
Automatic merge from submit-queue Fix flaky cluster-autoscaler e2e Ref: https://github.com/kubernetes/autoscaler/issues/89 Add pdb to allow cluster-autoscaler to drain nodes with some kube-system components (turns out there can be enough of them to deny scale-down even with 5 healthy nodes). Increased scaleDownTimeout to take into account the time it will take to re-schedule pods running on broken node (this may reset scale-down timer).