As I understand we will need separate container instances for running the registry sensor and the runtime sensor.
However, this does not seem to be a capability of the current qualys-tc chart.
This means we now need 2 instances of the same helm chart.
Because there is no option to deviate from the default naming used for the generated resources, the two charts have to be deployed to two separate namespaces.
This seems a bit unfortunate.