SALIDO DevOps Challenge

This challenge is provided to help assess a candidate's development skills. There is no time limit, but we ask that you be prompt in submitting your solution.

Submission Instructions

1. Complete the challenge according to the Challenge Instructions below.
2. Deploy your code on the internets so we can see it in action.
3. Post your solution in a Github repo. Be sure to include a README to explain and document your solution.
4. Send a link to your solution to [email protected].

Resources

• AWS provides free and low-cost resources for anything you may need to complete this challenge. (https://aws.amazon.com)
• mLab provides free hosted instances of MongoDB. (https://mlab.com)
• Heroku provides free containers for hosting apps and services. (https://www.heroku.com)
• You are encouraged to make use of available DevOps tools as you see fit for configuration, automation, orchestration, containerization, etc.

Challenge Instructions

Objective

Create a "one click" process for cloning production data into other environments.

Intro

SALIDO has three environments to which it deploys code: Staging, Sandbox and Production. We need to clone data from our production MongoDB into other environments so we can test and play with data that is consistent with production.

Each customer is stored in our database as an "organization". Most documents in the database are associated with an organization via direct or indirect relationships. For example, an organization has many brands; a brand has many locations; a location has many checks; a check has many payments. Use these relationships to find all of the documents that ultimately belong to an organization.

When cloning data from production, the current environment's data is purged (except for certain things as described in the business requirements), and the production data is pulled in to replace it. We need the ability to specifically select which organizations should be cloned from production, and which organizations (if any) should be retained in the current environment.

Business Requirements

• Data that cannot be associated (directly or indirectly) with an organization should never be purged or cloned.
• The user must be able to select which organizations to retain in the current environment. Data associated with all other organizations will be purged.
• The user must be able to select which organizations to clone from production data. Only the data associated with the selected organizations will be cloned.
• Cloned production data must be sanitized as follows:
  • All email addresses must be replaced with mock email addresses: fake.email+{UUID}@salido.com (where {UUID} is a UUID with no dashes).
  • Credit card gateway configurations must all be sandboxed. This simply means setting sandbox: true for all documents in the cc_gateways collection.
• The following collections must be explicitly excluded from the cloned production data:
  • banks
  • cc_gateway_transactions
  • checks
  • shifts

Requirements for Mock Email Addresses

Email addresses are found in two places:

• In the employees collection in the personal_email property.
• In the emails collection in the address property.

If a value at employees.personal_email matches a value at emails.address in the production data, then the mock email addresses must also match.

To Do

1. Use the MongoDB dumps provided to populate databases for each environment: Staging, Sandbox and Production.

2. Review the database collections to familiarize yourself with how things are generally structured, and how documents are associated with each other.

3. Create a tool that performs the cloning process according to the business requirements. This tool should allow the user to:

   • Select the environment into which Production data will be cloned (either Staging or Sandbox).
   • Select organizations to clone from production.
   • Select organizations to retain in the destination environment.

   This tool can take any reasonable form. Some ideas:

   • A command line program that takes various config arguments.
   • A command line program that takes a config file.
   • A command line program that prompts each step.
   • A web app.

4. Display some sort of progress indicator while the cloning process is running. Some ideas:

   • Progress bar.
   • Percentage complete.
   • A "live log" that displays console output as it's happening.

5. When cloning is complete, produce a log that includes pertinent information about the cloning process, and what data was cloned, retained, purged, etc.

6. Display appropriate and informative success and failure messages as needed.