Circuit breaker for outlier service node | sentinel能否对故障实例熔断的隔离

现在有A服务调用B服务，B服务有三个节点，其中有个节点发生了异常，我们希望在请求A服务调用B服务的时候，不会把请求打到异常节点上，sentinel能支持在正常节点上负载，对异常节点进行隔离，这样保证主调方迅速自动屏蔽错误率高或故障的服务实例，并启动定时任务对熔断实例进行探活。在达到恢复条件后对其进行半开恢复。在半开恢复后，释放少量请求去进行真实业务探测。并根据真实业务探测结果去判断是否完全恢复正常。

The issue describes isolating unhealthy service instances during A-to-B calls, probing them on a schedule, and restoring them through half-open testing. No source files, tests, or entry points are identified, so first clarify the intended Sentinel component and recovery semantics; done would include routing away from failing nodes and validating recovery before full restoration.

Written by the indexing model from the issue text.