High Availability and Horizontal Scalability

Version:Starting with 2.414.2.2

Who needs it?

  • CloudBees CI Admins and Team Managers who need to ensure that CloudBees CI is always up and running and services remain uninterrupted during infrastructure updates so developers can use the service to build and test code.

  • Development teams who rely on CloudBees CI to build and test their code need the service up and running to do their job.

The problem

Overloaded monolith controllers slow the software delivery life cycle. The performance is slow in terms of the controller having a slow startup, slow UI, longer running builds, and even outages and prolonged downtime that affect many users and halt the build process.

Downtime stops the software delivery lifecycle and causes delays in delivery to customers. Services are down. Work stops. Business outcomes are missed.

The system is seen as unstable if this downtime is a regular occurrence. Unstable infrastructure results in downtime. And without fast recovery, this results in work loss.

CloudBees can help

We’ve all experienced the challenges of downtime - planned or unplanned. Restarting a Jenkins controller takes time, causing builds to stall, webhooks to vanish, and web browsing to halt. Administrators feel the pressure to minimize downtime by avoiding controller restarts, setting high resource limits, and carefully scheduling maintenance windows. Downtime is no joke.

At CloudBees, we get it. CloudBees CI is built on Jenkins, that is why we’re uniquely positioned to tackle this downtime challenge. Our solution transforms controllers into stable, reliable, and high-performing powerhouses that eliminate downtime and keep your services up and running when you need them. Say goodbye to after-hours work and weekends - it’s time for uninterrupted productivity.

The solution

CloudBees CI High Availability Mode eliminates downtime and ensures continuity to keep your software delivery pipelines running during unexpected losses or routine infrastructure updates. Active-active HA means that the operational state of your controller is always up and running regardless of these losses. All you see is uptime.

How we do it better

Though versions of HA do exist with older versions of CloudBees CI and third-party implementations with Jenkins, this is Active-Passive HA, and there is still downtime to contend with. Having Active-Active HA built into CloudBees CI provides zero downtime Jenkins and resilience in services not seen with previous solutions.

Use cases

  • HA removes the single point of failure and provides automatic recovery from loss without downtime.

  • HA provides automatic recovery from Kubernetes’ maintenance with minimum downtime and gives the freedom of scheduling routine maintenance without affecting software delivery.

  • HA allows automation at scale by automatically spreading workloads across controller replicas to improve controller stability, resilience, and performance.