JNSimba opened a new pull request, #61389:
URL: https://github.com/apache/doris/pull/61389

   ### What problem does this PR solve?
   
   #### Background                                                              
                                                          
                                                                                
                                         
     StreamingJob currently supports two offset modes:                          
                                                          
     - `initial`: full snapshot + continuous incremental replication            
                                                          
     - `latest` / `earliest`: incremental replication only (no snapshot)        
                                                          
                     
     There is no way to perform a one-time full sync and stop. This is
     needed for data migration scenarios where only a point-in-time full
     copy is required, without ongoing replication.
   
     #### Usage
   
     Set `offset=snapshot` when creating a StreamingJob:
   
     ```sql
   CREATE JOB mysql_db_sync
   ON STREAMING
   FROM MYSQL (
       ...
       "user" = "root",
       "password" = "",
       "database" = "db",
       "include_tables" = "user_info,student", 
       "offset" = "snapshot"
   )
   TO DATABASE target_test_db (
   )
   ```
   
     The job will perform a full table snapshot and automatically transition
     to FINISHED once all data is synced. No binlog/WAL subscription is
     established.
   
    #### Design
   
     The implementation centers on a hasReachedEnd() signal in
     SourceOffsetProvider:
   
     - FE: JdbcSourceOffsetProvider returns null from getNextOffset()
     when all snapshot splits are consumed in snapshot-only mode.
     StreamingInsertJob.onStreamTaskSuccess() checks hasReachedEnd()
     before creating the next task — if true, the job is marked FINISHED.
     - BE (cdc_client): snapshot maps to StartupOptions.snapshot() for
     both MySQL and PostgreSQL connectors. The chunk-split path is reused
     from initial mode.
     - Crash recovery: if FE crashes before persisting FINISHED, the job
     auto-resumes via PAUSED→PENDING. getNextOffset() returns null,
     the task no-ops safely, and onStreamTaskSuccess() re-checks
     hasReachedEnd() to correctly transition to FINISHED.
   
   #### Testing
   
     Added regression tests for both MySQL and PostgreSQL:
     - test_streaming_mysql_job_snapshot.groovy
     - test_streaming_postgres_job_snapshot.groovy
   
     Both tests verify:
     1. All existing data is synced correctly after job finishes
     2. Job status transitions to FINISHED
     3. Data inserted after the job finishes is not consumed
    
   
   ### Check List (For Author)
   
   - Test <!-- At least one of them must be included. -->
       - [ ] Regression test
       - [ ] Unit Test
       - [ ] Manual test (add detailed scripts or steps below)
       - [ ] No need to test or manual test. Explain why:
           - [ ] This is a refactor/code format and no logic has been changed.
           - [ ] Previous test can cover this change.
           - [ ] No code files have been changed.
           - [ ] Other reason <!-- Add your reason?  -->
   
   - Behavior changed:
       - [ ] No.
       - [ ] Yes. <!-- Explain the behavior change -->
   
   - Does this need documentation?
       - [ ] No.
       - [ ] Yes. <!-- Add document PR link here. eg: 
https://github.com/apache/doris-website/pull/1214 -->
   
   ### Check List (For Reviewer who merge this PR)
   
   - [ ] Confirm the release note
   - [ ] Confirm test cases
   - [ ] Confirm document
   - [ ] Add branch pick label <!-- Add branch pick label that this PR should 
merge into -->
   
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to