Closed shubifeng closed 6 years ago
Please update your benchmark result and more details.
Usually it is because that the database cannot handle the deleting/loading quickly. Extend you database cluster and test again.
应该不是数据库集群问题.
所有的binlog解析到maxwell-topic, 要么是kafka队列读的慢, 因为有一部分数据还能快速更新, 有一部分表的数据很慢很慢.
没数据的时候也建立临时库? 为什么会没有自动删除?
当库表较多和kafka消息队列中 消息数 过于庞大时, 当maxwell.properties配置同步一个表时, bireme入库时间会很长.