Repository navigation
Commit e9001c9
committed
test: stabilize <...>_sort_data_test::test_gc
This commit adds waiting for checkpoint GC results after
`box.snapshot()` in the `test_gc` test case in
`box-luatest/gh_10847_memtx_sort_data_test`. The checkpoint GC is
asynchronous.
I see less frequent `test_box_backup` failures of the same kind. I'll
let the test author to work on it: there are more checks there and I
don't want to break anything.
Statistics over three weeks of our CI logs (from 2026-09-08 to
2026-09-29, includes re-runs made by test-run itself):
| test case | arm64 | x86_64 | overall |
| --------------- | ----- | ------ | ------- |
| test_gc | 74 | 103 | 177 |
| test_box_backup | 8 | 18 | 26 |
My recipe to reproduce the instability locally:
```sh
$ ./test/test-run.py $(yes box-luatest/gh_10847_memtx_sort_data_test.lua | head -n 32)
```
NO_DOC=test adjustment
NO_CHANGELOG=same1 parent 8655911 commit e9001c9
1 file changed
Lines changed: 28 additions & 16 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
586 | 586 | | |
587 | 587 | | |
588 | 588 | | |
589 | | - | |
590 | | - | |
591 | 589 | | |
592 | | - | |
593 | | - | |
594 | | - | |
595 | | - | |
| 590 | + | |
| 591 | + | |
| 592 | + | |
| 593 | + | |
596 | 594 | | |
597 | | - | |
598 | | - | |
599 | | - | |
600 | | - | |
601 | | - | |
602 | | - | |
603 | | - | |
604 | | - | |
605 | | - | |
| 595 | + | |
| 596 | + | |
| 597 | + | |
| 598 | + | |
| 599 | + | |
| 600 | + | |
| 601 | + | |
| 602 | + | |
| 603 | + | |
| 604 | + | |
| 605 | + | |
| 606 | + | |
| 607 | + | |
| 608 | + | |
| 609 | + | |
| 610 | + | |
| 611 | + | |
| 612 | + | |
| 613 | + | |
606 | 614 | | |
607 | 615 | | |
608 | 616 | | |
| 617 | + | |
609 | 618 | | |
610 | 619 | | |
611 | 620 | | |
612 | 621 | | |
613 | | - | |
| 622 | + | |
| 623 | + | |
| 624 | + | |
| 625 | + | |
614 | 626 | | |
615 | 627 | | |
616 | 628 | | |
| |||
0 commit comments