DanielLeens commented on issue #10899:
URL: https://github.com/apache/seatunnel/issues/10899#issuecomment-4523789997

   This is not random in the sense of "it sometimes happens without a pattern".
   
   The problematic path is conditional, but deterministic once that condition 
is met:
   
   - the job is using `startup.mode = timestamp`
   - the job enters the restore / checkpoint recovery path
   - and the current implementation re-resolves the binlog position from the 
original `startup.timestamp` instead of reusing the restored split offset
   
   If that original timestamp still maps to a valid current binlog position, 
the problem may stay hidden. Once that historical position is purged or 
otherwise unavailable, recovery fails in exactly this path.
   
   So the core issue is not occasional behavior inside one code path, but that 
the restore path is currently using the wrong recovery anchor for `timestamp` 
mode.


-- 
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]

Reply via email to