Mattermost Message Monitor

A Go-based application that connects to Mattermost's WebSocket API to monitor and log messages from specified channels. It captures "posted" events, extracts relevant information including channel names, and writes the data to a JSON file.

Features

• Real-time Monitoring: Connects to Mattermost's WebSocket API to listen for events in real-time.
• Event Filtering: Processes only "posted" events from specified channels.
• Structured Logging: Structured logs with configurable log levels (DEBUG, INFO, WARN, ERROR).
• JSON Output: Writes captured messages to a file, including details like channel names.
• File Rotation: Rotate the output file on given intervals.

Requirements

• Go: Version 1.16 or higher.
• Mattermost Server: Access to a Mattermost server with appropriate permissions.
• Access Token: A valid Mattermost access token with access to the monitored channels.

Installation

Docker

A docker image of the application is available at ghcr.io/vfaergestad/mattermost-logger.

Run with docker compose:

services:
  mattermost-logger:
    image: ghcr.io/vfaergestad/mattermost-logger:latest
    container_name: mattermost-logger
    volumes:
      - ./config.json:/app/config.json
      - ./out:/app/out

Build the binary

1. Clone the Repository:

git clone https://github.com/vfaergestad/mattermost-logger.git

2. Navigate to the Project Directory:

cd mattermost-logger

3. Install Dependencies:

go mod tidy

4. Build the Application:

go build -o mattermost_logger cmd/main.go

Configuration

Create a config.json file in the project directory with the following structure:

{
"mattermost_domain": "your-mattermost-domain.com",
"mattermost_port": 443,
"use_tls": true,
"auth_token": "your_access_token",
"channel_ids": [
  "channel_id_1",
  "channel_id_2"
],
"output_dir": "out",
"output_file_prefix": "messages",
"rotation_interval": "5m",
"insecure_skip_tls_verify": false,
"log_level": "INFO"
}

Configuration Fields

• mattermost_domain: The domain of your Mattermost server.
• mattermost_port: (Optional) The port of your Mattermost server. Defaults to 443 for tls, else 80.
• use_tls: Set to true if using TLS (wss), otherwise false (ws).
• auth_token: Your Mattermost access token.
• channel_ids: An array of channel IDs to monitor.
• output_dir: (Optional) The directory to save the output of the bot. Defaults to out.
• output_file_prefix: (Optional) The prefix of the files created by the bot. Defaults to messages.
• rotation_interval: (Optional) The interval that the bot will rotate its output files. Supported: 1m, 5m, 24h, 168h (1 week), 720h (1 month). Default: 24h
• insecure_skip_tls_verify: (Optional) Set to true to skip TLS certificate verification (not recommended for production).
• log_level: (Optional) Set the logging level (DEBUG, INFO, WARN, ERROR). Defaults to INFO.

Logging

The application uses the zap library for structured logging. Logs are categorized into different levels:

• DEBUG: Detailed information, useful during development and troubleshooting.
• INFO: General operational messages.
• WARN: Indicators of potential issues.
• ERROR: Critical problems that may prevent some operations from functioning correctly.

Log Configuration

Adjust the log_level in config.json to control the verbosity of the logs.

"log_level": "INFO"

Available levels:

• DEBUG
• INFO
• WARN
• ERROR

Output

Captured messages are written to the specified output_file in JSON format. Each entry includes:

• id: Message ID.
• created_at: Timestamp when the message was created on Mattermost.
• user_id: ID of the user who posted the message.
• channel_id: ID of the channel where the message was posted.
• channel_name: Name of the channel.
• message: The message content.
• username: Username of the sender.
• processed_at: Timestamp when the message was processed by the application.

Sample messages.json Entry

{
"id": "mpno5dgjp3fhxjy7smsx4cjttc",
"created_at": "2024-09-14T12:32:04Z",
"user_id": "ww98uzzdtfy4mmca7wfy96gp1w",
"channel_id": "pjpk8d8qopg9cowtggf1bbqkknz3h",
"channel_name": "social",
"message": "heyhey",
"username": "nordmann",
"processed_at": "2024-09-14T12:32:04Z"
}