Search <book_title>...

Important Update: Cohesity Products Documentation

All Cohesity product documentation are now managed via the Cohesity Docs Portal: https://docs.cohesity.com/HomePage/Content/home.htm. Some documentation available here may not reflect the latest information or may no longer be accessible.

Arctera InfoScale™ for Kubernetes 8.0.410 - Linux

Last Published: 2025-08-21

Product(s): InfoScale & Storage Foundation (8.0.410)

Platform: Linux

Removing nodes from an existing cluster

The remove node operation is supported only for shared storage.

It is not supported in the following cases:

When other nodes (excluding those targeted for removal) in the cluster are in the 'out of cluster' state.
During an upgrade.
In the FSS environment.

Note:

When performing this operation on OpenShift, replace kubectl with oc in the following procedure and run the commands on the bastion node instead of master node.

Complete the following steps to remove nodes from an existing InfoScale cluster:

Ensure that all the nodes (excluding the nodes that are being removed) must be in a Ready state. Run the following command on the master node to check whether the nodes are ready.
kubectl get nodes
Ensure that nodes not being removed are in a Joined state. Run the following command on the master node and then review the output.
kubectl describe <Name of the InfosScale Cluster> -n <Namespace>
For example, if worker-node-3 is planned for removal, verify that worker-node-1 and worker-node-2 are in the Joined state by checking their Role status:
```
Cluster Name: <Name of the InfosScale Cluster>    
	  Cluster Nodes: 
	    Node Name:    worker-node-1 
	    Role:         Joined,Master 
	    Node Name:    worker-node-2 
	    Role:         Joined,Slave 
	    Node Name:    worker-node-3 
	    Role:         Joined,Slave 
```

The InfoScale cluster must be in a running state. Run the following command on the master node to verify.

kubectl get infoscalecluster -n <Namespace>

See the State in the output similar to the following:

NAME           VERSION   CLUSTERID      STATE     DISKGROUPS   STATUS    AGE
<Name of       8.0.410   <Cluster ID>   Running   <DG Name>    <Status>  3d13h
the InfoScale                                    
Cluster>

Edit the clusterInfo section of the sample /infoscale-yamls-v8.0.410/kubernetes/cr.yaml to remove information about the nodes that are planned for removal.
Run the following command on the master node to initiate the remove node operation.
kubectl apply -f /infoscale-yamls-v8.0.410/kubernetes/cr.yaml
You can run the following command on the master node to check the InfoScale cluster state.
kubectl get infoscalecluster -n <Namespace>
See the state in the output as displayed below. ProcessingRemoveNode indicates the node is getting removed.
```
NAME      VERSION  CLUSTERID     STATE                 DISKGROUPS  STATUS   AGE 
<Name     8.0.410  <Cluster ID>  ProcessingRemoveNode  <DG Name>   Healthy  3d13h
of the               
InfosScale           
Cluster>
```
Run the following command to verify if pods are being removed. It may take some time for the pods to get deleted.
kubectl get pods -n <Namespace> -o wide

Run the following command on the master node to verify if the cluster is 'Running' & 'Healthy'.

kubectl get infoscalecluster -n <Namespace>

Verify the output:

NAME                VERSION   CLUSTERID     STATE     DISKGROUPS  STATUS    AGE
<Name of the 
InfoScale cluster>  8.0.410   <Cluster ID>  Running   <DG Name>   Healthy   3d13h