Fix MNG scale command missing $

aws-samples · Aug 30, 2024 · 26dba28 · 26dba28
1 parent 3e5ab4a
commit 26dba28
Show file tree

Hide file tree

Showing 2 changed files with 12 additions and 19 deletions.
diff --git a/website/docs/fundamentals/managed-node-groups/basics/adding-nodes/index.md b/website/docs/fundamentals/managed-node-groups/basics/adding-nodes/index.md
@@ -14,7 +14,7 @@ $ eksctl get nodegroup --name $EKS_DEFAULT_MNG_NAME --cluster $EKS_CLUSTER_NAME
 We'll scale the nodegroup in `eks-workshop` by changing the node count from `3` to `4` for **desired capacity** using below command:
 
 ```bash
-aws eks update-nodegroup-config --cluster-name $EKS_CLUSTER_NAME \
+$ aws eks update-nodegroup-config --cluster-name $EKS_CLUSTER_NAME \
   --nodegroup-name $EKS_DEFAULT_MNG_NAME --scaling-config minSize=4,maxSize=6,desiredSize=4
 ```
 
@@ -24,29 +24,21 @@ After making changes to the node group it may take up to **2-3 minutes** for nod
 $ eksctl get nodegroup --name $EKS_DEFAULT_MNG_NAME --cluster $EKS_CLUSTER_NAME
 ```
 
-To wait until the node group update operation is complete you can run this command:
-
-```bash hook=wait-node
-$ aws eks wait nodegroup-active --cluster-name $EKS_CLUSTER_NAME --nodegroup-name $EKS_DEFAULT_MNG_NAME
-```
-
-Once the command above completes we can review the changed worker node count with following command, which lists all nodes in our managed node group by using the label as a filter:
+Monitor the nodes in the cluster using the following command with the `--watch` argument until there are 4 nodes:
 
 :::tip
 It can take a minute or so for the node to appear in the output below, if the list still shows 3 nodes be patient.
 :::
 
-```bash
-$ kubectl get nodes -l eks.amazonaws.com/nodegroup=$EKS_DEFAULT_MNG_NAME
-NAME                                          STATUS     ROLES    AGE     VERSION
-ip-10-42-104-151.us-west-2.compute.internal   Ready      <none>   2d23h   vVAR::KUBERNETES_NODE_VERSION
-ip-10-42-144-11.us-west-2.compute.internal    Ready      <none>   2d23h   vVAR::KUBERNETES_NODE_VERSION
-ip-10-42-146-166.us-west-2.compute.internal   NotReady   <none>   18s     vVAR::KUBERNETES_NODE_VERSION
-ip-10-42-182-134.us-west-2.compute.internal   Ready      <none>   2d23h   vVAR::KUBERNETES_NODE_VERSION
+```bash hook=wait-node
+$ kubectl get nodes --watch
+NAME                                          STATUS     ROLES    AGE  VERSION
+ip-10-42-104-151.us-west-2.compute.internal   Ready      <none>   3h   vVAR::KUBERNETES_NODE_VERSION
+ip-10-42-144-11.us-west-2.compute.internal    Ready      <none>   3h   vVAR::KUBERNETES_NODE_VERSION
+ip-10-42-146-166.us-west-2.compute.internal   NotReady   <none>   18s  vVAR::KUBERNETES_NODE_VERSION
+ip-10-42-182-134.us-west-2.compute.internal   Ready      <none>   3h   vVAR::KUBERNETES_NODE_VERSION
 ```
 
-Notice that the node shows a status of `NotReady`, which happens when the new node is still in the process of joining the cluster. We can also use `kubectl wait` to watch until all the nodes report `Ready`:
+Once 4 nodes are visible you can exit the watch using `Ctrl+C`.
 
-```bash hook=add-node
-$ kubectl wait --for=condition=Ready nodes --all --timeout=300s
-```
+You may see a node shows a status of `NotReady`, which happens when the new node is still in the process of joining the cluster.
diff --git a/website/docs/fundamentals/managed-node-groups/basics/adding-nodes/tests/hook-wait-node.sh b/website/docs/fundamentals/managed-node-groups/basics/adding-nodes/tests/hook-wait-node.sh
@@ -14,6 +14,7 @@ after() {
 
   if [ $EXIT_CODE -ne 0 ]; then
     >&2 echo "Node count did not increase to 4 as expected"
+    kubectl get node
     exit 1
   fi
 }