Open up2neck opened 1 month ago
APM Server is responsible for translating data that it receives from agents into a format that is compatible with Elasticsearch, and generally speaking mapping exceptions should not be caused by upgrading an agent. If that is happening, then we are likely doing something wrong in APM Server.
Could you please share some specifics about errors you have seen when upgrading agents? i.e.
APM Server is handy tool, widely used in our team, but so far we've realized it lacks validation between APM integration installed in Fleet and APM agents which sends data (correct me if I'm wrong).
It would be nice to have a way to configure version constraints for APM agents to receive data by the APM Server. It will reduce amount of invalid requests made to Elasticsearch, for example mapping exception errors, due to up-to-date APM integration and old APM agent.