本篇文章只讲Airflow的部署以及再部署中遇到的坑和解决方式
-
环境准备
-
Python的安装
python安装的过程中 你可能会遇到各种各样的问题,上网搜各种问题的解法也不尽相同,最关键的是基本没啥效果。在我安装的过程中总结了几点,再执行我下面的流程的时候,一定要一步不落,并且保证环境一定要干净,如果在执行某个步骤的时候出现 已存在的字眼,一定要删掉然后重新执行这一步。(这都是血与泪的教训)
#python依赖 yum -y install zlib zlib-devel yum -y install bzip2 bzip2-devel yum -y install ncurses ncurses-devel yum -y install readline readline-devel yum -y install openssl openssl-devel yum -y install openssl-static yum -y install xz lzma xz-devel yum -y install sqlite sqlite-devel yum -y install gdbm gdbm-devel yum -y install tk tk-devel yum install gcc #安装wget命令 yum -y install wget #使用wget下载Python源码压缩包到/root目录下 wget -P /root https://www.python.org/ftp/python/3.6.5/Python-3.6.5.tgz #在当前目录解压Python源码压缩包 tar -zxvf Python-3.6.5.tgz #进入解压后的文件目录下 cd /root/Python-3.6.5 #检测及校验平台 ./configure --with-ssl --prefix=/service/python3 #编译Python源代码 make #安装Python make install #备份原来的Python软连接 mv /usr/bin/python /usr/bin/python2.backup #制作新的指向Python3的软连接 ln -s /service/python3/bin/python3 /usr/bin/python #建立pip的软连接 ln -s /service/python3//bin/pip3 /usr/bin/pip###注意这一步执行的时候就可能会出现已经存在,这时候你要把/usr/bin/pip删掉然后再执行这一步错误:ln: failed to create symbolic link ‘/usr/bin/pip’: File exists
#查看Python版本 python -V #检测pip是否可用 pip #升级pip pip install --upgrade pip #获取yum命令所在位置 whereis yum #yum: /usr/bin/yum /etc/yum /etc/yum.conf /usr/share/man/man8/yum.8 #编辑yum文件 vi /usr/bin/yum /etc/yum /etc/yum.conf /usr/share/man/man8/yum.8 #进入编辑模式 i #修改第一行内容(看系统版本,centos7对应2.7,centos6对应2.6) #修改前: #!/usr/bin/python #修改后: #!/usr/bin/python2.7 #退出编辑模式 esc #保存文件 :wq #按上述方式编辑以下文件,修改第一行内容 /usr/libexec/urlgrabber-ext-down 注意在编辑yum文件这一步一定要谨慎不要弄错。否则后边用yum命令的时候可能会出问题,有包下不下来的情况
Error downloading packages:
python3-rpm-generators-6-2.el7.noarch: [Errno 5] [Errno 2] No such file or directory
对了 在以上步骤都执行完毕以后请输入pip命令看一下是否正常
如下就是正常的
-
MySQL的安装
mysql的安装有两个方式一个是RPM安装,另一种是tar包部署
本文中给出简单的RPM部署
如果不成功或者想保证一次性安装成功可以用TAR包部署,TAR包部署可参考
https://www.cnblogs.com/xuziyu/p/10353968.html
#卸载mariadb rpm -qa | grep mariadb rpm -e --nodeps mariadb-libs-5.5.52-1.el7.x86_64 #sudo rpm -e --nodeps mariadb-libs-5.5.52-1.el7.x86_64 rpm -qa | grep mariadb
#下载mysql的repo源 wget -P /root http://repo.mysql.com/mysql-community-release-el7-5.noarch.rpm #通过rpm安装 rpm -ivh mysql-community-release-el7-5.noarch.rpm #安装mysql yum install mysql-server #授权 chown -R mysql:mysql /var/lib/mysql #开启Mysql服务 service mysqld start #用root用户连接登录mysql: mysql -uroot #重置mysql密码 use mysql; update user set password=password(‘root‘) where user=‘root‘; flush privileges; #为Airflow建库、建用户 #建库: create database airflow; #建用户: create user ‘airflow‘@‘%‘ identified by ‘airflow‘; create user ‘airflow‘@‘localhost‘ identified by ‘airflow‘; #为用户授权: grant all on airflow.* to ‘airflow‘@‘%‘; grant all on airflow.* to ‘root‘@‘%‘; flush privileges; exit;
官网:http://airflow.apache.org/
Airflow2019年1月成为了Apache的顶级项目,它是由Python编写的一个任务调度框架。
-
接下来安装airflow
#设置临时环境变量 export SLUGIFY_USES_TEXT_UNIDECODE=yes #添加编辑环境变量 vi /etc/profile #在最后添加以下内容: ----》 export PS1="[\[email protected]\h \w]\$ " #Python环境变量 export PYTHON_HOME=/service/python3 export PATH=$PATH:$PYTHON_HOME/bin #Airflow环境变量 export AIRFLOW_HOME=/root/airflow export SITE_AIRFLOW_HOME=/service/python3/lib/python3.6/site-packages/airflow export PATH=$PATH:$SITE_AIRFLOW_HOME/bin ----》 #生效环境变量 source /etc/profile #安装apache-airflow并且指定1.10.0版本 pip install apache-airflow===1.10.0(这一步你若能顺利执行下来,你就可以欢呼一会了,太难了)
airflow会被安装到python3下的site-packages目录下,完整目录为:
${PYTHON_HOME}/lib/python3.6/site-packages/airflow #绝对路径/service/python3/lib/python3.6/site-packages/airflow
执行airflow
命令做初始化操作
解决:参考博客:https://www.cnblogs.com/wang-li/p/7620483.html
https://blog.csdn.net/yingkongshi99/article/details/52658538
airflow #### [2019-07-17 04:40:01,565] {__init__.py:51} INFO - Using executor SequentialExecutor usage: airflow [-h] {backfill,list_tasks,clear,pause,unpause,trigger_dag,delete_dag,pool,variables,kerberos,render,run,initdb,list_dags,dag_state,task_failed_deps,task_state,serve_logs,test,webserver,resetdb,upgradedb,scheduler,worker,flower,version,connections,create_user} ... airflow: error: the following arguments are required: subcommand #### #到此,airflow会在刚刚的AIRFLOW_HOME目录下生成一些文件。当然,执行该命令时可能会报一些错误,可以不用理会! #报错如下: [2019-07-17 04:40:01,565] {__init__.py:51} INFO - Using executor SequentialExecutor usage: airflow [-h] {backfill,list_tasks,clear,pause,unpause,trigger_dag,delete_dag,pool,variables,kerberos,render,run,initdb,list_dags,dag_state,task_failed_deps,task_state,serve_logs,test,webserver,resetdb,upgradedb,scheduler,worker,flower,version,connections,create_user} ... airflow: error: the following arguments are required: subcommand #生成的文件logs如下所示: [[email protected] ~]$ cd airflow/ [[email protected] ~/airflow]$ ll total 28 -rw-r--r--. 1 root root 20572 Jul 17 04:40 airflow.cfg drwxr-xr-x. 3 root root 23 Jul 17 04:40 logs -rw-r--r--. 1 root root 2299 Jul 17 04:40 unittests.cfg
#为airflow安装mysql模块 pip install ‘apache-airflow[mysql]‘
#出现报错: ERROR: Complete output from command python setup.py egg_info: ERROR: /bin/sh: mysql_config: command not found Traceback (most recent call last): File "<string>", line 1, in <module> File "/tmp/pip-install-dq81ujxt/mysqlclient/setup.py", line 16, in <module> metadata, options = get_config() File "/tmp/pip-install-dq81ujxt/mysqlclient/setup_posix.py", line 51, in get_config libs = mysql_config("libs") File "/tmp/pip-install-dq81ujxt/mysqlclient/setup_posix.py", line 29, in mysql_config raise EnvironmentError("%s not found" % (_mysql_config_path,)) OSError: mysql_config not found ---------------------------------------- ERROR: Command "python setup.py egg_info" failed with error code 1 in /tmp/pip-install-dq81ujxt/mysqlclient/ #解决方案,查看是否有mysql_config文件 [[email protected] ~]$ find / -name mysql_config #如果没有 [[email protected] ~]$ yum -y install mysql-devel #安装完成后再次验证是否有mysql_config find / -name mysql_config #采用mysql作为airflow的元数据库 pip install mysqlclient #安装MySQLdb pip install MySQLdb #报错不支持 Collecting MySQLdb ERROR: Could not find a version that satisfies the requirement MySQLdb (from versions: none) ERROR: No matching distribution found for MySQLdb #所以使用python-mysql pip install pymysql pip install cryptography #避免之后产生错误 #airflow.exceptions.AirflowException: Could not create Fernet object: Incorrect padding #需要修改airflow.cfg (默认位于~/airflow/)里的fernet_key #修改方法 python -c "from cryptography.fernet import Fernet; print(Fernet.generate_key().decode())" #这个命令生成一个key,复制这个key然后替换airflow.cfg文件里的fernet_key的值, #可能出现报错 Traceback (most recent call last): File "<string>", line 1, in <module> ModuleNotFoundError: No module named ‘cryptography‘ #处理方式: pip install cryptography #文件中进行fernet_key值修改 cd ${AIRFLOW_HOME} vi airflow.cfg #查找fernet_net /fernet_net #编辑替换fernet值
#修改airflow.cfg文件中的sql_alchemy_conn配置 sql_alchemy_conn = mysql+mysqldb://airflow:[email protected]:3306/airflow #保存文件 #为避免初始化数据库时有如下报错 #Global variable explicit_defaults_for_timestamp needs to be on (1) for mysql #修改MySQL配置文件my.cnf #查找my.cnf位置 mysql --help | grep my.cnf
#修改my.cnf vi /etc/my.cnf #在[mysqld]下面(一定不要写错地方)添加如下配置: explicit_defaults_for_timestamp=true
#重启mysql服务使配置生效 service mysqld restart #检查配置是否生效 mysql -uroot -proot mysql> select @@global.explicit_defaults_for_timestamp; +------------------------------------------+ | @@global.explicit_defaults_for_timestamp | +------------------------------------------+ | 1 | +------------------------------------------+ 1 row in set (0.00 sec)
通过修改airflow.cfg调整配置
1.修改webserver地址
base_url = http://IP:8080 web_server_port = 8080
2.修改executor
3.时区
#修改airflow.cfg中
default_timezone = Asia/Shanghai
其他配置
#同时需要修改另外三个文件 #修改webserver页面上右上角展示的时间: vi ${PYTHON_HOME}/lib/python3.6/site-packages/airflow/www/templates/admin/master.html ----------------------------------- {% block tail_js %} {{ super() }} <script src="{{ url_for(‘static‘, filename=‘jqClock.min.js‘) }}" type="text/javascript"></script> <script> x = new Date() // var UTCseconds = (x.getTime() + x.getTimezoneOffset()*60*1000);##修改的内容 var UTCseconds = x.getTime();##修改的内容 $("#clock").clock({
#修改webserver lastRun时间: vi ${PYTHON_HOME}/lib/python3.6/site-packages/airflow/models.py -----------------------------------》 #在指定位置添加如下内容,可以借助get_last_dagrun定位 def utc2local(self,utc): import time epoch = time.mktime(utc.timetuple()) offset = datetime.fromtimestamp(epoch) - datetime.utcfromtimestamp(epoch) return utc + offset
vi ${PYTHON_HOME}/lib/python3.6/site-packages/airflow/www/templates/airflow/dags.html #在图中指定位置修改为如下内容 dag.utc2local(last_run.execution_date).strftime("%Y-%m-%d %H:%M") dag.utc2local(last_run.start_date).strftime("%Y-%m-%d %H:%M")
4.添加用户认证(暂时不做这一步,还没懂)
在这里我们采用简单的password认证方式 #(1)安装password组件: sudo pip install apache-airflow[password] #(2)修改airflow.cfg配置文件: [webserver] authenticate = True auth_backend = airflow.contrib.auth.backends.password_auth #(3)编写python脚本用于添加用户账号: #编写add_account.py文件: import airflow from airflow import models, settings from airflow.contrib.auth.backends.password_auth import PasswordUser user = PasswordUser(models.User()) user.username = ‘airflow‘ user.email = ‘[email protected]‘ user.password = ‘airflow‘ session = settings.Session() session.add(user) session.commit() session.close() exit() #执行add_account.py文件: python add_account.py #你会发现mysql元数据库表user中会多出来一条记录的。
5修改scheduler线程数控制并发量
parallelism = 32
6修改检测新DAG间隔
min_file_process_interval = 5
-
初始化源数据库及启动组件
#初始化元数据库信息(其实也就是新建airflow依赖的表) pip install celery pip install apache-airflow[‘kubernetes‘] airflow initdb #或者使用airflow resetdb
#准备操作 #关闭linux防火墙 systemctl stop firewalld.service systemctl disable firewalld.service #同时需要关闭windows防火墙 #数据库设置 mysql -uroot -proot mysql> set password for ‘root‘@‘localhost‘ =password(‘‘); Query OK, 0 rows affected (0.00 sec) mysql> grant all on airflow.* to ‘airflow‘@‘%‘; Query OK, 0 rows affected (0.00 sec) mysql> grant all on airflow.* to ‘root‘@‘%‘; Query OK, 0 rows affected (0.01 sec) mysql> flush privileges; Query OK, 0 rows affected (0.00 sec) mysql> exit; #启动组件: airflow webserver -D #airflow scheduler -D #airflow worker -D #airflow flower -D
-
Web页面查看
#地址 192.168.150.128:8085/admin/ #测试 可以选择airflow_db数据库简单查询进行测试 select * from log;
https://blog.csdn.net/yingkongshi99/article/details/52658538
原文地址:https://www.cnblogs.com/xuziyu/p/11696549.html