Rebooting the UCP Advisor VMs

Unified Compute Platform (UCP) Advisor Administration Guide

Version

4.5.x

Audience

anonymous

Part Number

MK-92UCP119-13

ft:lastEdition

2024-03-11

If you must reboot the UCP Advisor VMs for maintenance purposes, perform the following steps to reboot them safely.

Note:

Use the sudo command if you are rebooting the UCP Advisor VMs as a non-root user.
If you must shut down the UCP Advisor VMs, you can do so safely by running the following command on the Worker Node VMs first, and then running it on the Master Node VM:
```
/opt/ekco/shutdown.sh && shutdown now
```

Using SSH, log on to a Worker Node VM, and run the following command:
```
/opt/ekco/shutdown.sh && reboot now
```
Repeat the previous step for the other Worker Node VM, and then again for the Master Node VM.
After the VMs are powered on, verify the status of the UCP Advisor pods.
1. Using SSH, log on to the Master Node VM as a ucpadmin user, and then run the following command:
```
kubectl get pods -n ucp
```
2. Verify that the status of the pods are Running or Completed.
3. If the status of the UCP Advisor-kong pod is Init:0/1, then restart the UCP Advisor-kong pod. Enter:
```
kubectl rollout restart -n ucp deployment/ucpadvisor-kong
```
4. Wait for at least one minute, and then verify the status of the pods. Enter:
```
kubectl get pods -n ucp
```
  The status of the pods must be Running or Completed.
In a new browser window, log on to UCP Advisor, and then verify the status of each UCP system.
If the status of a UCP system is Failed, then do one of the following:
- Wait for at least fifteen minutes for the UCP systems to recover.
- Click the Refresh System icon on the Summary tab to refresh the status information.
- Wait for the UCP systems status to become Normal before you perform any UCP-related operations.