We have to recalculate for all containers to allow e.g. over provisioning models.
:return:
"""
- for d in self._allocated_compute_instances.itervalues():
+ for d in self._allocated_compute_instances.values():
if not self.deactivate_cpu_limit:
self._apply_cpu_limits(d)
if not self.deactivate_mem_limit: