Hi Jenkins community,
I need some help figuring out how we can achieve the desired configuration on Kubernetes. We use the Jenkins Helm chart and the Kubernetes plugin, each build starts in a new pod and ends by shutting down the pod. We also use Cluster Autoscaler to scale the node group on which our agents are scheduled.
We want to achieve the following:
- Keep one agent pod always running and ready for builds.
- Have other agent pods start when needed for builds and shut down after.
Does anyone know if we can achieve this with the current version of the Jenkins Helm chart and Kubernetes plugin?
Thanks for any tips!