[Bug]: Initializing Podman containers is slow

Testcontainers version

0.39.0

Using the latest Testcontainers version?

Yes

Host OS

Linux

Host arch

x64

Go version

1.24.6

Docker version

Client: Docker Engine - Community
 Version:           28.1.1
 API version:       1.41 (downgraded from 1.49)
 Go version:        go1.23.8
 Git commit:        4eba377
 Built:             Fri Apr 18 09:52:29 2025
 OS/Arch:           linux/amd64
 Context:           default

Server: linux/amd64/debian-unknown
 Podman Engine:
  Version:          5.4.2
  APIVersion:       5.4.2
  Arch:             amd64
  BuildTime:        2025-07-08T21:45:18Z
  Experimental:     false
  GitCommit:        
  GoVersion:        go1.24.4
  KernelVersion:    6.12.35-1rodete1-amd64
  MinAPIVersion:    4.0.0
  Os:               linux
 Conmon:
  Version:          conmon version 2.1.12, commit: unknown
  Package:          conmon_2.1.12-4_amd64
 OCI Runtime (runc):
  Version:          runc version 1.2.5
commit: v1.2.5-0-g59923ef
spec: 1.2.0
go: go1.23.7
libseccomp: 2.6.0
  Package:          containerd.io_1.7.27-1_amd64
 Engine:
  Version:          5.4.2
  API version:      1.41 (minimum version 1.24)
  Go version:       go1.24.4
  Git commit:       
  Built:            Tue Jul  8 21:45:18 2025
  OS/Arch:          linux/amd64
  Experimental:     false

Docker info

Client: Docker Engine - Community
 Version:    28.1.1
 Context:    default
 Debug Mode: false
 Plugins:
  buildx: Docker Buildx (Docker Inc.)
    Version:  v0.23.0
    Path:     /usr/libexec/docker/cli-plugins/docker-buildx
  compose: Docker Compose (Docker Inc.)
    Version:  v2.30.1
    Path:     /[...]/home/polarczyk/.docker/cli-plugins/docker-compose

Server:
 Containers: 81
  Running: 1
  Paused: 0
  Stopped: 80
 Images: 2
 Server Version: 5.4.2
 Storage Driver: overlay
  Backing Filesystem: extfs
  Supports d_type: true
  Native Overlay Diff: true
  Using metacopy: false
  Supports shifting: false
  Supports volatile: true
 Cgroup Driver: systemd
 Cgroup Version: 2
 Plugins:
  Volume: local
  Network: bridge macvlan ipvlan
  Log: k8s-file none passthrough journald
 Swarm: inactive
 Runtimes: crun-vm crun-wasm kata runc runsc youki crun krun ocijail runj
 Default Runtime: runc
 Init Binary: 
 containerd version: 
 runc version: 
 init version: 
 Security Options:
  apparmor
  seccomp
   Profile: default
  rootless
 Kernel Version: 6.12.35-1rodete1-amd64
 Operating System: debian
 OSType: linux
 Architecture: amd64
 CPUs: 24
 Total Memory: 94.29GiB
 Name: polarczyk.c.googlers.com
 ID: a107fe65-de51-4ad3-bcad-dedc3ff9df4a
 Docker Root Dir: /[...]/home/polarczyk/.local/share/containers/storage
 Debug Mode: false
 Experimental: true
 Live Restore Enabled: false
 Product License: Apache-2.0

What happened?

When using Podman (even with reusable containers) the startup time for a test is long, around 12s on average. Tested with Docker, this time is in milliseconds.

I've done some investigation and I assume that communication with Podman in itself is slow, but I found there is a log, that takes over 5s, being 50% of the total time. Removing this log significantly improves the developer experience when running a single test.

Here are the logs from before the change, time annotated:

2025-10-13T14:42:06.119468638Z: Running Test

2025-10-13T14:42:06.119544362Z: provider.go:123 - GetProvider - Getting Podman provider
2025-10-13T14:42:06.119550348Z: provider.go:134 - NewDockerProvider - Entering NewDockerProvider()
2025-10-13T14:42:08.688498747Z: docker_client.go:43 - Info - Entering Info()
2025/10/13 14:42:16 github.com/testcontainers/testcontainers-go - Connected to docker:
          Server Version: 5.4.2
          API Version: 1.41
          Operating System: debian
          Total Memory: 96555 MB
          Testcontainers for Go Version: v0.39.0
          Resolved Docker Host: unix:///run/user/1360621/podman/podman.sock
          Resolved Docker Socket Path: /run/user/1360621/podman/podman.sock
          Test SessionID: 43b345006bb22e17803242bbbad62aa1be3a28abeab83400070bfa4983b50ef1
          Test ProcessID: b1ce92a5-42fd-4f27-a6ea-07371b1f906a

2025-10-13T14:42:16.319430256Z: docker_client.go:67 - Info - Exiting Info()
2025-10-13T14:42:16.319475087Z: docker.go:1327 - ReuseOrCreateContainer - Entering ReuseOrCreateContainer
2025/10/13 14:42:16 ✅ Container started: a51ead6f9f3c
2025/10/13 14:42:16 ⏳ Waiting for container id a51ead6f9f3c image: docker.io/library/postgres:16-alpine. Waiting for: &{timeout:<nil> deadline:0xc000b2e540 Strategies:[0xc0000ff200 0xc0010b3980]}
2025/10/13 14:42:16 🔔 Container is ready: a51ead6f9f3c

And here are the logs after the change:

2025-10-13T14:42:55.768227221Z: provider.go:123 - GetProvider - Getting Podman provider
2025-10-13T14:42:55.768232798Z: provider.go:134 - NewDockerProvider - Entering NewDockerProvider()
2025-10-13T14:42:58.296600686Z: docker_client.go:43 - Info - Entering Info()
2025-10-13T14:43:01.210343414Z: docker_client.go:67 - Info - Exiting Info()
2025-10-13T14:43:01.210594013Z: docker.go:1327 - ReuseOrCreateContainer - Entering ReuseOrCreateContainer
2025/10/13 14:43:01 ✅ Container started: a51ead6f9f3c
2025/10/13 14:43:01 ⏳ Waiting for container id a51ead6f9f3c image: docker.io/library/postgres:16-alpine. Waiting for: &{timeout:<nil> deadline:0xc000a7db60 Strategies:[0xc001397440 0xc000e37c50]}
2025/10/13 14:43:01 🔔 Container is ready: a51ead6f9f3c

I assume much more optimization like caching could be done, but removing this log for Podman runs (or depending on log level) would be sufficient for now for us, improving the total initialization time from 11 to 6s

Relevant log output

Additional information

No response

Start with the provider.go and docker_client.go:43 Info entry points shown in the timing logs, then inspect how NewDockerProvider handles Podman. Reproduce the initialization timing with the reported Podman setup and compare the relevant log behavior. Done means the slow log no longer adds substantial startup time while the provider still initializes correctly and tests pass.