[BUG] random missing network when service has more than one

[shencangsheng]

Description

I am encountering an issue while utilizing an external network, named some-network, in the docker-compose.yml file. I have defined this external network and applied it to a specific service. However, when I use the command docker compose up -d, there are intermittent occurrences where the service starts successfully without loading the specified external network. Upon inspecting the service using docker inspect, I noticed that the some-network entry is not present under the Networks section.

I have already upgraded the version of Docker Compose to above the version that fixes this problem, but this issue still occurs.

Steps To Reproduce

1. docker network create some-network

version: "3.0"
services:
  nginx:
    image: nginx:latest
    environment:
      TZ: Asia/Shanghai
    restart: always
    networks:
      - some-network
      - nginx-network
networks:
  nginx-network:
  some-network:
    external: true

2. docker compose up -d

Expected behavior

docker inspect nginx Networks appear some-network

Compose Version

Docker Compose version v2.24.6

Docker Environment

Client: Docker Engine - Community
 Version:    25.0.3
 Context:    default
 Debug Mode: false
 Plugins:
  buildx: Docker Buildx (Docker Inc.)
    Version:  v0.12.1
    Path:     /usr/libexec/docker/cli-plugins/docker-buildx
  compose: Docker Compose (Docker Inc.)
    Version:  v2.24.6
    Path:     /usr/libexec/docker/cli-plugins/docker-compose

Server:
 Containers: 23
  Running: 23
  Paused: 0
  Stopped: 0
 Images: 18
 Server Version: 25.0.3
 Storage Driver: overlay2
  Backing Filesystem: xfs
  Supports d_type: true
  Using metacopy: false
  Native Overlay Diff: true
  userxattr: false
 Logging Driver: json-file
 Cgroup Driver: cgroupfs
 Cgroup Version: 1
 Plugins:
  Volume: local
  Network: bridge host ipvlan macvlan null overlay
  Log: awslogs fluentd gcplogs gelf journald json-file local splunk syslog
 Swarm: inactive
 Runtimes: io.containerd.runc.v2 runc
 Default Runtime: runc
 Init Binary: docker-init
 containerd version: ae07eda36dd25f8a1b98dfbf587313b99c0190bb
 runc version: v1.1.12-0-g51d5e94
 init version: de40ad0
 Security Options:
  seccomp
   Profile: builtin
 Kernel Version: 3.10.0-1160.49.1.el7.x86_64
 Operating System: CentOS Linux 7 (Core)
 OSType: linux
 Architecture: x86_64
 CPUs: 16
 Total Memory: 62.36GiB
 Name: Genecast_Workstation
 ID: 916dd817-6126-402c-8520-0a0d893f7b49
 Docker Root Dir: /data/lib/docker
 Debug Mode: false
 Experimental: false
 Insecure Registries:
  127.0.0.0/8
 Live Restore Enabled: false

Anything else?

moby/moby#46155

#10778

[f1amy]

Experiencing the same. Messes up my workflow very much.
Could not find any workaround.

[f1amy]

@shencangsheng maybe the same issue as #11563 (comment)?

[tralafiti]

Had the same bug and can confirm that this was fixed in #11510 included in v2.24.7. However, docker desktop still delivers the broken compose version v2.24.6

[ndeloof]

I'm closing this issue as Docker Desktop 4.29.0 comes with v2.26.1 which has the fix for this bug