Why are Istio components missing?

Virtual Private Cloud Classic infrastructure

One or more of the Istio control plane components, such as istiod, does not exist in your cluster.

Review the following possible causes:

You deleted one of the Istio deployments that is installed in your cluster Istio managed add-on.
You changed the default IstioOperator (IOP) resource. When you enable the managed Istio add-on, you can't use IstioOperator (iop) resources to customize the Istio control plane installation. Only the IstioOperator resources that are managed by IBM for the Istio control plane are supported. Changing the control plane settings might result in an unsupported control plane state. If you create an IstioOperator resource for custom gateways in your data plane in Istio version 1.23 and earlier, you are responsible for managing those resources.

To verify the control plane components installation:

Check the values of any customizations that you specified in the customization ConfigMap. For example, if the value of the istio-components-pilot-requests-cpu setting is too high, control plane components might not be scheduled.
```
kubectl describe cm managed-istio-custom -n ibm-operators
```sh
{: pre}
```
Check the logs of the Istio operator. If any logs indicate that the operator is failing to reconcile your customization settings, verify your settings for the customized Istio installation again.
```
kubectl logs -n ibm-operators -l name=managed-istio-operator
```
Optional: To refresh your managed-istio-custom ConfigMap resource, delete the ConfigMap from your cluster. After about 5 minutes, a default ConfigMap that contains the original installation settings is created in your cluster. The Istio operator reconciles the installation of Istio to the original add-on settings, including the core components of the Istio control plane.
```
kubectl delete cm managed-istio-custom -n ibm-operators
```
If Istio control plane components or other Istio resources are still unavailable, refresh the cluster master.
```
ibmcloud ks cluster master refresh -c <cluster_name_or_ID>
```