I've had a problem since I started using Unraid where it seems that every single drive I put into my array has increasing CRC error counts. I'm using a 36-bay supermicro server, and I've tried swapping bays and have only used brand new drives in the array, but every single disk shows increasing crc errors whenever I run backups to Unraid from my synology boxes. What should I do to correct this problem? I have tried re-seating the SATA cables, but that didn't seem to do anything. Do I need to get all new cables? Do I need to replace the LSI controller? Can I just disable that error somehow? Everything else seems to work just fine, so I'm trying to find a solution. I have uploaded my diagnostics file here for reference. Thank you!

mpwc-unraid-diagnostics-20230626-1013.zip

Syslog rotated so cannot see current LSI firmware, but if not the latest start by updating it.

2 hours ago, JorgeB said:

Syslog rotated so cannot see current LSI firmware, but if not the latest start by updating it.

Ok I'm trying to update the firmware to see if it works. I'm trying to follow your instructions for above for my AOC-S3008L-L8E HBA card, but when following them I can't seem to get my sasflash.efi file to execute. I did the chmod +x command on it after copying it to /lsi like you mentioned, but after doing that and trying to run the -listall command I get an error saying "no file found". I'm still relatively very new to linux, so I was wondering if you had any idea why please?

If you follow the instructions correctly it should work, you can post below every step you are doing to see if the issue can be spotted.

I was able to successfully update my firmware, thank you for the instructions! However, I still seem to be getting increasing error counts after the update. What would be the next possible thing to check to remedy the error counts, please?

Next would be cables or the HBA itself, whichever is easier for you to test.

If rear 12 bay was connect to front 24 bay expander BP, then try to connect rear BP directly to LSI and test problem stop or not. ( Bypass front BP ).

Edited June 27, 20233 yr by Vr2Io

Well something seems to have worked. After the firmware update, I replaced the thermal paste on the heatsink of the LSI with a new thermal pad, moved it to a different PCI slot, and reseated the cables. Since then I transferred 2 TB of new data to the array, and where before I would've had a lot of new CRC errors, I haven't had a single new one pop up! Thanks so much for all of your help!