[SERVER-40048] Tests should not spawn a replica set just to determine if majority read concern is supported Created: 08/Mar/19 Updated: 23/Oct/23 |
|
| Status: | Open |
| Project: | Core Server |
| Component/s: | Testing Infrastructure |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Task | Priority: | Major - P3 |
| Reporter: | Judah Schvimer | Assignee: | [DO NOT ASSIGN] Backlog - DevProd Correctness |
| Resolution: | Unresolved | Votes: | 0 |
| Labels: | tig-mongorunner | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Assigned Teams: |
Correctness
|
| Participants: |
| Description |
|
This function is used in many tests to determine if the storage engine supports majority read concern: https://github.com/mongodb/mongo/blob/4cc82e953bcfcb355cccf47ba5bb696e7aa7a057/jstests/multiVersion/libs/causal_consistency_helpers.js#L72-L80. This starts replica set and then stops it, just to start a new one for the test. This is wasteful. |
| Comments |
| Comment by Judah Schvimer [ 11/Mar/19 ] |
|
max.hirschhorn, that seems reasonable. Using a tag here also could maybe make sense. |
| Comment by Max Hirschhorn [ 09/Mar/19 ] |
|
Should we just hardcode a list of storage engines that support majority read concern and rely on TestData.storageEngine? |