I have setup a FreeNAS (11.2-U8) server at home containing 8 drives comprised of 2TB Seagate Barracuda ST2000DM006 drives in a RAID-Z2 configuration.

The rest of the system hardware is as follows:

  • CPU: Intel(R) Core(TM) i3-4170 CPU @ 3.70GHz
  • Power Supply: EVGA 100-W1-0500-KR
  • Motherboard: ASRock C226 (ECC capable)
  • RAM: NEMIX RAM 16GB (2 x 8GB) DDR3-1066MHz PC3-8500 (ECC)

This setup has been running for 2.5 years without issues after which I started to have SATA errors on a hard disk connected to port ADA7. 2.5 years being short in my opinion, I decided to replace the drive a "NAS grade" drive that has a longer waranty, so I bought a WD Red Pro WD2002FFSX.

After successfully resilvering the pool, I started to receive the same type of SATA errors with this new drive after only few days of operations.

I started suspecting the cables so I switched one end of the SATA cable from the drive on ADA6 and ADA7, connecting the problematic drive on ADA6. The errors kept being flagged, but on ADA6 now; eliminating the possibility of a bad cable or even a problem with the motherboard port.

After replugging everything on their respective ports, I then proceeded to switch the power cables of ADA6 and ADA7. The errors were still presents on ADA7 meaning the power cable is clear.

This being said, what's my next step of debugging? I am not an expert on the matter and it seems like either I am very unlucky and got two bad drives or something else is happening and I have no idea what it is.

The error I get are as follow:

Device: /dev/ada7, ATA error count increased from 2960 to 2961
Device: /dev/ada7, not capable of SMART self-check
Pool main state is DEGRADED: One or more devices has experienced an unrecoverable error. An attempt was made to correct the error. Applications are unaffected.

I must mention that all hard drives are SMART capable. Sometimes the NAS is running without problems for few days, then it starts sending me emails repeatedly reporting issues. It really is sporadic behavior.

  • What RAID implementation? (Or does FreeNAS only come with one?) I'm guessing you're using something built into FreeNAS, rather than hardware RAID. RAID 2 seems like an interesting choice. en.wikipedia.org/wiki/RAID#Standard_levels says "s of 2014 it [RAID 2] is not used by any commercially available system." My understanding is this is mainly because RAID 5 doesn't have significant drawbacks compared to RAIDs 2, 3, and 4, and RAID 5 is typically faster in implementation (perhaps typically hardware-based implementations).
    – TOOGAM
    Nov 29 '20 at 2:49
  • Could be the SATA controller / SATA ports. (I understand that testing this theory may mean replacing the motherboard. It is a sensible thing to check, logically. Whether you find that easily feasible or not might be a different story.)
    – TOOGAM
    Nov 29 '20 at 2:50
  • It's a Raid-Z2. Similar to raid-5 but with 2 redundant disk. Raid-Z is the software Raid implemented in freenas. A moderator removed the Z thinking it was a mistake ;). I'll fix this Nov 29 '20 at 2:51
  • @Toogam I connected another disk on the same SATA port, the errors followed the hard drive and changed port Nov 29 '20 at 2:52
  • @Pier-YvesLessard I’m no moderator. Just a community member with high rep who edits questions. Anyone can edit questions. Nov 29 '20 at 2:56

I believe this was a software issue. I can't pinpoint exactly, but running all possible updates on my NAS (Switching to lastest update train 11.3 and running all pool upgrade) seems to have done the trick.

I have been running for almost 2 months without issues since I updated everything.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.