ubuntu-phone team mailing list archive

[Paul Larson <paul.larson@xxxxxxxxxxxxx>]

177 looks like it could have some pulseaudio issue with mako in the
smoke tests.  During two of the tests, systemsettle-after failed and
it looked like pulseaudio was eating ~1.5% cpu. The two tests were
online_accounts_ui and dropping_letters_app.  I'm not sure if it's
truly the cause, but I did notice a couple of gstreamer updates,
including gstreamer1.0-pulseaudio.

On a positive note, the autopilot fix is in now, and
ubuntu-system-settings was able to run all the tests, bumping the test
total on mako to 521 tests run. I'd like to point out that the dbus
crash [1] is still present though.

Thanks,
Paul Larson

[1] http://jenkins.qa.ubuntu.com/job/trusty-touch-mako-smoke-daily/34/artifact/clientlogs/ubuntu_system_settings/_usr_sbin_system-image-dbus.32011.crash/*view*/

On Wed, Feb 12, 2014 at 12:04 PM, Didier Roche <didrocks@xxxxxxxxxx> wrote:
> Hey,
>
> First, I want to thanks Lukasz for his excellent work on the last couple of
> days. I know it was tough, but we keep getting work done!
> Since the last email, we got 2 images built:
>
> 175:
> * This one was all about nested Mir mode being back. That's aligned as well
> with the 4.4 android port.
>
> 176:
> * Fix for the qmlscene crash that we discussed a while back, blocking
> applications that failed on startup to be able to register a surface on Mir
> if they fail early. Note that this doesn't fix the application itself
> failing at startup, but at least, you are able then to relaunch it.
> * New Unity8, mostly with more support for the incoming scope preview work.
> * New webapps container bug fixes.
>
> This image is getting promoted!
>
> We see that the system settle is higher than what it was in previous images.
> Multiple components seems to sit on top, unity-system-compositor,
> ophono-simd... That can be nested mir related... Oliver discussed about it with
> upstream and a bug (https://bugs.launchpad.net/mir/+bug/1279391) was filed.
> This won't be considered as a blocker for now, but we'll keep it closely on
> our radar with upstream's help.
>
> We still have the regular flaky test which seems to be similar than the
> application not starting first (but we don't have the same issue that all
> following tests are failing then). Also, the previously accepted unity8 test
> failure is still around. I see more Unity 8 work getting in in the incoming
> images, let's hope they are going to fix this.
>
> We had some troubles due to Google App Scripts failing and creating random
> issues on the CI Train. We spent quite some time to debug and try to
> workaround it. However, the issue is not present anymore after few hours
> (and this service seems to be known to have some transient issues).
> Everything should be back to normal. We tried to recreate (the spreadsheet
> was reverted) the same state than this morning and let then more and more
> work getting in for the next image... which we are just kicking in!
>
> Cheers,
> Didier
>
> --
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to     : ubuntu-phone@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp