On 1/29/2024 at 10:41 AM, JorgeB said:

Disk dropped offline so there's no SMART, check/replace cables and post new diags after array start.

Hi, so I waited a bit, it just so happens I had new HDDs coming in and was running preclear, I waited for that disk to complete pre-clear first in case I need it to step in to replace disk 2.

so per your suggestion, what I found was that disk 2 is fine, or so I think. it's a shucked WD HDD with those 3.3V pin, the tape that I put on to block the pin got loose somehow, once I checked the cables and reapplied new tape, Disk 2 becomes available again upon boot up. Even though I don't know what else to look for before jumping to conclusion but I think Disk 2 is fine...if it is, could you update the link to procedure for "re-enable the drive" you originally posted at 

or do their reddit discussion also cover the correct procedure?

I really don't want to rebuild the entire parity if I could avoid it...thanks so much for everything!

unryzend-diagnostics-20240131-0029.zip

Edited January 31, 20242 yr by superfans124

SMART looks fine, and the emulated disk is mounting, assuming contents look correct you can rebuild on top.

I'm having a similar issue: I have a drive now marked as disabled. The server has been running well for months without issue. I reviewed some of the documentation from the links provided in this thread. The two courses of action I see are to replace the drive or to do a rebuild on itself. Could an expert please review my attached diagnostics and suggest a course of action? I am still a noob to Unraid, so any advice would greatly appreciate it. 

Thank you.

tower-smart-20240504-1134.zip tower-diagnostics-20240504-1140.zip

Disk1 appears to currently not be online so there is no SMART information for that drive so cannot give any view on its health.   Looks like several days ago (may 1st) you started getting continual resets on that drive of the type we frequently associate with power or connection issues, although it also possible it was the drive itself starting to play up.

 

You should power down the server, carefully check the cabling (both power and SATA) to that drive, and then power it back up to see if that drive comes back online and then post new diagnostics.   
 

It looks as if the ‘emulated’ disk1 is mounting OK but you might want to check it has the content you expect.

 

BTW:  The diagnostics includes SMART reports for all drives so no need to post them separately (although it can be useful to mention which disk has problems to help us focus on the correct drive).

Thank you for getting back to me so promptly. I apologize for not including the disk number. I appreciate the clarity of your instructions. I have followed your directions, checked the cabling, replaced the power cable connection, restarted the array, and run a new diagnostic, which is attached. Disk 1 is still disabled. I look forward to your assessment. 

tower-diagnostics-20240504-1353.zip

These new diagnostics show disk1 back online, and nothing obviously bad in the SMART information for the drive.  There are a few CRC errors but since these only indicate connection issues and not drive issues they are not a concern unless they continue to increase (they never reset to 0)  If you want to do a confidence test then the extended SMARt test is a good indication that the drive is fine.

 

You can only clear the ‘disabled’ status by rebuilding either the drive or parity.  You would normally want to rebuild the drive so that any writes to that ‘emulated’ drive are not lost.   Whatever is currently showing on the ‘emulated’ disk1 is what you would end up with after a rebuild.
 

Ideally you would rebuild onto a different drive keeping the current physical disk1 intact just in case something goes wrong with the rebuild (to give other recovery options) but if you are happy that the contents of the emulated drive look OK then rebuilding the drive onto itself as documented here in the online documentation accessible via the ‘Manual’ link at the bottom of the GUI or the DOCS link at the top of each forum page should be safe enough.

Thank you for being so thorough. This information is very helpful. I think I will go with the rebuild on the self option. Just one thing: I cannot seem to find directions on how to check whether the "emulated disk" contents are okay. There are references to checking this check, including before running the rebuild procedure, but no specific directions on what this check entails. If you could provide a link to this information, I will review it before proceeding. 

Thank you. 

Just now, brownsugar said:

There are references to checking this check, including before running the rebuild procedure, but no specific directions on what this check entails. If you could provide a link to this information, I will review it before proceeding. 

There is no definitive way to check the results - it is a manual process.   You can open the ‘emulated’ drive and browse its contents.    If you have backups you can do some checks against those.   If you have checksums for the files they can be checked.

Thank you so much for your help. I will proceed with the rebuild process. Wish me luck. 

On 1/7/2021 at 10:11 AM, ChatNoir said:

JorgeB did link to the procedure like 10 posts above.

Another dead link in the forums related to the most common issue I have with Unraid which is still not resolved. Drives dropping out for no reason and switching designations so they are seen as new disks. Unraid has started to become unstable to the point I'm rebuilding my disks every 3 to 6 months. I've been using this for 6 years now and I'm about to switch back to Windows Server because the support here is horrific. I paid good money for this OS but I get Microsoft stuff for free in my lab environment. I like so many things that Unraid has to offer but stability is not among them. I can run docker hosts as VMs and get more stability out of a Windows platform. 

15 minutes ago, jagame said:

the most common issue I have with Unraid which is still not resolved.

Please create a new thread and post the diagnostics after that happens.

On 11/16/2017 at 8:40 AM, JorgeB said:

SMART looks mostly fine, but a non zero Raw_Read_Error_Rate is never a good sign on WD disks, still it should be OK for now, to rebuild:

 

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

 

Outdated link:

https://wiki.lime-technology.com/Troubleshooting#Re-enable_the_drive

The links are outdated.

Potential SOLUTION:

 

One drive of my array was disabled. Smart Tests could be run on it.

 

The solution that worked for me was to reinstall the Flash drive form a recent backup.

 

I figured that the disk information within the unraid settings must have been faulty / damaged because of some reason.

 

I have to mention, I run an enterprise setup with hotswapable drives and I did a hotwap whicht Unraid doesnt seem to take so well.

 

I hope this will help in some cases. Please give a response if it does.

2 hours ago, kAI53r said:

The links are outdated.

Not sure if you noticed that you are quoting a post from 2017

shure, but if the problem is still an issue, it might be worth updating??

 

I dont have the overview most of your Moderators have, but if there is a solution in the documentation and some posts in this block have active links which work or even show the conent you were aiming at.

 

Its great if the needy find/use working links fast. So you could just take these out and it would be so time consuming ...

 

No offense just a thought. Thanks for your contribution.

12 hours ago, kAI53r said:

shure, but if the problem is still an issue, it might be worth updating??

You can create a new thread and ask for help, or find the new links in recent posts or in the Unraid docs site, not sure if that's what you are suggesting, but it would be impossible for us to update links in old posts.

Was wondering if you can explain what is the error with my disk. I got a discord alert telling me disk 5 is in error. I ran smart, xfs repair and found no problem.

 

Thank you

servraid-diagnostics-20250201-0054.zip

2 hours ago, Nodiaque said:

Was wondering if you can explain what is the error with my disk.

Looks like a power/connection issue, most likely the SATA cable.

that'S weird. I switch the power cable for a new one about 3 months (new psu) and the sata cable is a new cable from HBA card. hmmmm

3 minutes ago, Nodiaque said:

that'S weird. I switch the power cable for a new one about 3 months (new psu) and the sata cable is a new cable from HBA card. hmmmm

It is easy for a cable to become imperfectly mated (particularly the SATA one) over time, possibly due to vibration.

I'll recheck. I reconnected everything about 2 weeks ago when I cleaned the place (took the server down and vacuumed the dust in and out).

Having the same issue. I would appreciate anyone telling me if the drive is ok? Or letting me know how I can check for myself.

ST10000NM000G-2MU103_ZS51FVM6-20260205-1032 disk4 (sdd) - DISK_DSBL.txt

1 hour ago, essjayeire said:

would appreciate anyone telling me if the drive is ok?

SMART looks fine, full diags when the errors occurred may give a better idea.

25 minutes ago, JorgeB said:

SMART looks fine, full diags when the errors occurred may give a better idea.

Appreciate the assistance. Where would I check for the above (which file, any particular string to search for etc)?

Edited February 5Feb 5 by essjayeire

1 hour ago, essjayeire said:

Or letting me know how I can check for myself.

ST10000NM000G-2MU103_ZS51FVM6-20260205-1032 disk4 (sdd) - DISK_DSBL.txt

Normally the best indication of whether a disk is OK is whether it can complete the Extended SMART test without error.