Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. And see here for current known issues.

New Member

VCS Cluster Upgrade from X.7.2.2 to X.8.1 only one way or no Video . Is Dangerous!!!

Hi,

 

we had an upgrade from X.7.2.2 to X.8.1 and run into the problem with one way Video or established call without video or no incoming video, if the SIP Client is registered on the VCS-E Cluster.

If the client is regisred on the VCS-C Cluster , than we have no probleme, only if the registering on the VCS-E Cluster.

I found the Bug CSCun38958

Symptom:
Where the customer has a VCSe Cluster and external endpoints are registered on the VCSe cluster peers, SIP calls between internal and external clients can result in one-way video.

Conditions:
One way video is as a result of SIP signalling routing between peers on the VCSe cluster, e.g. Internal EP -> VCSc -> VCSe1 -> VCSe2 -> External EP.

Workaround:
Proxy registrations from the external clients to the VCS-C cluster.

 

The strange point is, that since X.8.1 you could not switch to the proxy registration from VCS-E to VCS-C Cluster. It doesnt matter if you remove the Sip Domain on the VCS-E and set proxy registration to " proxy knwo only/any. You cant switched. Is there a limitation on with X.8.1. i check also the device authentication guide and to my suprise, this model( proxy registration to the vcs-s) is no longer available.

What going on?

On one hand cisco says please switch the registration Model , and then the switch does not work.

 

6 REPLIES

If you could tell us a bit

If you could tell us a bit more about your deployment.

If you really have an issue, I would recomend you to open a TAC case for it.

 

I think a college had some issue with proxy registration but if I recall it, that was a search rule

issue. In addition I read about some trouble with negative regex matching, so maybe thats related.

 

Regards no media, I do not want to rule out that you hit some bug, but also check if your

call flow really looks like described. If you run the second cluster peer for redundancy and

not regards scaling, maybe setting the second peer into maintanance mode could at least be

a workaround for you (with a manual failover if needed).

 

Besides that, all your traversal zone VCS (the -e and the -c) are on X8.1 right? If not thats a requirement which is said to cause media issues.

 

Also the port range was increased. So check that the used firewall ports are open. For regular media, multiplexing, turn server if used, ... Check the updated X8.1 firewall guide and the open port list on your VCS.

 

 

Please remember to rate helpful responses and identify

New Member

Hi, I think a college had

Hi,

 

I think a college had some issue with proxy registration but if I recall it, that was a search rule

issue. In addition I read about some trouble with negative regex matching, so maybe thats related.

I tried with the simple Search Rule any to any rom VCS-E to VCS-C

 

Besides that, all your traversal zone VCS (the -e and the -c) are on X8.1 right? If not thats a requirement which is said to cause media issues.

All peers was on X.8.1

 

Also the port range was increased. So check that the used firewall ports are open. For regular media, multiplexing, turn server if used, ... Check the updated X8.1 firewall guide and the open port list on your VCS

It was not a new installation, it was a upgrade. We did not change the FW ports ( release notes X.8.1)

 

VIP Green

The media demux ports change

The media demux ports change with X8.1 (mentioned in the release notes).  You may need to update your firewall rules to allow for this.

Wayne
--
Please remember to rate responses and to mark your question as answered if appropriate.

Wayne
--
Please remember to rate responses and to mark your question as answered if appropriate.
VIP Purple

We did an upgrade from X7.2.2

We did an upgrade from X7.2.2. to X8.1 like you.  I just changed the registration proxy on our Expressway to send them to the Control, and it's working without issue.  I didn't even remove the SIP domain from the Expressway yet.

New Member

I know also the configuration

I know also the configuration and without or with removing the sip domain and set the proxy regist..... to the vcs-c but it does not work. It could not be to change his deployment and call policy rules ect.. for this.

I must  change my registration Modell (all register on the VCS-C)of this bug CSCud17952 to seperate external on the vcs-e and internal on the vcs-c. 

And now cisco says  changing again your registration model ogf this bug CSCun38958.

Not nice

VIP Purple

Haha, oh very funny, it's my

Haha, oh very funny, it's my bug.  We had this issue even before X8.1 running X7.2.2.  We haven't changed our registration settings on the Expressway yet, the issue only occures a portion of the time, depending on if the Control cluster traverses to the Expressway cluster to the same peer as the registered endpoint.  We don't have many external users at this time, so we're kind of playing it by year at the moment, and if we happen to have several external calls that just fail over and over, then we'll make the change.

243
Views
0
Helpful
6
Replies
CreatePlease login to create content