该方法的问题是，和为错误的构造函数调用报告错误一样，程序员将无法预知其程序的某一次运行将崩溃。

The problem with this approach is that, like signaling errors for bad constructor calls, the programmer will have no way of predicting that a given run of his program will crash.

youdao

在生产环境中，将运行多个Catalog Server来提供故障转移，但是，在此示例中只运行了一个实例，因此，如果您删除它，该应用程序将无法再运行。

In a production environment, multiple Catalog Servers would be run to provide failover, but you are only running one instance in this example, so if you kill it, the application will no longer work.

youdao