Netflix Spinnaker: multi-cloud continuous integration tool

With the release of the open source Netflix Spinnaker delivery platform, the cloud-savvy video on demand company signals that the cloud is bigger than AWS

Video on demand provider Netflix has been successfully running much of their operation in the AWS cloud for six years now, and the relationship has been mutually beneficial. If people are talking about AWS, then the odds are that you’ll hear something about Netflix: whether it’s how they migrated their huge workload to AWS, their Simian Army resilience testing tool, or how well Netflix manages AWS outages. Now you can add Netflix Spinnaker to the conversation.

In a blog post a few months ago, Netflix first mentioned Spinnaker, an open source, multi-cloud deployment tool that was then under development. The post mentioned how the company was planning to move to Spinnaker from its own Asgard. The Netflix Spinnaker code is now publicly available.

What is Netflix Spinnaker?

According to Netflix:

Netflix Spinnaker is an open source multi-cloud Continuous Delivery platform for releasing software changes with high velocity and confidence. Spinnaker is designed with pluggability in mind; the platform aims to make it easy to extend and enhance cloud deployment models. To create a truly extensible multi-cloud platform, the Spinnaker team partnered with Google, Microsoft and Pivotal to deliver out-of-the-box cluster management and deployment. As of today, Spinnaker can deploy to and manage clusters simultaneously across both AWS and Google Cloud Platform with full feature compatibility across both cloud providers. Spinnaker also features deploys to Cloud Foundry; support for its newest addition, Microsoft Azure, is actively underway.

It’s very interesting that Netflix, which has always relied on AWS for its infrastructure, has now built a tool to migrate or deploy on AWS or Google Cloud Platform and Pivotal. Are they trying to reduce their dependency on a sole vendor?

Netflix Spinnaker is an open source, multi-cloud continuous delivery platform for releasing software changes. It provides both Cluster and Deployment management.

Cluster Management

Cluster Management deals with cloud resources like:

  • Server Groups. Collections of cloud-based virtual machines (VMs) built from AMIs or images. A server group can also be associated with a load balancer, Security Group, user information, region and availability zones.
  • Load Balancers. Protocols for managing traffic loads associated with a server group.
  • Security Groups. Rules defining network traffic access and firewall rules according to IP range (CIDR), communication protocol (e.g., TCP) and port range.
  • Clusters. User-defined, logical groupings of Server Groups within Spinnaker.

Deployment Management

Netflix Spinnaker deployment management provides for continuous integration and workflow. It facilitates the creation of pipelines that represent a delivery process that can begin with the creation of some deployable asset (such as a machine image, JAR file, or Docker image) and end with a deployment.

  • Pipeline. The key deployment management construct in Netflix Spinnaker. Pipelines are defined by a sequence of stages, triggers that kick off the pipeline, and parameters that get passed between pipeline stages and can be configured to issue notifications (via email, SMS or HipChat) as the pipeline executes. Triggers can be either automatic – like a Jenkins job, CRON schedule, or another pipeline – or manual.
  • Stage. A Stage in Spinnaker represents different phases in a pipeline’s lifecycle. Spinnaker comes pre-packaged with a number of stages, including:
    • Bake: Bakes an image within the specified region.
    • Deploy: Deploys a previously baked or found image.
    • Destroy Server Group: Destroys a server group.
    • Disable Server Group: Disables a server group.
    • Enable Server Group: Enables a server group.
    • Find Image: Finds a previously-baked image to deploy into an existing cluster.
    • Jenkins: Runs a Jenkins job.
    • Manual Judgment: Waits for user approval before continuing.
    • Modify Scaling Process: Suspend or resume server group scaling processes.
    • Pipeline: Runs a pipeline. This allows pipelines to be composed hierarchically.
    • Quick Patch Server Group: Quick patches a server group. Used for emergency patches.
    • Resize Server Group: Resizes a server group.
    • Script: Runs an arbitrary shell script.
    • Shrink Cluster: Shrinks a cluster.
    • Wait: Waits a specified period of time.

Netflix Spinnaker Components:

Spinnaker includes the following micro services:

  • Clouddriver: encapsulates all cloud operations.
  • Deck: user interface.
  • Echo: the event service that is responsible for triggering pipeline executions and forwarding pipeline events to listeners.
  • Front50: data store for pipelines, notifications and applications.
  • Gate: service gateway responsible for providing an API and UI to end users.
  • Igor: interface to Jenkins, Stash, and GitHub.
  • Orca: orchestration engine responsible for running Spinnaker pipelines and tasks.
  • Rosco: bakery responsible for creating images for deployment.
  • Rush: general purpose scripting engine.
Netflix Spinnaker

(Netflix Spinnaker components – courtesy: Netflix)

Setting Up Netflix Spinnaker

Setting up Spinnaker for development takes just a few steps. These instructions cover pulling Spinnaker from source and setting it up to run locally against Amazon Web Services and/or Google Cloud Platform accounts. We will then see how Spinnaker is configured for the AWS Cloud from a developer machine.
Before jumping in, make sure you have an AWS account. We’ll follow the steps from the Spinnaker getting started guide:

Create an EC2 role

  • Create a New Role in AWS Identity & Access Management (IAM) and set the Role Name to BaseIAMRole. EC2 instances launched with Spinnaker will be associated with this role.

Create an EC2 Key Pair

  • Create a Key Pair and name the key pair <aws_account_name>-keypair (note: this must match your account name plus “-keypair”).
  • Download the private key file <aws_account_name>-keypair.pem and change its permissions:
chmod 400 aws_account_name-keypair.pem

Create AWS credentials for Spinnaker

  • Create a new user in IAM. Create an access key for the user. Save the access key and secret key into the ~/.aws/credentials file on your workstation.
  • Click the checkbox next to PowerUserAccess in IAM, then click Attach Policy.
  • Create an inline policy with “Select for Policy Generator”.
  • From the IAM Policy Generator, select “AWS Identity and Access Management” as an AWS Service, and then “PassRole” for Actions.
  • Type * (the asterisk character) and “Apply Policy”.

Bootstrap

The bootstrap scripts are meant to run on Ubuntu 14.04 LTS and Mac OS X 10.11. Here’s now it works on Ubuntu:

  • Clone the Spinnaker from GitHub. Make sure you have git version 2.0 and above.
mkdir -p $SPINNAKER_HOME
cd $SPINNAKER_HOME
git clone git@github.com:spinnaker/spinnaker.git
  • Set the path as follows:
export SPINNAKER_HOME=/path/to/your/Spinnaker/workspace

Configure the Environment:

  • Before setting up Spinnaker, make sure your workstation meets the following dependencies.
    • JDK8
    • Redis
    • Cassandra
    • Packer
  • Run the following scripts:
cd $SPINNAKER_HOME
spinnaker/dev/install_development.sh
spinnaker/dev/bootstrap_dev.sh

Configure Spinnaker:

  • Create a directory for Spinnaker configuration overrides, copy the default configuration template there, and edit to select the appropriate cloud provider(s). Here we will have for AWS.
cd $SPINNAKER_HOME
mkdir -p $HOME/.spinnaker spinnaker/config/default-spinnaker-local.yml
$HOME/.spinnaker/spinnaker-local.yml
chmod 600 $HOME/.spinnaker/spinnaker-local.yml
  • If you are deployed on a supported cloud provider (e.g., AWS or Google) and you only wish to deploy on that platform, then you do not need to perform additional configuration. Otherwise, edit $HOME/.spinnaker/spinnaker-local.yml, enable the cloud provider(s) of your choice, and configure the attributes for that provider. Here we will enable only AWS.

Start the Netflix Spinnaker Service

  • To start the Spinnaker Service, execute following:
cd $SPINNAKER_HOME/build
../spinnaker/dev/run_dev.sh [service]
  • If a service is provided, then just that one service will be started. If no service is provided, then all the services will be started.

Stop the Netflix Spinnaker Service

  • To Stop Spinnaker (except Redis and Cassandra), run following:
cd $SPINNAKER_HOME/build
../spinnaker/dev/stop_dev.sh [service]

Conclusion

This has been just a quick introduction to Netflix Spinnaker. With support for many cloud vendors and even containers like Docker, this looks like a very promising package. We hope to bring you more posts on Spinnaker in coming weeks to demonstrate how it can be installed on AWS and Google Cloud Platform, and how it can manage clusters and workflows.

Avatar

Written by

Chandan Patra

Cloud Computing and Big Data professional with 10 years of experience in pre-sales, architecture, design, build and troubleshooting with best engineering practices. Specialities: Cloud Computing - AWS, DevOps(Chef), Hadoop Ecosystem, Storm & Kafka, ELK Stack, NoSQL, Java, Spring, Hibernate, Web Service


Related Posts

Connie Benton
Connie Benton
— April 1, 2020

How To Build a Career with AWS Certifications

From Iaas and PaaS solutions to digital marketing, cloud computing reshapes the world of technology. As the influence of this technology grows, so does investment. Tens of billions of dollars are being spent on cloud computing-related services each year. This influx is continuing to inc...

Read more
  • AWS
  • Certifications
Vijayakumar Athithan
Vijayakumar Athithan
— March 27, 2020

What is Cognito in AWS?

Web applications usually allow a valid username and password combination for successful sign in to the application. Modern authentication flows incorporate more approaches to ensure user authentication. When using AWS, this is no exception, thanks to the abilities and features offered b...

Read more
  • AWS
  • AWS Cognito
  • Solutions Architect
Avatar
Andrew Larkin
— March 20, 2020

The 12 AWS Certifications: Which is Right for You and Your Team?

As companies increasingly shift workloads to the public cloud, cloud computing has moved from a nice-to-have to a core competency in the enterprise. This shift requires a new set of skills to design, deploy, and manage applications in cloud computing. As the market leader and most ma...

Read more
  • AWS
  • AWS Certifications
Alisha Reyes
Alisha Reyes
— March 17, 2020

Cloud Academy’s Blog Digest: How Do AWS Certifications Increase Your Employability, How to Become a Microsoft Certified Azure Data Engineer, and more

With everything going on right now, it's likely that the only thing you've been reading lately is related to the coronavirus pandemic. It's important to stay informed during these times, but it's also good to jump into something that can take your mind off of the current situation for j...

Read more
  • AWS
  • Azure
  • blog digest
  • Certifications
  • Cloud Academy
  • programming
  • Security
Avatar
Cloud Academy Team
— March 13, 2020

Which Certifications Should I Get?

As we mentioned in an earlier post, the old AWS slogan, “Cloud is the new normal” is indeed a reality today. Really, cloud has been the new normal for a while now and getting credentials has become an increasingly effective way to quickly showcase your abilities to recruiters and compan...

Read more
  • AWS
  • Azure
  • Certifications
  • Cloud Computing
  • Google Cloud Platform
Alisha Reyes
Alisha Reyes
— March 7, 2020

New on Cloud Academy: Intro to GitOps; AWS Courses; Java, Python, Amazon Linux 2, Ubuntu, & Docker Playgrounds; and much more

New Lab Playgrounds This month, our Content Team released six new "playground labs." Our playground labs provide a safe and secure sandbox environment for you to explore your own ideas, follow along with Cloud Academy courses, or answer your own questions — all without having to instal...

Read more
  • AWS
  • Azure
  • gitops
  • Google Cloud Platform
  • lab playground
  • programming
Alisha Reyes
Alisha Reyes
— March 6, 2020

New on Cloud Academy: Intro to GitOps; AWS Courses; Java, Python, Amazon Linux 2, Ubuntu, & Docker Playgrounds; and much more

New Lab Playgrounds This month, our Content Team released six new "playground labs." Our playground labs provide a safe and secure sandbox environment for you to explore your own ideas, follow along with Cloud Academy courses, or answer your own questions — all without having to instal...

Read more
  • AWS
  • Azure
  • gitops
  • Google Cloud Platform
  • lab playground
  • programming
Patrick Navarro
Patrick Navarro
— March 4, 2020

AWS Certifications: How Do They Increase Your Employability and Progress Your Career?

AWS certifications are no walk in the park. They’re designed to validate in-depth, specialist knowledge and comprehensive experience, often requiring months of dedicated studying to earn even for those already working with the cloud platform. But the rewards that AWS professionals ca...

Read more
  • AWS
  • AWS certification
  • certification
Avatar
Chandan Patra
— February 21, 2020

Elasticsearch vs. CloudSearch: AWS Cloud Search Choices

Elasticsearch vs. CloudSearch: What's the main difference? Let's compare AWS-based cloud tools: Elasticsearch vs. CloudSearch. While both services use proven technologies, Elasticsearch is more popular, open source, and has a flexible API to use for customization; in comparison, CloudS...

Read more
  • AWS
  • Azure
  • cloudsearch
  • elasticsearch
Avatar
Andrew Larkin
— February 13, 2020

Cloud Academy Content Roadmap Updates

Welcome to our Q1 2020 roadmap. This is the content we plan to build over the next three months, between February 1 - and April 30, 2020. Let's look at some of our roadmap highlights. Atlassian Bamboo for CI/CD We had a lot of requests for practical guides on how to apply DevOps tool...

Read more
  • Artificial Intelligence
  • AWS
  • Azure
  • Docker
  • Google Cloud Platform
  • Kubernetes
  • Machine Learning
Alisha Reyes
Alisha Reyes
— February 7, 2020

New on Cloud Academy: Git Labs, CKA and CKAD Lab Challenges, AWS and Azure Learning Paths, AGILE, and Much More

We just kicked off our first Free Weekend of 2020. This means we've unlocked our Training Library for just 72 hours. Until Sunday at 11:59 pm (PST), you can get unlimited access to our industry-leading learning paths, courses, certification prep exams, and our most popular hands-on labs...

Read more
  • agile
  • AWS
  • Azure
  • Google Cloud Platform
  • Linux
  • OWASP
  • programming
  • red hat
  • scrum
Avatar
Stuart Scott
— February 6, 2020

How to Encrypt an EBS Volume

Keeping data and applications safe in the cloud is one of the most visible challenges facing cloud teams in 2020. Cloud storage services where data resides are frequently a target for hackers, not because the services are inherently weak but because they are often improperly configured....

Read more
  • AWS
  • EBS
  • Encryption