Closed Bug 990202 Opened 6 years ago Closed 6 years ago

Intermittent test_framedhistoryframes.html | /tests/docshell/test/test_bug384014.html finished in a non-clean fashion, probably because it didn't call SimpleTest.finish()

Categories

(Core :: DOM: Navigation, defect)

Product:
Core
Component:
DOM: Navigation
Platform:
x86
Windows 7
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED FIXED
Milestone:
mozilla31
Tracking Flags:
Tracking Status
firefox29 --- fixed
firefox30 --- fixed
firefox31 --- fixed
firefox-esr24 --- unaffected
b2g-v1.3 --- fixed
b2g-v1.3T --- fixed
b2g-v1.4 --- fixed
b2g-v2.0 --- fixed

People

(Reporter: RyanVM, Assigned: ehsan)

References

Details

(Keywords: intermittent-failure)

Attachments

(1 file, 1 obsolete file)

Move the test page to a new window to prevent it from navigating the test runner frame back; r=bzbarsky
2.45 KB, patch
bzbarsky
: review+
Details | Diff | Splinter Review 
https://tbpl.mozilla.org/php/getParsedLog.php?id=37009356&tree=Mozilla-Inbound

Windows 7 32-bit mozilla-inbound debug test mochitest-1 on 2014-03-31 09:20:20 PDT for push 2c78aa5b01ff
slave: t-w732-ix-119

10:10:14     INFO -  6306 INFO TEST-START | /tests/docshell/test/test_framedhistoryframes.html
10:10:14     INFO -  ++DOMWINDOW == 73 (3C691898) [pid = 584] [serial = 6865] [outer = 19C931A8]
10:10:14     INFO -  ++DOCSHELL 420BD308 == 15 [pid = 584] [id = 1439]
10:10:14     INFO -  ++DOMWINDOW == 74 (3C6A6EC8) [pid = 584] [serial = 6866] [outer = 00000000]
10:10:14     INFO -  [Parent 584] WARNING: NS_ENSURE_TRUE(NS_SUCCEEDED(rv) && subjPrincipal) failed: file c:/builds/moz2_slave/m-in-w32-d-0000000000000000000/build/docshell/base/nsDocShell.cpp, line 8687
10:10:14     INFO -  ++DOMWINDOW == 75 (27CA65A8) [pid = 584] [serial = 6867] [outer = 3C6A6EC8]
10:10:14     INFO -  ++DOCSHELL 420BA888 == 16 [pid = 584] [id = 1440]
10:10:14     INFO -  ++DOMWINDOW == 76 (27CA5120) [pid = 584] [serial = 6868] [outer = 00000000]
10:10:14     INFO -  [Parent 584] WARNING: NS_ENSURE_TRUE(NS_SUCCEEDED(rv) && subjPrincipal) failed: file c:/builds/moz2_slave/m-in-w32-d-0000000000000000000/build/docshell/base/nsDocShell.cpp, line 8687
10:10:14     INFO -  ++DOMWINDOW == 77 (27CA8EB8) [pid = 584] [serial = 6869] [outer = 27CA5120]
10:10:14     INFO -  ++DOMWINDOW == 78 (27CA9A20) [pid = 584] [serial = 6870] [outer = 27CA5120]
10:10:14     INFO -  ++DOMWINDOW == 79 (38BF1960) [pid = 584] [serial = 6871] [outer = 27CA5120]
10:10:14     INFO -  ++DOMWINDOW == 80 (27C99CE8) [pid = 584] [serial = 6872] [outer = 19C931A8]
10:10:14     INFO -  ++DOCSHELL 420B8B50 == 17 [pid = 584] [id = 1441]
10:10:14     INFO -  ++DOMWINDOW == 81 (27CA0A68) [pid = 584] [serial = 6873] [outer = 00000000]
10:10:14     INFO -  [Parent 584] WARNING: Subdocument container has no frame: file c:/builds/moz2_slave/m-in-w32-d-0000000000000000000/build/layout/base/nsDocumentViewer.cpp, line 2418
10:10:14     INFO -  ++DOMWINDOW == 82 (27CA1EF0) [pid = 584] [serial = 6874] [outer = 27CA0A68]
10:10:14     INFO -  [Parent 584] WARNING: NS_ENSURE_TRUE(NS_SUCCEEDED(rv) && subjPrincipal) failed: file c:/builds/moz2_slave/m-in-w32-d-0000000000000000000/build/docshell/base/nsDocShell.cpp, line 8687
10:10:14     INFO -  ++DOMWINDOW == 83 (38C068B8) [pid = 584] [serial = 6875] [outer = 27CA0A68]
10:10:14     INFO -  ++DOMWINDOW == 84 (27C7F498) [pid = 584] [serial = 6876] [outer = 27CA0A68]
10:10:14     INFO -  ++DOMWINDOW == 85 (27C9ECC0) [pid = 584] [serial = 6877] [outer = 27CA0A68]
10:10:14     INFO -  6307 INFO TEST-INFO | MEMORY STAT vsize after test: 1330233344
10:10:14     INFO -  6308 INFO TEST-INFO | MEMORY STAT vsizeMaxContiguous after test: 355532800
10:10:14     INFO -  6309 INFO TEST-INFO | MEMORY STAT residentFast after test: 476356608
10:10:15     INFO -  6310 ERROR TEST-UNEXPECTED-FAIL | /tests/docshell/test/test_framedhistoryframes.html | /tests/docshell/test/test_bug384014.html finished in a non-clean fashion, probably because it didn't call SimpleTest.finish()
10:10:15     INFO -  6311 INFO TEST-END | /tests/docshell/test/test_framedhistoryframes.html | finished in 1231ms
Summary: Intermittent test_bug384014.html finished in a non-clean fashion, probably because it didn't call SimpleTest.finish() → Intermittent test_framedhistoryframes.html | /tests/docshell/test/test_bug384014.html finished in a non-clean fashion, probably because it didn't call SimpleTest.finish()
:(  Now this test navigates the frame back!
On the bright side, http://mxr.mozilla.org/mozilla-central/source/docshell/test/historyframes.html?force=1#19 already knows how to handle window.opener, so this is dead simple to fix!
Assignee: nobody → ehsan
Attachment #8399606 - Flags: review?(bzbarsky)
Comment on attachment 8399606 [details] [diff] [review]
Open the test in a new window to prevent it from navigating the test runner frame back; r=bzbarsky

Doesn't this just make the test identical to docshell/test/test_windowedhistoryframes.html ?  There's a reason historyframes.html handles both cases; we wanted to test both.

This test should definitely not be navigating anything outside itself.  If it is, we have a problem somewhere (possibly in the test itself, of course).
Attachment #8399606 - Flags: review?(bzbarsky) → review-
(In reply to Boris Zbarsky [:bz] from comment #8)
> Comment on attachment 8399606 [details] [diff] [review]
> Open the test in a new window to prevent it from navigating the test runner
> frame back; r=bzbarsky
> 
> Doesn't this just make the test identical to
> docshell/test/test_windowedhistoryframes.html ?  There's a reason
> historyframes.html handles both cases; we wanted to test both.

Oh, right.

> This test should definitely not be navigating anything outside itself.  If
> it is, we have a problem somewhere (possibly in the test itself, of course).

The thing is, this <http://mxr.mozilla.org/mozilla-central/source/docshell/test/historyframes.html?force=1#92> and maybe some of the other back() calls in this test seem to race with the session history items being injected somehow in a way that they navigate the test runner iframe itself back, so we really need to move this test into a separate window which lives outside of the test runner iframe.  Should I make this test open a new window which hosts an iframe pointing to historyframes.html?
Flags: needinfo?(bzbarsky)
> seem to race with the session history items being injected somehow 

That sounds like a bug we should fix!

For now, opening a new window that hosts the iframe is ok, I guess.
Flags: needinfo?(bzbarsky)
Blocks: 919208
(In reply to Boris Zbarsky [:bz] from comment #11)
> > seem to race with the session history items being injected somehow 
> 
> That sounds like a bug we should fix!

FWIW we've had this bug for as long as I can remember.  I don't actually know how this stuff works and have never looked into it very deeply but basically every time a test does history.back() off of an event/timeout on something loaded inside the test runner frame, the entire frame gets navigated back intermittently.  That is why I added the "finished in a non-clean fashion" warning back in bug 668728.
Attachment #8399606 - Attachment is obsolete: true
Attachment #8400259 - Flags: review?(bzbarsky)
Comment on attachment 8400259 [details] [diff] [review]
Move the test page to a new window to prevent it from navigating the test runner frame back; r=bzbarsky

r=me
Attachment #8400259 - Flags: review?(bzbarsky) → review+
Filed bug 991198 for this failure.
Relanded with the test disabled on b2g-desktop: https://hg.mozilla.org/integration/mozilla-inbound/rev/2c4c218a7a09
https://hg.mozilla.org/mozilla-central/rev/2c4c218a7a09
Status: NEW → RESOLVED
Closed: 6 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla31
You need to log in before you can comment on or make changes to this bug.