Currently entrada expects an explicit list of nameservers to process, which is not always ideal if said list is not under the user's control.
You might risk missing out on new data when the actual list changes.
Or alternatively by combining all files together under a single nameserver you miss out on useful functionality; you lose for example granularity of nameserver/location in the resulting parquet files, or the spool/temp files for queries with unmatched answer expire before it gets to the correct "next" sequential file for a specific nameserver.
Currently entrada expects an explicit list of nameservers to process, which is not always ideal if said list is not under the user's control.
You might risk missing out on new data when the actual list changes.
Or alternatively by combining all files together under a single nameserver you miss out on useful functionality; you lose for example granularity of nameserver/location in the resulting parquet files, or the spool/temp files for queries with unmatched answer expire before it gets to the correct "next" sequential file for a specific nameserver.