connectors

A flexible data integration tool to help nonprofits connect to their data collection tools and ERP systems

Available Connectors (Airflow DAGs)

1. SurveyCTO
2. ComCare
3. ONA

With all the above connectors, the pipeline does the following operations:

• Fetch data from the servers
• Dump the data into Postgres or MongoDB
• Sync server and database everytime the pipeline runs (remove stale data)

Obtaining the DAGs

After installing Airflow, or setting it up on your environment,

clone the DAGs into your dags folder by using these commands:

• For HTTPs: https://github.com/hikaya-io/connectors.git dags
• For SSH: [email protected]:hikaya-io/connectors.git dags

Enable the DAGs

Navigate to your DAGs page on theAirflow UI and refresh to see the new cloned DAGs from connectors repo

Turn on whichever DAG you want to run.

Configurations

Our connectors uses configurations that are provided by the user on the Airflow UI interface.

Setting the Variables Manually

To add these Variables, naviagte to the Admin -> Variables - ADD/+

Depending on which DAG you've activated, add the following Variables:

Postgress Settings

POSTGRES_DB_PASSWORD = ''

POSTGRES_DB = ''

POSTGRES_HOST = ''

POSTGRES_USER = ''

POSTGRES_PORT = ''

MongoDB Settings

MONGO_DB_USER = ''

MONGO_DB_PASSWORD = ''

MONGO_DB_HOST = ''

MONGO_DB_PORT = ''

SurveyCTO Variables

SURV_SERVER_NAME = '' :- SurveyCTO server name

SURV_USERNAME = '' :- SurveyCTO login username (email address)

SURV_PASSWORD = '' :- SurveyCTO login password

SURV_DBMS = '' :- Database management system to dump the data (mongo/mongodb, postgres/postgresdb)

Upload the Variables from a json file

Download the variables_template.json on this repo and update it with the correct values.

Navigate to the Admin -> Variables

Click on Import Variables after selecting the valid JSON file.

For this option to work, the file must be a valid flat JSON file as the one in the template.

Trigger the DAG

To manually trigger the configured DAG:

• Click on the DAG to open the DAG details
• Click on Trigger DAG
• Confirm triggering by clicking ok on the prompt

If all the settings are correct, then your DAG should run successfully.

Docker Airflow Setup (Optional)

Even though it is posible to install Apache Airflow pip, we have to perform more configurations to ensure it is production ready. Deploying Airflow via Docker is definititely faster especially using the image puckel/docker-airflow; for more information check out docker-airflow Github repo.

Requirements

• docker and docker-compose installed. (Create a Docker Ubuntu DO Droplet)
• create directory structure /home/hikaya/dags

Setup

• Copy all files in the DAGs directory into the dags directory created above

• Navigate to the Docker directory and run the command below to deploy Airflow:

  docker-compose -f docker-compose-CeleryExecutor.yml up -d

• To create an admin user:

  • execute an interactive bash shell on the Airflow webserver container:

  docker exec -it docker-airflow_webserver_1 bash

  • use airflow create_user CLI to create an admin user

  airflow create_user --role Admin --username admin --email admin --firstname admin --lastname admin --password admin

• Login using the user created above and load the variables