系列文章目录

线上问诊：业务数据采集
线上问诊：数仓数据同步
线上问诊：数仓开发(一)
线上问诊：数仓开发(二)
线上问诊：数仓开发(三)
线上问诊：可视化展示

---

---

前言

这次博客我们就可以完成线上问诊的离线数仓项目。

---

一、全流程调度

正常应该使用DolphinScheduler进行全流程调度，但是由于博主的电脑硬件的极限。所以为了完成整个项目，我们自己写一个bash脚本，来进行流程调度

1.生产新数据

vim /opt/module/maxwell/config.properties

medical_mock.sh 1

在数据库中查看是否出现10号数据。

2.编写脚本

这个脚本是我自己写的，项目中原本没有
vim ~/bin/medical_run.sh

#!/bin/bash 
/bin/bash /home/atguigu/bin/medical_mysql_to_hdfs_full.sh all $1
/bin/bash /home/atguigu/bin/medical_hdfs_to_ods.sh all $1
/bin/bash /home/atguigu/bin/medical_ods_to_dwd.sh all $1
/bin/bash /home/atguigu/bin/medical_ods_to_dim.sh all $1
/bin/bash /home/atguigu/bin/medical_dwd_to_dws_1d.sh all $1
/bin/bash /home/atguigu/bin/medical_dws_1d_to_dws_nd.sh all $1
/bin/bash /home/atguigu/bin/medical_dws_1d_to_dws_td.sh all $1
/bin/bash /home/atguigu/bin/medical_dws_to_ads.sh all $1
/bin/bash /home/atguigu/bin/medical_hdfs_to_mysql.sh all

添加权限
chmod +x ~/bin/medical_run.sh

3.脚本测试

先打开之前创建的所有通道

myhadoop.sh start
zk.sh start
kf.sh start
mxw.sh start
medical-f1.sh start
hiveservices.sh start

medical_run.sh 2023-05-10

当数据库中出现10号，代表成功。

二、可视化展示

这里我们就不需要通道了，如果硬件不太够，可以把通道关闭。

1.安装Miniconda3

minconda3清华源

bash Miniconda3-latest-Linux-x86_64.sh

安装地址修改一下。
/opt/module/miniconda3
激活环境
source ~/.bashrc
取消默认激活base
conda config --set auto_activate_base false
换源
conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free
conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
conda config --set show_channel_urls yes

2.安装python3

conda create --name superset python=3.8.16
激活环境
conda activate superset

3.Superset部署

1.安装依赖

sudo yum install -y gcc gcc-c++ libffi-devel python-devel python-pip python-wheel python-setuptools openssl-devel cyrus-sasl-devel openldap-devel

2.加载模块

编写一个base.txt文件
vim base.txt

aiohttp==3.8.1
    # via slackclient
aiosignal==1.2.0
    # via aiohttp
alembic==1.6.5
    # via flask-migrate
amqp==5.1.0
    # via kombu
apispec[yaml]==3.3.2
    # via flask-appbuilder
async-timeout==4.0.2
    # via aiohttp
attrs==21.2.0
    # via
    #   aiohttp
    #   jsonschema
babel==2.9.1
    # via flask-babel
backoff==1.11.1
    # via apache-superset
billiard==3.6.4.0
    # via celery
bleach==3.3.1
    # via apache-superset
brotli==1.0.9
    # via flask-compress
cachelib==0.4.1
    # via apache-superset
celery==5.2.2
    # via apache-superset
cffi==1.14.6
    # via cryptography
charset-normalizer==2.0.4
    # via aiohttp
click==8.0.4
    # via
    #   apache-superset
    #   celery
    #   click-didyoumean
    #   click-plugins
    #   click-repl
    #   flask
    #   flask-appbuilder
click-didyoumean==0.3.0
    # via celery
click-plugins==1.1.1
    # via celery
click-repl==0.2.0
    # via celery
colorama==0.4.4
    # via
    #   apache-superset
    #   flask-appbuilder
convertdate==2.3.2
    # via holidays
cron-descriptor==1.2.24
    # via apache-superset
croniter==1.0.15
    # via apache-superset
cryptography==3.4.7
    # via apache-superset
deprecation==2.1.0
    # via apache-superset
dnspython==2.1.0
    # via email-validator
email-validator==1.1.3
    # via flask-appbuilder
flask==2.0.3
    # via
    #   apache-superset
    #   flask-appbuilder
    #   flask-babel
    #   flask-caching
    #   flask-compress
    #   flask-jwt-extended
    #   flask-login
    #   flask-migrate
    #   flask-sqlalchemy
    #   flask-wtf
flask-appbuilder==4.1.3
    # via apache-superset
flask-babel==1.0.0
    # via flask-appbuilder
flask-caching==1.10.1
    # via apache-superset
flask-compress==1.10.1
    # via apache-superset
flask-jwt-extended==4.3.1
    # via flask-appbuilder
flask-login==0.4.1
    # via flask-appbuilder
flask-migrate==3.1.0
    # via apache-superset
flask-sqlalchemy==2.5.1
    # via
    #   flask-appbuilder
    #   flask-migrate
flask-talisman==0.8.1
    # via apache-superset
flask-wtf==0.14.3
    # via
    #   apache-superset
    #   flask-appbuilder
frozenlist==1.3.0
    # via
    #   aiohttp
    #   aiosignal
func-timeout==4.3.5
    # via apache-superset
geographiclib==1.52
    # via geopy
geopy==2.2.0
    # via apache-superset
graphlib-backport==1.0.3
    # via apache-superset
gunicorn==20.1.0
    # via apache-superset
hashids==1.3.1
    # via apache-superset
holidays==0.10.3
    # via apache-superset
humanize==3.11.0
    # via apache-superset
idna==3.2
    # via
    #   email-validator
    #   yarl
isodate==0.6.0
    # via apache-superset
itsdangerous==2.1.1
    # via
    #   flask
    #   flask-wtf
jinja2==3.0.3
    # via
    #   flask
    #   flask-babel
jsonschema==3.2.0
    # via flask-appbuilder
kombu==5.2.4
    # via celery
korean-lunar-calendar==0.2.1
    # via holidays
mako==1.1.4
    # via alembic
markdown==3.3.4
    # via apache-superset
markupsafe==2.0.1
    # via
    #   jinja2
    #   mako
    #   wtforms
marshmallow==3.13.0
    # via
    #   flask-appbuilder
    #   marshmallow-enum
    #   marshmallow-sqlalchemy
marshmallow-enum==1.5.1
    # via flask-appbuilder
marshmallow-sqlalchemy==0.23.1
    # via flask-appbuilder
msgpack==1.0.2
    # via apache-superset
multidict==5.1.0
    # via
    #   aiohttp
    #   yarl
numpy==1.22.1
    # via
    #   apache-superset
    #   pandas
    #   pyarrow
packaging==21.3
    # via
    #   bleach
    #   deprecation
pandas==1.3.4
    # via apache-superset
parsedatetime==2.6
    # via apache-superset
pgsanity==0.2.9
    # via apache-superset
polyline==1.4.0
    # via apache-superset
prison==0.2.1
    # via flask-appbuilder
prompt-toolkit==3.0.28
    # via click-repl
pyarrow==5.0.0
    # via apache-superset
pycparser==2.20
    # via cffi
pyjwt==2.4.0
    # via
    #   apache-superset
    #   flask-appbuilder
    #   flask-jwt-extended
pymeeus==0.5.11
    # via convertdate
pyparsing==3.0.6
    # via
    #   apache-superset
    #   packaging
pyrsistent==0.16.1
    # via jsonschema
python-dateutil==2.8.2
    # via
    #   alembic
    #   apache-superset
    #   croniter
    #   flask-appbuilder
    #   holidays
    #   pandas
python-dotenv==0.19.0
    # via apache-superset
python-editor==1.0.4
    # via alembic
python-geohash==0.8.5
    # via apache-superset
pytz==2021.3
    # via
    #   babel
    #   celery
    #   convertdate
    #   flask-babel
    #   pandas
pyyaml==5.4.1
    # via
    #   apache-superset
    #   apispec
redis==3.5.3
    # via apache-superset
selenium==3.141.0
    # via apache-superset
simplejson==3.17.3
    # via apache-superset
six==1.16.0
    # via
    #   bleach
    #   click-repl
    #   flask-talisman
    #   holidays
    #   isodate
    #   jsonschema
    #   polyline
    #   prison
    #   pyrsistent
    #   python-dateutil
    #   sqlalchemy-utils
    #   wtforms-json
slackclient==2.5.0
    # via apache-superset
sqlalchemy==1.3.24
    # via
    #   alembic
    #   apache-superset
    #   flask-appbuilder
    #   flask-sqlalchemy
    #   marshmallow-sqlalchemy
    #   sqlalchemy-utils
sqlalchemy-utils==0.37.8
    # via
    #   apache-superset
    #   flask-appbuilder
sqlparse==0.3.0
    # via apache-superset
tabulate==0.8.9
    # via apache-superset
typing-extensions==3.10.0.0
    # via apache-superset
urllib3==1.26.6
    # via selenium
vine==5.0.0
    # via
    #   amqp
    #   celery
    #   kombu
wcwidth==0.2.5
    # via prompt-toolkit
webencodings==0.5.1
    # via bleach
werkzeug==2.0.3
    # via
    #   flask
    #   flask-jwt-extended
wtforms==2.3.3
    # via
    #   flask-appbuilder
    #   flask-wtf
    #   wtforms-json
wtforms-json==0.3.3
    # via apache-superset
yarl==1.6.3
    # via aiohttp

# The following packages are considered to be unsafe in a requirements file:
# setuptools

4.Superset部署

1.安装Superset

pip install apache-superset==2.0.0 -i https://pypi.tuna.tsinghua.edu.cn/simple -r base.txt

2.配置Superset元数据库

CREATE DATABASE superset DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
create user superset@'%' identified WITH mysql_native_password BY 'superset';
grant all privileges on *.* to superset@'%' with grant option;
flush privileges;

3.修改superset配置文件

vim /opt/module/miniconda3/envs/superset/lib/python3.8/site-packages/superset/config.py

SQLALCHEMY_DATABASE_URI = ‘mysql://superset:superset@hadoop102:3306/superset?charset=utf8’

4.安装python msyql驱动

conda install mysqlclient

5.初始化superset元数据

export FLASK_APP=superset
superset db upgrade

5.SupersetSet初始化

1.创建管理员用户

superset fab create-admin

密码自己输入一下，其他的默认就行了。

2.初始化superset

superset init

3.启动脚本

pip install gunicorn -i https://pypi.douban.com/simple/
vim ~/bin/superset.sh

#!/bin/bash

superset_status(){
    result=`ps -ef | awk '/gunicorn/ && !/awk/{print $2}' | wc -l`
    if [[ $result -eq 0 ]]; then
        return 0
    else
        return 1
    fi
}
superset_start(){
        source ~/.bashrc
        superset_status >/dev/null 2>&1
        if [[ $? -eq 0 ]]; then
            conda activate superset ; gunicorn --workers 5 --timeout 120 --bind hadoop102:8787 --daemon 'superset.app:create_app()'
        else
            echo "superset正在运行"
        fi

}

superset_stop(){
    superset_status >/dev/null 2>&1
    if [[ $? -eq 0 ]]; then
        echo "superset未在运行"
    else
        ps -ef | awk '/gunicorn/ && !/awk/{print $2}' | xargs kill -9
    fi
}


case $1 in
    start )
        echo "启动Superset"
        superset_start
    ;;
    stop )
        echo "停止Superset"
        superset_stop
    ;;
    restart )
        echo "重启Superset"
        superset_stop
        superset_start
    ;;
    status )
        superset_status >/dev/null 2>&1
        if [[ $? -eq 0 ]]; then
            echo "superset未在运行"
        else
            echo "superset正在运行"
        fi
esac

添加权限
chmod +x ~/bin/superset.sh
启动superset
superset.sh start
访问http://hadoop102:8787

用之前设置的密码登录。

6.Superset使用

1.连接MYSQL

2.配置Table

用这三张表进行数据可视化。

3.制作仪表盘

可以和数据库数据对比一下。









检查一下数据




自己拖拽三个框

直接拖进去

自己调整一下大小就可以保存了。


把得到的连接复制到浏览器即可使用。

---

总结

至此线上问诊离线数仓项目完成。可以考虑抓个快照。