Let's say I want to upgrade a data or parity drive and I have an extra slot, can I use DD to first copy the old drive to the new larger drive and then run a check only parity sync after re-assigning the freshly copied disk? Would this minimize the time spent in the parity or data rebuild phase and minimize risks? Or am I over thinking it?

Let's say I want to upgrade a data or parity drive and I have an extra slot, can I use DD to first copy the old drive to the new larger drive and then run a check only parity sync after re-assigning the freshly copied disk? Would this minimize the time spent in the parity or data rebuild phase and minimize risks? Or am I over thinking it?

You are definitely over thinking it!  Just follow the defined unRAID process for upgrading a disk to a larger one.

 

Where a free slot does come in useful is to enable you to plug in a new disk and put it through an initial stress test before trying to use in the array and trust your data o it.  This is typically done by running the pre_clear script against the disk.  This test can be done in parallel with the array being used as normal so causes no downtime.

 

Another point to note that when you do initiate the replacement process unRAID remains useable while a parity sync and/or disk rebuild is in progress.  There are those who like to make sure they do not write any new data to the array until this is finished, and although a reasonable precaution in cases rebuild fails it is not mandatory.

yes you are overthinking it.

 

preclear and stress test the new drive until you are comfortable.  Highly recommended.

Clear the statistics, Do a parity check, make sure there are no errors.  Highly recommended.

 

After you are done with both, examine the smart data on the drives (for that extra level of comfort). 

 

Make sure there no no large pending sectors or reallocated sectors.

People do not normally do this, I do since I had a dual drive failure once due to a pending sector.

 

rebuild and expand to the new drive.