Best practices when using vSAN and non-vSAN disks with the same storage controller
book
Article ID: 327054
calendar_today
Updated On:
Products
VMware vSAN
Issue/Introduction
This article provides best practices when both VMware vSAN (formerly known as Virtual SAN) and non-vSAN disks are attached to the same storage controller.
Notes:
The physical storage aspects such as vSAN disks managing I/O, retries, and fault declaration are managed differently when the same elements are handled by non-vSAN workloads.
This applies only in circumstances where the non-vSAN disks are attached to the same storage controller as the vSAN disks.
Environment
VMware vSAN 6.x VMware vSAN 7.x VMware vSAN 8.x
Resolution
To avoid conflicts or other issues with the vSAN infrastructure, consider these points when the same storage controller is backing both vSAN and non-vSAN disks:
Do not mix the controller mode for vSAN and non-vSAN disks.
If the vSAN disks are in Pass-through mode, the non-vSAN disks must also be in Pass-through mode.
If the vSAN disks are in RAID mode, the non-vSAN disks must also be in RAID mode.
Mixing the controller mode will mean that various disks will be handled in different ways by the storage controller. This introduces the possibility that issues affecting one configuration could also affect the other, with possible negative consequences for vSAN.
There should only ever be one non-vSAN disk and that disk should ONLY be used for ESXi boot and or scratch, the (VMFS) OSDATA datastore should ONLY be used for scratch, logging, and coredumps. VMs should never reside on the OSDATA datastore.
Virtual machines should not be running from a disk or RAID group that shares its controller with vSAN disks or RAID groups.
ESXi host installation is permitted on non-vSAN disks attached to the same controller for vSAN 7.0 and higher. This is NOT supported for vSAN 6.x and lower.
Do not pass through non-vSAN disks to virtual machine guests as Raw Device Mappings (RDMs).
Notes:
If you use vSAN and non-vSAN disks for high-volume operations on the same storage controller data might become unavailable due to disks erroneously reporting as failed. In a worst-case scenario, this could lead to potential data loss on the vSAN datastore.
If you are removing any VMFS on non-vSAN disks, you must ensure that the coredump target partition on the ESXi host is large enough. For more information, see: