Cron Scheduler

Summary

This repository contains the microservice responsible for the handling of the Mia-Platform cronjob.

Local Development

To develop the service locally you need:

• Node 8+

To setup node, please if possible try to use nvm, so you can manage multiple versions easily. Once you have installed nvm, you can go inside the directory of the project and simply run nvm install, the .nvmrc file will install and select the correct version if you don’t already have it.

Once you have all the dependency in place, you can launch:

npm i
npm run coverage

This two commands, will install the dependencies and run the tests with the coverage report that you can view as an HTML page in coverage/lcov-report/index.html.
After running the coverage you can create your local copy of the default values for the env variables needed for launching the application.

cp ./default.env ./local.env

From now on, if you want to change anyone of the default values for the variables you can do it inside the local.env file without pushing it to the remote repository.

Once you have all your dependency in place you can launch:

set -a && source ./local.env
npm start

After that you will have the service exposed on your machine.

Configuring the Cron Scheduler

The Cron Scheduler needs a configuration file specifying the scripts to run and when to run them. Currently, only Node scripts are supported.
The configuration file must be placed in the root folder of the service having and specify its name via the env variable CONFIG_PATH. The config-example.json file of this repository is an example showing how it should be structured.

[
  {
    "name": "YOUR_DESIRED_SCRIPT_NAME",
    "filePath": "FULL_PATH_TO_THE_SCRIPT",
    "crontab": "CRON_TAB_STRING (Ex: * * * * * 30)",
    "timezone": "THE_DESIRED_TIMEZONE",
    "env": {
      "ENV_VARIABLE_NAME_1": "ENV_VARIABLE_VALUE_1",
      "ENV_VARIABLE_NAME_2": "ENV_VARIABLE_VALUE_2"
    }
  },
  {
  }
]

Configuring cron scheduler for another client

You can use cron scheduler as is, or integrating your scripts. In the latter case, you should create your own dockerfile inheriting from the docker image at nexus.mia-platform.eu/cron-scheduler. Through the dockerfile, copy the scripts you want to run in the created container, and create your own configuration file that you should also copy in the docker image.

Scripts available

Cron scheduler offers some script ready to be used simply setting them in the configuration. These services are:

• send-curl

Send curl

This is a script that try to send POST requests reading continuously from mongo the information to send. The script connects to the MONGO_COLLECTION specified in the env and reads rows so composed:

• url: a string representing to url to send the request to
• body: a stringified object containg the body to attach to the request (json body is only supported now)
• headers: a stringified object containing the headers to attach to the request
• state: the current processing state of the request to send among:
  • PENDING: the request need to be processed
  • SENDING: the request is being processed
  • SENT: the request was processed succesfully
• attempt: the number of attempts the script did for this request

For each read row in a PENDING state, the script attempts to send a POST request with the read information. If the POST is sent succesfully and the status code of the response is 204 or 200 the request is not sent anymore. Otherwise a new attempt is done after the period set in the cron settings. A configuration example is the following:

{
  "name": "Send Curl",
  "filePath": "/cron-scheduler/scripts/send-curl/index.js",
  "crontab": "*/5 * * * * *",
  "timezone": "Europe/Rome",
  "env": {
    "MONGO_URL": "mongodb://localhost:27017/curl-db-name",
    "MONGO_COLLECTION": "curl-collection-name"
  }
}

Where:

• MONGO_URL is the url to reach mongo with the db containing the MONGO_COLLECTION collection where curls are stored
• MONGO_COLLECTION - Is the name of the collection from where the curl are retrieved

Contributing

To contribute to the project, please be mindful for this simple rules:

1. Don’t commit directly on master
2. Start your branches with feature/ or fix/ based on the content of the branch
3. If possible, refer to the Jira issue id, inside the name of the branch, but not call it only fix/BAAST3000
4. Always commit in english
5. Once you are happy with your branch, open a Merge Request

Run the Docker Image

If you are interested in the docker image you can get one and run it locally with this commands:

docker pull nexus.mia-platform.eu/core/cron-scheduler:latest
docker run --name cron-scheduler \
           --detach \
           --mount type=bind,source=$(pwd)/tests/config.json,target=/home/node/app/config.json
           --env LOG_LEVEL=info \
           --env CONFIG_PATH=./config.json \
           nexus.mia-platform.eu/core/cron-scheduler