search

sabre1041 / ose-jenkins-cluster

[DEPRECATED] Jenkins slave cluster on OpenShift
17 stars 64 forks source link

one of maxSurge or maxUnavailable must be specified #6

Closed itewk closed 8 years ago

itewk commented 8 years ago

When I try to run a build of the jenkins-slave i get the following:

[root@master-1 ~]# oc logs jenkins-slave-2-deploy                                                                       
I0320 14:40:48.492010       1 deployer.go:198] Deploying from jenkins/jenkins-slave-1 to jenkins/jenkins-slave-2 (replicas: 0)                                                                                                                  
I0320 14:40:49.634951       1 rolling.go:232] RollingUpdater: Continuing update with existing controller jenkins-slave-2.                                                                                                                       
F0320 14:40:49.678825       1 deployer.go:65] one of maxSurge or maxUnavailable must be specified
sabre1041 commented 8 years ago

@itewk I've updated the blog post to indicate that the deploymentConfig should be removed instead of scaling down the number of running instances to avoid seeing the issue that you are experiencing.

If this is an acceptable resolution, would you be able to close this issue?

Thanks

itewk commented 8 years ago

Wasn't there an alternative option of setting the replica count to 0? Or was it something else. I know i got this working without removing the DC. Maybe worth listing both options?

sabre1041 commented 8 years ago

Scaling down to a replica count of 0 was the primary cause of the issue. The solution is to keep at least 1 pod running for the rolling deployment to work. If the Kubernetes plugin is to be used as the only method for spinning up slaves, the deploymentConfig should be removed

itewk commented 8 years ago

Hmm. I could have sworn I found a way to get it working with 0 replicas deployed but still with the DC. But in any case. yeah I can close this.