Currently, the ArgoCD integration does not work well for clusters that are known to be off somewhat frequently.
To handle this, it would be nice to either:
  • Adjust the ArgoCD integration, add handling that will silently continue when a request fails (or with shorter backoff), or
  • Add handling to the Ocean framework that will either: Silently continue, Reduce the number of retries, or Reduce the backoff period when a request fails