help request: Our APISIX is deployed in K8S and monitors K8S services through routes and upstreams. After the service is expanded, APISIX cannot perceive the expanded service Pod

[xiaoxiongxyy]

Description

Our APISIX is deployed in K8S and monitors K8S services through routes and upstreams. After the service is expanded, APISIX cannot perceive the expanded service Pod。
We have 10 APISIX Pods. However, 2 Pods have a path request error of 503, while the others are normal. This phenomenon occurs twice in a row, each time triggered by service expansion. What is the reason?

Environment

• APISIX version (run apisix version):3.3
• Operating system (run uname -a):K8S
• OpenResty / Nginx version (run openresty -V or nginx -V):1.19
• etcd version, if relevant (run curl http://127.0.0.1:9090/v1/server_info):
• APISIX Dashboard version, if relevant:
• Plugin runner version, for issues related to plugin runners:
• LuaRocks version, for installation issues (run luarocks --version):

[zhoujiexiong]

Similar to #10422