Inspired by how a team of ex-colleagues support, tame and maintain a 1’000’000 LOC legacy monster written in PHP, I wanted to see if it would be possible to build a tool for PHP apps that allows adding debug statements to production code on-the-fly, without altering the code and avoiding redeployments.<p>The goal was to allow support engineers to diagnose root cause of business logic errors much quicker, by adding invisible “echo”, “var_dump” and “print_r” like statements to the production code with just a few clicks.<p>- Sometimes rollback process is painful, involves manual work or the team just has a move-forward culture (fixing bugs quickly with new deployments instead of performing rollbacks).<p>- Occassionally there are bugs specific to production environment data and conditions that can’t be easily reproduced in test/staging environments.<p>- Due to the legacy nature of the codebase it is not always clear how and why the data ended in the place it did and it is not immediately clear which parts of the code were involved in handling the request.<p>Continuous debugging aims to solve these issues both for the hands-on debugging as well as enriching data that is automatically sent to error tracking services.<p>I’ve built a proof-of-concept, technical preview (pre-beta) version that demonstrates technical feasibility of such approach for PHP.<p>I am looking for pilot customers and early feedback to understand if it would be worthwile to invest time into further improvements and additions (if fellow developers would benefit from such debugging workflow).<p>I’m always humbled by the high quality of HN discussions and deep insights it generates. I would greatly appreciate any comments and thoughts.