The test was removed in r276009 as it was solved. However, r276009 was partially reverted in r276021 and since then this test is failing again. The test should be reintroduced in glib TestExpectations. https://build.webkit.org/results/GTK-Linux-64-bit-Release-Tests/r276100%20%281194%29/imported/w3c/web-platform-tests/service-workers/service-worker/next-hop-protocol.https-diff.txt --- /home/buildbot/worker/gtk-linux-64-release-tests/build/layout-test-results/imported/w3c/web-platform-tests/service-workers/service-worker/next-hop-protocol.https-expected.txt +++ /home/buildbot/worker/gtk-linux-64-release-tests/build/layout-test-results/imported/w3c/web-platform-tests/service-workers/service-worker/next-hop-protocol.https-actual.txt @@ -1,4 +1,4 @@ -FAIL nextHopProtocol reports H1 correctly when routed via a service worker. assert_equals: nextHopProtocol is set on pass-through expected "http/1.1" but got "" +FAIL nextHopProtocol reports H1 correctly when routed via a service worker. assert_equals: nextHopProtocol is set on cached response expected "http/1.1" but got "" FAIL nextHopProtocol reports H2 correctly when routed via a service worker. assert_equals: nextHopProtocol is set on fallback expected "h2" but got "http/1.1"
Reintroduced test failure in r276122.
<rdar://problem/77061210>
I don't understand how this could work before or why it works in mac, because ServiceWorkerFetchTask::didFinish() always passes an empty metrics to Messages::WebResourceLoader::DidFinishResourceLoad(). Any idea Alex?
Fixed by r288088. Closing bug.