Add healthcheck

[skrul]

Temporary measure to restart TeslaBleHttpProxy until #76 is fixed. Adds a heathcheck that causes the container to go unhealthy when it is consuming more than 100MB of memory. This requires willfarrell/autoheal to be added to your docker-compose.yaml since docker itself does not restart unhealthy containers:

services:
  tesla-ble-http-proxy:
    image: wimaha/tesla-ble-http-proxy
    environment:
      - logLevel=debug
    container_name: tesla-ble-http-proxy
    volumes:
      - /root/key:/key
      - /var/run/dbus:/var/run/dbus
    restart: always
    privileged: true
    network_mode: host
    cap_add:
      - NET_ADMIN
      - SYS_ADMIN
    labels:
      autoheal-app: true
  autoheal:
    deploy:
      replicas: 1
    environment:
      AUTOHEAL_CONTAINER_LABEL: autoheal-app
    image: willfarrell/autoheal:latest
    network_mode: none
    restart: always
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock

Not sure if this should be merged, but could be useful to others experiencing this problem.

[wimaha]

Hi, thanks for the PR. It would be a good workaround, but not a real solution. Personally I don't want the FROM alpine:3.21 because the image would become very big file size.

If it's ok for you we wait for a real fix for this issue.

[skrul]

Yes, I was not expecting this to be accepted, was mostly just for me 😄 That said, it might be interesting to have some kind of health check since the proxy is intended to run all the time. I was first thinking of adding a check for "time since last command processed" but this makes a lot of assumptions about usage (having a predictable client like evcc helps) plus I wasn't sure if I could tell when the car was not present -- would the context deadline exceeded error from NewConnection work for this?