Deploying artifacts across the nodes
Use the following deployment synchronization recommendations based on the rate of change of artifacts that happen in your cluster:
- For a high rate of changes (i.e., if changes happen very frequently):
- network file share
- For a medium rate of change
- For a low rate of changes (i.e., if changes happen once a week):
- use the configuration management system to handle artifact
- other deployment options (e.g., Puppet, Chef etc.)
However, using a shared file system is the first preference that WSO2 recommends to synchronize the artifacts among the nodes, because you can publish synapse artifacts to any of the nodes, avoiding the vulnerability of a single point of failure that is present when using remote synchronization (rsync).
One common approach for synchronizing artifacts across all cluster nodes is to use rsync tool, which is a file copying tool. Therefore, you can first deploy artifacts in one node of the cluster and then use rsync to copy those artifacts to other nodes as described below.