github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-22 13:42:25 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #9655] Ollama on runpods.io: Network error: Request URL is missing an 'http://' or 'https://' protocol. #6300

New Issue
Closed
opened 2026-04-12 17:45:04 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 17:45:04 -05:00
Owner
Copy Link

Originally created by @itinance on GitHub (Mar 11, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9655

What is the issue?

We run Ollama inside a docker image that ships a FastAPI application and the Ollama service.
It is running on runpods.io using a A40 GPU.

We have set the ENV-Var OLLAMA_HOST to 0.0.0.0 as it is written in their official docs: https://docs.runpod.io/tutorials/pods/run-ollama

However, although it has worked for weeks without interruption, suddenly, we see after a reboot of the pod the following error message:

Network error communicating with Ollama: Request URL is missing an 'http://' or 'https://' protocol.

When SSH into the Pod, we can talk to Ollama both via CLI command and also via HTTP request:

curl http://localhost:11434/api/tags
curl http://0.0.0.0:11434/api/tags

But within the python application, it can't connect anymore this way, it always shows

Network error communicating with Ollama: Request URL is missing an 'http://' or 'https://' protocol. which doesn't come from the python application. It comes from Ollama according to the logs.

What has changed that we can talk to Ollama within a python application on such an environment?

Relevant log output

OS

Ubuntu 22.04

GPU

A40

CPU

No response

Ollama version

0.5.13

Originally created by @itinance on GitHub (Mar 11, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9655 ### What is the issue? We run Ollama inside a docker image that ships a FastAPI application and the Ollama service. It is running on runpods.io using a A40 GPU. We have set the ENV-Var `OLLAMA_HOST` to 0.0.0.0 as it is written in their official docs: https://docs.runpod.io/tutorials/pods/run-ollama However, although it has worked for weeks without interruption, suddenly, we see after a reboot of the pod the following error message: > Network error communicating with Ollama: Request URL is missing an 'http://' or 'https://' protocol. When SSH into the Pod, we can talk to Ollama both via CLI command and also via HTTP request: ``` curl http://localhost:11434/api/tags curl http://0.0.0.0:11434/api/tags ``` But within the python application, it can't connect anymore this way, it always shows `Network error communicating with Ollama: Request URL is missing an 'http://' or 'https://' protocol.` which doesn't come from the python application. It comes from Ollama according to the logs. What has changed that we can talk to Ollama within a python application on such an environment? ### Relevant log output ```shell ``` ### OS Ubuntu 22.04 ### GPU A40 ### CPU _No response_ ### Ollama version 0.5.13
GiteaMirror added the bug label 2026-04-12 17:45:04 -05:00
GiteaMirror commented 2026-04-12 17:45:07 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 11, 2025):

Request URL is missing an 'http://' or 'https://' protocol sounds like an httpx error. It's not something that the ollama server emits.

<!-- gh-comment-id:2715236250 --> @rick-github commented on GitHub (Mar 11, 2025): `Request URL is missing an 'http://' or 'https://' protocol` sounds like an httpx error. It's not something that the ollama server emits.
GiteaMirror commented 2026-04-12 17:45:08 -05:00
Author
Owner
Copy Link

@itinance commented on GitHub (Mar 11, 2025):

I found it :( We tested CursorAI and indeed it changed a file that it shouldn't touch and changed our custom ENV-Var from OLLAMA_API_HOST to OLLAMA_HOST. But we are using both. Because of a requirement of Runpod, we assigned 0.0.0.0 to OLLAMA_HOST, while the OLLAMA_API_HOST contains the full URL for the python service to talk to the ollama service (http://localhost:11434).
We review code, especially when the changes are coming from an AI. However, this change has been undetected, with harsh consequences.

<!-- gh-comment-id:2715468683 --> @itinance commented on GitHub (Mar 11, 2025): I found it :( We tested CursorAI and indeed it changed a file that it shouldn't touch and changed our custom ENV-Var from `OLLAMA_API_HOST` to `OLLAMA_HOST`. But we are using both. Because of a requirement of Runpod, we assigned `0.0.0.0` to OLLAMA_HOST, while the OLLAMA_API_HOST contains the full URL for the python service to talk to the ollama service (`http://localhost:11434`). We review code, especially when the changes are coming from an AI. However, this change has been undetected, with harsh consequences.
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#6300
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?