[SERVER-40181] Remove stmtId from transaction oplog entries Created: 17/Mar/19 Updated: 23/Sep/19 Resolved: 22/May/19 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Replication, Upgrade/Downgrade |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Improvement | Priority: | Major - P3 |
| Reporter: | Judah Schvimer | Assignee: | Jason Chan |
| Resolution: | Duplicate | Votes: | 0 |
| Labels: | todo_in_code | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||||||||||
| Sprint: | Repl 2019-06-03 | ||||||||||||||||||||
| Participants: | |||||||||||||||||||||
| Description |
|
It's currently not used internally at all, and generating it properly is untested. Since we do not plan to test that stmtIds are generated properly, it could make sense to remove the field so no one relies on behavior we do not want to guarantee. |
| Comments |
| Comment by Jason Chan [ 22/May/19 ] |
|
Closing this ticket in favor of |
| Comment by Judah Schvimer [ 09/May/19 ] |
|
How are we going to do this ticket in 4.2 with the above invariant? |
| Comment by Siyuan Zhou [ 17/Apr/19 ] |
|
Rollback via refetch code invariants that statement id is present once txnNumber is present, which may make this hard for downgrade. |
| Comment by Siyuan Zhou [ 17/Mar/19 ] |
|
This field is tested in the new oplog format, but I still think it's a good idea to remove the field as it's not used by transactions. |