configuration.md

Configuration

Overview

Configuration for each service is done primarily through environment variables:

PORT: 3000
DATABASE_NAME: postgres
DATABASE_USER: postgres
DATABASE_PASSWORD: postgres
DATABASE_HOST: postgres
DATABASE_PORT: 5432

Reading the configuration

The service should define a configuration type to hold all the configuration values needed. Those values can then be loaded when the service boots

export interface Config {
    port: number;
    knex: KnexConfig;
};

const appConfig: Config = {
    port: Number(process.env.PORT || 3000),
    knex: {
        client: 'pg',
        connection: {
            host: process.env.DATABASE_HOST,
            database: process.env.DATABASE_NAME,
            password: process.env.DATABASE_PASSWORD,
            user: process.env.DATABASE_USER,
            port: Number(process.env.DATABASE_PORT),
        },
    },
};


export default appConfig;

In some cases, a sensible value can be set if not provided by the environment

Docker Compose

When running with docker compose, use the environments section of the service to set the values

services:
  submission:
    environment:
      PORT: 3000
      DATABASE_NAME: postgres
      DATABASE_USER: postgres
      DATABASE_PASSWORD: postgres
      DATABASE_HOST: postgres
      DATABASE_PORT: 5432
...

Kubernetes

In a Deployment set the environment variables for each container

apiVersion: apps/v1beta2
kind: Deployment
metadata:
  name: {{ include "libero-reviewer.fullname" . }}-submission
  ...
  spec:
    ...
    template:
      spec:
        containers:
          ...
          env:
            - name: DATABASE_NAME
              value: "{{ include "libero-reviewer.fullname" . }}-postgresql"
            - name: DATABASE_PORT
              value: "5432"
            - name: DATABASE_USER
              value: "{{ .Values.postgresql.postgresqlUsername }}"
            - name: DATABASE_PASSWORD
              value: "{{ .Values.postgresql.postgresqlPassword }}"
            - name: DATABASE_NAME
              value: "{{ .Values.postgresql.postgresqlDatabase }}"

Service settings

Submission

The submission service uses the following configuration values

DATABASE_NAME: postgres database name
DATABASE_USER: postgres username
DATABASE_PASSWORD: postgres password
DATABASE_HOST: postgres hostname
DATABASE_PORT: postgres access port

S3_AWS_ENDPOINT: S3 service endpoint, only used for testing and dev
S3_ACCESS_KEY_ID: AWS access key id for S3
S3_SECRET_ACCESS_KEY: Aws secrety key for S3
S3_FORCE_PATH_STYLE: Always set to 'true'
S3_FILE_BUCKET: Name of the S3 Bucket for file uploads

NEW_RELIC_HOME: local path to newrelic.js config file

MAX_QL_COMPLEXITY: maximum GraphQL complexity (recommended: 100)
MAX_QL_DEPTH: maximum GraphQL query depth (recommended: 5)
MAX_FILE_SIZE_IN_BYTES: maximum file upload

AUTHENTICATION_JWT_SECRET: Libero reviewer authentication secret
USER_ADAPTER_URL: Continuum Adaptor endpoint for getting current user info (http://continuum-adaptor.url/current-user\).
SCIENCE_BEAM_URL: ScienceBeam endpoint for converting uploaded manuscript, for examplehttp://continuum-api.url/science-beam/convert. (from continuum)
SCIENCE_BEAM_TIMEOUT: Conversion timeout value (recommended: 20000)

Continuum Adaptor

PORT: The listening port for the service
RABBITMQ_URL: The url for rabbitmq.
LOGIN_URL: The url for continuum (e.g. https://elifesciences.org/submit\)
LOGIN_RETURN_URL: The login url to return to after token exchange (e.g. https://reviewer.elifesciences.org/login\)

DATABASE_NAME: postgres database name
DATABASE_USER: postgres username
DATABASE_PASSWORD: postgres password
DATABASE_HOST: postgres hostname
DATABASE_PORT: postgres access port

AUTHENTICATION_JWT_SECRET: Libero reviewer authentication secret
CONTINUUM_JWT_SECRET: The secret for continuum journal (from continuum)
CONTINUUM_API_URL: Continuum Api endpoint (from continuum)
ELIFE_API_GATEWAY_SECRET: Elife Api Gateway secret (from continuum)