awslabs / benchmark-ai

Anubis (formerly known as Benchmark AI), measures the goodness of machine learning workloads
Apache License 2.0
16 stars 6 forks source link

[feature] Add --terminate command to allow for smart-exiting of the watcher #1033

Closed ryansteakley closed 3 years ago

ryansteakley commented 4 years ago

Description of changes: --status -- watch will exit instead of looping infinitely, if local logs exist it will exit with a status of 1 in failure 0 in success. The watcher will read the local logs after 30seconds of continuous empty responses from the api. Then will exit based off of the local logs indications of success or failure

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

arjkesh commented 3 years ago

Hi @gavinmbell -- do you recommend we implement a separate feature for this instead of piggy backing off of watcher? Looking for a path forward here, as we want to have a way of exiting a CodeBuild job (or any automation unit) with the status of the benchmark job