osrf-migration
commented
7 years ago Original comment by Ian Chen (Bitbucket: Ian Chen, GitHub: iche033).
The dry run will mimic the conditions of the finals. The difference is that we'll only provide you 1 round to run for the dry-run.
In the finals you can request to run rounds in parallel and we'll start multiple constellations for you,
e.g.
-
First request: Round, 1 and 2 => We'll launch one constellation with these 2 Rounds in it.
-
Second request: Round 3 => we launch another constellation with only Round 3 (this request can come any time).
-
etc
We'll ask you for your team name, github URL, and deploy key in order to launch a constellation for you.
You should be able to modify your code between rounds. The field computer will clone and build the latest version of the code in your github repo.
Original report (archived issue) by Tom Tseemceeb Xiong (Bitbucket: Tom_Xiong).
I'm confused on how the dry run will work. Is it going to essentially be like the finals? Or will it be more like the current practice state where we would enter in our git repo and run that from there with the 0.7.0 fully built srcsim software? Also, in accordance to the finals, can we modify code after each attempt? Is there a set time which our code should be frozen for the competition? It says for us to email osrf about starting a constellation but within this email should we also provide our git repo so that the docker gets rebuilt every time or will all final 5 runs will be with the exact same repo?