Comments (1)
I have a couple questions/concerns about this, and the proposed implementation.
If a workload already has a VPA created, currently Goldilocks still creates a VPA for that workload, potentially overwriting the existing resource
This is false, unless for some reason you are manually creating VPAs that are prefixed with goldilocks-
, which would be... odd?
or creating a conflicting VPA.
Only in the case of using VPAs in non-off mode, which is not the original intended use-case for Goldilocks.
- Why would an end user enable Goldilocks in a namespace where they are managing their own VPAs? This seems like optimizing for an unintended use-case
- What about the case where a goldilocks managed vpa is in recommend (off) mode, and it is prefixed with goldilocks-. This has no impact to existing VPAs and can provide meaningful information on the dashboard.
- How will this change in behavior be reflected in the dashboard?
- We are planning to add this PR where the dashboard provides a warning to the user if they are using an HPA, this will overlap with or eliminate that functionality, and should be considered.
In general, I would like to work towards supporting more VPA use-cases, but this is a very large change to how Goldilocks currently functions and is intended to be used, so I want to be diligent about answering all these questions and considering all the tradeoffs before making such a large change.
from goldilocks.
Related Issues (20)
- Add controller flags to fine-tune the creation of VPA objects HOT 6
- Question: Can goldilocks handle already created VPA objects? HOT 3
- Goldilocks can not recommend size for "Rollout" Kind HOT 8
- basePath is not being honored for static assets, e.g. CSS and Images HOT 8
- Many completed pods cause high memory consumption of controller
- Question: Get summary HOT 1
- Display resource quota of the namespace HOT 1
- Wrong "current" shown for memory in dashboard HOT 1
- goldilocks-vpa-recommender restarts all the time HOT 3
- Allow to disable CPU limit recommandation
- [RFE] HTML anchors for specific workload in namespace HOT 1
- not possible to force only specific VPA to be in Auto
- For different Kinds with same name, only one VPA created HOT 2
- Setting `--enable-cost=false` breaks the display of the yaml recommendations in the dashboard
- istio-proxy not shown HOT 3
- getAllTopControllers generate too much load on api server
- VPA Mode: Auto HOT 2
- helm chart: wrong values for dashboard template
- Goldilocks makes delay on pods initialization
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from goldilocks.