Predictive failure on Smart Array P411 - once again

Hi,

our Server is an HP ProLiant DL160 G6, Smart Array P411/512 MB SAS RAID-Controller with BBU-Unit,  running Vmware ESXi 6.

Running hpssacli displays the following:

Smart Array P411 in Slot 4

   array A

      Logical Drive: 1
         Size: 931.5 GB
         Fault Tolerance: 1
         Heads: 255
         Sectors Per Track: 32
         Cylinders: 65535
         Strip Size: 128 KB
         Full Stripe Size: 128 KB
         Status: OK
         Caching:  Enabled
         Unique Identifier: 600508B10010543953595A3959520300
         Logical Drive Label: A04A2EA5PACCT9SYZ9YRCEB2
         Mirror Group 1:
            physicaldrive 2E:0:5 (port 2E:box 0:bay 5, SATA, 1 TB, OK)
         Mirror Group 2:
            physicaldrive 2E:0:6 (port 2E:box 0:bay 6, SATA, 1 TB, Predictive Failure)
         Drive Type: Data
         LD Acceleration Method: Controller Cache

What are the next steps right now?

Can the Harddisk be replaced while it's running or do we have to shutdown the server? Are there further steps required after inserting the new drive? Should the drive to be replaced definitely or is it possible to repair it by the controller?

Thanks for your suggestions.