Hello, guys,

 

This week end, i installed a new cache drive. But in the process, it bend the cables and the card  from the M.2 controller where 3 of my HDD were. when i started unraid, they did not show up, then after fiddling a bit, they showed up. But it was not stable and one disconnected in the middle of a write operations.

 

One of the drive became disabled (red cross on disk 1). I did a smart short test on the three drives and they seemed fine. So i rebuild the array with the disk 1 in place.

 

It finished this morning, but there was 504 errors. Mainly because of disk 2 and a current pending sector.

I did smart extended test on all three drives. Disk 3 came back fine, but there was errors on disk 1 and 2 (errors occured - check smart report).

I just launched a parity check (without correction).

 

I ordered a new m.2 controller just in case there are more errors (could mean that the controller was damaged).

 

What do i do for disk 1 and 2?  it's only media on it, i would not die if i lose some of it. Are the errors really bad? or if there is no more errors, i should be fine? Right now, the overall server is working fine.

 

Thanks for the help

 

WDC_WD40EFRX-68N32N0_WD-WCC7K6SKLNHN-20210830-1651.txt WDC_WD40EFRX-68N32N0_WD-WCC7K5VY4XY7-20210830-1604.txt WDC_WD40EFRX-68N32N0_WD-WCC7K5VY42E9-20210830-1727.txt

 

unraidchoupi-diagnostics-20210830-1727.zip

Edited August 30, 20214 yr by Troussdesoin

Can you just attach the zip file of the diagnostics ?

I doubt people will bother downloading every file one by one.

You're right,

 

Thanks for that, i edited the first post

1 hour ago, Troussdesoin said:

What do i do for disk 1 and 2?

Unless all your data is not important they should be replaced, don't forget that Unraid requires all other disks to be OK to rebuild a failed disk.

Yes i think you're right, i got disk1 read errors as well on the parity check

Edited August 30, 20214 yr by Troussdesoin

You should setup Unraid to monitor attributes 1 and 200 on your WD disks. Click on a disk to get to its settings page.

Ok,

 

I just did it. I got Raw read error rate of 0 on disk 2 and 3, and error rate of 5 on the disk 1. Actually after reading into it, it does not seem to matter much because the worst is at 199. which is far above the trehold.

Also the multi zone error rate on disk 1 is also at 1.

 

I was wondering if it could be that only the controller /Cable is bad? Could it provoke all those errors? Should i replace the drives now or wait for the new controller/ cables to arrive?

 

Edited August 30, 20214 yr by Troussdesoin

39 minutes ago, Troussdesoin said:

I was wondering if it could be that only the controller /Cable is bad?

The extended SMART test failed on both disks with a read error, this is a disk problem, can't be controller/cable related.

All right, it's settled then, i'll switch both of them. And how can i test the Cable/controller? So that i know if i need to replace them as well?

Edited August 30, 20214 yr by Troussdesoin

8 hours ago, Troussdesoin said:

And how can i test the Cable/controller?

Non correcting parity check is a good test.

I did a parity check (non correcting). It returned with 0 errors (with the two failing drives).

 

I am rebuilding the array with a new drive. And then i'll do some preclear on the old drive to see if it was really failing or it was from something else.

52 minutes ago, Troussdesoin said:

with the two failing drives

These type of read errors can be intermittent, i.e., they can work today and fail tomorrow.

Well, 

 

I switched Disk 1. It rebuilded with a few errors (most likely a few seconds of a film were lost). Did a few preclear on the dying disk, and it was really dyning. I'm waiting for a second disk to come in to replace the other failing drive.

I'm also waiting for a new controller, Just to be sure.

 

Overall it was a good learning experience. Thanks guys for the help.