One pending sector: Anything to worry about?

thatflashcat

Explorer
Joined
Aug 9, 2018
Messages
85
I'm running six Seagate Constellation ES.3 drives in RAIDZ2 and recently had to replace one due to a high count of unreadable sectors. Now another of my drives is showing a single unreadable sector. I'd feel bad exchanging another drive with the seller so is there any way of repairing the bad sector? Such as taking it offline, running bad blocks, and resilvering it? I can include smartctl output as well if need be.
 

Chris Moore

Hall of Famer
Joined
May 2, 2015
Messages
10,080
I'd feel bad exchanging another drive with the seller
If these drives are so new that the seller is still honoring an exchange, you should take advantage of that. In the US, and it varies by seller, but you can usually only exchange a product in the first 30 days. Some places give a bit more time, but after that you are down to the manufacturer warranty. Seagate Constellation drives should have a five year warranty, so that would be the next course of action, if this doesn't clear up in a day or two.
Here is a link to a guide:

Hard Drive Troubleshooting Guide (All Versions of FreeNAS)
https://forums.freenas.org/index.ph...bleshooting-guide-all-versions-of-freenas.17/
 

DrKK

FreeNAS Generalissimo
Joined
Oct 15, 2013
Messages
3,630
I understand where you're coming from , OP. I feel the same way. I know retailers are eating costs and/or passing them on, so I like to be very judicious about being a participant in overly zealous returns.

But, still, my tolerance for unreadable sectors in a new hard drive is 0.000000000000000000000000. A single unreadable sector in a new drive, and it's getting replaced.
 

thatflashcat

Explorer
Joined
Aug 9, 2018
Messages
85
I understand where you're coming from , OP. I feel the same way. I know retailers are eating costs and/or passing them on, so I like to be very judicious about being a participant in overly zealous returns.

But, still, my tolerance for unreadable sectors in a new hard drive is 0.000000000000000000000000. A single unreadable sector in a new drive, and it's getting replaced.
The strange thing is the error appears to have disappeared. I'm running a smart long test now and I'll check in the morning.
 

Chris Moore

Hall of Famer
Joined
May 2, 2015
Messages
10,080
The strange thing is the error appears to have disappeared. I'm running a smart long test now and I'll check in the morning.
On occasion, a drive will internally retest a pending sector and clear the error without user intervention. I have seen that a few times on new drives. A pending sector will be marked and later it will clear up on it's own. You might want to keep an eye on it, but it will likely run for years more before it has any further errors. We still must be vigilant, but modern drives are actually quite reliable.
I commissioned a new server at work back around the end of November 2018 that has 60 of the 10 TB Seagate Exos drives in it and through all the burn-in testing and after loading 350TB of data into it, I only had one drive pop a pending sector and, like yours, it cleared on it's own. Two years ago, I went through a similar process with a server that had 60 of the Western Digital 6TB Red Pro drives and in the first three months I had to replace three drives. It is always interesting to see what happens.
 

thatflashcat

Explorer
Joined
Aug 9, 2018
Messages
85
On occasion, a drive will internally retest a pending sector and clear the error without user intervention. I have seen that a few times on new drives. A pending sector will be marked and later it will clear up on it's own. You might want to keep an eye on it, but it will likely run for years more before it has any further errors. We still must be vigilant, but modern drives are actually quite reliable.

That's what I had hoped but I ran the smart long test and now I have another pending sector:

1549985812843.png


Bear in mind this drive has only 1200 power hours.
 

Chris Moore

Hall of Famer
Joined
May 2, 2015
Messages
10,080
That's what I had hoped but I ran the smart long test and now I have another pending sector:

Bear in mind this drive has only 1200 power hours.
What about Reallocated Sectors?
 

Chris Moore

Hall of Famer
Joined
May 2, 2015
Messages
10,080
Still at zero.
For now, I would just keep an eye on it as long as it is pending. If it becomes reallocated, that is time to get a warranty replacement for sure, but if it is still doing this after a few weeks, I would get to the warranty replacement.
 

thatflashcat

Explorer
Joined
Aug 9, 2018
Messages
85
For now, I would just keep an eye on it as long as it is pending. If it becomes reallocated, that is time to get a warranty replacement for sure, but if it is still doing this after a few weeks, I would get to the warranty replacement.

Will do, thank you!
 

thatflashcat

Explorer
Joined
Aug 9, 2018
Messages
85
Update: The number of pending sectors had decreased from 2 to 1, and I will continue to monitor it.
 

Chris Moore

Hall of Famer
Joined
May 2, 2015
Messages
10,080
Update: The number of pending sectors had decreased from 2 to 1, and I will continue to monitor it.
Did you get a reallocated sector out of that?
 

MSameer

Dabbler
Joined
Feb 12, 2019
Messages
14
I had the same issue with my shiny new WD RED during the initial burn in.
I got an alert from freenas about a 4 unreadable sectors
Code:
* Device: /dev/da0 [SAT], 4 Currently unreadable (pending) sectors

I canceled the burn in and restarted the process again. The sectors immediately disappeared.
The reallocated sectors is still at zero.
I highly doubt my reseller will accept returns so I am just keeping the drive. A bit unhappy with WD though :/
 
Top