本文阅读nacos-2.0.2的config源码,编写示例,分析推送配置、监听配置的原理。

客户端

创建NacosConfigService对象

Properties properties = new Properties();
properties.setProperty(PropertyKeyConst.SERVER_ADDR, NACOS_HOST);
NacosConfigService configService = new NacosConfigService(properties);

构造方法:

public NacosConfigService(Properties properties) throws NacosException {
    ValidatorUtils.checkInitParam(properties);
    
    initNamespace(properties);
    this.configFilterChainManager = new ConfigFilterChainManager(properties);
    ServerListManager serverListManager = new ServerListManager(properties);
    serverListManager.start();
    
    this.worker = new ClientWorker(this.configFilterChainManager, serverListManager, properties);
    // will be deleted in 2.0 later versions
    agent = new ServerHttpAgent(serverListManager);
}
  1. 创建ConfigFilterChainManager - 过滤器链
  2. 创建ServerListManager - 服务器列表管理
  3. 创建ClientWorker - 用来发送请求,内部封装了一个ConfigRpcTransportClient类型对象agent,它能够获取到RpcClient与服务端进行通信

推送配置

Properties properties = new Properties();
properties.setProperty(PropertyKeyConst.SERVER_ADDR, NACOS_HOST);

NacosConfigService configService = new NacosConfigService(properties);

StringWriter out = new StringWriter();
properties.store(out, "test config");

// 推送配置到nacos服务器
configService.publishConfig(
    ORDER_SERVICE, Constants.DEFAULT_GROUP, out.toString(), "properties");

推送配置到nacos服务器:

public boolean publishConfig(String dataId,
                             String group,
                             String content,
                             String type) throws NacosException {
    return publishConfigInner(namespace, dataId, group, null, null, null, content, type, null);
}

private boolean publishConfigInner(String tenant, String dataId, String group, String tag, String appName,
        String betaIps, String content, String type, String casMd5) throws NacosException {
    group = blank2defaultGroup(group);
    ParamUtils.checkParam(dataId, group, content);

    ConfigRequest cr = new ConfigRequest();
    cr.setDataId(dataId);
    cr.setTenant(tenant);
    cr.setGroup(group);
    cr.setContent(content);
    cr.setType(type);
    configFilterChainManager.doFilter(cr, null);
    content = cr.getContent();
    String encryptedDataKey = (String) cr.getParameter("encryptedDataKey");

    return worker.publishConfig(
        dataId, group, tenant, appName, tag, betaIps, content, encryptedDataKey, casMd5, type);
}

worker使用agent推送配置:

// 1. 封装ConfigPublishRequest对象
ConfigPublishRequest request = new ConfigPublishRequest(dataId, group, tenant, content);
request.setCasMd5(casMd5);
request.putAdditionalParam(TAG_PARAM, tag);
request.putAdditionalParam(APP_NAME_PARAM, appName);
request.putAdditionalParam(BETAIPS_PARAM, betaIps);
request.putAdditionalParam(TYPE_PARAM, type);
request.putAdditionalParam(ENCRYPTED_DATA_KEY_PARAM, encryptedDataKey);
// 2. 获取RpcClient对象
// 3. 使用RpcClient发请求
ConfigPublishResponse response = (ConfigPublishResponse) requestProxy(getOneRunningClient(), request);

监听配置

Properties properties = new Properties();
properties.setProperty(PropertyKeyConst.SERVER_ADDR, NACOS_HOST);

NacosConfigService configService = new NacosConfigService(properties);

// 添加监听器
configService.addListener(ORDER_SERVICE, Constants.DEFAULT_GROUP, new AbstractListener() {
  @Override
  public void receiveConfigInfo(String configInfo) {
    System.out.printf(">> config: n%snn", configInfo);
  }
});
  1. 将监听器注册到本地,本地使用CacheData作为监听器管理器,封装配置名、dataId和监听器集合,内部使用CopyOnWriteArrayList保存监听器,如果是第一次监听,会先拉取一次配置。本地注册表为Map结构,使用dataId+group+tenant作为key,value是CacheData对象
  2. 在client初始化阶段,会注册一个ServerRequestHandler,专门处理服务器端的ConfigChangeNotifyRequest请求,该请求只会推送变化了的配置的基本信息,而不包括内容,所以此处还会触发一次ConfigBatchListenRequest请求
  3. 之后查找到变化的配置后,再发一个查询请求拉取配置

服务端

推送配置处理器

ConfigPublishRequestHandler处理器

配置中心使用ConfigPublishRequestHandler类处理客户端配置推送请求:

  1. 验证请求参数

  2. 封装configAdvanceInfo配置扩展信息

  3. 创建ConfigInfo封装配置信息

  4. 使用持久层对象insert或者update配置

  5. 使用ConfigChangePublisher推送一个ConfigDataChangeEvent事件

    ConfigChangePublisher.notifyConfigChange(
        new ConfigDataChangeEvent(false, dataId, group, tenant, time.getTime()));
    

ConfigDataChangeEvent事件处理

ConfigDataChangeEvent事件会触发数据dump操作和集群同步操作。

此处先介绍数据dump操作:

dumpService.dump(syncRequest.getDataId(), syncRequest.getGroup(), syncRequest.getTenant(),
    syncRequest.getTag(), syncRequest.getLastModified(), NetUtils.localIP());

dumpService是Dump data service用于备份数据:

// Add DumpTask to TaskManager, it will execute asynchronously.
public void dump(String dataId, String group, String tenant, String tag, long lastModified, String handleIp,
        boolean isBeta) {
    String groupKey = GroupKey2.getKey(dataId, group, tenant);
    String taskKey = String.join("+", dataId, group, tenant, String.valueOf(isBeta), tag);
    // 推送一个DumpTask
    dumpTaskMgr.addTask(taskKey, new DumpTask(groupKey, tag, lastModified, handleIp, isBeta));
}

DumpTask将使用DumpProcessor类处理:

  1. 把数据写到磁盘
  2. 推送LocalDataChangeEvent事件

LocalDataChangeEvent事件会被以下几个类处理:

  • LongPollingService$1会触发DataChangeTask,响应长轮询订阅者
  • RpcConfigChangeNotifier将数据变化推送给rpc订阅者
  • InternalConfigChangeNotifier处理内部配置文件变化

监听配置处理器

ConfigChangeBatchListenRequestHandler处理器

处理客户端的配置监听请求:

public class ConfigChangeBatchListenRequestHandler
        extends RequestHandler<ConfigBatchListenRequest, ConfigChangeBatchListenResponse> {

    @Autowired
    private ConfigChangeListenContext configChangeListenContext;

    @TpsControl(pointName = "ConfigListen")
    @Secured(action = ActionTypes.READ, parser = ConfigResourceParser.class)
    public ConfigChangeBatchListenResponse handle(
            ConfigBatchListenRequest configChangeListenRequest, RequestMeta meta)
            throws NacosException {
        String connectionId = StringPool.get(meta.getConnectionId());
        String tag = configChangeListenRequest.getHeader(Constants.VIPSERVER_TAG);

        ConfigChangeBatchListenResponse configChangeBatchListenResponse = 
            new ConfigChangeBatchListenResponse();
        for (ConfigBatchListenRequest.ConfigListenContext listenContext : configChangeListenRequest
                .getConfigListenContexts()) {
            String groupKey = GroupKey2
                    .getKey(listenContext.getDataId(), listenContext.getGroup(), listenContext.getTenant());
            groupKey = StringPool.get(groupKey);

            String md5 = StringPool.get(listenContext.getMd5());

            // 监听
            if (configChangeListenRequest.isListen()) {
                // 注册监听器,维护groupKey->connectionId集关系和groupKey->md5关系
                configChangeListenContext.addListen(groupKey, md5, connectionId);
                // 判断变化
                boolean isUptoDate = ConfigCacheService.isUptodate(groupKey, md5, meta.getClientIp(), tag);
                if (!isUptoDate) {
                    // 把变化的配置基本信息添加到响应
                    configChangeBatchListenResponse
                        .addChangeConfig(listenContext.getDataId(), listenContext.getGroup(),
                            listenContext.getTenant());
                }
            } else {
                // 取消监听
                configChangeListenContext.removeListen(groupKey, connectionId);
            }
        }

        return configChangeBatchListenResponse;
    }
}
内容来源于网络如有侵权请私信删除

文章来源: 博客园

原文链接: https://www.cnblogs.com/xugf/p/17647070.html

你还没有登录,请先登录注册
  • 还没有人评论,欢迎说说您的想法!