Forum Discussion
tstack
9 years agoFormer Employee
A true production setup would require four nodes: 2 Feed-Masters and 2 JCCs. There should also be a load-balancer for the Feed-Masters. The Ultra Tasks should also be configured to have at least two instances. The redundancy is required to satisfy the high-reliability goals of Ultra. For example, if one of the nodes fails, the other node can handle the load until the failed machine can be restored.
Both nodes cannot be hosted on the same VM, they must be separate.