Hi Guys,

so ive had nothing but dramas with my server since i upgraded over the past few weeks, most likely user error by not understanding certain unraid requirements, i had files moving all over the show, i think the issue stemmed from most my drives being close to full.

 

I purchased a new 16TB drive to replace an 4TB. i rebuild the drive from parity, and all seemed good.

I then seemed to have had a power cut a week later. which i think may be the culprit of my server having issues, when i got home and booted it back up, its stated the new drive was unmounted.

So i stopped the server, re-mounted and started the rebuild. this took almost 2 days, the rebuild is done, but the drive still says its not mounted, ive stopped the array, if i put it into maintenance mode, it mounts with no errors, but when i start the server up as normal, i get the unmounted error. if i go into maintenance mode and click "Mount" on the unmounted device, i get this error : Failed: Device "/dev/sdd" Failed to mount. Check the syslog for details.

Ive attached the log, as im struggling to navigate where the issue is.

 

If i click on the "remove" partition option on the unasigned disk, and then re-asign, will this wipe the disk and then rebuild again, and hopefully work?

 

Im hoping my parity is fine, as i think it would be, the SMART report also attached i cant see any issues with the disk.

 

help appreciated.

 

matrix-diagnostics-20241216-1824.zip matrix-smart-20241216-1741.zip

Unmountable disks are usually fixed by checking filesystem, not rebuilding, the syslog rotated, reboot and post new diags after array start.

Hi JorgeB.

 

sorry i started a rebuild again before i seen this reply. takes over 2 days to rebuild lol so i was waiting.

i restarted the server just now, and still wont mount.

please see attatched logs

 

Regards

matrix-diagnostics-20241218-1806.zip

33 minutes ago, Matrix-B said:

Hi JorgeB.

 

sorry i started a rebuild again before i seen this reply. takes over 2 days to rebuild lol so i was waiting.

i restarted the server just now, and still wont mount.

please see attatched logs

 

Regards

matrix-diagnostics-20241218-1806.zip 206.93 kB · 0 downloads

If a disk is unmountable before you start a rebuild it will also be unmountable afterwards as all the rebuild process does is make the physical drive match the emulated one.

 

The syslog shows you have XFS level corruption on disk4 so you need to run a check filesystem against disk4 (from the GUI) and run it without -n if you want a repair to be done rather just a check be made.

2 hours ago, itimpi said:

If a disk is unmountable before you start a rebuild it will also be unmountable afterwards as all the rebuild process does is make the physical drive match the emulated one.

 

The syslog shows you have XFS level corruption on disk4 so you need to run a check filesystem against disk4 (from the GUI) and run it without -n if you want a repair to be done rather just a check be made.

Hi itimpi.

 

ive looked for the above, and read the link, but i dont have the option to do this via GUI. please see attatched screenshot

 

Click on "Check"

i did, but i just get an endless supply of dots that never finishes... is this normal?

That means there's no valid filesystem on that disk, we can't see what happened because the syslog rotated, but possibly the rebuild was incorrectly done, or there were errors.

the rebuild from parity said it completed with no error, but the disk wont mount.

whats the options? format and just lose the 4gb of data and just deal with it? if i biuy a new HDD will it rebuild the parity onto that or is this a parity issue? as im foine with buying a new HDD and putting that in and then wiping this one and using this in another slot.

Options are basically: format and restore from a backup, or use a file recovery app like UFS explorer to see if it can recover any data.

matrix-diagnostics-20241216-1757.zipi found this report, not sure if this says anything different? it was a few hours i think before the first report.

4 minutes ago, JorgeB said:

Options are basically: format and restore from a backup, or use a file recovery app like UFS explorer to see if it can recover any data.

i think the only backup i have is the parity. i do have the 4TB HDD i removed and replaced with the 16TB that is literally full, the new parity rebuilt that 4tb onto the 16TB, so im guessing the 4TB HDD has all the original files lost on the 16TB other than anything added in the past 2 weeks, which is manageable rather than losing z whole 16TB.

 

whats the best way of putting the 4TB back in and creating a new parity, and then putting out the 4TB and replacing with the 16? i dont want to put in the 4TB and then lose that data if possible

 

appreciate the help as always.

 

p.s i dont have enough cabling free to slot in the 4TB along side the 16TB at once.

 

Edited December 19, 20241 yr by Matrix-B
P.S

Diags don't really help since the syslog was being spammed.

 

If you have the old 4TB disk, easiest would be to format the new disk and copy the data from the old disk, but if you cannot have both connected together, you can also do a new config with the old disk and resync parity, and you can then try the replacement again.

1 minute ago, JorgeB said:

Diags don't really help since the syslog was being spammed.

 

If you have the old 4TB disk, easiest would be to format the new disk and copy the data from the old disk, but if you cannot have both connected together, you can also do a new config with the old disk and resync parity, and you can then try the replacement again.

awesome cheers mate. ive just carried out the format. at least now i have space lol, i think ill just get a sata PCI expansion tomorrow, and then plug in the 4TB and try and move the data from the drive to the 16TB.

 

what would be the best way of doing this? could i literally plug the 4TB in, and mount is like a USB drive to pull data from it through Krusader?

likely also to but a UPS tomorrow as this is a headache, and wont to minimise this happening again. lol

also. just check my log, and can see this message ive not seen before. do you know what this means?

i did a Smart can on my NVME cache drives and it came back good with no errors.  i also thought i had 2x500gb nvme's but looks like 1? has one failed?

 

 

You can use the UD plugin to mount it and your favorite tool to copy

 

Those nginx errors should be harmless