When running Elastic Agent through fleet with packages, debugging issues without having direct access to the cluster is tricky. If a user has problems, it is not just asking him for the log file of the agent but many other parts play a role.
In a recent discussion with @urso the idea of having a "diagnostics download button" would be nice. It would collect all the necessary information, pack it into a zip file and make it possible to share it. This zip file would contain informations like:
Last x log lines of the Elastic Agent and the processes
When running Elastic Agent through fleet with packages, debugging issues without having direct access to the cluster is tricky. If a user has problems, it is not just asking him for the log file of the agent but many other parts play a role.
In a recent discussion with @urso the idea of having a "diagnostics download button" would be nice. It would collect all the necessary information, pack it into a zip file and make it possible to share it. This zip file would contain informations like: