Actually 3 disks! 

 

First it started with Parity 1 with (UDMA CRC error count)

I changed the SATA cable and connected it to another SATA port in the motherboard, and started rebuilding.

While rebuilding, Disk 2 got disabled with +2000 IO errors. Rebuilding was not interrupted, I think because I have 2 Parities.

Rebuilding completed, Parity 1 disk became green again. Disk 2 has red ball.

After few hours, Disk 3 got disabled.

 

Disk 2 - ST8000DM004-2CX188_WCT0M8D4 (sdm) (errors 2)
Disk 3 - ST8000DM004-2CX188_WG8011AG (sdl) (errors 2048)

 

Parity 1 is connected like this: motherborad -> 5.25 to 3.5 converter -> HDD

Disk 2 and Disk 3 are connected to LSI SAS HBA through the same port.

 

What is the best way to proceed? I believe losing another disk would cause data lose.

I have a spare pre-cleared disk but it has (60 Reported uncorrect), not sure how it got pre-cleared without issues.

 

Please let me know if need more info. I know I did a mistake by shutting down the server before getting the logs, I wanted to check the connections and was too scared to lose another disk.

 

Thanks.

 

 

Unraid Version: 6.7.2

CPU: AMD Ryzen 7 2700X

Mobo: MSI X470 GAMING PLUS (MS-7B79)

Memory: 16 GiB DDR4

 

 

 

tower-diagnostics-20200528-1303.zip

Edited May 28, 20206 yr by HAMANY

You rebooted since the errors so we can't see what happened, for now I would recommend unassigning both disabled disks and starting the array, check that both emulated disk mount correctly and contents look OK.

31 minutes ago, johnnie.black said:

You rebooted since the errors so we can't see what happened, for now I would recommend unassigning both disabled disks and starting the array, check that both emulated disk mount correctly and contents look OK.

 

Thanks for the quick response. Will try this now.

Edit: Done, the files looks fine.

 

Is there anyway to prevent the logs from being removed with rebooting, and keep them for few days?

Edited May 28, 20206 yr by HAMANY

23 minutes ago, HAMANY said:

Is there anyway to prevent the logs from being removed with rebooting, and keep them for few days?

You can set up a syslog server, though that's used more for troubleshooting when the server keeps crashing or similar, for this kind of problem you just need to download the diagnostics before rebooting.

Just now, johnnie.black said:

You can set up a syslog server, though that's used more for troubleshooting when the server keeps crashing or similar, for this kind of problem you just need to download the diagnostics before rebooting.

Noted.

 

The server is up and my files looks fine.

Should I assign the 2 disks again and start rebuilding?

Yes, and if any more issues post new diags (before rebooting)

On 5/28/2020 at 2:56 PM, johnnie.black said:

Yes, and if any more issues post new diags (before rebooting)

 

Disk 2 got disabled again just after the parity check started. Diags attached. I've stopped the parity check.

 

tower-diagnostics-20200601-1107.zip

That looks more like a power/connection issue, swap/replace cables and try again.

On 6/1/2020 at 1:18 PM, johnnie.black said:

That looks more like a power/connection issue, swap/replace cables and try again.

Thank you Johnnie!

The drives are now more stable after changing one of the power cables (4pins to SATA).

 

However, I''m getting read and SMART errors in one of the 8TB Seagate drives (Disk 10 - sdi)

I did a full scan using SeaTools but it didn't report any errors! Also it passes the SMART extended self-test.

Could it be a connection issue also?

 

The diagnostics file is attached. 

Appreciate your support.

 

Thank you.

tower-diagnostics-20200706-0948.zip

Edited July 6, 20205 yr by HAMANY

This does not look too good:

Quote

  5 Reallocated_Sector_Ct   PO--CK   099   099   010    -    2352

While reallocated sectors are not necessarily a problem as long as the numbers stays constant anything other that a small number is often a good indication that the drive's health may be suspect.  You might want to run an extended SMART test on the drive to see how that goes.

2 hours ago, HAMANY said:

Also it passes the SMART extended self-test.

It passed the short test, you should run a long one, but if Seatools passed it should also pass, it did fail a long test before, so there were issues before.

4 hours ago, itimpi said:

This does not look too good:

While reallocated sectors are not necessarily a problem as long as the numbers stays constant anything other that a small number is often a good indication that the drive's health may be suspect.  You might want to run an extended SMART test on the drive to see how that goes.

 

3 hours ago, johnnie.black said:

It passed the short test, you should run a long one, but if Seatools passed it should also pass, it did fail a long test before, so there were issues before.

 

Thank you both for your reponse.

I will re-run the extended SMART and SeaTools again just to make sure.

 

The drive is still under warranty, but I can't create an RMA without the SeaTools log showing the problem.

1 hour ago, HAMANY said:

The drive is still under warranty, but I can't create an RMA without the SeaTools log showing the problem

With that many reallocated sectors I would expect them to accept a RMA regardless of the result of the Extended SMART test.

SMART extended self-test has completed without errors!

I'm sure SeaTools would give me the same results same as before. Very weird.

Logs attached. Appreciate your suggestions.

 

@johnnie.black @itimpi 

tower-smart-20200711-0246.zip tower-diagnostics-20200711-0250.zip

For ZCT1BN14 disk, a weird thing happen, SMART clear previous error. I haven't look deep in previous diagnostics. But you got trouble on 3 Seagate disk.

Many years ago, 3 of 4 ( same lot ) Seagate 3TB disk got problem less in 2 yrs, after that, I never buy Seagate again, mainly because other brand have good quality and low price source.

 

Error 12 [11] log entry is empty
Error 11 [10] log entry is empty
Error 10 [9] log entry is empty
Error 9 [8] log entry is empty
Error 8 [7] log entry is empty
Error 7 [6] log entry is empty
Error 6 [5] log entry is empty
Error 5 [4] log entry is empty

Edited July 11, 20205 yr by Benson

6 hours ago, HAMANY said:

SMART extended self-test has completed without errors!

That means disk is fine for now, just keep an eye on it.