Impala负载均衡

如有不妥之处，欢迎随时留言沟通交流，谢谢~

Impala分为是三个组件，statestored/catalogd和impalad，其中statestored和catalogd是单点的，没有高可用的需求，因为这两个实例是无状态的，本身不存储任何数据，例如catalogd的数据存储在第三方数据库(例如mysql中)，statestore的数据全都存储在内存中，可以通过简单的主备的方式来实现高可用，本文最后会提到。正常情况下只有master提供服务，slave只是运行状态但是不接受任何请求，当master出现问题之后再slave提升为master提供服务。

而对于impalad节点，每一个节点都可以提供jdbc和thrift等服务，并且对于连接到该impalad的查询作为coordinator节点（需要消耗一定的内存和CPU）存在，为了保证每一个节点的负载的平衡需要对于这些impalad做一下均衡，负载均衡分为四层负载均衡和七层负载均衡，前者是针对运输层的，后者是针对应用层的，区别在于前者不需要了解应用协议，只需要对传输层收到的IP数据包进行转发，而后者需要了解应用协议的，而对于impalad这种SQL服务器，就需要使用SQL协议的代理，所以七层代理对于impalad是有点不切实际的。

主要用的就是haproxy四层交换机的特性，讲所有指向haproxy主机和端口的请求，转发到相应的主机：端口上。

1、安装haproxy

yum install haproxy

2、配置文件

vim /etc/haproxy/haproxy.cfg

文件内容

global
    # To have these messages end up in /var/log/haproxy.log you will
    # need to:
    #
    # 1) configure syslog to accept network log events.  This is done
    #    by adding the '-r' option to the SYSLOGD_OPTIONS in
    #    /etc/sysconfig/syslog
    #
    # 2) configure local2 events to go to the /var/log/haproxy.log
    #   file. A line like the following can be added to
    #   /etc/sysconfig/syslog
    #
    #    local2.*                       /var/log/haproxy.log
    #
    log         127.0.0.1 local0
    log         127.0.0.1 local1 notice
    chroot      /var/lib/haproxy
    pidfile     /var/run/haproxy.pid
    maxconn     4000
    user        haproxy
    group       haproxy
    daemon

    # turn on stats unix socket
    #stats socket /var/lib/haproxy/stats

#---------------------------------------------------------------------
# common defaults that all the 'listen' and 'backend' sections will
# use if not designated in their block
#
# You might need to adjust timing values to prevent timeouts.
#---------------------------------------------------------------------
defaults
    mode                    http
    log                     global
    option                  httplog
    option                  dontlognull
    option http-server-close
    option forwardfor       except 127.0.0.0/8
    option                  redispatch
    retries                 3
    maxconn                 3000

    #连接时间需要修改，因为如果时间较短的话会出现，任务在执行，但是连接已经断开所以获取不到结果的情况。
    timeout connect 3600000ms
    timeout client 3600000ms
    timeout server 3600000ms

#
# This sets up the admin page for HA Proxy at port 25002.
#
listen stats :25002
    balance
    mode http
    stats enable
    stats auth admin:admin

#配置修改
listen impala-shell :25003
    mode tcp
    option tcplog
    balance leastconn

    server impala2 impala-host02:21000
    server impala3 impala-host03:21000
    server impala4 impala-host04:21000
    server impala5 impala-host05:21000

# This is the setup for Impala. Impala client connect to load_balancer_host:25003.
# HAProxy will balance connections among the list of servers listed below.
# The list of Impalad is listening at port 21000 for beeswax (impala-shell) or original ODBC driver.
# For JDBC or ODBC version 2.x driver, use port 21050 instead of 21000.

#配置修改
listen impala-jdbc:25001

    # impala负载均衡需要第四层的，所以填tcp
    mode tcp
    option tcplog
    balance leastconn

    #主机列表
    #impala-host02是impala的主机列表
    #由于是要对jdbc的请求进行转发，所以端口设置的是21050
    server impala2 impala-host02:21050
    server impala3 impala-host03:21050
    server impala4 impala-host04:21050
    server impala5 impala-host05:21050
                                    

3、启动

/usr/sbin/haproxy  -f   /etc/haproxy/haproxy.cfg

4、负载连接

haproxy-host-ip是haproxy的ip地址，主机名也可以，负载连接时用

impala-shell:haproxy-host-ip:25003  
impala-jdbc:haproxy-host-ip:25001


在haproxy上，使用impala-shell（localhost可以改成haproxy-host-ip）：

[email protected]:~$ impala-shell -i localhost:25003
Starting Impala Shell without Kerberos authentication
Connected to localhost:25003
Server version: impalad version 2.9.0-cdh5.12.0 RELEASE (build 03c6ddbdcec39238be4f5b14a300d5c4f576097e)
***********************************************************************************
Welcome to the Impala shell.
(Impala Shell v2.11.0-cdh5.14.2 (ed85dce) built on Tue Mar 27 13:39:48 PDT 2018)

You can run a single query from the command line using the '-q' option.
***********************************************************************************
[localhost:25003] > select * from t_sec_hurst_bs_dim limit 1;
Query: select * from t_sec_hurst_bs_dim limit 1
Query submitted at: 2018-07-25 20:16:07 (Coordinator: http://bigdata04:25000)
ERROR: AnalysisException: Could not resolve table reference: 't_sec_hurst_bs_dim'