Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Use multiple clusters for checking out upgrade path at least.

Plan regular upgrade so it's moved in small chunks and so that teams are aware of where they stand with APIs etc.



This is garbage path compared to controlled quantity. In 2008 no one would have countenanced this for good reason. What is the reason now?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: