Welcome to the Linux Foundation Forum!

Lab 7.1 can't be passed, require some clarification

Options

In lab 7.1 of LFW212 having trouble fully passing the last validation test which should kill the brand service and confirm aggregator returns 500. It works as expected when tested manually, but running node validate.js gives me different sorts of errors almost every time. 2 common ones - status code 200 instead of 500 (running services manually first then running validate.js) and GET http://localhost:3000/1 failed to respond.

It's unclear how exactly it should be passed.
The lab is based on fastify which now includes sensible from the box, whereas the course suggests installing it and configuring it every time.

Answers

  • gicontz
    Options

    same sentiments, dunno how it should be passed. I manually terminate 1 of the service for the (status code 500) yet it cannot still be passed using the validate.js

  • gicontz
    Options

    ⛔️ GET http://localhost:3000/1 must respond 500 when brand service is down, got 200

    Seems like the service failed to terminate it by itself, i wish someone could help me out how to figure this out :(

  • gicontz
    Options

    Ok, haha, I've fixed it!

    do not start the boat and brand service :p
    Let the validator do that

    yey!

  • davidmarkclements
    Options

    thanks everyone, I'm adding a note in this exercise to be sure to stop the services

  • javier.thelegend
    Options

    Hey I don't understand this requirement "If either service is not available, respond with a 500 status code with any response body." when brand service is down it returns 500 however it is not passing the test case, I get failed to response as you can see in the next screenshot.

  • javier.thelegend
    javier.thelegend Posts: 2
    edited July 2022
    Options

    I fixed it increasing the timeout to the get method in validate.js file, now I got PASSED as result

  • xdxmxc
    xdxmxc Posts: 139
    Options

    ah yeah @javier.thelegend this is unfortunately one those situations that involve getting the timeout threshhold right for as many machines and scenarios as possible. Out of curiosity what did you set it to?

  • jhonny111s
    Options

    to me, we need some advice because it is more related to the request library.

    https://forum.linuxfoundation.org/discussion/858890/labs-ch-7-labs-1-servererror-validate#latest

  • xdxmxc
    xdxmxc Posts: 139
    Options

    @jhonny111s like all scenarios where timeouts must be used, the problem is intractable - in the next course update I'll think of ways to highlight this

Categories

Upcoming Training