Adds fabric worker mgmt
@ mention of reviewers`
@jimmykodes`
A brief description of the purpose of the changes contained in this PR.
Resolves #340 (closed) adding a script to maintain compute workers
A checklist for hand testing
-
prereq: pip install fab-classic
outside of docker, uses your SSH key on host machine -
create server_config.yaml
and add the gpus to it (I'll just send you this), runfab -R autodl-gpu status
-
update some workers with fab -R autodl-gpu update
(may be overkill, can wait until we have relevant changes)
Checklist
-
Code review by me -
Hand tested by me -
I'm proud of my work -
Code review by reviewer -
Hand tested by reviewer -
Ready to merge