Hello.

 

So I made an oopsie yesterday 😰  and I am in need of your help.

 

I wanted to install a new graphics card in my server. In order to do so I've had to move my cache SSDs (there is four of them mounted in a ASUS Hyper M.2 X16 Card) to make room for the GPU

Fyi: these cards work if you split an 16x PCIe lane into 4x4x4x4x in the motherboard bios.

 

I did not remember this when I moved the card with the SSD. So when I turned my server back on, the cache wasn't working (unmountable) since it recognised only 1 of the SSD as the PCIe was in x16 mode.

 

So I tuned my server back off and went into the bios to transform the lane where it sits on from 16x to 4x4x4x4x. Then back into unraid.

 

Now it sees all of my cache drives but only assigns automatically cache n°4 into the cache pool (the disk that was recognised when in 16x mode). If I put them manually in my cache pool, it says that as soon as I start the array, it will erase the data on the drives. 

 

And... you guessed it.... I would like to avoid that.

 

Is there any way of rebuilding my cache array without erasing all it's content.

This array is a btrfs raid5.

 

Thank you for your help.

Edited March 30, 20215 yr by yaneck
Solved

If Docker/VM services are using the cache pool disable them, unassign all cache devices, start array to make Unraid "forget" current cache config, stop array, reassign all cache devices (there can't be an "All existing data on this device will be OVERWRITTEN when array is Started" warning for any cache device), re-enable Docker/VMs if needed, start array.

SOLVED

 

Well well, @JorgeB It seems you are my hero of the day!

 

You got anything where i can tip you a drink?

 

Thank you very much. It worked exactly as you said.