Troubleshooting your Local Runner

You may find yourself:

• Troubleshooting problems with your AI Teammates

• Troubleshooting problems with your local runner

• Wanting to check the health of your local runner

The often boil down to checking the health of your local runner and in some cases, restarting certain deployments. In this doc, you'll learn how to:

1. Check the health of your local runner

2. Restart deployments

We'll start by reviewing how to do this from the Kubiya web app. If you're interested in doing this manually from Kubernetes (kubectl), jump to this section for our guide on that.

From Kubiya Web App

Checking runner health

1. Go to the Runners page

2. Next to each runner, you will see it's overall health, which is the aggregate health of all deployments. If any of its deployments are unhealthy, the health will appear is Unhealthy . If it says Healthy, that means all deployments are healthy.

3. To see the health of specific deployments, select a runner

4. In the right-side panel you will see additional information about your it, including the specific deployments and their health

Restarting deployments

1. Go to the Runners page

2. Click on a specific runner

3. In the right-side panel, next to each deployment is a Restart button.

4. Click the Restart button and the deployment will restart

From Kubernetes (kubectl)

1. Run the following command :

$ kubectl get deployments -n kubiya

1. Locate your runner in one of the deployments ( the runner name is kubiya-operator ).

2. A runner with a healthy status should have as follows:

   NAME                                 READY   UP-TO-DATE   AVAILABLE   AGE
   kubiya-operator                       1/1     1            1           15d

Check the runner deployment logs

1. Stream the logs from the runner deployment

   $ kubectl logs -f deployment/kubiya-operator -n kubiya   
   2024/04/19 15:27:20 INFO waiting for messages...
   2024/04/19 15:27:20 INFO subscribe subject=kubiya-solutions-engineering.dev-cluster.>
   
   

2. In case you are seeing an error in the streamed logs try to rollout restart the deployment and repeat steps 4-5 and 1.

   # example error
   ---------------
   $ kubectl logs -f deployment/kubiya-operator -n kubiya   
   2024/04/18 17:14:40 ERROR failed to connect to NATS with credentials error="nats: Authorization Violation"
   
   # rollout restart
   -----------------
   $ kubectl rollout restart deployment kubiya-operator -n kubiya
   deployment.apps/kubiya-operator restarted