uorb top in the px4 console does not show the #LOST column

[sstroobants]

I assume this has been removed in an update but it has not been removed from the documentation (https://docs.px4.io/master/en/middleware/uorb.html)

I am also interested to know why it was removed actually. When I am using the ROS2 RTPS bridge and the sensor_combined_listener example for instance, ros2 topic hz /fmu/sensor_combined/out shows around 230Hz, while uorb top shows that it is published at 250Hz. It seems like once every so many messages, one gets lost/dropped and this causes the lower publish "rate" (in fact the rate is around 250Hz, only some messages seem to get lost along the way).
I am looking for a way to see where this message is lost..

[dagar]

It was removed from uorb top because the numbers often didn't make sense with consumers like logger only grabbing a subset of messages.

Instead we've started monitoring it from the Subscriber side. If there's a particular case where you can't tolerate any gaps then you can check the generation before and after an update.

https://github.com/PX4/PX4-Autopilot/blob/cc4152e10dd9645bfac3bf87b667e6f3c31f53e8/src/modules/navigator/navigator_main.cpp#L232-L238

[hamishwillee]

If one of you has an image we can use to update the docs that would be useful? https://docs.px4.io/master/en/middleware/uorb.html#uorb-top-command

[sstroobants]

@dagar Thanks, that is very clear and makes sense. I will look into checking the generation.
@hamishwillee, I could definitely make a picture, but my terminal does not have the same graphic settings as the one in the docs. An example output for me is as follows:

Also in the simulation environment I have a lot of instances for some topics as you can see in the picture above. Is that normal, or might this suggest something is wrong?