flink cdc接入备库B，类似这种情况咋解决的呢?

背景：
MySQL数据库的架构是一主（A）两备（B、C），整套架构都已经启用GTID 模式，MySQL 版本5.7

备库B当前的binlog文件号是 mysql-bin.022043
备库C当前的binlog文件号是 mysql-bin.022003

两个备库都和主库保持数据同步，无延迟。

flink cdc接入备库B，通过拉取b该库inlog实时采集数据流到kafka，待下游消费使用

当备库B不可对外提供服务器， 需要把flink 采集任务切到备库C，这时候flink会报找不到binlog的错误：
2024-04-09 19:09:37,005 INFO com.ververica.cdc.connectors.mysql.debezium.task.context.StatefulTaskContext [] - Connector requires binlog file 'mysql_bin.022043', but MySQL only has mysql_bin.021983, mysql_bin.021984, mysql_bin.021985, mysql_bin.021986, mysql_bin.021987, mysql_bin.021988, mysql_bin.021989, mysql_bin.021990, mysql_bin.021991, mysql_bin.021992, mysql_bin.021993, mysql_bin.021994, mysql_bin.021995, mysql_bin.021996, mysql_bin.021997, mysql_bin.021998, mysql_bin.021999, mysql_bin.022000, mysql_bin.022001, mysql_bin.022002, mysql_bin.022003
2024-04-09 19:09:37,006 ERROR org.apache.flink.connector.base.source.reader.fetcher.SplitFetcherManager [] - Received uncaught exception.
java.lang.RuntimeException: SplitFetcher thread 0 received unexpected exception while polling the records

就是已经启用了GTID，再验证备库C文件号呢，实际中可能每个备库binlog保留策略受本地磁盘空间大小影响会有差异，向我们这个案例中就比较大的差异。

类似这种情况咋解决的呢?