dify 本地源码启动

dify 本地源码启动

1. 环境准备

官网参考
保姆级步骤参考

1.1. 基础环境

我使用的是 Ubuntu 20.04 环境，首先需要安装 docker 27.3.1和 docker compose
我的 docker compose 安装的 v2 版本。

1.2. clone dify 代码

git clone https://github.com/langgenius/dify.git

1.3. 部署依赖服务

在启用业务服务之前，我们需要先部署 PostgreSQL / Redis / Weaviate（如果本地没有的话），可以通过以下命令启动：

cd docker
cp middleware.env.example middleware.env
docker compose -f docker-compose.middleware.yaml up -d
# 关闭时，使用stop
docker compose stop

2. 服务端部署

2.1. 创建 conda 环境

官方推荐：服务器启动需要 Python 3.11 或 3.12。建议使用 pyenv 快速安装 Python 环境。（安装失败）
使用 conda 解决。

conda create --name dify311 python=3.11
conda activate dify311

2.2. 部署后端服务

2.2.1. 修改环境配置

# 后端服务在api文件夹中
cd api
# 拷贝环境配置
cp .env.example .env

修改 SECRET_KEY

# 生成随机密钥，并替换 .env 中 SECRET_KEY 的值。
openssl rand -base64 42
sed -i 's/SECRET_KEY=.*/SECRET_KEY=<your_value>/' .env

也可以使用 vim 直接修改.env 文件中 SECRET_KEY 的值。

2.2.2. 安装依赖包

官方推荐：Dify API 服务使用 Poetry 来管理依赖。你可以执行 poetry shell 来激活环境。（安装失败）
使用 pip 解决。

# 首先进入 api 目录 
cd api
# 执行 poetry show 查看安装依赖的包
poetry show 
# 将依赖包重定向输出到 requirements.txt 文件中。并修改 requirements.txt 包的格式。
poetry show >> requirements.txt
# 安装环境（时间漫长，使用清华源）
pip install -r requirements.txt -i https://pypi.tuna.tsinghua.edu.cn/simple

requirements.txt 文件有417行，也就是417个依赖包。

requirements.txt 实例如下：

aiofiles==24.1.0
aiohappyeyeballs==2.4.4
aiohttp==3.11.11
aiohttp-retry==2.9.1
aiomysql==0.2.0
aiosignal==1.3.2
alembic==1.14.0
alibabacloud-credentials==0.3.6
alibabacloud-endpoint-util==0.0.3
alibabacloud-gateway-spi==0.0.2
alibabacloud-gpdb20160503==3.8.3
alibabacloud-openapi-util==0.2.2
alibabacloud-openplatform20191219==2.0.0
alibabacloud-oss-sdk==0.1.0
alibabacloud-oss-util==0.0.6
alibabacloud-tea==0.4.0
alibabacloud-tea-fileform==0.0.5
alibabacloud-tea-openapi==0.3.12
alibabacloud-tea-util==0.3.13
alibabacloud-tea-xml==0.0.2
aliyun-python-sdk-core==2.16.0
aliyun-python-sdk-kms==2.16.5
amqp==5.3.1
aniso8601==9.0.1
annotated-types==0.7.0
anthropic==0.23.1
anyio==4.8.0
arxiv==2.1.0
asgiref==3.8.1
async-timeout==5.0.1
attrs==23.2.0
authlib==1.3.1
azure-ai-inference==1.0.0b6
azure-ai-ml==1.20.0
azure-common==1.1.28
azure-core==1.32.0
azure-identity==1.16.1
azure-mgmt-core==1.5.0
azure-storage-blob==12.13.0
azure-storage-file-datalake==12.8.0
azure-storage-file-share==12.20.0
backoff==2.2.1
bce-python-sdk==0.9.25
bcrypt==4.2.1
beautifulsoup4==4.12.2
billiard==4.2.1
blinker==1.9.0
boto3==1.35.74
botocore==1.35.94
build==1.2.2.post1
bottleneck==1.4.2
brotli==1.1.0
bs4==0.0.2
cachetools==5.3.3
celery==5.4.0
certifi==2024.12.14
cffi==1.17.1
chardet==5.1.0
charset-normalizer==3.4.1
chroma-hnswlib==0.7.6
chromadb==0.5.20
circuitbreaker==2.0.0
click==8.1.8
click-didyoumean==0.3.1
click-plugins==1.1.1
click-repl==0.3.0
clickhouse-connect==0.7.19
cloudpickle==2.2.1
cloudscraper==1.2.71
cohere==5.2.6
colorama==0.4.6
coloredlogs==15.0.1
contourpy==1.3.1
cos-python-sdk-v5==1.9.30
couchbase==4.3.4
crcmod==1.7
cryptography==44.0.0
cssselect==1.2.0
cycler==0.12.1
dashscope==1.17.1
dataclass-wizard==0.34.0
dataclasses-json==0.6.7
db-dtypes==1.3.1
decorator==5.1.1
defusedxml==0.7.1
deprecated==1.2.15
deprecation==2.1.0
dill==0.3.9
distro==1.9.0
docker==7.1.0
docstring-parser==0.16
duckdb==1.1.3
duckduckgo-search==6.3.7
durationpy==0.9
elastic-transport==8.17.0
elasticsearch==8.14.0
emoji==2.14.0
esdk-obs-python==3.24.6.1
et-xmlfile==2.0.0
eval-type-backport==0.2.2
fal-client==0.5.6
fastapi==0.115.6
fastavro==1.10.0
feedfinder2==0.0.4
feedparser==6.0.10
filelock==3.16.1
filetype==1.2.0
fire==0.7.0
flasgger==0.9.7.1
flask==3.1.0
flask-compress==1.17
flask-cors==4.0.2
flask-login==0.6.3
flask-migrate==4.0.7
flask-restful==0.3.10
flask-sock==0.7.0
flask-sqlalchemy==3.1.1
flatbuffers==24.12.23
fontmeta==1.6.1
fonttools==4.55.3
frozendict==2.4.6
frozenlist==1.5.0
fsspec==2024.12.0
future==1.0.0
gevent==24.11.1
gmpy2==2.2.1
google==3.0.0
google-ai-generativelanguage==0.6.9
google-api-core==2.18.0
google-api-python-client==2.90.0
google-auth==2.29.0
google-auth-httplib2==0.2.0
google-cloud-aiplatform==1.49.0
google-cloud-bigquery==3.27.0
google-cloud-core==2.4.1
google-cloud-resource-manager==1.14.0
google-cloud-storage==2.16.0
google-crc32c==1.6.0
google-generativeai==0.8.1
google-pasta==0.2.0
google-resumable-media==2.7.2
googleapis-common-protos==1.63.0
gotrue==2.11.1
greenlet==3.1.1
grpc-google-iam-v1==0.14.0
grpcio==1.67.1
grpcio-status==1.62.3
grpcio-tools==1.62.3
gunicorn==23.0.0
h11==0.14.0
h2==4.1.0
hiredis==3.1.0
hpack==4.0.0
html5lib==1.1
httpcore==1.0.7
httplib2==0.22.0
httptools==0.6.4
httpx==0.27.2
httpx-sse==0.4.0
huggingface-hub==0.16.4
humanfriendly==10.0
hyperframe==6.0.1
idna==3.10
importlib-metadata==6.11.0
importlib-resources==6.5.2
iniconfig==2.0.0
isodate==0.7.2
itsdangerous==2.2.0
jieba==0.42.1
jieba3k==0.35.1
jinja2==3.1.5
jiter==0.8.2
jmespath==0.10.0
joblib==1.4.2
jsonlines==4.0.0
jsonpath-ng==1.6.1
jsonpath-python==1.0.6
jsonschema==4.23.0
jsonschema-specifications==2024.10.1
kaleido==0.2.1
kiwisolver==1.4.8
kombu==5.4.2
kubernetes==31.0.0
langdetect==1.0.9
langfuse==2.51.5
langsmith==0.1.147
levenshtein==0.26.1
litellm==1.51.3
llvmlite==0.43.0
loguru==0.7.3
lxml==5.3.0
lz4==4.3.3
mailchimp-transactional==1.0.56
mako==1.3.8
markdown==3.5.2
markdown-it-py==3.0.0
markupsafe==3.0.2
marshmallow==3.24.1
matplotlib==3.8.4
mdurl==0.1.2
milvus-lite==2.4.11
mistune==3.1.0
mmh3==5.0.1
mock==4.0.3
monotonic==1.6
mplfonts==0.0.10
mpmath==1.3.0
msal==1.31.1
msal-extensions==1.2.0
msrest==0.7.1
multidict==6.1.0
multiprocess==0.70.17
multitasking==0.0.11
mypy-extensions==1.0.0
ndjson==0.3.1
nest-asyncio==1.6.0
newspaper3k==0.2.8
nltk==3.9.1
nomic==3.1.3
novita-client==0.5.7
numba==0.60.0
numexpr==2.9.0
numpy==1.26.4
oauthlib==3.2.2
oci==2.135.2
odfpy==1.4.1
olefile==0.47
onnxruntime==1.20.1
openai==1.52.2
opencensus==0.11.4
opencensus-context==0.1.3
opencensus-ext-azure==1.1.14
opencensus-ext-logging==0.1.1
opendal==0.45.13
openpyxl==3.1.5
opensearch-py==2.4.0
opentelemetry-api==1.29.0
opentelemetry-exporter-otlp-proto-grpc==1.15.0
opentelemetry-instrumentation==0.50b0
opentelemetry-instrumentation-asgi==0.50b0
opentelemetry-instrumentation-fastapi==0.50b0
opentelemetry-proto==1.15.0
opentelemetry-sdk==1.29.0
opentelemetry-semantic-conventions==0.50b0
opentelemetry-util-http==0.50b0
opik==1.3.4
oracledb==2.2.1
orjson==3.10.13
oss2==2.18.5
overrides==7.7.0
packaging==24.2
pandas==2.2.3
pandas-stubs==2.2.3.241126
pathos==0.3.3
peewee==3.17.8
pgvecto-rs==0.2.2
pgvector==0.2.5
pillow==11.1.0
platformdirs==4.3.6
plotly==5.24.1
pluggy==1.5.0
ply==3.11
portalocker==2.10.1
postgrest==0.17.2
posthog==3.7.5
pox==0.3.5
ppft==1.7.6.9
primp==0.10.0
prompt-toolkit==3.0.48
propcache==0.2.1
proto-plus==1.25.0
protobuf==4.25.5
psutil==6.1.1
psycogreen==1.0.2
psycopg2-binary==2.9.10
py==1.11.0
pyarrow==18.1.0
pyasn1==0.6.1
pyasn1-modules==0.4.1
pycparser==2.22
pycryptodome==3.19.1
pydantic==2.9.2
pydantic-core==2.23.4
pydantic-extra-types==2.9.0
pydantic-settings==2.6.1
pydash==8.0.4
pydub==0.25.1
pygments==2.19.1
pyjwt==2.8.0
pymilvus==2.5.3
pymochow==1.3.1
pymysql==1.1.1
pyobvector==0.1.18
pyopenssl==24.3.0
pypandoc==1.14
pyparsing==3.2.1
pypdf==5.1.0
pypdfium2==4.30.1
pypika==0.48.9
pypng==0.20220715.0
pyproject-hooks==1.2.0
pytest==8.3.4
python-calamine==0.3.1
python-dateutil==2.9.0.post0
python-docx==1.1.2
python-dotenv==1.0.1
python-iso639==2024.10.22
python-magic==0.4.27
python-oxmsg==0.0.1
python-pptx==1.0.2
pytz==2024.2
pyxlsb==1.0.10
pyyaml==6.0.2
qdrant-client==1.7.3
qrcode==7.4.2
rank-bm25==0.2.2
rapidfuzz==3.11.0
readabilipy==0.2.0
realtime==2.1.0
redis==5.0.8
referencing==0.35.1
regex==2024.11.6
replicate==0.22.0
requests==2.31.0
requests-file==2.1.0
requests-oauthlib==2.0.0
requests-toolbelt==1.0.0
resend==0.7.2
retry==0.9.2
rich==13.9.4
rpds-py==0.22.3
rsa==4.9
s3transfer==0.10.4
safetensors==0.4.5
sagemaker==2.231.0
sagemaker-core==1.0.16
schema==0.7.7
scikit-learn==1.5.2
scipy==1.15.0
sentry-sdk==1.44.1
setuptools==75.7.0
sgmllib3k==1.0.0
shapely==2.0.6
shellingham==1.5.4
simple-websocket==1.1.0
six==1.17.0
smdebug-rulesconfig==1.0.1
sniffio==1.3.1
socksio==1.0.0
soupsieve==2.6
sqlalchemy==2.0.35
sqlparse==0.5.3
starlette==0.41.0
storage3==0.8.2
strictyaml==1.7.3
supabase==2.8.1
supafunc==0.6.2
sympy==1.13.3
tabulate==0.9.0
tblib==3.0.0
tcvectordb==1.3.2
tenacity==9.0.0
tencentcloud-sdk-python-common==3.0.1298
tencentcloud-sdk-python-hunyuan==3.0.1298
termcolor==2.5.0
threadpoolctl==3.5.0
tidb-vector==0.0.9
tiktoken==0.8.0
tinysegmenter==0.3
tldextract==5.1.3
tokenizers==0.15.2
toml==0.10.2
tos==2.7.2
tqdm==4.67.1
transformers==4.35.2
twilio==9.0.5
typer==0.15.1
types-pytz==2024.2.0.20241221
types-requests==2.32.0.20241016
typing-extensions==4.12.2
typing-inspect==0.9.0
tzdata==2024.2
ujson==5.10.0
unstructured==0.16.12
unstructured-client==0.28.1
upstash-vector==0.6.0
uritemplate==4.1.1
uuid6==2024.7.10
uvicorn==0.34.0
uvloop==0.21.0
validators==0.21.0
vanna==0.7.5
vine==5.1.0
volcengine-compat==1.0.156
volcengine-python-sdk==1.0.103
watchfiles==1.0.3
wcwidth==0.2.13
weaviate-client==3.21.0
webencodings==0.5.1
websocket-client==1.7.0
websockets==13.1
werkzeug==3.1.3
wikipedia==1.4.0
wrapt==1.17.0
wsproto==1.2.0
xinference-client==0.15.2
xlrd==2.0.1
xlsxwriter==3.2.0
xmltodict==0.14.2
yarl==1.18.3
yfinance==0.2.51
youtube-transcript-api==0.6.3
zhipuai==2.1.5.20250106
zipp==3.21.0
zope-event==5.0
zope-interface==7.2
zstandard==0.23.0

执行数据库迁移

#  数据库迁移
flask db upgrade

2.2.3. 启动 API 服务

flask run --host 0.0.0.0 --port=5001 --debug
flask run --host 192.168.17.101 --port=5001 --debug

2.2.4. 启动 Worker 服务

用于消费异步队列任务，如知识库文件导入、更新知识库文档等异步操作。 Linux / MacOS 启动：

celery -A app.celery worker -P gevent -c 1 -Q dataset,generation,mail --loglevel INFO

3. 前端页面部署

3.1. 安装环境

这里注意一定要根据指南安装 Node.js v18.x (LTS) ，不然后续install会报错。

curl -sL https://deb.nodesource.com/setup_18.x | sudo -E bash -
apt-get install nodejs -y

安装依赖包

# 前端代码在 web 文件夹
cd web
npm install
# 如果安装很慢 切换源
npm config set registry https://registry.npm.taobao.org

错误参考 此处安装，没有使用改解决方案。
配置环境变量
在web目录下创建文件 .env.local，并复制 .env.example 中的内容。根据需求修改这些环境变量的值:

# For production release, change this to PRODUCTION
NEXT_PUBLIC_DEPLOY_ENV=DEVELOPMENT
# The deployment edition, SELF_HOSTED
NEXT_PUBLIC_EDITION=SELF_HOSTED
# The base URL of console application, refers to the Console base URL of WEB service if console domain is
# different from api or web app domain.
# example: http://cloud.dify.ai/console/api
#NEXT_PUBLIC_API_PREFIX=http://127.0.0.1:5001/console/api
NEXT_PUBLIC_API_PREFIX=http://192.168.17.101:5001/console/api
# The URL for Web APP, refers to the Web App base URL of WEB service if web app domain is different from
# console or api domain.
# example: http://udify.app/api
#NEXT_PUBLIC_PUBLIC_API_PREFIX=http://127.0.0.1:5001/api
NEXT_PUBLIC_PUBLIC_API_PREFIX=http://192.168.17.101:5001/api


# Disable Next.js Telemetry (https://nextjs.org/telemetry)
NEXT_TELEMETRY_DISABLED=1

# Disable Upload Image as WebApp icon default is false
NEXT_PUBLIC_UPLOAD_IMAGE_AS_ICON=false

# The timeout for the text generation in millisecond
NEXT_PUBLIC_TEXT_GENERATION_TIMEOUT_MS=60000

# CSP https://developer.mozilla.org/en-US/docs/Web/HTTP/CSP
NEXT_PUBLIC_CSP_WHITELIST=

# The maximum number of top-k value for RAG.
NEXT_PUBLIC_TOP_K_MAX_VALUE=10

# The maximum number of tokens for segmentation
NEXT_PUBLIC_INDEXING_MAX_SEGMENTATION_TOKENS_LENGTH=4000

## SENTRY
#NEXT_PUBLIC_SENTRY_DSN=http://192.168.17.101
#NEXT_PUBLIC_SENTRY_ORG=http://192.168.17.101
#NEXT_PUBLIC_SENTRY_PROJECT=http://192.168.17.101
# SENTRY
NEXT_PUBLIC_SENTRY_DSN=
NEXT_PUBLIC_SENTRY_ORG=
NEXT_PUBLIC_SENTRY_PROJECT=

最后构建代码

npm run build

3.2. 启动服务

npm run start

3.3. 访问 dify

最后访问 http://192.168.17.101:3000 即可使用本地部署的 Dify了~

4. 启动总结

# 进入 conda 环境
conda activate dify311

# 依赖服务
cd docker
docker compose -f docker-compose.middleware.yaml up -d

# 后端
cd api
# 启动 API 服务
flask run --host 192.168.17.101 --port=5001 --debug
# 启动 worker 服务
celery -A app.celery worker -P gevent -c 1 -Q dataset,generation,mail --loglevel INFO

# 前端
cd web
# 启动前端
npm run start

# 访问
http://192.168.17.101:3000