Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
Bug: cluster mode is not enabled #2932
Issue type: Bug report
My Rundeck detail
In this response from the api we see the server uuid field "empty". But the two rundeck instances have a server UUID in /etc/rundeck/framework.properties file.
If you have something for initialize the server uuid when the rundeck engine is started up or any ideas to fix that?
How to reproduce Behavior
I also have this problem. I can't get the ServerUUID back in a REST call nor am I able to schedule jobs to the machine. It's not until after someone logs into rundeck do all of these configurations become available. I need to be able to automate the scheduling of jobs, to a new node, but can't because of this bug. :(
if what you want is to enable the cluster mode when the rundeck app is already started, i think it's not possible for now.
If you did that and the problem persist you can try to see if the server uuid is available when using another endpoint (ex: https://[MY_RUNDECK_HOST]/api/17/project/[PROJECT]/jobs).
@gschueler I just discovered I'm able to programmatically work around it by importing a junk job via the REST api into a project. This seems to "wake it up" so to speak and I can retrieve the serverUUID from the /system/info endpoint and do cluster operations via the /scheduler/* endpoints. I think I can use this workaround until the bug is fixed. :)