Skip to main content

Deployment Failures

Common issues when environments fail to provision or services don't come up healthy.

Environment Not Creating​

  • Label not recognized — Double-check the deploy label name. It must exactly match the configured label.
  • Webhook delivery — GitHub webhooks can occasionally fail to deliver. Check with your platform team if the environment doesn't appear after a few minutes.
  • YAML syntax errors — Invalid YAML in your config file will cause provisioning to fail. Validate your YAML before adding the label.

Pods Not Starting​

  • Image pull errors — The container image might not exist or greyhound might not have pull access. Check the build status in the PR comment.
  • Resource limits — If pods are being evicted or OOMKilled, increase the resources.limits in your service config.
  • Security context — Some images require specific user/group IDs. Set securityContext in your service config.

Services Not Accessible​

  • Missing ingress rule — Your service needs a rule to be accessible externally.
  • Port mismatch — The rule's target service must expose at least one containerPort.
  • DNS propagation — New hostnames may take a moment to resolve. Try again after a minute.

Environment Stuck​

If an environment is stuck in a provisioning state, contact your platform team. They can inspect the Temporal workflow for details.