What is our primary use case?
We use it as one of the DevOps tools in continuous implementation. We use the tool for continuous deployment. There are two parts to the tool: continuous integration and continuous deployment. We work for continuous deployments of applications. We have multiple pipelines in our environment and multiple applications. There are about 1,000 applications within our environment. Each application has multiple pipelines and branches configured. We pull the source code from Git. There's also something called a repository. We use Groovy scripts to provision pipelines. These pipelines have multiple stages, starting with SCM checkout, build stage and Sonar scan stage. Finally, it gets deployed to a particular server defined in the code. Before we execute the pipeline, we have different environments for development and production. For development, we test by executing directly. We hand it over to the production team for production to test on the specific branches.
What is most valuable?
It is a standalone tool, so no networking is possible by itself. However, it can be integrated with multiple vendors. CloudBees, in particular, supports integration with multiple vendors and offers enterprise support, including vendor support. Whenever we encounter issues, we can troubleshoot them with their help, as they provide continuous support. They also offer monthly knowledge transfer sessions for the team working with the tool. Jenkins is a standalone tool without cluster or networking capabilities; you can't integrate it similarly. CloudBees, however, supports different networking components, such as pods, Kubernetes, and Docker, which can all be integrated. CloudBees continues to develop and release new features.
What needs improvement?
Improvements in resolution time may vary since every environment has different needs. CloudBees is still a growing platform, but it has proven to be capable of supporting enterprise-level pipelines and CI/CD requirements.
They could increase the number of support channels to expedite resolutions or introduce a chat option similar to what Amazon and other vendors offer. A chat option would make resolving issues quickly and efficiently easier.
New features and updates in the market for tools like Jenkins and others are often developed as open-source projects. Since CloudBees is an enterprise-level platform, it ensures thorough testing for critical environments like production. While CloudBees is doing well in this regard, it's not yet adopted in all areas. We have a dedicated team for CloudBees and receive proper support whenever we deploy.
For how long have I used the solution?
I have been using CloudBees for seven to eight months.
What do I think about the stability of the solution?
We haven't encountered any performance issues with CloudBees. CloudBees is hosted on a particular server, either a clustered or standalone one. CloudBees' performance might also be affected if the server's performance is impacted. CloudBees has been stable.
What do I think about the scalability of the solution?
In our setup, the Jenkins server's hard disk, which hosts Jenkins, is used to host the NAS. The NAS contains the CloudBees files. We performed a migration where Jenkins files were copied to the NAS for CloudBees. Through CloudBees' GUI, we refresh or reload to retrieve the required files. This affects every pipeline.
Scaling can be done through the parameters we configure in the pipelines. The parameters are not limited to just one; whatever we need, we rely on plugins, and these can be integrated. As these scripts are further developed to meet future needs, scaling CloudBees becomes even more flexible. Teradata integration has been working perfectly for us.
Scalability depends on the number of pipelines we have. As the number of pipelines increases daily, the number of DevOps engineers required to handle the workload must match. This is because multiple teams are involved, often encountering issues that need troubleshooting. Multiple issues still arise even if we are disconnected, and engineers must resolve them. It depends on the environment and the current number of pipelines.
In our environment, we have almost 1,000 applications. Of those, around 45 to 50 applications are actively in use. Each pipeline is managed based on the environment, such as development or production. There are approximately 713 development pipelines, and the production pipelines are about half that number. The number of users is difficult to count, as different application teams use each pipeline. To be efficient, each pipeline requires at least one dedicated engineer, so it's typically a one-to-one ratio between pipelines and engineers.
How was the initial setup?
A DevOps engineer needs some knowledge to implement CloudBees. Once they approach the CloudBees team, they provide knowledge transfer during a scheduled conference call. They share comprehensive information using reference guides, making it easier to implement. However, engineers should have prior knowledge of Jenkins or CloudCI, which is essential for a smooth implementation.
We have implemented CloudBees in Kubernetes nodes, which are basically in the cloud only. In the initial migration stage; we have it in our local NAS. There'll be a future production, and one will be more for the CloudBees.
We are primarily involved in migrating pipelines from Jenkins to CloudBees. Migration is not difficult for us, but creating a new pipeline and ensuring it matches the Jenkins setup while executing can present challenges. One of the main issues we face is when new plugins are unavailable or when application support for plugins such as Java, Node.js, or Git.js versions is lacking. If the necessary versions aren't available, it becomes a problem.
What was our ROI?
There are costs involved, but the tool is useful and worth it. While we can rely on Jenkins, it lacks vendor support in a production environment. We are solely responsible for troubleshooting issues. This can be time-consuming, and if it's a production issue, it affects production time. With enterprise support, we benefit from immediate troubleshooting when issues arise, making it a valuable feature.
What other advice do I have?
If you use any plugins or whatever plugins you get, you'll have access to the latest versions of pipelines and branches. Everything, including the most recent versions, will be available. Whether it's a Java version or any other plugin, for example, if it's related to Java, they will provide the latest updates for the currently available versions.
It comes with vendor support and continuous assistance. CloudBees provides regular knowledge transfers about new and upcoming features and supports any issues that arise. For example, if new features are unavailable in the tool, we can raise a ticket or concern, and they promptly address it. They also communicate with us by scheduling calls with our tech leads to resolve any issues in real-time. Since different environments and development teams have varying needs, CloudBees ensures these are consistently met.
Similar to other infrastructure areas, it requires maintenance. Having the right options available is essential to maintaining it properly.
The core advantages of this tool are its enterprise-level vendor support and its ability to handle cluster-level and network-level configurations, which facilitate high availability in the environment—capabilities that other tools often do not provide. These are the two main pros: cluster concepts and enterprise-level support.
Overall, I rate the solution an eight out of ten.
Which deployment model are you using for this solution?
Public Cloud