Reports

What happens is that the error indicates that updating an unhealthy deployment with a working Docker image doesn't always result in a healthy deployment.

According to this document, despite the error, if you provide a good Docker image, the deployment might eventually become healthy. Updating an existing version with a new Docker image, though allowed, is not a good practice. Additionally, there is currently no rollback in case of version failure.

Here are the things that you might consider as well:

Extend app health check timeouts in app.yaml file. Add a /health endpoint and explicitly configure health checks and startup timeouts.

Update app.yaml file:

runtime: custom
env: flex

# Add health checks with higher timeout   wwwwwwwwwwwwws
readiness_check:
  path: "/health"
  check_interval_sec: 10
  timeout_sec: 4
  failure_threshold: 10
  success_threshold: 1
  app_start_timeout_sec: 300

liveness_check:
  path: "/health"
  check_interval_sec: 10
  timeout_sec: 4
  failure_threshold: 5

Retry the Deployment with a new version ID This ensures no internal state is cached, for instance:

gcloud app deploy --version=20250522

Check here if there were App Engine or GCE issues around May 21, 2025.

If the above suggestions do not help, then it might be observed that the issue is specifically for your project and requires a project inspection. It’s better to contact Google support for the same.

79634431