Closed Bug 1035172 Opened 6 years ago Closed 5 years ago

Intermittent test_touchcaret.py test_touchcaret.TouchCaretTest.test_contenteditable_move_caret_to_end_by_dragging_touch_caret_to_bottom_right_corner | TimeoutException: TimeoutException: Error loading page, timed out (onDOMContentLoaded)

Categories

(Testing :: Marionette, defect)

Product:
Testing
Component:
Marionette
Platform:
x86
Linux
Type:
defect
Priority:
Not set

Tracking

(Not tracked)

Status:
RESOLVED WORKSFORME

People

(Reporter: cbook, Assigned: automatedtester)

References

(
URL
)

Details

(Keywords: intermittent-failure, pi-marionette-intermittent)

Attachments

(2 files)

Disable touch caret test while being investigated
729 bytes, patch
mdas
: review+
Details | Diff | Splinter Review
Re-enable touch caret test.
582 bytes, patch
automatedtester
: review+
Details | Diff | Splinter Review 
b2g_emulator_vm b2g-inbound opt test marionette-webapi on 2014-07-06 21:06:12 PDT for push 506038ddee2c

slave: tst-linux64-spot-402

https://tbpl.mozilla.org/php/getParsedLog.php?id=43225014&tree=B2g-Inbound



TEST-UNEXPECTED-FAIL | test_touchcaret.py test_touchcaret.TouchCaretTest.test_contenteditable_move_caret_to_end_by_dragging_touch_caret_to_bottom_right_corner | TimeoutException: TimeoutException: Error loading page, timed out (onDOMContentLoaded)
Return code: 10
Not sure what made this start, but it's happening extremely frequently at this point.
Flags: needinfo?(tlin)
Assignee: nobody → dburns
Status: NEW → ASSIGNED
Attachment #8453008 - Flags: review?(mdas) → review+
I have disabled the test since the intermittent numbers are going off the chart and this will give tlin time to work on it.

https://hg.mozilla.org/integration/mozilla-inbound/rev/a3e4f653e475
Whiteboard: [leave open]
About two days ago, a new marionette test suite was introduced in bug 1000040 [1]. test_touchcaret.py is just run after those new tests. I've retriggered a bunch of Mnw in [2] to identify the regression.

[1] https://hg.mozilla.org/mozilla-central/rev/506038ddee2c
[2] https://tbpl.mozilla.org/?rev=085eea991bb9
Flags: needinfo?(tlin)
(In reply to Ting-Yu Lin [:TYLin] (UTC+8) from comment #86)
> About two days ago, a new marionette test suite was introduced in bug
> 1000040 [1]. test_touchcaret.py is just run after those new tests. I've
> retriggered a bunch of Mnw in [2] to identify the regression.
> 
> [1] https://hg.mozilla.org/mozilla-central/rev/506038ddee2c
> [2] https://tbpl.mozilla.org/?rev=085eea991bb9

My guess is that it is the disconnect test since that *may* kill leave marionette in a sad state since it might kill the networking we rely on.

I have pushed a patch to try https://tbpl.mozilla.org/?tree=Try&rev=2474de2a11ad

I might not be around to do the retriggers to check it until much later so anyone and everyone feel free to retrigger it
https://tbpl.mozilla.org/?tree=Try&rev=ccb6192a40e6(In reply to David Burns :automatedtester from comment #92)
> (In reply to Ting-Yu Lin [:TYLin] (UTC+8) from comment #86)
> > About two days ago, a new marionette test suite was introduced in bug
> > 1000040 [1]. test_touchcaret.py is just run after those new tests. I've
> > retriggered a bunch of Mnw in [2] to identify the regression.
> > 
> > [1] https://hg.mozilla.org/mozilla-central/rev/506038ddee2c
> > [2] https://tbpl.mozilla.org/?rev=085eea991bb9
> 
> My guess is that it is the disconnect test since that *may* kill leave
> marionette in a sad state since it might kill the networking we rely on.
> 
> I have pushed a patch to try
> https://tbpl.mozilla.org/?tree=Try&rev=2474de2a11ad
> 
> I might not be around to do the retriggers to check it until much later so
> anyone and everyone feel free to retrigger it

The bug still occurs after disabling test_ethernet_disconnect.js. I've push another try to disable the entire network test. I did not see the same syndrome so far.
I'm going to back out bug 1000040 first.
Depends on: 1000040
(In reply to Ting-Yu Lin [:TYLin] (UTC+8) from comment #102)
> https://tbpl.mozilla.org/?tree=Try&rev=ccb6192a40e6(In reply to David Burns
> :automatedtester from comment #92)
> > (In reply to Ting-Yu Lin [:TYLin] (UTC+8) from comment #86)
> > > About two days ago, a new marionette test suite was introduced in bug
> > > 1000040 [1]. test_touchcaret.py is just run after those new tests. I've
> > > retriggered a bunch of Mnw in [2] to identify the regression.
> > > 
> > > [1] https://hg.mozilla.org/mozilla-central/rev/506038ddee2c
> > > [2] https://tbpl.mozilla.org/?rev=085eea991bb9
> > 
> > My guess is that it is the disconnect test since that *may* kill leave
> > marionette in a sad state since it might kill the networking we rely on.
> > 
> > I have pushed a patch to try
> > https://tbpl.mozilla.org/?tree=Try&rev=2474de2a11ad
> > 
> > I might not be around to do the retriggers to check it until much later so
> > anyone and everyone feel free to retrigger it
> 
> The bug still occurs after disabling test_ethernet_disconnect.js. I've push
> another try to disable the entire network test. I did not see the same
> syndrome so far.

Disconnect/Disable operation is executed in every ethernet test cases (except cases don't do enable/connect ). However, the these operations are supposed executing on eth1 instead of eth0.
Comment on attachment 8453664 [details] [diff] [review]
Re-enable touch caret test.

Review of attachment 8453664 [details] [diff] [review]:
-----------------------------------------------------------------

ship it
Attachment #8453664 - Flags: review?(dburns) → review+
Keywords: checkin-needed
Whiteboard: [leave open]
I can not reproduce locally... ;( (Run ethernet test and touch caret test sequentially and they work fine)
Need more investigation.
No longer blocks: CopyPasteLegacy
[Mass closure] intermittent has not been seen in 2 months and has not been disabled so closing
Status: ASSIGNED → RESOLVED
Closed: 5 years ago
Resolution: --- → WORKSFORME
Removing leave-open keyword from resolved bugs, per :sylvestre.
Keywords: leave-open
You need to log in before you can comment on or make changes to this bug.