Gateway Configuration Properties

The properties listed are the attributes for the cache.xml gateway-hub element and its subelements. The gateway APIs in com.gemstone.gemfire.cache.util provide corresponding getter and setter methods.
<gateway-hub ...> 
  <gateway ...> 
    <gateway-endpoint .../> 
    <gateway-queue .../> 
</gateway-hub> ... 
<gateway-hub> Description
id Identifier for the hub, usually an identifier associated with a physical location. Default: null.
Note: A primary hub and any secondary hubs must have the same ID, as they are the same logical thing.
maximum-time-between-pings Maximum time in milliseconds the gateway allows to pass between messages or pings indicating the remote site is healthy. Default: 60000.
Note: A setting of 0 or less disables monitoring of the remote site’s health. Be careful not to do this accidentally.
port Port the hub is to listen on for communication from other sites. Default: null.
Note: If possible, use a single port number for all hubs in your WAN installation. This helps distinguish your gateway hubs from other services. If you have two hubs running on one machine, however, you must assign them different port numbers.
socket-buffer-size Size of the socket buffer used to receive messages from remote sites. This socket buffer size should match the size of the socket buffers on all remote gateways that have an endpoint specifying this hub’s host and port. Default: 32768.
startup-policy Specifies whether the hub should attempt to start as primary or secondary hub. Used for load balancing. Valid values are primary, secondary, and none. Default: none.

<gateway-hub> <gateway> Description
concurrency-level Defines how many parallel gateway queues and event processors to create. Default: none.
id Identifier for the gateway, must match the exact name of the remote hub represented by this gateway. Each gateway must be assigned a unique ID within the hub. Default: null.
order-policy Defines the ordering of events being distributed by the gateways. The valid policies are key,thread and partition. Key ordering means that updates to the same key are sent in order and are added to the same gateway queue. Thread ordering means that order is preserved by initiating thread, and updates by the same thread are sent in order to the same gateway queue. Partion-based ordering can be used when applications have implemented custom partitioning by using the PartitionResolver. Setting the order-policy to "partition" means that all gateway events that share the same "partitioning key" (RoutingObject) are guaranteed to be dispatched in order. Default: key.
socket-buffer-size Size of the socket buffers used to send messages to remote sites. Default: 32768.
Note: This should match the socket-buffer-size setting for the remote gateway-hubs this gateway connects to.


Milliseconds that this gateway will block waiting to receive an acknowledgment from a remote site. Default 10000.

<gateway-hub> <gateway> <gateway-endpoint> Description
host The host to connect to. The host/port pair must correspond to location where a remote host is listening for incoming communication. Default: null.
id Identifier for the endpoint, usually matching the name used for the remote hub that the endpoint connects to. Each endpoint must be unique within the gateway. Default: null.
port Port to connect to. he host/port pair must correspond to location where a remote host is listening for incoming communication. Default: null.

<gateway-hub> <gateway> <gateway-queue> Description
batch-conflation Boolean specifying whether to conflate messages. Default false.
batch-size Limit on the number of messages a batch can contain. Default: 100.
batch-time-interval Limit on the milliseconds that can elapse between sending batches. Default: 1000.
disk-store-name Used if enable-persistence is true. Assigns this queue to the disk store with this name. By default this is not defined, which causes all disk activities to be done through the system’s default disk store. If specified, this is the name of a store in the pool of disk stores serving the cache.
enable-persistence Boolean specifying whether to persist the queue to disk. If enabled, the queue is persisted to the disk store specified in disk-store-name. Default: false.
Note: This causes the queue to be configured like a persistent replicated region in the gateway’s cache.
maximum-queue-memory Maximum memory in megabytes that the queue can occupy before overflowing to disk. Default: 100.