部署步骤
1. 下载最新发布版#
2. 解压配置#
目录结构
README.mdbin/ # 启动脚本conf/ # 配置目录ext/ # 默认的外部jar包路径lib/ # 本身依赖share/ # java doc
⚠️ 目前对 Zookeeper 有强依赖, 需要提前安装好, 修改配置文件
conf/application.conf
必须要改的
pea.zk.connectString改成Zookeeper 的链接串地址. 其他都有默认值
# https://www.playframework.com/documentation/latest/Configurationinclude "framework.conf"
play.http.secret.key = "wuae_/xG6QUxPLWvXneCm8TH:b]Ki`?Hm0mOom`uahFh3xgTg8[9R_dfjCdpkVPG"play.http.secret.key = ${?APPLICATION_SECRET}
pea { // 注册到 zk 上的ip, 默认会选择第一个网卡的ip, 可指定, 可以通过 ADDRESS 环境变量配置 address = ${?ADDRESS} // 同上, 端口号 port = ${?PORT} // 机器标签名 label = "Pea" label = ${?LABEL} simulations { // 是否在启动时, 编译脚本 compileAtStartup = true webEditorBaseUrl = "https://github.com/ojlm/pea-simulations/blob/master/src/main/scala/" webEditorBaseUrl = ${?WEB_EDITOR_BASE} } results { // 报告生成的目录 folder = "./logs" folder = ${?RESULTS_FOLDER} report { logo.href = "https://github.com/ojlm/pea" desc.href = "https://github.com/ojlm/pea" desc.content = "https://github.com/ojlm/pea" } } zk { enabled = true // 当前节点为 `worker` 角色, 发压节点 role.worker = true role.worker = ${?ROLE_WORKER} // 当前节点为 `reporter` 角色, 聚合发压节点报告, 监控状态. `reporter` 节点会保存聚合报告 role.reporter = true role.reporter = ${?ROLE_REPORTER} // 注册到 zk 上的目录 path = "/pea" // zk 连接串 connectString = "localhost:2181" // zk 用户名 username = "" // zk 密码 password = "" } worker { // 现在 `reporter` 通过 http 接口控制 `worker`, 看自己的部署环境可以改成 `https` protocol = "http" // simulation source folder, 脚本的源码路径 source = "./test/simulations" source = ${?SIMULATIONS_SOURCE} // 脚本编译后的输出目录 output = "./logs/output" output = ${?SIMULATIONS_OUTPUT} // 资源目录, 用户数据, csv, json, 等文件 resources = "." resources = ${?RESOURCES_FOLDER} // 额外的jar包路径, 默认是 ext 目录 classpath = "" classpath = ${?EXTRA_CLASSPATH} }}启动
前端运行: windows 用户执行:
./bin/pea.bat, linux:./bin/pea
后台运行(linux shell): 启动 :
./bin/pea.sh start, 停止 :./bin/pea.sh stop