When you mentioned, that, what will happen if you lost one host - it means, that you already lost 1/6 of the RAW capacity and thus you will not have 200TB capacity :-) , so it will try to rebuild as much as possible objects with the RAID 5 policy to make most of the object compliant again, but with 2+1 scheme. And make the objects again redundant with FTT=1.
------------------------------
Jiri Viktorin
VCIX-DCV, VCIX-NV, VCF-VCI, vExpert
------------------------------
Original Message:
Sent: Dec 13, 2024 02:19 AM
From: abckdc
Subject: vSAN ESA Adaptive RAID Failures effect on usable capacity.
Hi All,
We are trying to analyze vSAN ESA as our HCI architecture for our greenfield environment. I have some confusion around vSAN Adpative RAID during failure. Lets say we run 6 node cluster with host rebuild reserve turned on (n+1). Our storage policy is set to be RAID 5. And our usable storage is coming 200 TB. Since its a 6 node cluster the RAID schema is supposed to be 4+1. Lets assume one of our node fails, and we have a host rebuild set to true so, that node will help in rebuilding the data. As per the document vSAN will then wait for 24 hrs and shift the cluster to RAID 5 2+1 schema.
RAID 2+1 will have a storage efficency of 150% in compared to 4+1 which has 125%. How will this effect my usuable storage capacity of 200 TB. Lets say I am fully utilizing my 200 TB(just assuming) of available storage, what imapct will this have on the cluster.