Can not delete empty node pool that is using a VM SKU that has limited availability in my region

BrianWrestler-7912 0 Reputation points
2025-07-31T15:11:51.67+00:00

The node pool uses Standard_D2ds_v5 which we no longer have quota for in our region. We have emptied the pool but are unable to delete it due to the quota issue. We are also unable to upgrade AKS cluster due to this issue.

Azure Kubernetes Service
Azure Kubernetes Service
An Azure service that provides serverless Kubernetes, an integrated continuous integration and continuous delivery experience, and enterprise-grade security and governance.
{count} votes

2 answers

Sort by: Most helpful
  1. Jose Benjamin Solis Nolasco 4,986 Reputation points
    2025-07-31T15:20:50.5033333+00:00

    Hi Brian,

    Thanks for reaching out — I understand you're unable to delete an empty node pool in your AKS cluster because it's using a VM SKU (Standard_D2ds_v5) that currently has no quota in your region. This is preventing you from both deleting the node pool and upgrading the cluster.

    Even if a node pool is empty, AKS still tries to validate the VM SKU (for quota and availability) during the deletion or upgrade process. If the SKU is no longer available or has zero quota, this operation fails — which is a known limitation.

    If possible, request a temporary quota increase for the Standard_D2ds_v5 SKU in that region via the Azure portal:

    Go to: Azure Portal > Subscriptions > Usage + quotas

    Search for Standard_D2ds_v5 and request quota for your region.

    Once granted, delete the node pool and optionally reduce the quota again

    Test this script first can help you make a force delete Sometimes using the --no-wait or --force flags via CLI can bypass some validation logic (not always guaranteed). Try:

    az aks nodepool delete \
      --resource-group <your-resource-group> \
      --cluster-name <your-aks-cluster-name> \
      --name <nodepool-name> \
      --no-wait
    

  2. BrianWrestler-7912 0 Reputation points
    2025-08-11T15:00:28.2366667+00:00

    The issue ended up being with subnet settings in the nodepool. Thanks.

    0 comments No comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.