服务端重启后可能会造成瞬时单节点高网络负载

[zz630]

Issue Description

Type: bug report or feature request
the same as #8317

Describe what happened (or what feature you want)

想确定#8317的问题在最新的版本有没有解决？
我在#8310没有找到相关代码。在develop分支也没找到。那我提交一个commit修复该问题？
I want to make sure whether the issue #8317 has been resolved in the latest version?
I did not find the relevant code in #8310. I also did not find it in the develop branch. Should I submit a commit to fix the issue?

Describe what you expected to happen

How to reproduce it (as minimally and precisely as possible)

Tell us your environment

Anything else we need to know?

[zz630]

可能我的场景比较少见。搞了一层代理，去订阅所有服务。客户端通过访问代理层接口，来获取服务列表。
nacos集群只要有重启，这个节点的请求就打到下一个节点，并一直维持。
不确定大家为什么没遇到这个问题，难道是客户端数量多，上下线也多，流量后面逐渐均衡了？

Maybe my scenario is rare. I set up a proxy layer to subscribe to all services. The client obtains the service list by accessing the proxy layer interface.
As long as the nacos cluster is restarted, the request of this node will be sent to the next node and maintained.
I'm not sure why everyone has not encountered this problem. Is it because there are many clients and many online and offline connections, and the traffic is gradually balanced later?

[KomachiSion]

可以自行拓展ServerListProvider 让不同的nacos-client获取到部分nacos-server节点，相当于自行进行分片。

ServerListProvider可以控制nacos-client可连接的server列表， 但nacos-client依旧会在ServerListProvider返回的列表中进行轮训。