Hi,

 

My parity disk suddenly stop after seing 4 writes errors, short log below

Aug 21 01:07:46 MNAS kernel: sd 4:0:0:0: [sde] tag#17 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=0x00
Aug 21 01:07:46 MNAS kernel: sd 4:0:0:0: [sde] tag#17 CDB: opcode=0x8a 8a 00 00 00 00 00 e8 e8 74 b0 00 00 00 08 00 00
Aug 21 01:07:46 MNAS kernel: print_req_error: I/O error, dev sde, sector 3907548336
Aug 21 01:07:46 MNAS kernel: print_req_error: I/O error, dev sde, sector 3907548336
Aug 21 01:07:46 MNAS kernel: md: disk0 write error, sector=3907548272
Aug 21 01:07:46 MNAS kernel: program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO

If I try to execute SMART tests, it fail with

Short INQUIRY response, skip product id
A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options.

After rebooting Unraid, SMART tests were OK, log file attached

 

So  I did a new preclear of the disk and result was good and parity rebuild too

 

It's the second time that I have this issue, I don't know what to think

where is the problem? hard drive? motherboard? unraid? advice would be appreciated

 

and is there a better/faster way to reactivate parity without preclear and full rebuild?

 

thanks

WDC_WD40PURX-64NZ6Y0_WD-WCC7K3XXUPH6-20200821-0913.txt

SMART shows a recent UNC at LBA error, if it happens again you should run an extended SMART test first.

 

 

24 minutes ago, Meldrak said:

and is there a better/faster way to reactivate parity without preclear and full rebuild?

You can skip the preclear.

a full rebuild is mandatory? when it occurs I only had 4 write errors

unplugging, unassigning disk and rebooting unraid didn't restart parity process, disk always came back with its 4 errors

I had to clear the disk to get parity again, I was not so easy

1 hour ago, Meldrak said:

a full rebuild is mandatory? when it occurs I only had 4 write errors

unplugging, unassigning disk and rebooting unraid didn't restart parity process, disk always came back with its 4 errors

I had to clear the disk to get parity again, I was not so easy

Not entirely sure I understand what you are saying in this post.

 

You must rebuild a disabled disk, whether the disabled disk is parity or data.

 

There is no need to clear a disk you are going to rebuild, because the rebuild will completely overwrite the disk anyway.

 

Are you saying that you had to clear the disk again before it would let you rebuild? That is the part I am not sure I understand.

 

The correct procedure to rebuild to the same disk, whether parity or data

1. Stop array
2. Unassign disabled disk
3. Start array with disabled disk unassigned
4. Stop array
5. Reassign disabled disk
6. Start array to begin rebuild

 

7 hours ago, trurl said:

Are you saying that you had to clear the disk again before it would let you rebuild? That is the part I am not sure I understand.

Yes and sorry for my english 

I did the same as your procedure and at step 6 my disk came back without rebuild in fail state with the same 4 errors

 

That suggests you still had some problem with the disk or its connections. Couldn't have anything to do with preclear because

12 hours ago, trurl said:

There is no need to clear a disk you are going to rebuild, because the rebuild will completely overwrite the disk anyway.

 

Hi,

 

I still have this issue, but I noticed that it only occur after a reboot

I don't understand why, so here I join my full log if someone would check

 

diagnostics-20210215-0832.zip

Cache device dropped offline:

 

Feb 15 01:04:15 MNAS kernel: ata3: hard resetting link
Feb 15 01:04:20 MNAS kernel: ata3: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
Feb 15 01:04:30 MNAS kernel: ata3.00: qc timeout (cmd 0xec)
Feb 15 01:04:30 MNAS kernel: ata3.00: failed to IDENTIFY (I/O error, err_mask=0x4)
Feb 15 01:04:30 MNAS kernel: ata3.00: revalidation failed (errno=-5)
Feb 15 01:04:30 MNAS kernel: ata3.00: disabled

 

Check/replace cables.

so if cache device go offline parity disk fail? it's not a good behavior

in anycase I will change my cables

 

3 minutes ago, Meldrak said:

so if cache device go offline parity disk fail?

No, unrelated.

but the only errors was on parity when I saw it

Cache errors don't appear on the GUI:

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=700582

 

good to know 👍