Hello. On my shared Plex server, I was trying to replace a 4 TB drive with a 6 TB recently pulled from my home server but 1% into the data rebuild the operation paused with read errors on the replacement disk. Obviously I assume I need to cancel the read check now but what are my next steps from there? Is it possible to put the old 4 TB drive back in or does it have to be a new (healthy) 6 TB drive? Diagnostics attached and any help appreciated.

jbox-diagnostics-20230129-1022.zip

Not clear there is anything wrong with the disk. Check connections, both ends, power and SATA, including splitters. Try rebuild again.

2 hours ago, trurl said:

Not clear there is anything wrong with the disk. Check connections, both ends, power and SATA, including splitters. Try rebuild again.

Thanks  The physical connections seemed fine and a SMART self test of the disk came back with no errors too (SMART report attached) However on the Dashboard under SMART it says error and on the main page there is a red X and the message when you hover over it "device is disabled, contents emulated", and when I toggled on the drop down to No device and then back again I had the following notification pop up:

 

Unraid device dev2 SMART health [199]: 29-01-2023 14:04

Warning [JBOX] - udma crc error count is 1398238
WDC_WD60EFRX-68MYMN1_WD-WX91D65DCDSN (dev2)

 

Also at the bottom of Main with the drive selected again it says Configuration Valid and starting will simply bring the array on line again with no mention of any data rebuild / parity sync If I start the array like this, with or without the replacement drive, am I not risking the loss of the emulated date from the previous drive? I'm still stumped 

jbox-smart-20230129-1356.zip

The SMART report was already in the diagnostics. UDMA CRC Errors are recorded in the drive firmware by the drive firmware when inconsistent data is received by the drive based on checksum. These are almost always connection problems.

 

Emulated disk4 was mounted in those Diagnostics.

 

If you think you have fixed connection problems you can try again.

 

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

2 hours ago, trurl said:

The SMART report was already in the diagnostics. UDMA CRC Errors are recorded in the drive firmware by the drive firmware when inconsistent data is received by the drive based on checksum. These are almost always connection problems.

 

Emulated disk4 was mounted in those Diagnostics.

 

If you think you have fixed connection problems you can try again.

 

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

Thanks. I followed the linked instructions, starting the array with the drive unassinged, stopping, and starting it again with it reassigned to start the data rebuild / party check again. The result was the same as last time. The rebuild stops itself at 1% saying read errors on the drive and array errors. I'm sure there's nothing wrong with the physical SATA or power connections to the drive which are the same as they were with the replaced drive that was working normally, but I guess I'll have to try to re-rig different connections somehow, or maybe try a new 6TB drive even though SMART is saying there's nothing wrong with this one. Am I right in assuming though that putting the old 4 TB drive back in is no longer an option at this point? If you have any other suggestions please let me know.

That is a large number of crc so maybe the problem is somewhere on the disk side of the connection. 

 

You can New Config the original back into the array and rebuild parity. 

18 hours ago, trurl said:

That is a large number of crc so maybe the problem is somewhere on the disk side of the connection. 

Yes it would have to be. Both the SATA and power connections are on the same cables connecting 3 other data drives to the LSI sata controller and the PSU respectively, and until now there's never been an issue with any drive connected to them. Anyway, the server is at my folks and I won't have a chance to work on it again until the weekend but when I do I'll check the connectors on the drive itself and see if maybe some dust got in there or something else that can be cleaned or repaired. If that fails I'll do the new config with the previous drive and maybe try replacing a different 4 TB drive with different connections that might be more forgiving (assuming no apparent issue with the drive's connectors). 

 

I'll keep the topic open just in case more questions come up then but I expect I'll be marking it solved one way or another at that point. Thanks again for your help.