Read errors on parity rebuild..

[DevXen]

So I'm rebuilding my parity on one of my 2 drives that was disabled earlier today for free l read errors and not 3a 3rs one has had read errors. Not sure what to do..

[DevXen]

Oh here's the diagnostics

mediaxen-diagnostics-20231205-1818.zip

[Sardine8207]

I've had read errors multiple times, in my experience it's always been due to those fragile SAS breakout cables going bad.  I had my server wall mounted, and the drive trays would slip down occasionally, pinching/chafing the data cables.  You will get some better advice here soon enough.

 

Have you run a smart report on the disk?

[DevXen]

11 minutes ago, Sardine8207 said:

I've had read errors multiple times, in my experience it's always been due to those fragile SAS breakout cables going bad.  I had my server wall mounted, and the drive trays would slip down occasionally, pinching/chafing the data cables.  You will get some better advice here soon enough.

 

Have you run a smart report on the disk?

 

No. I tried with the first ones that had write errors and the smart tests stuck At 100% and would never finish. But the diagnostics have the smart info for each drive I saw so that's good. 

 

Here's a little back story... (Didn't put it here cause I already posted a different post about it. But here...

 

 

On Sat I had 2 drives disabled for write errors. Then I swapped them out and rebuilt data separately on them. Took one out ran an extensive chkdsk looking for bad sectors. It didn't find any. The second one still has like 6 days to go on it's check but so far no errors. 

And then today I woke up to 2 drives disabled due to read errors. One brand New one I replaced on Sat and a different drive in the array.

 

About 3 hours into the rebuild on one of the drives I got 2 more drives with read errors.

 

So at that point I turned my server off. Thinking it's the hba controller card or the cables. But it would be strange for multiple cables die at the same time. They are really thick cables. Here's a pic. 

[JorgeB]

Looks like the same issue as in your other thread, controller problems:

 

Dec  5 18:15:05 MediaXen kernel: aacraid: Host adapter abort request.
Dec  5 18:15:05 MediaXen kernel: aacraid: Outstanding commands on (4,1,21,0):
Dec  5 18:15:05 MediaXen kernel: aacraid: Host bus reset request. SCSI hang ?
Dec  5 18:15:05 MediaXen kernel: aacraid 0000:82:00.0: outstanding cmd: midlevel-0
Dec  5 18:15:05 MediaXen kernel: aacraid 0000:82:00.0: outstanding cmd: lowlevel-0
Dec  5 18:15:05 MediaXen kernel: aacraid 0000:82:00.0: outstanding cmd: error handler-5
Dec  5 18:15:05 MediaXen kernel: aacraid 0000:82:00.0: outstanding cmd: firmware-37
Dec  5 18:15:05 MediaXen kernel: aacraid 0000:82:00.0: outstanding cmd: kernel-0
Dec  5 18:15:05 MediaXen kernel: aacraid 0000:82:00.0: Controller reset type is 3
Dec  5 18:15:05 MediaXen kernel: aacraid 0000:82:00.0: Issuing IOP reset