1

Our application is currently running in standalone (on-premise), planning to move in to cluster(High Availability). Could be please let me know what all the possible Mule components and changes which i needs to be done in Developer point of view.

If you have already experience on migrating and any issues which should taken care. Your points really help me.Hope you understand the question. Please let me know if my questions needs further explanation. Thanks in advance.

Infinity
  • 484
  • 2
  • 8
  • 21

2 Answers2

0

In general, Mule automatically coordinates the resources and messages in a HA Cluster.

For example:

  • If you have an FTP connector, a File or a JDBC, etc. Mule will automatically coordinate the nodes, so only one will access at one time.

It's important to divide the flows in many steps and use VMs or JMS, so Mule can balance the messages across the nodes.

Take a look to the documentation: https://docs.mulesoft.com/mule-user-guide/v/3.5/mule-high-availability-ha-clusters To see al considerations and recomendations.

But in general you won't need to make major changes to your application, at most using VMs and JMS

Victor P.
  • 251
  • 1
  • 7
  • I have gone through the link earlier.Thanks for the response. I'm looking for realTime scenario's who might faced while migrating and things should i consider. – Infinity Jul 20 '16 at 20:55
0

You need to functionally test the flows which are using the Object store to maintain any kind of state. If you are retrieving any object from object store then Hazelcast (mule internally uses hazelcast in cluster) gives you copy back and you need to store it back again in the object store only then it will reflect back on all the cluster nodes

Naren
  • 11
  • 1