2017-12-12

codis安装及配置

一、环境安装

系统环境centos

Golang 环境搭建: yum install go
codis 下载和编译:

go get -u -d github.com/CodisLabs/codis
cd $GOPATH/src/github.com/CodisLabs/codis
make
make gotest
mkdir -p /usr/local/codis/{logs,conf,scripts}
cp –rf bin /usr/local/codis/
cp config.ini /usr/local/codis/conf/

Codis-HA 编译
Codis-HA。这是一个通过 Codis 开放的 api 实现自动切换主从的工具。该工具会在检测
到 master 挂掉的时候将其下线并选择其中一个 slave 提升为 master 继续提供服务。
go get github.com/ngaut/codis-ha
cd codis-ha
go build
codis-ha –codis-config=localhost:18087 –productName=test
Zookeeper集群搭建

首先安装开发工具及openjdk，zookeeper是由Java语言开发的，所以需要openjdk环境。

首先安装开发工具及openjdk，zookeeper是由Java语言开发的，所以需要openjdk环境。

1
2
3

yum groupinstall "Development tools" "Compatibility libraries" -y 
yum install openssl-devel openssl -y
yum install java-1.8.0-openjdk-devel java-1.8.0-openjdk -y

确定Java运行环境正常

java -version
  openjdk version "1.8.0101"
  OpenJDK Runtime Environment (build 1.8.0101-b13)
  OpenJDK 64-Bit Server VM (build 25.101-b13, mixed mode)

安装二进制版本的zookeeper

tar xvf zookeeper-3.4.9.tar.gz -C /usr/local/ 
ln -s /usr/local/zookeeper-3.4.9/ /usr/local/zookeeper 
cd /usr/local/zookeeper/conf 
cp zoo_sample.cfg zoo.cfg

编译zookeeper配置文件/usr/local/zookeeper/conf/zoo.cfg

maxClientCnxns=60
tickTime=2000
initLimit=10
syncLimit=5
dataDir=/data/zookeeper/db
dataLogDir=/data/zookeeper/log
clientPort=2181
# cluster configure
server.1=10.173.225.60:2888:3888
server.2=10.174.33.81:2888:3888
server.3=10.173.224.34:2888:3888

1	mkdir /data/zookeeper/{db,log} -p

其中2888表示zookeeper程序监听端口，3888表示zookeeper选举通信端口。

下面需要生成ID，这里需要注意，myid对应的zoo.cfg的server.ID，比如第二台zookeeper主机对应的myid应该是2，以此类推，三个主机分别为：

1
2
3

10.173.225.60 echo 1 > /data/zookeeper/db/myid
10.174.33.81 echo 2 > /data/zookeeper/db/myid
10.173.224.34 echo 3 > /data/zookeeper/db/myid

然后输出环境变量。

1 2	export PATH=$PATH:/usr/local/zookeeper/bin/ source /etc/profile

然后就可以启动zookeeper了。

1	zkServer.sh start

查看各个zookeeper节点的状态（会有一个leader节点，两个follower节点）。

[root@node1 ~]# zkServer.sh status 
	Mode: follower
[root@node2 ~]# zkServer.sh status 
	Mode: leader
[root@node3 ~]# zkServer.sh status 
	Mode: follower

客户端连接，可以查看相关信息。

1	zkCli.sh -server 127.0.0.1:2181

至此，zookeeper已经搞定了。

二、启动服务

再编译后的codis文件夹下有启动服务的脚本

启动codis-dashboard

使用 codis-dashboard-admin.sh 脚本启动 dashboard，并查看 dashboard 日志确认启动是否有异常。

dashboard只需要启动一个

配置文件dashboard.toml

# Set Coordinator, only accept "zookeeper" & "etcd" & "filesystem".
# for zookeeper/etcd, coorinator_auth accept "user:password"
# Quick Start
coordinator_name = "zookeeper"
coordinator_addr = "10.173.225.60:2181,10.174.33.81:2181,10.173.224.34:2181"
#coordinator_name = "zookeeper"
#coordinator_addr = "127.0.0.1:2181"
#coordinator_auth = ""

# Set Codis Product Name/Auth.
product_name = "codis-demo"
product_auth = ""

# Set bind address for admin(rpc), tcp only.
admin_addr = "0.0.0.0:18080"

1 2	./admin/codis-dashboard-admin.sh start tail -100 ./log/codis-dashboard.log.2017-04-08

启动codis-fe

1	./bin/codis-admin --dashboard-list --zookeeper=127.0.0.1:2181 \| tee $Gopath/src/github.com/CodisLabs/codis/config/codis.json

codis.json

[
{
    "name": "codis-demo",
    "dashboard": "10.174.33.81:18080"
 }
 ]

启动codis-fe

1 2	nohup `which codis-fe` --ncpu=2 --log=/data/codis/log/fe.log --log-level=WARN --dashboard-list=$Gopath/src/github.com/CodisLabs/codis/config/codis.json --listen=0.0.0.0:8080 &

启动codis-proxy

使用 codis-proxy-admin.sh 脚本启动 codis-proxy，并查看 proxy 日志确认启动是否有异常。(代理可以启动一个，也可以启动多个，但是启动的多个代理的配置必须是一样的，是同一个dashboard的地址)

配置文件proxy.toml

19 # Set bind address for admin(rpc), tcp only.

20 admin_addr = “10.173.225.60:11080”
21
22 # Set bind address for proxy, proto_type can be “tcp”, “tcp4”, “tcp6”, “unix” or “unixpacket”.
23 proto_type = “tcp4”
24 proxy_addr = “10.173.225.60:19000”
26 # Set jodis address & session timeout
27 # 1. jodis_name is short for jodis_coordinator_name, only accept “zookeeper” & “etcd”.
28 # 2. jodis_addr is short for jodis_coordinator_addr
29 # 3. jodis_auth is short for jodis_coordinator_auth, for zookeeper/etcd, “user:password” is accepted.
30 # 4. proxy will be registered as node:
31 # if jodiscompatible = true (not suggested):32 # /zk/codis/db{PRODUCT_NAME}/proxy-{HASHID} (compatible with Codis2.0)
33 # or else
34 # /jodis/{PRODUCT_NAME}/proxy-{HASHID}
35 jodis_name = “”
3 # jodis_addr不能写地址，不然启动报错，不知道为啥，可能没有安装jodis,
36 jodis_addr = “”
37 jodis_auth = “”
38 jodis_timeout = “20s”

1 2	./admin/codis-proxy-admin.sh start tail -100 ./log/codis-proxy.log.2017-04-08

要启动多个代理需要修改脚本

vi codis-proxy-admin.sh

1	CODIS_DASHBOARD_ADDR="10.173.225.60:18080"

启动codis-server

使用 codis-server-admin.sh 脚本启动 codis-server，并查看 redis 日志确认启动是否有异常。

配置redis.conf 和sentinel.conf

redis.conf

48 # bind 127.0.0.1 ::1
49 #
50 # WARNING If the computer running Redis is directly exposed to the
51 # internet, binding to all the interfaces is dangerous and will expose the
52 # instance to everybody on the internet. So by default we uncomment the
53 # following bind directive, that will force Redis to listen only into
54 # the IPv4 lookback interface address (this means Redis will be able to
55 # accept connections only from clients running into the same computer it
56 # is running).
57 #
58 # IF YOU ARE SURE YOU WANT YOUR INSTANCE TO LISTEN TO ALL THE INTERFACES
59 # JUST COMMENT THE FOLLOWING LINE.
60 # ~~
61 bind 0.0.0.0
…….
…….
82 # Accept connections on the specified port, default is 6379 (IANA #815344).
83 # If port 0 is specified Redis will not listen on a TCP socket.
84 port 6379

sentinel.conf

48 # bind 127.0.0.1 ::1
49 #
50 # WARNING If the computer running Redis is directly exposed to the
51 # internet, binding to all the interfaces is dangerous and will expose the
52 # instance to everybody on the internet. So by default we uncomment the
53 # following bind directive, that will force Redis to listen only into
54 # the IPv4 lookback interface address (this means Redis will be able to
55 # accept connections only from clients running into the same computer it
56 # is running).
57 #
58 # IF YOU ARE SURE YOU WANT YOUR INSTANCE TO LISTEN TO ALL THE INTERFACES
59 # JUST COMMENT THE FOLLOWING LINE.
60 #~~
61 bind 0.0.0.0
…….
…….
82 # Accept connections on the specified port, default is 6379 (IANA #815344).
83 # If port 0 is specified Redis will not listen on a TCP socket.
84 port 26379

1 2	./admin/codis-server-admin.sh start tail -100 /tmp/redis_6379.log

redis.conf 配置中 pidfile、logfile 默认保存在 /tmp 目录，若启动失败，请检查当前用户是否有该目录的读写权限。

通过fe添加group

通过web浏览器访问集群管理页面(fe地址:127.0.0.1:8080) 选择我们刚搭建的集群 codis-demo，在 Proxy 栏可看到我们已经启动的 Proxy，但是 Group 栏为空，因为我们启动的 codis-server 并未加入到集群添加 NEW GROUP，NEW GROUP 行输入 1，再点击 NEW GROUP 即可添加 Codis Server，Add Server 行输入我们刚刚启动的 codis-server 地址，添加到我们刚新建的 Group，然后再点击 Add Server 按钮即可，如下图所示

通过fe初始化slot

新增的集群 slot 状态是 offline，因此我们需要对它进行初始化（将 1024 个 slot 分配到各个 group），而初始化最快的方法可通过 fe 提供的 rebalance all slots 按钮来做，如下图所示，点击此按钮，我们即快速完成了一个集群的搭建。

每次增加组之后就需要重新执行Rebalance All Slots

本文标题:codis安装及配置

文章作者:四娃

发布时间:2017-12-12, 16:21:21

最后更新:2018-08-10, 16:27:07

原始链接:http://zhqqqy.github.io/2017/12/12/codis安装及配置/

许可协议: "署名-非商用-相同方式共享 4.0" 转载请保留原文链接及作者。