Autoscaler 1.25 or later: If a node fails to be deleted, the lastScaleDownFailTime is not refresh.

[yaohuatj]

Which component are you using?:

cluster-autoscaler

What version of the component are you using?:
autoscaler 1.25

cluster-autoscaler-1.25.0

Component version:

What k8s version are you using (kubectl version)?:

kubectl version Output

$ kubectl version

What environment is this in?:
hws

What did you expect to happen?:
If a node fails to be deleted, the lastScaleDownFailTime will refresh.

What happened instead?:

If the go routine fails to delete a node, the error is not detected and the function still returns nil. Then the lastScaleDownFailTime is not refresh.

This indicates that the scale-down-delay-after-failure parameter does not take effect, but the scale-down-delay-after-delete parameter takes effect.

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

Anything else we need to know?:

[tarishij17]

/assign

[tarishij17]

I don't see the attached function in here: https://github.com/kubernetes/autoscaler/blob/master/cluster-autoscaler/core/scaledown/actuation/actuator.go#L96
Can you please help me locate which go routine is referred here?

[yaohuatj]

I don't see the attached function in here: https://github.com/kubernetes/autoscaler/blob/master/cluster-autoscaler/core/scaledown/actuation/actuator.go#L96 Can you please help me locate which go routine is referred here?

these two function will never collect errors, even if the node fails to be deleted.

[Shubham82]

/triage accepted

[Bharadwajshivam28]

Hey @gjtempleton Can i work on this issue?

[Bharadwajshivam28]

@towca @gjtempleton Hey can I work on this?

[avirlrma]

/assign