Introduction to Snyk Code Local Engine
Snyk Code Local Engine is a fully contained version of the Snyk Code Engine, allowing users to avoid uploading their code to the internet.
This high-level architecture diagram shows the different components and their interactions.
Snyk Code Local Engine Architecture
When you use the Local Engine, only the scan is performed locally. Your scan results are uploaded to Snyk, so you can view them in the Snyk Web UI.
The core requirements to deploy the Snyk Code Local Engine are:
- Kubernetes – version 1.19.0 - 1.23.5:
- Recommended: a dedicated Kubernetes cluster
- Outbound HTTPS connections supporting websockets from the cluster to *.snyk.io
- Kubernetes – one of the following:
- Managed public cloud Kubernetes service - EKS, AKS, GKE - or -
- Unmanaged Kubernetes (a self-hosted cluster)
- PR Checks and Snyk CLI support requires network access:
- From your Kubernetes cluster to your Git and CI/CD tooling
- From users running Snyk CLI to the Kubernetes cluster
- Helm – version 3.8.0 or newer
- 3 Nodes – each one with the following:
- Disk: 500 GB (>300GB Ephemeral Storage)
- CPU and RAM:
- [For AWS] instance type m5.8xlarge
- [For GCP] instance type e2-standard-32
- 32 Core CPU with 120GB RAM
Last modified 2mo ago