Presto连接MySQL

准备

准备两台机器，机器名为 bd1 和 bd2，，并且在这两个节点上安装 hadoop，hive，并且准备一个mysql数据库。

bd1 - coordinator节点bd2 - worker节点

安装 presto server

wget -c https://repo1.maven.org/maven2/com/facebook/presto/presto-server/0.211/presto-server-0.211.tar.gz tar zxvf presto-server-0.211.tar.gz cd /opt/presto-server-0.211 mkdir etc

coordinator 节点配置

etc/node.properties

node.environment=production node.id=ffffffff-ffff-ffff-ffff-fffffffffff1 node.data-dir=/opt/presto-server-0.211/data

etc/config.properties

coordinator=true node-scheduler.include-coordinator=false http-server.http.port=8080 query.max-memory=8GB query.max-memory-per-node=1GB query.max-total-memory-per-node=2GB discovery-server.enabled=true discovery.uri=http://bd1:8080

etc/jvm.config

-server -Xmx8G -XX:+UseG1GC -XX:G1HeapRegionSize=32M -XX:+UseGCOverheadLimit -XX:+ExplicitGCInvokesConcurrent -XX:+HeapDumpOnOutOfMemoryError -XX:+ExitOnOutOfMemoryError

worker 节点配置

etc/node.properties

node.environment=production node.id=ffffffff-ffff-ffff-ffff-fffffffffff2 node.data-dir=/opt/presto-server-0.211/data

etc/config.properties

coordinator=false http-server.http.port=8080 query.max-memory=8GB query.max-memory-per-node=1GB query.max-total-memory-per-node=2GB discovery.uri=http://bd1:8080

etc/jvm.config

-server -Xmx8G -XX:+UseG1GC -XX:G1HeapRegionSize=32M -XX:+UseGCOverheadLimit -XX:+ExplicitGCInvokesConcurrent -XX:+HeapDumpOnOutOfMemoryError -XX:+ExitOnOutOfMemoryError

运行Presto

分别在两个节点上运行下面的启动命令

# 启动 Presto bin/launcher start # 停止 Presto bin/launcher stop # 前台运行 Presto，建议刚开始的时候使用这种方式，如果配置有错误，可以立刻在console上看到错误信息，方便调试。 bin/launcher run

服务启动后可以通过下面的地址来查看服务状态

http://<coordinator_ip>:8080/ui/

设置 MySQL Connector

创建数据表

由于当前版本的 MySQL Connector 不支持创建数据表的操作，因此我们需要先通过MySQL的终端来创建数据表。

$ mysql -u root -p MariaDB [(none)]> use test; MariaDB [test]> create table user(id int not null, username varchar(32) not null, password varchar(32) not null); MariaDB [test]> insert into user values(1,'user1','password1'); MariaDB [test]> insert into user values(2,'user2','password2'); MariaDB [test]> insert into user values(3,'user3','password3'); MariaDB [test]> select * from user; +----+----------+-----------+ | id | username | password | +----+----------+-----------+ | 1 | user1 | password1 | | 2 | user2 | password2 | | 3 | user3 | password3 | +----+----------+-----------+

MySQL Connector 配置

在etc目录下创建catalog目录，然后创建 mysql.properties 文件,编辑保存后将此文件复制的集群中的其它节点相同目录下。

etc/catalog/mysql.properties

connector.name=mysql connection-url=jdbc:mysql://bd1:3306 connection-user=root connection-password=<password>

获取 Presto 命令行工具

# 下载 wget -c https://repo1.maven.org/maven2/com/facebook/presto/presto-cli/0.211/presto-cli-0.211-executable.jar # 这个jar文件是一个自运行的jar包，因此为了使用方便，我们可以将其改名为presto mv presto-cli-0.211-executable.jar presto

通过 Presto 操作 MySQL 数据库

./presto --server bd1:8080 --catalog mysql --schema test presto:test> show schemas from mysql; Schema -------------------- information_schema performance_schema test (4 rows) Query 20181024_011707_00040_szyec, FINISHED, 2 nodes Splits: 19 total, 19 done (100.00%) 0:00 [4 rows, 64B] [9 rows/s, 147B/s] presto:test> show tables from mysql.test; Table ------- user (1 row) Query 20181024_011716_00041_szyec, FINISHED, 2 nodes Splits: 19 total, 19 done (100.00%) 0:01 [1 rows, 18B] [0 rows/s, 16B/s] presto:test> select * from mysql.test.user; id | username | password ----+----------+----------- 1 | user1 | password1 2 | user2 | password2 3 | user3 | password3 (3 rows) Query 20181024_011724_00042_szyec, FINISHED, 1 node Splits: 17 total, 17 done (100.00%) 0:00 [3 rows, 0B] [7 rows/s, 0B/s] presto:test> insert into mysql.test.user values(4,'user4','password4'); INSERT: 1 row Query 20181024_011938_00046_szyec, FINISHED, 2 nodes Splits: 35 total, 35 done (100.00%) 0:03 [0 rows, 0B] [0 rows/s, 0B/s] presto:test> select * from mysql.test.user; id | username | password ----+----------+----------- 1 | user1 | password1 2 | user2 | password2 3 | user3 | password3 4 | user4 | password4 (4 rows) Query 20181024_011943_00047_szyec, FINISHED, 1 node Splits: 17 total, 17 done (100.00%) 0:01 [4 rows, 0B] [6 rows/s, 0B/s]