CMS redundancy question

[n.mulvihill]

My company is setting up Cisco Meeting server in a 3 server redundant configuation with expressway as the edge. Each server will run callbridge webbridge webadmin and xmpp with xmpp, callbridge and webridge being clustered. Only the second server is setup at the moment, the behavior I'm noticing is when I take down cms2 and try to access a meeting via webrtc on cms1 it will tell me the meeting is not found, as soon as I boot cms then I can access the meeting, this behavior is also seen when shutting down cms1. I'm sure its called out in the documentation somewhere but I've read every document at least once and never saw anything specifically calling out that behavior, is this to be expected with clustered webbridges?

[Anurag Srivastava1]

Hello,

You can not cluster webbridge. There are different ways to achieve webbridge redundancy.
Please see the below deployment guide under section"Deploying multiple Web Bridges"

https://www.cisco.com/c/dam/en/us/td/docs/conferencing/ciscoMeetingServer/Deployment_Guide/Version-2-6/Cisco-Meeting-Server-2-6-Scalable-and-Resilient-Deployment.pdf

Regarding your issue it seems you are missing some configuration .

Please refer the below link-

https://www.cisco.com/c/en/us/support/docs/conferencing/meeting-server/213481-configure-verify-and-troubleshoot-multi.html

Hope this helps.

Thanks,

Please rate if it is helpful....

[n.mulvihill]

I used the wrong wording, I know you cannot "cluster" a webbridge but it preforms a similar function in the CMS environment in that you can load balance calls across them, they just don't explicitly know about each other as that function is offloaded on the callbridge. I have read that scalable/redundancy guide all the way through multiple times and my environment has been setup based on this guide. It does not explicitly state what the expected behavior is in certain outage scenarios.

However I have not seen that troubleshooting guide before, I've been looking everywhere for something like that, thank you. I'll report back once I figure it out.

[n.mulvihill]

That guide didn't give me the information I was hoping for, the issue I'm having occurs on internal and external queries when one server is down so that eliminates the expressway from being the problem.

[Anurag Srivastava1]

Hi,

Is your database clustering is working properly?

and is all the spaces are being showed on all servers properly?

When you try to join the space making a server down did you see that in cospaces that meeting number is there or not?

One more thing did you added all call bridges in all xmpp servers?

You can check thing using command xmpp callbridge list.

Also if you can take sip, DNS and webbridge detailed logs probably it will be more clear.

Thanks

Please rate if it is helpful...

[abhispal]

Hello,

When you say "Only the second server is setup at the moment". what do you mean?  the behavior I'm noticing is when I take down cms2 and try to access a meeting via webrtc on cms1 it will tell me the meeting is not found, as soon as I boot cms then I can access the meeting, this behavior is also seen when shutting down cms1.

The above scenario is for both internal and external?  

I think some improper configuration might be causing the issue. When you see "meeting not Found" i think it is not even taking you to the page where you enter the Name?