Confusing SMART results

Status
Not open for further replies.

NiceTry

Explorer
Joined
Jun 8, 2011
Messages
62
I am seeing a repeating notice in the system log:
"Aug 22 10:39:00 ServerA smartd[2336]: Device: /dev/ada2, 8 Currently unreadable (pending) sectors
Aug 22 10:39:00 ServerA smartd[2336]: Device: /dev/ada2, 8 Offline uncorrectable sectors"

I ran a smartctl -t long /dev/ada2 that shows no errors.

See the attached smartctl -a and smartctl -x results.

Log still reports problems. Anyone have an idea whats going on here?
 

Attachments

  • smarttest-ada2.txt
    5.5 KB · Views: 226
  • smarttest-x-ada2.txt
    7.5 KB · Views: 259

Robert Trevellyan

Pony Wrangler
Joined
May 16, 2014
Messages
3,778
The system is warning you about the values of attributes #197 and #198. I don't have a good explanation for why the extended test doesn't complain about this.

If you want to force the issue, and you have enough redundancy in your pool, you could offline that drive, move it to another system and run badblocks on it. Otherwise, you can just keep a close eye on that drive's SMART data to see if it gets worse, especially attribute #5.
 

NiceTry

Explorer
Joined
Jun 8, 2011
Messages
62
Thank you Robert.
What exactly am I looking for in #5 and what do I do about it? I think I see that no sectors have been reallocated, right?

The error message is coming from smartd. Does it do anything differently than can be done through smartctl?

I do have a spare disk. Maybe I should just replace the suspicious one and get it over with...
 
Last edited:

joeschmuck

Old Man
Moderator
Joined
May 28, 2011
Messages
10,994
You are looking for any non-zero value in ID 5, or if ID's 197 or 198 start to go up. If this happens then I'd replace the drive.

In my tagline is a link to the Wiki for SMART IDs. Scroll down to the chart and the items highlighted in light red are key things to look for with respect to a failure. There is some nice explanations as well.

I'd also run a scrub for the hell of it for peace of mind.
 

NiceTry

Explorer
Joined
Jun 8, 2011
Messages
62
Thanks Joe and Robert.

After reading Joe's referred chart, I understand what Robert was suggesting and why.

I can see how badblocks -w would force a write and reallocation of affected blocks.

Is it worth reformatting after that and then reinstall/resilver it into the array?

Or did I confuse myself...
 
Last edited:

BigDave

FreeNAS Enthusiast
Joined
Oct 6, 2013
Messages
2,479
Is your spare drive burned in (tested)?
My first effort here would be to protect my volume by replacing the drive with errors.
Then once it's removed, run whatever tests on it you can to determine it's condition
for replacement, RMA or reformat for a spare.
Your volume's health comes first, find out whats wrong with this drive after that...
 

NiceTry

Explorer
Joined
Jun 8, 2011
Messages
62
Good point, Big. I have another available SATA port in my rig. I can replace the suspect drive and put it on the spare SATA to work on it. I don't really have another machine I can do byte-banging with. I'll have to use my FreeNAS OS.

As you say, I might be able to convince myself it is good enough as a spare. BTW, the drive is already a warranty replacement for a failed drive. Not sure I can do the trick twice.
 

joeschmuck

Old Man
Moderator
Joined
May 28, 2011
Messages
10,994
OMG, @BigDave beat me to the answer. I'm still giving it. Your drive has very few hours on it and I'd RMA it. Go for an advanced replacement where they ship you a drive first and you can use the shipping packaging to return the failing drive. I'm being distracted by TV, sorry. I'ts taken me about 20 minutes to get in this reply.
 

joeschmuck

Old Man
Moderator
Joined
May 28, 2011
Messages
10,994
Not sure I can do the trick twice.
Sure you can, look up the Serial Number on the website, should be good to go if you are in the warranty period of the original drive.
 
Last edited:

BigDave

FreeNAS Enthusiast
Joined
Oct 6, 2013
Messages
2,479
OMG, @BigDave beat me to the answer. I'm still giving it. Your drive has very few hours on it and I'd RMA it. Go for an advanced replacement where they ship you a drive first and you can use the shipping packaging to return the failing drive. I'm being distracted by TV, sorry. I'ts taken me about 20 minutes to get in this reply.
I didn't mean to step on your toes Old Man:p I figured you were havin' a nap on this Sunday afternoon;)
I'm watchin' Tigers vs Rangers (box score, not live TV) with one eye of the forums and zzzz'n out from time to time myself:D
 

joeschmuck

Old Man
Moderator
Joined
May 28, 2011
Messages
10,994
I know it's a bit off topic but...

This old man got some work done around the house today and it wore me out, oh how I wish I were in my mid 20's again. I'd be happy to take a nap but I can only do that if I'm sick. Very hard to sleep when the sun is out. The recliner is comfortable but I think it's time to light up the BBQ for the second time for the day. Lunch was good (Chili Cheese Dogs), dinner should be even better (Steak). Yea, I'm a carnivore.
 

NiceTry

Explorer
Joined
Jun 8, 2011
Messages
62
I'll replace the drive with a spare and try to RMA it. The drive is a "refurbished" drive I got from RMA'ing another one. Seagate does not recognize the serial number of the current drive as valid. That's why I'm not sure I will be successful with the RMA. I'll call them instead of just pounding through the website.

Thanks for the discussion, gentlemen.
 

joeschmuck

Old Man
Moderator
Joined
May 28, 2011
Messages
10,994
Good luck on getting the drive replaced.
 

NiceTry

Explorer
Joined
Jun 8, 2011
Messages
62
FYI, gentlemen. Replacement drive installed, resilvered, passed SMART test. No reported errors anywhere. Life is good.

Thanks for the pointers.
 
Status
Not open for further replies.
Top