-
Notifications
You must be signed in to change notification settings - Fork 64
Support for exponential requeuing time and user supplied requeuing time #263
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Merged
metalcycling
merged 11 commits into
project-codeflare:main
from
metalcycling:priorities-and-preemption
Feb 14, 2023
Merged
Support for exponential requeuing time and user supplied requeuing time #263
metalcycling
merged 11 commits into
project-codeflare:main
from
metalcycling:priorities-and-preemption
Feb 14, 2023
Conversation
This file contains hidden or 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
… to specify how long to wait for the first requeuing period after dispatching.
…ntially. This time is set by default to 5 minutes but can be modified from the AppWrapper schedulingSpec field.
…m with user supplied initial time. This tests uses init containers to trick the requeuing mechanism into thinking the PODs have failed and are not going to complete.
asm582
requested changes
Feb 7, 2023
… time (not location in the list of conditions). Users can now control different ways to requeue the job and to stop it from requeuing.
asm582
reviewed
Feb 14, 2023
asm582
reviewed
Feb 14, 2023
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice work, can you please address one comment?
asm582
previously approved these changes
Feb 14, 2023
asm582
approved these changes
Feb 14, 2023
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.
In this PR, code changes are intended to fix the requeuing mechanism from triggering an AppWrapper requeuing at the beginning of the execution due to PODs taking a long time to become
Running
. This has been observed to happen when pulling images takes too long or in the presence of init containers that take a long time to run. The new exponential waiting time growth allows for long running starting PODs to eventually have enough time to get ready. In cases where the user wants to specify this by themselves, we expose therequeuingTimeMinutes
field in theschedulingSpec
stanza. By default this value is set to 5 minutes which, in our experience, is usually enough time for most applications. The new updates also check the requeuing time with respect to the last condition as opposed to when the AppWrapper wasDispatched
.