Loading...

VONDS ©

Lab 157: Docker Networking

Back to Lab Index

Create an isolated user-defined Docker bridge network, attach two containers to it, verify they can resolve and reach each other by name, capture network evidence for the ticket, then remove all resources cleanly.

containers networking troubleshooting

Scenario

Ops ticket: “Stand up an isolated Docker network for app containers. Create a user-defined bridge network, attach two containers, and verify they can resolve each other by name. Then clean up everything.”

Operator context

User-defined bridge networks provide built-in DNS for container name resolution. This is the preferred pattern for multi-container apps that need predictable connectivity without hardcoding IP addresses.

Objective

  • List existing Docker networks.
  • Create a user-defined bridge network named appnet.
  • Run two containers (app1, app2) attached to appnet.
  • Verify containers are running.
  • Verify name resolution and connectivity by pinging app2 from app1.
  • Capture evidence using docker network inspect.
  • Remove containers and delete the network.

Concepts

  • Docker networks define how containers connect and communicate.
  • User-defined bridge networks include automatic DNS-based name resolution for containers on the same network.
  • Container names become resolvable hostnames within the network scope.
  • docker network inspect is your evidence tool for showing attached containers, IP assignments, and network configuration.
  • Cleanup is part of the ticket: remove containers first, then remove the network.

Walkthrough

Step 1 : List existing Docker networks.
Command 
docker network ls

This is your baseline. It proves current state and helps you confirm the default networks exist before creating anything new. 

NETWORK ID     NAME      DRIVER    SCOPE
a1b2c3d4e5f6   bridge    bridge    local
b2c3d4e5f6a1   host      host      local
c3d4e5f6a1b2   none      null      local
Step 2 : Create a user-defined bridge network named appnet.
Command 
docker network create appnet

User-defined bridge networks are the standard approach for multi-container connectivity because they provide scoped DNS and predictable service discovery by name. 

7aa1bb2cc3dd4ee5ff6677889900aabbccddeeff0011223344556677889900
Step 3 : Verify the network exists.
Command 
docker network ls

Confirm the new network appears in the list. This is your proof that creation succeeded before you attach workloads. 

NETWORK ID     NAME      DRIVER    SCOPE
a1b2c3d4e5f6   bridge    bridge    local
b2c3d4e5f6a1   host      host      local
c3d4e5f6a1b2   none      null      local
d4e5f6a1b2c3   appnet    bridge    local
Step 4 : Start two containers on appnet.
Commands 
docker run -d --name app1 --network appnet alpine:latest sleep 1d
docker run -d --name app2 --network appnet alpine:latest sleep 1d

These containers are intentionally minimal and long-running. Using sleep 1d keeps them alive so you can test name resolution and connectivity. 

# Example container IDs:
111122223333444455556666777788889999aaaabbbbccccddddeeeeffff0000
0000ffffeeeeddddccccbbbbaaaa999988887777666655554444333322221111
Step 5 : Verify both containers are running.
Command 
docker ps

Confirm both containers are up before testing network connectivity. 

CONTAINER ID   IMAGE          COMMAND        STATUS          NAMES
111122223333   alpine:latest  "sleep 1d"    Up 3 seconds    app1
0000ffffeeee   alpine:latest  "sleep 1d"    Up 1 seconds    app2
Step 6 : From app1, ping app2 by name.
Command 
docker exec -it app1 ping -c 3 app2

This is the key test. Success proves that user-defined network DNS is working and the containers can reach each other within the bridge network. 

PING app2 (172.19.0.3): 56 data bytes
64 bytes from 172.19.0.3: seq=0 ttl=64 time=0.12 ms
64 bytes from 172.19.0.3: seq=1 ttl=64 time=0.10 ms
64 bytes from 172.19.0.3: seq=2 ttl=64 time=0.11 ms
--- app2 ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
Step 7 : Inspect appnet for ticket evidence.
Command 
docker network inspect appnet

This shows network properties and attached containers. In real ops work, this is the screenshot or copy-paste proof you include back into the ticket. 

[
  {
    "Name": "appnet",
    "Driver": "bridge",
    "Containers": {
      "111122223333...": { "Name": "app1" },
      "0000ffffeeee...": { "Name": "app2" }
    }
  }
]
Step 8 : Remove both containers.
Command 
docker rm -f app1 app2

Remove containers first. Networks cannot be removed while containers are attached. 

app1
app2
Step 9 : Remove the user-defined network.
Command 
docker network rm appnet

This returns the system to baseline and closes out the ticket requirement for full cleanup. 

appnet

Common breakpoints

docker network create fails

The name may already exist or the daemon may be unhealthy. Verify Docker is running and confirm the network name is not in use.

ping by name fails: “bad address” or “unknown host”

Containers may not be attached to the same user-defined network. Confirm both are on appnet and retry.

docker network rm fails: “has active endpoints”

One or more containers are still attached. Remove containers (or disconnect them) before removing the network.

Cleanup checklist

This lab includes cleanup as part of the workflow. Confirm the containers are removed and the network no longer exists.

Commands 
docker ps
docker network ls | grep appnet || true
Success signal

app1 and app2 are gone, and appnet no longer appears in docker network ls.

Reference

  • docker network ls : List Docker networks.
  • docker network create <name> : Create a user-defined network.
  • docker run -d --name <name> --network <net> <image> <cmd> : Start a container attached to a specific network.
  • docker ps : List running containers.
  • docker exec -it <container> ping -c <n> <name> : Test name resolution and connectivity from inside a container.
  • docker network inspect <name> : Show network configuration and attached containers.
  • docker rm -f <container...> : Stop and remove one or more containers.
    • -f : Force stop then remove.
  • docker network rm <name> : Remove a Docker network.