Enis and Greg;
This sounds like the availability zone issue. I am assuming you are restarting the same cluster (i.e., using the same name) and the instance that gets created may be created in a zone different from the one that the volume from the previous instantiation of the cluster was in.
You are correct, that I'm restarting a cluster.
BioCloudCentral does try to consistently start in the same availability zone to avoid this issue. I added some fixes at the end of December that try to handle it. Can you describe the process you initially used to create the cluster before re-starting it with BioCloudCentral? I can think of a couple of ways this could be a problem: - If you start the initial instance of 'cluster name' via the Console or some other method, then subsequently start it up via BioCloudCentral. - If you start the initial instance of 'cluster name' via BioCloudCentral as one image type (saw micro), then subsequently restart it as a different image type (say large). The availability zone issue is a bit tricky, since not all image types are available in all availability zones. Additionally this differs between users. Greg, if you're consistently starting the same instance type via BioCloudCentral maybe something else is wrong and we can dig into it more to try and figure this out. We'd definitely like to make it as work as smoothly as possible. Thanks for the feedback, Brad