Comments (9)
Ok, I see your point. Let's get the change as suggested in #111 merged.
from kube-score.
This was fixed in #111.
from kube-score.
You can disable the requirement of CPU limits with the --ignore-container-cpu-limit
flag. See https://github.com/zegl/kube-score#configuration
Does this match your needs?
from kube-score.
We also want to disable memory limits.
from kube-score.
We typically set requests for cpu and memory, but not limits as if we set limits Kubernetes will set aside at least that amount of memory resulting in poor memory utilization. We instead allow applications to float and if they use to many resources to get terminated.
from kube-score.
if they use to many resources to get terminated.
I'm curious, how are you achieving this without limits?
but not limits as if we set limits Kubernetes will set aside at least that amount of memory resulting in poor memory utilization
Isn't it the opposite? The request is reserved for the pod, and the limit is what's causing the pod to be OOMKilled if the limit is reached.
from kube-score.
If you use too many resources the kernel or the kubelet will kill you once the server near 100% memory.
Isn't it the opposite? The request is reserved for the pod, and the limit is what's causing the pod to be OOMKilled if the limit is reached.
This is not fully true. If you set a limit the schedule will set aside that much memory. If I have a service with a 1gb request and a 5gb limit and that service 99% of the time only uses 1gb the schedule will still set aside 5gb of memory for my service resulting in extremely poor utilization. In our case we would prefer to not have the limit and allow the service to risk going OOM to improve our overall utilization.
from kube-score.
If you set a limit the schedule will set aside that much memory.
I don't think that this is true, a node is allowed to be overcommitted, i.e. the sum of limits is over the amount of resources of the node.
from kube-score.
Yes, if overcommitment is set on your system, but it will still schedule based off of the limits if they are set. If the limits are not set it will schedule off of the requests and those will also leverage overcommitment. Overcommitment has no impact on my overall assertion of our desire to be able to disable limit checks.
from kube-score.
Related Issues (20)
- feature request: topologySpreadConstraints HOT 2
- Feature request: Optional config file instead of CLI arguments HOT 5
- Please submit a PR to update version in krew-index HOT 1
- amd64 docker images have not been published HOT 5
- Newest release(s) not respecting kube-score/ignore annotations. HOT 3
- Support colored output in Github Actions HOT 5
- Add flag --run-test strings HOT 8
- Documentation on the use of livenessProbe
- Gatekeeper policy compatibility HOT 2
- Facing an issue in kube-score code HOT 1
- Can we set rules to ignore init-Container HOT 2
- API for easier embedding into other applications
- Helm chart is reporting a missing PodDisruptionBudget when replicas count is not defined HOT 2
- Ignoring rule container-resources makes every rule exceptional HOT 4
- Support sidecarContainers HOT 1
- ignore-test test pods HOT 4
- lack of awareness for ResourceQuota for ephemeral-storage check HOT 2
- SARIF output is not valid HOT 4
- Add check for replicas > 1 HOT 1
- Add check for update strategy HOT 1
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 kube-score.