Problem initializing EM solver error

I installed AODT on a VM in a colocated environment, but when I execute “start UE mobility,” I encounter an error like the one shown in the attached image. What could be the cause of this issue?


@takanori.yatsunami021 Can you restart the back end containers?

I restarted backend_bundle-connector-1, backend_bundle-gis-1, and backend_bundle-clickhouse-1, then executed “start UE mobility” again, but the same error occurred. Are these the correct backend containers to restart? Additionally, I changed the SM from 89 to 86; could this change be causing the issue?

What about switching to the Kyoto.usd map?

I switched, but the same error occurred.

From the error report, maybe some parameters in scenario is incorrect.

@takanori.yatsunami021 How did you start the backend container? Can you. please list your steps?
Compute capability 86 is correct for RTXA6000. By default the script will pick capability of 80.
If possible, can you reboot?

Thank you for the advice. The backend container is started using the following steps:

First, I modified the docker-compose.yml file as follows:
nvcr.io/esee5uzbruax/aodt-sim:1.1.1_runtime_$GEN_CODEnvcr.io/esee5uzbruax/aodt-sim:1.1.1_runtime_SM86

Next, I ran the following command in the terminal to start the backend container:
docker-compose -f docker-compose.yml up

After that, when I executed docker ps -a, I found that some containers were not running.
So, I ran the following command to start all exited containers:
docker ps -a --filter “status=exited” -q | xargs docker start

@takanori.yatsunami021 I see that there are two aodt front ends, one on each GPU. By default, the front end runs on GPU 0. How and why is it runnin on GPU 1? Are you able to remove the PID 12559?

When I use kill 12559 to terminate the process, the process on GPU0 is also terminated, and the AODT screen closes. Is it possible to terminate only the process on GPU1? Also, could this issue be related to the fact that I am using a VM?

@takanori.yatsunami021 Are you able to solve the issue?

I solved it by changing from vgpu to passthrough.

thank you.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.