Skip to content

Latest commit

 

History

History
16 lines (14 loc) · 1.37 KB

File metadata and controls

16 lines (14 loc) · 1.37 KB
implementing jenkins pipeline in a microservice architecture

NOTE:

microservice components:
  • Loader(Python) --> responsible for reading a JSON file containing a list of movies and pushing each movie item to Amazon SQS
  • Parser(Golang) --> responsible for consuming movies by subscribing to SQS and scraping movies information from the IMDb website and storing the metadata into MongoDB
  • Store(Node.js) --> responsible for servig a RESTful API with endpoints to fetch a list of movies and insert new movies into the watch list database in the MongoDB
  • Marketplace(Angular & TypeScript) --> responsible for serving a frontend to browse movies by calling the store RESTful API
key approaches:
  • using an aws api gateway to invoke a lambda function (written in js) and trigger github webhook
  • statis code analysis for Angular and the marketplace with SonarQube and perform webhook to jenkins for completion or failure
  • CVE (Common Vulnerability and Exposure) scanning of source code using Nancy
  • baking Nexux Repository OSS machine image with Packer to host our private docker registery.
  • using Anchore Engine for continuous visibility into supply chain security risks