Hi everyone, I have a question: I'm deploying a Harvester cluster with 3 nodes. Suppose each node has 1TB of storage, so the total is 3TB. If I create a VM with a 300GB volume on one node, the data will also be replicated to the other 2 nodes, meaning it will consume 900GB in total (300GB on each node). Now, if I deploy a SAN and configure it so that Harvester writes data to the SAN instead, will it store just 300GB, or will it store the full 900GB (including replication)? Thank you all for reading my question!

In case you didn’t know Longhorn is the default storage class for Harvester. In the GUI you can create a storageclass with 1 replica. You are correct, the default is 3. 1 copy per server. When using a SAN you should use the CSI instead of connecting the drives directly to the nodes. Are you FC or Iscsi?

Hi mr.andy @thousands-advantage-10804 tks for answer my question . I use IScSi for SAN . Would you mind sharing any documentation or best practices for configuring Harvester to work with a CSI driver, if available? Thank you in advance!

Take a look at https://docs.harvesterhci.io/v1.5/advanced/csidriver. There are others that are not validated yet.

tks sir