Container Sentry

Introduction

This is a Docker monitoring application that searches for patterns in the logs of running Docker containers and sends a message to a Telegram chat if a pattern is found. The application requires a Docker daemon to be running on the host machine and a Telegram bot to be set up with a valid token and chat ID.

Configuration

docker-config.json

This file contains the configuration for connecting to the Docker daemon and the Telegram bot. The contents of the file are as follows:

{
  "docker_socket": "/var/run/docker.sock",
  "telegram": {
    "token": "your_token_here",
    "chat_id": "your_chat_id_here"
  }
}

docker_socket: the path to the Docker daemon socket file. telegram.token: the Telegram bot token. telegram.chat_id: the ID of the Telegram chat to send messages to.

containers-config.json

This file contains the configuration for the containers to monitor and the patterns to search for in their logs. The contents of the file are as follows:

{
  "containers": {
    "container_name_1": {
      "patterns": [
        {
          "pattern": "error",
          "message": "Error occurred in container_name_1"
        }
      ]
    },
    "container_name_2": {
      "patterns": [
        {
          "pattern": "warning",
          "message": "Warning occurred in container_name_2"
        },
        {
          "pattern": "critical",
          "message": "Critical occurred in container_name_2"
        }
      ]
    }
  }
}

containers.container_name_x: the name of the container to monitor, where x is a number. containers.container_name_x.patterns: a list of patterns to search for in the container’s logs. containers.container_name_x.patterns.pattern: the string pattern to search for in the logs. containers.container_name_x.patterns.message: the message to send to the Telegram chat if the pattern is found.

requirements.txt

This file lists the Python dependencies required by the application. The contents of the file are as follows:

docker
requests

docker: the Python Docker API library

requests: the Python HTTP library used to send messages to Telegram.

Running the Application

Create a Telegram bot and obtain a token and chat ID.

1. Create the docker-config.json and containers-config.json files and populate them with the required configurations.
2. Create a Docker image of the application using the provided Dockerfile.
3. Start the application container with the following command:

docker run -v /var/run/docker.sock:/var/run/docker.sock your_image_name

Limitations

• The application only searches for patterns in the logs of running Docker containers.

• The application only sends messages to a single Telegram chat.

• The application only supports the string pattern matching.