[Bug] dubbo3.3.0 + zookeeper 消费者本地服务列表缓存刷新失败问题

[skotty-open]

Pre-check

• I am sure that all the content I provide is in English.

Search before asking

• I had searched in the issues and found no similar issues.

Apache Dubbo Component

Java SDK (apache/dubbo)

Dubbo Version

dubbo-spring-boot-starter 3.3.0
dubbo-zookeeper-curator5-spring-boot-starter 3.3.0

Steps to reproduce this issue

1、A服务（消费者）调用B服务（服务提供者）（A、B服务各1个pod）
2、B服务重启（k8s pod）
3、A服务仍然调用旧的B服务地址(Pod IP)

消费者错误日志：
{"dateTime":"2025-01-23T20:41:30.030+0800","level":"WARN","traceId":"TID: N/A", "serviceName":"atlas-platform-user-center", "thread":"Dubbo-framework-SD-address-refresh-0-thread-1", "logger":"org.apache.dubbo.registry.client.ServiceDiscoveryRegistryDirectory", "message":" [DUBBO] Received url with EMPTY protocol from registry ServiceDiscoveryRegistryDirectory(registry: common-test-zookeeper-1a-1.aws-jp-2.yorkidc.com:2181, subscribed key: [atlas-platform-account])-Directory(invokers: 1[10.3.137.225:8080], validInvokers: 1[10.3.137.225:8080], invokersToReconnect: 0[]), will clear all available addresses., dubbo version: 3.3.0, current host: 10.3.139.21, error code: 4-1. This may be caused by , go to https://dubbo.apache.org/faq/4/1 to find instructions. "}
{"dateTime":"2025-01-23T20:41:30.030+0800","level":"INFO","traceId":"TID: N/A", "serviceName":"atlas-platform-user-center", "thread":"Dubbo-framework-SD-address-refresh-0-thread-1", "logger":"org.apache.dubbo.registry.client.migration.DefaultMigrationAddressComparator", "message":" [DUBBO] No instance address available, stop compare., dubbo version: 3.3.0, current host: 10.3.139.21 "}
{"dateTime":"2025-01-23T20:41:30.030+0800","level":"INFO","traceId":"TID: N/A", "serviceName":"atlas-platform-user-center", "thread":"Dubbo-framework-SD-address-refresh-0-thread-1", "logger":"org.apache.dubbo.registry.client.ServiceDiscoveryRegistryDirectory", "message":" [DUBBO] Received invokers changed event from registry. Registry type: instance. Service Key: com.avenir.atlas.module.account.api.PortfolioServiceApi. Urls Size : 1. Invokers Size : 0. Available Size: 0. Available Invokers : empty, dubbo version: 3.3.0, current host: 10.3.139.21 "}
{"dateTime":"2025-01-23T20:41:30.030+0800","level":"DEBUG","traceId":"TID: N/A", "serviceName":"atlas-platform-user-center", "thread":"Apollo-RemoteConfigLongPollService-1", "logger":"sun.net.www.protocol.http.HttpURLConnection", "message":"sun.net.www.MessageHeader@5e9ea39b10 pairs: {null: HTTP/1.1 304}{Date: Thu, 23 Jan 2025 12:41:30 GMT}{Connection: keep-alive}{X-Content-Type-Options: nosniff}{X-XSS-Protection: 1; mode=block}{Cache-Control: no-cache, no-store, max-age=0, must-revalidate}{Pragma: no-cache}{Expires: 0}{X-Frame-Options: DENY}{Request-Id: a372318842cc72901902dba4982fe1ef} "}
{"dateTime":"2025-01-23T20:41:30.030+0800","level":"DEBUG","traceId":"TID: N/A", "serviceName":"atlas-platform-user-center", "thread":"Apollo-RemoteConfigLongPollService-1", "logger":"com.ctrip.framework.apollo.internals.RemoteConfigLongPollService", "message":"Long polling response: 304, url: http://apollo-configservice.system-service.common-test.yorkapp.com/notifications/v2?cluster=test-3&appId=atlas-platform-user-center&ip=10.3.139.21&notifications=%5B%7B%22namespaceName%22%3A%22RD1.portfolio%22%2C%22notificationId%22%3A263609%7D%2C%7B%22namespaceName%22%3A%22application%22%2C%22notificationId%22%3A2234%7D%2C%7B%22namespaceName%22%3A%22application.yaml%22%2C%22notificationId%22%3A263611%7D%5D "}

重日志可以看出，A服务消费正常接收B服务提供者状态变更（包括旧的服务和接口下线和新服务和接口上线），但是仍然请求B服务旧的pod ip地址

What you expected to happen

1、A服务（消费者）调用B服务（服务提供者）
2、B服务重启（k8s pod）
3、A服务调用新的B服务地址(Pod IP)

Anything else

No response

Are you willing to submit a pull request to fix on your own?

• Yes I am willing to submit a pull request on my own!

Code of Conduct

• I agree to follow this project's Code of Conduct

[wcy666103]

Please re-submit this pr in English only.