Hi,

I have two nodes, both written in Python.

One, the tf node, is listening for the tf's from the opennitracker, analyzing them and then publishing messages based on its findings. The other one, the main node, is then acting on these messages while subscribing to an rgb topic and a depth topic from opennicamera.

The problem is, when the main node is running the tf node prints a lot of warnings with text like this: Inbound TCP/IP connection failed: connection from sender terminated before handshake header received. 0 bytes were received. Please check sender for additional details.

This means a lot of tf's were never received. Running rviz and subscribing to the same image topics does not result in any errors. Running both rviz and the main node still results in errors. In the main node, I also have a dynamic reconfigure client: cfgkinect = dynamic_reconfigure.client.Client("/camera/driver", timeout=30) Disabling this (cfgkinect = None) reduces the frequency of the TCP/IP errors.

This problem is also relatively new; I didn't have the problem in Electric.