Vic Iglesias
Vic Iglesias
Jenkins doesn't scale out so in this case Deployment or StatefulSet of size 1 will have the same behavior since Jenkins doesn't require a static hostname.
Can you try running with the pod explicitly defined in your Jenkinsfile and letting the Kuberentes plugin inject the jnlp container itself? For example: https://github.com/GoogleCloudPlatform/continuous-deployment-on-kubernetes/blob/master/sample-app/Jenkinsfile#L7
In the latest update to the tutorial #96 there is an example on how to define your agent config in your Jenkinsfile. Can you give it a try: https://github.com/GoogleCloudPlatform/continuous-deployment-on-kubernetes/blob/master/sample-app/Jenkinsfile
Hey @MarkPare this should not be the case. Can you post the minimal Jenkinsfile that reproduces this and also how your Kubernetes plugin is configured? What version of K8s and...
Were you able to get the pod to restart? It can take a couple of minutes for volumes to detach from the old node and reattach to a new one.
This is automated, it should not loop. If it is looping forever then there is a bug.
What version is your cluster?
Fixed in #139
Hey @marioparris-qless! Those volumes are provisioned on the fly and connected to the Jenkins master via GCE persistent disk. Can you show `kubectl describe pod `?
It could also be a quota issue that youve hit in your project.