Server name

The CCM server name that you give under System > Server should correspond to the actual name of the server used during install. You cannot change this later.

MoH Scenarios

A. Unicast G711 everywhere

1. Create region for MoH. G711 with all other regions.

2. Do not enable multicast in audio source, server, MRG

3. Enable only G711 in service parameter (Default)

B. G711 multicast everywhere

1. Create region for MoH. G711 with all other regions.

2. Enable multicast in audio source, server, MRG. Set hop count to 5.

3. Enable only G711 in service parameter (Default)
4. Enable multicast in infrastructure (igmp snooping in switches, pim sparse-dense in routers)

C. G711 multicast everywhere, remote site uses MoH from local SRST flash

1. Create region for MoH. G711 with all other regions.

2. Enable multicast in audio source, server, MRG. Set hop count to 1.

3. Enable only G711 in service parameter (Default)
4. Enable multicast in HQ and Remote LAN infrastructure (igmp snooping in switches, pim sparse-dense in routers), Disable multicast in WAN infrastructure.

5. Enable SRST MOH feature from flash.

6. Enable loopback interface in SRST router for PSTN users to hear MoH.

D. G711 multicast at HQ, G711 unicast at Remote

1. Create region for MoH. G711 with all other regions.

2. Enable multicast in audio source, server, MRG. Set hop count to 1.

3. Enable only G711 in service parameter (Default)
4. Enable multicast in HQ and Remote LAN infrastructure (igmp snooping in switches, pim sparse-dense in routers), Disable multicast in WAN infrastructure.

5. Create two MRG, one with multicast enabled for HQ and one with no mulitcast for BR1.

E. G711 unicast at HQ site, G729 unicast at remote

1. Enable Service parameter for G711 and G729

2. Use the HQ device pool for MOH server (assuming HQ device pool uses g729 for remote)

3. Do not enable multicast

F. G711 multicast at HQ, g729 multicast at remote

1. Enable Service parameter for G711 and G729

2. Use the HQ device pool for MOH server (assuming HQ device pool uses g729 for remote)

3. Enable multicast in audio source, server, MRG

4. Enable multicast in infrastructure

G. g729 unicast at remote using transcode

1. Enable only G711 in service parameter

2. Put MoH server in HQ device pool (assuming HQ uses G729 for remote)

3. Add a transcoder to MRGL of MoH server

I. HQ uses G711 multicast, remote uses G729 unicast, all PSTN uses hear tone on hold

1. Use HQ device pool for MoH server

2. Enable multicast on Audio source and server.

3. Create 2 MRG, one with multicast, one without multicast

4. Enable g711 and g729 in the service parameter

5. Create a third MRG and MRGL with no MOH and apply that directly to gateways

MoH server using Trancoder

If the remote device does not support the codec of the MOH stream, then MOH server can trigger a transcoder. You would need to configure a MRGL with transcoder resource in the MoH servers device pool. In this case, CCM does not use the trancoder of the remote device. It will only use the transcoder of MOH server.

Unity TRAP

TRAP uses the Default Outdial Restriction table.

Meetme conference with announcing callers name facility

a. setup meetme conference, conf bridge as necessary. Let meetme number be 1900

b. setup a route point for calls from external and internal callers as the dial-in bridge number. Let this be  1800. Cfwdall this CTI route point to voice mail.

c. Setup a Call handler meetme in Unity. Under transfer settings, tell it to ring subscriber at extn 1900. Use Supervise transfer and Check Ask Caller’s name option.

d. Create a call routing rule under forwarded calls in Unity, and set teh Forwarding station to be 1800 (CTIRP). Attempt a transfer to callhandler meetme that you just created in step c.

Unity will ask the caller his name and transfer him to the conference bridge. Conference bridge members who already have joined the conference will hear, “Call from <callername>”

Allow caller input – Unity

Under caller input page, lock a key (0 through 9, * #) to a particular action if you want Unity to perform that action as soon as you press that key. The “Allow callers to dial an extension durin greeting” has no effect if the keys are locked. To allow callers to dial an extension during the greeting, dont lock the key to a particular action

 If you have any of your keys set to Ignore key, the greeting that is played is usually System greeting “I do not recognize that as a valid selection”. This is the Error greeting. Expose the Error greeting using Advanced Settings tool from Tools Depot and you can re-record the Error greeting to a message of your choice.

Automatic Available vs Automatic Work

Automatic Work setting is at the CSQ level (Or at Agent Based Routing settings level) (Default Disabled).

Automatic Available setting is available for each agent. (Default Enabled)

Default settings make the agent go to Ready status immediately.

Automatic Work Settings override Automatic available setting set at agent level, when Automatic work is enabled.  No matter what the Automatic available setting is, if Automatic Work is set to enabled, agent goes into Work state after terminating the call. The length of the time he/she will be in Work state is determined by the Wrapup time period. (Upto 7200 seconds)

If Automatic Work is disabled, and Automatic available enabled under that agent, the agent is pushed to Ready status after the call

If Automatic Work is disabled, and Automatic available disabled under the agent,  the agent is pushed to Not-ready status after the call.

Agent based routing

a. Use GetUser step to create a User object from a userid or extension. Store the result in a User variable called studentagent.

b. Use SelectResourcestep and change Routing target type to Resource. Specify Resource Target as studentagent. Set the timeout to a value < Call Forward No answer timeout set for the agent’s phone in Callmanager. One ring is 4 seconds.

c. With agent based routing, you route the call to a specific agent. Say caller calls into IPCC and selects option 1, the call is routed to agent1. Caller selects option 2, call is routed to agent 2 and so on and so forth.

** No matter how many agents are available (Ready status), call will be routed only to the particular agent specified in the script.

** With agent based routing, if the agent doesnt pickup the call, the agent is set to not-ready state and even if he logs back into the queue, call is not routed back to the  specified agent. So there is no point in putting a queueloop inside the Failed step.

** Agent based routing has separate Wrapup time and Automatic Work settings compared to  the CSQ. No matter what you set at the CSQ level, it doesnt affect the agents behaviour after he terminates the call. To force the agent to work state when doing Agent based routing, go under RmCm->Agent Based Routing settings -> set the Automatic Work to enabled and Wrapup time to X number of seconds.

**

How to enable media cut through for H323 clients in CCM?

Set the Service Parameter, H225 Device Connect Timer to either 0 or 1.

Default is 0 and media cut through is enabled. To disable set it to 2.

How to setup CCM to accept calls from H323 gateways that are not defined locally?

Change the service parameter “Accept Unknown TCP Connection” to True

Default is False

Gatekeeper trunk redundancy

If you have two gatekeeper trunks, one in a G711 region and the other in a G729 region and you setup the Route list with G711 as the first priority and G729 as the second priority and the gatekeeper does not have enough bandwidth to process a G711 call, then it will fallback and use the G729 trunk to process the call.

You do not need to enable BRQ to make this happen.

Time of Day routing

Rule of thumb.

a. First analyze and group the time-period for which you want the partition to be active. This could be business hours (M-F, 8-5 for say).

b. If the time slots cannot be grouped into one time period, define multiple time periods.

c. Group all timeperiods into one time schedule.

d. apply time schedule to partitions and partitions to CSS.

e. calling devices should use this CSS (Gateways/ phones etc)

f. Note that when you want differential call treatment based on ToD, you need to create a tie on the dialed number. To create the tie, you have to define the same number (say 1005) twice  one in each partition P1 and P2. P1 may be a TOD enabled partition, while P2 may be a normal partition. P1 will be listed above P2 in the calling device CSS.

Unity – Tips

The following tip may be utilized when you are asked to forward a call from CM / CME to Voicemail and have the caller leave a message. Then the message should be delivered to multiple user inboxes and also  light up mwi on the appropriate phones.

a. Setup CM/CME to forward calls to VM. (say  a route point with number  1005)

b. Create a PDL in Unity with members as the individual subscribers who needs to receive the message.

c. Create a Callhandler CH with number 1005. Go to Messages – > Message recipient and select this PDL you created in step 2.

Gateway registration to gatekeeper

For a gateway to register with a gatekeeper, there should be at least one voip dial-peer defined in the gateway.

If it is a CME router, then when you define ephones, that will create dial-peer and hence that will work fine.