chore: fix a flaky test #9699
chore: fix a flaky test #9699
Conversation
|
Auto Cherry-pick Instructions |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #9699 +/- ##
==========================================
+ Coverage 59.68% 59.70% +0.02%
==========================================
Files 551 551
Lines 59782 59788 +6
==========================================
+ Hits 35679 35697 +18
+ Misses 20844 20835 -9
+ Partials 3259 3256 -3
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
pkg/testutil/apps/common_util.go
Outdated
| if pobj.GetDeletionTimestamp().IsZero() { | ||
| d, _ := json.Marshal(pobj) | ||
| panic("expected DeletionTimestamp is not nil, obj: " + string(d)) | ||
| ginkgo.GinkgoWriter.Printf("Warning: deleted obj's DeletionTimestamp is nil. This may due to an old reconciliation process is not done. obj: %s", string(d)) |
There was a problem hiding this comment.
If we remove the panic, nobody will notice or care about the warning. Cases may fail due to the residual resources, which are also flaky and hard to debug.
@leon-inf what's your opinion
There was a problem hiding this comment.
+1, we shouldn't simply remove the panic here. Regarding the specific issue, we should add a mechanism to coordinate the timing of deleting the component and its sub-resources.
There was a problem hiding this comment.
I can try delete Component object without removing the finalizer, so that we can make sure no component's reconciliation is still in progress.
This reverts commit b14b436.
587b223 to
6463d11
Compare
fixes #9448