[SERVER-26408] Methods that were formerly called only in the replication executor should be noexcept Created: 30/Sep/16 Updated: 06/Dec/22 |
|
| Status: | Backlog |
| Project: | Core Server |
| Component/s: | Replication |
| Affects Version/s: | 3.3.15 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Andy Schwerin | Assignee: | Backlog - Replication Team |
| Resolution: | Unresolved | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Assigned Teams: |
Replication
|
| Operating System: | ALL |
| Participants: |
| Description |
|
Now that the topology mutex has replaced execution in the replication executor as the way to serialize access to topology coordinator data structures, we have introduced an exception safety problem. Exceptions that escaped replication executor callbacks led to system termination, but exceptions that escape topo lock blocks do not. We should review all such methods, and either mark them noexcept or ensure that they are exception safe. |
| Comments |
| Comment by Judah Schvimer [ 17/Apr/20 ] |
|
This code has changed significantly since 3.4, and I am not sure this audit is still worthwhile. tess.avitabile, do you think it is, or should we close this? |