The ClusterResourceManager PolicyEngine

The primary goal of the PolicyEngine is to encapsulate the potential complex computation of the TransitionGraph from the CurrentClusterState.

The PolicyEngine does not do any network IO, it only runs this algorithm.

Main success scenario

1. The ClusterResourceManager starts the PolicyEngine on demand, or when becomes DesignatedCoordinator.

2. The PolicyEngine starts up and signs in with the ClusterResourceManager. This stage verifies the version numbers et cetera.

3. The PolicyEngine waits for orders from the ClusterResourceManager.

4. An order is received:

   1. Either a ClusterInformationBase dump requesting a computation of a transition graph.

   2. Or a request to shutdown.
5. The requested operation is performed.

   1. PolicyEngineComputation

6. The answer is sent back to the ClusterResourceManager.

   1. In case of the shutdown request, this message signs us out from the ClusterResourceManager.

   2. Otherwise, we sent back the TransitionGraph.

7. If applicable, we go back to waiting for orders, unless this was the last iteration.
8. We shutdown cleanly with exit code 100.

Extensions

• To ease testing, the PolicyEngine shall have a commandline mode which only reads the ClusterInformationBase dump from stdin and outputs the TransitionGraph on stdout.

Open issues

AndrewBeekhof: It would be nice if we could give you a ClusterInformationBase dump (possibly minus StateInformation) and the PolicyEngine gave a yes or no answer as to whether it was solvable/valid. We could then decide if an AdministrationUpdate was ok to apply to the ClusterInformationBase.

Linux-HA site built by the Linux-HA crowd. HTML4 / CSS2