Does this mean this drive is dying?

[Mortalic]

Woke up this morning and noticed this disk has thrown 6974 errors.  It still shows green.

https://wiki.unraid.net/Troubleshooting#Hard_drive_failures

This FAQ suggests nothing to worry about, but it says no-zero.  6974 seems a bit more worrisome than non-zero.

Should I get a replacement?

 

For what it's worth, I'm copying a ton of data to the array right now (and for the last few days).

[ChatNoir]

It would help the guys give you an accurate answer if you could provide your diagnostics in your next post.

[trurl]

Impossible to give any more information until you give us more information.

 

Go to Tools - Diagnostics and attach the complete Diagnostics ZIP file to your NEXT post in this thread.

[Mortalic]

Thanks I am currently running the short SMART test on the drive, it looks like the diagnostics collects that data so I'll wait for it to finish.

[trurl]

Diagnostics will give us a lot more information than just the results of that SMART test. It might even tell us more about the problems on your other thread(s). Be sure to get us the diagnostics without rebooting since syslog resets on reboot.

[Mortalic]

19 minutes ago, trurl said:

Diagnostics will give us a lot more information than just the results of that SMART test. It might even tell us more about the problems on your other thread(s). Be sure to get us the diagnostics without rebooting since syslog resets on reboot.

Well the SMART test has been stuck at 90% for quite a while now, so I'm just going to put the diagnostics here now.  Syslog looks like it's a bunch of read errors.  What does that mean?

vault-diagnostics-20200729-1251.zip

[trurl]

Some of that might be connection issues, but there are these mixed in:

Jul 28 18:29:04 vault kernel: ata6.00: cmd 25/00:28:a8:08:72/00:03:06:00:00/e0 tag 5 dma 413696 in
Jul 28 18:29:04 vault kernel:         res 51/40:17:b0:09:72/00:02:06:00:00/e0 Emask 0x9 (media error)

Also, this SMART attribute is something to watch out for on WD Reds, in fact, I have it added to the notifications for those disks I have:

ID# ATTRIBUTE_NAME          FLAGS    VALUE WORST THRESH FAIL RAW_VALUE
  1 Raw_Read_Error_Rate     POSR-K   200   200   051    -    266

Do you have a replacement?

[Mortalic]

5 minutes ago, trurl said:

Some of that might be connection issues, but there are these mixed in:

Jul 28 18:29:04 vault kernel: ata6.00: cmd 25/00:28:a8:08:72/00:03:06:00:00/e0 tag 5 dma 413696 in
Jul 28 18:29:04 vault kernel:         res 51/40:17:b0:09:72/00:02:06:00:00/e0 Emask 0x9 (media error)

Also, this SMART attribute is something to watch out for on WD Reds, in fact, I have it added to the notifications for those disks I have:

ID# ATTRIBUTE_NAME          FLAGS    VALUE WORST THRESH FAIL RAW_VALUE
  1 Raw_Read_Error_Rate     POSR-K   200   200   051    -    266

Do you have a replacement?

Regarding the connection issues, like a sketchy SATA cable or something?

Regarding the replacement, yeah I've got a spare drive.  Should I swap it out, or take a wait and see approach?

[trurl]

Did the short SMART test ever complete? I would replace. Do you know how?

[Mortalic]

14 minutes ago, trurl said:

Did the short SMART test ever complete? I would replace. Do you know how?

It's still stuck at 90% actually.  I do know how.  I had one drive I knew was throwing smart errors I used to practice with before I sent it for recycling.

Thanks for asking though.

Edited July 29, 2020 by Mortalic
typo

[Mortalic]

Short smart test completed without error

[trurl]

Run an extended SMART test on it, maybe replace it first.

[Mortalic]

14 hours ago, trurl said:

Run an extended SMART test on it, maybe replace it first.

 

WDC_WD20EFRX-68EUZN0_WD-WCC4M2ZPD2U2-20200729-2059.txt

[Mortalic]

Extended SMART test said it had errors in the extended test but unraid still shows the drive is healthy.  Very strange.

Anyway I've got low activity on the system this morning so I'm going to replace it.

 

Any chance someone could explain what I should be looking at in that Extended SMART test? 

[JorgeB]

4 minutes ago, Mortalic said:

Extended SMART test said it had errors in the extended test but unraid still shows the drive is healthy.

That's normal, you'd likely have more errors on the next check/read, that disk should be replaced.

[Mortalic]

I replaced the disk.  Thanks everyone for the help.

[trurl]

1 hour ago, Mortalic said:

unraid still shows the drive is healthy

Do you mean healthy as not being disabled, or healthy as in not showing any SMART warnings?

 

You should click on each WD Red drive you have and enable notifications for these SMART attributes:

ID# ATTRIBUTE_NAME          FLAGS    VALUE WORST THRESH FAIL RAW_VALUE
  1 Raw_Read_Error_Rate     POSR-K   200   200   051    -    266
200 Multi_Zone_Error_Rate   ---R--   200   200   000    -    33