Disabled disk (Red X)

[Boo-urns]

So I logged into my Unraid UI (v6.8.0) the other day and had some notifications that a parity check (run weekly) had failed, and some errors were found in disk 1. This disk is now disabled.

Server hasn't been moved or touched for a few months, just came out of nowhere. I did have some temperature warnings as it's been stupid hot here lately.

Diagnostics and SMART report for disk 1 is attached, if anyone could some assistance please?

Cheers.

skynet-smart-20200104-2249.zip skynet-diagnostics-20200104-2344.zip

[Decto]

Looks to me like the disk is toast.

 

187 Reported_Uncorrect       -O--CK   001   001   000    -    65535

197 Current_Pending_Sector  -O--C-   001   001   000    -    17480
198 Offline_Uncorrectable       ----C-   001   001   000    -    17480

 

Compared to one of my seagates - ST4000LM024

 

187 Reported uncorrect                  100      100   000      -  0

197 Current pending sector            100      100   000      -  0

198 Offline uncorrectable                100      100   000      -  0

 

That model drive is know to have issues https://en.wikipedia.org/wiki/ST3000DM001

Edited January 5, 2020 by Decto

[trurl]

Diagnostics are after reboot so can't see anything about that failed parity check. Was that a correcting parity check? Why are you doing weekly parity checks anyway?

 

You should always try to get diagnostics before rebooting, or setup Syslog Server so you can retain syslog after reboot.

 

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=781601

 

Yes, that disk must be replaced. But keep it handy in case there are problems with the rebuild since it's not entirely clear what the state of parity is.

[Boo-urns]

2 hours ago, Decto said:

Looks to me like the disk is toast.

 

187 Reported_Uncorrect       -O--CK   001   001   000    -    65535

197 Current_Pending_Sector  -O--C-   001   001   000    -    17480
198 Offline_Uncorrectable       ----C-   001   001   000    -    17480

 

Compared to one of my seagates - ST4000LM024

 

187 Reported uncorrect                  100      100   000      -  0

197 Current pending sector            100      100   000      -  0

198 Offline uncorrectable                100      100   000      -  0

 

That model drive is know to have issues https://en.wikipedia.org/wiki/ST3000DM001

Bugger. Probably explains why it's relatively cheap also.

1 hour ago, trurl said:

Diagnostics are after reboot so can't see anything about that failed parity check. Was that a correcting parity check? Why are you doing weekly parity checks anyway?

 

You should always try to get diagnostics before rebooting, or setup Syslog Server so you can retain syslog after reboot.

 

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=781601

 

Yes, that disk must be replaced. But keep it handy in case there are problems with the rebuild since it's not entirely clear what the state of parity is.

I couldn't get diagnostics before reboot as I believe the system had rebooted automatically at least once due to overheating. Plus I don't have the Syslog server setup as you mentioned, but will set that up in the future.

The parity checks were set weekly when I was setting various scheduled tasks, weekly seemed adequate. How often would you suggest parity checking?

 

Looks like I'll have to replace the disk then, doesn't seem like there's any doubt there. Thanks for your prompt assistance.

 

A further question on that, I have another Seagate drive (same model) as the parity, and the other 2 drives are WD Red (which I understand are designed for NAS use). So I would assume the best option would be to have the WD Red drives used for Parity and disk 1, with the other Seagate as a spillover.

How can I rearrange the drives to this format (ie. Parity = WD1, Disk 1 = WD2, Disk 2=Remaining Seagate)

 

[trurl]

8 minutes ago, Boo-urns said:

the system had rebooted automatically at least once due to overheating

You really must not allow this to happen. Fix this cooling problem or it will only cause more problems.

 

8 minutes ago, Boo-urns said:

How often would you suggest parity checking?

Most people only do monthly parity checks. Some even do less frequently for servers that aren't always on. Parity checks aren't required to maintain parity, they are just checks. Parity is always updated in realtime.

 

You didn't answer this question:

1 hour ago, trurl said:

Was that a correcting parity check?

 

9 minutes ago, Boo-urns said:

So I would assume the best option would be to have the WD Red drives used for Parity

Some people mistakenly think parity is the most important disk. In order to rebuild a missing disk, parity PLUS ALL remaining disks must be reliably read in order for the parity calculation to get the data to rebuild. Parity by itself can't recover anything. Also, parity doesn't actually contain any of your data, unlike the data disks. So arguably parity is the least important disk.

 

Have you ever replaced a disk before? Do you know how? Please ask for more advice if you are unsure.

 

[Boo-urns]

Re the cooling, I have added a few extra fans and improved the ventilation around the system (it's a mITX system) in a Node 304 case, so reasonably tight for space. Temps have improved a lot though.

1 hour ago, trurl said:

You didn't answer this question: Was that a correcting parity check?

I'm not sure, and not sure how to check either.

1 hour ago, trurl said:

Have you ever replaced a disk before? Do you know how? Please ask for more advice if you are unsure.

 

I haven't, as I haven't ever had the need so far. However it seems like shrinking the array would be the best option for me using this guide?

I don't need to replace the failed drive currently, I have enough space.

 

Using the above guide, can I re-allocate disk 3 (empty) to replace disk 1? I'm assuming i'd have to follow a specific process to do so...

[trurl]

You can copy the data from the emulated disk1 to other disks then remove disk1. I recommend the "Remove and Rebuild Parity Method" from that link. It is simpler and just as fast. At step 4 New Config you can assign the disks however you want. Just make sure you don't assign a data disk to the parity slot.