Job failed (system failure): Internal error occurred - Kubernetes Executor on Runner 15.8.0

Status update 2023-02-08

The bug was introduced by the MR !3751 (merged) (as @DarrenEastman suspected). It occurs during the Uploading artifacts for successful job stage. Runner Manager tries to write the script on the helper container. That is the reason for the job failure.

We are working on a fix MR !3900 (merged)

Summary

After upgrading to Runner 15.8.0 CI jobs on a Kubernetes cluster for building a Docker image started failing with the following error:

ERROR: Job failed (system failure): Internal error occurred: error executing command in container: failed to exec in container: failed to start exec "6c84886ceae6d793c210f623a27567c2aa6d9e20fab607b667cd45d71f13e986": OCI runtime exec failed: exec failed: unable to start container process: chdir to cwd ("/workspace") set in config.json failed: no such file or directory: unknown[0;m

Steps to reproduce

.gitlab-ci.yml

Add the job definition that is failing here

Actual behavior

The CI job to create a container image fails and generates the job failed (system failure) error.

Expected behavior

CI job to create a container image completes with no errors.

Relevant logs and/or screenshots

job log

Add the job log

Environment description

config.toml contents

Add your configuration here