consumer 调用并发高，对接口做了短时间（100ms）熔断。consumer服务cpu飙升

A服务调用B服务的接口
A服务单机并发500，对调用B服务的接口做了timeout设置100ms。并查看日志发现了大部分超时日志。并且A服务cpu狂飙。
B服务如果100ms只支持并发200左右。是否是由于A服务调用B服务的执行线程实际是没有断开还在执行的原因？

Start by reproducing the A-service to B-service scenario described in the issue: 500 concurrent calls, a 100ms consumer timeout, and B handling about 200 concurrent requests. Inspect the timeout logs and consumer execution behavior to determine whether timed-out calls continue running; done means explaining the CPU spike and confirming the relevant timeout semantics.

Written by the indexing model from the issue text.