Is this the reason why there was no update today on nightly-latest? On Thu, Feb 4, 2016 at 6:39 PM, Hubert Figuière <h...@mozilla.com> wrote:
> On 04/02/16 11:16 AM, Alberto Pastor wrote: > > Gij11 and Gij18 are broken since Jan 26th [1]. Is anybody looking at > > it? I tried to check what’s going on, but both pass locally without > > problems. I haven’t found a gaia commit that could have caused this > > by the time it started failing. > > > > Does anybody has a clue? Is treeherder somehow running differently > > than when running the test locally? It seems that both are related to > > the AudioChannels. I created [2] and [3] to track the progress. > > I'm gonna be on a hunch here: > > 1. I believe the change is in the platform, aka Gecko > > 2. It pass locally but not on automation: we have seen discrepancies > with this on the music app, caused be the difference in the sound > management at the OS level: on automation the sound devices are > "dummies" and cause a different set of error codes. At least they were > when we tried to properly deal with playback errors and everything > started failing. > > 3. I don't think there is a need to change the test. The test check that > we show the playback icon which is done in response to sound events > (and AudioChannel). It is how these events are sent that could have > changed. > > Again, this is just a hunch and I could be totally wrong. > > Hub > _______________________________________________ > dev-fxos mailing list > dev-fxos@lists.mozilla.org > https://lists.mozilla.org/listinfo/dev-fxos >
_______________________________________________ dev-fxos mailing list dev-fxos@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-fxos
