Hi,

I bought Unraid years ago but haven't used it until now. I now have a lot of different hard drives lying around and wanted to start using them with Unraid. I had to create a new installation USB drive because there were issues with the old one. The parity sync ran at about 20 MB/s. I have now created a new USB drive. Unfortunately, I ran into problems during the creation process. It didn’t work with the default tool. I then used Rufus and Debian to create a USB drive and copied the UNRAID files from the ZIP file onto it. That way, I was at least able to get a bootable USB drive.

With this new USB drive, I was able to set up my new array, and the parity sync also completed at about 120 MB/s. So I thought the speed issue was now resolved.

Since I had already filled the disks with the “old” stick initially, there was already various data on the hard drives. I’ve now started an rsync process to copy the missing data. That worked well at first, until the cache was full and the data had to be written to the array. Now I’m back down to a write speed of about 20 MB/s on the array. Does anyone have any idea what the problem might be?

I’m ruling out a communication issue with the controller, since the parity sync ran quite smoothly at 120 MB/s.

I know that one or two of the drives have a SMART error. However, I don't think these errors are what's slowing down the array.

I'd appreciate any advice.

unraid-diagnostics-20260320-1339.zip

Edited March 20Mar 20 by Susi

Looks like you are probably trying to run mover at the same time.

Mover is intended for idle time. Default schedule is once per day in the middle of the night.

If you need to write more than cache can hold, don't cache.

Haven't examined SMART. Which disks specifically?

Hi, and thx for your reply.

I canceled the rsync job and still got around 20mb/sec without any additional jobs. There is no active shared, so no activity. Only the the mover is running.

Disk 2 and 3 have CRC error count > 0. But currently the data will be written to drive 5, where all the smart values are good.

6 minutes ago, Susi said:

Disk 2 and 3 have CRC error count > 0.

The CRC count never resets to 0, so all you can ever do is stop it from increasing.

I know, but there was a question about the smart issues and disks. So I added these to the post.

Some of your disks are SMR, try writing to a non-SMR disk, like disk 9, for example

16 hours ago, Susi said:

Only the the mover is running.

Looked like you might be moving a lot of small files. That would have more overhead than a few large files.

You might want to turn off mover logging.

14 hours ago, JorgeB said:

Some of your disks are SMR, try writing to a non-SMR disk, like disk 9, for example

Do you know which one is non-SMR. Maybe I have to remove it at all.

These are SMR: disk2, disk5, the unassigned 8TB Seagate, the other ones are CMR.

1 hour ago, JorgeB said:

These are SMR: disk2, disk5, the unassigned 8TB Seagate, the other ones are CMR.

thx