Kubernetes ReplicaSet: Difference between revisions

From NovaOrdis Knowledge Base
Jump to navigation Jump to search
No edit summary
 
(One intermediate revision by the same user not shown)
Line 6: Line 6:


=Overview=
=Overview=
A ReplicaSet is a [[Kubernetes_Workload_Resources#Overview|workload resource]] aimed at maintaining a stable set of replica Pods running at any given time, guaranteeing the availability of a specified number of identical Pods.


=Replicaset Controller=
=Replicaset Controller=
 
<font color=darkkhaki>
The replicaset controller is part of the [[Kubernetes_Control_Plane_and_Data_Plane_Concepts#Controller_Manager|controller manager]]. The controller implements a watch loop that is constantly observing the cluster making sure that the current state matches the desired state - the desired number of pods.
The replicaset controller is part of the [[Kubernetes_Control_Plane_and_Data_Plane_Concepts#Controller_Manager|controller manager]]. The controller implements a watch loop that is constantly observing the cluster making sure that the current state matches the desired state - the desired number of pods.
</font>
=Scaling=
=Scaling=
<font color=darkkhaki>
A ReplicaSet has a Scale sub-resource, allowing it to be controlled by a [[Kubernetes Horizontal Pod Autoscaler#Overview|horizontal pod autoscaler]].
A ReplicaSet has a Scale sub-resource, allowing it to be controlled by a [[Kubernetes Horizontal Pod Autoscaler#Overview|horizontal pod autoscaler]].
</font>

Latest revision as of 00:08, 13 July 2023

External

Internal

Overview

A ReplicaSet is a workload resource aimed at maintaining a stable set of replica Pods running at any given time, guaranteeing the availability of a specified number of identical Pods.

Replicaset Controller

The replicaset controller is part of the controller manager. The controller implements a watch loop that is constantly observing the cluster making sure that the current state matches the desired state - the desired number of pods.

Scaling

A ReplicaSet has a Scale sub-resource, allowing it to be controlled by a horizontal pod autoscaler.