Thanks for your answers.
I moved the persistent scratch location to another LUN/Datastore.
(Maintenace mode and reboot for every ESXi host)
I tought this would be the best solution for me.
The persistent scratch location uses the UUID from the volume as part of the folder path.
By adding new storge a new UUID will be created. This was the reason for me to move the persistent scratch location to another LUN/Datastore.
In my case for the same LUN:
/vmfs/volumes/482ad53f-150a1320-86cf-001e4f38cb49 (old UUID)
/vmfs/volumes/4fc76096-933e0970-7aa3-d067e5e901cd (new UUID)
Thanks.