I manage a couple of storages at work. I got a few Huawei OceanStor storages. v2 as wel v3 storages. There is some commercial tooling from Huawei available to gather metrics, but that didn’t fit my needs. I found out that all or almost all metrics that I need are available via SNMP. So it was not hard to setup a Grafana dashboard with all metrics.

Pre install

Make sure you have installed InfluxDB as the time-series database Telegraf as the collector first.

Quick Start

Get the all files, at My GitHub page.

  • Enable SNMP on your Huawei OceanStor storage.
  • Put hw_stor.conf in your /etc/telegraf/telegraf.d directory.
  • Edit the community string as appropriate.
  • Edit the SNMP version as appropriate, please note that verion 2 is disabled by default on the Dorado storage, check with show snmp version on the Storage.
  • Edit the ‘agents’ list to include all of your monitored OceanStor storages.
  • Put the files in the mibs dir on the server running telgraf. With Ubuntu it is in /usr/share/snmp/mibs other Linux distributions can differ.
  • Import the Grafana dashboard in Grafana.

Screenshot

After you setup all, you should have some dashboard like this.

Note, on the latest version on GitHub CIFS and NFS stats has been added, not on this screenshot yet.

Grafana screenshot zswap