Adding local mount in workflow pod to speed up operation?

[StanislavGlebik]

Hey, thanks for the project, we are using Actions Runner Controller with containerMode: kubernetes with a network volume mount (to allow runner and workflow pod to be on different nodes) and noticed that it can take a while to run.

We are wondering if having a separate local mount in the workflow pod is an option? Is there a recommended path to mount it on?

[vvanouytsel]

As far as I understand they have recently released v.0.8.0 which contains #244.

This should remove the dependency on the volume and would allow the workflow pods to run on other nodes than the runner. However I am currently struggling implementing this.

As my 'Initialize Containers' step always fails.

Run '/home/runner/k8s/index.js'
(node:66) [DEP0005] DeprecationWarning: Buffer() is deprecated due to security and usability issues. Please use the Buffer.alloc(), Buffer.allocUnsafe(), or Buffer.from() methods instead.
(Use `node --trace-deprecation ...` to show where the warning was created)
Error: Error: cpToPod failed after 30 attempts: {}
Error: Process completed with exit code 1.
Error: Executing the custom container implementation failed. Please contact your self hosted runner administrator.

[nikola-jokic]

What is the debug output of it? I tested this change myself multiple times, so I'm wondering how to reproduce this issue.

[vvanouytsel]

What is the debug output of it? I tested this change myself multiple times, so I'm wondering how to reproduce this issue.

I've split it into its own issue.
#257

It is related to permissions and can be reproduced by using a 'container' block that does not run as root.