This update fixes the following issues that may occur when you manage a Microsoft HPC Pack 2008 R2 cluster that contains Microsoft Azure nodes:
- When you try to use an availability policy to schedule Microsoft Azure nodes that are connected to a high-availability head node, the nodes may not come online. The deployment may fail because both head nodes try to start the Microsoft Azure nodes at the same time. This hotfix enables the active head node to perform this action.
- When you try to take active Microsoft Azure nodes offline, those nodes may get stuck in the "draining" state. This behavior may occur if the head node or the server that is hosting the cluster's SQL database is under high load when the operations are tried. This hotfix makes the cluster more resistant to time-out related issues.
- Microsoft Azure nodes may enter the "error" state because of an issue with the HPC diagnostics/monitoring service. In this case, the node is not listed as online and available for use until it is restarted. This hotfix resolves the problem in which the diagnostic service blocks the node.