[BUG] --wait shouldn't have any effect when using a single service

[leonardoheld]

Description

Pretty much the title, if you use "--wait" with a single service, it hangs indefinitely. This makes sense: if you --wait and you only have one service, it'll hang forever because it won't receive a Healthy or Running from anyone else (I think...).

Steps To Reproduce

Run docker compose up --wait and docker compose up on the provided docker-compose.yml.
Adding a restart always or adding another service that will emit a Healthy or Running status clears the problem, hence my conjecture.

Compose Version

Docker Compose version v2.14.1

Docker Environment

➜  test docker info
Client:
 Context:    default
 Debug Mode: false
 Plugins:
  app: Docker App (Docker Inc., v0.9.1-beta3)
  buildx: Docker Buildx (Docker Inc., v0.9.1-docker)
  compose: Docker Compose (Docker Inc., v2.14.1)
  scan: Docker Scan (Docker Inc., v0.23.0)

Server:
 Containers: 20
  Running: 1
  Paused: 0
  Stopped: 19
 Images: 107
 Server Version: 20.10.22
 Storage Driver: overlay2
  Backing Filesystem: extfs
  Supports d_type: true
  Native Overlay Diff: false
  userxattr: true
 Logging Driver: json-file
 Cgroup Driver: systemd
 Cgroup Version: 2
 Plugins:
  Volume: local
  Network: bridge host ipvlan macvlan null overlay
  Log: awslogs fluentd gcplogs gelf journald json-file local logentries splunk syslog
 Swarm: inactive
 Runtimes: io.containerd.runc.v2 io.containerd.runtime.v1.linux runc
 Default Runtime: runc
 Init Binary: docker-init
 containerd version: 5b842e528e99d4d4c1686467debf2bd4b88ecd86
 runc version: v1.1.4-0-g5fd4c4d
 init version: de40ad0
 Security Options:
  seccomp
   Profile: default
  rootless
  cgroupns
 Kernel Version: 5.15.0-58-generic
 Operating System: Ubuntu 22.04.1 LTS
 OSType: linux
 Architecture: x86_64
 CPUs: 20
 Total Memory: 15.37GiB
 Name: toradex
 ID: R7MP:XSGO:BLYX:7HHW:OFHR:U3S4:B4AN:B4AL:54BV:KRXB:MKXX:3MSF
 Docker Root Dir: /home/ljh/.local/share/docker
 Debug Mode: false
 Username: leonardoheldattoradex
 Registry: https://index.docker.io/v1/
 Labels:
 Experimental: false
 Insecure Registries:
  127.0.0.0/8
 Live Restore Enabled: false

WARNING: No cpu cfs quota support
WARNING: No cpu cfs period support
WARNING: No cpu shares support
WARNING: No cpuset support
WARNING: No io.weight support
WARNING: No io.weight (per device) support
WARNING: No io.max (rbps) support
WARNING: No io.max (wbps) support
WARNING: No io.max (riops) support
WARNING: No io.max (wiops) support

Anything else?

No response

[laurazard]

This isn't quite accurate, although it's a bit complicated –

If using a Compose file such as:

services:
 a:
  image: alpine
  command: top

Running compose up --wait does work, as the condition running_or_healthy is applied, and the container stays alive (since top doesn't exit) long enough for it's status to be running when we make the inspect call.

However, if using a Compose file such as:

services:
 a:
  image: alpine
  command: echo hello world

Running compose up --wait indeed hangs forever, as the container exits almost immediately and by the time Compose makes the container inspect call, it's already dead.

This is a racy issue, and we'll have to look into whether we can rearchitect this to fix it.

In the meantime however, compose up --wait does work and has value, even for single service Compose files.

[ndeloof]

thanks to @laurazard diagnostic, I've proposed #10209

[leonardoheld]

Hi y'all!
Laura's explanation makes a lot more sense than mine, thanks for understanding the problem even with my bad description. I had to deal with a hellish Yocto recipe situation to test it, but Nicolas' fixes seem to work fine for the cases I tried :-)

Shall we keep this issue open until #10209 is merged?

Thank you again!