Skip to content

Latest commit

 

History

History
 
 

go

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
embedded
embedded
 
 
infra/docker
infra/docker
 
 
internal
internal
 
 
types
types
 
 
README.md
README.md
 
 
__init__.py
__init__.py
 
 
main.go
main.go
 
 
main_test.go
main_test.go
 
 

README.md

[Update 10/31/2024] This Go feature server code is updated from the Expedia Group's forked Feast branch (https://github.com/ExpediaGroup/feast.git) on 10/22/2024. Thanks the engineers of the Expedia Groups who contributed and improved the Go Feature Server.

Build and Run

To build and run the Go Feature Server locally, create a feature_store.yaml file with necessary configurations and run below commands:

    go build -o feast-go ./go/main.go
    # start the http server (metrics on port 9090 by default)
    ./feast-go --type=http --port=8080 --metrics-port=9090
    # or start the gRPC server
    #./feast-go --type=grpc  --port=[your-choice] --metrics-port=9091

Prometheus Metrics

The server exposes Prometheus metrics at the /metrics endpoint on a dedicated port (default :9090).

  • HTTP Mode: Metrics server runs on port 9090 (configurable via -metrics-port).
  • gRPC Mode: Metrics server runs on port 9090 (configurable via -metrics-port).

Key metrics include:

  • http_request_duration_seconds: Histogram of response latency.
  • http_requests_total: Counter of HTTP requests by status, method, and path.
  • Standard Go and Process metrics.

A /health endpoint is available on the main application port (default :8080) for readiness probes.

OTEL based observability

The OS level env variable ENABLE_OTEL_TRACING=="true"/"false" (string type) is used to enable/disable this service (with Tracing only). You can also configure the service name using OTEL_SERVICE_NAME env variable (defaults to "FeastGoFeatureServer").

The default exporter URL is "http://localhost:4318". The default schema of sending data to collector is HTTP. Please refer the following two docs about the configuration of the OTEL exporter:

  1. https://opentelemetry.io/docs/languages/sdk-configuration/otlp-exporter/
  2. https://pkg.go.dev/go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp#WithEndpoint

List of files have OTEL observability code

  1. internal/feast/transformation/transformation.go
  2. internal/feast/onlinestore/redisonlinestore.go
  3. internal/feast/server/grpc_server.go
  4. internal/feast/server/http_server.go
  5. internal/feast/server/server_commons.go
  6. internal/feast/featurestore.go

Example monitoring infra setup

  1. docker compose file to setup Prometheus, Jaeger, and OTEL-collector.
services:
  prometheus:
    image: prom/prometheus
    volumes:
      - ./prometheus.yaml:/etc/prometheus/prometheus.yaml
    ports:
      - 9090:9090   # web UI  http://localhost:9090
  jaeger:
    image: jaegertracing/all-in-one:latest
    ports:
      - 16686:16686  # Web UI: http://localhost:16686
      - 14268:14268  # http based receiver
      - 14250:14250  # gRPC based receiver
  otel-collector:
    image: otel/opentelemetry-collector-contrib
    volumes:
      - ./otel-collector-config.yaml:/etc/otelcol-contrib/config.yaml
    ports:
      #- 1888:1888 # pprof extension
      - 8888:8888 # Prometheus metrics exposed by the Collector
      - 8889:8889 # Prometheus exporter metrics
      #- 13133:13133 # health_check extension
      #- 4317:4317 # OTLP gRPC receiver
      - 4318:4318 # OTLP http receiver
      - 55679:55679 # zpages extension. check http://localhost:55679/debug/tracez
    depends_on:
      - jaeger
      - prometheus
  1. OTEL collector configure file.
receivers:
  otlp:
    protocols:
      http:
        endpoint: 0.0.0.0:4318

exporters:
  prometheus:
    endpoint: 0.0.0.0:8889
    namespace: feast-go
  otlp/jaeger:  # this is a gRPC based exporter. use "otelhttp" for http based exporter.
    endpoint: jaeger:4317
    tls:
      insecure: true

extensions:
  zpages:
    endpoint: 0.0.0.0:55679

processors:
  batch:

service:
  extensions: [zpages]
  pipelines:
    metrics:
      receivers: [otlp]
      exporters: [prometheus]
    traces:
      receivers: [otlp]
      exporters: [otlp/jaeger]
  1. Prometheus config.
 #https://github.com/prometheus/prometheus/blob/release-3.6/config/testdata/conf.good.yml
 scrape_configs:
  - job_name: 'otel-collector'
    scrape_interval: 1m
    scrape_timeout: 30s # Increase this if needed
    static_configs:
      # Check the IP address of or Docker host network. 
      # Refer: https://stackoverflow.com/questions/48546124/what-is-the-linux-equivalent-of-host-docker-internal
      - targets: ['172.17.0.1:8888'] # Replace with the Collector's IP and port
  - job_name: 'otel-collected'
    scrape_interval: 1m
    scrape_timeout: 30s # Increase this if needed
    static_configs:
      - targets: ['172.17.0.1:8889'] # Replace with the Collector's IP and port
  1. Jaeger config file is not used in this setup.