http/tests/appcache/cyrillic-uri.html timed out on Snow Leopard Release Bot
--- /Volumes/Data/WebKit-BuildSlave/snowleopard-intel-release-tests/build/layout-test-results/http/tests/appcache/cyrillic-uri-expected.txt 2010-03-31 22:17:55.000000000 -0700
+++ /Volumes/Data/WebKit-BuildSlave/snowleopard-intel-release-tests/build/layout-test-results/http/tests/appcache/cyrillic-uri-actual.txt 2010-03-31 22:17:55.000000000 -0700
@@ -1,3 +1,4 @@
+FAIL: Timed out waiting for notifyDone to be called
Test that non-ASCII URIs work correctly in cache manifests.
Should be a series of PASS messages, followed with DONE.
@@ -9,5 +10,4 @@
I'm not sure why we need bugs for each test timeout. There is nothing possibly flaky in most of these tests - it seems to be some general issue, so we could as well end up with a bug for every test.
Bugs give us a permanent record. We can always dup them to some root cause once we have some clue what that cause is.
How would a person fixing a flaky bit of code know which of the unreproducible bugs were caused by it? Or how would we know it if it's fixed by a CFNetwork or even lower level change?
Perhaps you could consider using wiki for keeping track of infrequent failures such as this.
Chromium has a flakiness dashboard which is automatically kept up to date based on real data from the bots. That would be a better solution than our current ad-hoc bug solution. However this ad-hoc bug solution is better than ignoring the bots. Note that the bots have become way greener since we've started filing failures about them in bugs.webkit.org. :)
timed out: http://build.webkit.org/results/SnowLeopard%20Intel%20Release%20%28Tests%29/r57141%20%287751%29/results.html