cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
4314
Views
85
Helpful
15
Replies

Extension Mobility Redundancy Issue

alfonso.cornejo
Participant
Participant

Hi to all,

I have an scenario of a publisher and a subscriber in my telephony infrastructure and i have configured EM for some extensions, the issue that i have is that the url for the EM service is pointing to my publisher server so when i turn it off for a HA test the EM service is no longer available.

I've searched for HA scenarios, there are suggestions about configuring a "redundant" EM service but this time the url should be pointing to the subscriber server, i did that and also the two services were showed in the ip phones and i was able to login using both EM services but when i turned off the publisher server to test it all the services were gone from my phones.

Is there something else besides the phone service for EM that i need to configure???? i also checked a dns solution using a name instead of an ip address in the url for the EM service but it doesn't work because of the behavior of dns that it doesn't know when a server is down so it could be resolving wrong ip addresses.

Thanks in advance for your help.

1 Accepted Solution

Accepted Solutions

All excellent replies! A round of 5's for everyone. I have actually done this using DNS solely. I had a customer that was planning to add SLB at a later time but wanted an interim solution. I leveraged DNS round-robin for non-enterprise services and associated URLs (like the actual Services button). I found that in 8.5(3)+ version of firmware on 79xx phones (yes, dated) that the phones will actually handle the scenario where DNS returned an IP address for a server that was actually off line. In my tests, the phones would simply bypass the "broken" host and use one of the other A records returned by DNS.

I have not tested this with conference stations, 69xx, 89xx, or 99xx stations. But I have tested several versions of firmware after 8.5(3) and the behavior was consistent.

I have more details here:

http://www.ucguerrilla.com/2012/05/redundancy-considerations-and-cisco-ip.html

As noted by others, the SRND, and the blog entry above - a SLB solution is recommended and ideal but I think DNS round robin is worth testing. Of course, if you don't trust your DNS infrastructure then maybe that won't work for you.

HTH

-Bill (@ucguerrilla)

HTH -Bill (b) http://ucguerrilla.com (t) @ucguerrilla

Please remember to rate helpful responses and identify

View solution in original post

15 Replies 15

Harmit Singh
Cisco Employee
Cisco Employee

Hi Alfonso,

Did you get a chance to review the following doc:

http://www.cisco.com/en/US/products/sw/voicesw/ps556/products_tech_note09186a008097cd93.shtml

It talks about configuring IP Phone service redundancy which is what you would need in this scenario.

HTH.

Regards,

Harmit.

Just to add to my earlier response, please refer to the SRND which talks in alot of detail on EM and IP Phone service redundancy:

http://www.cisco.com/en/US/docs/voice_ip_comm/cucm/srnd/7x/cmapps.html#wp1043795

http://www.cisco.com/en/US/docs/voice_ip_comm/cucm/srnd/7x/cmapps.html#wp1043648

Regards,

Harmit.

Hi guys,

I've read this docummentation but unfourtunately my customer doesn't have a load balancer to perform the best redundancy mechanism but for me it sounds like a "manual" redundancy the fact that you can configure a second EM service but with it's url pointing to one of the subscribers ip address and use this to log in to the phones when the publisher is down.

Do you agree??

My problem with this is that once i turn off the publisher no services are being showed in the phones, do you know if there is something else that has to be configured in order to have services that are being handled by the subscribers available after turning off the publisher???

Thanks in advance for your help.

Hi Alfonso

The problem is that the first thing the phone does is retrieves the 'services' URL, and then gets the list of services (e.g. EM and so on) from that URL.

So you need that first URL to be resilient; and you can only enter one in Enterprise Parameters.

Aaron

Aaron Please remember to rate helpful posts to identify useful responses, and mark 'Answered' if appropriate!

Alfonso you can use cisco IOS SLB feature for load balancing, rather than a load balancer appliance.

The link below shows how to configure SLB in IOS.

http://www.cisco.com/en/US/docs/ios-xml/ios/slb/configuration/12-4/Configuration_Examples_for_IOS_SLB.html

I must point out that if your Load balancer goes down, then your EM service will also not be available. So to have a full resiliency your load balancer has to be resilient.

If that is still not an option then you are left with using DNS. This is not reliable because DNS servers do not have a way to know if a resolved host is down or not.

Please rate all useful posts

"'Nature is too thin a screen, the glory of the omnipresent God bursts through it everywhere"-Ralph Waldo Emerson

Please rate all useful posts

All excellent replies! A round of 5's for everyone. I have actually done this using DNS solely. I had a customer that was planning to add SLB at a later time but wanted an interim solution. I leveraged DNS round-robin for non-enterprise services and associated URLs (like the actual Services button). I found that in 8.5(3)+ version of firmware on 79xx phones (yes, dated) that the phones will actually handle the scenario where DNS returned an IP address for a server that was actually off line. In my tests, the phones would simply bypass the "broken" host and use one of the other A records returned by DNS.

I have not tested this with conference stations, 69xx, 89xx, or 99xx stations. But I have tested several versions of firmware after 8.5(3) and the behavior was consistent.

I have more details here:

http://www.ucguerrilla.com/2012/05/redundancy-considerations-and-cisco-ip.html

As noted by others, the SRND, and the blog entry above - a SLB solution is recommended and ideal but I think DNS round robin is worth testing. Of course, if you don't trust your DNS infrastructure then maybe that won't work for you.

HTH

-Bill (@ucguerrilla)

HTH -Bill (b) http://ucguerrilla.com (t) @ucguerrilla

Please remember to rate helpful responses and identify