github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #6308] Getting Error: unexpected status code 200 when pulling a model from an internal registry v0.3.1 and above #29716

New Issue
Open
opened 2026-04-22 08:52:49 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-22 08:52:49 -05:00
Owner
Copy Link

Originally created by @killerwhile on GitHub (Aug 11, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6308

What is the issue?

Starting version 0.3.1, when pulling a model from an internal registry (https://distribution.github.io/distribution/), I'm getting the error unexpected status code 200.
Version up to 0.3.0 worked properly with this setup.

The line returning the error seems to be https://github.com/ollama/ollama/compare/v0.3.0...v0.3.1#diff-9e32d213fc229fc9c327863932f4fc8a875d854333b5ad2dffa9b43fd0848232R226

How to reproduce

Via docker-compose, I create a test environment with ollama and registry.

version: "3"

services:
  ollama:
    image: ollama/ollama:0.3.0
    ports:
      - "11434"

  registry:
    image: registry:2
    environment:
      REGISTRY_LOG_LEVEL: debug
      REGISTRY_LOG_ACCESSLOG_DISABLED: "false"
    ports:
      - "5000"

Start the test stack via docker compose up -d.

In the snippet above, I'm using ollama v0.3.0.

The following commands will pull a model (qwen2:0.5b for sake of size), and push it to the local registry.

docker compose exec ollama ollama pull qwen2:0.5b
docker compose exec ollama ollama cp qwen2:0.5b registry:5000/library/qwen2:0.5b
docker compose exec ollama ollama push registry:5000/library/qwen2:0.5b --insecure

Now the models can be remove from ollama:

docker compose exec ollama ollama rm qwen2:0.5b registry:5000/library/qwen2:0.5b

And re-downloaded from the local registry:

docker compose exec ollama ollama pull registry:5000/library/qwen2:0.5b --insecure

With ollama version up to 0.3.0 (included), this works.
With ollama version from 0.3.1 (included), I'm getting the following error:

Error: unexpected status code 200

OS

Docker

GPU

No response

CPU

No response

Ollama version

0.3.1 and above

Originally created by @killerwhile on GitHub (Aug 11, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6308 ### What is the issue? Starting version 0.3.1, when pulling a model from an internal registry (https://distribution.github.io/distribution/), I'm getting the error `unexpected status code 200`. Version up to 0.3.0 worked properly with this setup. The line returning the error seems to be https://github.com/ollama/ollama/compare/v0.3.0...v0.3.1#diff-9e32d213fc229fc9c327863932f4fc8a875d854333b5ad2dffa9b43fd0848232R226 # How to reproduce Via docker-compose, I create a test environment with ollama and registry. ``` version: "3" services: ollama: image: ollama/ollama:0.3.0 ports: - "11434" registry: image: registry:2 environment: REGISTRY_LOG_LEVEL: debug REGISTRY_LOG_ACCESSLOG_DISABLED: "false" ports: - "5000" ``` Start the test stack via `docker compose up -d`. In the snippet above, I'm using ollama v0.3.0. The following commands will pull a model (qwen2:0.5b for sake of size), and push it to the local registry. ``` docker compose exec ollama ollama pull qwen2:0.5b docker compose exec ollama ollama cp qwen2:0.5b registry:5000/library/qwen2:0.5b docker compose exec ollama ollama push registry:5000/library/qwen2:0.5b --insecure ``` Now the models can be remove from ollama: ``` docker compose exec ollama ollama rm qwen2:0.5b registry:5000/library/qwen2:0.5b ``` And re-downloaded from the local registry: ``` docker compose exec ollama ollama pull registry:5000/library/qwen2:0.5b --insecure ``` With ollama version up to 0.3.0 (included), this works. With ollama version from 0.3.1 (included), I'm getting the following error: ``` Error: unexpected status code 200 ``` ### OS Docker ### GPU _No response_ ### CPU _No response_ ### Ollama version 0.3.1 and above
GiteaMirror added the bug label 2026-04-22 08:52:49 -05:00
GiteaMirror commented 2026-04-22 08:52:50 -05:00
Author
Owner
Copy Link

@jorander commented on GitHub (Aug 17, 2024):

I experience the exact same issue still in version 0.3.6.

With version <= 0.3.0 pulling from my internal library setup in Artifactory works just fine, but in later versions I get the "unexpected status code 200".
I seems to me that the problem is that this code, added in PR #5962, supposes there will be at least one redirect, but when using an internal repository we will have the direct url from the start.

<!-- gh-comment-id:2294911119 --> @jorander commented on GitHub (Aug 17, 2024): I experience the exact same issue still in version 0.3.6. With version <= 0.3.0 pulling from my internal library setup in Artifactory works just fine, but in later versions I get the "unexpected status code 200". I seems to me that the problem is that [this code](https://github.com/ollama/ollama/commit/c8af3c2d969a99618eecf169bd75aa112573ac27#diff-9e32d213fc229fc9c327863932f4fc8a875d854333b5ad2dffa9b43fd0848232R183-R222), added in PR #5962, supposes there will be at least one redirect, but when using an internal repository we will have the direct url from the start.
GiteaMirror commented 2026-04-22 08:52:51 -05:00
Author
Owner
Copy Link

@yuebo commented on GitHub (Sep 4, 2024):

The same error for me, and I just setup a nginx proxy to resolve this.
You need to have two domain name in your private network. here I use hub-ollama.test.com and blobs-ollama.test.com, and related nginx config like this.

server {
    listen       80;
    listen  [::]:80;
    listen 443 ssl;
    listen [::]:443 ssl;

    server_name hub-ollama.test.com;
    client_max_body_size 409600m;
    ssl_certificate "test.pem";
    ssl_certificate_key "test.key";
    ssl_session_cache shared:SSL:1m;
    ssl_session_timeout  10m;
    ssl_ciphers EECDH+CHACHA20:EECDH+CHACHA20-draft:EECDH+AES128:RSA+AES128:EECDH+AES256:RSA+AES256:EECDH+3DES:RSA+3DES:!MD5;
    ssl_prefer_server_ciphers on;
    location / {
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto "https";
        proxy_pass http://127.0.0.1:5080;
        proxy_redirect http://127.0.0.1:5080/ https://hub-ollama.test.com;
    }
    location ~ ^/v2/.*/blobs/.*$ {
        return 307 https://blobs-ollama.test.com$request_uri;
    }

}

server {
    listen       80;
    listen  [::]:80;
    listen 443 ssl;
    listen [::]:443 ssl;

    server_name blobs-ollama.test.com;
    client_max_body_size 409600m;
    ssl_certificate "test.pem";
    ssl_certificate_key "test.key";
    ssl_session_cache shared:SSL:1m;
    ssl_session_timeout  10m;
    ssl_ciphers EECDH+CHACHA20:EECDH+CHACHA20-draft:EECDH+AES128:RSA+AES128:EECDH+AES256:RSA+AES256:EECDH+3DES:RSA+3DES:!MD5;
    ssl_prefer_server_ciphers on;
    location / {
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto "https";
        proxy_pass http://127.0.0.1:5080;
        proxy_redirect http://127.0.0.1:5080/ https://hub-ollama.test.com;
    }
}

The key configuration is below,

location ~ ^/v2/.*/blobs/.*$ {
        return 307 https://blobs-ollama.test.com$request_uri;
}

you can use ollama to pull private registry images like ollama pull hub-ollama.test.com/llama3:7b.

<!-- gh-comment-id:2328284162 --> @yuebo commented on GitHub (Sep 4, 2024): The same error for me, and I just setup a nginx proxy to resolve this. You need to have two domain name in your private network. here I use `hub-ollama.test.com` and `blobs-ollama.test.com`, and related nginx config like this. ``` server { listen 80; listen [::]:80; listen 443 ssl; listen [::]:443 ssl; server_name hub-ollama.test.com; client_max_body_size 409600m; ssl_certificate "test.pem"; ssl_certificate_key "test.key"; ssl_session_cache shared:SSL:1m; ssl_session_timeout 10m; ssl_ciphers EECDH+CHACHA20:EECDH+CHACHA20-draft:EECDH+AES128:RSA+AES128:EECDH+AES256:RSA+AES256:EECDH+3DES:RSA+3DES:!MD5; ssl_prefer_server_ciphers on; location / { proxy_set_header Host $host; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header X-Forwarded-Proto "https"; proxy_pass http://127.0.0.1:5080; proxy_redirect http://127.0.0.1:5080/ https://hub-ollama.test.com; } location ~ ^/v2/.*/blobs/.*$ { return 307 https://blobs-ollama.test.com$request_uri; } } server { listen 80; listen [::]:80; listen 443 ssl; listen [::]:443 ssl; server_name blobs-ollama.test.com; client_max_body_size 409600m; ssl_certificate "test.pem"; ssl_certificate_key "test.key"; ssl_session_cache shared:SSL:1m; ssl_session_timeout 10m; ssl_ciphers EECDH+CHACHA20:EECDH+CHACHA20-draft:EECDH+AES128:RSA+AES128:EECDH+AES256:RSA+AES256:EECDH+3DES:RSA+3DES:!MD5; ssl_prefer_server_ciphers on; location / { proxy_set_header Host $host; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header X-Forwarded-Proto "https"; proxy_pass http://127.0.0.1:5080; proxy_redirect http://127.0.0.1:5080/ https://hub-ollama.test.com; } } ``` The key configuration is below, ``` location ~ ^/v2/.*/blobs/.*$ { return 307 https://blobs-ollama.test.com$request_uri; } ``` you can use ollama to pull private registry images like `ollama pull hub-ollama.test.com/llama3:7b`.
GiteaMirror commented 2026-04-22 08:52:51 -05:00
Author
Owner
Copy Link

@GhostDog98 commented on GitHub (Feb 17, 2025):

I'm getting this exact same issue when pulling certain models from huggingface...
To replicate:
ollama pull hf.co/zetasepic/Qwen2.5-32B-Instruct-abliterated-v2-GGUF:IQ4_XS

Is there a workaround for this that doesn't involve setting up an entire freaking proxy server?

<!-- gh-comment-id:2662429385 --> @GhostDog98 commented on GitHub (Feb 17, 2025): I'm getting this exact same issue when pulling certain models from huggingface... To replicate: `ollama pull hf.co/zetasepic/Qwen2.5-32B-Instruct-abliterated-v2-GGUF:IQ4_XS` Is there a workaround for this that doesn't involve setting up an entire freaking proxy server?
GiteaMirror commented 2026-04-22 08:52:52 -05:00
Author
Owner
Copy Link

@swiftimundo commented on GitHub (Feb 21, 2025):

I was having this issue on v0.3.3. Saw the PR and redownloaded. On 0.5.11 I don't have this issue anymore. @GhostDog98 upgrade your client.

<!-- gh-comment-id:2675356336 --> @swiftimundo commented on GitHub (Feb 21, 2025): I was having this issue on v0.3.3. Saw the PR and redownloaded. On 0.5.11 I don't have this issue anymore. @GhostDog98 upgrade your client.
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#29716
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?