Balena OS Proxy Redirector setup killing docker internal network and not working properly

After configuring my Balena device to use the redsocks proxy redirector configured via Balena Supervisor I’m experiencing a series of issues…
To give some context:

  "network": {
    "proxy": {
      "type": "socks5",
      "ip": "",
      "port": 1080,
      "noProxy": [
    "hostname": ""
  • By executing the troubleshooting CURL command to use the redoscks redirector I’m getting healthy communication over my proxy
root@7c83344001c3:/src# curl -x socks5:// | jq
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100   316  100   316    0     0   1541      0 --:--:-- --:--:-- --:--:--  1549
  "ip": "${PROXY_IP}",
  "hostname": "ec2-${PROXY_IP}",
  "city": "Ashburn",
  "region": "Virginia",
  "country": "US",
  "loc": "39.0437,-77.4875",
  "org": "AS14618, Inc.",
  "postal": "20147",
  "timezone": "America/New_York",
  "readme": ""

So everything is configured as expected… And even like that I’m finding issues like:

  • If Proxy Settings are configured the host loses communication with Docker Internal bridge network containers… Containers that before had exposed a port and that could be reached from the host by using this setting they lost all communication… I suspect this is related to Balena Engine and a restart of it when applying proxy settings or a deeper issue where if proxy redirector is configured then the containers in the internal network do not work.
  • My supervisor cannot connect to Balena API, even if the proxy redirector is set properly… Obviously, updates also no not work and my device goes to VPN Only

Opened support ticket as well, have not got back any information yet, but it would be nice if this is a feature that actually works or if it has even been tested.