Describe the solution you'd like
Currently, when sampling rate is not 100%, trace_id is not added to logs. It makes impossible to connect logs to individual requests.
Describe alternatives you've considered
Having another "request-id" header moving between services, IMO, is just a duplicate - it would serve the same purpose as trace-id. If request-id and trace-id are different, it makes harder to connect traces with logs.
Maybe I'm not aware of something and there is already a way to achieve the goal?
Describe the solution you'd like Currently, when sampling rate is not 100%, trace_id is not added to logs. It makes impossible to connect logs to individual requests.
Describe alternatives you've considered Having another "request-id" header moving between services, IMO, is just a duplicate - it would serve the same purpose as trace-id. If request-id and trace-id are different, it makes harder to connect traces with logs.
Maybe I'm not aware of something and there is already a way to achieve the goal?