Aborted error exercise 5.1.2 and 5.1.3

[MustaphaJames]

When initialising the node on 5.1.2 it results in an error saying that the process was aborted. I figured out this can be fixed by changing the reliability policy to best effort (which is not mentioned in the notebook) of the initial pointcloud however then when adding the next two point clouds in 5.1.3 then the error returns when initialising those nodes so i do not believe this is the correct solution especially due to how inconsistent it is.
I have tried resetting the nodes muslitple times and 5.1.2 works occaisionally but 5.1.3 never works (connection is always aborted)
Is there a specific reason the process is being aborted?

[girishkumar.kannan]

Hi @MustaphaJames ,

The good news with the ABORTED message is that we know that the action is up and running.
So, if anything gets ABORTED from the action server, then it means there is some configuration issue in the server side.

In the basic_grasping_perception.cpp, check for the following:

1. cloud_sub_ parameter. Check if that topic matches the topic produced by the camera.
2. Try changing the value for clock Duration in the Get objects section under the execute() function from 3.0 seconds (default) to something like 5.0 or 10.0 seconds.

If you still have issues, do the following in order:

1. Terminate the action server.
2. Terminate the action client.
3. Re-launch the action server and wait for few seconds.
4. Send a goal to the action server and wait.
5. At this point 3 point cloud topics must be available in RViz2 - /points (or the points data from corresponding topic), /object_cloud, /support_cloud.
6. Send the goal again, then you should be able to visualize all the 3 point clouds on your RViz screen.

The above is the method I followed when I did this course. It should work for you too.

Let me know if you still have issues.

Regards,
Girish