DOH! Forgot the ‘cli’. I was just blindly following your advice from this post 
– New hostname and failed worker nodes – where it didn’t have the ‘cli’ in there! But I now see in THIS one Adding new hostname in user interface that you had it in there – I blame not enough coffee in thinking straight yet this morning! I think we’re good to go. I’ll check with the person that was trying to run the job. Mucho Thanks!
Yes, I think that solved THAT problem. But now there is another issue. I tried to run ‘cryosparcm test workers P1’, but that failed with “Error: [CUresult.CUDA_ERROR_LAUNCH_FAILED] Call to cuDevicePrimaryCtxRetain results in CUDA_ERROR_LAUNCH_FAILED”. I’ll open a new issue if I can’t figure that out.