Can't connect to tws

[IlanKalendarov]

I am using the default template for tws-docker-compose.yml and replaced the 127.0.0.1 part for the paper port to 0.0.0.0

And when using the bellow script:

from ib_insync import IB

ib = IB()
ib.connect(host='127.0.0.1', port=7497, clientId=1)

print("Connected:", ib.isConnected())
ib.disconnect()

I get timeout error.

When using telnet to check if everything works I do get

Trying 127.0.0.1...
Connected to 127.0.0.1.
Escape character is '^]'.

Don't really know what to do here, any ideas?

[gnzsnz]

are you sure that your host is mapping into 127.0.0.1 if docker is running in a VM the IP might be different.

[IlanKalendarov]

This is my compose file:

name: algo-trader
services:
  tws:
    restart: unless-stopped
    devices:
      - /dev/dri:/dev/dri #recomended
    shm_size: "1gb" #recomended
    security_opt:
      - seccomp:unconfined #recomended
    build:
      context: ./latest
      dockerfile: Dockerfile.tws
      tags:
        - "ghcr.io/gnzsnz/tws-rdesktop:latest"
    image: ghcr.io/gnzsnz/tws-rdesktop:latest
    environment:
      PUID: 1000
      PGID: 1000
      PASSWD: ${PASSWD:-}
      TWS_USERID: ${TWS_USERID}
      TWS_PASSWORD: ${TWS_PASSWORD}
      TRADING_MODE: ${TRADING_MODE}
      TWS_USERID_PAPER: ${TWS_USERID_PAPER:-}
      TWS_PASSWORD_PAPER: ${TWS_PASSWORD_PAPER:-}
      TWS_SETTINGS_PATH: ${TWS_SETTINGS_PATH:-}
      TWS_ACCEPT_INCOMING: ${TWS_ACCEPT_INCOMING:-manual}
      READ_ONLY_API: ${READ_ONLY_API:-}
      TWOFA_TIMEOUT_ACTION: ${TWOFA_TIMEOUT_ACTION:-exit}
      BYPASS_WARNING: ${BYPASS_WARNING:-}
      AUTO_RESTART_TIME: ${AUTO_RESTART_TIME:-}
      AUTO_LOGOFF_TIME: ${AUTO_LOGOFF_TIME:-}
      TWS_COLD_RESTART: ${TWS_COLD_RESTART:-}
      SAVE_TWS_SETTINGS: ${SAVE_TWS_SETTINGS:-}
      RELOGIN_AFTER_TWOFA_TIMEOUT: ${RELOGIN_AFTER_TWOFA_TIMEOUT:-no}
      TWOFA_EXIT_INTERVAL: ${TWOFA_EXIT_INTERVAL:-60}
      TWOFA_DEVICE: ${TWOFA_DEVICE:-}
      EXISTING_SESSION_DETECTED_ACTION: ${EXISTING_SESSION_DETECTED_ACTION:-primary}
      ALLOW_BLIND_TRADING: ${ALLOW_BLIND_TRADING:-no}
      TIME_ZONE: ${TIME_ZONE:-Etc/UTC}
      TZ: ${TIME_ZONE:-Etc/UTC}
      CUSTOM_CONFIG: ${CUSTOM_CONFIG:-NO}
      JAVA_HEAP_SIZE: ${JAVA_HEAP_SIZE:-}
      SSH_TUNNEL: ${SSH_TUNNEL:-}
      SSH_OPTIONS: ${SSH_OPTIONS:-}
      SSH_ALIVE_INTERVAL: ${SSH_ALIVE_INTERVAL:-}
      SSH_ALIVE_COUNT: ${SSH_ALIVE_COUNT:-}
      SSH_PASSPHRASE: ${SSH_PASSPHRASE:-}
      SSH_REMOTE_PORT: ${SSH_REMOTE_PORT:-}
      SSH_USER_TUNNEL: ${SSH_USER_TUNNEL:-}
      SSH_RESTART: ${SSH_RESTART:-}
      SSH_RDP_PORT: ${SSH_RDP_PORT:-}
    volumes:
      - ./config:/config # optional
      - tws_tmp:/tmp # recommended for performance
    # custom certificate and key
    #- ${PWD}/keylock:/keylock # required to avoid key generation by container
    #- ${PWD/key.pem:/etc/xrdp/key.pem # key
    #- ${PWD}/cert.pem:/etc/xrdp/cert.pem # cert
    ports:
      - "7496:7498" # live
      - "7497:7499" # paper
      - "3370:3389" # xrdp

volumes:
  tws_tmp:

I must say, after a restart it did work for some weird reason.
But sometimes it will say that there is another TWS session active from a different IP while I checked and there is none.
Sometimes after I do a container restart the error is gone.

[gnzsnz]

my point is that from your host point of view docker ports might not be available on 127.0.0.1

from within the TWS instance, connect to RDP and test this

if you get "yes" two times, then from within docker everything works fine.

if you can't connect to RDP, see "my point". At this point is a problem of your local setup.

[gnzsnz]

this one might help #98 (comment)