Pablo Garrido

It is explained here: https://docs.vulcanexus.org/en/latest/rst/tutorials/micro/memory_management/memory_management.html#id1

Where are you running the `ros2` command? in another docker?

Are you setting some domain in the micro-ROS Client?

Does it works in a local ROS 2 installation or in a separate docker running ROS 2?

Well, it seems to be something related to Fast DDS communication between the micro-ROS Agent and ROS 2 because the communication between micro-ROS Client and Agent is correct as far...

In any case, try to add `--ipc=host` in the docker launch, to check if it related with Shared Memory discovery between Fast DDS Instances.

Maybe you can increase the priority of the stepper task to a greater number than the micro-ROS task

Does it increase if after reaching `AGENT_CONNECTED` you just run `rclc_executor_spin_some` without checking `EXECUTE_EVERY_N_MS(200, state = (RMW_RET_OK == rmw_uros_ping_agent(100, 1)) ? AGENT_CONNECTED : AGENT_DISCONNECTED;);` ?

What if you do not make a `imu__update` inside the timer?

Is the timer callback being called at 50 Hz if you keep it empty of operations?