Double disk failure?

[cobolstinks]

 

I got this email a couple of days ago and I'm just getting around to looking into it. 
 

Event: Unraid Disk 1 error
Subject: Alert [TOWER] - Disk 1 in error state (disk dsbl)
Description: MD3000GSA6472E_1_P8H1R6XR (sdd)
Importance: alert

 

When I web into the server I see this:

 

The problem is disks 1 and 5 are gone...

When I click on the main tab I see this:

 

When I download the syslog I get an empty zip file.  How do I proceed?  Did I just have a double disk failure?  I replaced Disk 1 about a month ago, if that matters.

[apandey]

3 hours ago, cobolstinks said:

When I download the syslog I get an empty zip file

how are you doing this? best if you can share diagnostics

 

3 hours ago, cobolstinks said:

Did I just have a double disk failure?

bad connections / bad cables are more prevalent than bad disks. check those first

 

[cobolstinks]

I powered down the server rebooted it and now I can see both the disks in the main tab.

 

I ran a SMART test on disk 1 last night.  It came back clean 0 errors.  I attached the diagnostics zip to this post.  Am I safe to re enable Disk 1?  How do you do that, I'm stuggling to find the "enable" button in the Disk menu.

 

tower-diagnostics-20230421-0859.zip

[JorgeB]

Post new diags after array start, to see if the emulated disk is mounting.

[cobolstinks]

5 hours ago, JorgeB said:

Post new diags after array start, to see if the emulated disk is mounting.

tower-diagnostics-20230421-1411.zip

 

It looks like it's mounting but i still see it as disabled.

 

[JorgeB]

Emulated disk is mounting so assuming contents look correct you can rebuild on top, recommend checking/replacing cables first to rule that out.

 

https://wiki.unraid.net/Manual/Storage_Management#Rebuilding_a_drive_onto_itself

 

P.S. reiserfs is deprecated, you should convert to xfs when possible.

[cobolstinks]

when you say convert to xfs is there where I would do that?
 

[JorgeB]

https://wiki.unraid.net/File_System_Conversion

[cobolstinks]

Ok I ended up replacing the sata cable and rebuilt the disk onto itself.  It was fine for a few weeks and now I'm back to a disabled disk1 .  I've attached the smart test.  This disk is only about 3 - 4 months old.  Did i get a bad replacement disk or is an issue with my sata controller?  Its a really old machine, wondering If I should replace the mobo/cpu/ram.  

 

tower-diagnostics-20230524-0834.zip

tower-smart-20230524-0831.zip

Edited May 24, 2023 by cobolstinks

[trurl]

SMART report looks fine. Bad connections are much more common than bad disks.

 

Attach diagnostics to your NEXT post in this thread.

[trurl]

On 4/21/2023 at 2:24 PM, JorgeB said:

P.S. reiserfs is deprecated, you should convert to xfs when possible.

 

[cobolstinks]

29 minutes ago, trurl said:

SMART report looks fine. Bad connections are much more common than bad disks.

 

Attach diagnostics to your NEXT post in this thread.

Thank you for the help.  I believe I attached the diagnostics zip file to my post along with the SMART test results, unless I'm posting the wrong diagnostics.

I hear you on the bad connections, I guess I'm not sure how to proceed here.  I can rebuild again on disk1, but if it fails again I'm not sure how to fix this long term.  I've replaced the sata cable and the disk is 3-4 old replacing what I thought was a previous disk failure.  So as far as hardware that leaves the disk and the sata port itself... right?  I don't have any spare sata ports on my mobo.  So if this isn't the disk but rather the port, I'm looking at a mobo/cpu/ram upgrade or maybe a PCI sata card, right?

Edited May 24, 2023 by cobolstinks

[trurl]

3 hours ago, cobolstinks said:

I believe I attached the diagnostics zip file to my post

You edited your post to attach them after I requested them.

 

Your syslog indicates you booted Dec 31, but NTP corrected that. So your server isn't keeping time after reboot. Check your CMOS battery.

 

You rebooted after the problems occurred, so syslog from diagnostics can't tell us anything that happened before.

 

You barely have enough RAM for just NAS duty, best if you just disable Docker and VM Manager and don't attempt that.

 

SMART for all drives looks OK except for the age of some. Why are your disk controllers IDE instead of AHCI?

 

Do you have a spare you can use to rebuild disk1?

 

 

[trurl]

Can't tell if emulated disk1 is mountable since those diagnostics were without the array started.

 

Start the array and attach diagnostics to your NEXT post in this thread.

[cobolstinks]

I started the array and have begun reb uilding disk1 onto itself.  

I can check if AHCI is an option in my mobo settings.  Yes I'm aware that my CMOS battery is dead, every time i loose power I have to reset the boot priority.  I should just replace the battery.

I do not user VMs or Docker I only use this machine as a NAS.  I do not have a spare disk.

tower-diagnostics-20230524-1314.zip