New USB drives arrived for my Ceph cluster, and they’re not reporting SMART data. Again. After solving this problem in my October 2025 article, I need to update the configuration with new device IDs and share the lessons learned from running this solution across my entire cluster.

The bottom line: This was absolutely the right decision. SMART monitoring has already caught failing drives before they damaged data, and the performance trade-off is negligible compared to the stability benefits.

One of the underrated features of Proxmox is the ability to leave notes on the individual resources like the Datacenter, individual nodes, and every VM or Container. However, these notes are tucked away in individual configuration files within the Proxmox Cluster File System (/etc/pve).

If you’re managing a cluster with nodes like quell, edgar, and tanaka, having a single consolidated Markdown file of all your documentation is incredibly helpful for disaster recovery or general reference.