Merge pull request #102739 from Karishma-Tiwari-MSFT/patch-142

(ForumsTeam) fixes MicrosoftDocs/azure-docs#47082

Author: PRMerger18

articles/aks/use-multiple-node-pools.md

@@ -393,7 +393,7 @@ For more information on how to use advanced Kubernetes scheduled features, see [
 In this example, apply a taint to your GPU-based node using the --node-taints command. Specify the name of your GPU-based node from the output of the previous `kubectl get nodes` command. The taint is applied as a *key:value* and then a scheduling option. The following example uses the *sku=gpu* pair and defines pods otherwise have the *NoSchedule* ability:
 
 ```console
-az aks nodepool --node-taints aks-gpunodepool-28993262-vmss000000 sku=gpu:NoSchedule
+az aks nodepool add --node-taints aks-gpunodepool-28993262-vmss000000 sku=gpu:NoSchedule
 ```
 
 The following basic example YAML manifest uses a toleration to allow the Kubernetes scheduler to run an NGINX pod on the GPU-based node. For a more appropriate, but time-intensive example to run a Tensorflow job against the MNIST dataset, see [Use GPUs for compute-intensive workloads on AKS][gpu-cluster].