I'm glad you figured this out.   There are some other places you can look for information about provider-specific configuration properties as well.   There are reference configuration files here: https://github.com/cloudera/director-scripts/tree/master/configs   From the help menu in the Director UI, you can access the API console, and the provider-metadata endpoints there will show you all the valid provider-specific configuration properties.   If you're comfortable reading code, the plugins that enable Director to work with multiple cloud providers are open source. If you search those repositories for ConfigurationPropertyToken, you will see the classes that define the metadata for the provider-specific configuration properties. The AWS repository, for example, is here: https://github.com/cloudera/director-aws-plugin/tree/v2   ... View more

Glad your problem is solved. Thanks for letting the community know the solution. Hopefully if someone else ends up in your situation they will find your answer helpful. ... View more

Director's deployment and cluster refreshers should detect this sort of thing and update the deployment and cluster representations (respectively) in the database with the updated version information. Can you please check the server application.log to see if either of them encountered problems?   ... View more

Yes, this is most likely a consequence of the other issue. We should try to address that one first. ... View more

The root cause should be a little earlier in the log. The log snippet you provided starts when Director is trying to clean up resources in response to the error.   ... View more

I'm sorry you're having trouble getting your cluster running. You are trying to create the cluster through the Cloudera Director UI, right? If so (or if you're using the bootstrap-remote command from the command line), you should look in Cloudera Director server's application.log for an earlier exception giving the specific failure during instance allocation.   ... View more

I'm glad your problem is resolved. Thank you for posting the details so other forum members can benefit from your experience.   ... View more

If you are using EBS volumes for your CM/CDH instances, please read the "Pausing a Cluster in AWS" section of the Cloudera Director User Guide. If you are using ephemeral drives, you cannot stop and restart instances, because the data on the ephemeral drives is lost when you restart the instances.   ... View more

I'm glad you found the answer. The byon and sandbox providers are blacklisted by default since they are not suitable for production use. Overriding the lp.plugin.configuration.blacklist property as you have done is the correct way to enable them for experimentation. If you implement your own production-quality BYON provider you should give it its own ID.   ... View more