Overview
During data consistency check, DTS compares the collection data between the source and target databases and outputs the comparison result and inconsistency details for you to perform a business cutover stably and reliably.
Notes
1. Data consistency check compares only the objects selected in the source database and objects migrated to the target database. If you write data into the target database during migration, then the written data will not be included in the consistency check.
2. A data consistency check task may increase the load in the source database instance. Therefore, you need to perform such tasks during off-peak hours.
3. A data consistency check task can be executed repeatedly, but one DTS instance can initiate only one such task at any time.
4. If you choose to complete or terminate a DTS task before a data consistency check task is completed, the check task will fail.
5. When creating a consistency check, the system will automatically create the dts_verify_result
library on the target end to record content related to the consistency check. The table styles created under the dts_verify_result
library are as follows:
diff_5xxxxxxxx4231: Saves inconsistent data detected
diff_meta_5xxxxxxxxx4231: Saves inconsistent metadata detected
result_5xxxxxxxxx4231: Records the results after phase validation
status_5xxxxxxxxx4231: Records validation progress
Restrictions
Currently, check tasks are imperceptible to the DDL operations. If you perform DDL operations in the source database during migration, the check result will be inconsistent with the actual data, and you need to initiate another check task to get the accurate comparison result.
Check Scheme
DTS checks and compares all the data migrated during full migration and incremental migration from the source database. A full data check compares the data in the source and target databases row by row. Once the thread of the incremental data check finds that the full data comparison is completed, it immediately starts the incremental data check to get the start timestamp of the full data check, get the incremental oplog in the source database in a loop, and compare the differences between the source and target databases. When the time lag of data in the source and target databases is below 10 seconds, the comparison ends, and the check result is output.
Was this page helpful?