In the process of moving my drives to my new unRaid build I noticed after starting a preclear that one drive is listed as having HPA. This drive is not yet part of the array so has not changed my parity or affected my data in any way.

 

I plan to use the telnet method to remove the HPA and have read up quite a bit on it. I also understand that preclear does not remove it. My question is will I have to preclear again after removing it?

 

Heres what I have planned

 

1. Let the preclear finish

2. Remove HPA using hdparm -N -p(NATIVE_SIZE) /dev/sdX

3. Add to array

 

Is that all I need to do?

If you are adding the drive to a set that already includes valid parity, then yes, you will need to preclear again to keep from having to wait on the array to clear it. If you haven't yet calculated parity, then no, unless you just want assurance that the drive is still ok.

 

Preclear is recommended for 2 reasons, one being a health check, the other to limit the array downtime when adding a drive to an already protected array. You've already accomplished the health check part, assuming the preclear report comes back ok. The HPA removal will invalidate the preclear signature that allows the drive to be added quickly to a parity protected array.

 

If you don't run preclear again, you will have to wait for the array to start while it writes zeros to the entire drive.

Dang thats what I was afraid of. I will stop the preclear when I get back home, remove the HPA, then start another preclear.

 

Im not too worried about the health of the drive as its been in another machine for about 2 years.

Use the option to skip the pre and post read so that it only writes zeros to the disk.  I believe it is -n (lower case N) but I'm at work and don't remember for sure.  Saves some time on a "known good" drive.

Thanks for the tip!

Im not too worried about the health of the drive as its been in another machine for about 2 years.

Keep in mind that almost any use besides unraid will not have touched parts of the drive, perhaps ever. You could have bad spots that have never been asked to store data and you would never know it. Just because a drive has been in long term use doesn't mean it's healthy enough for unraid. If a drive fails all your other drives are used from start to finish to reconstruct the bad drive, so you don't want any questionable areas on any drives. If it were me, I'd wait out the current preclear if it's almost done, or if it's not, do a full preclear cycle after you remove the HPA.