Search the VMware Knowledge Base (KB)
View by Article ID
Creating and managing extents on ESX/ESXi (2000643)
For a VMFS volume, there is a rule of one VMFS per LUN. SCSI-2 Reservation locking in ESX locks the entire LUN and not a specific partition. Therefore, the best practice is to have one LUN per VMFS only (exception being local storage). When using extents, gather multiple LUNs under one logical VMFS and not multiple partitions per LUN.
Mixing LUNs with different speeds can affect the performance of your extended VMFS. Therefore, it is better to have all extents presented from the same storage array.
When you want to create new virtual machines on a datastore or when the virtual machines running on a datastore require more space, you can dynamically increase the capacity of a VMFS datastore.
The current size limit for a VMFS extent is 2TB-512B. For large VMFS volumes, you must use spanning to concatenate multiple 2TB-512B extents. An extent is a partition on a storage device or LUN. As there is currently a limit of 32 extents in a VMFS volume, the size limit on any VMFS volume is 64TB. Spanning of multiple volumes (LUNs) is required to reach that upper limit and is needed for any VMFS volume that is greater than 2TB in size. For more information on the size limits, see the Configuration Maximums guide.
For large VMFS volumes, use one of these methods:
- Add a new extent. The spanned VMFS datastore can use any of its extents at any time. It does not need to fill up a particular extent before using the next one.
- Grow an extent in an existing VMFS datastore so that it fills the available adjacent capacity. Only extents with free space immediately after them are expandable. This can only be done on a LUN with size limit 2TB-512B.
For information on how to increase the capacity of a VMFS datastore, see the Changing VMFS Datastore Properties section of the ESX Configuration Guide. In vSphere, this can be done without having to power off the virtual machines.
Note: Ensure a complete rescan is done on all ESX hosts after adding an extent to a VMFS volume on one ESX host. Otherwise, the same extent might be inadvertently added by another node in the cluster, which could potentially cause loss of data. Best practice is to add extents to an existing VMFS volume from a single node and then rescan the storage resources from all ESX hosts capable of accessing that shared storage resource.
To troubleshoot issues when creating an extent, see:
- Cannot add a datastore or add an extent after increasing the size of the disk or LUN (1007026)
- The Add Extent wizard includes a warning message regarding data loss after extending a LUN (1002163)
- Adding an extent to a VMFS volume fails after increasing local storage space (1002821)
- Adding an extent fails with the error: Can't overwrite contents of active fb partition (1008938)
Once an extent is added, it is not possible to remove it without compromising the data residing on the extent. For more information, see Cannot remove a LUN extent (1003344).
All VMFS metadata is located on the master extent, which is typically the first LUN on which you created the datastore. If one of the extents goes offline, its data becomes inaccessible. For example, data on the entire .vmdk or the partial .vmdk files that span the extent. For more information, see Alert messages are logged when extents in a spanned volume go offline (1010314).
If the master extent goes offline, the entire datastore becomes inaccessible.
Request a Product Feature
To request a new product feature or to provide feedback on a VMware product, please visit the Request a Product Feature page.