milestone-testing-local.txt

This document explains how to run local milestone testing. 
We provide a couple convenience endpoints for you to do this:

You may want to make an `outputs` and `test_data` directory
to hold client outputs / diffs, and base generated test data respectively.
The following command does that for you (assuming you are running this in your base project directory,
and have these two folders existing there in relative locations already):
`make startcontainer outputdir=$(pwd)/outputs testdir=$(pwd)/test_data`

`make run_mile mile_id=1 server_wait=2`
Would be the corresponding command that runs milestone number 1, with waiting 2 seconds between each server restart 
(some client tests shutdown the server, thus a restart is needed before the next test).
Each milestone contains a collection of tests.
Note that you may want to vary server_wait if you anticipate that your server will recover persisted data more slowly than the default wait.
You can see the Make target `run_mile` and the corresponding script it calls, that is kept in `infra_scripts/test_milestone.sh`, to further
understand how the test battery for each milestone is structured under the hood.
You can see that there is a `infra_scripts/run_test.sh`, which the test_milestone.sh calls, for one test at a time. 

You may want to use these for convenience, when you manually trigger your client/server when you do specific debugging
in your docker container.

Remember you can `docker attach` to your container to run client/server commands directly inside.
Note that `exit` in a container will kill the container, consult the Docker documentation online 
to learn the detach and escape shortcuts (to keep container alive but leave the interactive mode).