Hi!

 

After running with 2 x LSI SAS9217-8i cards flawlessly for some years, I decided to switch to some ASM1166/ASM1064 cards for better power consumption (save on my electricity bill 🙂).

 

It's 2 x ASM 1166 and 1 x ASM 1064 and then eight on-board SATA ports on a Gigabyte C246M-WU4. Unraid 6.9.2.

 

The problem is, now when I run a parity check, I get 5 errors... for some reason on the same sectors.

 

I'm running a second parity check now and the errors haven't come up yet, and I don't think they will.

So it seems like it only occurs on a fresh power up (parity check after a restart).

 

I used to have a similar issue with a AOC-SAS2LP-MV8 (at least similar in appearance to me), where errors would always come up on first parity check after powering on - but I'm not sure if with that card it's always the same sectors.

 

Sep 20 03:09:37 nas2015 kernel: md: recovery thread: P corrected, sector=2743151176
Sep 20 03:09:37 nas2015 kernel: md: recovery thread: P corrected, sector=2743151184
Sep 20 03:09:37 nas2015 kernel: md: recovery thread: P corrected, sector=2743151192
Sep 20 03:09:37 nas2015 kernel: md: recovery thread: P corrected, sector=2743151200
Sep 20 03:09:37 nas2015 kernel: md: recovery thread: P corrected, sector=2743151208

 

Nov 11 02:09:59 nas2015 kernel: md: recovery thread: P corrected, sector=2743151176
Nov 11 02:09:59 nas2015 kernel: md: recovery thread: P corrected, sector=2743151184
Nov 11 02:09:59 nas2015 kernel: md: recovery thread: P corrected, sector=2743151192
Nov 11 02:09:59 nas2015 kernel: md: recovery thread: P corrected, sector=2743151200
Nov 11 02:09:59 nas2015 kernel: md: recovery thread: P corrected, sector=2743151208

 

I have this feeling that if I switch back to the 9217-8i cards the problem will go away.

 

Diagnostics from both runs is attached.

 

Can someone help explain what is happening?

 

Thanks.

 

unraid-diagnostics-20221112-1319.zip unraid-diagnostics-20220920-1809.zip

11 minutes ago, minority said:

but I'm not sure if with that card it's always the same sectors.

They were also always the same sectors.

 

11 minutes ago, minority said:

I have this feeling that if I switch back to the 9217-8i cards the problem will go away.

Mostly likely, looks like a controller problem.

 

Hello JorgeB!

 

Any tips to figure out which of the controllers it is acting up?

Everyone else reports great experiences with these controller cards.

 

Kind regards.

Easiest way would be to swap one of the with the LSI and test.

You can also read this thread:

 

        https://forums.unraid.net/topic/85085-solved-disk-errors/#comment-789393

 

I know that this is several years old now (and is based on an earlier thread that I can't access) but it is worth a try if you don't need IOMMU. 

Thanks both for the reply. I will do some reading and some testing, I was hoping for a less manual way (less time consuming) of diagnosing what is causing the issue. Will post at some point once I get to the bottom of it. The second parity-check (without reboot) completed without finding any errors.

Hey, sorry to push this old thread but I just noticed that I am getting the same error. At the same sectiors:

kernel: md: recovery thread: P corrected, sector=2743151176
kernel: md: recovery thread: P corrected, sector=2743151184
kernel: md: recovery thread: P corrected, sector=2743151192
kernel: md: recovery thread: P corrected, sector=2743151200
kernel: md: recovery thread: P corrected, sector=2743151208

Using an ASM1064 Serial ATA Controller under Unraid 7.2.3

As this issue is still apparent, does anyone have an idea how to solve this? Can a firmware update help in this regard?