Graceful Shutdown and Power-On Sequence for VMware Aria Suite Environment
Article ID: 419249
Updated On:
Products
Issue/Introduction
This article details the dependency-based graceful shutdown and subsequent power-on sequence required for a VMware Aria Suite environment. This process is necessary when planning a complete Data Center (DC) shutdown, such as for maintenance or power events.
The sequence must be strictly followed to protect the data integrity of core components, including the embedded databases (PostgreSQL, etcd) used by components like VMware Aria Automation (vRA) and VMware Identity Manager (vIDM).
The general order for shutdown is: Applications > Core Services > Infrastructure.
Environment
The environment components covered include:
- VMware Aria Automation (vRA) Cluster
- VMware Aria Operations (vRealize Operations (vROps))
- VMware Identity Manager (vIDM)
- VMware Aria Suite Lifecycle (ASL / vRealize Suite Lifecycle Manager ((vRSLCM))
- vCenter Server and ESXi Infrastructure
Cause
Potential service disruption upon startup is caused by the incorrect sequencing of power-off and power-on operations for the dependent applications and infrastructure. Applications must be gracefully stopped before their virtual machines (VMs) are powered off, and the startup must be in the reverse order to ensure identity, database, and core infrastructure services are fully operational before the main applications attempt to start. This prevents issues like database cluster failure or a quarantined vIDM node.
Resolution
1. Graceful Shutdown Sequence
Follow the steps below to gracefully shut down the environment:
| Step | Component | Action | Details |
|---|---|---|---|
| 1 | VMware Aria Automation (vRA) Cluster | Gracefully stop the application services. |
Recommended: Use the Power Off day 2 operation from Aria Suite Lifecycle against the vRA cluster. Secondary: SSH to any vRA appliance and run |
| Power off the VM. | Once services are stopped, use Shut Down Guest OS via vCenter for all vRA VMs. | ||
| 2 | vRealize Operations (vROps) | Gracefully stop the cluster services. | Log into the vROps Admin UI and click the "Take Offline" button. Wait for the cluster to report as offline. |
| Power off the VMs. | Use Shut Down Guest OS via vCenter. Power off data nodes first, then the master replica, and finally the master node. | ||
| 3 | VMware Identity Manager (vIDM) / Workspace ONE Access | Gracefully stop the application services. |
Recommended: Use the Power Off day 2 operation from Aria Suite Lifecycle from the global environment card. Secondary: Perform a graceful service shutdown of all vIDM nodes (e.g., stopping |
| Power off the VMs. | Use Shut Down Guest OS via vCenter for all vIDM VMs. | ||
| 4 | VMware Aria Suite Lifecycle (LCM) | Power off the VM. | Use Shut Down Guest OS via vCenter for the LCM VM. |
| 5 | vCenter Server | Power off the VM. | Shut down the vCenter Server VM. |
| 6 | Infrastructure | Power off the infrastructure. | Shut down the ESXi hosts and underlying storage (like vSAN). |
2. Power-On Sequence (Reverse Order)
The startup sequence is the exact reverse of the shutdown order, ensuring dependencies are met:
| Step | Component | Action | Verification |
|---|---|---|---|
| 1 | Infrastructure | Power on all ESXi Hosts and vSAN/storage. | Confirm storage is accessible and datastores are mounted. |
| 2 | vCenter Server | Power on the vCenter VM. | Wait until the vCenter UI is fully accessible. |
| 3 | VMware Aria Suite Lifecycle (LCM) | Power on the LCM VM. | Wait until the LCM UI is fully accessible. |
| 4 | VMware Identity Manager (vIDM) | Power on all vIDM nodes. |
Recommended: Use the Power On operation from Aria Suite Lifecycle. |
| 5 | vRealize Operations (vROps) | Power on the VMs (master first). | Power on the vROps master, then the master replica, and finally the data nodes. Log into the vROps Admin UI and click "Bring Online". |
| 6 | VMware Aria Automation (vRA) Cluster | Power on all vRA appliance VMs. |
Recommended: Use the Power On day 2 operation from Aria Suite Lifecycle. Secondary: Power on all vRA VMs, then SSH to any node and run the startup script: |
| Service verification. | Wait for the vRA cluster to stabilize. Check: kubectl get pods --all-namespaces should show all services as Running or Completed. |
Additional Information
Feedback
