Currently, the new las2peer logger enforces a "use local file-based logging only" approach. This is not good. Logging should be refactored to notify the monitoring as default.
Also the duplicate timestamps and the logging class entry itself could be removed from the files, since they take up a considerable amount of space without adding any information.
Also, service custom messages should maybe be put into a separate local file (additionally to them being also sent to the monitoring node) to allow service developers to quickly check their messages while developing a success model for their service.
All in all, we should put some effort in improving monitoring and logging, and then write a short article about this in our wiki.