alexsun/compose-anything
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: add frps/frpc/mineru-sgalng

Browse Source
This commit is contained in:
Sun-ZhenXing
2025-09-24 09:28:28 +08:00
parent 79ad907def
commit 232517b58f
11 changed files with 279 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
README.md
View File

@@ -4,10 +4,12 @@ Compose Anything helps users quickly deploy various services by providing a set
## Supported Services
| Service | Version |
| ---------------------------- | ------- |
| [Redis](./src/redis) | 8.2.1 |
| [MinerU v2](./src/mineru-v2) | 2.5.3 |
| Service | Version |
| ---------------------------------------- | ------- |
| [Redis](./src/redis) | 8.2.1 |
| [MinerU v2](./src/mineru-v2) | 2.5.3 |
| [Bifrost Gateway](./src/bifrost-gateway) | 1.2.15 |
| [Docker Registry](./src/docker-registry) | 3.0.0 |
## Guidelines

10
README.zh.md
View File

@@ -4,10 +4,12 @@ Compose Anything 通过提供一组高质量的 Docker Compose 配置文件，
## 已经支持的服务
| 服务 | 版本 |
| ---------------------------- | ----- |
| [Redis](./src/redis) | 8.2.1 |
| [MinerU v2](./src/mineru-v2) | 2.5.3 |
| 服务 | 版本 |
| ---------------------------------------- | ------ |
| [Redis](./src/redis) | 8.2.1 |
| [MinerU v2](./src/mineru-v2) | 2.5.3 |
| [Bifrost Gateway](./src/bifrost-gateway) | 1.2.15 |
| [Docker Registry](./src/docker-registry) | 3.0.0 |
## 规范

32
src/frpc/README.md Normal file
View File

@@ -0,0 +1,32 @@
# FRPC
内网穿透客户端。
以 SSH 服务穿透为例，新建 `frpc.toml` 配置文件，内容如下：
```toml
[common]
server_addr = {{ .Envs.FRP_SERVER_ADDR }}
server_port = {{ .Envs.FRP_SERVER_PORT }}
token = {{ .Envs.FRP_SERVER_TOKEN }}
[app_22]
type = "tcp"
remote_port = 23922
local_ip = 192.168.10.100
local_port = 22
```
配置远程 FRPS 服务地址到 `.env` 文件中：
```properties
FRP_SERVER_ADDR=frps.example.com
FRP_SERVER_PORT=9870
FRP_SERVER_TOKEN=password
```
启动服务，即可代理 `192.168.10.100:22` 到 `FRP_SERVER_ADDR:23922`。
```bash
docker compose up -d
```

30
src/frpc/docker-compose.yaml Normal file
View File

@@ -0,0 +1,30 @@
x-default: &default
restart: unless-stopped
volumes:
- &localtime /etc/localtime:/etc/localtime:ro
- &timezone /etc/timezone:/etc/timezone:ro
logging:
driver: json-file
options:
max-size: 100m
services:
frpc:
<<: *default
image: snowdreamtech/frpc:${FRPC_VERSION:-0.64.0}
volumes:
- *localtime
- *timezone
- ./frpc.toml:/etc/frp/frpc.toml:ro
environment:
FRP_SERVER_ADDR: ${FRP_SERVER_ADDR}
FRP_SERVER_PORT: ${FRP_SERVER_PORT}
FRP_SERVER_TOKEN: ${FRP_SERVER_TOKEN}
deploy:
resources:
limits:
cpus: '0.5'
memory: 128M
reservations:
cpus: '0.1'
memory: 64M

10
src/frpc/frpc.toml Normal file
View File

@@ -0,0 +1,10 @@
[common]
server_addr = {{ .Envs.FRP_SERVER_ADDR }}
server_port = {{ .Envs.FRP_SERVER_PORT }}
token = {{ .Envs.FRP_SERVER_TOKEN }}
[app]
type = "tcp"
remote_port = 22
local_ip = {{ .Envs.APP_HOST }}
local_port = 22

35
src/frps/README.zh.md Normal file
View File

@@ -0,0 +1,35 @@
# FRPS
内网穿透服务端。
新建 `frps.toml` 配置文件，内容如下：
```toml
[common]
bind_port = {{ .Envs.FRP_SERVER_PORT }}
token = {{ .Envs.FRP_SERVER_TOKEN }}
dashboard_addr = 0.0.0.0
dashboard_port = {{ .Envs.FRP_ADMIN_PORT }}
dashboard_user = {{ .Envs.FRP_ADMIN_USER }}
dashboard_pwd = {{ .Envs.FRP_ADMIN_PASS }}
```
配置密钥等信息到 `.env` 文件中：
```properties
FRP_SERVER_TOKEN=token
FRP_ADMIN_USER=admin
FRP_ADMIN_PASS=password
```
启动服务，即可代理客户端请求，注意需要将客户端的端口映射出来。可以通过 HOST 网络或共享网络的方式共享服务。
使用如下方法启用 HOST 模式：
```yaml
services:
frps:
# ...
network_mode: host
```

35
src/frps/docker-compose.yaml Normal file
View File

@@ -0,0 +1,35 @@
x-default: &default
restart: unless-stopped
volumes:
- &localtime /etc/localtime:/etc/localtime:ro
- &timezone /etc/timezone:/etc/timezone:ro
logging:
driver: json-file
options:
max-size: 100m
services:
frps:
<<: *default
image: snowdreamtech/frps:${FRPS_VERSION:-0.64.0}
volumes:
- *localtime
- *timezone
- ./frps.toml:/etc/frp/frps.toml:ro
ports:
- ${FRP_PORT_OVERRIDE_SERVER:-9870}:${FRP_SERVER_PORT:-9870}
- ${FRP_PORT_OVERRIDE_ADMIN:-7890}:${FRP_ADMIN_PORT:-7890}
environment:
FRP_SERVER_TOKEN: ${FRP_SERVER_TOKEN}
FRP_SERVER_PORT: ${FRP_SERVER_PORT:-9870}
FRP_ADMIN_PORT: ${FRP_ADMIN_PORT:-7890}
FRP_ADMIN_USER: ${FRP_ADMIN_USER:-admin}
FRP_ADMIN_PASS: ${FRP_ADMIN_PASS:-password}
deploy:
resources:
limits:
cpus: '0.5'
memory: 128M
reservations:
cpus: '0.1'
memory: 64M

8
src/frps/frps.toml Normal file
View File

@@ -0,0 +1,8 @@
[common]
bind_port = {{ .Envs.FRP_SERVER_PORT }}
token = {{ .Envs.FRP_SERVER_TOKEN }}
dashboard_addr = 0.0.0.0
dashboard_port = {{ .Envs.FRP_ADMIN_PORT }}
dashboard_user = {{ .Envs.FRP_ADMIN_USER }}
dashboard_pwd = {{ .Envs.FRP_ADMIN_PASS }}

26
src/mineru-sgalng/README.md Normal file
View File

@@ -0,0 +1,26 @@
# MinerU SGLang
SGLang backend server:
```bash
docker compose --profile sglang-server up -d
```
Document parse API:
```bash
docker compose --profile api up -d
```
Gradio WebUI:
```bash
docker compose --profile gradio up -d
```
Test SGLang backend:
```bash
pip install mineru
mineru -p demo.pdf -o ./output -b vlm-sglang-client -u http://localhost:30000
```

90
src/mineru-sgalng/docker-compose.yaml Normal file
View File

@@ -0,0 +1,90 @@
x-default: &default
restart: unless-stopped
volumes:
- &localtime /etc/localtime:/etc/localtime:ro
- &timezone /etc/timezone:/etc/timezone:ro
logging:
driver: json-file
options:
max-size: 100m
x-mineru-sglang: &mineru-sglang
<<: *default
image: ${MINERU_DOCKER_IMAGE:-alexsuntop/mineru-sglang:2.2.2}
environment:
MINERU_MODEL_SOURCE: local
ulimits:
memlock: -1
stack: 67108864
ipc: host
deploy:
resources:
limits:
cpus: '8.0'
memory: 4G
reservations:
cpus: '1.0'
memory: 2G
devices:
- driver: nvidia
device_ids: [ '0' ]
capabilities: [ gpu ]
services:
mineru-sglang-server:
<<: *mineru-sglang
profiles: ["sglang-server"]
ports:
- ${MINERU_PORT_OVERRIDE_SGLANG:-30000}:30000
entrypoint: mineru-sglang-server
command:
- --host 0.0.0.0
- --port 30000
# If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode.
# - --data-parallel-size 2
# If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter,
# if VRAM issues persist, try lowering it further to `0.4` or below.
# - --gpu-memory-utilization 0.5
healthcheck:
test: ["CMD-SHELL", "curl -f http://localhost:30000/health || exit 1"]
mineru-api:
<<: *mineru-sglang
profiles: ["api"]
ports:
- ${MINERU_PORT_OVERRIDE_API:-8000}:8000
entrypoint: mineru-api
command:
- --host 0.0.0.0
- --port 8000
# If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode.
# - --data-parallel-size 2
# If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter,
# if VRAM issues persist, try lowering it further to `0.4` or below.
# - --gpu-memory-utilization 0.5
mineru-gradio:
<<: *mineru-sglang
profiles: ["gradio"]
ports:
- ${MINERU_PORT_OVERRIDE_GRADIO:-7860}:7860
entrypoint: mineru-gradio
command:
- --server-name 0.0.0.0
- --server-port 7860
# Enable the vllm engine for Gradio
- --enable-vllm-engine true
# If you want to disable the API, set this to false
# - --enable-api false
# If you want to limit the number of pages for conversion, set this to a specific number
# - --max-convert-pages 20
# If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode.
# - --data-parallel-size 2
# If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter,
# if VRAM issues persist, try lowering it further to `0.4` or below.
# - --gpu-memory-utilization 0.5

2
src/mineru-v2/docker-compose.yaml
View File

@@ -23,7 +23,7 @@ x-mineru-vllm: &mineru-vllm
cpus: '8.0'
memory: 4G
reservations:
cpus: '2.0'
cpus: '1.0'
memory: 2G
devices:
- driver: nvidia