Replies: 2 comments 1 reply
-
I have the same question, even I tried setting the resources but unfortunately I couldn't. |
Beta Was this translation helpful? Give feedback.
0 replies
-
Okay I tried using |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
We are using ARC to run our own self hosted runners on EKS. So far everything is working great, however some jobs are getting killed by the cluster due to not requesting enough resources for the container:
I've tried a couple ways to configure this without success. Is it possible to specify in the Workflow file or elsewhere the resources required for the job? Here's what I've tried:
Setting the resources in the deployment for the runner, i.e.
This results in the pod requesting these resources (i.e. the number of running pods will equal
node_resources / ephemeral-storage-requests
), but I still see that therunner
container requests 0.Setting the resources in the Github Workflow file, i.e.
however as far as I know there isn't a Docker CLI option for storage? https://docs.docker.com/config/containers/resource_constraints/
Would love a pointer on how to resolve this, preferably at the job level since not all of our jobs need the same resources, and would be great not to lose parallelism due to unused resources being locked up.
Beta Was this translation helpful? Give feedback.
All reactions