Open Source Radar
Flink-Agent 智能服务器性能分析与 DeepSeek AI 运维平台,使用 Django、WebSocket 实时推送指标并提供运维助手功能。支持本机 psutil 采集、远程 SSH/Paramiko 采集、历史队列、AIReport 巡检等。
21stars
1forks
0issues
2026since
Star historydaily snapshots by VibeCrowd
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
Reviewgenerated from repository data · Jul 18, 2026
What it is
Flink-Agent is a Django-based server monitoring platform that collects local and remote server metrics and pushes updates via WebSocket to an ECharts dashboard. It includes an AI operations assistant named @Jix for alerts, fault analysis, and optimization suggestions, with DeepSeek integration for intelligent analysis and 12-hour automated inspections.
How it works
- Local agent collects CPU, memory, disk, I/O, network, and top processes data every second via a psutil-based collector running in a background thread.
- Remote servers are monitored via SSH (Paramiko) and collection occurs in parallel managed by a backend collector, with data stored in a bounded history queue and SQLite.
- Data is sent through Django Channels to a WebSocket endpoint, updating the dashboard in real time.
- AI analysis uses DeepSeek for chat completions and local degradation handling when DeepSeek is unavailable or key is missing.
- 12-hour automated inspections produce AIReport entries and can be triggered from the dashboard.
Getting started
-
- Get the project and navigate into it.
-
- Install dependencies: "python -m pip install -r requirements.txt".
-
- Initialize the database: "python manage.py migrate".
-
- Configure DeepSeek by copying the template and editing .env, e.g. DEEPSEEK_API_KEY, DEEPSEEK_MODEL.
-
- Start the project: "python manage.py runserver". Access at http://127.0.0.1:8000/ (Daphne provides HTTP and WebSocket services).
Remote server access is via dashboard, server addition, SSH connection tests, and remote node selection.
Recent releases
- No releases listed in the repository.
Traction
- Stars: 21
Behind the repo
- No linked startup or company information provided.
Caveats
- License: none listed.
- Language: 主体文档为中文,README 提供了详细配置与 API 概览。
- Created: 2026-07-15, Last push: 2026-07-15
