/usr/lib/arm-linux-gnueabihf/qt5/bin/qmlscene:6:qt_message_fatal:QMessageLogger::fatal: UbuntuClientIntegration::UbuntuClientIntegration:UbuntuMirClientIntegrationPlugin::create:loadIntegration

[errors.ubuntu.com bug bridge]

The Ubuntu Error Tracker has been receiving reports about a problem regarding qtdeclarative-opensource-src. This problem was most recently seen with version 5.3.0-3ubuntu10, the problem page at https://errors.ubuntu.com/problem/4618cad07e610ca3c59a3e07c582e963d4780359 contains more details.

steps to reproduce:
run the lrt switch test

1. open a bunch of apps
2. swipe the screen to show the app selector
3. pick an app at random
4. swipe the screen again to show the app selector
5. pick an app at random
6. repeat steps 2-5 until crash shows in /var/log

[Chris Gagnon (chris.gagnon)]

logs for https://errors.ubuntu.com/oops/3c04129c-3429-11e4-8804-fa163e4ccdf2

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[kevin gunn (kgunn72)]

marked this as duplicate
https://bugs.launchpad.net/ubuntu/+source/qtubuntu/+bug/1365673

[Gerry Boland (gerboland)]

This is message from a client trying to start, and unity8 either not expecting it and thus rejecting the client connection, or due to bug 1368101. Hard to know for certain

[Chris Gagnon (chris.gagnon)]

The fix from bug #1368101 is in the 20141008.1 rtm image, it did not fix the issue.
see:
https://errors.ubuntu.com/oops/5eb5f232-4f2d-11e4-9ea7-fa163e4aaad4

[Victor Tuson Palau (vtuson)]

not able to reproduce in my phone..

[Gerry Boland (gerboland)]

I've spent some time at this bug, running automated and random input tests to try reproduce it. I've found 1 probable cause for this issue: url-dispatcher starts apps if they're not running - and it appears to be possible that AppMan is told to respawn the same app around the same time - causing 2 instances of the same app to be executed - one of which AppMan will reject. I had this happen once anyway.

I suspect there's another issue causing this bug too (going by log file output anyway). Am still investigating.

So far, I've been unable to reproduce any user-facing fault related to this issue - AppMan is only rejecting a second instances of an app running.

[Chris Gagnon (chris.gagnon)]

I didn't see this in the 20141010.1 image on the krillin, the tests ran over a 3 day weekend.

[Olli Ries (ories)]

marked incomplete based on comment #10, reopen when reproducable again

[Chris Gagnon (chris.gagnon)]

I was able to reproduce this by putting the phone in flight mode and repeatedly opening the gmail app

https://errors.ubuntu.com/oops/e3efce40-5ae1-11e4-a164-fa163e339c81

[Chris Gagnon (chris.gagnon)]

this happened 19 times yesterday in the long running tests

[kevin gunn (kgunn72)]

marking as high per RTM bug review

[Ying-Chun Liu (paulliu)]

If using trunk version it doesn't crash at all.

[Chris Gagnon (chris.gagnon)]

This crashed cause failures in the dashboard on vivid mako https://jenkins.qa.ubuntu.com/job/vivid-touch-mako-smoke-daily/121/artifact/clientlogs/messaging_app/_usr_bin_messaging-app.32011.crash/*view*/

[Michał Sawicz (saviq)]

We'll need more time to investigate this, I'm thinking ww05-2015.

[Gerry Boland (gerboland)]

This bug consistently evades me. I've spent much time trying to reproduce this to no avail.

This fail will occur if mir client process are *not* launched via ubuntu-app-launch - as unity8 will not accept mir connections it doesn't expect, causing the client to crash with this error. The obvious possibility is a race where unity8 is informed of a client launch only after client tries to connect - but the logs don't indicate this.

There could be a bug somewhere in the stack causing a client to be launched twice, the first will be accepted, the second rejected and crashing. The logs do seem to indicate this possibility, but I cannot pinpoint where.

Logs from ubuntu-app-launch & upstart might help shed light.
Any suggestions on increasing the odds of reproducing this would be appreciated.

[Michał Sawicz (saviq)]

In light of this, and no known user impact, I'd drop this from the RTM list.

[Gerry Boland (gerboland)]

It has been a year, I've not heard this complaint in some time. I'm marking this as fixed, please re-open if it happens to you.