Rejected by CAC

[medoneit]

Why do I occaisionaly receive "Rejected by CAC" errors?

I tend to get HEAPS all at the one time... often at 6pm for some strange reason...

Where could it be comming from? What causes "CAC" errors?

[medoneit]

Also receiving on my M32:
MG rejected by cac System capability exceeded

What is MG and what is CAC?

[skyplonk]

Same. I get them on 2x, 4x, 16x, and now even 32x systems. No one can really give me a good reason why as often there is little system load when I get them.

Really odd

[davrays]

Well,

MG - is the "Megia Gateway" application on teh Quadro
CAC - is the "Call Admission Control" module of that application.

Every such event means that one call was rejected by the system because of the resourse allocation problems. This could be due to low bandwidth, or heavy usage of G.729 licenses, or just too many calls. If you get heaps of such events at 6pm, it means your users place too many calls at that time, so sytem has to reject some of them.

You would need to review the logs to see the exact reason... Or just check the bandwidth setting on the Quadro, and the usage og G.729 license. Those are the most often reasons of the CAC errors.

[tsim]

matter is now resolved and perhaps it might be relevant to share observations with others. We had several "rejected by cac System capability exceeded " event messages. When this happens, an incoming call will get an error message advising "RTP error". This was traced to RTP setting of port range, 6000 to 6005 being too low. We have now set this to 6000 to 6255 with similar port forwarding in router and so far working well.

We had the change done to RTP to diagnose a one-way voice issue; one earlier forum entry advised to lower the port range but this did not address the issue.

[tchancev]

We get this a lot and we have followed Epygi's documentation on NAT Traversal when PBX is behind firewall. Epygi recommends the full 6000-6255 RTP port range. Even with this setting, we sometimes get this same message of call rejection due to call capacity being exceeded. So even with the 255 RTP ports available, and only one call active, we have seen this error. Leads one to believe it is not always tied to RTP port range.

Love how they always say look at the logs. Sure would be nice to know what specific log file to review. I have reviewed the MG log and do not see a whole lot of useful data. And I cannot even find the CCA log that is mentioned in this thread.

Anyone else know of other possible causes of the event message "System capability exceeded" logged at the same time as the unsuccessful call history message of "System Overload - RTP port limit". Again, only one call active when this happens. Is that one call using all 255 RTP ports?

Would like to trace this down as we do see it on several qx models periodically. Not very often but we do see it.

[ChuckN]

We get this a lot and we have followed Epygi's documentation on NAT Traversal when PBX is behind firewall. Epygi recommends the full 6000-6255 RTP port range. Even with this setting, we sometimes get this same message of call rejection due to call capacity being exceeded. So even with the 255 RTP ports available, and only one call active, we have seen this error. Leads one to believe it is not always tied to RTP port range.

Love how they always say look at the logs. Sure would be nice to know what specific log file to review. I have reviewed the MG log and do not see a whole lot of useful data. And I cannot even find the CCA log that is mentioned in this thread.

Anyone else know of other possible causes of the event message "System capability exceeded" logged at the same time as the unsuccessful call history message of "System Overload - RTP port limit". Again, only one call active when this happens. Is that one call using all 255 RTP ports?

Would like to trace this down as we do see it on several qx models periodically. Not very often but we do see it.

Sorry for this rather late reply, I just came across this same "Rejected by CAC" error on my QX200 running 6.2.45, and came across this. In my case, I too only have one cal active at the time and received the error. I only have my RTP port range set to 6200 to 6299, and I have now extended it to 6200 - 6499.

Kind of makes me wonder, perhaps the QX200 picks a random port (up to 255) from the "start" port to use for its RTP streams, hence the error - maybe some kind of a bug? I may be completely out of whack with my thinking, but can't think of anything else as to why it would require that many RTP ports with just a single call...