Disk 1 disabled, log shows read and write errors

[oh-tomo]

I'm going to shut down and replace the disk.  Attached are the diagnostics.  Wish me luck.

server2018-diagnostics-20220618-0010.zip

[JorgeB]

Doesn't look like a disk problem, suggest running and extended SMART test before replacing it.

[oh-tomo]

8 hours ago, JorgeB said:

Doesn't look like a disk problem, suggest running and extended SMART test before replacing it.

 

What kind of problem does it look like?   14 hours left on new disk data rebuild.   Should I abort and put the old disk back in for an extended SMART test?   My impulse was that the array was vulnerable with a disabled disk exhibiting read/write errors so I wanted to get a functioning disk in there ASAP.  Old disk warranty expired in February.

[trurl]

Rebuilding to a new disk is always a little safer than rebuilding on top of the original. That way you have the original in case there are problems with rebuild. But of course you need a spare disk and they aren't free.

 

Let rebuild complete.

[oh-tomo]

Here's an updated diagnostics while it rebuilds.   Took  a while to generate with the rebuild going on.

server2018-diagnostics-20220618-1202.zip

[trurl]

That looks fine. I didn't check SMART for all other disks. Do any have SMART warnings on the Dashboard page?

[oh-tomo]

55 minutes ago, trurl said:

That looks fine. I didn't check SMART for all other disks. Do any have SMART warnings on the Dashboard page?

 

Disk 4 has error (UDMA CRC error count 1), the rest are healthy on dashboard.

[trurl]

33 minutes ago, oh-tomo said:

 

Disk 4 has error (UDMA CRC error count 1), the rest are healthy on dashboard.

That's OK. These are just logged by the drive when there is some inconsistency in the data it received. The data is resent.

 

Click on that warning and acknowledge it, it will warn again if it increases. If you get a lot of these you should try to figure out where the problem is, connection or such.

[oh-tomo]

I paused the rebuild and ten minutes later got an alert that Disk 1 (the new disk) is in error state.  So maybe JorgeB is onto something.  Running an extended SMART test now.  How long do they take?   Will the test reveal an actionable solution to the cause of the errors?

[oh-tomo]

33 minutes ago, oh-tomo said:

I paused the rebuild and ten minutes later got an alert that Disk 1 (the new disk) is in error state.  So maybe JorgeB is onto something.  Running an extended SMART test now.  How long do they take?   Will the test reveal an actionable solution to the cause of the errors?

 

Extended SMART test didn't finish -- "Interrupted (host reset)"  I'll try again.

 

Edited June 19, 2022 by oh-tomo

[oh-tomo]

1 hour ago, oh-tomo said:

 

Extended SMART test didn't finish -- "Interrupted (host reset)"  I'll try again.

 

 

Extended SMART test (second attempt) has been stuck at "self-test in progress, 10% complete" all this time.

[JorgeB]

Extended SMART test will take several hours, also make sure spin down is disabled for that disk.

[itimpi]

5 hours ago, oh-tomo said:

 

Extended SMART test (second attempt) has been stuck at "self-test in progress, 10% complete" all this time.

The test should take something like 1-2 hours per TB.   Progress only gets updated every 10%.

[oh-tomo]

"Completed without error"  Now what?

ST16000NE000-2RW103_ZL2P6V5T-20220619-2307 redacted.txt

Edited June 20, 2022 by oh-tomo

[oh-tomo]

I guess I could resume the parity check.

[JorgeB]

3 hours ago, oh-tomo said:

"Completed without error"  Now what?

It confirms the disk wasn't the problem, check/replace cables and if contents on the emulate disk1 look correct you can rebuild on top:

https://wiki.unraid.net/Manual/Storage_Management#Rebuilding_a_drive_onto_itself

[oh-tomo]

I couldn't see any problem with the cables.   Maybe they were arcing too high towards the upper case fan.  I tucked them a bit to the side away from the fan.  I guess I just wait and see if the problem returns.