Sentinel集群限流探索（二）

wg204wg

发布于 2022-6-13 17:37

5455浏览

0收藏

实现InitFunc接口，重写init方法，代码直接贴出来，这里整体依赖的是Apollo的配置方式，注释的部分是我在测试的时候写死代码的配置方式，也是可以用的。

public class DemoClusterInitFunc implements InitFunc {
    private final String namespace = "application";
    private final String ruleKey = "demo_sentinel";
    private final String ruleServerKey = "demo_cluster";
    private final String defaultRuleValue = "[]";

    @Override
    public void init() throws Exception {
        // 初始化 限流规则
        initDynamicRuleProperty();
        //初始化 客户端配置
        initClientConfigProperty();
        // 初始化 服务端配置信息
        initClientServerAssignProperty();
        registerClusterRuleSupplier();
        // token-server的传输规则
        initServerTransportConfigProperty();
        // 初始化 客户端和服务端状态
        initStateProperty();
    }

    /**
     * 限流规则和热点限流规则配置
     */
    private void initDynamicRuleProperty() {
        ReadableDataSource<String, List<FlowRule>> ruleSource = new ApolloDataSource<>(namespace, ruleKey,
                defaultRuleValue, source -> JSON.parseObject(source, new TypeReference<List<FlowRule>>() {
        }));
        FlowRuleManager.register2Property(ruleSource.getProperty());

        ReadableDataSource<String, List<ParamFlowRule>> paramRuleSource = new ApolloDataSource<>(namespace, ruleKey,
                defaultRuleValue, source -> JSON.parseObject(source, new TypeReference<List<ParamFlowRule>>() {
        }));
        ParamFlowRuleManager.register2Property(paramRuleSource.getProperty());
    }

    /**
     * 客户端配置，注释的部分是通过Apollo配置，只有一个配置我就省略了
     */
    private void initClientConfigProperty() {
//        ReadableDataSource<String, ClusterClientConfig> clientConfigDs = new ApolloDataSource<>(namespace, ruleKey,
//                defaultRuleValue, source -> JSON.parseObject(source, new TypeReference<ClusterClientConfig>() {
//        }));
//        ClusterClientConfigManager.registerClientConfigProperty(clientConfigDs.getProperty());

        ClusterClientConfig clientConfig = new ClusterClientConfig();
        clientConfig.setRequestTimeout(1000);
        ClusterClientConfigManager.applyNewConfig(clientConfig);
    }

    /**
     * client->server 传输配置，设置端口号，注释的部分是写死的配置方式
     */
    private void initServerTransportConfigProperty() {
        ReadableDataSource<String, ServerTransportConfig> serverTransportDs = new ApolloDataSource<>(namespace, ruleServerKey,
                defaultRuleValue, source -> {
            List<ClusterGroupEntity> groupList = JSON.parseObject(source, new TypeReference<List<ClusterGroupEntity>>() {
            });
            ServerTransportConfig serverTransportConfig = Optional.ofNullable(groupList)
                    .flatMap(this::extractServerTransportConfig)
                    .orElse(null);
            return serverTransportConfig;
        });
        ClusterServerConfigManager.registerServerTransportProperty(serverTransportDs.getProperty());
//        ClusterServerConfigManager.loadGlobalTransportConfig(new ServerTransportConfig().setIdleSeconds(600).setPort(transPort));
    }

    private void registerClusterRuleSupplier() {
        ClusterFlowRuleManager.setPropertySupplier(namespace -> {
            ReadableDataSource<String, List<FlowRule>> ds = new ApolloDataSource<>(this.namespace, ruleKey,
                    defaultRuleValue, source -> JSON.parseObject(source, new TypeReference<List<FlowRule>>() {
            }));
            return ds.getProperty();
        });
        ClusterParamFlowRuleManager.setPropertySupplier(namespace -> {
            ReadableDataSource<String, List<ParamFlowRule>> ds = new ApolloDataSource<>(this.namespace, ruleKey,
                    defaultRuleValue, source -> JSON.parseObject(source, new TypeReference<List<ParamFlowRule>>() {
            }));
            return ds.getProperty();
        });
    }

    /**
     * 服务端配置，设置server端口和IP，注释的配置是写死的方式，这个在服务端是不用配置的，只有客户端需要配置用来连接服务端
     */
    private void initClientServerAssignProperty() {
        ReadableDataSource<String, ClusterClientAssignConfig> clientAssignDs = new ApolloDataSource<>(namespace, ruleServerKey,
                defaultRuleValue, source -> {
            List<ClusterGroupEntity> groupList = JSON.parseObject(source, new TypeReference<List<ClusterGroupEntity>>() {
            });

            ClusterClientAssignConfig clusterClientAssignConfig = Optional.ofNullable(groupList)
                    .flatMap(this::extractClientAssignment)
                    .orElse(null);
            return clusterClientAssignConfig;
        });
        ClusterClientConfigManager.registerServerAssignProperty(clientAssignDs.getProperty());

//        ClusterClientAssignConfig serverConfig = new ClusterClientAssignConfig();
//        serverConfig.setServerHost("127.0.0.1");
//        serverConfig.setServerPort(transPort);
//        ConfigSupplierRegistry.setNamespaceSupplier(() -> "trade-center");
//        ClusterClientConfigManager.applyNewAssignConfig(serverConfig);
    }

    private Optional<ClusterClientAssignConfig> extractClientAssignment(List<ClusterGroupEntity> groupList) {
        ClusterGroupEntity tokenServer = groupList.stream().filter(x -> x.getState().equals(ClusterStateManager.CLUSTER_SERVER)).findFirst().get();
        Integer currentMachineState = Optional.ofNullable(groupList).map(s -> groupList.stream().filter(this::machineEqual).findFirst().get().getState()).orElse(ClusterStateManager.CLUSTER_NOT_STARTED);
        if (currentMachineState.equals(ClusterStateManager.CLUSTER_CLIENT)) {
            String ip = tokenServer.getIp();
            Integer port = tokenServer.getPort();
            return Optional.of(new ClusterClientAssignConfig(ip, port));
        }
        return Optional.empty();
    }

    /**
     * 初始化客户端和服务端状态，注释的也是写死的配置方式
     */
    private void initStateProperty() {
        ReadableDataSource<String, Integer> clusterModeDs = new ApolloDataSource<>(namespace, ruleServerKey,
                defaultRuleValue, source -> {
            List<ClusterGroupEntity> groupList = JSON.parseObject(source, new TypeReference<List<ClusterGroupEntity>>() {
            });
            Integer state = Optional.ofNullable(groupList).map(s -> groupList.stream().filter(this::machineEqual).findFirst().get().getState()).orElse(ClusterStateManager.CLUSTER_NOT_STARTED);
            return state;
        });
        ClusterStateManager.registerProperty(clusterModeDs.getProperty());

//            ClusterStateManager.applyState(ClusterStateManager.CLUSTER_SERVER);

    }

    private Optional<ServerTransportConfig> extractServerTransportConfig(List<ClusterGroupEntity> groupList) {
        return groupList.stream()
                .filter(x -> x.getMachineId().equalsIgnoreCase(getCurrentMachineId()) && x.getState().equals(ClusterStateManager.CLUSTER_SERVER))
                .findAny()
                .map(e -> new ServerTransportConfig().setPort(e.getPort()).setIdleSeconds(600));
    }

    private boolean machineEqual(/*@Valid*/ ClusterGroupEntity group) {
        return getCurrentMachineId().equals(group.getMachineId());
    }

    private String getCurrentMachineId() {
        // 通过-Dcsp.sentinel.api.port=8719 配置， 默认8719，随后递增
        return HostNameUtil.getIp() + SEPARATOR + TransportConfig.getPort();
    }
  
    private static final String SEPARATOR = "@";
}1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
149.
150.
151.

基础类，定义配置的基础信息。

@Data
public class ClusterGroupEntity {
    private String machineId;
    private String ip;
    private Integer port;
    private Integer state;
}1.
2.
3.
4.
5.
6.
7.

然后是Apollo中的限流规则的配置和server/client集群关系的配置。

需要说明一下的就是flowId，这个是区分限流规则的全局唯一ID，必须要有，否则集群限流会有问题。

thresholdType代表限流模式，默认是0，代表单机均摊，比如这里count限流QPS=20，有3台机器，那么集群限流阈值就是60，如果是1代表全局阈值，也就是count配置的值就是集群限流的上限。

demo_sentinel=[
    {
        "resource": "test_res", //限流资源名
        "count": 20, //集群限流QPS
        "clusterMode": true, //true为集群限流模式
        "clusterConfig": {
            "flowId": 111, //这个必须得有，否则会有问题
            "thresholdType": 1 //限流模式，默认为0单机均摊，1是全局阈值
        }
    }
]
demo_cluster=[
    {
        "ip": "192.168.3.20",
        "machineId": "192.168.3.20@8720",
        "port": 9999, //server和client通信接口
        "state": 1 //指定为server
    },
    {
        "ip": "192.168.3.20",
        "machineId": "192.168.3.20@8721",
        "state": 0
    },
    {
        "ip": "192.168.3.20",
        "machineId": "192.168.3.20@8722",
        "state": 0
    }
]1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.

OK，到这里代码和配置都已经OK，还需要跑起来Sentinel控制台，这个不用教，还有启动参数。

本地可以直接跑多个客户端，注意修改端口号：-Dserver.port=9100 -Dcsp.sentinel.api.port=8720这两个一块改，至于怎么连Apollo这块我就省略了，自己整吧，公司应该都有，不行的话用代码里的写死的方式也可以用。

-Dserver.port=9100 -Dcsp.sentinel.api.port=8720 -Dcsp.sentinel.dashboard.server=localhost:8080 -Dcsp.sentinel.log.use.pid=true 1.

文章转自公众号：艾小仙

分类

其他

标签

Sentinel

已于2022-6-13 17:37:51修改

51CTO

51CTO博客

51CTO学堂

Sentinel集群限流探索（二）

订阅鸿蒙技术特刊，精选内容抢先看