[autoscaler] Create Docker Command Runner#8806
[autoscaler] Create Docker Command Runner#8806richardliaw merged 6 commits intoray-project:masterfrom
Conversation
|
Can one of the admins verify this patch? |
| process_runner=subprocess, | ||
| use_internal_ip=False): | ||
| use_internal_ip=False, | ||
| docker_config=None): |
There was a problem hiding this comment.
Should we just pass the entire config in?
Currently provider_config, auth_config, cluster_name, and file_mounts (and now docker_config) are all parts
There was a problem hiding this comment.
No, let's pass in the docker_config like what you're doing now
There was a problem hiding this comment.
Okay--is there a specific reason for this?
There was a problem hiding this comment.
Well, it helps isolate what is needed and what is not needed. For example, initialization_commands is also part of the config, but there are times when we use the NodeUpdater where we want to run it and when we don't want to.
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
|
Test PASSed. |
|
Test PASSed. |
|
Test PASSed. |
|
Is this ready to merge? |
|
@richardliaw Yes it is! |
Why are these changes needed?
This PR is the first in a few PRs to refactor how the autoscaler interfaces with docker. This PR mainly focuses on how the CommandRunner is instantiated. The DockerCommandRunner mostly passes commands down to an enclosed SSHCommandRunner. The main functionality change
is that rsync actually sends the file into docker.
Related issue number
Closes #4183
Closes #4403
Replaces #8527
Checks
scripts/format.shto lint the changes in this PR.