nunet onboard-ml errors with minimum storage spec.
Summary
When running on a CPU only machine with minimum specs I was getting errors when running nunet onboard-ml command. I had ~13gb allocated to the machine at the time, I tried running the command numerous times but it kept saying error ran out of space on device. The minimum specs suggests 10gb, but it wasn't possible for me to run that command with less storage. I had to increase it in order to get the command to successfully run.
Steps to reproduce
-Install DMS on a 10-13 gb disk size Ubuntu 20.04 VM, and onboard machine/wallet to nunet. -Run the nunet onboard-ml command or sudo nunet onboard-ml. -It starts to download but eventually errors out.
What is the current bug behavior?
-It will start to download several things, some successful, but eventually after waiting it fails and says the install failed because the device has ran out of storage. I don't have the exact error message as I fixed it already and have not tried to reproduce after fixing it.
What is the expected correct behavior?
nunet onboard-ml successfully installs and at the end of running the command both pytorch and the other ml suite I am forgetting the name of will have green text stating they were installed and are ready to accept jobs.
Relevant logs and/or screenshots
I don't have the screenshots or useful logs unfortunately. The only thing I found in my dms logs was this line, but even after I fixed it this error still exists in the log. traces export: rpc error: code = Internal desc = grpc: error while marshaling: string field contains invalid UTF-8
Version number of components and SO (if applicable)
Latest DMS I believe. Ubuntu 20.04 LTS.
Possible fixes
All I had to do was add more storage to the local ubuntu--vg-ubuntu--lv and re run the command to fix the issue and get the command to run successfully.