lrt test found crash file _usr_bin_webbrowser-app.32011.crash

[Chris Gagnon]

Summary:
Crash file _usr_bin_webbrowser-app.32011.crash was found

Steps to reproduce:
1. Run lrt test in lp:lrt

Expected result:
No crashes happen

Actual result:
A crash happened

System info:
base: system-image.ubuntu.com
http_port: 80
https_port: 443
channel: ubuntu-touch/utopic-proposed
build_number: 87
version_detail: ubuntu=20140618,device=20140530,version=87
Device=mako

[Chris Gagnon (chris.gagnon)]

crash file

[Chris Gagnon (chris.gagnon)]

Unfortunatly the crash file isn't very good :(

[Olivier Tilloy (osomon)]

It looks like the stack is corrupted in the crash file, not much we can do with it. The only useful information I’m seeing is that the crash occured while webbrowser-app was browsing http://www.weather.com/weather/hourbyhour/graph/ARBA0009?par=ubu.

[Chris Gagnon (chris.gagnon)]

There are a lot of these crashes webbrowser crashes in lrt-crashes, so I thought I would open one so we can track it.

https://bugs.launchpad.net/lrt-crashes/?field.searchtext=_usr_bin_webbrowser-app.32011.crash

I haven't found one with a good stacktrace yet :/

[Olivier Tilloy (osomon)]

@Chris, can you elaborate on what those lrt tests do exactly? I’m gonna need to be able to reproduce the crash locally to investigate it.

[Chris Gagnon (chris.gagnon)]

lrt tests are long running tests of autopilot randomly clicking and swiping the phone until it finds a crash file.

[Olivier Tilloy (osomon)]

@Chris: any chance we can get a newer crash file if this is still happening? Perhaps we’ll be able to extract a more useful backtrace.

[Allan LeSage (allanlesage)]

Recognizing that we need a better trace to fix but tagging for our qa review--ChrisGagnon has a better trace surfaced?

[Chris Gagnon (chris.gagnon)]

https://bugs.launchpad.net/lrt-crashes/+bug/1347283 a better stack trace has surfaced here.

[Chris Gagnon (chris.gagnon)]

It's still not a full stacktrace :(

[Leo Arias (elopio)]

I saw a crash while testing krillin, image #109 14.09-proposed: bug #1381876.
Feel free to mark it as duplicate of this one if that is more appropriate.

[Olivier Tilloy (osomon)]

Marking invalid as the information in this bug is too old to be useful, and bug #1347283 seems to have a better stack trace anyway.