View on GitHub

NYU DevOps Troubleshooting Guide

A guide to troubleshooting common issues in the NYU DevOps class

Behaviour Driven Development Issues

Problem : Selenium WebDriver Not Found

Selenium tests in Docker fail with error: “WebDriverException: Message: ‘chromedriver’ executable needs to be in PATH.”

Solution

Use the rofrano/pipeline-selenium Docker image, which includes Chrome and chromedriver, in .devcontainer/Dockerfile.

Detailed Explanation:

This issue occurs when the Selenium WebDriver, specifically ‘chromedriver’, is not found in the Docker container’s PATH. To resolve this:

1.Switch Docker Image: Update the Dockerfile to use rofrano/pipeline-selenium. This image is pre-configured with Chrome and chromedriver.

  1. Rebuild Docker Container: After updating the Dockerfile, rebuild the container to ensure the new configuration is applied.
  2. Verify Installation: Check if ‘chromedriver’ is correctly installed and accessible by running a test command inside the container.

Problem: Behave Command Not Found

When running behave in a Docker container, an error indicates that the command is not found.

Solution

Ensure behave is listed in the requirements.txt file and the Docker container is rebuilt. Add the below

# Behavior Driven Development
behave==1.2.6
selenium==4.10.0
compare==0.2b0
requests==2.31.0
Detailed Explanation:

The error indicating that the ‘behave’ command is not found suggests it is not installed in the Docker container. To fix this:

1.Check requirements.txt: Ensure behave is listed in the requirements.txt file.

  1. Rebuild Container: Rebuild the Docker container to install behave from the updated requirements.txt.
  2. Test Behave Installation: Run a simple behave command to confirm it’s now recognized in the container.

Problem : Chromedriver Unstable or Failing

Chromedriver is inconsistent or fails in Docker environments when running Selenium tests.

Solution

Initially, test with export DRIVER=firefox. If this resolves the issue, make it permanent by setting DRIVER: firefox and WAIT_SECONDS: 3 in .devcontainer/docker-compose.yml.

Detailed Explanation:

To address Chromedriver’s inconsistent behavior across different systems, follow these steps:

1.Temporary WebDriver Switch:

  • Run export DRIVER=firefox in the Docker environment to temporarily switch to Firefox WebDriver.
  • This change applies only to the current session and helps determine if Firefox WebDriver resolves the issue.

2.Test the Change:

  • Conduct your Selenium tests again to check if the issue with Chromedriver is resolved using Firefox WebDriver.

3.Permanent Configuration:

  • If the issue is resolved with Firefox, then make this change permanent.
  • In your project’s .devcontainer/docker-compose.yml, add the following under the environment section: environment: DRIVER: firefox WAIT_SECONDS: 3
  • DRIVER: firefox sets Firefox as the default WebDriver.
  • WAIT_SECONDS: 3 reduces the wait time in case of errors, speeding up test execution.

4.Rebuild Docker Environment:

  • After updating the docker-compose.yml file, rebuild the Docker environment to apply these changes.
  • This ensures that all future tests will automatically use Firefox WebDriver.

5.Verify Stability:

  • Run the tests again in the updated Docker environment to confirm the stability and consistency of Selenium tests with Firefox WebDriver.

Problem : behave failes due to assertion error even with 200

Behavior-driven development tests with behave are failing due to an assertion error despite receiving a 200 HTTP status code.

Solution

Ensure the database is correctly set up by running flask db-create and then execute honcho start to run the application with the correct environment.

Detailed Explanation

Context

A user is encountering assertion errors when running behave tests, even though the application seems to be running correctly as indicated by the 200 HTTP status code received from a service.

Possible Causes

  1. Database Issues: As mentioned by the user, the error was related to the database (DB) table not being set up correctly.
  2. Environment Setup: The suggestion to run honcho start implies that the application’s environment variables might not have been initialized properly.

Steps to Resolve

  1. Database Initialization:

    • Run flask db-create to set up the database tables correctly.
    • Ensure that the database is properly migrated and seeded with any required initial data.

  2. Application Environment:

    • Use honcho start to run the application. Honcho reads from a .env file to set environment variables, ensuring that the application runs in the correct context.

Conclusion

The user’s problem was resolved by setting up the database tables correctly and ensuring that the application’s environment was correctly initialized using Honcho. This highlights the importance of having both the database and environment correctly configured for behavior-driven development tests to pass.