setting ashift when creating pool in napp-it?

[wkearney99]

I'm using the latest install of OI and napp-it (setup and upgraded yesterday).

I've got some WD 2TB EARS drives; one connected to an areca 1260 and the other to an M1015. When I create a pool on the one connected to the areca it's set up with an ashift of 9. When I create another pool with the same model drive connected to the M1015 it's set up with an ashift of 12. Why?

And how would I go about having napp-it allow me to pick the ashift?

 

[gea]

Correct value for WD EARS is ashift=12
ZFS use this automatically when the disk reports 4k physical sector size.
This seems the case only with the M1015

You can override this detection (mostly) in file sd.conf (see menu disk - details - sd.conf) to force ashift=12 or ashift=9

 

[MiniKnight]

Any other issues with the M1015? Does this always happen? Not good.

 

[gea]

Any other issues with the M1015? Does this always happen? Not good.

??
Very good. IBM 1015 sets the correkt value

 

[MiniKnight]

Oh yes. Is it only the m1015 that aligns correctly then?

 

[gea]

I would expect HBA's to work better than hardware raid controller with this and as a general rule with ZFS

 

[kendrick]

Correct value for WD EARS is ashift=12
ZFS use this automatically when the disk reports 4k physical sector size.
This seems the case only with the M1015

You can override this detection (mostly) in file sd.conf (see menu disk - details - sd.conf) to force ashift=12 or ashift=9

what commands can you use to get that information? I currently have both zfsguru and nappit running on my vm as I am trying to determine which would be best fit for my needs.

I have found quite a few posts here and there saying you can find the info but not much on how. also is there a way to globally force ashift=12 ? even if it really is a 512 drive i dont want to have to break things later due to new drives so i dont want any =9

 

[gea]

what commands can you use to get that information? I currently have both zfsguru and nappit running on my vm as I am trying to determine which would be best fit for my needs.

I have found quite a few posts here and there saying you can find the info but not much on how. also is there a way to globally force ashift=12 ? even if it really is a 512 drive i dont want to have to break things later due to new drives so i dont want any =9

You can request ashift values via zdb -C tank (or napp-it menu pools, click on a pool)
With ZFS Guru, you can manually force ashift
With Illumos based systems, ashift=12 is done automatically when disks reports 4k phisical sectorsize, otherwise you can (mostly) override the settings with sd.conf (asumed sd driver is used)