# Revision history [back]

To expand on the solution provided in the previous answer.

depth_image_proc::PointCloudXyzNode publisher has a "sensor data" QoS profile, which means it has a "best effort" reliability policy. This policy is incompatible with the default QoS profile that uses "reliable" as its reliability policy.

ros2 topic echo was still working because it defaults to "sensor data" as a QoS profile when no profile is provided.

To whoever may run in this same issue, you can check the profile of every publisher and subscriber on a topic by using ros2 topic info -v [topic_name].

Maybe a warning of some sort should be included somewhere to notify if two nodes are using incompatible QoS profiles?