- tl;dr - My cluster-wide prometheus scraping config was picking up port 25 as a good place to send GET /metrics requests, and broke my surprisingly fragile email setup (Haraka -> Maddy), simply by inundating it with invalid HTTP requests to a port meant for SMTP. My mailer setup makes use of two pretty good F/OSS tools that I rarely have problems with: Haraka Maddy They work great, and I don’t spend much time thinking about them – I often set up a new maddy instance with separate backups...
