June 2018 Summaries

The blog post by Matt Richardson discusses the development and improvements made to Octopus Cloud by using Octopus Deploy extensively within the company, a practice known as "dogfooding." Through this approach, the team has been able to identify and address several issues, such as improving the OctopusDSC module for better high-availability scenarios, enhancing security features to prevent sensitive data leaks, and optimizing script execution by introducing Workers for greater isolation. The team also focused on metrics collection using tools like Elasticsearch and Grafana to monitor performance, leading to better resource management and cost efficiency. The post highlights usability enhancements and bug fixes that have been implemented based on user testing and real-world deployment experiences, such as simplifying multi-tenant deployments and refining user interface elements for clarity and efficiency. Despite the progress, the team acknowledges there is still a long list of potential improvements to pursue, emphasizing their commitment to continual enhancements of the Octopus Cloud platform.

In the development of the Octopus Cloud platform, the team initially faced a decision between using Amazon Web Services (AWS) or Microsoft Azure, ultimately choosing AWS due to their familiarity with its services. Despite initial attempts to implement Docker and Kubernetes on Windows, challenges led them to opt for a more traditional setup using individual EC2 instances for each customer, influenced by the concept of a Minimum Viable Product. The team employed Terraform for creating a custom Virtual Private Cloud (VPC) setup, designed to support significant scalability, and prioritized security by engaging Insomnia Security for penetration testing, resulting in a robust network configuration. Octopus Cloud servers currently run on t2.medium EC2 instances, with plans for high availability and regional expansion to include Sydney, London, and Frankfurt. The company is also looking to integrate Docker and Kubernetes in the future to enhance server provisioning, security, and cost efficiency.

Flexible working has significantly transformed the life of Jason Brown, a Cloud Architect at Octopus Deploy, allowing him to escape the stressful and costly lifestyle of Sydney for a more relaxed and fulfilling existence in Bright, North-East Victoria. After struggling with long commutes and high rent in Sydney, Jason embraced the remote-first culture of his workplace and relocated to a picturesque town known for its cycling trails and natural beauty. Despite facing some initial challenges, such as setting up internet access, Jason found that the slower pace of life, along with the ability to enjoy outdoor activities like cycling, greatly enhanced his well-being. He appreciates the tranquility and simplicity of rural living, despite the occasional inconvenience, and credits the flexibility offered by Octopus Deploy for enabling this positive change.

The post by Robert Erez discusses the integration of Octopus Deploy with Bitbucket Pipelines to streamline continuous deployment processes using containerized build systems. Since its original publication, Octo.exe has been renamed to Octopus CLI, and up-to-date container images are now available on DockerHub, facilitating more efficient deployments. This integration allows for precise tailoring of build processes to application requirements without conflicts, using Octopus Deploy to manage deployments and Bitbucket Pipelines to automate builds from Bitbucket repositories. The use of containers, such as the octopusdeploy/octo:4.37.0-alpine image, simplifies the deployment process, allowing commands like "octo pack" and "octo push" to be run efficiently. The approach can also be applied to other containerized systems like CircleCI, making it easier to set up and manage CI/CD pipelines in a streamlined and containerized environment.

In 2018, Octopus Deploy announced the launch of Octopus Cloud, marking a significant shift from traditional on-premises deployment to cloud-based solutions, driven by the changing landscape where applications predominantly run in the cloud. The development of Octopus Cloud involved overcoming numerous challenges related to cloud architecture, security, and infrastructure, requiring the team to adopt creative solutions and break down knowledge silos among team members. The transition aimed to provide seamless deployment experiences by managing cloud instances via a special instance of Octopus, incorporating robust alerting systems for continuous monitoring. As Octopus Cloud entered the invite-only early access phase, the focus was on optimizing the platform for improved performance and cost efficiency, with ongoing efforts to enhance usability and responsiveness.

Node.js, a popular JavaScript runtime, has gained significant traction as a server-side technology due to JavaScript's widespread use and developer familiarity. However, transitioning from compiled languages to Node.js can lead to poor deployment practices, such as directly pulling code from GitHub and encountering issues with dependency versioning due to the semver range in packages.json. These practices can cause unintended production issues, like running untested dependencies and facing deployment failures if npm servers are down. To mitigate these problems, it is recommended to package Node.js applications like compiled languages, using tools like Octopus Deploy's octojs CLI to create deployable artifacts. This approach involves installing production dependencies, building the application once on a build server, and packaging it for consistent deployment across environments, ensuring stability and reliability in continuous delivery pipelines.

Octopus Deploy 2018.6 introduces several user-requested improvements and foundational changes aimed at enhancing scalability for larger teams. The release includes updates such as expanded control types for prompted variables and the ability to set sensitive output variables, enhancing security for dynamic provisioning scenarios. It lays the groundwork for future features like "Workers," which will improve deployment performance by offloading tasks from the Octopus Server, and "Spaces," which will allow users to create customized environments within Octopus, beneficial for large organizations with numerous projects. Licensing changes affect only new installations after March 31, 2018, requiring a commercial license for those who began trials after this date. Additionally, the release removes certain commands related to external workers due to the integration of the new worker feature, while ensuring no significant breaking changes for existing users. The update also includes backend enhancements and optimizations that may require additional upgrade time.