Skip to content

INSTALL.md

Latest commit

 

History

History
44 lines (30 loc) · 3.84 KB

INSTALL.md

File metadata and controls

44 lines (30 loc) · 3.84 KB

SPEKE Reference Server Installation

The following page guides the user through deployment and configuration of the SPEKE Reference Server.

Prerequisites

  • An AWS account with administrator rights and access to the AWS console
  • Note that this solution can be deployed to any region that supports API Gateway, Lambda, and S3. You need to consider the packager or encoder's location relative to the API Gateway endpoint used to create encryption keys. The encoder, packager and SPEKE services should be in the same region or as geographically close as possible to reduce the request/response latency in key generation.

Note

For ease of deployment, please use the pre-built template and Lambda binaries hosted by the project sponsors. The CloudFormation template located in the src/ folder cannot be deployed to a stack without modification. Review the build_and_host.sh to understand how to prepare the template for deployment with dependencies.

Deploy changes to CloudFormation template and lambda function

  1. Create a virtual environment for this project using python3.
  2. Install dependencies within the virtual environment using pip3 install -r requirements.txt.
  3. In zappa_settings.json under src, replace aws_region with the region this lambda will be deployed.
  4. Create a build folder if it doesn't exist and then run local_build.sh.
  5. The script will generate required artifacts under build folder.
  6. Create a new bucket in S3 (For example: speke-us-east-1). Create a folder called speke and upload the generated speke-reference lambda zip file.
  7. In the generated speke_reference.json, replace rodeolabz with the name of your created bucket (speke is used in this example).
  8. Use the speke_reference.json template in CloudFormation to deploy the speke reference server following the instructions give below.

Deploying from Mac/Windows

On platforms other than AmazonLinux, one of the required dependencies: cffi is generated differently and so running the reference server might result in an error: No module named '_cffi_backend'. To resolve this, create a requirements.txt file with cffi==<version>, replacing version with the desired version number and follow the steps outlined here to create a layer and update the speke reference lambda function.

Deploy using prepared CloudFormation template

The hosted template is supported in regions: ap-northeast-1 ap-northeast-2 ap-south-1 ap-southeast-1 ap-southeast-2 eu-central-1 eu-west-1 eu-west-3 sa-east-1 us-east-1 us-west-1 us-west-2

  1. Sign in to the AWS console.
  2. Choose a region such as us-east-1 or us-west-2 to start.
  3. Navigate to the AWS CloudFormation console.
  4. Create a new stack.
  5. On the Select Template page, provide the hosted copy of the CloudFormation template. The prepared CloudFormation template is hosted here by the project sponsors: https://s3.amazonaws.com/rodeolabz-us-east-1/speke/speke_reference.json and and can be launched in several AWS regions.
  6. At the Specify Details pages, provide a stack name, like SPEKE.
  7. Provide a value for the KeyRetentionDays parameter. This is the amount of time to retain a key in the S3 bucket for client playback. Keys older than this amount will be automatically removed by S3. The default is 2 days, which is usually enough for live content across multiple time zones.
  8. The Options page does not require any input, although you can choose to be notified after the template completes.

When the template is complete you will have an operational reference SPEKE server that can be used for HLS encryption. You can review the Resources tab of the template to see what was created or updated, and the Outputs tab for the URL of the SPEKE server and the role ARN that permits MediaPackage access.

Home