odc
odc copied to clipboard
[Bug]: The record details of the executed archiving task show an imprecise count of scanned records.
ODC version
4.3.0
OB version
ALL
What happened?
The statistics base scans primary key ranges, and currently, the default batchSize for scanning is set to 200. If the number of rows scanned is less than 200, it is considered as one batch, which can lead to imprecise row count estimates.
统计信息基于通过扫描主键范围,目前扫描的 batchSize 默认是 200,小于 200 就是一个 batch,导致扫描行数不精确
What did you expect to happen?
Provide accurate row count for scans.
提供准确的扫描行数
How can we reproduce it (as minimally and precisely as possible)?
Archive a table with a number of records to be processed that is not divisible by 200.
归档一个要处理的记录数不能被200整除的表
Anything else we need to know?
No response