Podman Compose 1.3.0 up -d command never returns/finishes/ends

[ArthoPacini]

Describe the bug
When running podman-compose up -d the command never returns, it stalls / freezes

To Reproduce

1. docker-compose.yaml:

services:
  service_a:
    image: docker.io/crccheck/hello-world
    container_name: "service_a"
    depends_on:
      service_b:
        condition: service_healthy
    healthcheck:
      test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost:8000"]
      interval: 10s
      start_interval: 15s
      timeout: 5s
      retries: 3

  service_b:
    image: docker.io/crccheck/hello-world
    container_name: "service_b"
    healthcheck:
      test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost:8000"]
      interval: 10s
      start_interval: 15s
      timeout: 5s
      retries: 3

networks:
  default:
    driver: bridge

2. Run docker-compose up -d:

podman-compose up -d
9a61f6744544dd5343a3e497579e2aadfbac156936d735e89749d95b520832f2
8d43c893be1ddbda564692e8e7e28f27f8a04c559ac1d38a121cf6b6736e4d29
7a350b27a718f4bda86977f51ef32ae75da0b78c38b7d5c84a1e3a62ad767254

Expected behavior
The command should end

Actual behavior
It requires CTRL-C to be called to terminate the program in which it outputs:

^CTraceback (most recent call last):
  File "/usr/lib/python3.11/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.11/asyncio/base_events.py", line 653, in run_until_complete
    return future.result()
           ^^^^^^^^^^^^^^^
  File "/usr/local/bin/podman-compose", line 3707, in async_main
    await podman_compose.run()
  File "/usr/local/bin/podman-compose", line 1861, in run
    retcode = await cmd(self, args)
              ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/bin/podman-compose", line 2705, in compose_up
    await run_container(compose, cnt["name"], cnt["_deps"], ([], "start", [cnt["name"]]))
  File "/usr/local/bin/podman-compose", line 2654, in run_container
    await check_dep_conditions(compose, deps)
  File "/usr/local/bin/podman-compose", line 2643, in check_dep_conditions
    await asyncio.sleep(1)
  File "/usr/lib/python3.11/asyncio/tasks.py", line 639, in sleep
    return await future
           ^^^^^^^^^^^^
asyncio.exceptions.CancelledError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/local/bin/podman-compose", line 3715, in <module>
    main()
  File "/usr/local/bin/podman-compose", line 3711, in main
    asyncio.run(async_main())
  File "/usr/lib/python3.11/asyncio/runners.py", line 190, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.11/asyncio/runners.py", line 123, in run
    raise KeyboardInterrupt()
KeyboardInterrupt

Output

$ podman-compose version
podman-compose version 1.3.0
podman version 4.3.1

podman --version
podman version 4.3.1

$ podman-compose up -d
...

Environment:

• OS: Debian 12
• podman version: 4.3.1
• podman compose version: 1.3.0

[schnell18]

This seems to be related to the podman version. I can run your example successfully w/ a more recent podman:

❯ podman-compose -f docker-compose.yml up -d
0e02d346cfd19d2fee3c5a60caaff6bc32aef82395b40c4cf644d70102dee358
9e16fbcbdcd87f16bffc34186db47e329ddf9518c9082a80a2229726ee6861de
service_b
65bc84d794f06496cbf9605243aca2abca532b535513f1168da2f5b8bc6d6f01
service_a
❯ podman-compose -f docker-compose.yml ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
9e16fbcbdcd8 docker.io/crccheck/hello-world:latest /bin/sh -c echo "... 3 seconds ago Up 3 seconds (healthy) service_b
65bc84d794f0 docker.io/crccheck/hello-world:latest /bin/sh -c echo "... 3 seconds ago Up 3 seconds (healthy) service_a
❯ podman-compose -f docker-compose.yml down
service_a
service_b
service_a
service_b
0e02d346cfd19d2fee3c5a60caaff6bc32aef82395b40c4cf644d70102dee358
foo_default

The podman-compose and podman versoins are as follows:

❯ podman-compose --version
podman-compose version 1.3.0
podman version 5.4.1

So you may try to work around this issue by upgrade to podman 5.x. Hope this helps!

[schnell18]

Just to supply additional findings regarding this issue. It is indeed bug which fails to take into account the fact that podman prior to 4.6.0 doesn't support "podman wait --conditioni=healthy" and enters a dead loop. This bug is fixed by the PR #1184.

@ArthoPacini, do you have any chance to validate the PR #1184? Thanks!

[p12tic]

Fixed in #1184.