During the Disk 2 rebuild, there were 1151 read errors on Disk 7.  I did not cancel.  Subsequent to that rebuild (with the read errors), we ran the check disk actions per your notes above.  Per your most recent instructions, I have now rebooted, and attached the Diags.

 

Next steps?  I don't know what caused the read errors.  If we need to rebuild again, I have no problem with doing that...  Just being "slow and steady" through this part.

whitenas-diagnostics-20230417-2247.zip

You can try again, after checking/replacing cables on disk7, but if you do I would do it using a new disk2, in case it doesn't work or it gets worse.

I have the "new disk 7" that I could overwrite...  It was built with the 18k or so of read errors, and the "better" parity state...  Do you think that at this point, we are better off on this route with "old disk 7"?  I kinda think we are, but I also acknowledge that I'm not sure we will get another chance at reading from "old disk 2" again...

Possibly, difficult to say for sure, 1151 errors is also not many, especially if they were all together, could affect a single file.

They happened close in time.  I'll try to rebuild overwriting the "new disk 7" and see if they pop up again.

The re-wrotento disk 2 is nearly complete. Couple next step questions:

 

Will I need to redo the check disk activities for the now new disk 2 given this update? 

 

Any recommended tools for recovering potentially lost files?  (Thinking checksum evaluators, etc that can work from Windows to see if downloaded backups match files in Unraid...) Backup solution is CrashPlan running on the Unraid box itself.  Once array is running can do a restore there, but seems wasteful. Should only restore damage / corrupted files if possible...

Finished, no errors, awaiting next steps, but going to go with this...

 

Another question (3rd one)...  I plan to use one of the blank 4tb drives as a parity drive.  What order of operations should I follow?  Get the parity drive cleared and added then recover stuff? Or other way around...

 

Thank you so much!!!

8 hours ago, jrhamilt said:

Will I need to redo the check disk activities for the now new disk 2 given this update? 

Should not be needed but you can do it.

 

8 hours ago, jrhamilt said:

Any recommended tools for recovering potentially lost files?

It won't help for this time but recommend using btrfs or zfs, builtin checksums.

 

5 hours ago, jrhamilt said:

Another question (3rd one)...  I plan to use one of the blank 4tb drives as a parity drive.  What order of operations should I follow?  Get the parity drive cleared and added then recover stuff? Or other way around...

If the last rebuild finished without errors you can just replace parity.

Understand that another file system type would be better...  And longer term I can head that way.  In the mean time, is there anything that you know of that will help identify and replace corrupt files?

 

As for parity, my intent is to add a second parity disk. Does it need pre-cleared?  Can I just add it to the array?  I feel like I should do that before changing file system types.

 

9 minutes ago, jrhamilt said:

In the mean time, is there anything that you know of that will help identify and replace corrupt files?

Only if you have pre-exisitng checksums for all files or have the same files elsewhere, like a backup server, so you could run a comparison.

 

10 minutes ago, jrhamilt said:

Does it need pre-cleared?

No, preclear is always optional.

Huh.  Drive 7 now throwing a few more errors and has something in the SMART file.  Next steps?  (Haven't finished building second parity because of the errors...  Disk 7 had 128 errors during that rebuild.

WDC_WD40EFAX-68JH4N0_WD-WX42D308JR81-20230422-1004.txt

It's showing a pending sector, so likely a disk problem, you can run an extended SMART test to confirm.

Turns out, I have the "Dynamic File Integrity" plugin...  Should this be able to find any of the file corruption?

Yes, but first need to fix the disk issues.

Disk issues resolved. I hope. Disk 7 rebuilt on new drive. Disk 7 going back with RMA, failed in under 3 years.  Disk 7 doesn't appear to have been selected in the plugin settings somehow?  Dont know. Isn't clear that it has been running recently...

 

What do those steps look like to ID the corrupt files?  Any help I can get finding them here is awesome.

 

Will get a other drive running as second parity after the data restore...

14 hours ago, jrhamilt said:

What do those steps look like to ID the corrupt files?

If you have the file integrity plugin installed and it was creating checksums for all files you just need to run a check.

Sorry, this seems stupid...  How do I "run a check"?  No button or other mechanism here or in the settings (as far as I can see).  Do I need to run an export first?  The exports are ancient...  (It probably has not been running...)

 

Sorry, it's been a few years since I've used the plugin, recommend asking in the existing plugin support thread.