persistant disable disk

[wilsonhomelab]

Hi, 

 

May I ask for help for the problem I have recently RE disable disk?

 

I have been transferring my UnRaid server from old hardware (Dell R410 with Perc H200a HBA controller) to a new desktop hardware (Ryzen 2600, Gigabyte B450 Aorus Pro, 8GBx4 DDR4 Corsair 3200 LPX, Quadro P1000 4GB GPU, same Perc H200a HBA controller in IT mode). I recently purchased a new 8TB Ironwolf and planning to upgrade my parity drive.

 

I first problem I had is the pre-clear the new 8TB drive. The post read wasn't successful. After hours of googling, I found it is due to my memory error. At this point, I do realise how reliable of the old R410 with ECC memory which I used to pre-clear many disks without any problems. The solution is to swap slots for those 4 DDR4 memories, and I managed to pass the memtest64. (Note: I performed the pre-clear procedure in another mini PC I have without any problems. So I didn't test the pre-clear after I fixed the memory problem)

 

2nd problem, the disable disk. After I reorganise the disks in my hotswap bays (5 hot swap bays x2, plus two internal slots, so total 12 HDD slots), I had the disk2 showing "disable" error.  SMART report here.MB3000FCWDH_Z1Y3A9NA0000C533HAY2_35000c500833efff7-20210822-1133.txt  I couldn't figure it out what went wrong. So I went ahead to rebuild the disable drive following this wiki steps.

 

After 8hrs of disk rebuild, I have got 4 green lights back for my array. To make sure everything is right, I clicked the parity check. Somehow the "disable disk" error came back for the same disk 2. I download the diagnostics unraid-ryzen-diagnostics-20210822-1110.zip straight away. Please give me some helps and point me out what went wrong?

 

I suspect the mini SAS 8087 to 4 SATA cable I purchased off eBay may be to blame. But I wasn't sure. It looks quite good quality cable. I also have another mini SAS 8087 to 4 SATA cable I originally used in R410, which it doesn't seem to have problem last  couple of years.

 

 

 

Edited August 22, 2021 by wilsonhomelab

[wilsonhomelab]

Please give me some advice for what to check in the diagnostics. I have been using Unraid for two years now, but consider myself still new to this OS. 

[JorgeB]

Swap cables/slot on disk2 and see if the problems stays with the disk or the cables/slot.

[wilsonhomelab]

48 minutes ago, JorgeB said:

Swap cables/slot on disk2 and see if the problems stays with the disk or the cables/slot.

Thanks JorgeB. I swap the disk 2 to the old "reliable" cable and running a parity check at the moment. I am just wonder if the "new" cable is the culprit, would the disk rebuild I performed over this cable be ok ? There is no error reported after the rebuild.

[trurl]

9 hours ago, wilsonhomelab said:

running a parity check

So did you rebuild disk2 again after those diagnostics?

[wilsonhomelab]

3 hours ago, trurl said:

So did you rebuild disk2 again after those diagnostics?

I am currently rebuilding disk2. I am quite happy with 166MB/s speed.  About 3hr left. Will report back the result.

 

[trurl]

If you were running that parity check with a disabled disk and only single parity, then it was doing a read-check, not a parity check, since there aren't enough disks to do the parity calculation to compare to parity.

[wilsonhomelab]

So the rebuild of disk2 finished.

Next, I will upgrade the 4TB parity to 8TB drive. Before I move on, will I need to make sure that the disk2 is rebuilt properly? please advise

 

Both disk 1 and disk 2 still have SMART errors.

disk 1

 

disk 2

 

[itimpi]

Without the Diagnostics/SMART reports is is difficult to guess what the errors might be.  For instance they could be CRC errors which were recovered and you simply need to acknowledge them to stop being prompted about them if no more occur.

[JorgeB]

1 hour ago, wilsonhomelab said:

Both disk 1 and disk 2 still have SMART errors.

Those are normal with SAS devices, it's a GUI issue, you can still download the SMART report to check if all looks good.

[wilsonhomelab]

22 minutes ago, itimpi said:

SMART reports

 

3 minutes ago, JorgeB said:

SMART report

 

disk 2 unraid-ryzen-smart-20210823-1956.zip

 

disk 1 unraid-ryzen-smart-20210823-1954.zip

 

diagnostic unraid-ryzen-diagnostics-20210823-1812.zip

 

I don't have enough knowledge for the SMART report. but this section caught my attention.

 

In disk 1 report, it says

Quote

Background scan results log
  Status: scan is active
    Accumulated power on time, hours:minutes 21942:06 [1316526 minutes]
    Number of background scans performed: 564,  scan progress: 16.72%
    Number of background medium scans performed: 564

 

I did click the "SMART extended self-test" before. Is the SAS drive performing an extended self-test? The spin-down is enable. Looks like the self-test may be interrupted when spin-down.

 

11 minutes ago, JorgeB said:

it's a GUI issue

It makes me quite worried. So it is a false alert!

 

So, am I safe to go ahead for parity drive upgrade?

 

[JorgeB]

Both disks look fine, but there are no complete long SMART tests to confirm.

[wilsonhomelab]

34 minutes ago, JorgeB said:

Both disks look fine, but there are no complete long SMART tests to confirm.

The UnRaid behaves as normal. I just backup a few important folders onto my desktop and went ahead the parity drive upgrade. (reported 19hrs at 113MB/s). 

 

RE the suspected faulty mini SAS cable, what should I do to test it after the parity drive upgrade? It could be a faulty SAS controller port (port A) it connected to. 

[JorgeB]

1 hour ago, wilsonhomelab said:

the suspected faulty mini SAS cable, what should I do to test it after the parity drive upgrade?

Just replace or swap with another and see if the problem goes away/follows the cable.