Possible xfs filesystem issues

[David13858]

Hi All,

 

I'm not quite sure what issues I am having but my system isn't behaving as it should, I think it could be an issue with a drive or the file system.

 

A few weeks ago I had a drive failure, did the standard procedure and replaced the failed drive. No issues from what I could see. This week I had another failure. Stopped the array and replaced the drive. When I turn the server on the array automatically starts and the array says I am missing disk 11 (The failed drive) I cannot stop the array as it will just error out indefinitely.

 

When I enter the command : xfs_db /dev/md1

 

I get the response

/dev/md1: No such file or directory

fatal error -- couldn't initialize XFS library

 

I tried to run the xfs_repair command today and the only response I got back was something like "no secondary super blocks found" I forgot to note down the exact message...

 

Diagnostics are attached below.

 

plex-diagnostics-20230826-1925.zip

[JorgeB]

Emulated disk11 is mounting, why to you think you have a filesystem problem?

 

If this unassigned disk was disk11 it needs to be replaced, it's failing badly:

 

Model Family:     Seagate Barracuda 7200.14 (AF)
Device Model:     ST1000DM003-1ER162
Serial Number:    Z4YB2L45

 

 

[David13858]

I forgot to include more information that is probably important.

 

Yesterday I ran into an LZMA error, data is corrupted. I followed a forum post reinstalling UNRAID and copying the config files. That got the server back up and running again.

 

The drive you mentioned above ending in 45 is the drive that has just been put in the server. That should be in full working order. When I stop the array and try to select the drive the option suddenly disappears (As seen in the screenshot). This is the second drive I have tried as well so it is unlikely that both ‘new’ drives would be failing. (That said they are technically used so it is possible)

[JorgeB]

7 hours ago, David13858 said:

Yesterday I ran into an LZMA error, data is corrupted

That's usually flash drive related.

 

7 hours ago, David13858 said:

The drive you mentioned above ending in 45 is the drive that has just been put in the server.

Taat disk is not good, you need a new one, then rebuild disk11.

[itimpi]

10 hours ago, David13858 said:

The drive you mentioned above ending in 45 is the drive that has just been put in the server. That should be in full working order.

I would disagree.   The SMART information for the drive shows:

5 Reallocated_Sector_Ct   PO--CK   061   061   010    -    7864

and

197 Current_Pending_Sector  -O--C-   001   001   000    -    63632
198 Offline_Uncorrectable   ----C-   001   001   000    -    63632

On a ‘healthy’ disk these should ideally be 0’s.

[David13858]

Ok so I guess you guys were right, both the drives I tried appeared to be DOA (Inherited from a few old machines). Inserted another new drive and rebuilt the data.

 

I am now back to 23 drives all functional.

 

The part I still don't understand is the responses I get when I type the below command.

 

I know Defragging doesn't matter that much (if at all) but It's still something that makes me think there are some issues.

 

 

Is this something irrelevant or should I be worried about it ?

 

I have attached the most recent diagnostics.

plex-diagnostics-20230827-1957.zip

[itimpi]

4 minutes ago, David13858 said:

Ok so I guess you guys were right, both the drives I tried appeared to be DOA (Inherited from a few old machines). Inserted another new drive and rebuilt the data.

 

I am now back to 23 drives all functional.

 

The part I still don't understand is the responses I get when I type the below command.

 

I know Defragging doesn't matter that much (if at all) but It's still something that makes me think there are some issues.

 

 

Is this something irrelevant or should I be worried about it ?

 

I have attached the most recent diagnostics.

plex-diagnostics-20230827-1957.zip 273.37 kB · 0 downloads

You are on a 6.12.x release so the correct device name would be /dev/md1p1

[David13858]

Ah, that works. I had no idea the device name changed.

 

It looks like that is me up and running again, Thank you.

[itimpi]

6 minutes ago, David13858 said:

Ah, that works. I had no idea the device name changed.

 

It looks like that is me up and running again, Thank you.

I believe that is part of getting ready for later releases when Unraid will not be restricted to using partition 1 on a drive.