It is not uncommon to get emails like these from Call Manager RTMT alerts:
Number of MediaResourceListExhausted events exceeds configured threshold during configured interval 0 within 60 minutes on cluster StandAloneCluster.
There are 1 MediaResourceListExhausted events (up to 30) received during the monitoring interval From Tue Sep 23 14:47:32 EDT 2014 to Tue Sep 23 15:47:32 EDT 2014:
MediaResourceListName : NULL_LIST
MediaResourceType : 7
AppID : Cisco CallManager
ClusterID : StandAloneCluster
NodeID : ABCD-CCM04
TimeStamp : Tue Sep 23 14:48:24 EDT 2014
What does this Alert mean?
The sort answer is that CallManager has gone through all devices in the Media Resource Group and was not able to assign a device to use. For example, someone wanted MOH (music on hold), and CallManager went down each MRG (media resource group) inside of the MRGL (media resource group list), and was not able to find an available resource.
At this point, you’ll want to look at the MediaResourceListName – that is the MRGL that ran out of resources. You could simply add more resources to this list, such as more DSPs for a router, or adding additional MRGs to the MRGL. I my case, the MediaResourceListName is the NULL_LIST
What is the NULL_LIST?
The NULL_LIST is the default media resource group list. When you install CallManager, there are no Media Resource Group Lists – everything is in a catch all bucket called the NULL_LIST. When you add additional resources, such as Call Manager subscribers, transcoders, or the like, if they are not assigned to a specific MRGL then they become part of the default list AKA NULL_LIST.
When a media resource is requested, Call Manager first checks the assigned MRGL, and if no resources are found, it checks the NULL_LIST, and still nothing is found, it throws an alert.
So, what can we do about this?
The way to resolve this is to make sure that every media resource is assigned to a Media Resource Group List, that way nothing is using the NULL_LIST. This is harder than you may think, since there are so many different resources in a normal Call Manager deployment, but by assigning everything to a list you’ll eventually weed through these errors.