Rev4 MacOSX Lion 10.7 mozilla-inbound opt test mochitests-1/5 on 2012-09-05 06:58:57 PDT for push 6afc8e389883 slave: talos-r4-lion-063 https://tbpl.mozilla.org/php/getParsedLog.php?id=14980679&tree=Mozilla-Inbound { 57373 INFO TEST-INFO | /tests/content/canvas/test/webgl/test_webgl_conformance_test_suite.html | [conformance/textures/gl-teximage.html] (WebGL mochitest) Starting test page 57374 INFO TEST-PASS | /tests/content/canvas/test/webgl/test_webgl_conformance_test_suite.html | [conformance/textures/gl-teximage.html] Test passed - getError was expected value: NO_ERROR : Should be no errors from setup. 57375 ERROR TEST-UNEXPECTED-FAIL | /tests/content/canvas/test/webgl/test_webgl_conformance_test_suite.html | [conformance/textures/gl-teximage.html] Timeout in this test page }

Android Tegra 250 mozilla-inbound opt test mochitest-1 on 2012-09-05 17:51:16 PDT for push 72c85979e852 slave: tegra-262 https://tbpl.mozilla.org/php/getParsedLog.php?id=14999107&tree=Mozilla-Inbound

bjacob, please may you look at this? (Or if you don't have the cycles, let me know and we can just disable the test on Android).

It's not that I don't have the cycles --- it's that I have looked into this kind of intermittent WebGL test timeouts on Android before and they can only be explained as an infra issue, most likely the Android slaves don't have enough memory. So, disabling a single page in the WebGL mochitest tends to displace the problem to the next big WebGL test page. The better compromise, rather than disabling the WebGL mochitest altogether, would be to disable only say 80% of it and keep a smart 20% subset that would be enough to catch most regressions. Let me try to see if I can make such a patch.

Resolving WFM any keyword:intermittent-failure bug where: * Changed: (is less than or equal to) -3m * Whiteboard: (contains none of the strings) random disabled marked fuzzy todo fails failing annotated time-bomb * Whiteboard: (does not contain the string) leave open There will inevitably be some false positives; for that (and the bugspam) I apologise, but at least this will clear out the open cruft (and thus reduce risk of mis-starring) on TBPL's annotated summary bug suggestions. Filter on orangewfm.