Difference between Connection Timeout and Read timeout

[Lasse Thorst]

We're have a big data app with few users ( < 15) which we don't want to time out in the browser for a long period. So, the users hops on the app, and does their stuff - then they should be able to hop back and view this after x-time. X being at least 1 hour. 

The description for Connection Timeout and Read Timeout is kinda vague and hard to understand: 

" Connection Timeout: (900 secs) Seconds of inactivity before a browser connection to a worker process is considered to be idle and is closed. "

" Read Timeout: (3600 secs) Seconds of browser to worker inactivity after which the connection is considered to be idle and is closed. Use a value of 0 for non-interactive applications, e.g. a dashboard."  

What is the difference between these two? They both talk about browser to worker connection? 

[Joshua Spiewak]

The documentation gives a slightly longer explanation, but is essentially the same.

The connection timeout looks at traffic in both directions, while the read timeout is the application reading from the browser, thus if you have a dashboard you want to set that to 0.

The higher you set the timeouts, the more hours your application will consume, so make sure you are on an appropriate plan.