[serve] Don't change deployment status when autoscaling #36520
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Cindy Zhang <cindyzyx9@gmail.com>
Signed-off-by: Cindy Zhang <cindyzyx9@gmail.com>
edoakes
approved these changes
Jun 20, 2023
| @@ -1289,6 +1289,19 @@ def _set_target_state(self, target_info: DeploymentInfo) -> None: | |||
|
|
|||
| logger.info(f"Deploying new version of deployment {self._name}.") | |||
|
|
|||
| def _set_target_state_autoscaling(self, num_replicas: int) -> None: | |||
| """Update the target number of replicas based on an autoscaling decision.""" | |||
There was a problem hiding this comment.
please update this comment to indicate why we need a separate codepath and what is different from _set_target_state
| def wait_for_condition_raise( | ||
| condition_predictor, timeout=10, retry_interval_ms=100, **kwargs: Any | ||
| ): | ||
| """Wait until a condition is met. If exception occurs, raise it.""" |
There was a problem hiding this comment.
can you just make this behavior a flag passed to existing wait_for_condition? would reduce the likelihood of someone else rewriting this themselves in the future
There was a problem hiding this comment.
Good point, I've added this as a parameter to wait_for_condition
Signed-off-by: Cindy Zhang <cindyzyx9@gmail.com>
Signed-off-by: Cindy Zhang <cindyzyx9@gmail.com>
Signed-off-by: Cindy Zhang <cindyzyx9@gmail.com>
sihanwang41
approved these changes
Jun 21, 2023
|
Thanks @sihanwang41, I've addressed your comment! @edoakes Tests are passing, should be ready to merge! |
8 tasks
arvind-chandra
pushed a commit
to lmco/ray
that referenced
this pull request
Aug 31, 2023
…36520) The deployment state UPDATING should only be used during redeployment. Right now the state is updating during autoscaling, which can be confusing for users. This PR makes it so that the state doesn't change during autoscaling. This usually means that a deployment's status will remain HEALTHY while it's autoscaling. Signed-off-by: e428265 <arvind.chandramouli@lmco.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Why are these changes needed?
The deployment state UPDATING should only be used during redeployment. Right now the state is updating during autoscaling, which can be confusing for users. This PR makes it so that the state doesn't change during autoscaling. This usually means that a deployment's status will remain HEALTHY while it's autoscaling.
Related issue number
Closes #35948
Checks
git commit -s) in this PR.scripts/format.shto lint the changes in this PR.method in Tune, I've added it in
doc/source/tune/api/under thecorresponding
.rstfile.