[core] add option for raylet to inform whether a task should be retried #31230
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: Clarence Ng <clarence.wyng@gmail.com>
Signed-off-by: Clarence Ng <clarence.wyng@gmail.com>
|
tests look ok gentle ping: @rkooo567 @stephanie-wang @scv119 |
|
^I will review this tmrw! |
Signed-off-by: Clarence Ng <clarence.wyng@gmail.com>
scv119
approved these changes
Jan 3, 2023
AmeerHajAli
pushed a commit
that referenced
this pull request
Jan 12, 2023
…ed (#31230) Add plumbing for raylet to inform whether the worker that died should be retried. This is a no-op change for now, to be used in a follow up PR once we implement group-by-owner-id policy that will report task failure on deadlock. change oom killer to expose a bit on whether to retry the task. A future policy that can detect deadlock will set this bit to false. For now, the current policy will set this bit to true and allow the task to use its retry. if the task manager fetches this bit it will fail the task immediately, ignoring available retry. Signed-off-by: Clarence Ng <clarence.wyng@gmail.com>
tamohannes
pushed a commit
to ju2ez/ray
that referenced
this pull request
Jan 25, 2023
…ed (ray-project#31230) Add plumbing for raylet to inform whether the worker that died should be retried. This is a no-op change for now, to be used in a follow up PR once we implement group-by-owner-id policy that will report task failure on deadlock. change oom killer to expose a bit on whether to retry the task. A future policy that can detect deadlock will set this bit to false. For now, the current policy will set this bit to true and allow the task to use its retry. if the task manager fetches this bit it will fail the task immediately, ignoring available retry. Signed-off-by: Clarence Ng <clarence.wyng@gmail.com> Signed-off-by: tmynn <hovhannes.tamoyan@gmail.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.
Signed-off-by: Clarence Ng clarence.wyng@gmail.com
Why are these changes needed?
Add plumbing for raylet to inform whether the worker that died should be retried. This is a no-op change for now, to be used in a follow up PR once we implement group-by-owner-id policy that will report task failure on deadlock.
Related issue number
#30900
Checks
git commit -s) in this PR.scripts/format.shto lint the changes in this PR.