[Solved] Is my new disk dying?

[raymix]

Hi folks

 

EDIT: Not sure if matters, but drive is brand new Toshiba E300 . I got it because it's supposedly economic on power requirements.

 

 

My disk has been acting very weird lately. 2 days ago this message popped up:

Event: unRAID Disk 1 SMART health [197] Subject: Warning [UNRAID] – current pending sector is 8 Description: TOSHIBA_HDWA120_679U9KHAS (sdd) Importance: warning

Along with this message:

Event: unRAID Disk 1 SMART health [198] Subject: Warning [UNRAID] – offline uncorrectable is 1 Description: TOSHIBA_HDWA120_679U9KHAS (sdd) Importance: warning

 

Minutes later current pending sector became 16 and few minutes later 24. Offline uncorrectable jumped up to 3.

 

After running parity check with correct unchecked I received this:

Event: unRAID array errors Subject: Warning [UNRAID] – array has errors Description: Array has 1 disk with read errors Importance: warning

Disk 1 – TOSHIBA_HDWA120_679U9KHAS (sdd) (errors 224)

No idea where the 224 errors came from, but whatevs. After running it again, this is the error it settled with:

Event: unRAID Disk 1 SMART health [5] Subject: Warning [UNRAID] – reallocated sector ct is 2 Description: TOSHIBA_HDWA120_679U9KHAS (sdd) Importance: warning

Offline uncorrectable and Current pending sectors went back to 0, but I was still stuck with Reallocated sector count 2. I've read on this forum that this warning is ok?

I decided to run yet another (third) array check, but with corrections enabled. 5 hours later nothing changed - still 2 errors.

I approved the reallocated 2 warning, but it has jumped to 3 during last array check and been sitting like that for a day with error below:

Event: unRAID Parity check Subject: Notice [UNRAID] – Parity check finished (2 errors) Description: Duration: 5 hours, 56 minutes, 6 seconds. Average speed: 93.6 MB/s Importance: warning

This warning does not seem to go away, I thought that reallocated sectors warning is not that serious? Yet parity check report says 2 errors, hmm. I am so confused as to what's going on..

I have brand new replacement drive same exact as this toshiba, cables disconnected. Should I go ahead, replace this drive with the new one and send the broken one back to amazon? I've only had it for 20 days. If so, what do I tell them? Send them SMART logs?

 

 

 

unraid-diagnostics-20180328-1832.zip

Edited March 29, 2018 by raymix
solved

[JorgeB]

Pending sectors were reallocated, and currently they are relatively few, but never a good sing on a new disk, depending where you bought it, if you can trade for a new disk do it, if it's RMA only, at least run an extended test, since it might be OK for now and refurbished disks are a crapshoot.

 

 

 

 

[raymix]

4 minutes ago, johnnie.black said:

Pending sectors were reallocated, and currently they are relatively few, but never a good sing on a new disk, depending where you bought it, if you can trade by a new disk do it, if it's RMA only, at least run an extended test, since it might be OK for now and refurbished disks are a crapshoot.

 

 

Hi Johnnie

 

Thanks for suggestion, I have few more quick questions: 

I have "acknowledged" the error, but it still shows up as yellow warning under realocated error count. Is this normal?

Is it safe to swap drives while Parity reports errors?

Does extended test get rid of warnings/fix errors or is it used for SMART logs only?

Oh and - I really love preclear plugin, do you guys think it's worth downgrading to 6.4 just to get it back?

 

Edited March 28, 2018 by raymix

[JorgeB]

6 minutes ago, raymix said:

I have "acknowledged" the error, but it still shows up as yellow warning under realocated error count. Is this normal?

Yes, it should be green on the dashboard.

 

7 minutes ago, raymix said:

Is it safe to swap drives while Parity reports errors?

Not ideal, but it can be necessary, if all pending sectors were reallocated you can do a correcting check, but do that only if the disk doens't fail an extended SMART test.

 

8 minutes ago, raymix said:

Does extended test get rid of warnings/fix errors or is it used for SMART logs only?

It won't get rid of the current errors, but you'll know if the disk is currently OK or not.

 

9 minutes ago, raymix said:

Oh and - I really love preclear plugin, do you guys think it's worth downgrading to 6.4 just to get it back?

IMO no.

[Frank1940]

17 minutes ago, raymix said:

I really love preclear plugin

 

You can always use the preclear script but you will be learning a bit about the command line...

 

P.S.  There are a few of us can still use the preclear plugin but you have to have a rather minimal NAS type setup.  Plus, the Fix Common Problems plugin will complain about it.  

[raymix]

Thanks for replies, guys. Extended test finished with errors (see attached), looks like HDD is indeed dying. Any tips on how to proceed? Should I just swap it like I'd normally do or do I have to be extra careful due to parity errors? It's only a PLEX server.

unRAID Disk 1 SMART health [197]: 28-03-2018 20:16
Warning [UNRAID] - current pending sector is 8
TOSHIBA_HDWA120_679U9KHAS (sdd)

EDIT: Amazon is great, replacement drive is on its way, just need to take this one out, wipe it (wish I had preclear now lol) and send it away.

 

TOSHIBA_HDWA120_679U9KHAS-20180328-2025.txt

Edited March 28, 2018 by raymix

[JorgeB]

Quote

Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error
# 1  Extended offline    Completed: read failure       70%       504         942107520

 

SMART test failed, so it needs to be replaced ASAP.

[raymix]

Thanks for help, guys, I really appreciate your patience with me.

I have replaced drives in the array with a spare and it's rebuilding from parity right now. Faulty drive sits under unassigned devices and keeps spitting errors non stop.

Apart from offline uncorrectable errors there's actually a new one, too lol

seek error rate (failing now) is 106

Super happy to be part of this community and looking forward to find myself on the other side of issue some day helping someone else.

 

I might need to find some external HDD adapter case (the one where you can slot the drive in without screws) and find some windows application that allows me to clear drives outside of array. If you guys know of any software that also writes "preclear" data after zeroing it for speedy drive replacements, I'd be eternally greatful. If not - at least something that allows me to completely wipe them before shipping back for warranty. I see a lot of people are against preclear on this forum and don't care about stressing drives, but for me personally drive preparation and wiping outside of a working array was the reason to have it. Oh well.

[Frank1940]

You might want to look here:

 

       https://lime-technology.com/forums/topic/54648-preclear-plugin/?page=77&tab=comments#comment-646999

 

 

Preclear is back in for ver6.5.0  and I am presently running it (+4 hours in on 3TB)  on version 6.5.1-rc2 and it seems OK at this point.

[raymix]

Thank you so much for heads up, Frank! I am so relieved right now, lol!