The setup issues and the reason why Ngrok is being used.

[maxandersen]

I installed it as I have a s10 vacuum configured in switchbot app. But nothing happens when trying to fetch devices.

I see this error around ngrok (why is that used?) in logs:

ERROR:  You may not specify any endpoint or tunnel names with the --none switch
start - start endpoints or tunnels by name from the configuration file
USAGE:
  ngrok start [flags]
AUTHOR:
  ngrok - <support@ngrok.com>
COMMANDS: 
  config          update or migrate ngrok's configuration file
  http            start an HTTP tunnel
  tcp             start a TCP tunnel
  tunnel          start a tunnel for use with a tunnel-group backend
EXAMPLES: 
  ngrok http 80                                                 # secure public URL for port 80 web server
  ngrok http --url baz.ngrok.dev 8080                           # port 8080 available at baz.ngrok.dev
  ngrok tcp 22                                                  # tunnel arbitrary TCP traffic to port 22
  ngrok http 80 --oauth=google --oauth-allow-email=foo@foo.com  # secure your app with oauth
Paid Features: 
  ngrok http 80 --url mydomain.com                              # run ngrok with your own custom domain
  ngrok http 80 --cidr-allow 2600:8c00::a03c:91ee:fe69:9695/32  # run ngrok with IP policy restrictions
  Upgrade your account at https://dashboard.ngrok.com/billing/subscription to access paid features
Upgrade your account at https://dashboard.ngrok.com/billing/subscription to access paid features

[hsakoh]

The reason why the add-on uses Ngrok is explained in the Wiki (there is also an alternative to avoid using it).
As for why Ngrok is used in your setup, it’s because it was configured that way.

The logs you provided are too limited to identify the root cause of the issue, and I couldn’t fully understand what specific problem you are facing.

Could you clarify the following points?

• Are you intentionally configuring the setup to use Ngrok?
• Did you follow the installation guide provided in the Wiki?
• How did you set up the system? (e.g., as an HA-OS add-on, directly in Docker, as a .NET application, etc.)
• What values have you set in the add-on configuration?
• Could you provide a broader range of logs to help pinpoint the issue?
• What specific actions are you performing, and how does the outcome differ from your expectations?

Let me know so I can assist further!

[hsakoh]

Let’s tackle the issues one at a time:

• Unable to enumerate devices
• Error when starting Ngrok
  These appear to be two separate issues.

First, let’s focus on resolving the connection issue with SwitchBot:

To suppress the Ngrok issue, configure the add-on so that the webhook does not start:

• Set AutoStartServices to false in the add-on configuration.
  • Set WebhookService:UseWebhook to false in the add-on configuration.
  • Set WebhookService:UseNgrok to false in the add-on configuration.
• Next, set the LogLevel in the add-on configuration to Trace.
• Start the add-on after applying these settings.
• On the Ingress page, go to the Device Configuration page and press the Fetch Devices By SwitchBotApi button. What happens?
• Are any error logs output?