The errors and logs can now have log rolling if provided by the user.
The current customisable parameter of it are:
The maximal size of the file before rolling.
The maximal age/time of the file before rolling.
The number of backups to keep.
If we do not use a pointer here, the startup function that opens the log file stores the log file in a copy of the handler, not the same instance of the handler, causing panics during requests, which is bad, especially when the response is gzipped: the next recover() is beyond the gzip handler, so the browser downloads a gz file instead.
