VMware Cloud Foundation 9.1: Safely Shutting Down an Entire VCF Services Runtime Cluster

With VMware Cloud Foundation 9.1, Broadcom introduced a new automated method for safely shutting down all nodes within a VCF Services Runtime cluster. While this may sound like a small operational improvement, anyone who has ever performed a full management domain maintenance window knows how important a coordinated shutdown process can be.

As more VCF services are delivered through the VCF Services Runtime platform, traditional „power off the VMs“ approaches become increasingly risky. Kubernetes workloads, platform services, databases, and supporting components all have dependencies that should be handled gracefully.

To address this, Broadcom now provides a supported shutdown utility that orchestrates the entire process.

Why a Graceful Shutdown Matters

Modern VMware Cloud Foundation environments contain multiple interconnected services that continuously exchange data and maintain cluster state.

An unplanned or improperly coordinated shutdown can potentially result in:

Incomplete database transactions
Kubernetes workload inconsistencies
Service startup delays
Failed platform services after reboot
Additional troubleshooting during recovery

For planned maintenance windows, power outages, datacenter migrations, or infrastructure upgrades, a controlled shutdown process is strongly recommended.

Introducing the VCF Services Runtime Shutdown Utility

Starting with VCF Services Runtime 9.1, administrators can use the following utility:

.\vcf_services_runtime_shutdown.sh

The script performs an orchestrated shutdown of the entire runtime cluster by communicating directly with the VCF Services Runtime APIs.

During execution, the utility:

Performs pre-flight validation checks
Discovers runtime cluster nodes
Identifies active workloads
Handles service dependencies automatically
Preserves recovery information
Coordinates shutdown sequencing
Powers down nodes in the correct order

This significantly reduces the operational risk associated with manual shutdown procedures.

Where to Download the Script

The shutdown utility is provided directly by Broadcom and can be downloaded from the official Knowledge Base article:

Broadcom KB 440874

The article contains:

The latest version of the shutdown script
Detailed prerequisites
Usage examples
Supported command-line parameters
Recovery guidance

After downloading the script, make it executable:

chmod +x vcf_services_runtime_shutdown.sh

Prerequisites

Before executing the script, ensure the following tools are available:

curl
jq
govc

You will also need:

Network access to a VCF Services Runtime Control Plane node
Connectivity to port 5480
Appropriate administrative credentials
Access to the runtime kubeconfig file

The script can be executed from any management workstation that satisfies these requirements.

Performing a Dry Run

Before initiating an actual shutdown, Broadcom recommends validating the environment using the dry-run option.

Example:

./vcf_services_runtime_shutdown.sh \
  --node-ip <CONTROL_PLANE_IP> \
  --dry-run \
  --kubeconfig <kubeconfig-file>

A dry run allows administrators to verify connectivity, permissions, and cluster discovery without impacting running services.

This is especially useful when performing the procedure for the first time or when preparing for a large maintenance event.

Typical Use Cases

The shutdown utility is particularly useful for:

Datacenter power maintenance
UPS replacement projects
Infrastructure migrations
Hardware refresh activities
Full-environment backup operations
Disaster recovery testing
Planned maintenance windows

In each scenario, the automated workflow provides a safer alternative to manually shutting down runtime nodes.

Benefits Compared to Manual Shutdown Procedures

Historically, VMware administrators often relied on lengthy shutdown runbooks that required services to be powered down in a very specific sequence.

These procedures were:

Time-consuming
Error-prone
Difficult to troubleshoot
Dependent on administrator experience

The new Runtime Shutdown Utility simplifies operations considerably.

Traditional Approach	Automated Runtime Shutdown
Manual service sequencing	Fully orchestrated workflow
Dependency tracking required	Dependencies handled automatically
Higher risk of human error	Consistent execution
Longer maintenance windows	Faster operational workflow
More troubleshooting	Predictable recovery process

For organizations operating production VCF environments, this can significantly reduce operational complexity.

Operational Best Practices

Before initiating a shutdown, verify that:

No lifecycle operations are currently running
SDDC Manager tasks have completed
Backup jobs have finished successfully
Active maintenance activities are complete
Stakeholders have been notified

Remember that all services hosted within the VCF Services Runtime cluster will become unavailable during the shutdown process.

Proper communication and planning remain essential.

Recovery Considerations

Broadcom has also published a companion article that explains how to recover services if the VCF Management Services Runtime cluster does not fully recover after startup.

Recovery KB:

Broadcom KB 440862

While most environments should start cleanly after a controlled shutdown, it is always worth bookmarking the recovery documentation before beginning any maintenance activity.

Final Thoughts

The introduction of the vcf_services_runtime_shutdown.sh utility is one of those operational improvements that administrators will immediately appreciate.

Instead of relying on manual runbooks and shutdown sequences, VMware Cloud Foundation 9.1 now provides a supported, automated, and repeatable process for safely shutting down an entire VCF Services Runtime cluster.

If you are operating VCF 9.1 or later, this procedure should become part of your standard maintenance and datacenter shutdown runbook. It reduces operational risk, simplifies maintenance activities, and helps ensure a predictable startup experience when services are brought back online.