[SERVER-51830] mirrored_reads.js doesn't wait for all mirrored reads to resolve Created: 23/Oct/20 Updated: 06/Dec/22 Resolved: 04/Feb/21 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Critical - P2 |
| Reporter: | Benjamin Caimano (Inactive) | Assignee: | Backlog - Service Architecture |
| Resolution: | Duplicate | Votes: | 0 |
| Labels: | servicearch-wfbf-day | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||
| Assigned Teams: |
Service Arch
|
||||||||||||
| Operating System: | ALL | ||||||||||||
| Participants: | |||||||||||||
| Linked BF Score: | 126 | ||||||||||||
| Description |
|
When we wait for read mirroring to complete here, we never check that our "sent" value is equal to our "resolved" value. This means that occasionally, we have achieved the desired level of mirroring but there are still operations being mirrored to secondaries. This becomes a problem when we run 0% mirroring tests on slow variants, because even a single late operation breaks the statistical bounds in the test. |
| Comments |
| Comment by Daniel Gottlieb (Inactive) [ 21/Jan/21 ] |
|
Upping the priority as this has been a persistent failure with a proposed fix. |