Open Ali-Kazmi opened 3 years ago
We'll take into consideration your case. Feel free to submit your solution over the weekend.
some of our code is not compiling on the docker image due to not having access to the vorc_gazebo on the image
Not sure if this helps, but you should be able to git clone
the vorc repo from within your Dockerfile, like we do here:
https://github.com/osrf/vrx-docker/blob/7a047437f4a4589a9c2598286626427f83b284ef/vorc_server/vorc-server/Dockerfile#L111
If you still have problems related to package access, make sure you install everything and not just compile, e.g. colcon build
or catkin_make install
(I've only tried the former, can't guarantee the latter works too).
Thank you so much for the extension! We were able to resolve the Docker issues today and test our system against the vorc docker. I believe the update to our branch should be included in the pull request. The tag for our submission should be v9.2020.
If you would like to audit our code to check for alterations in functionality, please let me know. Everything we changed was with regards to getting our system to work in the docker environment, and not our system's functionality.
One source of confusion in the process was the docker entrypoint. In the tutorial page https://github.com/osrf/vorc/wiki/docker_repeat, it suggests to set the entrypoint with the --entrypoint flag. Due to our unfamiliarity with docker, we were unaware that this change could be transmitted in docker commits, which ultimately resulted in bash being used as the entrypoint for our first few attempts.
Thank you again for the extension and for hosting this competition. We learned a lot from it, and look forward to participating in the future.
@seantfish Thanks for this helpful feedback regarding the Docker tutorial. I can see why that could have caused confusion. If we keep this process going forward we'll add a note to warn about accidentally changing the entrypoint this way.
Hi everyone! I am with gt-mrg (Georgia Tech's marine robotics group). We have all of our files ready for submission (task 1-3 are working locally, but we didn't have time for task 4) but are having last minute issues with docker. Specifically, some of our code is not compiling on the docker image due to not having access to the vorc_gazebo on the image. We will upload a preliminary docker with the results we have so far, and can provide git timestamps for proof that the code is all there and working. Is it possible for us to do some docker stuff tomorrow for our submission? Thanks, and thank you to the competition organizers for a great experience! We learned a lot from the competition.