Configuring and Running a Cluster
gfsh command-line utility to configure your Pivotal GemFire cluster (also called a “distributed system”). The cluster configuration service persists the cluster configurations and distributes the configurations to members of the cluster. There are also several additional ways to configure a cluster.
gfsh to configure regions, disk stores, members, and other GemFire objects.
You can also use
gfsh to start and stop locators, servers, and GemFire monitoring tools. As you execute these commands, the cluster configuration service persists the configuration. When new members join the cluster, the service distributes the configuration to the new members.
gfsh is the recommended means of configuring and managing your Pivotal GemFire cluster, however you can still configure many aspects of a cluster using the older methods of the cache.xml and gemfire.properties files. See cache.xml and the Reference for configuration parameters. You can also configure some aspects of a cluster using a Java API. See Managing Pivotal GemFire.
The Pivotal GemFire cluster configuration service persists cluster configurations created by
gfshcommands to the locators in a cluster and distributes the configurations to members of the cluster.
A short walk-through that uses a single computer to demonstrate how to use
gfshto create a cluster configuration for a GemFire cluster.
You can dynamically deploy your application JAR files to specific members or to all members in your distributed system. GemFire automatically keeps track of JAR file versions; autoloads the deployed JAR files to the CLASSPATH; and auto-registers any functions that the JAR contains.
Pivotal GemFire allows you to organize your distributed system members into logical member groups.
The cluster configuration service exports and imports configurations created using
gfshfor an entire Pivotal GemFire cluster.
When you use the cluster configuration service in GemFire, you can examine the generated configuration files in the
cluster_configdirectory on the locator.
gfshsaves configuration files at the cluster-level and at the individual group-level.
To load an existing cache.xml or gemfire.properties configuration file into a new cluster, use the
--load-cluster-configuration-from-dirparameter when starting up the locator.
You can connect
gfshvia HTTP or HTTPS to a remote cluster and manage the cluster using
You can deploy your Pivotal GemFire configuration files in your system directory structure or in jar files. You determine how you want to deploy your configuration files and set them up accordingly.
Determine the proper startup and shutdown procedures, and write your startup and shutdown scripts.
The locator is a GemFire process that tells new, connecting members where running members are located and provides load balancing for server use.
A GemFire server is a process that runs as a long-lived, configurable member of a client/server system.
GemFire output files are optional and can become quite large. Work with your system administrator to determine where to place them to avoid interfering with other system activities.
You can configure and limit port usage for situations that involve firewalls, for example, between client-server or server-server connections.