Iterative Request Router Sequence

Demonstrates sequential sub-request execution where each step completes before the next begins

Category: Practical
Task: Demonstrates sequential sub-request execution where each step completes before the next begins

Prerequisites: The product runtime and any backend services referenced by this configuration.

This configuration comes from the selected release. The example has not been run here; external services are not bundled.

Download the source file.

# Iterative Request Router: Sequential Steps
#
# Demonstrates sequential sub-request execution
# where each step completes before the next begins.
# Step A's response feeds into step B, which
# produces the final response to the client.
#
# ⚠️  EXPERIMENTAL: This example uses the `iterative-request-router` experimental feature.
# Not recommended for production use. Build with --features iterative-request-router.
#
# Usage:
#   cargo run -p praxis-proxy --features iterative-request-router -- -c examples/configs/pipeline/iterative-request-router-sequence.yaml
#   curl http://localhost:8080/

listeners:
  - name: gateway
    address: "127.0.0.1:8080"
    filter_chains:
      - sequence

filter_chains:
  - name: sequence
    filters:
      - filter: iterative_request_router
        initial_step: step-a
        steps:
          - name: step-a
            filters:
              - filter: router
                routes:
                  - path_prefix: "/"
                    cluster: service-a
              - filter: load_balancer
                clusters:
                  - name: service-a
                    endpoints:
                      - "127.0.0.1:3000"
            on_result:
              - default: true
                next: step-b
          - name: step-b
            filters:
              - filter: router
                routes:
                  - path_prefix: "/"
                    cluster: service-b
              - filter: load_balancer
                clusters:
                  - name: service-b
                    endpoints:
                      - "127.0.0.1:3001"
            on_result:
              - default: true
                done: true

insecure_options:
  allow_private_endpoints: true # example proxies to local backends