Hi, community,
This is a new plan to support showing the execution process for running SQLs, especially for DML/DDL always needing much time to run. With my thinking, there are three main tasks for this new feature. ### Tasks #### Generate Job IDs - Add configuration item to allow users to set instance ID for each proxy - Generate distributed IDs by snowflake for this job's trunk and branches - Pass these IDs through the whole execution #### Report execution process from one Proxy - `ExecutorEngine` reports the context (IDs, beginTime, etc.) of each query to `governance` - Each branch job reports its status and ID to `governance` once it finishes in `ExecutorCallback` - `Governance` handles these requests synchronously #### Query execution process from other Proxies - It is triggered by `show processlist` - Get the execution status from all the jobs from `governance` - Make up the result based on `show processlist`' original result with the collected data If you’re interested in this one, please watch [1]. Regards, Trista [1] https://github.com/apache/shardingsphere/issues/9568 ------------------------------------------------------- Email：[email protected] Juan Pan(Trista) Apache ShardingSphere
