Hi there, I was on 6.4 (I believe) for the longest time and recently updated about a month ago. I started noticing disk errors starting to pop up, got concerned and updated to the latest (6.6.6). But the errors are still continuing to increase but the SMART data is showing no errors logged. The parity drive is a bit more than a year old (8tb wd red), while the remainder are nearly 4 years old (1-3 enterprise drives, 4-6 nas drives). Should I be concerned and if so, what should I do. It literally went from 0 across all drives to all drives showing errors. Thanks in advance!

 

System: 

Dell LSI 9210-8i (not in use).

I/O crest Sata 3, 2 port ASM-1061 (gives connection to ssd and parity drive).

the remaining 5 drives are powered through the board: supermicro x8dtl-if

 

 

Post up the Diagnostic Report.         Tools   >>>  Diagnostics   

 

I am assuming that you have not rebooted since the last Parity check.  Hopefully, there will be some clues as to the source of the problem.  The Parity check speed has dropped significantly also.   

2 hours ago, Frank1940 said:

Post up the Diagnostic Report.         Tools   >>>  Diagnostics   

 

I am assuming that you have not rebooted since the last Parity check.  Hopefully, there will be some clues as to the source of the problem.  The Parity check speed has dropped significantly also.   

I haven't rebooted since the last parity check. Attaching the report. 

tower-diagnostics-20190123-2047.zip

First thing to do is to change the onboard controller from IDE to SATA, though likely that's not what's causing the errors, still run a non correcting check after that, if errors continue post new diags, but it could be a hardware problem, like a bad PSU, bad board, etc.

10 hours ago, johnnie.black said:

First thing to do is to change the onboard controller from IDE to SATA, though likely that's not what's causing the errors, still run a non correcting check after that, if errors continue post new diags, but it could be a hardware problem, like a bad PSU, bad board, etc.

I'll hunt around for a vga cable to change the setting. Would it make a difference if I just shift everything over to the LSI card and then did a test (effectively cut out the mobo sata connectors from the equation)? I have a spare board but not a psu. I'm not understanding how it can just start happening all of a sudden, like a switch was flipped. 

1 hour ago, bbyf16 said:

Would it make a difference if I just shift everything over to the LSI card and then did a test (effectively cut out the mobo sata connectors from the equation)?

That's also a good test.

On 1/24/2019 at 1:31 PM, johnnie.black said:

That's also a good test.

So I tried running a parity check without corrections one more time before doing anything and it was running slower than molasses. Did a diagnostic report. Pulled out all the sata cables, and connected two breakout cables from the lsi sas card to the 8 drives. Ran a parity check with no error correction and it's coming back clean with the fastest time to boot. 

 

Questions:

1.) was it the sata cables? I don't think its the board since two cables were coming from the sata 3 card and not the board. 

2.) Am i getting my hopes up and will the errors come back?

3.) Any idea what went on?

 

 

Pre-LSI diag.zip

Post-LSI diag.zip

Looks to me more like a motherboard (or just the board controller) problem, yes, there were also some errors on parity disk, which was connect to the Asmedia controller, but after a bunch of NMI events, and very few compared to the other disks, also log was filled with call traces and other errors, it's completely clean with the LSI.

 

You could for example move a couple of disks to the Asmedia controller and run another check, if there are no issues move the same disks to the onboard controller, if errors start again that's the problem.

23 hours ago, johnnie.black said:

Looks to me more like a motherboard (or just the board controller) problem, yes, there were also some errors on parity disk, which was connect to the Asmedia controller, but after a bunch of NMI events, and very few compared to the other disks, also log was filled with call traces and other errors, it's completely clean with the LSI.

 

You could for example move a couple of disks to the Asmedia controller and run another check, if there are no issues move the same disks to the onboard controller, if errors start again that's the problem.

Thanks for the assistance!