Bug 203171

Summary: Layout Test inspector/layers/layers-for-node.html is flaky
Product: WebKit Reporter: Truitt Savell <tsavell>
Component: Tools / TestsAssignee: Nobody <webkit-unassigned>
Status: NEW ---    
Severity: Normal CC: ap, graouts, hi, simon.fraser, webkit-bot-watchers-bugzilla, webkit-bug-importer, yurys
Priority: P2 Keywords: InRadar
Version: WebKit Nightly Build   
Hardware: Unspecified   
OS: Unspecified   
See Also: https://bugs.webkit.org/show_bug.cgi?id=202649

Description Truitt Savell 2019-10-18 16:40:56 PDT
The following layout test is flaky on macOS

inspector/layers/layers-for-node.html

Probable cause:

Started failing around r250654

Flakiness Dashboard:

https://webkit-test-results.webkit.org/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=inspector%2Flayers%2Flayers-for-node.html

Diff:
--- /Volumes/Data/slave/mojave-debug-tests-wk1/build/layout-test-results/inspector/layers/layers-for-node-expected.txt
+++ /Volumes/Data/slave/mojave-debug-tests-wk1/build/layout-test-results/inspector/layers/layers-for-node-actual.txt
@@ -109,14 +109,14 @@
             "x": 0,
             "y": 0,
             "width": "number",
-            "height": "number"
-        },
-        "memory": "number",
-        "compositedBounds": {
-            "x": 0,
-            "y": 0,
-            "width": "number",
-            "height": "number"
+            "height": 72
+        },
+        "memory": "number",
+        "compositedBounds": {
+            "x": 0,
+            "y": 0,
+            "width": "number",
+            "height": 72
         },
         "paintCount": "number"
     }
Comment 1 Truitt Savell 2019-10-18 16:46:35 PDT
marked test as failing https://trac.webkit.org/changeset/251311/webkit
Comment 2 Radar WebKit Bug Importer 2019-10-18 16:47:03 PDT
<rdar://problem/56424548>
Comment 3 Yury Semikhatsky 2019-10-21 09:38:24 PDT
@graouts: Antoine, can you have a look at the test and check that original assumptions since you added the test still hold true? After playing for a while with the test I feel like with the current set of events in LayerTree domain it may be hard to reliably detect when the layers get to the expected state but I don't have enough knowledge in this domain to tell for sure.