[SERVER-85858] Revert burn_in_test's inclusion of transitively-affected tests Created: 29/Jan/24 Updated: 29/Jan/24 Resolved: 29/Jan/24 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | 8.0.0-rc0 |
| Type: | Improvement | Priority: | Major - P3 |
| Reporter: | Steve Gross | Assignee: | Steve Gross |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||
| Assigned Teams: |
Correctness
|
||||||||
| Backwards Compatibility: | Fully Compatible | ||||||||
| Sprint: | Build and Correctness OnDeck | ||||||||
| Participants: | |||||||||
| Description |
|
In https://jira.mongodb.org/browse/SERVER-82200 , we amended burn_in_test's logic for determining which tests should run. Previously, the test looked at directly affected files. After, the test looked at indirectly affected files. This had the positive effect of including more likely-to-be-broken tests. It also had the negative effect of exponentially increasing the volume of tests-to-be-run, which in turn resulted in timeouts and resource exhaustion in Evergreen. As a result, we've decided (https://mongodb.slack.com/archives/C06FEUUJW2U/p1706293634610059 ) that the drawbacks outweigh the benefits, so we'll revert the change. |