Pivotal Knowledge Base

Follow

cluster_config directory is not created when starting locator in Pivotal GemFire

Environment

Product Version
Pivotal GemFire 8.x and later

Symptom

The cluster_config directory does not always get created at locator start up; the purpose of this article is to explain why this happens. 

Cause

When starting a cluster with cluster-config enabled, it may happen that the cluster_config directory under the locator's working directory is not created. Moreover, even after stopping the cluster, it is still not created. This may occur even when the locator is configured with the following:

-Dgemfire.enable-cluster-configuration=true    (note that enable-cluster-configuration default value is true)

and the locator log contains the following message:

   [info 2015/09/25 11:28:51.356 EDT loc1-d9635  tid=0x56] Cluster configuration service start up completed successfully and is now running ...

So, under what circumstances is the cluster_config directory created?

The cluster_config directory and files are not created unless they need to be. If you start up a cluster using a cache.xml file to define regions, the cluster_config directory will not get created; cluster configuration is tied to the use of Gfsh. When you define regions, configure properties, or deploy jars, the cluster_config directory will be created with the following content:

  • cluster.xml -- A cache.xml file containing configuration common to all members
  • cluster.properties -- a gemfire.properties file containing properties common to all members
  • Jar files that are intended for deployment to all members

Note: the cluster can be setup using an existing cluster_config directory as described here.

Comments

Powered by Zendesk