currently, caboose does not provide a way for it's state of which backends are good to be saved / restored.
in cases of restart / redeploy it would be less disruptive if caboose could pick up where it left off, and immediately start with 'good' nodes rather than figuring that out again.
currently, caboose does not provide a way for it's state of which backends are good to be saved / restored.
in cases of restart / redeploy it would be less disruptive if caboose could pick up where it left off, and immediately start with 'good' nodes rather than figuring that out again.