On H2 Clustering / High Availability documentation, it says Clustering can only be used in the server mode (the embedded mode does not support clustering). Does this mean CreateCluster tool can not be used in embedded mode? Or we just should not use H2 embedded mode in clustering environment?
Running the cluster is only supported in server mode, not when using the embedded mode.
The CreateCluster tool is used to copy a database and initialize clustering. It can in theory be used in embedded mode: Either the source or the target or both can be embedded databases. However, if both source and target are embedded databases, then the database files need to be moved after the target is created, because running the cluster doesn't make sense if both databases are on the same machine:
The main idea behind the H2 cluster feature is to provide high availability, that means the application should continue to run even if one of the database servers fails. If both databases are running within the same machine, then you don't have high availability any more.