With respect to the one hard drive data you posted, the only thing I see that may be alarming is 195 Hardware_ECC_Recovered -O-RC- 008 007 000 where the 000 is the THRESHOLD value for failure, your drive is currently at 008 but has been at 007. The current value can change and go back up, however it is very low right now. It is not a failure, yet, but to me it is an indication that your drive could be failing.
The fact that is passed an Extended test is a very good thing, not ZFS errors either. And what are the results for the other two drives?
I’ve got to say, this thread sounds familiar, like I had a very similar discussion with a person with almost the same exact issue.
My advice, if you have not already done so, backup any important data you have, sooner is better. This way the data will be safe just in case a second drive fails.
@PhilD13
The commonality was the fact that is was three drives and the problematic drive had the similar, if not identical ID 195 data. Very strange and I’m not accusing the OP of posting this twice, it is just strange. But at the end of that conversation, the VALUE was moving up and recovering slowly. I honestly expect that drive to fail. So I would currently advise replacing the drive here as well, but backing up the data is the first step.
I’m not sure it could be RMA’d since it has not actually failed yet.
truenas_output_2.txt (38.5 KB)
It all passed…should I be worrying anyway? They all have that Hardware_ECC_Recovered line, all a bit higher than before.
While the Hardware_ECC_Recovered values are not an actual failure, I find them suspect.
What is the output of smartctl -l farm /dev/sdd and we can compare to see if there is an obvious issue with it being a heavily used drive.
Something else you can do, check the warranty of each drive to ensure they have the correct warranty expiration date. If it doesn’t match, that is a red flag. And you could also contact Seagate and provide them some information (definitely the smartctl -a of the drives) and hope they get back to you promptly. Tell them you are trying to rule out if these were previously used drives relabeled. And
I don’t want to tell you that your drives are not new, I honestly do not know yet, and maybe I will not be able to tell. You may end up using them until one of them fails and send it in for RMA.
