github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 13:19:40 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #5940] 配置命令都是什么啊 - how to keep models loaded for 24h #3706

New Issue
Closed
opened 2026-04-12 14:31:11 -05:00 by GiteaMirror · 13 comments
GiteaMirror commented 2026-04-12 14:31:11 -05:00
Owner
Copy Link

Originally created by @673092756 on GitHub (Jul 25, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/5940

Originally assigned to: @dhiltgen on GitHub.

我想要让ollama在服务器运行24小时不掉机，我查了所有的相关资料没有找到，希望官方出一些ollama所有的配置命令手册， 这样可以更好的使用，感谢

Originally created by @673092756 on GitHub (Jul 25, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/5940 Originally assigned to: @dhiltgen on GitHub. 我想要让ollama在服务器运行24小时不掉机，我查了所有的相关资料没有找到，希望官方出一些ollama所有的配置命令手册， 这样可以更好的使用，感谢
GiteaMirror added the question label 2026-04-12 14:31:11 -05:00
GiteaMirror commented 2026-04-12 14:31:12 -05:00
Author
Owner
Copy Link

@liuia commented on GitHub (Jul 25, 2024):

https://github.com/ollama/ollama/blob/main/docs/linux.md#adding-ollama-as-a-startup-service-recommended

<!-- gh-comment-id:2249724144 --> @liuia commented on GitHub (Jul 25, 2024): https://github.com/ollama/ollama/blob/main/docs/linux.md#adding-ollama-as-a-startup-service-recommended
GiteaMirror commented 2026-04-12 14:31:12 -05:00
Author
Owner
Copy Link

@673092756 commented on GitHub (Jul 25, 2024):

https://github.com/ollama/ollama/blob/main/docs/linux.md#adding-ollama-as-a-startup-service-recommended

抱歉，不知道是我的原因还是什么，我在这篇文档里面没有看到类似的操作手册，只看到如何安装

<!-- gh-comment-id:2249828647 --> @673092756 commented on GitHub (Jul 25, 2024): > https://github.com/ollama/ollama/blob/main/docs/linux.md#adding-ollama-as-a-startup-service-recommended 抱歉，不知道是我的原因还是什么，我在这篇文档里面没有看到类似的操作手册，只看到如何安装
GiteaMirror commented 2026-04-12 14:31:13 -05:00
Author
Owner
Copy Link

@handsomejustin commented on GitHub (Jul 25, 2024):

难道你是windows系统？下载ollama windows版本也可以保证24h不掉线的。

<!-- gh-comment-id:2250022765 --> @handsomejustin commented on GitHub (Jul 25, 2024): 难道你是windows系统？下载ollama windows版本也可以保证24h不掉线的。
GiteaMirror commented 2026-04-12 14:31:13 -05:00
Author
Owner
Copy Link

@eust-w commented on GitHub (Jul 25, 2024):

使用ollama serve即可

<!-- gh-comment-id:2251553961 --> @eust-w commented on GitHub (Jul 25, 2024): 使用ollama serve即可
GiteaMirror commented 2026-04-12 14:31:14 -05:00
Author
Owner
Copy Link

@673092756 commented on GitHub (Jul 26, 2024):

难道你是windows系统？下载ollama windows版本也可以保证24h不掉线的。

我是Linux系统

<!-- gh-comment-id:2251773709 --> @673092756 commented on GitHub (Jul 26, 2024): > 难道你是windows系统？下载ollama windows版本也可以保证24h不掉线的。 我是Linux系统
GiteaMirror commented 2026-04-12 14:31:14 -05:00
Author
Owner
Copy Link

@673092756 commented on GitHub (Jul 26, 2024):

使用ollama serve即可

我想要的不是ollama不掉线，想要的是运行的模型不掉线，启动一次就行了，后续可以更方便访问

<!-- gh-comment-id:2251775372 --> @673092756 commented on GitHub (Jul 26, 2024): > 使用ollama serve即可 我想要的不是ollama不掉线，想要的是运行的模型不掉线，启动一次就行了，后续可以更方便访问
GiteaMirror commented 2026-04-12 14:31:14 -05:00
Author
Owner
Copy Link

@jlvihv commented on GitHub (Jul 26, 2024):

You can send a message to the Ollama model at regular intervals, such as every 1 minute, similar to a heartbeat mechanism, to prevent the model from disconnecting.

<!-- gh-comment-id:2251840062 --> @jlvihv commented on GitHub (Jul 26, 2024): You can send a message to the Ollama model at regular intervals, such as every 1 minute, similar to a heartbeat mechanism, to prevent the model from disconnecting.
GiteaMirror commented 2026-04-12 14:31:15 -05:00
Author
Owner
Copy Link

@eust-w commented on GitHub (Jul 26, 2024):

使用ollama serve即可

我想要的不是ollama不掉线，想要的是运行的模型不掉线，启动一次就行了，后续可以更方便访问

你可以用keepalive参数 最大持续时间可以设置为 2562047h47m16.854775807s

我想这已经满足了你的需求

<!-- gh-comment-id:2251859357 --> @eust-w commented on GitHub (Jul 26, 2024): > > 使用ollama serve即可 > > 我想要的不是ollama不掉线，想要的是运行的模型不掉线，启动一次就行了，后续可以更方便访问 你可以用keepalive参数 最大持续时间可以设置为 2562047h47m16.854775807s 我想这已经满足了你的需求
GiteaMirror commented 2026-04-12 14:31:15 -05:00
Author
Owner
Copy Link

@673092756 commented on GitHub (Jul 26, 2024):

You can send a message to the Ollama model at regular intervals, such as every 1 minute, similar to a heartbeat mechanism, to prevent the model from disconnecting.

Do you mean setting it inside the configuration item ollama.server or writing a program?

<!-- gh-comment-id:2251870903 --> @673092756 commented on GitHub (Jul 26, 2024): > You can send a message to the Ollama model at regular intervals, such as every 1 minute, similar to a heartbeat mechanism, to prevent the model from disconnecting. Do you mean setting it inside the configuration item ollama.server or writing a program?
GiteaMirror commented 2026-04-12 14:31:15 -05:00
Author
Owner
Copy Link

@673092756 commented on GitHub (Jul 26, 2024):

使用ollama serve即可

我想要的不是ollama不掉线，想要的是运行的模型不掉线，启动一次就行了，后续可以更方便访问

你可以用keepalive参数 最大持续时间可以设置为 2562047h47m16.854775807s

我想这已经满足了你的需求

image
这是我设置的，但是呢，我这里不起作用，这两种方式我都试过了，模型没起来，是我设置的问题还是说写错了。你那里设置之后启动成功了吗，或者说生效了吗？

<!-- gh-comment-id:2251872360 --> @673092756 commented on GitHub (Jul 26, 2024): > > > 使用ollama serve即可 > > > > > > 我想要的不是ollama不掉线，想要的是运行的模型不掉线，启动一次就行了，后续可以更方便访问 > > 你可以用keepalive参数 最大持续时间可以设置为 2562047h47m16.854775807s > > 我想这已经满足了你的需求 ![image](https://github.com/user-attachments/assets/8e256f2f-40e5-45f1-9f30-fdb780c124cf) 这是我设置的，但是呢，我这里不起作用，这两种方式我都试过了，模型没起来，是我设置的问题还是说写错了。你那里设置之后启动成功了吗，或者说生效了吗？
GiteaMirror commented 2026-04-12 14:31:15 -05:00
Author
Owner
Copy Link

@eust-w commented on GitHub (Jul 26, 2024):

使用ollama serve即可

我想要的不是ollama不掉线，想要的是运行的模型不掉线，启动一次就行了，后续可以更方便访问

你可以用keepalive参数 最大持续时间可以设置为 2562047h47m16.854775807s
我想这已经满足了你的需求

image 这是我设置的，但是呢，我这里不起作用，这两种方式我都试过了，模型没起来，是我设置的问题还是说写错了。你那里设置之后启动成功了吗，或者说生效了吗？

像这样运行
ollama run llama3.1 --keepalive 2562047h47m16.854775807s 你就会看到model loaded 是forever的

<!-- gh-comment-id:2252171674 --> @eust-w commented on GitHub (Jul 26, 2024): > > > > 使用ollama serve即可 > > > > > > > > > 我想要的不是ollama不掉线，想要的是运行的模型不掉线，启动一次就行了，后续可以更方便访问 > > > > > > 你可以用keepalive参数 最大持续时间可以设置为 2562047h47m16.854775807s > > 我想这已经满足了你的需求 > > ![image](https://private-user-images.githubusercontent.com/138642126/352350865-8e256f2f-40e5-45f1-9f30-fdb780c124cf.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3MjE5ODAyNDUsIm5iZiI6MTcyMTk3OTk0NSwicGF0aCI6Ii8xMzg2NDIxMjYvMzUyMzUwODY1LThlMjU2ZjJmLTQwZTUtNDVmMS05ZjMwLWZkYjc4MGMxMjRjZi5wbmc_WC1BbXotQWxnb3JpdGhtPUFXUzQtSE1BQy1TSEEyNTYmWC1BbXotQ3JlZGVudGlhbD1BS0lBVkNPRFlMU0E1M1BRSzRaQSUyRjIwMjQwNzI2JTJGdXMtZWFzdC0xJTJGczMlMkZhd3M0X3JlcXVlc3QmWC1BbXotRGF0ZT0yMDI0MDcyNlQwNzQ1NDVaJlgtQW16LUV4cGlyZXM9MzAwJlgtQW16LVNpZ25hdHVyZT0yZWQxOWYyNDVjNDVlZmRhZmZjY2E3YjU5YjU0M2IwNmMyOWQyMzY4MzUwNDExNDY2OGVkNjRhMzdhMGQwZDM0JlgtQW16LVNpZ25lZEhlYWRlcnM9aG9zdCZhY3Rvcl9pZD0wJmtleV9pZD0wJnJlcG9faWQ9MCJ9.FztlefxEqcMKz5NqXEZxj_GC0-ZdGxb2lyupmSrBNRE) 这是我设置的，但是呢，我这里不起作用，这两种方式我都试过了，模型没起来，是我设置的问题还是说写错了。你那里设置之后启动成功了吗，或者说生效了吗？ 像这样运行 `ollama run llama3.1 --keepalive 2562047h47m16.854775807s ` 你就会看到model loaded 是forever的
GiteaMirror commented 2026-04-12 14:31:15 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Jul 26, 2024):

In addition to specifying the --keepalive flag to the run command, you can adjust the default for the server with OLLAMA_KEEP_ALIVE on the server. On linux you configure the server through the systemd file described here https://github.com/ollama/ollama/blob/main/docs/faq.md#setting-environment-variables-on-linux

<!-- gh-comment-id:2253491732 --> @dhiltgen commented on GitHub (Jul 26, 2024): In addition to specifying the `--keepalive` flag to the run command, you can adjust the default for the server with OLLAMA_KEEP_ALIVE on the server. On linux you configure the server through the systemd file described here https://github.com/ollama/ollama/blob/main/docs/faq.md#setting-environment-variables-on-linux
GiteaMirror commented 2026-04-12 14:31:16 -05:00
Author
Owner
Copy Link

@673092756 commented on GitHub (Jul 29, 2024):

In addition to specifying the --keepalive flag to the run command, you can adjust the default for the server with OLLAMA_KEEP_ALIVE on the server. On linux you configure the server through the systemd file described here https://github.com/ollama/ollama/blob/main/docs/faq.md#setting-environment-variables-on-linux

This method works, thank you so much, it didn't work before should be my version.

<!-- gh-comment-id:2255052383 --> @673092756 commented on GitHub (Jul 29, 2024): > In addition to specifying the `--keepalive` flag to the run command, you can adjust the default for the server with OLLAMA_KEEP_ALIVE on the server. On linux you configure the server through the systemd file described here https://github.com/ollama/ollama/blob/main/docs/faq.md#setting-environment-variables-on-linux This method works, thank you so much, it didn't work before should be my version.
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 question
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#3706
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?