Version 1.12. This version is no longer supported. For more information, see the version support policy. For information about how to upgrade to version 1.13, see Upgrading Anthos on bare metal in the 1.13 documentation.

Available supported versions: 1.28 | 1.16 | 1.15

Edit VM configuration in VM Runtime on Google Distributed Cloud

This document is intended for application owners that run GKE on Bare Metal. This document shows you how to edit virtual machines (VMs) that run using VM Runtime on Google Distributed Cloud. For example, you can edit resource allocation like CPU and memory, or change the network a VM connects to.

You can modify any field in the spec section of a VirtualMachine resource. You can also edit the labels for a VM. Other fields, such as the VM name (metadata.name), can't be edited. Your VM must be in a Stopped state before you edit the resource.

If your edits to the VirtualMachine resource contain errors when you save them, the changes are rejected and you see a notification. Correct the error(s) and try to save the VirtualMachine resource again. If your VM doesn't start after you make changes, use the kubectl describe gvm VM_NAME command to view troubleshooting information and correct the error(s).

Before you begin

To complete this document, you need access to the following resources:

A VM that runs in one of your clusters. If needed, create a VM in GKE on Bare Metal.
The virtctl client tool installed as a plugin for kubectl. If needed, install the virtctl client tool.

Edit compute resources

If your compute workload demands change, you can update the number of virtual CPUs and amount of virtual memory you assign to your VMs. To edit your compute workloads, complete the following steps:

Stop the VM that you want to edit:
```
kubectl virt stop VM_NAME
```
Replace VM_NAME with the name of the VM that you want to stop.
Use kubectl to edit the VM:
```
kubectl edit gvm VM_NAME
```
Replace VM_NAME with the name of the VM that you want to edit.
In your editor, update the compute resource values that you want to change.

For example, the following sample VirtualMachine manifest shows that the VM resource has 2 vCPUs assigned:
```
apiVersion: vm.cluster.gke.io/v1
kind: VirtualMachine
metadata:
  name: vm1
spec:
  ...
  compute:
    cpu:
      vcpus: 2
...
```
If you want to update the number of assigned vCPUs, change the value in your editor, as shown in the following example:
```
apiVersion: vm.cluster.gke.io/v1
kind: VirtualMachine
metadata:
  name: vm1
spec:
  ...
  compute:
    cpu:
      vcpus: 4
...
```
Save and close the VM manifest.
Start the VM that you edited:
```
kubectl virt start VM_NAME
```
Replace VM_NAME with the name of the VM that you edited.
Check the STATUS of your VM:
```
kubectl get gvm VM_NAME
```
Make sure that your VM is in a Running state. If your VM requests more compute resources than the host can provide, your VM can't start. If the VM isn't in a Running state, check your VirtualMachine resource manifest and the compute resource availability on the host.

The following example output shows the VM in a Running state:
```
NAME   STATUS    AGE   IP
vm1    Running   1m    192.168.2.72
```
View detailed information about a VM using kubectl describe gvm:
```
kubectl describe gvm VM_NAME
```
Replace VM_NAME with the name of the VM that you edited.

The following example output shows the condensed information for a VM, with a change to the number of vCPUs successfully applied:
```
Name:         vm1
Namespace:    default
Labels:       <none>
Annotations:  <none>
API Version:  vm.cluster.gke.io/v1
Kind:         VirtualMachine
...
Spec:
  Compute:
    Cpu:
      Vcpus:  4
...
```

For more information, see how to create VMs with specific vCPU and memory compute configurations.

Edit disk resources

If your storage requirements change, you can add or remove virtual disks to a VM. To edit the disks attached to your VM, complete the following steps:

Stop the VM that you want to edit:
```
kubectl virt stop VM_NAME
```
Replace VM_NAME with the name of the VM that you want to stop.
Use kubectl to edit the VM:
```
kubectl edit gvm VM_NAME
```
Replace VM_NAME with the name of the VM that you want to edit.

In your editor, update the spec.disks section to attach or detach disks.

For example, the following sample VirtualMachine manifest shows that only a boot disk is attached to the VM:

apiVersion: vm.cluster.gke.io/v1
kind: VirtualMachine
metadata:
  name: vm1
spec:
  ...
  disks:
    - boot: true
      virtualMachineDiskName: vm1-boot-dv
  ...

If you want to add an existing empty disk, update the disk configuration in your editor, as shown in the following example:

apiVersion: vm.cluster.gke.io/v1
kind: VirtualMachine
metadata:
  name: vm1
spec:
  ...
  disks:
    - boot: true
      virtualMachineDiskName: vm1-boot-dv
    - virtualMachineDiskName: DISK_NAME
  ...

Replace DISK_NAME with the name of the disk to attach.

Save and close the VM manifest.
Start the VM that you edited:
```
kubectl virt start VM_NAME
```
Replace VM_NAME with the name of the VM that you edited.
Check the STATUS of your VM:
```
kubectl get gvm VM_NAME
```
Make sure that your VM is in a Running state. If your VM requests a StorageClass or disk assignment that the host can't provide, your VM can't start. If the VM isn't in a Running state, check your VirtualMachine and VirtualMachineDisk resource manifests and the host storage support.

View detailed information about a VM using kubectl describe gvm:

kubectl describe gvm VM_NAME

Replace VM_NAME with the name of the VM that you edited.

The following example output shows the condensed information for a VM, with a change to the attached disks successfully applied:

Name:         vm1
Namespace:    default
Labels:       <none>
Annotations:  <none>
API Version:  vm.cluster.gke.io/v1
Kind:         VirtualMachine
...
Spec:
  Disks:
    Name:  vm1-boot-dv
    Name:  data-disk-01
...

For more information, see how to create and manage disks.

Edit network resources

As your infrastructure changes, you might want to change the network configuration for a VM. For example, you might want to connect the VM to a different virtual network or manually assign an IP address. To edit the network configuration for a VM, complete the following steps:

Stop the VM that you want to edit:
```
kubectl virt stop VM_NAME
```
Replace VM_NAME with the name of the VM that you want to stop.
Use kubectl to edit the VM:
```
kubectl edit gvm VM_NAME
```
Replace VM_NAME with the name of the VM that you want to edit.
In your editor, update the network configuration settings that you want to change.

For example, the following sample VirtualMachine manifest shows that the VM connects to a network named backend-vlan100 and uses DHCP since no IP address is defined:
```
apiVersion: vm.cluster.gke.io/v1
kind: VirtualMachine
metadata:
  name: vm1
spec:
  ...
  interfaces:
    - name: eth0
      networkName: backend-vlan100
      default: true
...
```
If you want to change the network that the VM connects to or assign a manual IP address, update the network configuration in your editor, as shown in the following example:
```
apiVersion: vm.cluster.gke.io/v1
kind: VirtualMachine
metadata:
  name: vm1
spec:
  ...
  interfaces:
    - name: eth0
      networkName: NETWORK_NAME
        ipAddresses:
          - IP_ADDRESS
      default: true
...
```
Replace NETWORK_NAME with the name of the network to connect to, or add the ipAddresses section and specify the IP_ADDRESS for your VM to use.
Save and close the VM manifest.
Start the VM that you edited:
```
kubectl virt start VM_NAME
```
Replace VM_NAME with the name of the VM that you edited.
Check the STATUS of your VM:
```
kubectl get gvm VM_NAME
```
Make sure that your VM is in a Running state. If your VM requests a network connection that the host can't provide, your VM can't start. If the VM isn't in a Running state, check your VirtualMachine resource manifest and the host network support.