A few of our users occasionally spin up pods that do a lot of number crunching. The front end is a web app that queries the pod and waits for a response.

Some of these queries exceed the default 30s timeout for the pod ingress. So, I added an annotation to the pod ingress to increase the timeout to 60s. Users still report occasional timeouts.

I asked how long they need the timeout to be. They requested 1 hour.

This seems excessive. My gut feeling is this will cause problems. However, I don’t know enough about ingress timeouts to know what will break. So, what is the worst case scenario of 3-10 pods having 1 hour ingress timeouts?

  • rglullis@communick.news
    link
    fedilink
    arrow-up
    1
    ·
    3 hours ago

    I asked how long they need the timeout to be. They requested 1 hour.

    That’s outright insane. Does this mean that if they connection has any type of hiccup, all their work is lost?

    Instead of having web apps working directly out of request-response cycle, these long running jobs need to be treated as a proper separate task, which gets a proper record entry in their database and could be queried for the results later.