Shared API Gateway

When to use

When your API Gateway is shared across multiple Serverless services, use sharedApiGateway: true in the secondary services. This prevents them from overwriting the cache cluster settings configured by the primary service.

Setup

Primary service (owns the API Gateway)

# Creates the cache cluster
custom:
  interlaceCaching:
    enabled: true
    clusterSize: '1.6'
    ttlInSeconds: 300

Secondary services (share the API Gateway)

custom:
  interlaceCaching:
    enabled: true
    sharedApiGateway: true
    restApiId: ${cf:api-gateway-${self:provider.stage}.RestApiId}
    basePath: /animals

functions:
  getCats:
    handler: getCats.handle
    events:
      - http:
          path: /cats
          method: get
          caching:
            enabled: true

How it works

When sharedApiGateway is true, the plugin:

1. Skips stage-level operations (cache cluster enable/disable, cluster size)
2. Still applies per-endpoint settings (TTL, encryption, invalidation, cache keys)
3. Uses the provided restApiId to find the correct API Gateway

Cross-stack references

If the API Gateway is in a different CloudFormation stack, export its RestApiId and reference it:

# In the primary stack
resources:
  Outputs:
    RestApiId:
      Value:
        Ref: ApiGatewayRestApi
      Export:
        Name: api-gateway-${self:provider.stage}-RestApiId

# In the secondary stack
custom:
  interlaceCaching:
    restApiId: ${cf:api-gateway-${self:provider.stage}.RestApiId}

If the API Gateway is in the same stack, you don't need to specify restApiId — the plugin finds it automatically.

Base path

When the shared gateway has a default base path that is not part of your endpoint configuration:

custom:
  interlaceCaching:
    sharedApiGateway: true
    basePath: /animals  # prepended to all endpoint paths

Your endpoint /cats becomes /animals/cats when looking up the API Gateway resource.