test(server): Fix flakiness in TestServer_ReloadWorkerTags #3810
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This attempts to address some flakiness by replacing hard-coded sleeps with some smarter polling similarly used in TestServer_ReloadInitialUpstreams.
moduli
commented
Oct 6, 2023
| require.True(ok) | ||
| tags := w.CanonicalTags() | ||
| v, ok := tags[key] | ||
| require.True(ok, fmt.Sprintf("EXPECTED: map[%s:%s], ACTUAL: %+v", key, values, tags)) |
There was a problem hiding this comment.
Updated logging to get more information
Comment on lines
+122
to
+149
| timeout := time.NewTimer(15 * time.Second) | ||
| poll := time.NewTimer(0) | ||
| var w *server.Worker | ||
| var lastStatusTime time.Time | ||
| serversRepo, err := testController.Controller().ServersRepoFn() | ||
| require.NoError(err) | ||
| pollController: | ||
| for { | ||
| select { | ||
| case <-timeout.C: | ||
| t.Fatalf("timeout waiting for worker to connect to controller") | ||
| case <-poll.C: | ||
| w, err = serversRepo.LookupWorkerByName(testController.Context(), "test") | ||
| require.NoError(err) | ||
| if w != nil { | ||
| switch { | ||
| case lastStatusTime.IsZero(): | ||
| lastStatusTime = w.GetLastStatusTime().AsTime().Round(time.Second) | ||
| default: | ||
| if !lastStatusTime.Equal(w.GetLastStatusTime().AsTime().Round(time.Second)) { | ||
| timeout.Stop() | ||
| break pollController | ||
| } | ||
| } | ||
| } | ||
| poll.Reset(time.Second) | ||
| } | ||
| } |
There was a problem hiding this comment.
This is from TestServer_ReloadInitialUpstreams
| timeout.Reset(15 * time.Second) | ||
| poll.Reset(time.Second) | ||
| lastStatusTime = time.Time{} | ||
| startTime := time.Now() |
There was a problem hiding this comment.
This is a slightly modified version of the loop. TestServer_ReloadInitialUpstreams is different in that it uses a second controller and can set lastStatusTime to zero. In this case, the idea is to wait for 2 status updates starting from this point in time.
|
Just got a failure when testing this locally... back to the drawing board... |
|
Looks like there might be actually something in the boundary application causing this. More investigation is needed. Closing this PR. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR attempts to address flakiness in
TestServer_ReloadWorkerTagsby replacing hard-coded sleeps with some smarter polling similarly used inTestServer_ReloadInitialUpstreams. This test would occasionally fail for the following reason...The test is expecting the controller to get updated worker tags, but sometimes, it's not getting the expected values. The initial theory is that the hard-coded sleeps didn't wait long enough for the changes to propagate to the controller.