Is there anything available I can use to track the 'Wearout' value of disks in my PVE 8.4 cluster over time? I've got a few 'chatty' k8s VMs that I'd like to see what impact they're having on the NVMe drives they're running from over a period of weeks and months. With the cost of flash storage going insane for the forseeable future I'd build a HDD-based cluster to house these if they're going to chew them up quickly. Any ideas for tracking the wearout? Thanks.
Monitoring SSD/NVMe 'Wearout' on PVE Disks Over Time
- Thread starter smoosmoo
- Start date
https://pve.proxmox.com/pve-docs-8/pve-admin-guide.html#disk_health_monitoring There is also the Show S.M.A.R.T. values in the web GUI in the Disks section of the Proxmox node.
Last edited:
Thanks for the reply. I'm aware of these adhoc commands and S.M.A.R.T. values but was wondering if there's something like a Prometheus module or something that can show a graph of the wearout values changing over weeks and months. Just keen to see if there's a VM that is shredding an NVMe drive without constantly manually checking the wearout value.
