A MESA Lab repo (shihaoyue/yy_deploy_script) contains an organized suite of active attack tooling specifically targeting encrypted DNS: DoH/DoT downgrade attacks, DoH/DoT DDoS including a CVE-2024-4487 (HTTP/2 Rapid Reset) exploit script, IPv6 DNS response spoofing/injection (fakedns6 using forged-source-address "saddns"), a DNSSEC-downgrade proxy, DNSSEC-DDoS tooling, and an active DNS-fingerprinting probe pair (fpdns_client/fpdns_server under a folder named "TargetGZ"). This is direct evidence of GFW-adjacent R&D that specifically attacks the encrypted-DNS protocols (DoH/DoT) circumvention tools often rely on for resolver privacy.
10_doh_injection/degrade_phase1.py, 11_dot_injection/degrade_phase2.py, 6_dot_DDoS/att_pending_cookie.py, 7_doh_DDoS/cve44487.py, 4_v6_injection/src/ucr.edu/fakedns6/saddns, 12_TargetGZ/fpdns_client, 12_TargetGZ/fpdns_server, 2_dnssec_Downgrade/proxy.py (shihaoyue/yy_deploy_script.bundle)
Defense implications
- Do not assume DoH/DoT alone provides censorship resistance for a circumvention tool's bootstrap/resolver traffic — this repo shows active downgrade and DDoS (incl. HTTP/2 Rapid Reset, CVE-2024-4487) attacks purpose-built against encrypted DNS resolvers; pair encrypted DNS with a fallback resolution path that doesn't depend on DoH/DoT availability.
- Treat forged/off-path DNS responses over IPv6 (source-address-spoofed "saddns"-style injection) as a live threat model alongside IPv4 poisoning.
Related findings
A 2023 MESA Lab monthly report describes the 'TF' project's active-defense work: test cases for serial ('串联') HTTP hijack/tamper and parallel ('并联') DNS race-injection ('DNS抢答'), performance tuning that scaled active-defense capacity from 3 to 5 units, converting two existing TSG boxes to active-defense mode, and rewriting the active-defense flow-control logic to no longer depend on sapp.
An internal experiment using BurpSuite as a MITM proxy tests injecting a forged QUIC ServerHello carrying a connection-close frame ahead of the real server response, to make the client abandon its QUIC handshake; the author separately proposes recording every site known to support QUIC and simply blocking plain TCP connections to those sites, on the theory that this indirectly forces QUIC off since a censor cannot otherwise tell in advance which TCP flow would have upgraded to QUIC.
handingkang/ohmydns2 is a CoreDNS-fork DNS server built by a MESA Lab/IIE engineer ([email protected]) that bundles a "prober" active-probing plugin, a "v64dns" module, and an "atk" plugin with dedicated branches (atk_DDoS, atk_DDoS_resolver, atk_qp) implementing DNS response amplification and an attempted DNS injection/tampering feature ("注入篡改功能实现尝试"). This combines active DNS probing with resolver-based amplification/attack tooling in one codebase.
TSG's firewall 'Deny' security-policy action is confirmed implemented via three interchangeable mechanisms: TCP RST injection, forged HTTP 404 response-page injection, and DNS-redirect. Confirmed via commits to MESA_Platform/sapp and tsg/tsg-os-buildimage GitLab repos (git.mesalab.cn).
An internal TSG operations/troubleshooting manual lays out TSG's full traffic pipeline (NIC -> mrzcpd/marsio capture driver -> sapp DPI engine -> firewall/proxy(KNI->TFE)/active-defense/WAN-NAT policy branches) and shows engineers using maat_redis_tool to pull the live Redis-synced blocking policy tables (TSG_SECURITY_COMPILE, TSG_OBJ_IP_ADDR, TSG_OBJ_APP_ID) and filter them by numeric policy/object ID to debug why a block rule isn't firing.
An internal TSG functional-requirements spec ('加密协议JG') defines device support for identifying and blocking ECH, ESNI and QUIC traffic via per-connection SNI/region-matching tables (e.g. DF_QUIC_REGION), plus a companion 'FD报文全流程感知' feature that both passively monitors and actively injects synthetic verification traffic end-to-end through the network path to compute a live per-rule 'CT率' (breakthrough/penetration rate) — the system self-measures how often its own QUIC/ESNI/ECH blocks fail to take effect, and separately throttles logging/blocking detail for rule IDs receiving unusually high hit counts (DF_ATTACK_PROTECTION, 'targeted attack detection').