VMware ESXi and vSphere Cluster Management

VMware ESXi Maintenance Mode Explained

Learn what VMware ESXi maintenance mode does, how to evacuate virtual machines, enter it in vSphere, troubleshoot failures, and compare it with standby mode.

ESXi maintenance mode is a host state used to safely prepare a VMware ESXi host for servicing, patching, upgrades, hardware work, or other administrative tasks. An ESXi host is a physical server running the VMware ESXi hypervisor and hosting virtual machines (VMs).

When a host enters maintenance mode, it remains available for management, but it is removed from normal workload use. Running VMs must first be shut down or moved to other eligible hosts. The host then cannot receive newly deployed VMs or power on VMs until it leaves maintenance mode.

What Is ESXi Maintenance Mode?

Maintenance mode is an ESXi host state that removes the host from workload use so maintenance operations can be performed safely. It is commonly used before:

ESXi patches and upgrades
Hardware replacement or inspection
Host configuration changes
Certain installation and update workflows, including live-install workflows that require a clustered host to be in maintenance mode

A host in maintenance mode is still visible and manageable through vCenter Server or another management interface. However, it is not treated as an available destination for normal VM operation. The objective is to create a predictable condition in which the host has no running workload while maintenance is performed.

Why a Host Must Be Drained First

Before maintenance mode can complete, all running VMs must be removed from the host. There are two applicable ways to handle them:

Migrate the VMs to other eligible ESXi hosts.
Shut down the VMs before entering maintenance mode.

Virtual machine migration means moving a VM from one ESXi host to another so the original host can be drained. In a cluster, migration is often preferred because it can preserve application availability, depending on the migration method, VM configuration, storage, networking, licensing, and workload requirements.

Migration is possible only when the environment provides a suitable destination. The destination must have enough CPU, memory, storage access, networking, and other required resources. The hosts and VM must also support the available migration capability, such as vMotion where applicable.

If migration is not possible or the workload can tolerate downtime, shut down the VMs cleanly. A powered-off VM is no longer a running workload on the host, but it may still be registered in inventory.

VM condition	Required action	Result for maintenance-mode entry
Running and suitable for migration	Migrate it to another eligible host	The source host can be drained without shutting down the VM
Running but migration is unavailable or unsuitable	Shut down the VM according to the application requirements	The host can proceed after the VM is no longer running
Already powered off	No shutdown is required; verify that no other running VMs remain	The VM does not prevent maintenance-mode entry

Maintenance Mode and vSphere Clusters

A vSphere cluster is a group of ESXi hosts managed together. Clusters make it possible to move workloads among eligible hosts, but they do not guarantee that every VM can move to every host.

Before evacuating a host, check the cluster's remaining capacity. The other hosts must be able to run the migrated VMs without creating a CPU or memory shortage. Also check that the destination hosts can access the VM's datastores and networks and that cluster policies do not prevent placement.

If the cluster does not have enough capacity, the choices may be to shut down selected VMs, move workloads manually, add capacity, or postpone maintenance. Entering maintenance mode without planning for the displaced workloads can cause failed migrations or unnecessary downtime.

Host Behavior While in Maintenance Mode

Once an ESXi host is in maintenance mode:

It cannot accept newly deployed VMs.
VMs cannot be powered on on that host.
Running VMs should not remain there as the host completes the transition.
The host remains available for management and maintenance operations.
Administrators can perform the planned update, installation, or hardware work.

This distinction is important: a host can be available to administrators while being unavailable to workload placement. Maintenance mode is therefore different from a management outage and different from a powered-off host.

Entering Maintenance Mode in the vSphere Interface

Open the vSphere inventory and locate the ESXi host that requires maintenance.
Review the VMs currently running on that host.
Decide whether to migrate those VMs to other hosts or shut them down.
Confirm that destination hosts have adequate capacity and the required migration access.
Select the target ESXi host.
Open the host context menu and choose Enter Maintenance Mode.
Complete or monitor the evacuation or shutdown actions requested by the vSphere interface.
Wait for the operation to finish before beginning host maintenance.

The operation completes only after the VM requirements have been satisfied. If VMs are still running and cannot be migrated or shut down, the host remains outside maintenance mode or the operation reports a problem that requires administrator action.

After the host is safely in maintenance mode, perform the update or service task. When the work is complete, return the host to normal operation using the appropriate vSphere action, then verify connectivity, storage, networking, cluster membership, and workload placement.

Practical Example: Preparing for an ESXi Update

Identify the ESXi host that requires the update.
Inspect its running VMs and identify suitable destination hosts.
Move each eligible VM to another host, or shut down VMs that cannot be migrated.
Confirm that no running VM still depends on the host.
Use the host context menu and select Enter Maintenance Mode.
Wait until vSphere confirms the host has entered maintenance mode.
Perform the ESXi update while the host is unavailable for VM placement.

If a live-install update workflow requires maintenance mode, the update should not be started until the host has successfully reached that state.

Practical Example: Active Workloads Prevent Entry

Suppose an administrator starts the maintenance-mode operation while several VMs are still running. Those VMs must be addressed before the operation can complete.

If the cluster has sufficient resources and migration is supported, migrate the VMs to eligible hosts.
If a VM cannot be migrated, shut it down if the application and maintenance plan permit downtime.
Retry or continue the maintenance-mode operation after all running VMs have been handled.

Do not assume that a VM can move simply because another host is visible in inventory. Capacity, shared storage or equivalent storage access, network configuration, compatibility, permissions, and migration support all affect the result.

Maintenance Mode vs. Standby Mode

Standby mode is a separate host state in which the ESXi host is powered off. It is a power-saving state rather than a preparation state for keeping the host available to administrators during servicing.

vSphere Distributed Power Management (DPM) is a vSphere capability that can place hosts into standby mode when cluster demand is low, helping reduce power consumption. When more capacity is needed, the environment can bring suitable hosts out of standby according to its configuration and operational policies.

Host state	Primary purpose	Virtual machine placement behavior	Host power state	Relevant vSphere feature
Normal operation	Run and manage workloads	Eligible VMs can be deployed or powered on, subject to capacity and policy	Powered on	Routine vSphere host operation
Maintenance mode	Prepare the host for servicing, updates, or installation work	New VMs cannot be deployed there, and VMs cannot be powered on there	Powered on	Host maintenance workflows; may be used with cluster evacuation features
Standby mode	Reduce power use when host capacity is not needed	The powered-off host cannot run or receive VMs	Powered off	vSphere Distributed Power Management (DPM)

Troubleshooting Maintenance-Mode Problems

The host does not enter maintenance mode

Likely cause: One or more VMs are still running on the host.

Resolution: Review the host's VM list. Shut down the remaining VMs or migrate them to other eligible hosts, then retry entering maintenance mode. If migration fails, investigate destination capacity, datastore access, network requirements, compatibility, and cluster policy restrictions.

A VM cannot be deployed or powered on on the selected host

Likely cause: The host is already in maintenance mode.

Resolution: Select another eligible host for the workload, or return the host to normal service after maintenance is complete. Do not treat a maintenance-mode host as a normal placement target.

A host is unavailable after being placed into standby mode

Likely cause: Standby mode powers off the host as part of power optimization.

Resolution: Determine whether the intended action was maintenance mode or standby mode. Maintenance mode leaves the host powered on for management and servicing; standby mode is a powered-off state. Manage the host's power state according to the maintenance or capacity plan.

Operational Checklist

Identify the correct ESXi host and the reason for maintenance.
List all running VMs on the host.
Confirm that suitable destination hosts have enough capacity.
Verify migration prerequisites before starting evacuation.
Shut down VMs that cannot be migrated and for which downtime is acceptable.
Select Enter Maintenance Mode from the host context menu.
Wait for successful completion before performing maintenance.
After servicing, return the host to normal operation and verify cluster health.
Use standby mode only when powering off the host for power management is the intended result.