DNS WORKBENCH / SYNTHETIC DEMO / NO REAL NETWORK TRAFFIC

Follow the query trail.

An evidence report for synthetic-dns.pcap. Long, varied DNS labels can carry data—or ordinary application tokens. This workbench shows the observations behind each review signal.

178packets inspected
89query questions parsed
1client/domain groups to review
0coverage-gap occurrences

Why these groups stand out

example.test

Client 192.0.2.10
REVIEW SUGGESTED

At least 24 distinct names with long, high-entropy subdomain labels occur within 60s; 100% of queries in that window meet the label criteria.

24distinct encoded-looking names in window
3.871peak eligible-label entropy (bits/char)
48longest subdomain label
100% of 24 window queries meet the label criteria. Capture time 1700000000.000000–1700000023.000000 seconds.

Alternative explanation: Telemetry IDs, anti-abuse tokens, trackers or CDN traffic can produce similar labels. Review the packet evidence and context before drawing a conclusion.

Inspect packet evidence
  • Packet 1: 36ca6564787eb7dc8beadbe7cbf1f42f584c37fd28407d35.channel.example.test
  • Packet 9: 52ccf91583d0a59af6be19c408933d2aa99786b3a115994a.channel.example.test
  • Packet 15: 538aff05c47e8d7e80b982288461effbaecda1dedda383d8.channel.example.test
  • Packet 23: bc68bd18ca9da19b930fcf6d02ec094ecb06887ceab648a4.channel.example.test
  • Packet 29: fb86d1bdf859e17ff667d38f0ec16fbb9cf462ec11e2c4bc.channel.example.test
  • Packet 37: 3e7230f3873b296daf18a0cb27418a3aad134044304795da.channel.example.test
  • Packet 43: 2ca3879226febc6f829c48bda2a2fb4e57f8198ea8eadbaf.channel.example.test
  • Packet 51: 6616479d6829a7316a9b885bb3373efb94994c1b201fc4c6.channel.example.test

Up to eight sample questions from the strongest observed window. RCODE counts: {'3': 24}. 24 / 24 questions have a matched response within 30 seconds; missing responses may be a capture gap.

Showing 1 of 1 flagged groups. Full results are available in the JSON export.

All observed groups

ClientDomainQueriesDistinct namesMax labelEntropyResult
192.0.2.10example.test2424483.871Review
192.0.2.10school.test20240.0No pattern flagged
192.0.2.20service.test201483.692No pattern flagged
192.0.2.10app.test10130.0No pattern flagged
192.0.2.30tenant0.github.io1100.0No pattern flagged
192.0.2.30tenant1.github.io1100.0No pattern flagged
192.0.2.30tenant10.github.io1100.0No pattern flagged
192.0.2.30tenant11.github.io1100.0No pattern flagged
192.0.2.30tenant12.github.io1100.0No pattern flagged
192.0.2.30tenant13.github.io1100.0No pattern flagged
192.0.2.30tenant14.github.io1100.0No pattern flagged
192.0.2.30tenant2.github.io1100.0No pattern flagged
192.0.2.30tenant3.github.io1100.0No pattern flagged
192.0.2.30tenant4.github.io1100.0No pattern flagged
192.0.2.30tenant5.github.io1100.0No pattern flagged
192.0.2.30tenant6.github.io1100.0No pattern flagged
192.0.2.30tenant7.github.io1100.0No pattern flagged
192.0.2.30tenant8.github.io1100.0No pattern flagged
192.0.2.30tenant9.github.io1100.0No pattern flagged

Showing the first 19 of 19 groups. 0 questions excluded by the configured allowlist. A no-pattern result is specific to these thresholds and this capture.

Coverage & reproducibility

UDP port 53 DNS over Ethernet/VLAN or raw IPv4/IPv6. No TCP DNS, IP reassembly, DoH, DoT or checksum verification.

Allowlist (exact domains and their subdomains): []

Thresholds and processing limits
  • window_seconds: 60
  • min_unique: 10
  • min_label_length: 20
  • min_entropy: 3.5
  • min_suspicious_fraction: 0.5
  • max_packets: 250000
  • max_questions: 100000
  • max_file_mib: 128

Entropy is Shannon entropy of label characters, not cryptographic randomness. At least one subdomain label must meet both the length and entropy thresholds; distinct suspicious names and their fraction are then measured in a sliding window separately for each client and registrable domain. This is a configurable heuristic, not a validated classifier.