ORA-16660: fast-start failover operation attempted in absence of a broker configuration

文档解释

ORA-16660: fast-start failover operation attempted in absence of a broker configuration

Cause: An attempt was made to enable or disable fast-start failover when connected to a standby database for which broker configuration details are currently unavailable. For instance, the standby database may currently require re-creation (or flashback reinstantiation) before it may respond to broker client commands.

Action: 1) An attempt to enable or disable (non-FORCE) fast-start failover at this standby database will be rejected until such time that the broker configuration details have been made available to that standby database from the primary database. This normally occurs when the standby database is successfully re-created or flashed back, and then reenabled at the primary database. 2) Use the FORCE option to override fast-start failover that has been enabled at the standby database even when the broker configuration details are currently unavailable to the standby database. In this case, this status message is only a warning. Note that fast-start failover is not formally disabled in the broker configuration. The effect of this command issued under these circumstances may or may not be permanent, depending upon when the primary and standby databases regain full communication between each other and if the state of fast-start failover has been altered at the primary database in the meantime.

ORA-16660 错误属于 Oracle 资源管理器中的一个报错。只有在使用 Oracle 的高可用性设置，如 Data Guard (数据守卫）时，这个错误才会出现。

该错误指出 Fast-Start Failover 操作尝试在缺少 Broker 配置的情况下进行，从而要求将它置顶位置必须有配置的 Broker 才能进行操作。

官方解释

ORA-16660 异常发生在 Data Guard 数据守卫环境中，指出您在缺少 Fast-Start Failover Broker 配置的情况下尝试执行 Fast-Start Failover 操作。如果不存在 Broker，那么 Data Guard 将无法用于替代数据库功能。

常见案例

错误 ORA-16660 往往发生在维护 Data Guard 失败后，用户尝试通过命令主动执行 Fast-Start Failover 操作，但在缺少 Fast-Start Failover Broker 服务配置的情况下，而报错 ORA-16660 就产生了。

一般处理方法及步骤

1. 启动 Broker 服务，确保主机上已设置 Fast-Start Failover。

2. 检查使用的 Data Guard 字符集，确保与目标站点的字符集相匹配。

3. 检查是否存在 REDO 负载不平衡的情况。

4. 确认有着功能资格的 Data Guard 目标站点是否全部处于可用状态。

5. 如果在处理期间仍然无法解决，请考虑更改数据库属性，包括：fra_max_space、fra_max_retention、fra_min_retention 等，以释放数据库空间资源，保证在发生故障时，不会导致 Fast-Start Failover 无法成功复制。