I have a scraper running on Heroku. It has been running for a while (+- 2 months) and it has days where it does great and reaches its 1,000 maximum and days during which it just magically restarts.
Does anyone know what the reason could be for such a restart? The scraper is showing no errors, the only thing I can find is the message below in the Heroku logs:
Feb 05 03:02:55 scraper heroku/web.1: Cycling
Feb 05 03:02:55 scraper heroku/web.1: State changed from up to starting
Feb 05 03:02:57 scraper heroku/web.1: Stopping all processes with SIGTERM
Feb 05 03:02:57 scraper heroku/web.1: Process exited with status 143
Feb 05 03:03:16 scraper heroku/web.1: Starting process with command `npm start`
The
Cyclingbit of the log here is the interesting one.Heroku will restart dynos every 24h, and this process is called "cycling". This is what you're seeing here.