My system is currently running a parity check and I have just received a notification that my parity drive has read errors.

2 things:

This disk is only 3 weeks old

It is a Seagate drive which always reports millions of raw read errors, even when first plugged in

Is this likely just a quirk of the Seagate reporting confusing things? All of the "important" SMART attributes are still reporting 0

I know I should probably run an extended smart test - but not until the parity check is finished.

Or is that not actually necessary with Seagate's quirky reporting?

Thanks

Most likely unrelated, where are the errors reported? Also check syslog for drive errors.

Can be caused by a bad SATA connection and not the drive itself.

Edited July 1Jul 1 by Kilrah

You are likely to get more informed feedback if you attach your system's diagnostics zip file to your next post in this thread.

It is always a good idea to do this to allow us to see the current state of your system and so we can see logs and configuration information.

Sorry, I knew I'd forgotten something - my brain is fried

bob-diagnostics-20260701-1105.zip

It's logged as a disk error, but SMART looks OK. Recommend running another parity check or an extended SMART test.

Erm......help. The parity disk has now gone into error state and says disabled

Latest diagnotics attached. I have absolutely no idea what to do now :-/

bob-diagnostics-20260701-1357.zip

You can still run

2 hours ago, JorgeB said:

an extended SMART test.

Presumably stop the parity check (which seems to be paused) first? That might be a stupid question but I don't want to do anything wrong!

You can, but it's not required to run the SMART test.

OK, thanks. I presume let that run and post the results in about 16 hours.......unless it blows up sooner :-/

SMART test claims to have passes, logs attached

bob-smart-20260701-1430.zip

What do I do now?

Swap both cables with a different disk, re-enable the disk, and see if there are any more errors:

https://docs.unraid.net/unraid-os/using-unraid-to/manage-storage/array/replacing-disks-in-array/#re-enabling-a-disabled-disk-rebuilding-onto-itself

at the risk of asking a stupid question............if it is a dodgy cable, and that is now attached to another disk, will that not mess up the parity build?

1 hour ago, Stannis33 said:

at the risk of asking a stupid question............if it is a dodgy cable, and that is now attached to another disk, will that not mess up the parity build?

At the moment the parity already appears it could be messed up so will need rebuilding (or at least checking) anyway. I think the key point at this stage is to identify if you have an identifiable external issue such as cabling or power since passing the extended SMART test normally mean the drive is healthy.

I think the parity needs rebuilding to get it out of error state........what I do not understand is......if it tries to rebuild, and it can't complete because it now cannot read one of the other disks (because of the potentially wonky cable) and I end up with both a parity and data disk in error state am I screwed?

Or can that mess be got out of as there is nothing wrong with my data disk or its contents (I hope!) - presumably by getting new cables and then trying again?

Edited July 2Jul 2 by Stannis33
Typo

7 minutes ago, Stannis33 said:

Or can that mess be got out of as there is nothing wrong with my data disk or its contents (I hope!) - presumably by getting new cables and then trying again?

This would be the approach. Note that sometimes new cables are not needed - existing cables might just need seating more firmly.

The other possible issue relates to power to the drives. Do you by any chance use power splitters with the drives - if so give some details.

Erm this is where I run into a slight snag........ I did not build this machine myself, I don't really know what is plugged in, where or how. I am trying to figure it out but I am a long way from understanding! :-/

it does appear to use power splitters though.

26 minutes ago, Stannis33 said:

it does appear to use power splitters though.

If using SATA->SATA splitters then no SATA connection from the PSU should be split more than 2 way. If using Molex->SATA you can normally get away with 4-way.

Thanks for the help everyone - I found some unused cables inside the sever, plugged those into the disk and the parity sync suceeded

As this disk has now passed 2 extended SMART tests and built parity twice in the last month is it likely that the disk itself is probably fine and this was a connection issue - be it bad cables, me not plugging it in properly when I swapped the disk, other?

Most likely a power/connection issue, but keep monitoring for any more errors for now.