Drive faulted i 10-drive Z2 pool

[ChrisReeve]

Hi,

I am running TrueNAS Core with a single vdev of 10x10TB White Label RED drives in RaidZ2.

I recently had read errors on one of my drives, so it got faulted by TrueNAS. It has now been replaced with a new drive. I also replaced the cable, just to be sure.

My question is: is the drive that was faulted really bad, or is it possible due to the cable, or any other factors?

I am asking, because it still passes SMART-scan in windows. I have added a log from a smart scan of the drive, that was done after it faulted. It also passed smart scan in windows (tested it after removing it from the server). It is also worth mentioning that the same drive that I now replaced failed (froze) during a resilver a few months ago. After a reset, it was able to resilver successfully.

Please see attachment for SMART-scan.

 

[Heracles]

Hi,

Read Errors + Passed Smart Test = Cable / Connector problems (or at least, most of the time...).

So it looks like it is not the drive itself that has problem but the cable you used to connect it, the port it is connected to, the controller driving it or any other hardware component like that.

 

[blanchet]

You should test the hard drive with the manufacturer diagnostic tool on another computer to know the health of the disk.
Ultimate Boot CD holds the diagnostic tools for many vendors (including Seagate and WD)

If the diagnostic tool reports no error, it means that you have a connector, cable or power supply issue on your storage server.

 

[kherr]

My rule of thumb :

If it gradually goes bad ...... probably the disk.

If it goes "wongckers" all of a sudden or very short period of time .... probably hardware

If multiple go at once, especially on same card ..... most likely hardware ....

In any event I usually reboot to see if it stays the same or clears ..... usually it clears and never have a further problem ...
I also reseat cables before booting up again ..

YMMV