github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 05:05:37 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #1179] Why is this fixed to localhost? #599

New Issue
Closed
opened 2026-04-12 10:18:41 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-12 10:18:41 -05:00
Owner
Copy Link

Originally created by @oderwat on GitHub (Nov 17, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/1179

I want to run the server on one of my local machines and use if from from other ones in the same private network. I wonder why this (and some other similar software) is only running on localhost? I can rewrite it to use another host. I just wonder why it is how it is?

Originally created by @oderwat on GitHub (Nov 17, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/1179 I want to run the server on one of my local machines and use if from from other ones in the same private network. I wonder why this (and some other similar software) is only running on localhost? I can rewrite it to use another host. I just wonder why it is how it is?
GiteaMirror commented 2026-04-12 10:18:43 -05:00
Author
Owner
Copy Link

@oliverbob commented on GitHub (Nov 17, 2023):

It can actually run on any server. Not just localhost. Here for instance is a customized version. You can run this on any server of your choice.

You simply just rebuild it from source or do however you like with docker and then expose your running installation port to a reverse proxy.

Just do:

OLLAMA_HOST=0.0.0.0 OLLAMA_ORIGINS=* ollama serve

Or with docker, you can specify a port.

docker run -d -v ./ollama -p 11434:11434 -e OLLAMA_ORIGINS="*" --name ollama ollama/ollama

Then from there do a reverse proxy to forward 11434 to https via 443 or standard ssl.

You can verify whether your port is exposed with netstat -tunl

Look for the 0.0.0.0:11434 or :::11434 because that means that your ollama server is exposed to your specified port, otherwise, if you just run it with ollama serve, you'll just have 127.0.0.1:11434, meaning your ollama server is not exposed.

<!-- gh-comment-id:1817099547 --> @oliverbob commented on GitHub (Nov 17, 2023): It can actually run on any server. Not just localhost. Here for instance is a customized version. You can run this on any server of your choice. You simply just rebuild it from source or do however you like with docker and then expose your running installation port to a reverse proxy. Just do: `OLLAMA_HOST=0.0.0.0 OLLAMA_ORIGINS=* ollama serve` Or with docker, you can specify a port. `docker run -d -v ./ollama -p 11434:11434 -e OLLAMA_ORIGINS="*" --name ollama ollama/ollama` Then from there do a reverse proxy to forward 11434 to https via 443 or standard ssl. You can verify whether your port is exposed with netstat -tunl Look for the 0.0.0.0:11434 or :::11434 because that means that your ollama server is exposed to your specified port, otherwise, if you just run it with ollama serve, you'll just have 127.0.0.1:11434, meaning your ollama server is not exposed.
GiteaMirror commented 2026-04-12 10:18:44 -05:00
Author
Owner
Copy Link

@oderwat commented on GitHub (Nov 17, 2023):

OK. I missed the OLLAMA_HOST / OLLAMA_ORIGINS env variables. I just found them in the FAQ. I think this belongs into the documentation.

<!-- gh-comment-id:1817104027 --> @oderwat commented on GitHub (Nov 17, 2023): OK. I missed the OLLAMA_HOST / OLLAMA_ORIGINS env variables. I just found them in the FAQ. I think this belongs into the documentation.
GiteaMirror commented 2026-04-12 10:18:45 -05:00
Author
Owner
Copy Link

@Hi-Angel commented on GitHub (Mar 25, 2025):

Just do:

OLLAMA_HOST=0.0.0.0 OLLAMA_ORIGINS=* ollama serve

Doesn't seem to be working, see:

# OLLAMA_HOST=0.0.0.0 OLLAMA_ORIGINS="*" ollama serve 2>&1 | grep Listening
time=2025-03-25T09:30:13.740+03:00 level=INFO source=routes.go:1238 msg="Listening on 127.0.0.1:11434 (version 0.0.0)"

It says it's listening on 127.0.0.1 instead of 0.0.0.0. And expectedly, nc -v LAN_IP 11434 results in "Connection refused".

That is with 0.5.7 released a few months ago.

<!-- gh-comment-id:2750226524 --> @Hi-Angel commented on GitHub (Mar 25, 2025): > Just do: > > `OLLAMA_HOST=0.0.0.0 OLLAMA_ORIGINS=* ollama serve` Doesn't seem to be working, see: ```haskell # OLLAMA_HOST=0.0.0.0 OLLAMA_ORIGINS="*" ollama serve 2>&1 | grep Listening time=2025-03-25T09:30:13.740+03:00 level=INFO source=routes.go:1238 msg="Listening on 127.0.0.1:11434 (version 0.0.0)" ``` It says it's listening on `127.0.0.1` instead of `0.0.0.0`. And expectedly, `nc -v LAN_IP 11434` results in "Connection refused". That is with `0.5.7` released a few months ago.
GiteaMirror commented 2026-04-12 10:18:45 -05:00
Author
Owner
Copy Link

@Hi-Angel commented on GitHub (Apr 1, 2025):

Found out it may not work on Ubuntu due to a bug snap Ollama package. Reported it the last week to the package maintainer, but got no reply.

A workaround is to use:

OLLAMA_HOST=0.0.0.0 OLLAMA_ORIGINS="*" sh -c 'echo "ollama serve" | snap run --shell ollama'
<!-- gh-comment-id:2769219909 --> @Hi-Angel commented on GitHub (Apr 1, 2025): Found out it may not work on Ubuntu due to a bug `snap` Ollama package. Reported it the last week to the package maintainer, but got no reply. A workaround is to use: ```bash OLLAMA_HOST=0.0.0.0 OLLAMA_ORIGINS="*" sh -c 'echo "ollama serve" | snap run --shell ollama' ```
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
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#599
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?