feat(gitea): add Anubis anti-crawler proxy

Anubis (https://anubis.techaro.lol) is a lightweight proof-of-work
challenge that protects web services from AI crawlers and scrapers.

home/baseline/ssh.nix

@@ -13,25 +13,23 @@
     includes = [ "config.local" ];
     enableDefaultConfig = false;
 
-    matchBlocks =
+    settings =
       let
         nas = {
-          user = "admin";
+          User = "admin";
         };
         owned = {
-          user = "greg";
+          User = "greg";
         };
       in
       {
         inherit nas;
 
         "*" = {
-          dynamicForwards = [ { port = 10240; } ];
+          DynamicForward = [ "10240" ];
-          serverAliveInterval = 60;
+          ServerAliveInterval = 60;
-          extraOptions = {
           LogLevel = "error";
-            SetEnv = "TERM=xterm-256color";
+          SetEnv = { TERM = "xterm-256color"; };
-          };
         };
 
         "10.42.1.4" = lib.hm.dag.entryBefore [ "10.42.*" ] nas;
@@ -41,46 +39,46 @@
         "chronicles.thehellings.lan" = lib.hm.dag.entryBefore [ "*.thehellings.lan" ] nas;
 
         gh = {
-          user = "git";
+          User = "git";
-          hostname = "github.com";
+          Hostname = "github.com";
         };
         "src" = {
-          user = "git";
+          User = "git";
-          hostname = "jeremiah.shire-zebra.ts.net";
+          Hostname = "jeremiah.shire-zebra.ts.net";
-          port = 32222;
+          Port = 32222;
         };
         srcpub = {
-          user = "git";
+          User = "git";
-          hostname = "src.thehellings.com";
+          Hostname = "src.thehellings.com";
-          port = 2222;
+          Port = 2222;
         };
         ivr = {
-          user = "git";
+          User = "git";
-          hostname = "gitlab.com";
+          Hostname = "gitlab.com";
         };
 
         "ivr.thehellings.lan" = lib.hm.dag.entryBefore [ "ivr" ] {
-          user = "gregory.hellings";
+          User = "gregory.hellings";
         };
 
         "*.thehellings.lan" = owned;
         "10.42.*" = owned;
 
         "host.crosswire.org crosswire" = {
-          hostname = "host.crosswire.org";
+          Hostname = "host.crosswire.org";
-          user = "ghellings";
+          User = "ghellings";
         };
 
         fedpeople = {
-          hostname = "fedorapeople.org";
+          Hostname = "fedorapeople.org";
-          user = "greghellings";
+          User = "greghellings";
         };
 
         "src.fedoraproject.org pkgs.fedoraproject.org" = {
-          user = "greghellings";
+          User = "greghellings";
         };
 
-        "127.*".extraOptions = {
+        "127.*" = {
           PubkeyAcceptedAlgorithms = "+ssh-rsa";
           HostkeyAlgorithms = "+ssh-rsa";
         };

hosts/unstable/genesis/networking.nix

@@ -115,10 +115,15 @@ in
                 "@\tIN\tSOA\t${config.networking.hostName}\tgreg@thehellings.com (1 1m 1m 1m 1m)"
                 "\tIN\tNS\t${config.networking.hostName}"
               ];
-              makeHost = host: "${host.name}\tIN\tA\t${host.address}";
+              makeHost =
+                host:
+                [ "${host.name}\tIN\tA\t${host.address}" ]
+                ++ lib.map (a: "${a}\tIN\tA\t${host.address}") (
+                  if builtins.hasAttr "aliases" host then host.aliases else [ ]
+                );
             in
             pkgs.writeText "${domain}" (
-              builtins.concatStringsSep "\n" (preamble ++ (lib.map makeHost hosts) ++ [ "" ])
+              builtins.concatStringsSep "\n" (preamble ++ (lib.flatten (lib.map makeHost hosts)) ++ [ "" ])
             );
         in
         lib.mapAttrs

hosts/unstable/linode/git.nix

@@ -1,4 +1,4 @@
-{ ... }:
+{ config, ... }:
 
 let
   srcDomain = "src.thehellings.com";
@@ -6,15 +6,23 @@ let
 in
 {
   greg.proxies."${srcDomain}" = {
-    target = "https://gitea.shire-zebra.ts.net";
+    target = "http://unix:${config.services.anubis.instances.git.settings.BIND}";
     ssl = true;
     genAliases = false;
     extraConfig = ''
-      proxy_ssl_verify off;
+      #proxy_ssl_verify off;
-      proxy_ssl_server_name on;
+      #proxy_ssl_server_name on;
-      proxy_set_header X-Forwarded-Proto $scheme;
+
-      proxy_set_header X-Forwarded-Ssl on;
+      proxy_set_header X-Real-IP $remote_addr;
+      proxy_set_header X-Http-Version $server_protocol;
+      proxy_set_header User-Agent $http_user_agent;
       client_max_body_size 100000m;
+
+      #proxy_set_header Host $host;
+      #proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+      #proxy_set_header X-Forwarded-Proto $scheme;
+      #proxy_set_header X-Forwarded-Ssl on;
+
       # Ultimate AI Block List v1.7 20250924
       # https://perishablepress.com/ultimate-ai-block-list/
 
@@ -87,18 +95,22 @@ in
 
   networking.firewall.allowedTCPPorts = [ sshPort ];
 
-  systemd.services = {
+  services = {
-    haproxy = {
+    anubis = {
-      after = [
+      instances = {
-        "network-online.target"
+        git = {
-      ];
+          enable = true;
-      wants = [
+          settings = {
-        "network-online.target"
+            BIND = "/run/anubis/anubis-git/anubis.sock";
-      ];
+            COOKIE_DOMAIN = "thehellings.com";
+            SERVE_ROBOTS_TXT = true;
+            TARGET = "https://gitea.shire-zebra.ts.net/";
+          };
+        };
       };
     };
 
-  services.haproxy = {
+    haproxy = {
       enable = true;
       config = ''
         global
@@ -119,4 +131,18 @@ in
           server git-zeke zeke.shire-zebra.ts.net:32222
       '';
     };
+  };
+
+  systemd.services = {
+    haproxy = {
+      after = [
+        "network-online.target"
+      ];
+      wants = [
+        "network-online.target"
+      ];
+    };
+  };
+
+  users.users.nginx.extraGroups = [ config.users.groups.anubis.name ];
 }

hosts/unstable/linode/matrix.nix

@@ -2,12 +2,24 @@
 # them by the following commands:
 # nix run nixpkgs.matrix-synapse
 # register_new_matrix_user -k "B9EoPr2WV9hzwc7uL2Sx1JmvCeKDEOGCpB0uginQcQtEH4wzRtkSIdo7lltrjSQa" http://localhost:8448
-{ config, lib, ... }:
+{ config, ... }:
 let
   domain = "${config.networking.domain}";
   fqdn = "matrix.${domain}";
 in
 {
+  greg.proxies."${fqdn}" = {
+    extraConfig = ''
+      error_log /var/log/nginx/debug.log debug;
+      proxy_ssl_verify off;
+      proxy_ssl_server_name on;
+      proxy_set_header X-Forwarded-Proto $scheme;
+      proxy_set_header X-Forwarded-Ssl on;
+    '';
+    genAliases = false;
+    ssl = true;
+    target = "https://matrix.shire-zebra.ts.net";
+  };
   services.nginx = {
     virtualHosts = {
       # Server the '.well-known' files to find the Matrix API server
@@ -44,36 +56,6 @@ in
             return 200 '${builtins.toJSON client}';
           '';
       };
-
-      # Reverse proxy in front of the actual Matrix server
-      "${fqdn}" = {
-        enableACME = true;
-        forceSSL = true;
-
-        extraConfig = ''
-          error_log /var/log/nginx/debug.log debug;
-        '';
-
-        # Not the appropriate place for the chat client
-        locations =
-          (builtins.listToAttrs (
-            builtins.map
-              (
-                val:
-                lib.nameValuePair "/_${val}" {
-                  proxyPass = "https://matrix.shire-zebra.ts.net";
-                }
-              )
-              [
-                "matrix"
-                "synapse"
-                "dendrite"
-              ]
-          ))
-          // {
-            "/".extraConfig = "return 404;";
-          };
-      };
     };
   };
 

hosts/unstable/linode/nginx.nix

@@ -34,10 +34,5 @@ in
       ssl = true;
       genAliases = false;
     };
-    "doubles.thehellings.com" = {
-      target = "http://localhost:8081";
-      ssl = true;
-      genAliases = false;
-    };
   };
 }

lib/default.nix

@@ -15,6 +15,7 @@
       (name: value: {
         inherit name;
         address = builtins.getAttr netAttr value;
+        aliases = lib.optionals (builtins.hasAttr "aliases" value) (builtins.getAttr "aliases" value);
       })
       (
         lib.filterAttrs (

manifests/gitea/chart.yaml

@@ -119,6 +119,53 @@ spec:
         cpu: "100m"
         memory: "2Gi"
 
+    # Anubis anti-crawler sidecar
+    # Anubis listens on :8080 and proxies to Gitea at http://localhost:3000
+    # Ingress is updated to route to the anubis port instead of http
+    extraContainers:
+      - name: anubis
+        image: ghcr.io/techarohq/anubis:latest
+        imagePullPolicy: Always
+        env:
+          - name: BIND
+            value: ":8080"
+          - name: DIFFICULTY
+            value: "4"
+          - name: METRICS_BIND
+            value: ":9090"
+          - name: POLICY_FNAME
+            value: "/etc/anubis/policy.yaml"
+          - name: SERVE_ROBOTS_TXT
+            value: "true"
+          - name: TARGET
+            value: "http://localhost:3000"
+        volumeMounts:
+          - name: anubis-policy
+            mountPath: /etc/anubis
+            readOnly: true
+        resources:
+          limits:
+            cpu: "750m"
+            memory: "256Mi"
+          requests:
+            cpu: "250m"
+            memory: "256Mi"
+        securityContext:
+          runAsUser: 1000
+          runAsGroup: 1000
+          runAsNonRoot: true
+          allowPrivilegeEscalation: false
+          capabilities:
+            drop:
+              - ALL
+          seccompProfile:
+            type: RuntimeDefault
+
+    extraVolumes:
+      - name: anubis-policy
+        configMap:
+          name: anubis-policy
+
     strategy:
       type: RollingUpdate
       rollingUpdate:

manifests/mariadb-operator/chart.yaml

@@ -19,6 +19,7 @@ spec:
         kind: HelmRepository
         name: mariadb-operator
       interval: "1h"
+      version: "26.3.0"
 ---
 apiVersion: helm.toolkit.fluxcd.io/v2
 kind: HelmRelease
@@ -33,6 +34,7 @@ spec:
         kind: HelmRepository
         name: mariadb-operator
       interval: "1h"
+      version: "26.3.0"
   dependsOn:
     - name: mariadb-operator-crds
     - name: longhorn

manifests/mariadb-operator/databases.yaml

@@ -11,6 +11,7 @@ spec:
         kind: HelmRepository
         name: mariadb-operator
       interval: "1h"
+      version: "26.3.0"
   dependsOn:
     - name: mariadb-operator-crds
     - name: mariadb-operator

network.json

@@ -13,7 +13,8 @@
       "ip": "10.42.1.4",
       "pubkey": "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIEBecZUva9OnZuXLaBun6/1ITo5f9p0YMLPD+q0egLRS",
       "external": true,
-      "ts": "100.119.228.115"
+      "ts": "100.119.228.115",
+      "aliases": [ "s3" ]
     },
     "exodus": {
       "ip": null,