Stop SDDC Manager from deleting a failed appliance deployment for troubleshooting purposes.

Products

• VMware Cloud Foundation
• VMware SDDC Manager
• VCF - MCOE - SDDC

Issue/Introduction

By default, SDDC Manager deletes deployed virtual machines if a workflow fails. This makes it difficult to perform live troubleshooting or collect logs for deployment and upgrade issues.

Environment

VMware Cloud Foundation 5.x/9.x 

Resolution

The deployment of appliances typically takes place either during Workload Domain (WLD) creation or during vCenter upgrades/patching via Lifecycle Management.

Scenario A: Prevent Appliance Deletion During WLD Creation

1. Edit or create file

vi /etc/vmware/vcf/domainmanager/application-prod.properties

2. Add the following line to the file.

orchestrator.task.stopOnFirstFailure=true

3. Save the changes and exit

:wq

4, Cycle the domainmanager service

systemctl restart domainmanager

5. Retry the vCenter deployment.

6. Troubleshoot the vCenter deployment issue/collect vCenter logs.

7. Remove the flag from the application-prod.properties.

vi /etc/vmware/vcf/domainmanager/application-prod.properties
orchestrator.task.stopOnFirstFailure=true

8. Cycle the domainamanger service

systemctl restart domainmanager

9. Run the workflow again.

10. Once the required logs/troubleshooting steps have been performed make sure to remove the line from the application-prod.properties file.

 

Scenario B: Prevent Appliance Deletion During Upgrades

1. Take a snapshot of the SDDC manager or a backup of the application prod properties. 

cp /opt/vmware/vcf/lcm/lcm-app/conf/application-prod.properties /opt/vmware/vcf/lcm/lcm-app/conf/application-prod.properties.bck

2. Edit the application-prod.properties and add the following line: 

vi /opt/vmware/vcf/lcm/lcm-app/conf/application-prod.properties

lcm.vc.upgrade.delete.target.appliance:false

3. Save the changes and exit. 

:wq!

4. Cycle the lcm service

systemctl restart lcm

4. Troubleshoot the vCenter deployment issue/collect vCenter logs.

5. Once the issue is resolved and the upgrade completes successfully, revert the configuration

lcm.vc.upgrade.delete.target.appliance:true

6. Cycle the service

systemctl restart lcm