A lot more disk space than I have being shown (ZFS artifact?)


I’ve an installaton of node-exporter and prometheus on my NAS, and I get nice graphs on my grafana cloud dashboard - but the “Disk Space Usage” graph shows that I have 146TiB available. There is a lot of disk (compared to what I have in other machines, I have colleagues that have boxes with even more disk space) in that machine, but even before redundancy it only comes out to 70-80TiB, nowhere near the reåorted 146TiB.

Is that because I use ZFS on those disks, and it reports the pool’s free space for every file system on it. I.e. Is those 146TiB the sum of the reported free space for every file system? Is there a way to stop grafana from doing that?

If possible, I would use telegraf to monitor usage and availability. On Linux and FreeBSD,

poolMetrics = true
datasetMetrics = true

should work. On FreeBSD 13 and above, you can also add

  command = ["/usr/libexec/zfs/zpool_influxdb", "--execd"]
  name_override = "zpool_libexec_stats"
  signal = "STDIN"
  restart_delay = "10s"
  data_format = "influx"
  interval = "1m"

for even more metrics