STABForge

Software Tech. Accelerator Blog

Docker Part-4

Posted on by Karishma
Docker

Docker Part-4

Docker Swarm Visual is a powerful tool that makes cluster management simple and efficient.

Docker Swarm. an in-depth introduction to… | by Mesut Oezdil | Medium

1. Prerequisites

  • At least 2 or 3 servers (can be virtual or real)

  • Network access between the machines

  • Linux-based OS (e.g., Ubuntu)

2. Terminology Quick Look

Term Description
Manager Node Controls the swarm cluster (scheduler, orchestration)
Worker Node Executes the containers/services
Service Task definition to be run on the swarm
Task A container running as part of a service

 3. Step-by-Step Setup

Step 1: Initialize the Swarm (on Manager Node)

bashdocker swarm init --advertise-addr <MANAGER-IP>

Step 2: Join Worker Nodes to the Swarm

Run the given command (from manager output) on each worker node:

bashdocker swarm join --token <token> <MANAGER-IP>:2377

Run this on all worker nodes.

Step 3: Verify Nodes (on Manager Node)

bashdocker node ls

 Shows all nodes and their status (Leader, Reachable, Active).

Step 4: Deploy a Service

bashdocker service create --replicas 3 --name web nginx

This will run 3 NGINX containers across the swarm.

Step 5: Check Service Status

service ls # Lists running services

 

Step 6: Scale the Service

 service scale web=5

Increases replicas to 5.

Step 7: Remove the Service

 service rm web

What It Does

  • Shows real-time container placement on each node

  • Displays manager and worker nodes

  • Updates automatically when containers are added/removed

How to Deploy Swarm Visualizer

Step 1: Create an Overlay Network (if needed)


Step 2: Deploy Visualizer on the Manager Node

--publish=8080:8080
 --constraint=node.role==manager

What this does:

  • --publish=8080:8080: Exposes it on port 8080

  • --constraint=node.role==manager: Ensures it only runs on manager node

Step 3: Access the Visualizer UI

  • Open browser:
    http://<MANAGER-IP>:8080

You will see a graphical view of:

  • All nodes in the cluster

  • Running services (containers) with color-coded boxes

What It Looks Like

css

[ Manager Node ]
├── nginx-task-1
└── nginx-task-2

[ Worker Node 1 ]
├── nginx-task-3

[ Worker Node 2 ]
└── nginx-task-4

 Each container appears as a small box under a node.

How services work | Docker Docs

1. What Is a Service?

A Docker Swarm service ensures that the desired number of replicas of a container is always running in the swarm. It also manages:

  • Scaling

  • Load balancing

  • Container placement

  • Rolling updates

2 Types of Services

Type Description
Replicated Run a specific number of container replicas (e.g., 5 nginx containers)
Global Run one container on every node (e.g., monitoring agents like Prometheus Node Exporter)

3. Creating a Swarm Service

 service create
 --name web
 --replicas 3
 --publish 80:80
 nginx

Explanation:

  • --name web: Name of the service

  • --replicas 3: Run 3 containers

  • --publish 80:80: Expose service on port 80

  • nginx: Image to r

5. Load Balancing (Built-in)

The Swarm automatically load balances incoming requests across all replicas.
So if you have:

bash--publish 8080:80

And 5 replicas running, incoming traffic on port 8080 will be distributed across all 5 containers.

6. Updating a Service

 service update
 --image nginx:alpine web

7. Service vs Container (in Swarm)

Docker Swarm Service commands

1. Create a Service

service create --name web --replicas 3 -p 80:80 nginx

Runs 3 replicas of the NGINX container exposed on port 80.

2. List All Services

bashdocker service ls

Shows all running services in the swarm.

3. Inspect a Service

bashdocker service inspect web

Displays detailed information in JSON format.

For readable format:


4. View Tasks (Containers) in a Service

bashdocker service ps web

Lists all containers running as part of the service.

5. Update a Service

bashdocker service update --image nginx:latest web

Updates the image used by the service.

Change replicas:


Scales the service to 5 containers.

6. Roll Back a Service Update

bashdocker service rollback web

Reverts the service to the previous version.

7. Remove a Service

bashdocker service rm web

Deletes the service and stops its containers.

8. Logs of a Service

bashdocker service logs webShows logs from all tasks/containers of the service.

9. Force an Update (even without changes)

bashdocker service update --force web

Restarts all tasks even if nothing has changed.

10. Create Global Service

bashdocker service create
--name monitor
--mode global
prom/node-exporter

Runs one container on every node in the swarm.

Think of it as an advanced way to deploy and manage multiple services (like web, db, cache) together as one unit.

Docker Swarm Stack: Creating A Multi-Service Architecture From a Docker- Compose File | by Shomarri Romell Diaz | FAUN — Developer Community 🐾

Why Use?
Multi-Service Runs multiple services together (e.g., WordPress + MySQL)
Benefit Description
Reusable Uses docker-compose.yml – version-controlled and shareable

Scalable

 Easy to scale services
Declarative Infrastructure as code (IAC)
Organized Groups services logically under one stack name

Structure of a Stack

yamlversion: "3.9"
services:
web:
image: nginx
ports:
- "80:80"
deploy:
replicas: 3db:
image: mysql
environment:
MYSQL_ROOT_PASSWORD: root
deploy:
placement:
constraints: [node.role == manager]

Deploy a Stack (Step-by-Step)

Step 1: Create a docker-compose.yml

yamlversion: "3.9"
services:
web:
image: nginx
ports:
- "8080:80"
deploy:
replicas: 2

Step 2: Deploy the Stack

bashdocker stack deploy -c docker-compose.yml mystack

This creates a stack named mystack with the defined services.

Backup & Restore Docker swarm. This content deals with the backup and… | by Raghav D | Medium


Tips & Best Practices

Tip Description
 Automate Set up cron jobs to backup volumes and swarm state regularly
 Replicate Keep at least 3 manager nodes to protect raft consensus
 Store Offsite Save backups to cloud or remote server
 Secure Protect backups – they may include sensitive data

Summary

Task How
Backup Swarm Backup /var/lib/d/swarm (manager nodes only)
Backup Volumes Use tar inside a container to backup volume contents
Restore Swarm Extract swarm data to the same location
Restore Data Create volume and restore tar into it

 

Conclusion

In conclusion, Docker Swarm Visualizer is an excellent solution for developers and system administrators who want to manage clusters more effectively. With  Swarm Visualizer, it becomes easy to monitor services, scale applications, and deploy stacks in a user-friendly way. Unlike traditional command-line operations,  Swarm Visualizer provides clear insights into cluster activity, making the workflow faster and more reliable.

By integrating  Swarm Visual into your development or production environment, you can ensure better visibility, quicker troubleshooting, and enhanced efficiency. Whether you are setting up new services or restoring backups,r Swarm Visualizer simplifies every task. Overall, adopting  Swarm Visualizer is a smart choice for anyone who wants to take full advantage of clustering power.

For More Learn Click Here

For more related references Click Here

Learn the DevOps to start click here

DevOps Part -3

Previous