This topic describes how Enhanced Location CAC works in an intercluster deployment.
- Each cluster is configured with its local network model.
- All clusters exchange their network models with each other.
- In general, location names have to be globally unique across all clusters.
- Locations that exist in multiple clusters are called common locations.
- By overlaying common locations, individual cluster network models are merged into one overall network model including all locations of all clusters.
When Enhanced Location CAC is used across clusters, each LBM server has a complete view of the entire network model. From a configuration perspective, however, each cluster is only configured with its local network model (that is, the locations and links that belong to the local cluster).
All clusters exchange their local network model with each other so that the entire network model can be assembled at each cluster.
Because of the entire merged network model, locations must be globally unique across all clusters. If the same location, with the same, case-sensitive name, exists in multiple clusters, this location is referred to as a common location. Common locations are required to enable the individual clusters to put together the entire network model. Common locations are overlaid and thus “glue” the individual cluster network models together.
By default, each cluster has a location called Hub_None. If you do not want all locations to share this location as a common location, you must rename the Hub_None location at all clusters that should not share this location with other clusters as a common location.
The figure shows an example of three clusters. Cluster A has four locations (A-1, A-2, A-3, and CL_AB), cluster B has three locations (B-1, CL_AB, and CL_BC), and cluster C has three locations (CL_BC, C-1, and C-2). CL_AB is a common location of clusters A and B. CL_BC is a common location that is shared by clusters B and C. By overlaying the common locations, the three individual cluster network models are combined into a single network model of the entire intercluster topology. This combined network model has eight locations: three locations that exist only in cluster A, one location that exists only in cluster B, two locations that exist only in cluster C, and two locations that are each shared by two clusters.
As each cluster replicates its locations and links, along with the link configuration parameters (weight and bandwidth limits), all clusters are aware of all of the network model information.
LBM Replication Network Characteristics
This section describes how replication occurs in an intercluster Enhanced Location CAC deployment.
Replication within a cluster:
- Performed automatically
- Fully meshed communication between all LBMs of a cluster
Replication across clusters:
- One or more LBMs of a cluster can be designated as LBM hubs.
- LBM hubs are responsible for replicating information to other clusters.
- Each cluster replicates with all other clusters (full mesh of clusters).
- To avoid the need for full-mesh configuration, one or more replication network PoCs are configured at each cluster.
- Each cluster queries its configured PoCs to learn about other clusters.
- PoCs are configured by LBM intercluster replication groups.
- LBMs are designated as LBM hubs by assigning an LBM intercluster replication group to the server running the LBM service.
Replication within a cluster is performed automatically. It occurs in a fully meshed fashion between all LBMs of a cluster.
Replication across clusters does not occur by default. For intercluster replication to work, each cluster must have at least one LBM hub. One or more LBMs in a cluster can be designated as LBM hubs.
An LBM hub is responsible for replicating information to other clusters. The actual replication is performed in a fully meshed fashion between all clusters. From a configuration standpoint, there is no need to configure the full mesh of clusters. Instead, one or more replication network PoCs are configured at each cluster. Each cluster queries its configured PoCs to learn about the LBM hubs of other clusters.
PoCs are configured by LBM intercluster replication groups. It is recommended to configure the same PoCs at all clusters.
LBM hubs are designated as LBM hubs by assigning an LBM intercluster replication group to the server that runs the LBM service. If a server that has the Cisco LBM service activated is not configured with an LBM intercluster replication group, then the LBM service that is running on this server is an LBM spoke and will not exchange information with other clusters. Due to the internal replication within a cluster, LBM spokes will nevertheless learn all information via the local LBM hubs.
LBM Replication Network Example
This section shows an example of an intercluster Enhanced Location CAC replication network.
The figure shows three clusters. One cluster has four LBMs, of which one is designated as an LBM hub. Another cluster has four LBMs. Two of them are LBM hubs. The third cluster has three LBMs and two of these LBMs act as LBM hubs.
An LBM intercluster replication group (LBM_Intercluster-Replication_Group) is configured, and two out of the three possible PoCs are configured as members: POC-1 and POC-2. This LBM intercluster replication group is configured at each of the three clusters. This group is then applied to the local Cisco Unified Communications Manager servers that have the LBM service running and that should be performing the role of an LBM hub.
An LBM service is an LBM spoke if no LBM intercluster replication group is applied to the Cisco Unified Communications Manager server that provides the LBM service.
If a cluster has multiple LBM hubs, the LBM hub with the lowest IPv4 (entire) address will function as the sender of messages to other remote clusters. The LBM hub that functions as the sender for the network model information picks the first LBM hub of each other cluster to send replication information to this other cluster.
The LBM hub that receives information from other clusters forwards the received network model information to all other LBM servers within its cluster.
This section describes the so-called shadow location.
Intercluster Enhanced Location CAC requires a SIP ICT between clusters:
- SIP ICT must be in a special shadow location.
- Shadow location exists by default and cannot be deleted.
- Shadow location does not have links to other locations.
- Shadow location is required to exchange location and bandwidth information between clusters.
- If a device other than the SIP ICT is put into the shadow location, Cisco Unified Communications Manager will treat the device as if it were in the location Hub_None.
The shadow location is used to enable a SIP trunk to pass Enhanced Location CAC information, such as location name, call type, and other things, which are required for Enhanced Location CAC to function between clusters. To pass this location information across clusters, the SIP ICT must be assigned to the shadow location.
The shadow location cannot have a link to other locations. Therefore, no bandwidth can be reserved within the shadow location and between the shadow location and other locations.
Any device other than a SIP ICT that is assigned to the shadow location will be treated as if it was associated to Hub_None. This fact is important to know because if a device other than a SIP ICT resides in the shadow location, bandwidth deductions will be made from this device as if it were in Hub_None. That situation could have varying effects depending on the location and link configuration.
When the SIP ICT is enabled for Enhanced Location CAC, it passes information in the SIP Call-Info header that allows the originating and terminating clusters to process the location bandwidth deductions end-to-end.
Intercluster Enhanced Location CAC requires SIP ICTs. H.323 trunks are not supported by intercluster Enhanced Location CAC.
The shadow location exists by default. It cannot be renamed or deleted.
Intercluster Enhanced Location CAC: Configuration Procedure
This section describes the configuration steps that are required when enabling intercluster Enhanced Location CAC.
Configure common locations:
- Ensure that all locations, except common locations, have globally unique names.
- Ensure that common locations have identical names (case-sensitive).
Configure LBM intercluster replication:
- Configure LBM intercluster replication groups.
- Add LBM intercluster replication group.
- Configure PoCs.
- Apply LBM intercluster replication group to LBM servers that should act as LBM hubs.
The two main configuration areas are the configuration of common locations and the LBM intercluster replication.
The configuration of common locations does not include any new steps. Actually, you are simply configuring locations. The only difference between Enhanced Location CAC and traditional Location CAC is that you must carefully choose location names in Enhanced Location CAC. Ensure that locations that should not be shared by multiple clusters as a common location have globally unique location names.
On the other hand, make sure that locations that should act as common locations have the same name.
Location names are case-sensitive.