myna

Concurrent actors for the turaco API, written using Ruby, powered by Kafka.

Getting Started

We recommend you set up the project on your local machine, using docker-compose for services like ES, Zookeeper, and Kafka. However, for deployment, we recommend using our Dockerfile to ensure a correctly set up environment: our images will always contain the correct native extensions that they use through FFI.

Basic Dependencies

• MRI 2.4.0
• PocketSphinx
  • apt, brew, and the Arch Linux aur has this as a package
• Install elasticsearch, docker-compose, docker, and redis-server using your distribution's package manager.

Kafka

We use Apache Kafka as our message bus. Advanced users can skip this and manually set up zookeeper and kafka to their liking -- we currently do not do anything exotic. Everybody else is highly encouraged to use this docker-compose.yml definition for their Kafka setup.

version: '2'
services:
  zookeeper:
    restart: always
    image: wurstmeister/zookeeper
    ports:
      - "2181:2181"

  kafka:
    restart: always
    image: wurstmeister/kafka
    links:
      - zookeeper:zk
    ports:
      - "9092:9092"
    depends_on:
      - zookeeper
    environment:
      KAFKA_ADVERTISED_HOST_NAME: (docker container's IP address)
      KAFKA_ADVERTISED_PORT: (9092, but whatever you want if you NAT it)
      KAFKA_ZOOKEEPER_CONNECT: zk:2181
      KAFKA_CREATE_TOPICS: "sample_speech_recognition:1:1:compact,es_manage:1:1:compact"
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock

Running everything

• Use systemctl or whatever you use for services to start elasticsearch and redis-server
• In the same directory as your docker-compose.yml file, docker-compose up -d will run your Kafka/Zookeeper containers in the background.
  • To make sure they're running, please docker ps as root.
• bundle exec karafka w to start Sidekiq workers (required).
• bundle exec karafka s to start the Karafka server.

Deploying

To separate concerns in production, we do not recommend you package Kafka, Zookeeper, and Elasticsearch in one container (nor are these intended to be used this way). To build our container, just docker build -t talkbirdy-myna-image . in the same directory as the Dockerfile. Our CI configuration below illustrates what your CD steps look like for probably every PaaS with basic containerization support.

deployment:
  production:
    branch: dockerize-me-captain
    commands:
      - heroku plugins:install heroku-container-registry
      - docker login --username=_ --password=$NOT_DOCKER_HEROKU_API_KEY registry.heroku.com
      - heroku container:push worker -a talkbirdy-myna

OK, fine, that environment variable is totally an API key.

Data flow

We want our API to be fast and do as little work as possible outside of presentation and basic CRUD operations, we spawn actors to do the rest of the work. The best way to add a new feature (read: do non-API related work) would be to modify the PostgresSink in Turaco to dispatch the correct payloads to a specialized topic.

Topics

Topic	Purpose
es_manage	Notified whenever there is a database change to update Elasticsearch with latest data.
sample_speech_recognition	Notified when a sample row is inserted, passes audio through CMU Sphinx for inference processing.

Credits & License

• David Stancu
• Paulina Levit

Copyright (c) 2017 Awesome IT LLC.

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.