Hello. I am trying to set up workflow with cluster so that my master process would be on my local PC. In that way for example data would be immediately saved locally, and with a simple change for include statements (Sending code to workers) I would also be able to propagate code from my development environment to the workers. That would be uber awesome as would eliminate the need to maintain these usual sync scripts (at least for usual cases)!!! But I am stuck:
addprocs([("firstname.lastname@example.org",1),1],tunnel=true) @everywhere uisng ClusterManagers fetch(@spawnat 2 addprocs_pbs(5))
ERROR: On worker 2: Only process 1 can add and remove workers
Any ideas how to get around that?