github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 08:30:05 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #10110] Ollama does not recognize models that were copied #53143

New Issue
Closed
opened 2026-04-29 02:04:52 -05:00 by GiteaMirror · 6 comments
GiteaMirror commented 2026-04-29 02:04:52 -05:00
Owner
Copy Link

Originally created by @zaredh on GitHub (Apr 3, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10110

What is the issue?

I'm trying to copy the contents of the models on my host machine into a docker container. This is how I'm constructing the docker-compose.yml:

ollama:
    build:
      context: ./ollama
      dockerfile: Dockerfile
    container_name: ollama
    restart: always
    ports:
      - "8006:11434"
    volumes:
      - ollama_data:/root/.ollama
      - ${MODELS_DIR}:/usr/share/ollama/.ollama
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]

This is the contents of the Dockerfile:

FROM ollama/ollama

EXPOSE 11434

CMD []

Yes, I'm aware that I don't need a Dockerfile for this, I've obfuscated some things from it.

MODELS_DIR is set in the .env to /usr/share/ollama/.ollama. When I exec into the container after startup, the contents of my host machine are exactly replicated in /usr/share/ollama/.ollama.

ollama list returns an empty list. Restarting the container does not resolve the issue either.

There is no Modelfile by default stored under manifests/registry.ollama.ai/library/

For example with llama3.1, the library is library/llama3.1/latest.

Relevant log output

OS

Linux

GPU

Nvidia

CPU

No response

Ollama version

0.6.2

Originally created by @zaredh on GitHub (Apr 3, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10110 ### What is the issue? I'm trying to copy the contents of the models on my host machine into a docker container. This is how I'm constructing the docker-compose.yml: ``` ollama: build: context: ./ollama dockerfile: Dockerfile container_name: ollama restart: always ports: - "8006:11434" volumes: - ollama_data:/root/.ollama - ${MODELS_DIR}:/usr/share/ollama/.ollama deploy: resources: reservations: devices: - driver: nvidia count: 1 capabilities: [gpu] ``` This is the contents of the Dockerfile: ``` FROM ollama/ollama EXPOSE 11434 CMD [] ``` Yes, I'm aware that I don't need a Dockerfile for this, I've obfuscated some things from it. MODELS_DIR is set in the .env to /usr/share/ollama/.ollama. When I exec into the container after startup, the contents of my host machine are exactly replicated in /usr/share/ollama/.ollama. `ollama list` returns an empty list. Restarting the container does not resolve the issue either. There is no Modelfile by default stored under manifests/registry.ollama.ai/library/ For example with llama3.1, the library is library/llama3.1/latest. ### Relevant log output ```shell ``` ### OS Linux ### GPU Nvidia ### CPU _No response_ ### Ollama version 0.6.2
GiteaMirror added the bug label 2026-04-29 02:04:52 -05:00
GiteaMirror commented 2026-04-29 02:04:59 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 3, 2025):

    volumes:
      - ollama_data:/root/.ollama
      - ${MODELS_DIR}:/root/.ollama/models

Set MODELS_DIR to /usr/share/ollama/.ollama/models.

<!-- gh-comment-id:2775737690 --> @rick-github commented on GitHub (Apr 3, 2025): ```yaml volumes: - ollama_data:/root/.ollama - ${MODELS_DIR}:/root/.ollama/models ``` Set MODELS_DIR to /usr/share/ollama/.ollama/models.
GiteaMirror commented 2026-04-29 02:05:00 -05:00
Author
Owner
Copy Link

@zaredh commented on GitHub (Apr 3, 2025):

Tried that just now. The only difference is that ollama/.ollama no longer has id_ed25519 and id_ed25519.pub files that are on my host machine. ollama list still returns nothing before and after a container restart.

Edit: Including picture from inside the container of the llama3.1 blob files and ollama list returning nothing:

Image

<!-- gh-comment-id:2775761325 --> @zaredh commented on GitHub (Apr 3, 2025): Tried that just now. The only difference is that ollama/.ollama no longer has id_ed25519 and id_ed25519.pub files that are on my host machine. `ollama list` still returns nothing before and after a container restart. Edit: Including picture from inside the container of the llama3.1 blob files and `ollama list` returning nothing: ![Image](https://github.com/user-attachments/assets/d7c781e4-f250-4eb9-99ba-60b6c8f39bc3)
GiteaMirror commented 2026-04-29 02:05:01 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 3, 2025):

Is the screen shot from inside or outside the container?

<!-- gh-comment-id:2775786910 --> @rick-github commented on GitHub (Apr 3, 2025): Is the screen shot from inside or outside the container?
GiteaMirror commented 2026-04-29 02:05:01 -05:00
Author
Owner
Copy Link

@zaredh commented on GitHub (Apr 3, 2025):

Inside. Will edit original.

If I run ollama list from the host machine, it returns

NAME               ID              SIZE      MODIFIED     
llama3.1:latest    46e0c10c039e    4.9 GB    2 months ago
<!-- gh-comment-id:2775794448 --> @zaredh commented on GitHub (Apr 3, 2025): Inside. Will edit original. If I run `ollama list` from the host machine, it returns ``` NAME ID SIZE MODIFIED llama3.1:latest 46e0c10c039e 4.9 GB 2 months ago ```
GiteaMirror commented 2026-04-29 02:05:03 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 3, 2025):

Inside the container, the models need to be mounted at /root/.ollama/models, see https://github.com/ollama/ollama/issues/10110#issuecomment-2775737690

<!-- gh-comment-id:2775805342 --> @rick-github commented on GitHub (Apr 3, 2025): Inside the container, the models need to be mounted at /root/.ollama/models, see https://github.com/ollama/ollama/issues/10110#issuecomment-2775737690
GiteaMirror commented 2026-04-29 02:05:04 -05:00
Author
Owner
Copy Link

@zaredh commented on GitHub (Apr 3, 2025):

My mistake. Completely misread that location. Thank you so much.

<!-- gh-comment-id:2775826755 --> @zaredh commented on GitHub (Apr 3, 2025): My mistake. Completely misread that location. Thank you so much.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#53143
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?