[Bug] Nginx 無法反代本地的GPT4模型 about chatgpt-next-web HOT 5 CLOSED

Bot detected the issue body's language is not English, translate it automatically.

Title: [Bug] Nginx cannot reverse the local GPT4 model

Bug Description

In order to solve the CORS problem on my server, I made the following settings:
Reverse app.mydomain.com/gui to nextweb (front-end GUI), nginx settings are as follows:
location ^~ /gui {
proxy_pass http://172.17.0.4:3000/;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header REMOTE-HOST $remote_addr;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
proxy_set_header X-Forwarded-Proto $scheme;
proxy_http_version 1.1;
add_header X-Cache $upstream_cache_status;
add_header Strict-Transport-Security "max-age=31536000";
add_header Cache-Control no-cache;
}

The root at app.mydomain.com is the endpoint that backends to my backend model.
location ^~ / {
proxy_pass http://172.17.0.9:7077;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header REMOTE-HOST $remote_addr;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
proxy_set_header X-Forwarded-Proto $scheme;
proxy_http_version 1.1;
add_header X-Cache $upstream_cache_status;
add_header Strict-Transport-Security "max-age=31536000";
add_header Cache-Control no-cache;
}

I found that the next request after the first one, requesting app.mydomain.com instead of app.mydomain.com/gui, resulted in a 404 error. How can I ensure that the next request is still sent to app.mydomain.com/gui and that nginx proxies it to the correct internal IP and port?
During deployment, base url = app.mydomain.com/gui is written.

Steps to Reproduce

I used docker to deploy nextweb: docker compose file is as follows:

networks:
    1panel-network:
        external: true
services:
    chatgpt-next-web:
        container_name: ${CONTAINER_NAME}
        deploy:
            resources:
                limits:
                    cpus: ${CPUS}
                    memory: ${MEMORY_LIMIT}
        environment:
            - OPENAI_API_KEY=${API_KEY}
            - CODE=${SECRET_KEY}
            - PROXY_URL=${PROXY}
            - BASE_URL=${API_BASE_URL}
        image: yidadaa/chatgpt-next-web:v2.12.2
        labels:
            createdBy: Apps
        networks:
            - 1panel-network
        ports:
            - ${HOST_IP}:${PANEL_APP_PORT_HTTP}:3000
        restart: always

Then just set nginx
Configure Nginx to reverse proxy to the front-end GUI: In the Nginx configuration file, set a location block for app.mydomain.com/gui and reverse-proxy it to the front-end GUI (located at http://172.17. 0.4:3000/). The following are the relevant Nginx settings:

location ^~ /gui {
proxy_pass http://172.17.0.4:3000/;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header REMOTE-HOST $remote_addr;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
proxy_set_header X-Forwarded-Proto $scheme;
proxy_http_version 1.1;
add_header X-Cache $upstream_cache_status;
add_header Strict-Transport-Security "max-age=31536000";
add_header Cache-Control no-cache;
}

Configure Nginx reverse proxy to the backend: In the Nginx configuration file, set a location block for the root directory of app.mydomain.com and reverse proxy it to the backend (located at http://172.17. 0.9:7077). The following are the relevant Nginx settings:

location ^~ / {
proxy_pass http://172.17.0.9:7077;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header REMOTE-HOST $remote_addr;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
proxy_set_header X-Forwarded-Proto $scheme;
proxy_http_version 1.1;
add_header X-Cache $upstream_cache_status;
add_header Strict-Transport-Security "max-age=31536000";
add_header Cache-Control no-cache;
}

Expected Behavior

Nextweb is functioning normally

Screenshots

Deployment Method

• Docker
• Vercel
• Server

Desktop OS

Windows 11

Desktop Browser

Chrome

Desktop Browser Version

125.0.6422.41

Smartphone Device

No response

Smartphone OS

No response

Smartphone Browser

No response

Smartphone Browser Version

No response

Additional Logs

No response

from chatgpt-next-web.

from chatgpt-next-web.

Bot detected the issue body's language is not English, translate it automatically.

https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/assets/40236765/97eb28f7-bfdc-48cc-8766-204ce63c4a36

On Thu, 9 May, 2024 at 1:59PM, Robert Black @.***>
wrote:

Bug Description

In order to solve the CORS problem on my server, I made the following settings:
Reverse app.mydomain.com/gui to nextweb (front-end GUI), nginx settings are as follows:
location ^~ /gui {
proxy_pass http://172.17.0.4:3000/;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header REMOTE-HOST $remote_addr;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
proxy_set_header X-Forwarded-Proto $scheme;
proxy_http_version 1.1;
add_header X-Cache $upstream_cache_status;
add_header Strict-Transport-Security "max-age=31536000";
add_header Cache-Control no-cache;
}

The root at app.mydomain.com is the endpoint that backends to my backend model.
location ^~ / {
proxy_pass http://172.17.0.9:7077;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header REMOTE-HOST $remote_addr;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
proxy_set_header X-Forwarded-Proto $scheme;
proxy_http_version 1.1;
add_header X-Cache $upstream_cache_status;
add_header Strict-Transport-Security "max-age=31536000";
add_header Cache-Control no-cache;
}

I found that the next request after the first request, requesting app.mydomain.com instead of app.mydomain.com/gui, resulted in a 404
Error. How can I ensure that the next request is still sent to app.mydomain.com/gui and that nginx proxies it to the correct internal IP and port?
During deployment, base url = app.mydomain.com/gui is written.
Steps to Reproduce

I used docker to deploy nextweb: docker compose file is as follows:

networks:
1panel-network:
external: true
services:
chatgpt-next-web:
container_name: ${CONTAINER_NAME}
deploy:
resources:
limits:
cpus: ${CPUS}
memory: ${MEMORY_LIMIT}
environment:

• OPENAI_API_KEY=${API_KEY}
• CODE=${SECRET_KEY}
• PROXY_URL=${PROXY}
• BASE_URL=${API_BASE_URL}
  image: yidadaa/chatgpt-next-web:v2.12.2
  labels:
  createdBy: Apps
  networks:
  -1panel-network
  ports:
• ${HOST_IP}:${PANEL_APP_PORT_HTTP}:3000
  restart: always

Then set nginx
Configure Nginx reverse proxy to the front-end GUI: In the Nginx configuration file, set a location for app.mydomain.com/gui
block and reverse-proxy it to the front-end GUI at http://172.17.0.4:3000/. The following are the relevant Nginx settings:

location ^~ /gui {
proxy_pass http://172.17.0.4:3000/;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header REMOTE-HOST $remote_addr;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
proxy_set_header X-Forwarded-Proto $scheme;
proxy_http_version 1.1;
add_header X-Cache $upstream_cache_status;
add_header Strict-Transport-Security "max-age=31536000";
add_header Cache-Control no-cache;
}

Configure Nginx reverse proxy to the backend: In the Nginx configuration file, set a location for the root directory of app.mydomain.com
block and reverse-proxy it to the backend at http://172.17.0.9:7077. The following are the relevant Nginx settings:

location ^~ / {
proxy_pass http://172.17.0.9:7077;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header REMOTE-HOST $remote_addr;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
proxy_set_header X-Forwarded-Proto $scheme;
proxy_http_version 1.1;
add_header X-Cache $upstream_cache_status;
add_header Strict-Transport-Security "max-age=31536000";
add_header Cache-Control no-cache;
}
Expected Behavior

Nextweb is functioning normally
Screenshots

Screenshot.2024-05-09.155434.png (view on web)
https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/assets/168748301/32dd395b-6b7e-4c1d-b7a6-3a061db075e2
Screenshot.2024-05-09.155631.png (view on web)
https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/assets/168748301/1c3f410a-5c0c-4019-a1c6-11927f0227a0
Deployment Method

• Docker
  -Vercel
  -Server

Desktop OS

Windows11
Desktop Browser

Chrome
Desktop Browser Version

125.0.6422.41
Smartphone Device

No response
Smartphone OS

No response
Smartphone Browser

No response
Smartphone Browser Version

No response
Additional Logs

No response

—
Reply to this email directly, view it on GitHub
#4645, or
unsubscribe
https://github.com/notifications/unsubscribe-auth/BGQMJGEHRWDQJ3IAP22RGCTZBMUEVAVCNFSM6AAAAABHOLMKBKVHI2DSMVQWIX3LMV43ASLTON2WKOZSGI4DOMJSGM4DEOI
.
You are receiving this because you are subscribed to this thread.Message
ID: @.***>

from chatgpt-next-web.

你这里报错静态资源 not found ，原因是由于静态资源打包到根目录了，因此，html中的静态资源地址指向的是根目录，而不是/gui，如果在不修改nextchat 静态资源打包目录的情况下，你需要对静态资源的资源路径也进行rewrite才能解决这个问题

from chatgpt-next-web.

Bot detected the issue body's language is not English, translate it automatically.

The error reported here that static resources are not found is because the static resources are packaged in the root directory. Therefore, the static resource address in HTML points to the root directory, not /gui. If the nextchat static resource packaging directory is not modified, , you need to rewrite the resource path of the static resources to solve this problem.

from chatgpt-next-web.