tor-docker-compose/docker-compose.yml at main · PacketCipher/tor-docker-compose · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
version: "2.4"

services:
  # tor:
  #   image: wizardrysteamworks/tor-snowflake:latest
  #   container_name: tor
  #   restart: always
  #   # Expose only to host's loopback
  #   ports:
  #     - "9050:9050"   # Tor SOCKS
  #     - "9053:9053"   # Tor control (optional)
  #   volumes:
  #     - ./tor/etc:/run/tor/config:Z
  #     - ./tor/data:/run/tor/data:Z

  multitor:
    # Note: rebuild this frequently as newer snowflake clients offer more stealth
    build: ./multitor
    container_name: multitor
    ports:
      - "16378:16378" # Custom Load-balancer
      - "16379:16379" # Maps host port 16379 to container port 16379 (HAProxy)
      - "16380:16380" # HAProxy Stats
    volumes:
      - ./multitor/templates:/usr/src/app/templates:Z # Maps custom templates from host to container
    environment:
      ## More Instances = More Untracability + More Chance Of Good Circuits & Less Top_N_Proxies = Better Ping & Balanced Top_N_Proxies = More Speed Due To Good Circuits Concurrency
      ## For Time-Sensitive Stuff Such As Gaming And Realtime Messaging Better To Use Ping Check & For Speed-Sensitive Stuff Better To Use Download Check (TODO: Two Separate Socks5 Server)
      - TOR_INSTANCES=44 # Default number of Tor instances, can be overridden in .env file or by CLI, 2x number of templates for default is preferred, 4x number of templates for high-restricted networks is preferred, for custom cases such as snowflake only or snowflake + relay scanner only you would assign high quantity for maximum chance of connection, regardless of number of instances the maximum bootstrap time is 30 minutes.
      - PROXY_MODE=custom_lb # Proxy Mode Options: socks, custom_lb
      - LB_Top_N_Proxies=4 # number of proxies to roundrobin for custom_lb, for networks with high interrupts and dpi the higher number is better but it may affect speed inversely
      # - LB_Algorithm=sequential # sequential algorithm will retry each request sequentially sorted by check mode
      - LB_Check_Mode=1 # 0 = ping check, 1 = download check, ping might be a better choice for networks with high interrupts and dpi
    cap_add:
      - NET_ADMIN # Required by Tor for some operations, and potentially by HAProxy/Privoxy.
    restart: always
  # cpuset: "0"
  # deploy:
  #   resources:
  #     limits:
  #       # Hard Ceiling: Never Exceed N CPU Cores (Default is No Limit)
  #       cpus: '2' # Fix Race Condition
  #     # reservations:
  #     #  # Lower Priority Than Default Processes (Default is 1024)
  #     #    cpu_shares: 512

  # dpyproxy:
  #   build: ./dpyproxy
  #   ports:
  #     - "4433:4433" # Proxy-Server
  #   restart: always
  #   command: ["--frag_size", "20", "--debug", "--port", "4433", "--host", "0.0.0.0"]

  psiphon:
    image: swarupsengupta2007/psiphon:latest
    container_name: psiphon
    depends_on:
      - multitor
    environment:
      - PUID=1000
      - PGID=1000
    volumes:
      - ./psiphon:/config:Z
    ports:
      - "1080:1080"   # Psiphon SOCKS
      - "8080:8080"   # Psiphon HTTP
    restart: always

  # Note 1: Doesn't Work As Tor Socks5 Doesn't Support UDP Protocol.
  # Note 2: But It May Work With Psiphon
  # warp:
  #   build: ./warp
  #   container_name: warp
  #   depends_on:
  #     - psiphon
  #   ports:
  #     - "8086:8086"   # WARP SOCKS5 Proxy
  #   cap_add:
  #     - NET_ADMIN     # Required to create TUN device and manage routing
  #   devices:
  #     - /dev/net/tun  # Required to mount the TUN device
  #   restart: always

  ### Note: UDP VPN Services Such As Wireguard And OpenVPN UDP Can't Be Used Due to Tor Socks5 Lack Of UDP Support; Only TCP VPNs Such As OpenVPN TCP And SSH And TCP Proxies Will Work. ###