Comments (6)
Might not be super helpful, but env.reset()
stuck on my computer and the whole system is frozen (might be a sign of running out of memory) when I used Carla 0.9.9.3
. After I switched to Carla 0.9.4
it was resolved and the test script proceeded perfectly. (I was using Ubuntu 18.04 + Python 3.7.3)
from macad-gym.
Hi @tbienhoff , my apologies for the delayed response. This past week was a bit crazy.
Can you share which version of CARLA server and client you are using?
Listed below are the steps you could try to identify/debug the issue:
-
To eliminate other sources of errors, the first thing to try will be to see if you are able to launch the CARLA server binary on the P100 node without a display. This will make sure there are no OpenGL/Vulkan related rendering issues on the headless node.
-
If the CARLA server binary runs without issues, the next thing to would be to launch the CARLA server binary from a singularity container on the node. This will make sure there are no permission issues while accessing the GPU and running CARLA.
-
If 1. & 2. passed without issues, the next thing to check will be the CARLA client and server compatibility. Once you get to this step, please post your CARLA client and server versions.
-
I all the above goes well, there could be some previously unknown bug in MACAD-Gym which I can fix to resolve the whole issue.
Feel free to reach out if you need any assistance and I will try to respond soon.
from macad-gym.
I am able to complete steps 1, 2, and 3. I can run a CARLA server in singularity using the provided Docker images from CARLA (https://hub.docker.com/r/carlasim/carla/), and I can connect a CARLA client to the server. I have successfully done this for CARLA versions 0.8.2, 0.9.4, 0.9.5, and 0.9.6.
I am using version 0.9.5 with MACAD-Gym within a docker image built from the Dockerfile provided in Macad-Agent.
I believe the error might have something to do with not having a display for the server to output to.
from macad-gym.
Hmmm. From the last few lines of the error log you posted which contains the following lines:
sh: 1: xdg-user-dir: not found
error: XDG_RUNTIME_DIR not set in the environment.
The first hunch is that the user account inside the Singularity container doesn't seem to have the correct permissions or access to the X11/Xorg display server or to the video card (in case of SDL offscreen rendering) on the host.
from macad-gym.
Since you are able to launch a CARLA server inside the container and connect to it from a client (on the host?), I believe, this issue can be sorted out.
From the MACAD-Gym log you posted, it looks like, both multi-GPU mode and headless mode failed for launching the CARLA server failed on your node and therefore the CARLA server process was launched with rendering enabled which also eventually failed
Can you post the CARLA server log generated by MACAD-Gym (can be found in LOG_DIR/server_*
) when you run into this issue?
If you don't have the server log , you can also get the stdout console log from running the following commands inside the container and post it:
export CARLA_SERVER=FULL_PATH_TO_CARLA_SERVER_BINARY_ON_THE_CONTAINER
$(CARLA_SERVER) /Game/Carla/Maps/Town05 -windowed -ResX=600 -ResY=800 -benchmark -fps=20 -carla-server -carla-world-port=8376 -carla-streaming-port=0
from macad-gym.
This issue has been automatically closed because there has been no response to the request for more information from the original author. With only the information that is currently in the issue, there is not enough information to take action. Please reach out if you have or find the answers/request-information needed to investigate further.
from macad-gym.
Related Issues (20)
- Unable to use RL algorithms with continuous action space HOT 5
- Increasing number of steps per episode/ iteration HOT 3
- Multiprocess pickle Problem HOT 3
- How to set the spectator on the agent HOT 1
- Does macad-gym support multi-agent algos training? HOT 2
- Also stuck in env.reset() in example HOT 3
- How to create communicating environment?
- How to visualize the learning environment? HOT 2
- Support the library HOT 3
- `multi_view_render` will pop new display window on each frame with latest version of Pygame HOT 6
- The latest pull request is incomplete HOT 3
- gym version will affect the usage of ray[rllib] HOT 2
- PathTracker generate wrong route HOT 2
- when I use urban_signal_intersection_3c env ,happen error,please help me HOT 8
- How to customize a learning environment? HOT 2
- How do we port an existing multi-agent leanring algorithm such as IDDPG, IPPO? HOT 4
- Communication Mechanism
- v0.1.3 carla serve can't get connection HOT 1
- Implement of IMPALA Agent Examples HOT 2
- No support for other sensors? HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from macad-gym.