Here is the status of one of my nodes in terms of allocations (based on requests)

Allocated resources:
  (Total limits may be over 100 percent, i.e., overcommitted.)
  Resource                   Requests    Limits
  --------                   --------    ------
  cpu                        374m (4%)   3151m (39%)
  memory                     493Mi (1%)  1939Mi (7%)
  ephemeral-storage          0 (0%)      0 (0%)
  attachable-volumes-gce-pd  0           0

Despite the low usage I would expect it to be scaled down by the Cluster Autoscaler (enabled).

However it is not.

Here are the pods running

Non-terminated Pods:         (7 in total)
  Namespace                  Name                                                              CPU Requests  CPU Limits  Memory Requests  Memory Limits  AGE
  ---------                  ----                                                              ------------  ----------  ---------------  -------------  ---
  extra-services             external-dns-cfd4bb858-fvpfj                                      0 (0%)        0 (0%)      0 (0%)           0 (0%)         149m
  istio-system               istio-galley-65987fccb-prxk6                                      10m (0%)      0 (0%)      0 (0%)           0 (0%)         121m
  istio-system               istio-policy-76ddd9fc97-pkxhh                                     110m (1%)     2 (25%)     128Mi (0%)       1Gi (3%)       149m
  kube-system                fluentd-gcp-v3.2.0-7mndl                                          100m (1%)     1 (12%)     200Mi (0%)       500Mi (1%)     5h20m
  kube-system                kube-proxy-gke-my-node-name   100m (1%)     0 (0%)      0 (0%)           0 (0%)         5h20m
  kube-system                metrics-server-v0.3.1-8675cc4d57-xg9qt                            53m (0%)      148m (1%)   145Mi (0%)       395Mi (1%)     120m
  kube-system                prometheus-to-sd-n2jfq                                            1m (0%)       3m (0%)     20Mi (0%)        20Mi (0%)      5h20m

and here are my daemonsets:

➢  k get ds --all-namespaces
NAMESPACE     NAME                       DESIRED   CURRENT   READY   UP-TO-DATE   AVAILABLE   NODE SELECTOR                                  AGE
kube-system   fluentd-gcp-v3.2.0         14        14        14      14           14          beta.kubernetes.io/fluentd-ds-ready=true       226d
kube-system   metadata-proxy-v0.1        0         0         0       0            0           beta.kubernetes.io/metadata-proxy-ready=true   226d
kube-system   nvidia-gpu-device-plugin   0         0         0       0            0           <none>                                         226d
kube-system   prometheus-to-sd           14        14        14      14           14          beta.kubernetes.io/os=linux                    159d

Why isn't the node scaling down?

edit: This is what I get when trying to drain the node manually:

cannot delete Pods with local storage (use --delete-local-data to override): istio-system/istio-policy-76ddd9fc97-pkxhh