Skip to content
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

Remove commit mechanism for pipeline in allocate action #540

Closed
sivanzcw opened this issue Nov 20, 2019 · 3 comments
Closed

Remove commit mechanism for pipeline in allocate action #540

sivanzcw opened this issue Nov 20, 2019 · 3 comments
Labels
kind/feature Categorizes issue or PR as related to a new feature. lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale.

Comments

@sivanzcw
Copy link
Contributor

/kind feature

Description:

The allocate action uses the commit mechanism. Only when the job meets the requirements of gang, the operations performed in the allocate action will be executed. Otherwise, all actions will be rolled back.

The commit mechanism in allocate action for pipeline action should be canceled:

  • When judging whether the job meets the requirements of gang, it does not calculate the task with pipelined state. That is to say, in the commit mechanism, th pipeline action is not considered.

  • When the preempt action or the reclaim action is enabled, in the previous scheduling period, the pod named poda with pending state successfully evicts a running pod named podb, and poda was pipelined to the node where the podb is running. In the next cycle, when the allocate action is performed. Because the job where poda is located does not meet the requirements of gang, and cannot perform the pipelined action, in this scheduling period, poda is still in the pending state, and will evict other pods.

@k82cn
Copy link
Member

k82cn commented Nov 20, 2019

I think we should cancel the pipeline if gang scheduling failed; the releasing resource should be reused by other jobs.

@stale
Copy link

stale bot commented Aug 18, 2020

Hello 👋 Looks like there was no activity on this issue for last 90 days.
Do you mind updating us on the status? Is this still reproducible or needed? If yes, just comment on this PR or push a commit. Thanks! 🤗
If there will be no activity for 60 days, this issue will be closed (we can always reopen an issue if we need!).

@stale stale bot added the lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. label Aug 18, 2020
@stale
Copy link

stale bot commented Oct 17, 2020

Closing for now as there was no activity for last 60 days after marked as stale, let us know if you need this to be reopened! 🤗

@stale stale bot closed this as completed Oct 17, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
kind/feature Categorizes issue or PR as related to a new feature. lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale.
Projects
None yet
Development

No branches or pull requests

3 participants