Logs on scaleSet with k8s mode

[FR3DD221]

Why are you starting this discussion?

Question

What GitHub Actions topic or product is this about?

General

Discussion Details

I have an k8s cluster where i have an arc implementation, i used to use the legacy versión, but recently i upgraded to 0.13.0 version, i am using the k8s mode in that new implementation, but i want to know if is possible to setup in some way the initialize process logs for my gh workflows, like the legacy version, i know that is related to the dind mode, but i dont use that anymore.

So, basically my question is, exist a way of enable logs for the initialize proccess on the gh UI for my workflows?
Tbh, the logs that im looking for, are the docker related one, like the image thats being pulled/used or the digest (i know this is posible with the debug mode that gh provides, but i want something more permanent)

[rafaelxo]

Since you moved to the new ARC version (autoscaling runner set) and are using the Kubernetes mode without DinD, the behavior of 'Set up runner' logs has changed.

In the new architecture, the runner container is often already prepared or managed differently by the controller. To get those Docker-related logs (like image pull metadata and digests) permanently visible in the GitHub UI without manually enabling step debug logging, you can try the following steps:

1. Runner Scale Set Configuration: Check if your values.yaml for the runner scale set has the DEBUG environment variable enabled. You can set ACTIONS_STEP_DEBUG: true as a default environment variable for the runners in your configuration. This makes the debug-level information (which includes the initialization logs) permanent for all jobs.
2. Controller Logs: Since you are in K8s mode, the actual pod orchestration logs (the 'pulling image' event) stay at the Kubernetes level. You can use a log aggregator (like Fluentd or ELK) to forward your pod events to a central place if you need a permanent record outside the GH UI.
3. Check Workflow Hooks: Some users use containerMode: kubernetes and find that adding a small 'pre-step' to output environment info helps, but for the actual 'Initialize' process, setting the environment variable in the Helm chart is usually the most 'permanent' way to force those logs to appear.

If this helps solve your problem, please consider marking it as the answer to help other community members find this solution more easily!

[github-actions[bot]]

🕒 Discussion Activity Reminder 🕒

This Discussion has been labeled as dormant by an automated system for having no activity in the last 60 days. Please consider one the following actions:

1️⃣ Close as Out of Date: If the topic is no longer relevant, close the Discussion as out of date at the bottom of the page.

2️⃣ Provide More Information: Share additional details or context — or let the community know if you've found a solution on your own.

3️⃣ Mark a Reply as Answer: If your question has been answered by a reply, mark the most helpful reply as the solution.

Note: This dormant notification will only apply to Discussions with the Question label. To learn more, see our recent announcement.

Thank you for helping bring this Discussion to a resolution! 💬