Github user StephanEwen commented on the issue:
https://github.com/apache/flink/pull/5371
Two more questions about this:
1. There is code in the shell scripts that rotates log file each time you
start / stop the cluster, with .0 /.1/.2/etc suffixes to the log files. Have
you tested whether that still works with the changed setup?
2. I think rolling by time is something that users expect more commonly
that rolling by size. What do you think here?---
