-
Notifications
You must be signed in to change notification settings - Fork 19
Open
Description
Hi there!
I am running a crazy experiment in order to replicate what I have with ceph in Proxmox, shared storage replicated on 3 different nodes! And since both gluster, ceph AND ZFS fail at tiering for us mere mortals here I am !!!
For the moment I see very little movement of the VMs - can this be because the file is open pretty much all the time ? Any tips there ?
I am seeing loads of conflicts on the hidden .glusterfs folders - is there a way to ignore some folders ? Pinning them ?
I also am wondering what an optimal "Copy Buffer Size = 1 MiB" should be between a NVME ZFS and a HDD ZFS both with ARC enabled ?
Thank you so much for this project, very very useful!
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
No labels