Stop scaling by hand: follow the load.
You'll learn to
create an HPA and read its target
Recap
- 1Needs metrics-server + CPU requests
- 2desired = ⌈current × usage ÷ target⌉
- 3Scales down slowly (5 min)
Check yourself
Four pods average 90% CPU, and the target is 45%. How many pods does the HPA want?
4.11
Stamp this lesson
Watched it and passed the check? Mark it in your logbook.