I've always thought the PuppetDB fact blocklist was implemented in the wrong location. It should not be in the

puppetdb

service, it should be in the

puppetdb-terminii

that are used by Puppet server. Moving the blocklist to the terminii would have a couple of advantages: • Implemented in Ruby, same as Facter. Could even use a common library of logic to eliminate differences in capability and behavior. • Filtering fact data before sending the payload to PuppetDB means less network bandwidth and less work for PuppetDB to do. Only disadvantage I can really think of is that if PuppetDB is receiving data from a source that is not the terminii, that data would avoid the filter.