服务器原先有ollama，想要重装，遇到一系列问题

安装下载连接：https://github.com/ollama/ollama/blob/main/docs/linux.md模型下载链接：https://ollama.com/library/deepseek-r1:1.5b

一、安装新的ollama

在root用户下操作

1.卸载已安装的ollama

# Remove the ollama service:sudo systemctl stop ollama
sudo systemctl disable ollama
sudo rm /etc/systemd/system/ollama.service# Remove the ollama binary from your bin directory (either /usr/local/bin, /usr/bin, or /bin):sudo rm $(which ollama)# Remove the downloaded models and Ollama service user and group:sudo rm -r /usr/share/ollama
sudo userdel ollama
sudo groupdel ollama# Remove installed libraries:sudo rm -rf /usr/local/lib/ollama# Rremove the old libraries
sudo rm -rf /usr/lib/ollama

2.安装

curl -fsSL https://ollama.com/install.sh | sh

该命令国内下载缓慢，采取手工安装方式

Download and extract the package:curl -L https://ollama.com/download/ollama-linux-amd64.tgz -o ollama-linux-amd64.tgz
# 如果在服务器无法下载，直接在本地电脑浏览器打开 https://ollama.com/download/ollama-linux-amd64.tgz 下载文件，然后将文件移到服务器上sudo tar -C /usr -xzf ollama-linux-amd64.tgz# Start Ollama:ollama serve
# 可以正常启动，但是显示日志，所以使用以下命令 
# ollama serve &# In another terminal, verify that Ollama is running:ollama -v

3.设置启动服务

# Adding Ollama as a startup service (recommended)
# Create a user and group for Ollama:sudo useradd -r -s /bin/false -U -m -d /usr/share/ollama ollama
sudo usermod -a -G ollama $(whoami)# Create a service file in /etc/systemd/system/ollama.service:[Unit]
Description=Ollama Service
After=network-online.target[Service]
ExecStart=/usr/bin/ollama serve
User=ollama
Group=ollama
Restart=always
RestartSec=3
Environment="PATH=$PATH"[Install]
WantedBy=multi-user.target# Then start the service:sudo systemctl daemon-reload
sudo systemctl enable ollama

4.对外提供模型服务

便于dify，ragflow平台调用

vim /etc/systemd/system/ollama.service

[Unit]
Description=Ollama Service
After=network-online.target[Service]
ExecStart=/usr/bin/ollama serve
User=ollama
Group=ollama
Restart=always
RestartSec=3Environment="PATH=$PATH"
Environment="OLLAMA_HOST=0.0.0.0"
Environment="OLLAMA_PORT=11434"
Environment="OLLAMA_ORIGINS=*"[Install]
WantedBy=multi-user.target

sudo systemctl daemon-reloadsudo systemctl restart ollama

二、遇到的问题

1.Error: listen tcp 127.0.0.1:11434: bind: address already in us

# 命令
ollama serve
# 问题
Error: listen tcp 127.0.0.1:11434: bind: address already in us

查找占用 11434 端口的进程并终止进程

# 查看
sudo lsof -i :11434
# 终止
kill -9 <PID>
# 重新启动
ollama serve

若问题没有解决

root@user-NF5280M6:/home/lzm/Downloads# sudo lsof -i :11434
COMMAND     PID USER   FD   TYPE    DEVICE SIZE/OFF NODE NAME
ollama  3558609 root    3u  IPv4 274643309      0t0  TCP localhost:11434 (LISTEN)
root@user-NF5280M6:/home/lzm/Downloads# kill -9 3558609
root@user-NF5280M6:/home/lzm/Downloads# sudo lsof -i :11434
COMMAND     PID USER   FD   TYPE    DEVICE SIZE/OFF NODE NAME
ollama  3563660 root    3u  IPv4 274711630      0t0  TCP localhost:11434 (LISTEN)

ps aux | grep ollama# 强制终止所有 Ollama 进程
pkill ollama

2.下载模型超时，无法下载

ollama run deepseek-r1:1.5b
[GIN] 2025/06/27 - 10:42:02 | 200 |      42.146µs |       127.0.0.1 | HEAD     "/"
[GIN] 2025/06/27 - 10:42:02 | 404 |     209.992µs |       127.0.0.1 | POST     "/api/show"
pulling manifest ⠏ time=2025-06-27T10:42:12.655+08:00 level=INFO source=images.go:713 msg="request failed: Get \"https://registry.ollama.ai/v2/library/deepseek-r1/manifests/1.5b\": dial tcp: lookup registry.ollama.ai on 127.0.0.53:53: read udp 127.0.0.1:36213->127.0.0.53:53: i/o timeout"
[GIN] 2025/06/27 - 10:42:12 | 200 |  10.00749667s |       127.0.0.1 | POST     "/api/pull"
pulling manifest
Error: pull model manifest: Get "https://registry.ollama.ai/v2/library/deepseek-r1/manifests/1.5b": dial tcp: lookup registry.ollama.ai on 127.0.0.53:53: read udp 127.0.0.1:36213->127.0.0.53:53: i/o timeout

选用国内镜像

阿里云：https://registry.ollama.ai
DeepSeek 官方镜像：https://ollama.deepseek.com
浙江大学镜像站：https://ollama.zju.edu.cn
魔搭社区：https://ollama.modelscope.cn

mkdir -p ~/.ollama
cat << EOF > ~/.ollama/config.json
{"registry": {"mirrors": {"registry.ollama.ai": "https://ollama.deepseek.com"}}
}
EOF

sudo systemctl restart ollamasudo systemctl status ollama

参考链接：Ollma通过国内源实现模型本地化部署_ollama国内镜像源-CSDN博客

通义

三、运行情况展示

1.服务器端（UBUNTU OLLAMA）

2.平台端（DIFY）