The author reflects on the shutdown of their self-hosted git server, which had been operational since 2011. This decision stems from persistent attacks by AI scrapers that overwhelmed the server with excessive requests, rendering it unusable. Despite previously having a public CVS server, the situation deteriorated to the point where the author chose not to rebuild the git server, leaving the management of such issues to others better equipped to handle them. Most of the repositories are now hosted on larger platforms like GitLab and GitHub, which the author has redirected users to. The author recently updated links to point to these larger repositories, marking a significant shift in their approach to hosting. Now, only the blog and a few other minor services remain on their self-hosted setup, which uses Jekyll for static page generation. This change minimizes the risk of overload from AI scrapers, unlike the previous cgit setup. However, the scrapers still caused one outage, filling up the Apache logs with 404 responses due to the absence of the cgit service, which the bots failed to recognize. The author managed to fix the logging configuration to prevent future issues.