Performance metrics over SNMP

I reached this post looking for information about reading the performance metrics on a qnap qts hero, my version is the current 6.0.2.3591.

On the pre hero you could use the OID family .1.3.6.1.4.1.24681.1.4.1.11.5.6.2.1. to retrieve performance data.

With the hero, since it’s zfs this no longer applies, and there’s no equivalent I could find to get the values I need.

So the remarks on 6.0 beta are still there? Did anyone find a way to obtain real performance data? We are having performance issues and it’s really difficult to get to the root of the problem without the data.

Thanks

What performance issues are you currently running into, and which metrics are you mainly looking for? Could you let us know? Thanks!

Sure, we have a TS-HX77AXU-RP, with 8 discs forming part of a pool, in this pool there are two iscsi LUNs one for backups and another one for a proxomox test environment

We are doing a backup to an external media and it’s hurting the performance of the proxmox LUN to a point that’s unusable

I would like to have ideally:

  • metrics of this performance (read/s, writes/s, ..)
  • metrics of the zfs performance

If I look to the resource monitor everyhing seems to be going great (and it’s not)

Could you share a screenshot of the Resource Monitor performance page?

Hi today the task finished so the metrics are near zero (we moved part of the load to nfsv2), but these metrics you show seem to be interesting to have montitored. Are they exposed via SNMP?

Currently, these performance metrics aren’t available via SNMP on QuTS hero. The existing SNMP performance OIDs were designed for the QTS volume structure and don’t cover the ZFS-based architecture of QuTS hero. I’ve passed this along to our internal team, and we’ll look into improving this in the future. Thanks!