SMART error (OfflineUncorrectableSector) AND (CurrentPendingSector)

[Clinderw]

Hello,

I have SMART monitoring set up and received two emails just recently that i'm not sure how to interpret. I looked around the forums but I couldn't find a great match. Does this drive need replaced?

First Email: Device: /dev/ada3, 8 Currently unreadable (pending) sectors
Second Email: Device: /dev/ada3, 8 Offline uncorrectable sectors

Context information:

• Build: FreeNAS-9.2.1.3-RELEASE-x64
• Zpool Consists of 2 RaidZ2 VDevs
• System still come through as "healthy"

Thanks,
Chris

 

[joeschmuck]

Well good thing you were smart enough to setup the emails. Your ada3 drive has failed and the second message states it's offline. Good thing you have a RAIDZ2 but you will need to replace the failing drive. Follow the user guide and my advice is to not hot-swap the drive. You are safer to shut down the system and replace the drive.

If you do not know which drive is ada3, use the serial number of the drive. At the shell window type

Code:

smartctl -a /dev/ada3 | more

This will give you the drive serial number. Make sure that it the only drive you replace.

 

[Clinderw]

Thank you for the prompt response. You mentioned it is Offline, it currently states in the UI that it is online - i just wanted to see if that changes your diagnosis at all.

I have preemptively ordered another drive tonight to address this.

Thanks,
Chris

 

[gpsguy]

I think joeschmuck misread your message (as I originally did). My guess, is that your email messages are looking at SMART codes 197 & 198 and look something like the lines below, albeit with 8 problem sectors.

Code:

197 Current_Pending_Sector  0x0032  200  200  000    Old_age  Always      -      20
198 Offline_Uncorrectable  0x0030  200  200  000    Old_age  Offline      -      22

In any event, you should replace the drive. If you'd like to post the results of the "smartctl -a /dev/ada3" here for review, we'll look at it. Please use code tags to preserve the formatting.

 

[Clinderw]

Thank you. Attached is the results - it says the test result:passed. Not sure what else you could help gleam, thanks in advance.

Chris

 

[joeschmuck]

ID 5, 197 and 198 should all be zero values. Anything in these fields are indicators of a pending drive failure. If you leave this drive installed you will have more failure messages. It might be today or a few weeks from now. If you need to order a new drive, today is the day to do it. If you had a RAIDZ1 I would recommend you shut the system down until you have a new drive in hand, depending on how bad you need your NAS to operate. You can also try to RMA this drive and get an advanced replacement if you're willing, assuming the drive is still under warranty. And make sure you have a backup of anything that is important to you.

Yes, I did misread the original message. Thankfully I didn't give bad advice.

 

[Clinderw]

Thank you for the prompt responses and for confirming the diagnosis. I have ordered a new drive off Amazon that will be here tomorrow (slowing moving all HDs to NAS drives). Per Cyberjock's strong recommendation I set up the box using RaidZ2 vDevs so i will leave it on, but thanks for the cautionary recommendation.

Thanks again

 

[joeschmuck]

Cyberjock is a good fella and has very good recommendations based on personal experience. I won't say I agree with everything but 99% of the time I do.

 

[cyberjock]

I disagree with everything joeschmuck says. I mean, who trusts an old man with high tech computers? ;)