orleans cluster problem

I used Adonet as the cluster table, created two different Silo A and B, and according to the official documentation, both clusters were able to start and work properly.But I had a tricky problem.There is no graceful exit of SiloA, the state in the cluster is active at that time, now I want to start SiloB, I got some errors during startup, and finally the program exited abnormally, because silo abnormal exit factors are not controllable (l example: server power failure), what can be done to deal with this problem?


Can we use the start write IAmAliveTime time as the auxiliary judgment condition?
Or make a policy in the configuration cluster, whether to use the newly added silo host as the centralized state, or to ensure that the old cluster silo host is the primary state of the cluster.

Thank you very much for your answer

No source files, tests, or entry points are named. Reproduce the failure using an ADO.NET cluster table with two silos, then read the official cluster membership documentation and inspect the reported startup errors. Done means establishing and documenting a supported policy for recovering from an abruptly terminated silo without abnormal startup failure.

Written by the indexing model from the issue text.