What usually leaks
- Pod and node IPs such as 10.42.3.17
- Cluster DNS names ending in .svc.cluster.local and .compute.internal
- Service account tokens and JWTs in auth errors
- Cloud keys passed in through environment variables
Tips
- The same IP always gets the same placeholder, so <IP_1> → <IP_2> still tells the story of a failing connection.
- Turn on "Keep private IPs" if the internal addresses matter for debugging and you trust the audience.