[SERVER-4884] Make parallel suite easier to debug Created: 06/Feb/12 Updated: 19/Jun/23 Resolved: 19/Jun/23 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Testing Infrastructure |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Task | Priority: | Major - P3 |
| Reporter: | Tad Marshall | Assignee: | Backlog - Server Tooling and Methods (STM) (Inactive) |
| Resolution: | Won't Do | Votes: | 0 |
| Labels: | sdp-backlog-purge, stm | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||
| Assigned Teams: |
Server Tooling & Methods
|
||||||||||||
| Backwards Compatibility: | Fully Compatible | ||||||||||||
| Participants: | |||||||||||||
| Description |
|
The parallel suite runs several tests simultaneously against an instance of mongod. It randomizes the grouping and order of the tests on each run, which helps to exercise the code. However, when announcing failure of the master test suite, it doesn't identify which test failed, or what other tests were running at the same time. Moreover, there is no way to reproduce a particular failure, because the random seed is not reported in the log and there is no way to tell the test to use a particular seed. This test suite would be more useful if it summarized the failure at the end of the run and identified the failing test(s), and it would be more useful if it gave developers a way to re-run a failed test so that a fix could be verified. As it is, the test might work (or fail) on the next run and we don't know that anything has changed, because the actual test run is different. |
| Comments |
| Comment by Alex Neben [ 19/Jun/23 ] |
|
This has been identified as work that the SDP team won't do in the near term. Please reopen with a comment if you feel this work should be reprioritized and explain why. |
| Comment by Brooke Miller [ 30/Jun/20 ] |
|
robert.guo is this something we should do? If so, I'd propose moving this from Blocked to Open and leaving it on the Backlog. |
| Comment by Ryan Timmons [ 11/May/20 ] |
|
This ticket has grown stale and is not currently on this team's radar for prioritization. Is there still sufficient interest or value in doing this? If not I'd like to close as wont-fix; we can always re-open or re-evaluate it in the future. |
| Comment by Kamran K. [ 11/Dec/15 ] |
|
File name reporting is covered by |