As discussed in the SIG meeting, we should review the most popular/critical Kafka clients for every cluster and check if they are running an up-to-date version of their libraries or not. To better segment the scope of the task and avoid spending hours and hours on it, we should probably pick up max 2/3 clients for each cluster (jumbo, logging, main) and report their status in the task.

Most of the clients should use librdkafka behind the scenes, let's make sure that the versions that we run are not too out-of-date.

We can use the Google doc that Balthazar created as starting point (there is a section dedicated to the clients): https://docs.google.com/document/d/1eHqkgKZitERH3M4NkJPA3qW3qWqF4haIkyMbZLJtATw/edit?tab=t.0