The logging mechanism of 4.x introduced new mechanics for performance tests where a test log is produced at the completion of each test, with large improvement within the handlers .

A similar behaviour should be achieved by not performance oriented tests while using, for instance, the httpjson and filelog handlers.

In order to understand the actual outcome, one should also think about taking into account when a test fails the sanity stage. In this circumstance, the handlers should provide a detailed log, independently of the nature of the test.

This addition would allow a more concise log management and greater integration with remote end data storage services.