I have ICD application based on CRA 3.5.1 and installed it in
Pubslisher server. I read a documentation that we need to put both
publisher and subscriber IP to provide redundancy. With this info, i
have put both publisher and subscriber ip in the JTAPI and RMJTAPI
setup page. But I only installed and setup the application on
Publisher server (coresident) and nothing at Subscriber server. When
the Publisher server down, the ICD application was also failed. Anyway
to prevent this? Should I install and setup CRA application on
Subscriber server as well? Please send me reference page if any.Thanks

[ Post a follow-up to this message ]

What you are configuring is redundancy only in the event of CTI Manager
service failure.  There are some things you will need to configure in
CallManager to complete this portion of the configuration.  I would
suggest reading the high availability section of the IPCC Express SRND,
found here: http://www.cisco.com/go/srnd

The SRND also describes how redundancy works and the limitations  There
is a section that describes what you need to do for CRS server
redundancy as it is not automatic in 3.5.  Basically you will create a
mirror image of the CRS server to use in an offline server in case of
failure.  The expectation for this option is that the CRS server is
running on a standalone server.

There are also some limitations if the Publisher goes down DC Directory
is being used.  This URL provides steps to configure redundant LDAP
URL.
http://www.cisco.com/en/US/products/sw/custcosw/ps1846/
products_tech_note09186a00804b4982.shtml

Kevin

On Feb 27, 2006, at 1:52 AM, Robin H wrote:

> I have ICD application based on CRA 3.5.1 and installed it in
> Pubslisher server. I read a documentation that we need to put both
> publisher and subscriber IP to provide redundancy. With this info, i
> have put both publisher and subscriber ip in the JTAPI and RMJTAPI
> setup page. But I only installed and setup the application on
> Publisher server (coresident) and nothing at Subscriber server. When
> the Publisher server down, the ICD application was also failed. Anyway
> to prevent this? Should I install and setup CRA application on
> Subscriber server as well? Please send me reference page if any.Thanks
>
>  ________________________________________
_______
> cisco-voip mailing list
> cisco-voip@puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-voip
>

[ Post a follow-up to this message ]

For the most redundancy you will want separate CRA (CRS, IPCCx, whatever
they are calling it these days) servers and upgrade to 4.x.  4.x has the
clustering ability and off box from the CMs is the only way to get true
redundancy from the CM cluster.

Just my 2 cents.

Scott

-----Original Message-----
From: cisco-voip-bounces@puck.nether.net
[mailto:cisco-voip-bounces@puck.nether.net] On Behalf Of Kevin Thorngren
Sent: Monday, February 27, 2006 4:55 AM
To: Robin H
Cc: cisco-voip@puck.nether.net
Subject: Re: [cisco-voip] CRS 3.5.1

What you are configuring is redundancy only in the event of CTI Manager

service failure.  There are some things you will need to configure in
CallManager to complete this portion of the configuration.  I would
suggest reading the high availability section of the IPCC Express SRND,

found here: http://www.cisco.com/go/srnd

The SRND also describes how redundancy works and the limitations  There

is a section that describes what you need to do for CRS server
redundancy as it is not automatic in 3.5.  Basically you will create a
mirror image of the CRS server to use in an offline server in case of
failure.  The expectation for this option is that the CRS server is
running on a standalone server.

There are also some limitations if the Publisher goes down DC Directory

is being used.  This URL provides steps to configure redundant LDAP
URL.
http://www.cisco.com/en/US/products/sw/custcosw/ps1846/
products_tech_note09186a00804b4982.shtml

Kevin

On Feb 27, 2006, at 1:52 AM, Robin H wrote:

> I have ICD application based on CRA 3.5.1 and installed it in
> Pubslisher server. I read a documentation that we need to put both
> publisher and subscriber IP to provide redundancy. With this info, i
> have put both publisher and subscriber ip in the JTAPI and RMJTAPI
> setup page. But I only installed and setup the application on
> Publisher server (coresident) and nothing at Subscriber server. When
> the Publisher server down, the ICD application was also failed. Anyway
> to prevent this? Should I install and setup CRA application on
> Subscriber server as well? Please send me reference page if any.Thanks
>
>  ________________________________________
_______
> cisco-voip mailing list
> cisco-voip@puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-voip
>

 ________________________________________
_______
cisco-voip mailing list
cisco-voip@puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-voip

[ Post a follow-up to this message ]

I looked at IPCCx SRND and saw a single point of failure where
Publisher is as DC Directory. I kept thinking how to optimize the
usage on both CCM servers.
I have tried to setup the ICD application on both Pub and SUB servers.
When I use same CTI port/RP, the JTAPI subsystem went to Partial
Service on the second server.
When I use different profile between the first and second CRA server,
it worked based on call forwarding method at CTI RP. With this, the
primary ICD service is at Subscriber and secondary service is at
Publisher. Looks like this is very useful to handle overflow calls to
the first server and so far I didn;t enccounter any problem. Any of
you guys have tried this method as well? Just want to confirm if this
kind of setup is ok for live environment.

thanks,
Robin


On 2/28/06, Voll, Scott <Scott.Voll@wesd.org> wrote:
> For the most redundancy you will want separate CRA (CRS, IPCCx, whatever
> they are calling it these days) servers and upgrade to 4.x.  4.x has the
> clustering ability and off box from the CMs is the only way to get true
> redundancy from the CM cluster.
>
> Just my 2 cents.
>
> Scott
>
> -----Original Message-----
> From: cisco-voip-bounces@puck.nether.net
> [mailto:cisco-voip-bounces@puck.nether.net] On Behalf Of Kevin Thorngr
en
> Sent: Monday, February 27, 2006 4:55 AM
> To: Robin H
> Cc: cisco-voip@puck.nether.net
> Subject: Re: [cisco-voip] CRS 3.5.1
>
> What you are configuring is redundancy only in the event of CTI Manager
>
> service failure.  There are some things you will need to configure in
> CallManager to complete this portion of the configuration.  I would
> suggest reading the high availability section of the IPCC Express SRND,
>
> found here: http://www.cisco.com/go/srnd
>
> The SRND also describes how redundancy works and the limitations  There
>
> is a section that describes what you need to do for CRS server
> redundancy as it is not automatic in 3.5.  Basically you will create a
> mirror image of the CRS server to use in an offline server in case of
> failure.  The expectation for this option is that the CRS server is
> running on a standalone server.
>
> There are also some limitations if the Publisher goes down DC Directory
>
> is being used.  This URL provides steps to configure redundant LDAP
> URL.
> http://www.cisco.com/en/US/products/sw/custcosw/ps1846/
> products_tech_note09186a00804b4982.shtml
>
> Kevin
>
> On Feb 27, 2006, at 1:52 AM, Robin H wrote:
> 
>
>  ________________________________________
_______
> cisco-voip mailing list
> cisco-voip@puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-voip
>

[ Post a follow-up to this message ]