(..continue): Looking at OMV Destroing all My Arrays, another 2 drives failed

Hi,
night here. I have reconnected to OMV GUI to look at array reconstruction and it was slow, a timeout accurred. Connected again, it worked.
I moved to the ARRAY section and found it has just started recovering the second one: 0,0% here.

It seems my login caused the rebuild.

I can only suppose here: my drives where sleepy and loggin in woke up them but some kind of timeout accurred.

I can see all kicked off drives from the array in the drives section. They where addeded at the end of the list like they have been recognized as new.

I'll leave the system recovering and I pray it will able to end. The second array has a spare too but TWO drives have failed and 16TB takes a long time to recover

Please remember I have just upgraded to Kernel 4.9

A last note: as going to the SMART section times out since a couples of week, which command like could I use to disable SLEEP in all of my drives ?

A last question: is there a way to instruct MDADM to not never kick off a drive from the array and leave the array "READ ONLY" if drive still readable ?

How it eneded ? I have checked all the cables and moved fans to another line to lower the load of the HD one.

I have then removed a memory stick that from time to time give a CRC error.

I have reset all the excluded drives to factory settings and added again to the arrays.

Still unable to access the S.M.A.R.T section listing all the drives for "Communication Error"