Drive Critical, questions before replacing it

Currently reading
Drive Critical, questions before replacing it

2 days ago DSM messaged that one of the drives is in Critical state (DS920+, DSM7.2, 2x 4Tb ironwolf, SHR).

However storage pool is "Healthy".

I ordered a new drive, that just arrived, now the message has repeated and storage pool status is "Warning".

On the drive health:
  • SMART quick test returns passed
  • SMART extended test returns "Failing"
  • Ironwolf health return "Normal"

This is my first drive that is failing, I think I how to replace it: disable drive in DSM, remove, insert new one, repair array

I have two doubts however:
  • Should I wait until the array crashes or replace immediately?
  • The failing drive is inside warranty from Seagate (checked through SN in their website). What "evidence" should I collect before removing the drive to sustain the claim for the warranty?
 
Solution
Don't wait, why should you?
Typically, the extended smart printout showing the fail should be sufficient evidence for a replacement. And you still have the disk....
Don't wait, why should you?
Typically, the extended smart printout showing the fail should be sufficient evidence for a replacement. And you still have the disk....
Thanks, repair is already ongoing.
I've made dome Screenshots of DSM Messages to send to Seagate.
-- post merged: --

I connect it to a PC and test it using the vendor's diagnostic software.
Will do, thanks!
 
Upvote 0
So, the drive replacement and rebuild went just fine.

As suggested by @Telos I removed the failing drive from the NAS and connected to the desktop PC and ran the "Seatools" software (which is available on linux, nice!)

Sure enough, a long self test and long generic test generated errors. Then I ran the long "Fix All" tool, which completed fine, and after that the same tests that used to fail now do not fail.

I've reinstalled the drive on the DS920 and now the drive shows healthy after a long SMART test (which is what used to fail).

I've left the drive on the NAS but, for now, I haven't added it to any storage pool. I'm not sure if I can trust it, but it seems wasteful just to not use it, as the drive seems to work. And I don't know if Seagate would accept a warranty claim in this situation...

Any thoughts? Has anyone had a similiar experience?
 
Upvote 0
Last edited:
Had same thing happen with 1 of 2 WD drives in SHR years ago, except that a couple months after ‘fixing’ it, then re-using it…. It failed again.
On removal a 2nd time WD Software found errors yet again.
So…..

You have to balance errors and ‘fixes’ in comparison with the cost of new drives…..

What Synology should do is re-enable SSD as an upgrade to HDD’s, as in the next few years, what I ran into ( no SSD upgrade path for 2 drive system's: you must start over as if new!)….

THIS IS GOING TO AFFECT EVERYONE IN THE FUTURE!!
 
Upvote 0
HDD errors can repaired and they will be marked bad on the disk. In my experience, most of the time, the disk will start developing errors again within a short time. I would have returned the disk.
 
Upvote 0
Last edited:
I would have returned the disk.
Maybe for hard sector errors but as no SMART details were provided, it could have simply been soft errors. A SMART report would indicate if sectors were reassigned.

Warranty returns are a hassle as you often get refurbished drives, and then must forensically wipe the drive you are returning... meanwhile redundancy is unavailable for quite some time. Often it's best to have a spare on hand until things are sorted out.
 
Upvote 0
If its me, I would go the RMA route with insistence on "do not send me a refurb drive replacement" - we all have opinions, IMO once bad sectors start to appear, it is a matter of time before the situation gets worse. As Telos suggests, an on hand cold spare drive is always a good idea.
 
Upvote 0
Maybe for hard sector errors but as no SMART details were provided, it could have simply been soft errors. A SMART report would indicate if sectors were reassigned.
In this particular case I don't think sectors have been reassigned.

Current SMART data for the failed drive has the attribute "Reallocated_sector_Ct" at the RAW value of 100, this is the same value as the replacement (third) disk has, and also the same value the other older disk in the array which has been running for over 2 years also. All of them are Seagate Ironwolfs 4Tb (non pro)

The seatools log file returned two types of error depending on the test performed:
  • For long self test: rive /dev/sg3, DST failed status 7, returnValue 0
  • For long generic test: Device /dev/sg3, Read LBA Failed for LBA block 5022592 - 5022720
 
Upvote 0
Last edited:
I can tell you that back ‘When’ Drive Smart data was allowed, WD Support would not trust Synology’s prompts about drive errors. I had to go through a second WD Support person in order to get a warranty replacement drive.
All 3 RED drives eventually failed. Yes, even when tested with WD Software.

This was after I retired from TV, where I had a good report with Seagate Support, and cost for storage was not an issue.
I have since given up on relying on Cost of Drive, even because I’m now retired….. That is no longer an issue for me, and Wife is in total agreement with me!
Yes, I’ve become a Seagate fanboy!

I have learned the hard way!
Yes! I have said it before, I’ll say it again:
Seagate has been making drives for 35 or more Years! Synology has been making drives for 35 or so Months!!!
 
Upvote 0
I can tell you that back ‘When’ Drive Smart data was allowed, WD Support would not trust Synology’s prompts about drive errors. I had to go through a second WD Support person in order to get a warranty replacement drive.
All 3 RED drives eventually failed. Yes, even when tested with WD Software.

This was after I retired from TV, where I had a good report with Seagate Support, and cost for storage was not an issue.
I have since given up on relying on Cost of Drive, even because I’m now retired….. That is no longer an issue for me, and Wife is in total agreement with me!
Yes, I’ve become a Seagate fanboy!

I have learned the hard way!
Yes! I have said it before, I’ll say it again:
Seagate has been making drives for 35 or more Years! Synology has been making drives for 35 or so Months!!!
Synology does not "make" drives. They source from other manufacturers, apply proprietary firmware, and then rebrand and sell those drives at a marked up cost.
 
Upvote 0
I say again, Synology does NOT manufacture drives. applying custom firmware to a 3rd party drive and rebranding it does not mean they "made" the drive.

Same thing applies to RAM. Synology does not manufacture that either.
 
Upvote 0
But no one else will take ownership once firmware is changed. At that point the finger pointing begins, and like it or not, They own the drive by default
There is no difference if it is Avid, or Discrete Logic, or Synology… no drive mfg will stand behind the drive if it’s not their firmware.
 
Upvote 0

Create an account or login to comment

You must be a member in order to leave a comment

Create account

Create an account on our community. It's easy!

Log in

Already have an account? Log in here.

Similar threads

Glad to hear that your system is running fine :) and, yes, Synology can be slowish when it comes to disc...
Replies
8
Views
726
You can add the new drive and change the RAID type to RAID1 (which is what SHR is in a 2 drive array).
Replies
3
Views
1,106
  • Question
if you add the disk in bay 2 to your existing storage pool 1, you will end up with SHR two disks, and the...
Replies
1
Views
420
Thanks - Good thinking on the ps. Both drives happened at the same time - so you may be 100% correct. I...
Replies
2
Views
1,241
https://old.reddit.com/r/synology/comments/17tvsiv/synology_drive_not_working_after_moving_homes/
Replies
1
Views
1,418
Does browsing for Synology Web Assistant find your NAS? That’s one way to try to find it on the LAN. Other...
Replies
3
Views
1,129

Welcome to SynoForum.com!

SynoForum.com is an unofficial Synology forum for NAS owners and enthusiasts.

Registration is free, easy and fast!

Trending threads

Back
Top