Description of problem:

The only guidance we give around RAID is to disable it.

"If you have RAID (not recommended), configure your RAID controllers to RAID 0 (JBOD)."

We could stand to elaborate a bit. Something along the lines of:


RAID Controllers

If a RAID controller with 1-2 GB of cache is installed on the host, then enabling writeback caches might result in increased small IO write throughput. In order for this to be done safely, the cache must be non-volatile. Modern RAID controllers usually have super capacitors that provide enough power to drain volatile memory to NAND during a power loss event. It is important to understand how your particular controller and firmware behave after power is restored, some may require manual intervention by an operator. Hard drives typically advertise to the operating system whether their disk caches should be enabled or disabled by default. Some RAID controllers or firmwares may not pass these hints to the OS, so the operator must be verify that disk level caches are disabled to avoid filesystem corruption.

For each OSD data drive, a single RAID 0 volume should be created, with writeback enabled. If SAS or SATA connected SSDs are also present on the controller, it is worth investigating whether your controller and firmware support a passthrough mode. This will avoid the caching logic, and generally result in much lower latencies for fast media.


Version-Release number of selected component (if applicable):

https://access.redhat.com/webassets/avalon/d/Red_Hat_Ceph_Storage-1.3-Installation_Guide_for_RHEL_x86_64-en-US/Red_Hat_Ceph_Storage-1.3-Installation_Guide_for_RHEL_x86_64-en-US.pdf