Quick Start Tutorial
This quick tutorial walks you through running your first MSight nodes on a single machine.
It works on Windows, Linux, or macOS. You can see the tutorial folder with all resource in it.
π§° Step 1. Install MSight Locally
Follow the Pip Installation guide to install MSight on your machine.
π§± Step 2. Set Up a Redis Server
MSight uses Redis Pub/Sub for communication and configuration.
You need to have a Redis server running locally or accessible in your network.
Option A β Install Redis Natively
You can install Redis directly on your system by following the official guide:
π https://redis.io/docs/latest/operate/oss_and_stack/install/install-redis/
After installation, start the Redis server:
redis-server
Option B β Run Redis via Docker
If you have Docker installed, you can start a Redis container easily:
docker run -d --name redis-server -p 6379:6379 redis:latest
βοΈ Step 3. Set Environment Variable
Set the environment variable MSIGHT_EDGE_DEVICE_NAME to identify your device.
For this tutorial, weβll use testing.
On Linux/macOS:
export MSIGHT_EDGE_DEVICE_NAME=testing
On Windows (PowerShell):
set MSIGHT_EDGE_DEVICE_NAME=testing
π‘ In deployment, replace
testingwith the actual device name (e.g.,intersection_mcity_north).
β Step 4. Check MSight Status
Verify that your environment and Redis connection are working:
msight_status
If everything is set up correctly, you should see a table like this:
--------------------------NODES--------------------------
+-------------+-----------------+-------------------+--------+------------------+----------+
| Node Name | Publish Topic | Subscribe Topic | Type | Last Heartbeat | Status |
+=============+=================+===================+========+==================+==========+
+-------------+-----------------+-------------------+--------+------------------+----------+
If you see this table (even empty), π youβre ready to launch nodes!
π§© Step 5. Launch Two Simple Nodes
In this example, weβll create two nodes:
- A source node that reads an image and publishes it.
- A viewer node that subscribes to the same topic and displays the image.
(Illustration: two nodes connected through Redis Pub/Sub)

πΌοΈ Prepare the Folder
- Create a new folder anywhere on your system.
- Place your favorite
.jpgimage inside it (e.g.,myphoto.jpg).
π Launch the Source Node
Navigate to that folder in your console, start the local image source node:
msight_launch_local_image -n test_local -pt local --sensor-name local_image -p myphoto.jpg
Explanation:
-n: node name (test_local) β must be unique across the system-pt: publishing topic name (local) β subscribers must use the same topic to receive messages--sensor-name: identifies which sensor this data came from-p: path to the image file
If successful, youβll see something like:
Starting the dummy local image source node test_local.
π Launch the Viewer Node
Open a new terminal window and start the viewer node:
msight_launch_image_viewer -n image_viewer -st local
Since both nodes share the same topic (local), the viewer will receive and display the image.
If everything works, a window will open showing your image π
Youβve just built a simple data pipeline β one node publishes, the other subscribes and visualizes.
πΌοΈ Check Current Status
You can check the status of your nodes at any time using:
msight_status
You should see something like this:
--------------------------NODES--------------------------
+--------------+-----------------+-------------------+----------------------+---------------------+----------+
| Node Name | Publish Topic | Subscribe Topic | Type | Last Heartbeat | Status |
+==============+=================+===================+======================+=====================+==========+
| image_viewer | | local | ImageViewerSinkNode | 2025-12-13 21:36:03 | RUNNING |
+--------------+-----------------+-------------------+----------------------+---------------------+----------+
| test_local | local | | LocalImageSourceNode | 2025-12-13 21:36:03 | RUNNING |
+--------------+-----------------+-------------------+----------------------+---------------------+----------+
--------------------------TOPICS--------------------------
+--------------+----------------------------------+---------------+
| Topic Name | Data Type | Description |
+==============+==================================+===============+
| local | msight_core.data.image.ImageData | |
+--------------+----------------------------------+---------------+
π‘ What You Learned
You now understand:
- How MSight nodes communicate through Pub/Sub topics
- How to set up and verify your environment
- How to launch a simple source β sink data flow
This is the foundation for building complex MSight pipelines.
π Continue Exploring
Try these extended tutorials next: