kubernetes: DualStack: incorrect service state after pod deletion

What happened: Service IP to pod ip mapping is not getting removed from ipvs rules after corresponding pod deletion.

What you expected to happen:

How to reproduce it (as minimally and precisely as possible): Create a tcp ipv6 loadbalacer service with couple of pods. Then delete pod deployment alone. List ipvs rules using “sudo ipvsadm -Ln”. Even after pod deletion, service ip maps to deleted pod ipv6 ips.

Anything else we need to know?:

Environment:

  • Kubernetes version (use kubectl version): 1.17
  • Cloud provider or hardware configuration: Azure
  • OS (e.g: cat /etc/os-release): ubuntu16.04
  • Kernel (e.g. uname -a):
  • Install tools:
  • Network plugin and version (if this is a network-related bug): kubenet
  • Others:

About this issue

  • Original URL
  • State: closed
  • Created 4 years ago
  • Comments: 28 (19 by maintainers)

Most upvoted comments

@uablrek I’ll try it again and see if I can reproduce. If yes, then I would love to take a stab at the fix. If I can’t by this week, I’ll let you take it 😃