Disk I/O Statistics in "Reporting" seem way off for HGST disks

TrueNAS General
1

I’m upgrading my array from 8x 4TB Seagate ST4000DM000 disks in a RAIDz2 pool to 8TB disks from both Seagate (ST8000VN004 Ironwolf) and HGST (HUH721008ALN600). All are SATA 7200 RPM disks.

Under Reporting → Disk, the HGST disks are reporting wildly higher throughput than the Seagates. Seems to be almost 7-10x as much. For example, during the last disk replacement/resilver:

image
image1332×827 158 KB

All the Seagates (2x ST8000VN004 and 2x ST4000DM000) are topping out at about 170 Mebibytes/s, while all 4 of the HGST disks are approaching 1.5 Gibibytes/s. SATA III tops out at 6 Gbit, so I know that the read/write stats for the HGST drives can’t be right. The Seagate throughput seems reasonable though. Is this a known issue? I searched the forum, and I couldn’t find anything similar being mentioned.

The hardware is:
TrueNAS 25.10.3 - Goldeye
Netgear ReadyNAS 3312 (12 bay rackmount - all SATA interfaces)
16GB of ECC DDR4 UDIMM
Intel Xeon E3-1240 v5 CPU

(Edited to add TrueNAS version, RAID level and CPU model)

2

Bump. This is still rather confusing. As another data point, I have a ReadyNAS RN428 enclosure with 8x 10TB HGST drives and I don’t see anything like this behavior on that array.

3

If you think it is a bug, it’s best to use Report a Bug in the GUI and submit a debug dump.

Looks like a smile icon on the upper right of the TrueNAS GUI for Feedback / Report a Bug. You can also use Report a Bug at the top, upper right of the forum.

You should check your statistics for the drives using the command line to see if it report there. Guessing just the chart is incorrect

1 Like
4

Ah, cool, I’ll check with iostat and see where that leads me.