linode-cloud-controller-manager

Troubleshooting

Common Issues and Solutions

CCM Pod Issues

Pod Won't Start

kubectl get pods -n kube-system -l app=ccm-linode
kubectl describe pod -n kube-system -l app=ccm-linode

Common causes:

• Invalid API token
• Missing RBAC permissions
• Resource constraints

Pod Crashes

Check the logs:

kubectl logs -n kube-system -l app=ccm-linode

Common causes:

• API rate limiting
• Network connectivity issues
• Configuration errors

LoadBalancer Service Issues

Service Stuck in Pending

kubectl describe service <service-name>

Check for:

• API token permissions
• NodeBalancer quota limits
• Network configuration

Health Checks Failing

Verify:

• Backend pod health
• Service port configuration
• Health check path configuration

Node Issues

Missing Node Labels

kubectl get nodes --show-labels

Verify:

• CCM node controller logs
• Node annotations
• API permissions

Network Problems

Check:

• Private IP configuration
• VPC/VLAN setup
• Firewall rules

Gathering Information

Useful Commands

# Get CCM version
kubectl get pods -n kube-system -l app=ccm-linode -o jsonpath='{.items[0].spec.containers[0].image}'

# Check events
kubectl get events -n kube-system

# Get CCM logs with timestamps
kubectl logs -n kube-system -l app=ccm-linode --timestamps

Debug Mode

Set the following environment variable in the CCM deployment:

env:
  - name: LINODE_DEBUG
    value: "1"

Getting Help

If issues persist:

1. Join #linode on Kubernetes Slack
2. Check GitHub Issues
3. Submit a new issue with:
   • CCM version
   • Kubernetes version
   • Relevant logs
   • Steps to reproduce

This site is open source. Improve this page.