风险提示
- 以
v3.0.0-beta.1
版本为例进行部署; - 以 CentOS-7 为例,系统基础配置要求 4C-8G;
- 部署完成后,可通过浏览器:
IP:PORT
进行访问,默认端口是8080
,系统默认账号密码:admin
/admin2022_
。 v3.0.0-beta.2
版本开始,默认账号密码为admin
/admin
;- 本文为单机部署,如需分布式部署,请联系我们
软件依赖
软件名 | 版本要求 | 默认端口 |
---|---|---|
MySQL | v5.7 或 v8.0 | 3306 |
ElasticSearch | v7.6+ | 8060 |
JDK | v8+ | - |
CentOS | v6+ | - |
Ubuntu | v16+ | - |
在线安装
# 在服务器中下载安装脚本, 该脚本中会在当前目录下,重新安装MySQL。重装后的mysql密码存放在当前目录的mysql.password文件中。
wget https://s3-gzpu.didistatic.com/pub/knowstreaming/deploy_KnowStreaming-3.0.0-beta.1.sh
# 执行脚本
sh deploy_KnowStreaming.sh
# 访问地址
127.0.0.1:8080
离线安装
# 将安装包下载到本地且传输到目标服务器
wget https://s3-gzpu.didistatic.com/pub/knowstreaming/KnowStreaming-3.0.0-beta.1-offline.tar.gz
# 解压安装包
tar -zxf KnowStreaming-3.0.0-beta.1-offline.tar.gz
# 执行安装脚本
sh deploy_KnowStreaming-offline.sh
# 访问地址
127.0.0.1:8080
环境依赖
-
Kubernetes >= 1.14 ,Helm >= 2.17.0
-
默认依赖全部安装,ElasticSearch(3 节点集群模式) + MySQL(单机) + KnowStreaming-manager + KnowStreaming-ui
-
使用已有的 ElasticSearch(7.6.x) 和 MySQL(5.7) 只需调整 values.yaml 部分参数即可
安装命令
# 相关镜像在Docker Hub都可以下载
# 快速安装(NAMESPACE需要更改为已存在的,安装启动需要几分钟初始化请稍等~)
helm install -n [NAMESPACE] [NAME] http://download.knowstreaming.com/charts/knowstreaming-manager-0.1.5.tgz
# 获取KnowStreaming前端ui的service. 默认nodeport方式.
# (http://nodeIP:nodeport,默认用户名密码:admin/admin2022_)
# `v3.0.0-beta.2`版本开始(helm chart包版本0.1.4开始),默认账号密码为`admin` / `admin`;
# 添加仓库
helm repo add knowstreaming http://download.knowstreaming.com/charts
# 拉取最新版本
helm pull knowstreaming/knowstreaming-manager
环境依赖
安装命令
# `v3.0.0-beta.2`版本开始(docker镜像为0.2.0版本开始),默认账号密码为`admin` / `admin`;
# https://hub.docker.com/u/knowstreaming 在此处寻找最新镜像版本
# mysql与es可以使用自己搭建的服务,调整对应配置即可
# 复制docker-compose.yml到指定位置后执行下方命令即可启动
docker-compose up -d
验证安装
docker-compose ps
# 验证启动 - 状态为 UP 则表示成功
Name Command State Ports
----------------------------------------------------------------------------------------------------
elasticsearch-single /usr/local/bin/docker-entr ... Up 9200/tcp, 9300/tcp
knowstreaming-init /bin/bash /es_template_cre ... Up
knowstreaming-manager /bin/sh /ks-start.sh Up 80/tcp
knowstreaming-mysql /entrypoint.sh mysqld Up (health: starting) 3306/tcp, 33060/tcp
knowstreaming-ui /docker-entrypoint.sh ngin ... Up 0.0.0.0:80->80/tcp
# 稍等一分钟左右 knowstreaming-init 会退出,表示es初始化完成,可以访问页面
Name Command State Ports
-------------------------------------------------------------------------------------------
knowstreaming-init /bin/bash /es_template_cre ... Exit 0
knowstreaming-mysql /entrypoint.sh mysqld Up (healthy) 3306/tcp, 33060/tcp
访问
http://127.0.0.1:80/
docker-compose.yml
version: "2"
services:
# *不要调整knowstreaming-manager服务名称,ui中会用到
knowstreaming-manager:
image: knowstreaming/knowstreaming-manager:latest
container_name: knowstreaming-manager
privileged: true
restart: always
depends_on:
- elasticsearch-single
- knowstreaming-mysql
expose:
- 80
command:
- /bin/sh
- /ks-start.sh
environment:
TZ: Asia/Shanghai
# mysql服务地址
SERVER_MYSQL_ADDRESS: knowstreaming-mysql:3306
# mysql数据库名
SERVER_MYSQL_DB: know_streaming
# mysql用户名
SERVER_MYSQL_USER: root
# mysql用户密码
SERVER_MYSQL_PASSWORD: admin2022_
# es服务地址
SERVER_ES_ADDRESS: elasticsearch-single:9200
# 服务JVM参数
JAVA_OPTS: -Xmx1g -Xms1g
# 对于kafka中ADVERTISED_LISTENERS填写的hostname可以通过该方式完成
# extra_hosts:
# - "hostname:x.x.x.x"
# 服务日志路径
# volumes:
# - /ks/manage/log:/logs
knowstreaming-ui:
image: knowstreaming/knowstreaming-ui:latest
container_name: knowstreaming-ui
restart: always
ports:
- '80:80'
environment:
TZ: Asia/Shanghai
depends_on:
- knowstreaming-manager
# extra_hosts:
# - "hostname:x.x.x.x"
elasticsearch-single:
image: docker.io/library/elasticsearch:7.6.2
container_name: elasticsearch-single
restart: always
expose:
- 9200
- 9300
# ports:
# - '9200:9200'
# - '9300:9300'
environment:
TZ: Asia/Shanghai
# es的JVM参数
ES_JAVA_OPTS: -Xms512m -Xmx512m
# 单节点配置,多节点集群参考 https://www.elastic.co/guide/en/elasticsearch/reference/7.6/docker.html#docker-compose-file
discovery.type: single-node
# 数据持久化路径
# volumes:
# - /ks/es/data:/usr/share/elasticsearch/data
# es初始化服务,与manager使用同一镜像
# 首次启动es需初始化模版和索引,后续会自动创建
knowstreaming-init:
image: knowstreaming/knowstreaming-manager:latest
container_name: knowstreaming-init
depends_on:
- elasticsearch-single
command:
- /bin/bash
- /es_template_create.sh
environment:
TZ: Asia/Shanghai
# es服务地址
SERVER_ES_ADDRESS: elasticsearch-single:9200
knowstreaming-mysql:
image: knowstreaming/knowstreaming-mysql:latest
container_name: knowstreaming-mysql
restart: always
environment:
TZ: Asia/Shanghai
# root 用户密码
MYSQL_ROOT_PASSWORD: admin2022_
# 初始化时创建的数据库名称
MYSQL_DATABASE: know_streaming
# 通配所有host,可以访问远程
MYSQL_ROOT_HOST: '%'
expose:
- 3306
# ports:
# - '3306:3306'
# 数据持久化路径
# volumes:
# - /ks/mysql/data:/data/mysql
部署流程
- 安装
JDK-11
、MySQL
、ElasticSearch
等依赖服务 - 安装 KnowStreaming
yum 方式安装
# 配置yum源
wget https://dev.mysql.com/get/mysql57-community-release-el7-9.noarch.rpm
rpm -ivh mysql57-community-release-el7-9.noarch.rpm
# 执行安装
yum -y install mysql-server mysql-client
# 服务启动
systemctl start mysqld
# 获取初始密码并修改
old_pass=`grep 'temporary password' /var/log/mysqld.log | awk '{print $NF}' | tail -n 1`
mysql -NBe "alter user USER() identified by 'Didi_km_678';" --connect-expired-password -uroot -p$old_pass
rpm 方式安装
# 下载安装包
wget https://s3-gzpu.didistatic.com/knowsearch/mysql5.7.tar.gz
# 解压到指定目录
tar -zxf mysql5.7.tar.gz -C /tmp/
# 执行安装
yum -y localinstall /tmp/libaio-*.rpm /tmp/mysql-*.rpm
# 服务启动
systemctl start mysqld
# 获取初始密码并修改
old_pass=`grep 'temporary password' /var/log/mysqld.log | awk '{print $NF}' | tail -n 1`
mysql -NBe "alter user USER() identified by 'Didi_km_678';" --connect-expired-password -uroot -p$old_pass
# 下载安装包
wget https://s3-gzpu.didistatic.com/pub/jdk11.tar.gz
# 解压到指定目录
tar -zxf jdk11.tar.gz -C /usr/local/
# 更改目录名
mv /usr/local/jdk-11.0.2 /usr/local/java11
# 添加到环境变量
echo "export JAVA_HOME=/usr/local/java11" >> ~/.bashrc
echo "export CLASSPATH=/usr/java/java11/lib" >> ~/.bashrc
echo "export PATH=$JAVA_HOME/bin:$PATH:$HOME/bin" >> ~/.bashrc
source ~/.bashrc
- ElasticSearch 用于存储平台采集的 Kafka 指标;
- 以下安装示例为单节点模式,如需集群部署可以参考:Elasticsearch 官方文档
# 下载安装包
wget https://s3-gzpu.didistatic.com/pub/elasticsearch.tar.gz
# 创建ES数据存储目录
mkdir -p /data/es_data
# 创建ES所属用户
useradd arius
# 配置用户的打开文件数
echo "arius soft nofile 655350" >> /etc/security/limits.conf
echo "arius hard nofile 655350" >> /etc/security/limits.conf
echo "vm.max_map_count = 655360" >> /etc/sysctl.conf
sysctl -p
# 解压安装包
tar -zxf elasticsearch.tar.gz -C /data/
# 更改目录所属组
chown -R arius:arius /data/
# 修改配置文件(参考以下配置)
vim /data/elasticsearch/config/elasticsearch.yml
cluster.name: km_es
node.name: es-node1
node.master: true
node.data: true
path.data: /data/es_data
http.port: 8060
discovery.seed_hosts: ["127.0.0.1:9300"]
# 修改内存配置
vim /data/elasticsearch/config/jvm.options
-Xms2g
-Xmx2g
# 启动服务
su - arius
export JAVA_HOME=/usr/local/java11
sh /data/elasticsearch/control.sh start
# 确认状态
sh /data/elasticsearch/control.sh status
# 下载安装包
wget https://s3-gzpu.didistatic.com/pub/knowstreaming/KnowStreaming-3.0.0-beta.1.tar.gz
# 解压安装包到指定目录
tar -zxf KnowStreaming-3.0.0-beta.1.tar.gz -C /data/
# 修改启动脚本并加入systemd管理
cd /data/KnowStreaming/
# 创建相应的库和导入初始化数据
mysql -uroot -pDidi_km_678 -e "create database know_streaming;"
mysql -uroot -pDidi_km_678 know_streaming < ./init/sql/ddl-ks-km.sql
mysql -uroot -pDidi_km_678 know_streaming < ./init/sql/ddl-logi-job.sql
mysql -uroot -pDidi_km_678 know_streaming < ./init/sql/ddl-logi-security.sql
mysql -uroot -pDidi_km_678 know_streaming < ./init/sql/dml-ks-km.sql
mysql -uroot -pDidi_km_678 know_streaming < ./init/sql/dml-logi.sql
# 创建elasticsearch初始化数据
sh ./bin/init_es_template.sh
# 修改配置文件
vim ./conf/application.yml
# 监听端口
server:
port: 8080 # web 服务端口
tomcat:
accept-count: 1000
max-connections: 10000
# ES地址
es.client.address: 127.0.0.1:8060
# 数据库配置(一共三处地方,修改正确的mysql地址和数据库名称以及用户名密码)
jdbc-url: jdbc:mariadb://127.0.0.1:3306/know_streaming?.....
username: root
password: Didi_km_678
# 启动服务
cd /data/KnowStreaming/bin/
sh startup.sh