[BUG] koordlet collect evicted pod PSI metric

[j4ckstraw]

What happened:
when there are many evicted pods, koordlet will OOM frequently. koordlet logs:

E0417 09:41:06.477243  202881 performance_collector_linux.go:329] collect pod <namespace>/<pod> psi err: open /host-cgroup/cpuacct/kubepods.slice/kubepods-besteffort.slice/kubepods-besteffort-pod3833b19a_b26c_4225_a6a9_6dd35d935185.slice/cpu.pressure: no such file or directory
E0417 09:41:06.477249  202881 performance_collector_linux.go:329] collect pod <namespace>/<pod> psi err: open /host-cgroup/cpuacct/kubepods.slice/kubepods-besteffort.slice/kubepods-besteffort-pod2dca0a37_d138_422d_bd97_10035123c003.slice/cpu.pressure: no such file or directory
E0417 09:41:06.477257  202881 performance_collector_linux.go:329] collect pod <namespace>/<pod> psi err: open /host-cgroup/cpuacct/kubepods.slice/kubepods-besteffort.slice/kubepods-besteffort-podb4cb6360_ab54_4b26_abc5_69c1a3335fee.slice/cpu.pressure: no such file or directory
E0417 09:41:06.477258  202881 performance_collector_linux.go:329] collect pod <namespace>/<pod> psi err: open /host-cgroup/cpuacct/kubepods.slice/kubepods-besteffort.slice/kubepods-besteffort-pode0b448ea_ef12_48e2_9db0_39e72fd7f719.slice/cpu.pressure: no such file or directory

What you expected to happen:
only collect running pod psi metric

How to reproduce it (as minimally and precisely as possible):

Anything else we need to know?:

Environment:

• App version:
• Kubernetes version (use kubectl version):
• Install details (e.g. helm install args):
• Node environment (for koordlet/runtime-proxy issue):
  • Containerd/Docker version:
  • OS version:
  • Kernal version:
  • Cgroup driver: cgroupfs/systemd
• Others:

[j4ckstraw]

we can check pod with IsPodTerminal, or more precise with IsPodTerminated

[songtao98]

/assign