Update troubleshooting.md

added updates per paul's comments

Author: alvinli222

articles/aks/troubleshooting.md

@@ -135,13 +135,24 @@ To help diagnose the issue run `az aks show -g myResourceGroup -n myAKSCluster -
 
 This error is due to the requested operation being blocked by a PodDisruptionBudget (PDB) that has been set on the deployments within the cluster. To learn more about how PodDisruptionBudgets work, please visit check out [the official Kubernetes example](https://kubernetes.io/docs/concepts/workloads/pods/disruptions/#pdb-example).
 
-If you'd like to proceed, either configure your PDB to be less restrictive by reducing the minAvailable pod count, or increasing the maxUnavailable pod count, or deleting the PDB completely before proceeding with the operation.
-
 You may use this command to find the PDBs applied on your cluster:
 
 ```
-kubectl get poddisruptionbudgets
+kubectl get poddisruptionbudgets --all-namespaces
+```
+or
+```
+kubectl get poddisruptionbudgets -n {namespace of failed pod}
 ```
+Please view the label selector to see the exact pods that are causing this failure.
+
+There are a few ways this error can occur:
+1. Your PDB may be too restrictive such as having a high minAvailable pod count, or low maxUnavailable pod count. You can change it by updating the PDB with less restrictive.
+2. During an upgrade, the replacement pods may not be ready fast enough. You can investigate your Pod Readiness times to attempt to fix this situation.
+3. The deployed pods may not work with the new upgraded node version, causing Pods to fail and fall below the PDB.
+
+>[!NOTE]
+  > If the pod is failing from the namespace 'kube-system', please contact support. This is a namespace managed by AKS.
 
 For more information about PodDisruptionBudgets, please check out the [official Kubernetes guide on configuring a PDB](https://kubernetes.io/docs/tasks/run-application/configure-pdb/).