How to configure Jenkins using HA?

We currently use Jenkins as our CI system, and there is one main server and slaves that are provided by Saltstack on Openstack. If our main Jenkins server goes down, we need to create a new wizard, and we need to extract the files from the old wizard and place it in the new ones, but it will take at least 30 minutes.

Is there a way to configure Jenkins with high availability?

I already checked with Gearman Plugin , however, if the Gearman server is down for some reason, we need to configure HA for Gearman as well.

Are there other ways to configure high availability for Jenkins?

+6
source share
2 answers

Jenkins doesn't have a big HA story; the best thing you can do with the open source version is to put all the files in $ JENKINS_HOME into a shared file system and then have a cold backup master machine that you can deploy if the active master goes down. This will reduce the time to switch to another resource if it is required that the wizard restarts, which usually takes several minutes.

You can also see the CloudBees Jenkins Enterprise offer, which includes a high availability plugin .

+4
source

I am using a doc cluster from scratch to create an active / passive Jenkins WAN-HA cluster. See the attached Architecture Diagram for Jenkins HA using a pacemaker .

/etc/init.d/jenkins will need to be converted as an ocf script agent. Currently, I manually run Jenkins through systemd on the pcmk-2 server when pcmk-1 not working.

+2
source

All Articles