| as requested in https://github.com/jenkinsci/jenkins/pull/4027#issuecomment-534558658 I'm deploying a new Jenkins from scratch on a single host using Docker all on top of AWS. Its authentication mode is set to SAML (using Okta) and we configure it using JCasC (Configuration as code). The deployment strategy we decided is to deploy a new instance each time a configuration change is made. However, in order to give a good experience to our end users, we want to make 2 steps before swapping between old and new release:
- Put the old instance in Quiet mode.
- Query running builds
- When running query builds = 0 then swap the instance. We have no problem on doing that with the API but... the problem is that we depend on one single thing: the API token!
Ideally we would like to add a fix token into JCasC file to be able to connect to the API once Jenkins the host has been configured. jenkins saml |