Very Slow Parity Rebuild - Logs Included

[rh535]

I have been having some issues with my server lately. I am trying to replace my parity drive with a larger drive (largest in the array) and it seems the parity rebuild has stalled. It's not looking good (time wise). Any idea what might be going on?
 

Total size: 18 TB

Elapsed time: 7 hours, 39 minutes

Current position: 1.46 TB (8.1 %)

Estimated speed: 72.8 KB/sec

Estimated finish: 2403 days, 14 hours, 45 minutes

tower-diagnostics-20230303-1615.zip

[JorgeB]

There's something reading from disks 3 and 7, stop all array activity and post new diags.

[rh535]

4 hours ago, JorgeB said:

There's something reading from disks 3 and 7, stop all array activity and post new diags.

Thank you for the response.

A few hours after I posted my 3TB drive started showing lots of errors (1,828). The 3TB drive has no data on the drive and it seems the parity sync is going much faster. ETA: 1 day, 1 hour, 42 minutes

I plan on replacing the 3TB drive after my parity rebuild with my old parity drive. I have noticed in the last couple of weeks that my server would hit close to 100% cpu usage and plex would sometimes buffer random files and without doing anything seem to hit high cpu load.

Where in the files did you see the info about the disk 3 and disk 7? (so I can learn for the future)

Since the ETA has improved so much - should I still stop the array/parity?

 

Edited March 4, 2023 by rh535

[JorgeB]

If there's no data on the 3TB drive and the sync is going well for now you can let it finish, but you should replace that drive once it's done, and if it slows down again better to cancel and remove/replace that disk.