构建Zeebe作为下一代工作流引擎，用于新兴用例，例如微服务编排用例，这些用例可能需要引擎每秒处理数十万（或数百万）个新工作流实例。为此，我们使用的图形建模标准已经存在了近15年：BPMN（业务流程模型和表示法）。可以预见，未来微服务编排将会是中台服务化最有力的落地工具之一。

借这个机会和大家分享一下基本的Zeebe部署和一个简单的范例。

准备环境

1.先克隆Git上的repository获得Zeebe + Operate in Docker的文件包
git clone https://github.com/zeebe-io/zeebe-docker-compose
2.解压下载的zeebe-docker-compose-master.zip
3.执行基于docker-compose.yml下载并创建相关组件容器
# change to directory of the profile to start, i.e
# cd operate/
docker-compose up

备注： docker-compose down 用来shutdown容器

例如：

(base) MacBookPro:operate zhengr$ docker-compose up

Pulling operate (camunda/operate:1.0.0-alpha11)...

1.0.0-alpha11: Pulling from camunda/operate

6f2f362378c5: Pull complete

494c27a8a6b8: Pull complete

7596bb83081b: Pull complete

1e739bce2743: Pull complete

4dde2a90460d: Pull complete

1f5b8585072c: Pull complete

883cfe76ed71: Pull complete

c025c61085ef: Pull complete

b060cd7d08ce: Pull complete

93da9e8063ec: Pull complete

Pulling elasticsearch (docker.elastic.co/elasticsearch/elasticsearch-oss:6.7.1)...

6.7.1: Pulling from elasticsearch/elasticsearch-oss

8ba884070f61: Pull complete

2202e0dee1cc: Pull complete

046487a6b47b: Pull complete

2bf203bafcf4: Pull complete

ecabf552c89a: Pull complete

93123b7bf511: Pull complete

2867102d5376: Pull complete

Pulling zeebe (camunda/zeebe:0.18.0)...

0.18.0: Pulling from camunda/zeebe

de8cd8ac2c2a: Pull complete

62def3ffad8c: Pull complete

1dafe0169f91: Pull complete

f4521a1bc960: Pull complete

0c2a2533a883: Pull complete

dfd212cc6b74: Pull complete

65e222ddee27: Pull complete

ce1bf1781edd: Pull complete

4c631900e34b: Pull complete

55bebb03879a: Pull complete

6b22452bf891: Pull complete

8e4a1a2f6114: Pull complete

Pulling kibana (docker.elastic.co/kibana/kibana-oss:6.7.1)...

6.7.1: Pulling from kibana/kibana-oss

8ba884070f61: Already exists

9de755da2607: Pull complete

a530113b5366: Pull complete

fcc058e5c07b: Pull complete

b01d58af14f6: Pull complete

74a53cb7bf61: Pull complete

12a05e912a4b: Pull complete

94203c8a1395: Pull complete

Creating operate_kibana_1        ... done

Creating operate_elasticsearch_1 ... done

Creating operate_operate_1       ... done

Creating operate_zeebe_1         ... done

Attaching to operate_kibana_1, operate_operate_1, operate_elasticsearch_1, operate_zeebe_1
备注：
  • bin/zbctl* - cli binary to interact with the broker
服务 / 端口

我们启动的docker容器将提供如下服务:

4.检查服务状态
(base) MacBookPro:bin zhengr$ ./zbctl.darwin status

Cluster size: 1

Partitions count: 1

Replication factor: 1

Brokers:

  Broker 0 - 172.18.0.5:26501

​    Partition 1 : Leader

以上容器部署完成后， Zeebe的服务框架就准备好了，接下来我们做范例流程部署。

流程部署

The zbctl binary is included to allow you to interact with the running broker. The binary is named:

Operating System zbctl binary
Linux bin/zbctl
OS X bin/zbctl.darwin
Windows bin/zbctl.exe 
cd zeebe-docker-compose
bin/zbctl deploy ../path/to/your-file.bpmn

这里我们用一个已经制作完成的BPMN文件部署测试。order-process.bpmn

order-process

这个order-process.bpmn可以用zeebe Modeler可视化编排

image-20190628115030145

(base) MacBookPro:bin zhengr$ ./zbctl.darwin deploy order-process.bpmn
{
  "key": 2251799813685259,
  "workflows": [
    {
      "bpmnProcessId": "order-process",
      "version": 1,
      "workflowKey": 2251799813685251,
      "resourceName": "order-process.bpmn"
    }
  ]
}

创建流程实例

./bin/zbctl create instance order-process --variables '{"orderId": XXX}'

例如：

(base) MacBookPro:bin zhengr$ ./zbctl.darwin create instance order-process --variables '{"orderId": 45678}'

{

  "workflowKey": 2251799813685258,

  "bpmnProcessId": "order-process",

  "version": 2,

  "workflowInstanceKey": 2251799813685260

}

打开 http://localhost:8080 通过Operate检查部署实例情况：

Jietu20190628-122341

image-20190628120128229