I’m using BalenaDash to display a Tempest weather info page. Generally the page runs fine for a couple days until it finally ends up with a Chrome error 64000. Restarting services gets things running again.
I’m guessing there might be a device variable that could be passed to the scheduler that would force a services restart, but I can’t seem to find any documentation on allowed variables. Or is there another approach I should be looking at?
Thanks
Hi there Chris,
Do you know if the page works again if it gets refreshed, rather than restarting the service?
If so there is an API you can use:
balenaDash uses the browser block under the hood, which as an API which you can use to control Chrome: GitHub - balenablocks/browser: A drop-in web browser block
You can therefore post to that endpoint, either from another service on the device or some other device. Either way you could either run a bash script to call the endpoint, or just deploy another block: balenaHub: an easier way to find and publish fleets, projects, and blocks for edge devices
and use that to CURL the command to refresh the page.
There is also an auto-refresh endpoint, which you would need to set each time the service started - but this is still a bit of an experimental feature. Let me know if you do use it, and any issues or feedback you have. 
If you find that you really do have to restart the whole KIOSK service, then you could use the supervisor API to restart the service with a cron job or the block mentioned above.
Hope that helps,
Phil
Hey… Did Phil’s suggestions above help?