[Personal Experience] The “5 Traps” I Fell Into When Migrating to Podman<\/h2>\n
When switching from Docker to Podman, I neatly fell into several traps. Knowing the following countermeasures should save you from wasting time.<\/p>\n
- \n
- Trap where the GPU is not recognized
GPU does not work with Docker syntax. Instead, use “CDI,” an official mechanism provided by NVIDIA to connect GPUs directly to containers.<\/li>\n - Trap where images cannot be downloaded
Podman is strict about security and rejects shorthand names. Always specify the full path (fully qualified name) like docker.io\/ollama\/ollama.<\/li>\n - Trap where chat history disappears
Directly specifying a Windows-side folder causes database corruption due to permission errors. Use “named volumes,” where Podman automatically creates a safe storage area.<\/li>\n - Trap where loading takes longer after 5 minutes of inactivity
By default, models are evicted from VRAM if there is no chat activity for 5 minutes. Set the environment variable OLLAMA_KEEP_ALIVE=-1 to disable timeout and keep them resident in VRAM at all times.<\/li>\n - Trap where models are not loaded simultaneously
Even if you try to switch and use two models assuming a multi-agent setup, the default specification is that “only one model can be loaded into VRAM at a time.” Explicitly specify the environment variable OLLAMA_MAX_LOADED_MODELS=2 to allow simultaneous standby of two models.<\/li>\n<\/ol>\n[Shortest Build Procedure] docker-compose.yml<\/h2>\n
As a prerequisite, it is assumed that a WSL2 (Ubuntu 22.04 or later) environment is prepared.<\/p>\n
Step 1: Installing Podman and GPU Integration Tools<\/h3>\n
Open the Ubuntu terminal and install the required packages.<\/p>\n
# Podman core and Compose tools introduction\nsudo apt update\nsudo apt install -y podman podman-compose podman-docker\n\n# Introduce NVIDIA-provided GPU passthrough tool\nsudo apt-get install -y nvidia-container-toolkit\n\n# Automatically generate configuration file for CDI (direct GPU connection mechanism)\nsudo nvidia-ctk cdi generate --output=\/etc\/cdi\/nvidia.yaml<\/pre>\n
Step 2: Creating docker-compose.yml<\/h3>\n
Create the following docker-compose.yml in an appropriate working folder. This is the definitive edition incorporating all countermeasures against the “5 traps based on personal experience” mentioned above.<\/p>\n
services:\n ollama:\n image: docker.io\/ollama\/ollama\n container_name: ollama\n ports:\n - "0.0.0.0:11434:11434"\n environment:\n - "OLLAMA_KEEP_ALIVE=-1" # Countermeasure for Trap 4: Disable timeout\n - "OLLAMA_MAX_LOADED_MODELS=2" # Countermeasure for Trap 5: Simultaneous standby of 2 models\n volumes:\n - ollama_data:\/root\/.ollama # Countermeasure for Trap 3: Named volume\n devices:\n - nvidia.com\/gpu=all # Countermeasure for Trap 1: GPU passthrough via CDI\n\n open-webui:\n image: ghcr.io\/open-webui\/open-webui:main\n container_name: open-webui\n ports:\n - "0.0.0.0:3000:8080"\n environment:\n - OLLAMA_BASE_URL=http:\/\/ollama:11434\n volumes:\n - open-webui_data:\/app\/backend\/data\n depends_on:\n - ollama\n\nvolumes:\n ollama_data:\n open-webui_data:<\/pre>\n
Step 3: Startup and Fetching AI Models<\/h3>\n
Do not use sudo under any circumstances; run the startup command as a regular user.<\/p>\n
podman-compose up -d<\/pre>\n
Once started, download the models you want to use within the Ollama container.<\/p>\n
podman exec -it ollama ollama run qwen3:4b\npodman exec -it ollama ollama run gemma3:4b<\/pre>\n
You can also add models via settings on the browser.<\/p>\n
![\"Acquiring](\"https:\/\/donguri3.net\/wp-content\/uploads\/2026\/03\/68736b887d7eef46240becfe5d523cb2.png\")
<\/a><\/p>\n