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 #3639] MacOS not saving 0.0.0.0 between hardware restarts #48755

New Issue
Closed
opened 2026-04-28 09:12:45 -05:00 by GiteaMirror · 12 comments
GiteaMirror commented 2026-04-28 09:12:45 -05:00
Owner
Copy Link

Originally created by @gwthompson on GitHub (Apr 14, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/3639

What is the issue?

When I set launchctl setenv OLLAMA_HOST "0.0.0.0" and restart the Ollama ap everything works as expected and I can access the API from other devices on my network.

However when I reboot my Mac the OLLAMA_HOST reverts back to 127.0.0.1 and I have to run launchctl setenv OLLAMA_HOST "0.0.0.0" again.

What did you expect to see?

When I run launchctl setenv OLLAMA_HOST "0.0.0.0" and restart my Mac I expected the OLLAMA_HOST to still be set at 0.0.0.0

Steps to reproduce

  1. Run launchctl setenv OLLAMA_HOST "0.0.0.0"
  2. Restart Ollama App
  3. Check the logs to make sure it says "Listening on [::]:11434 (version 0.1.31)"
  4. Reboot the Mac
  5. Check the logs again and it now says "Listening on 127.0.0.1:11434 (version 0.1.31)"

Are there any recent changes that introduced the issue?

No response

OS

macOS

Architecture

arm64

Platform

No response

Ollama version

0.1.31

GPU

Apple

GPU info

No response

CPU

Apple

Other software

No response

Originally created by @gwthompson on GitHub (Apr 14, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/3639 ### What is the issue? When I set launchctl setenv OLLAMA_HOST "0.0.0.0" and restart the Ollama ap everything works as expected and I can access the API from other devices on my network. However when I reboot my Mac the OLLAMA_HOST reverts back to 127.0.0.1 and I have to run launchctl setenv OLLAMA_HOST "0.0.0.0" again. ### What did you expect to see? When I run launchctl setenv OLLAMA_HOST "0.0.0.0" and restart my Mac I expected the OLLAMA_HOST to still be set at 0.0.0.0 ### Steps to reproduce 1. Run launchctl setenv OLLAMA_HOST "0.0.0.0" 2. Restart Ollama App 3. Check the logs to make sure it says "Listening on [::]:11434 (version 0.1.31)" 4. Reboot the Mac 5. Check the logs again and it now says "Listening on 127.0.0.1:11434 (version 0.1.31)" ### Are there any recent changes that introduced the issue? _No response_ ### OS macOS ### Architecture arm64 ### Platform _No response_ ### Ollama version 0.1.31 ### GPU Apple ### GPU info _No response_ ### CPU Apple ### Other software _No response_
GiteaMirror added the bugmacos labels 2026-04-28 09:12:45 -05:00
GiteaMirror commented 2026-04-28 09:12:48 -05:00
Author
Owner
Copy Link

@arsaboo commented on GitHub (Apr 14, 2024):

That is correct...OLLAMA_HOST needs to be set after every boot. It is the expected behavior.

<!-- gh-comment-id:2054121854 --> @arsaboo commented on GitHub (Apr 14, 2024): That is correct...`OLLAMA_HOST` needs to be set after every boot. It is the expected behavior.
GiteaMirror commented 2026-04-28 09:12:48 -05:00
Author
Owner
Copy Link

@dahjson commented on GitHub (Apr 14, 2024):

I think we could improve the user experience by including some user configuration settings directly within the Ollama app for connecting to websites or Chrome extensions. That way upon a restart these configurations can be reapplied automatically.

https://github.com/ollama/ollama/issues/3598

<!-- gh-comment-id:2054132061 --> @dahjson commented on GitHub (Apr 14, 2024): I think we could improve the user experience by including some user configuration settings directly within the Ollama app for connecting to websites or Chrome extensions. That way upon a restart these configurations can be reapplied automatically. https://github.com/ollama/ollama/issues/3598
GiteaMirror commented 2026-04-28 09:12:49 -05:00
Author
Owner
Copy Link

@gwthompson commented on GitHub (Apr 14, 2024):

In my particular scenario I am running the Ollama app on a standalone Mac mini and then accessing the API from other devices on both my internal network and my Tailscale network so ideally if the mini reboots for any reason it would have the 0.0.0.0 set properly.

I'm going to look into whether using a cron job to set the OLLAMA_HOST to 0.0.0.0 on reboot would solve my particular issue.

By the way OLLAMA is the best thing I've used in my LLM journey! Great work!

<!-- gh-comment-id:2054145466 --> @gwthompson commented on GitHub (Apr 14, 2024): In my particular scenario I am running the Ollama app on a standalone Mac mini and then accessing the API from other devices on both my internal network and my Tailscale network so ideally if the mini reboots for any reason it would have the 0.0.0.0 set properly. I'm going to look into whether using a cron job to set the OLLAMA_HOST to 0.0.0.0 on reboot would solve my particular issue. By the way OLLAMA is the best thing I've used in my LLM journey! Great work!
GiteaMirror commented 2026-04-28 09:12:49 -05:00
Author
Owner
Copy Link

@lin16303 commented on GitHub (May 22, 2024):

With a script open in the Script Editor app on your Mac, choose File > Export. Enter a name for the app. To save your app in a different location than shown, click the disclosure triangle, then select a folder. Click the File Format pop-up menu, then choose Application. Put the Application in the Open at login and get rid of Ollama

do shell script "launchctl setenv OLLAMA_HOST 0.0.0.0"
do shell script "OLLAMA_NUM_PARALLEL=2"
do shell script "OLLAMA_MAX_LOADED_MODELS=2"
do shell script "OLLAMA_KEEP_ALIVE=24h"
do shell script "open /Applications/ollama.app"

<!-- gh-comment-id:2124218699 --> @lin16303 commented on GitHub (May 22, 2024): With a script open in the Script Editor app on your Mac, choose File > Export. Enter a name for the app. To save your app in a different location than shown, click the disclosure triangle, then select a folder. Click the File Format pop-up menu, then choose Application. Put the Application in the Open at login and get rid of Ollama do shell script "launchctl setenv OLLAMA_HOST 0.0.0.0" do shell script "OLLAMA_NUM_PARALLEL=2" do shell script "OLLAMA_MAX_LOADED_MODELS=2" do shell script "OLLAMA_KEEP_ALIVE=24h" do shell script "open /Applications/ollama.app"
GiteaMirror commented 2026-04-28 09:12:50 -05:00
Author
Owner
Copy Link

@lizhen1412 commented on GitHub (Jun 2, 2024):

Is there a configuration file that can be changed to specify 0.0.0.0 directly?

<!-- gh-comment-id:2143750955 --> @lizhen1412 commented on GitHub (Jun 2, 2024): Is there a configuration file that can be changed to specify 0.0.0.0 directly?
GiteaMirror commented 2026-04-28 09:12:52 -05:00
Author
Owner
Copy Link

@tkoenig89 commented on GitHub (Jun 12, 2024):

For anyone coming here (like i did) because other settings were also reset after a reboot. I tried @lin16303 approach, which only worked half, so i modified it to have ollama be startet with keepAlive etc. configured:

do shell script "launchctl setenv OLLAMA_NUM_PARALLEL 2"
do shell script "launchctl setenv OLLAMA_MAX_LOADED_MODELS 2"
do shell script "launchctl setenv OLLAMA_KEEP_ALIVE 24h"
do shell script "open /Applications/ollama.app"

Setting the variables without launchctl did not work for me.

<!-- gh-comment-id:2162666711 --> @tkoenig89 commented on GitHub (Jun 12, 2024): For anyone coming here (like i did) because other settings were also reset after a reboot. I tried @lin16303 approach, which only worked half, so i modified it to have ollama be startet with keepAlive etc. configured: ```bash do shell script "launchctl setenv OLLAMA_NUM_PARALLEL 2" do shell script "launchctl setenv OLLAMA_MAX_LOADED_MODELS 2" do shell script "launchctl setenv OLLAMA_KEEP_ALIVE 24h" do shell script "open /Applications/ollama.app" ``` Setting the variables without launchctl did not work for me.
GiteaMirror commented 2026-04-28 09:12:53 -05:00
Author
Owner
Copy Link

@lizhen1412 commented on GitHub (Oct 6, 2024):

For anyone coming here (like i did) because other settings were also reset after a reboot. I tried @lin16303 approach, which only worked half, so i modified it to have ollama be startet with keepAlive etc. configured:

do shell script "launchctl setenv OLLAMA_NUM_PARALLEL 2"
do shell script "launchctl setenv OLLAMA_MAX_LOADED_MODELS 2"
do shell script "launchctl setenv OLLAMA_KEEP_ALIVE 24h"
do shell script "open /Applications/ollama.app"

Setting the variables without launchctl did not work for me.

Have you tried reinstalling it? In the new version, it seems that the OLLAMA_HOST variable can be configured only with sudo startup.

<!-- gh-comment-id:2395459970 --> @lizhen1412 commented on GitHub (Oct 6, 2024): > For anyone coming here (like i did) because other settings were also reset after a reboot. I tried @lin16303 approach, which only worked half, so i modified it to have ollama be startet with keepAlive etc. configured: > > ```shell > do shell script "launchctl setenv OLLAMA_NUM_PARALLEL 2" > do shell script "launchctl setenv OLLAMA_MAX_LOADED_MODELS 2" > do shell script "launchctl setenv OLLAMA_KEEP_ALIVE 24h" > do shell script "open /Applications/ollama.app" > ``` > > Setting the variables without launchctl did not work for me. Have you tried reinstalling it? In the new version, it seems that the OLLAMA_HOST variable can be configured only with sudo startup.
GiteaMirror commented 2026-04-28 09:12:53 -05:00
Author
Owner
Copy Link

@lin16303 commented on GitHub (Oct 7, 2024):

ollama.app.zip

For anyone coming here (like i did) because other settings were also reset after a reboot. I tried @lin16303 approach, which only worked half, so i modified it to have ollama be startet with keepAlive etc. configured:

do shell script "launchctl setenv OLLAMA_NUM_PARALLEL 2"
do shell script "launchctl setenv OLLAMA_MAX_LOADED_MODELS 2"
do shell script "launchctl setenv OLLAMA_KEEP_ALIVE 24h"
do shell script "open /Applications/ollama.app"

Setting the variables without launchctl did not work for me.

Have you tried reinstalling it? In the new version, it seems that the OLLAMA_HOST variable can be configured only with sudo startup.

ollama.app.zip
work with 0.3.12

<!-- gh-comment-id:2395761581 --> @lin16303 commented on GitHub (Oct 7, 2024): [ollama.app.zip](https://github.com/user-attachments/files/17271934/ollama.app.zip) > > For anyone coming here (like i did) because other settings were also reset after a reboot. I tried @lin16303 approach, which only worked half, so i modified it to have ollama be startet with keepAlive etc. configured: > > ```shell > > do shell script "launchctl setenv OLLAMA_NUM_PARALLEL 2" > > do shell script "launchctl setenv OLLAMA_MAX_LOADED_MODELS 2" > > do shell script "launchctl setenv OLLAMA_KEEP_ALIVE 24h" > > do shell script "open /Applications/ollama.app" > > ``` > > > > > > > > > > > > > > > > > > > > > > > > Setting the variables without launchctl did not work for me. > > Have you tried reinstalling it? In the new version, it seems that the OLLAMA_HOST variable can be configured only with sudo startup. [ollama.app.zip](https://github.com/user-attachments/files/17271937/ollama.app.zip) work with 0.3.12
GiteaMirror commented 2026-04-28 09:12:55 -05:00
Author
Owner
Copy Link

@jfbloom22 commented on GitHub (Apr 1, 2025):

this worked for me. Any idea how to set the default context-length?

<!-- gh-comment-id:2769259127 --> @jfbloom22 commented on GitHub (Apr 1, 2025): this worked for me. Any idea how to set the default context-length?
GiteaMirror commented 2026-04-28 09:12:56 -05:00
Author
Owner
Copy Link

@lizhen1412 commented on GitHub (Apr 1, 2025):

this worked for me. Any idea how to set the default context-length?

It should be in the source code.

<!-- gh-comment-id:2769265099 --> @lizhen1412 commented on GitHub (Apr 1, 2025): > this worked for me. Any idea how to set the default context-length? It should be in the source code.
GiteaMirror commented 2026-04-28 09:12:57 -05:00
Author
Owner
Copy Link

@jcastro commented on GitHub (Apr 1, 2025):

It's quite strange that we need to use a script to set this environment. Shouldn't there be a config file or something the app can read when starting?

<!-- gh-comment-id:2769393035 --> @jcastro commented on GitHub (Apr 1, 2025): It's quite strange that we need to use a script to set this environment. Shouldn't there be a config file or something the app can read when starting?
GiteaMirror commented 2026-04-28 09:12:57 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Jul 1, 2025):

I think we can close this out now that 0.9.4 has a settings UI to cover this capability which persists.

Image
<!-- gh-comment-id:3025843473 --> @dhiltgen commented on GitHub (Jul 1, 2025): I think we can close this out now that 0.9.4 has a settings UI to cover this capability which persists. <img width="650" alt="Image" src="https://github.com/user-attachments/assets/6656d225-f20d-404d-ae61-1b504f358977" />
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 macos
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#48755
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?