Hi, I just build my first Unraid server with new CPU, new mainboard, new drives, new all. At my first try I put the four 10TB into an array and started the parity sync. During the process the UDMA CRC error count on my Disk 2 increased. Not much but by 3 during the whole 16-17 hrs process. I read in this forum that it might be a faulty connection, cable, controller but not likely an error of the disk itself.

 

Just to be sure and since I hadn't used the server for anything yet, I wanted to run some tests. I swapped the Sata cable and used a different Sata port on my main board and assured that the connectors where sitting correctly. I unassigned the drives in Unraid and started a full Preclear plugin process on every of those four disks. Reading/Writing/Reading-Again went through successfully without any errors, warning and without any UDMA CRC value increasing.

 

So I just assigned the disks again, started the array and let the parity sync. And after a few hours I received the next warning that the UDMA CRC on Disk 2 increased again … and I don't understand why and what to do next.

 

The drive is currently only being read during the parity sync. But with Preclear it has been read during pre-read and post-read completely without any problems. I already switched the cable and port on the mainboard (and the port that was previously in use by Disk 2, is now being used by another disk that still has its pristine error count). I just finished a SMART self test which didn't show any errors.

 

I don't know if I should return disk and get a new one (in case there is something broken with the connection on the disk itself)? Or is there something that I can do to test or fix it? I think I don't feel comfortable to sometimes have the value occasionally increase while all other drives do work without such errors.

WDC_WD101EFBX-68B0AN0_VH0------20240127-1602 disk2 (sde).txt

In addition to data cable you should also consider power. Any splitters? 

 

Also, don't bundle data cables and make sure there is plenty of slack so nothing is pulling on the connection. 

 

I usually just acknowledge the occasional CRC, maybe replug next time I'm in the case if I think of it. 

 

If it isn't increasing a lot during normal use probably nothing to worry about. 

 

CRC just means received data didn't pass checksum, so it is resent. 

 

CRC often won't even be logged as an I/O error. And bad connections don't necessarily get CRC because no data was received to checksum. 

 

 

Good idea … I will check the power cable. I unplugged it from the drives when I was changing Sata cables but I'm not so sure if I may have plugged the same one in again.

 

I just checked my previous file server's disks, 8 disks multiple years running and still zero count. The other disks in my new system zero count. Only this one disk now...

Only 4 is not much more than zero for this. If it were 4 RAM errors that would be serious.