As a Software Engineer, I deploy critical application code using the critical infrastructure consisiting of Jenkins and Terraform. I also manage AWS services like EC2, RDS, and ELB. I am responsible for handling on-call issues, deploying data bundles to various environments, and I operate on a weekly or bi-weekly deployment schedule based on requirements. We follow the Agile methodology and tracking work with tools like Jira.

We avoid application downtime by using Kubernetes' scaling features, such as horizontal pod autoscalers and load balancing services. This ensures our application handles increased requests efficiently and remains robust and scalable.

In Kubernetes, we use node-based architecture with nodes and pods and follow practices like RBAC and rollback. Multiple pods can run concurrently. We benefit from Kubernetes' ability to autoscale pods and use horizontal pod autoscalers to adjust the number of pods based on metrics like CPU or memory usage, ensuring efficient resource allocation and stability under load.

We sometimes face challenges during version upgrades, such as failures when migrating Kubernetes versions. 

Additionally, changes made by AWS services, like those in CodeBuild, require investigation to assess impacts on our applications, which can lead to challenges.

I've been using Kubernetes for the last three years. Additionally, we received an email from AWS mentioning changes starting on January 30th.

We do robust testing before deploying to production, undergoing multiple phases like testing, staging, and acceptance, to ensure stability. We rarely encounter production bugs, focusing on enhancements and UI changes instead.

Kubernetes provides scalability by using horizontal pod autoscalers that adjust the number of pods based on CPU or memory usage. The load balancing service distributes traffic across multiple pods, ensuring scalability and availability without straining any single pod.

I rate Kubernetes eight out of ten. 

I would recommend it to others as it is widely used.

Mostly, the company is a product-based company, primarily in the hard disk and flash business. A majority of the work involves firmware development. We are using Jenkins for the build process like CI pipeline. We are not using the CD pipeline in this case, only for building the artifacts.

Jenkins is a CI/CD tool and is the most robust tool. Although it comes from the open-source family, most companies use it just for cost-saving purposes. We are using it for the company's own infrastructure, not to serve a client. The company is into R&D stuff over here in India. In R&D possibly, it could be a reason they did not want to invest much by buying the product from different vendors other than going for open source.

The major drawback with Jenkins is the lack of support. If any issues arise, we need to fix them on our own, either by upgrading, going with a new version of Jenkins, or rolling back to the previous version. 

Unlike my previous company where we used Azure DevOps, where issues could be logged directly to Microsoft for a fix, this kind of support is not available for Jenkins.

I have been using Jenkins for almost one and a half years since moving from my previous company.

Jenkins is a stable product. From a rating standpoint, I would give it around seven to eight out of ten.

In terms of scalability, we have to rely on in-house efforts and online resources since Jenkins is open-source and doesn't offer built-in support. I would rate scalability around seven out of ten.

Support for Jenkins is not available as it is an open-source product. We need to rely on our own resources or external information sources like Google.

Neutral

In my previous company, we used Azure DevOps primarily since it was a Microsoft partner company.

We took support from a different vendor to write a script using Ansible, which helps us deploy Jenkins using new versions and all that.

Jenkins is used in many companies to save money, especially within R&D divisions, by avoiding the expenses of proprietary tools.

Jenkins can be compared with Azure DevOps and CircleCI, but Azure DevOps comes with comprehensive project management and ticketing, unlike Jenkins.

I would rate Jenkins between seven and eight out of ten. It is a good product for those who choose it based on their research; however, it lacks support for troubleshooting issues.

We use the solution for continuous integration and deployment.

The deployment time has reduced considerably. Since we started using it, the solution has taken care of everything that we used to do manually. It helps us with the automation of the deployment process.

The most valuable feature of the solution is its integration between different tools.

The dashboard needs to be improved. Though the access management and authentication functionalities are present, the dashboard and UI could be more user-friendly. The product has many plug-ins. Users have to go through the documentation to be able to use the product.

The UI must be more user-friendly. The information should be available in the dashboard itself. The users shouldn’t have to refer to the documentation. When a user hovers over the elements on the dashboard, it should reveal information about them.

I have been using the solution for the past three years.

The tool is very stable. I rate the stability an eight out of ten.

The tool is highly scalable. I rate the scalability an eight out of ten. We have around 600 to 700 users in the organization. We are using the tool 24/7.

CloudBees provides us with support for the product. It takes time to get a response from the support team. The response time for support must be reduced.

Neutral

The initial setup of the solution was complicated. I rate the ease of setup a six out of ten. We had to refer to additional materials to understand how to incorporate and integrate plug-ins with different software and endpoints. The setup and planning process took around one to two days.

To deploy the solution, we need a Kubernetes cluster. We implemented the solution in-house. We need one or two DevOps administrators to deploy the solution. Depending on the workload, we need five to six people to maintain the product.

We have saved time on deployment. Once the developers add new features to the application, the deployment process takes only a few minutes. We have received an ROI of almost 80% to 90%.

The product is expensive. The open-source version is free, but small companies would not be able to afford the cloud-based version. I rate the pricing an eight out of ten. Additionally, we have to pay for the product’s support.

People must consider the solution for enterprise support if they find it easy to read the documentation. If an organization has a good number of people for maintenance, they can choose the product. Overall, I rate the tool an eight out of ten.