VMware ESXi and vSphere Cluster Management

Increase the Size of a VMFS Datastore in VMware ESXi

Learn how to expand a VMFS datastore by enlarging its backing LUN or adding an extent, including rescans, vSphere steps, limits, validation, and troubleshooting.

A VMFS datastore is a logical storage container formatted with VMware Virtual Machine File System (VMFS). ESXi uses it to store virtual machine configuration files, virtual disks, snapshots, templates, swap files, and other virtual machine data.

When a datastore approaches its capacity limit, you can usually increase its size without moving powered-on virtual machines. The two supported approaches are to expand the existing VMFS extent or add another extent.

Why Increase VMFS Datastore Capacity?

Datastore growth may be required because the original storage was sized incorrectly, virtual machine disks have been expanded, more virtual machines were added, snapshots consumed unexpected space, or templates and other files increased storage usage.

Increasing datastore capacity is different from reclaiming unused space inside a guest operating system. Expanding a datastore gives VMFS more storage for files. It does not automatically discover and reclaim unused blocks inside a guest filesystem. Guest-space reclamation requires separate guest, virtual-disk, and storage-array processes.

Capacity planning should include expected virtual machine growth, snapshots, virtual machine swap files, templates, ISO images, backups, and operational headroom. Avoid planning to use every available gigabyte because low free space can prevent snapshot operations, disk expansion, migrations, and normal virtual machine activity.

Important Storage Terms

VMFS: VMware's clustered filesystem for ESXi datastores.
LUN: A logical storage device presented by a SAN or storage array to an ESXi host.
Extent: A VMFS partition on a storage device that contributes capacity to a datastore.
Existing extent expansion: Enlarging the device or LUN that already contains a VMFS extent, then extending VMFS into adjacent free space.
Add extent: Adding another eligible device or partition to an existing datastore.
Contiguous free space: Unallocated capacity located immediately after the current VMFS partition on the same device.
Storage rescan: An ESXi discovery operation that refreshes storage adapters, devices, paths, and capacity information.
Multipathing: The use of multiple paths between an ESXi host and a storage device for availability and, where supported, performance.

VMFS Datastore Growth Methods

Method	Storage-layer requirement	VMFS requirement	Result	Typical use case	Operational considerations
Expand existing extent	Increase the size of the existing LUN or device.	Free capacity must be directly after the current VMFS partition.	The existing extent and datastore become larger.	Use when the storage array can enlarge the current LUN.	Usually the simplest layout because the datastore continues to use one extent.
Add datastore extent	Present a new compatible LUN or unused eligible device space.	Create another VMFS extent and attach it to the datastore.	The datastore reports pooled capacity across multiple extents.	Use when the original LUN cannot be enlarged or a separate LUN is required.	Every backing device becomes a dependency, so document the layout and maintain all paths.

Method 1: Expand the Existing VMFS Extent

This method enlarges the LUN or device that already backs the datastore. After the storage array reports the larger device, ESXi can extend the existing VMFS partition and filesystem into the newly available space.

The required sequence is:

Confirm the correct datastore and backing LUN.
Increase the LUN capacity at the storage-array or SAN layer.
Rescan the ESXi storage adapters.
Confirm that ESXi detects the device at its new size and that expected paths remain visible.
Use the vSphere datastore capacity-increase workflow to grow the existing extent.
Review and confirm the operation.
Validate the new datastore capacity, free space, extent mapping, and workload access.

In-place growth requires unallocated capacity to be contiguous with and immediately following the existing VMFS partition on the same device. A larger LUN alone is not sufficient if the partition layout does not expose usable adjacent space.

When the storage platform supports safe LUN expansion, this is generally the preferred method because it avoids adding another datastore dependency.

Method 2: Add a VMFS Datastore Extent

An extent is a VMFS partition that contributes capacity to a datastore. With this method, you present a new compatible LUN or use eligible unused device space, then attach it as an additional extent.

Provision or identify an unused compatible LUN.
Present it consistently to every required ESXi host using correct zoning, masking, and multipathing.
Rescan storage adapters.
Confirm that the new device is visible, unused, and eligible.
Start the datastore capacity-increase action and select the option to add an extent.
Select the intended device or free space.
Review the device, partition, and resulting capacity before confirming.
Verify that the datastore reports the combined capacity of all extents.

Capacity is pooled at the datastore level, but the datastore now spans multiple backing devices. If any required device becomes unavailable, the datastore may be affected. Keep an accurate record of each extent, its LUN identifier, storage array, host presentation, and path configuration.

vSphere Client Workflow

The exact labels can vary slightly by vSphere version and permissions, but the general workflow is consistent.

Open the vSphere Client and go to the datastore inventory view, commonly through Storage or Datastores.
Select the target VMFS datastore. Verify its name, hosts, current capacity, and backing device before continuing.
Open the datastore Actions menu and choose the capacity-increase or similar datastore expansion action.
Review the available choices. The wizard may offer growth of the current extent when a larger device with adjacent free space is detected, an additional extent when an eligible device is available, or both.
Select the intended device and the available capacity. Do not select an unrelated LUN.
Review the summary, including the datastore name, device identifier, partition, current size, new size, and extent layout.
Confirm the operation and wait for the task to complete.
Refresh the datastore view and verify the updated total and free capacity.

Storage Discovery and Rescanning

ESXi cannot use newly provisioned LUN capacity until it discovers the change. A storage rescan refreshes the adapters, devices, paths, and reported capacity. Rescanning is needed both after enlarging an existing LUN and after presenting a new LUN.

In the vSphere Client, select the relevant host or hosts and use the storage adapter rescan action. For shared storage, make sure every host that must access the datastore has refreshed its device information. Also refresh the datastore inventory after the device rescan.

For ESXi Shell verification, use commands appropriate to your permissions and support procedures:

esxcli storage core adapter rescan --all

Inspect discovered devices and their reported capacity:

esxcli storage core device list

Review mounted filesystems and datastore capacity:

esxcli storage filesystem list

Display the devices and partitions backing VMFS extents:

esxcli storage vmfs extent list

These commands help verify discovery and layout; perform the actual datastore growth through the supported vSphere workflow unless your documented environment requires another supported procedure.

Availability and Workload Impact

Virtual machines can normally remain powered on while VMFS datastore capacity is increased. The operation does not require relocating virtual machines solely because the datastore is being resized.

Use an approved maintenance window when possible, especially for production storage. Coordinate with the storage team, confirm monitoring coverage, and avoid combining the resize with unrelated storage changes. A maintenance window provides time to investigate path, array, or inventory problems without increasing operational risk.

Constraints and Limitations

VMFS datastore growth is one-directional. An existing datastore cannot be reduced in place.
Commonly referenced VMFS limits for relevant environments include a maximum of 32 extents and an overall datastore capacity limit of 64 TB.
Limits can depend on the installed ESXi and vCenter versions, VMFS version, storage protocol, device type, and vendor array. Confirm the limits supported by the specific environment before proceeding.
Adding an extent requires an unused compatible LUN or qualifying unallocated device space.
Expanding an existing extent requires free capacity directly after the current VMFS partition.
A datastore with multiple extents depends on every backing device and its storage paths.

Do not assume that a device's advertised size, partition layout, or storage-array capability is supported merely because ESXi can detect the device. Check the VMware compatibility information and storage-vendor guidance for the installed environment.

Pre-Expansion Checklist

Check	Why it matters	How to verify
Correct datastore and backing device identified	Prevents changes to the wrong datastore or LUN.	Compare the datastore name, host mappings, device identifier, LUN number, and array records.
Storage array has sufficient capacity	The array must support the requested LUN growth or new LUN.	Check array free space, health, volume settings, and vendor procedures.
New LUN capacity or new LUN is presented to all applicable hosts	Shared hosts need consistent visibility.	Review zoning, masking, host groups, paths, and device discovery on each host.
Storage rescan completed	ESXi must learn the new device size or new device.	Rescan adapters and inspect device capacity and path visibility.
Contiguous unallocated space exists for in-place growth	VMFS can grow only into adjacent free capacity.	Review the device and partition layout in the vSphere workflow and ESXi output.
No unsupported size or extent limits will be exceeded	VMFS, ESXi, protocol, and array limits may prevent the operation.	Check the installed-version and vendor documentation.
Backup and change controls are in place	Storage changes can affect many workloads.	Confirm recoverable backups, approval, rollback planning, and an implementation window.

Validation After Expansion

Validation item	Expected result	Where to check
Datastore total capacity	The total is larger by the expected amount.	Datastore summary in the vSphere Client and filesystem output.
Datastore free capacity	Free space reflects the newly available capacity.	Datastore summary, capacity charts, and monitoring tools.
Host visibility	Every host requiring access sees the datastore at the revised size.	Hosts, datastores, device paths, and recent tasks in vSphere.
Extent/device mapping	The expected resized extent or new extent is listed.	Datastore device details and `esxcli storage vmfs extent list`.
Virtual machine accessibility	Expected virtual machine files remain accessible and workloads continue normally.	Virtual machine power state, guest operation, datastore browser, and application monitoring.
Storage path health	Expected paths are active and no new path or device alarms appear.	Host storage views, multipathing status, array health, and alarms.

Record the changed LUN size, datastore name, extent mapping, date, change identifier, storage-array details, and hosts affected. This documentation is especially important for multi-extent datastores.

Troubleshooting VMFS Expansion

Expanded LUN Still Appears at Its Old Size

Confirm that the array-side resize was completed and committed. Rescan all relevant ESXi storage adapters, then inspect the device capacity and path visibility. If the size remains unchanged, investigate zoning, masking, multipathing, array presentation, and path health.

The Wizard Does Not Offer Free Space on the Original Device

Possible causes include an unexpanded LUN, free capacity that is not adjacent to the VMFS partition, another partition between VMFS and the free space, or an incompatible device layout. Do not alter existing VMFS partitions without a validated recovery plan. Consider adding a new extent or migrating workloads to a newly provisioned datastore.

A New LUN Cannot Be Selected as an Extent

Verify that the LUN is unused, visible to the selected host, presented to all required hosts, and supported by the configuration. A LUN that is already formatted, partitioned, mounted, or used by another system may not be eligible.

The Datastore Size Did Not Change

Review recent tasks and events for errors. Refresh the datastore view, then compare the vSphere capacity with esxcli storage filesystem list and esxcli storage vmfs extent list. Confirm that the intended LUN was resized and that the expected extent mapping is present.

An Administrator Wants to Shrink the Datastore

VMFS does not support in-place datastore shrinking. Provision a correctly sized destination datastore, migrate or copy the required workloads and files, validate backups and access, and retire the original datastore only after the migration is complete.

Practical Examples

Grow a Single-LUN Datastore

A datastore backed by one LUN is nearly full.
The storage administrator increases the LUN capacity at the array.
ESXi rescans storage and recognizes the larger device.
The administrator grows the existing extent through the datastore capacity-increase workflow.
The datastore gains capacity without adding another extent.

Add Capacity When the Original LUN Cannot Be Enlarged

The current LUN cannot be extended by the storage platform.
A new compatible LUN is presented to all required ESXi hosts.
The hosts rescan storage and discover the new device.
The administrator adds the new LUN as another VMFS extent.
The datastore reports the combined capacity of both extents.

Choose the Appropriate Method

Use in-place expansion when the existing LUN can be enlarged and adjacent free capacity will follow the VMFS partition. Use an additional extent when another LUN is necessary or the original device cannot be grown. Prefer a simple single-extent layout when feasible because it reduces storage-layout complexity and device dependencies.

Summary

To increase a VMFS datastore, either enlarge its existing backing LUN and extend the current extent, or add another compatible LUN as a new extent. Resize the storage at the array first, rescan ESXi, verify device size and paths, complete the vSphere capacity-increase workflow, and validate every affected host and workload. Protect the change with backups, approval, correct device identification, path checks, and accurate storage documentation.