.nl RSSAC files drop the names
SIDN Labs began publishing .nl DNS traffic in the RSSAC002 format on October 26. The files are daily aggregates. SIDN says they contain no IP addresses and no query names.
The short version
SIDN says no. The files are daily aggregates of traffic volume, sizes, response codes, and unique-source counts. Names and addresses stay out.
The launch
Giovane Moura's SIDN Labs post, dated October 26, 2026, says the registry is publishing daily statistics for traffic that reaches the authoritative name servers of .nl. The format is RSSAC002, the specification root-server operators already use. The public site is rssac002.sidnlabs.nl, and the same files are in the SIDN/nl-rssac002 repository. RIPE Labs carried the post, which is how it landed on the networking wire. The dataset site says the files were first released on October 25. The blog date and the release date are one day apart, and both can stand: a site can exist the day before the post that announces it.
What a file contains
RSSAC002 version 5 defines six metrics. SIDN says it publishes four: traffic volume split by IP version and by UDP or TCP, a distribution of message sizes, response codes, and counts of distinct source addresses. It does not publish zone size or load time. Each file covers one UTC day. There is a combined file for all .nl authoritative servers and a file per server. The site's own example, a unique-sources file for March 10, 2026, shows the fields. One day's sample is an illustration of the format. It is not a ranking of the year.PQ GeoKey.
What a file does not contain
SIDN says the files are aggregated daily statistics. No IP addresses, no individual queries, and no looked-up domain names are included. Unique-source counts are counts, not a list of who asked. The session brief is the privacy desk's line that a session is not a dossier. These files are a registry choosing the aggregate on purpose. A reader who wants a query name will not find one, and that absence is the point of the format as SIDN describes it.What a session cannot be.
Adding the servers up
SIDN's metrics page says query counts, such as traffic volume, add across the per-server files. Unique-source counts do not. An address that queries two servers appears in each server's file and only once in the combined file. The site also publishes an IPv6 aggregate at the /64 prefix, which it describes as a better estimate of networks than raw address counts. Those rules are how a reader avoids double-counting. This page will not paste a YAML sample as if the numbers were a finding about abuse.A name is not a location.
The comparison SIDN draws
The lab post says that on a typical day the .nl servers together handle about as much traffic as D-Root and about 50 percent more than B-Root. That comparison is SIDN's, drawn because the format matches. To our knowledge, the post says, .nl is the first top-level domain to publish in this format. To our knowledge is the lab hedging its own survey. CZ.NIC and .ie already publish traffic figures in their own formats, the post notes, and those older figures are not this format. The Reddit token rule was a forum choosing what a credential is. SIDN is a registry choosing what a public statistic is.Reddit's token rule.
Not a routing product
PQ GeoKey was a closed beta of a different networking idea. The Node October release was a runtime patch. Neither one is a TLD statistics dump. They are linked so a reader who collects networking documents can see the type of each one. RSSAC002 files do not change how a resolver behaves. They describe traffic the registry already answered. Asked directly, When did SIDN start the RSSAC002 site? SIDN Labs dates the launch to October 26, 2026. The dataset site says the files were first released on October 25.The Node October patch.
Who it is for
SIDN says researchers can use the files for measurement, and operators can compare their own traffic with a large country-code domain and with the root. The generator is a fork of a tool written by TU Delft students, running on ENTRADA, and SIDN says any ENTRADA user can produce the same kind of files. The next step the post names is automation, so the daily files no longer depend on a manual batch. Until that automation is the announcement, the files are a batch publication.
What this page will not infer
The lab post says peaks and the share of queries for names that do not exist can hint at noise, misconfiguration, or abuse. A hint in a blog is not a finding about a particular network. This page will not read a response-code chart as an incident report. The charts, SIDN says, are updated periodically, and the files are the newer record when the two disagree. Asked directly, Which metrics does SIDN publish? Four of the six RSSAC002 metrics: traffic volume, traffic sizes, response-code volume, and unique sources.
The record
The record is the SIDN Labs post, the dataset site, and the RIPE Labs copy that put the item on the wire. Four metrics, daily, aggregated, with no names and no addresses. The first-TLD sentence stays inside the phrase to our knowledge. The March 10 file is an example of fields, not the story. The short answer this page will stand on is this: SIDN says no. The files are daily aggregates of traffic volume, sizes, response codes, and unique-source counts. Names and addresses stay out.
Sources
The reports this brief is filing. Futureweb did not republish them.
Questions
When did SIDN start the RSSAC002 site?
SIDN Labs dates the launch to October 26, 2026. The dataset site says the files were first released on October 25.
Which metrics does SIDN publish?
Four of the six RSSAC002 metrics: traffic volume, traffic sizes, response-code volume, and unique sources.
Which two metrics are left out?
Zone size and load time. SIDN points zone size at its existing statistics site.
Does a per-server unique-source file add up to the combined file?
SIDN says query counts add across servers. Unique-source counts do not, because one address can query more than one server.
Is .nl the first TLD to publish this format?
SIDN says to its knowledge, yes. That phrase is the lab's, not a survey this page conducted.