Hello as the title says, I am wondering / worried that I have some discs that are corrupt / failing in my array. The following errors are the ones in question.

 

Aug 23 16:04:55 Tower kernel: I/O error, dev sdc, sector 406173464 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 2

And

Aug 27 14:16:53 Tower kernel: I/O error, dev sdb, sector 1169562704 op 0x0:(READ) flags 0x0 phys_seg 32 prio class 2

I ran a SMART extended test on /dev/sdc and it completed with no errors, although these errors keep appearing. Not too sure what additional information is required. I have attached my diagnostics as well. I appreciate any help / feedback.

Thank you!

tower-diagnostics-20240827-1850.zip

Looks more like a power/connection issue, and since both disks have a lot of UDMA CRC errors, the miniSAS cable would be the first suspect.

5 hours ago, JorgeB said:

Looks more like a power/connection issue, and since both disks have a lot of UDMA CRC errors, the miniSAS cable would be the first suspect.

Hello JorgeB, So the UDMA CRC errors as far as I know were caused by a faulty SFF -> SATA Cable, I had picked up an HBA a couple years ago now I wanna say and one of the cables I had grabbed wasn't working properly and would spit out the UDMA errors. I replaced the cable back then and as far as I know, the errors haven't gone up. I have been keeping an eye on the numbers as well.

I can try replacing the cable and see if that solves the issue if you suspect that it is causing the issues, Just thought i'd mention that I did in fact have issues with the cable at one point but replaced it back when I first installed the HBA.

 

Thanks again!

If the CRC errors have been stable, it could also be a power problem.

Hmm ok, what would be the best way to diagnose a power problem? I haven't made any changes to my setup. although I assume the power supply could be going bad or something along those lines? not too sure. I took out my GPU a few months ago and installed the MediaSonic Probox with an eSata card, I doubt that's causing any issues as the drives are powered by the Probox and the impacted drives aren't in the Probox, but thought I'd mention it. I was thinking about running xfs_repair on the impacted drives and seeing what it returns but not too sure if that is worth doing or not, Pretty lost at this point haha..

Thank you for the help JorgeB, I appreciate it!

Could be the power suplly, or a power connector, especially if splitters are involved, you can swap PSU or just the power cables and see if the problem follows the disk or the plugs.

5 hours ago, JorgeB said:

Could be the power suplly, or a power connector, especially if splitters are involved, you can swap PSU or just the power cables and see if the problem follows the disk or the plugs.

Ok I think I am going to try to swap the cables positions and if the errors come up on the drives that I swapped to its safe to say its a cable issue? The errors are back again (on the same drives) and seem to be climbing slowly. Checked syslog and there is this, but don't think it really tells me anything I didn't already know.
 

md: disk4 read error, sector=4571643616

I don't have a spare PSU at the moment, although if I suspect that is the issue I have no problems grabbing a new one, just don't want to lose any data.

Thank you for the suggestions JorgeB, I appreciate it!

@noraaWas the cause identified? I am having a similar errors.

@QuarantineDrive Sorry I am just seeing this as I actually just started to get them again... I am not entirely sure what is causing them either 😧, I had swapped the cables and everything has been fine for 2+ months and now all of a sudden they are back. I am again worried about losing data and wondering if I should just get rid of the drive at this point. I really am stuck on what to do here haha..

I am curious though, have you found a solution to the problem? Would really appreciate it if you could share the solution if so.

Sorry again for not seeing this reply originally 😧