Why are scheduling failure events still being reported for the PodGroup when there are no associated pod objects under it?

[guoqinwill]

Please provide an in-depth description of the question you have: Volcano ensures that podgroups and pods share the same lifecycle through the Kubernetes owner reference mechanism. When there is an issue with my kube-manager component or a failure in the GC controller, it can result in pods being deleted while the podgroup remains. In such cases, the podgroup still reports scheduling failure events. Is this normal behavior? I expect that when there are no pods under a podgroup, scheduling failure events should not be reported.

What do you think about this question?: Why are scheduling failure events still being reported for the PodGroup when there are no associated pod objects under it?

Environment:

• Volcano Version:
• Kubernetes version (use kubectl version):
• Cloud provider or hardware configuration:
• OS (e.g. from /etc/os-release):
• Kernel (e.g. uname -a):
• Install tools:
• Others:

[lowang-bh]

Please take a look at the comments at https://github.com/volcano-sh/volcano/pull/3115.