DEV Community

Sean Killeen
Sean Killeen

Posted on β€’ Originally published at seankilleen.com on

TIL: Kubernetes Auto-scaling and Requests vs Limits

I recently revised an incorrect mental model I had about Kubernetes as part of a strange experience, and I figured I’d share here in case it helps someone else.

Background / Challenge

  • I have a Horizontal Pod Auto-scaler set to scale at 80% CPU or 80% RAM, with a minimum of 2 pods and a max of 5.
  • I’ve given these pods limits of 1GB RAM (throwing some more resources at a problem temporarily πŸ˜‰ )
  • I recently saw my HPA set the pod count to 3. So I’m curious as to why – maybe these things are just hogging RAM?
  • I see the RAM threshold as 84/80 on the HPA, even with 3 pods running
  • However, I check our instance of Goldilocks which is giving us recommendations from a vertical pod auto-scaler (in observe-only mode) – and it’s telling me I can set our resources way lower.
  • So I run kubectl top pods --all-namespaces --sort-by=memory …and I see the pods are using 145mi, 118mi, 115mi – far from the 1024mi I specified

What gives?

Read more at SeanKilleen.com!

Heroku

Simplify your DevOps and maximize your time.

Since 2007, Heroku has been the go-to platform for developers as it monitors uptime, performance, and infrastructure concerns, allowing you to focus on writing code.

Learn More

Top comments (0)

Postmark Image

Speedy emails, satisfied customers

Are delayed transactional emails costing you user satisfaction? Postmark delivers your emails almost instantly, keeping your customers happy and connected.

Sign up