[SERVER-43843] Skip validate during the rollback fuzzer "restartNode" command Created: 04/Oct/19 Updated: 29/Oct/23 Resolved: 22/Oct/19 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Replication, Testing Infrastructure |
| Affects Version/s: | None |
| Fix Version/s: | 4.3.1, 4.2.2, 4.0.14 |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Judah Schvimer | Assignee: | Judah Schvimer |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||||||
| Backwards Compatibility: | Fully Compatible | ||||||||||||||||
| Operating System: | ALL | ||||||||||||||||
| Backport Requested: |
v4.2, v4.0
|
||||||||||||||||
| Sprint: | Repl 2019-10-21, Repl 2019-11-04 | ||||||||||||||||
| Participants: | |||||||||||||||||
| Linked BF Score: | 13 | ||||||||||||||||
| Description |
|
Alternatively we could make the "validate collections on shutdown" code ignore NotMasterOrSecondary errors when run from the rollback fuzzer's "restartNode" command. |
| Comments |
| Comment by Githook User [ 31/Oct/19 ] |
|
Author: {'username': 'judahschvimer', 'email': 'judah.schvimer@10gen.com', 'name': 'Judah Schvimer'}Message: (cherry picked from commit b6f1f7f3c3455425d6169e9f9222eb4b9291697a) |
| Comment by Githook User [ 31/Oct/19 ] |
|
Author: {'name': 'Judah Schvimer', 'username': 'judahschvimer', 'email': 'judah.schvimer@10gen.com'}Message: (cherry picked from commit b6f1f7f3c3455425d6169e9f9222eb4b9291697a) |
| Comment by Githook User [ 22/Oct/19 ] |
|
Author: {'username': 'judahschvimer', 'email': 'judah.schvimer@10gen.com', 'name': 'Judah Schvimer'}Message: |
| Comment by Robert Guo (Inactive) [ 11/Oct/19 ] |
|
Will's suggestion sounds good. The fuzzer doesn't skip validation after transitioning to steady state repl so we're still getting coverage after each rollback. |
| Comment by William Schultz (Inactive) [ 11/Oct/19 ] |
|
Note that we will still validate collections during RollbackTest's full consistency checks, so we would only be losing validation coverage before each restart. I would probably vote to just skip validation in the restartNode function if the fixture is running with shutdowns enabled. I think this would be easier since there is already a clear way to skip validation from inside that function. |
| Comment by Robert Guo (Inactive) [ 04/Oct/19 ] |
|
I would vote for the latter as it's more surgical. It might be also worth exploring if we should assert.soon() the validate command, in case the next rollback masks any data consistency issues. |
| Comment by Judah Schvimer [ 04/Oct/19 ] |
|
robert.guo and geert.bosch, do you have any opinions on which of the two alternatives I should do? CC max.hirschhorn, since you've done a lot of work on the rollback fuzzer in the past. |