kubeovn / kube-ovn / 24141905472

25%

push

github

fix(controller): clean up orphaned LBHCs during multi-SLR deletion (#6596)

* fix(controller): clean up orphaned LBHCs during multi-SLR deletion

When multiple SwitchLBRules sharing backends are deleted concurrently,
the service handler may remove the LB→LBHC reference before the SLR
handler runs. This leaves orphaned LBHCs (no LB references them) that
the VPC-scoping logic fails to recognize, preventing VIP cleanup and
blocking subnet deletion indefinitely.

Treat an LBHC with zero LB references as belonging to the current VPC
when its subnet matches the SLR's own subnet, so it gets properly
cleaned up. Also harden the E2E AfterEach to defensively delete
remaining SLRs before subnet removal.

Signed-off-by: Mengxin Liu <liumengxinfly@gmail.com>
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Signed-off-by: Mengxin Liu <liumengxinfly@gmail.com>

* fix(e2e): address review feedback on AfterEach SLR cleanup

- Log error when SLR List fails instead of silently skipping cleanup
- Assert WaitToDisappear success to surface stuck SLR failures early
- Filter SLRs by both name suffix and Spec.Namespace to narrow blast
  radius on cluster-scoped resources

Signed-off-by: Mengxin Liu <liumengxinfly@gmail.com>
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Signed-off-by: Mengxin Liu <liumengxinfly@gmail.com>

---------

Signed-off-by: Mengxin Liu <liumengxinfly@gmail.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

8 of 8 new or added lines in 1 file covered. (100.0%)

13751 of 56156 relevant lines covered (24.49%)

0.29 hits per line