Skip to content

SCHED-1290 Run more checks in one slurm job for active checks#2720

Open
asteny wants to merge 1 commit into
mainfrom
SCHED-1290
Open

SCHED-1290 Run more checks in one slurm job for active checks#2720
asteny wants to merge 1 commit into
mainfrom
SCHED-1290

Conversation

@asteny

@asteny asteny commented Jul 9, 2026

Copy link
Copy Markdown
Collaborator

Problem

Solution

Testing

Release Notes

@asteny asteny added the feature label Jul 9, 2026
@asteny asteny marked this pull request as ready for review July 10, 2026 09:26
- "manage-jail-state"
- "wait-for-soperatorchecks-srun-ready"
- "wait-for-topology"
schedule: "25 5-23/6 * * *"

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No schedule anymore? As I understand it, by default it will be run once a year:

	// +kubebuilder:default="0 0 1 1 *"
	Schedule string `json:"schedule,omitempty"`

Was it on purpose?

slurmJobSpec:
sbatchScriptFile: "scripts/all-reduce-perf-nccl-with-ib.sh"
eachWorkerJobs: true
maxNumberOfJobs: 200

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

maxNumberOfJobs: 200 was not added to gpu-checks, is it OK?

#!/bin/bash
#SBATCH --deadline="now+4hours"
#SBATCH --deadline="now+12hours"
#SBATCH --time=30:00

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Shouldn't we also increase the time together with the deadline?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants