I am trying to figure out if Fluentd is able to use the source filename as part of the output filename.

I am using the S3 output plugin in the deployment of fluentd to ship the logs of hundreds of different pods to S3. We would need to be able to identify which logs came from which pod. As such, we could have the ID of the pod as part of the log file name.

For some reason, it looks like it is impossible to accomplish that with fluentd. Is there any convenient way I am missing to accomplish what I am trying to do?

Thank you.