Closed Bug 1027125 Opened 10 years ago Closed 10 years ago

Intermittent browser_967000_button_sync.js "Non-local network connections are disabled and a connection attempt to token.services.mozilla.com (54.200.96.200) was made"

Categories

(Firefox :: Toolbars and Customization, defect)

x86
Windows 8.1
defect
Not set
normal
Points:
1

Tracking

()

RESOLVED FIXED
Firefox 33
Iteration:
33.2
Tracking Status
firefox31 --- fixed
firefox32 --- fixed
firefox33 --- fixed
firefox-esr24 --- unaffected
b2g-v1.3 --- unaffected
b2g-v1.3T --- unaffected
b2g-v1.4 --- unaffected
b2g-v2.0 --- fixed
b2g-v2.1 --- fixed

People

(Reporter: emorley, Assigned: Gijs)

References

Details

(Keywords: intermittent-failure)

Attachments

(1 file)

WINNT 6.2 mozilla-inbound pgo test mochitest-browser-chrome-1 on 2014-06-18 04:56:59 PDT for push 9cd95132bc9f

slave: t-w864-ix-100

https://tbpl.mozilla.org/php/getParsedLog.php?id=41956147&tree=Mozilla-Inbound

{
05:08:37     INFO -  TEST-START | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js
05:08:37     INFO -  TEST-INFO | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | Entering test
05:08:37     INFO -  TEST-INFO | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | Check Sync button functionality
05:08:37     INFO -  TEST-PASS | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | The Sync button was added to the Panel Menu
05:08:37     INFO -  TEST-INFO | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | Wait for tab event: load
05:08:37     INFO -  TEST-INFO | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | Console message: [JavaScript Warning: "Unknown pseudo-class or pseudo-element '-webkit-search-cancel-button'.  Ruleset ignored due to bad selector." {file: "chrome://browser/content/aboutaccounts/normalize.css" line: 366 column: 22 source: "input[type="search"]::-webkit-search-cancel-button,"}]
05:08:37     INFO -  TEST-INFO | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | Console message: [JavaScript Warning: "Error in parsing value for 'min-height'.  Declaration dropped." {file: "chrome://browser/content/aboutaccounts/main.css" line: 125 column: 16 source: "    min-height: none;"}]
05:08:37     INFO -  TEST-INFO | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | Console message: [JavaScript Warning: "Expected media feature name but found 'min-device-pixel-ratio'." {file: "chrome://browser/content/aboutaccounts/main.css" line: 144 column: 17 source: "only screen and (min-device-pixel-ratio: 2),"}]
05:08:37     INFO -  TEST-INFO | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | skipping spurious load event
05:08:37     INFO -  TEST-INFO | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | Tab event received: load
05:08:37     INFO -  TEST-PASS | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | Firefox Sync page opened
05:08:37     INFO -  Non-local network connections are disabled and a connection attempt to token.services.mozilla.com (54.200.96.200) was made.  You should only access hostnames available via the test networking proxy (if running mochitests) or from a test-specific httpd.js server (if running xpcshell tests).  Browser services should be disabled or redirected to a local server.TEST-PASS | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | The panel closed
05:08:37     INFO -  TEST-INFO | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | Leaving test
05:08:37     INFO -  TEST-INFO | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | Entering test asyncCleanup
05:08:37     INFO -  TEST-PASS | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | The panel UI is in default state again.
05:08:38     INFO -  TEST-INFO | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | Leaving test asyncCleanup
05:08:38     INFO -  TEST-INFO | Main app process: exit status 80000003
05:08:38  WARNING -  TEST-UNEXPECTED-FAIL | chrome://mochitests/content/browser/browser/components/customizableui/test/browser_967000_button_sync.js | application terminated with exit code 2147483651
}
Blocks: 967000
Gijs, can you suggest somebody to look at this test that's intermittently hitting the network? My guess is it doesn't like the about:accounts tab being loaded?
*sigh* Would have helped if I'd actually put the needinfo? on I meant to.
Flags: needinfo?(gijskruitbosch+bugs)
It's weird that it's intermittent, and also that it's corresponding to a timeout in an earlier test (literally every occurrence so far seems to have that issue). I don't really know what'd be going on. Will leave the needinfo for a bit. Hopefully I will have time to look at this in more detail later today.
Yeah, I still don't really understand why this would be intermittent. I would expect it to either fail completely, or not at all. However, looking at the other aboutAccounts test, it looks like we should set the "identity.fxaccounts.remote.signup.uri" pref to something non-remote, cf. http://mxr.mozilla.org/mozilla-central/source/browser/base/content/test/general/browser_aboutAccounts.js#37
Assignee: nobody → gijskruitbosch+bugs
Status: NEW → ASSIGNED
Iteration: --- → 33.2
Points: --- → 1
Flags: needinfo?(gijskruitbosch+bugs) → firefox-backlog+
Tim, does this look right to you?
Attachment #8445771 - Flags: review?(ttaubert)
Marco, can you add this to this iteration?
Flags: needinfo?(mmucci)
Whiteboard: [qa-]
Would adding the pref to prefs_general.js mean we avoid hitting this again in another test that decides to use about:accounts?
Or to be clearer: Would it be better to add to prefs_general.js, that way we don't have to play whac-a-mole when new tests are added?
(In reply to Ed Morley [:edmorley UTC+0] from comment #12)
> Or to be clearer: Would it be better to add to prefs_general.js, that way we
> don't have to play whac-a-mole when new tests are added?

Yes. I would imagine the other about:accounts referenced URIs should go in there too. Tim would probably have a more complete list than I do!
Flags: needinfo?(ttaubert)
Comment on attachment 8445771 [details] [diff] [review]
set sync signup pref for button test,

Review of attachment 8445771 [details] [diff] [review]:
-----------------------------------------------------------------

Sorry, I don't have a complete list, looking at what the about:accounts tests do seems fine. Having the pref set globally for all tests might indeed make sense, we hit the same problem in bug 1027084 where about:home tries to load snippets.
Attachment #8445771 - Flags: review?(ttaubert) → review+
Added to Iteration 32.2
QA Whiteboard: [qa-]
Flags: needinfo?(mmucci)
Whiteboard: [qa-]
remote:   https://hg.mozilla.org/integration/fx-team/rev/f57d24c3e843
Flags: needinfo?(ttaubert)
Keywords: leave-open
pref("identity.fxaccounts.auth.uri", "https://api.accounts.firefox.com/v1");
pref("identity.fxaccounts.remote.signup.uri", "https://accounts.firefox.com/signup?service=sync&context=fx_desktop_v1");
pref("identity.fxaccounts.remote.force_auth.uri", "https://accounts.firefox.com/force_auth?service=sync&context=fx_desktop_v1");
pref("identity.fxaccounts.remote.signin.uri", "https://accounts.firefox.com/signin?service=sync&context=fx_desktop_v1");
pref("identity.fxaccounts.settings.uri", "https://accounts.firefox.com/settings");

Where some of these are only in all.js (auth.uri) and some only in firefox.js (the other 4). Which doesn't make any sense to me. rnewman, do you know why?

In any case, I guess we can set all 5 of them to example.com in prefs_general.js... patch + try push coming up...
Flags: needinfo?(rnewman)
(In reply to Tim Taubert [:ttaubert] (away July 7th-18th) from comment #14)
> Having the pref set globally for all tests might indeed
> make sense, we hit the same problem in bug 1027084 where about:home tries to
> load snippets.

Filed bug 1030149.
(In reply to Ed Morley [:edmorley UTC+0] from comment #18)
> (In reply to Tim Taubert [:ttaubert] (away July 7th-18th) from comment #14)
> > Having the pref set globally for all tests might indeed
> > make sense, we hit the same problem in bug 1027084 where about:home tries to
> > load snippets.
> 
> Filed bug 1030149.

Then we can close this once this has landed, I think.
Keywords: leave-open
Whiteboard: [qa-]
Whiteboard: [qa-]
Attachment #8445771 - Flags: checkin+
https://hg.mozilla.org/mozilla-central/rev/f57d24c3e843
Status: ASSIGNED → RESOLVED
Closed: 10 years ago
Resolution: --- → FIXED
Target Milestone: --- → Firefox 33
(In reply to :Gijs Kruitbosch from comment #17)

> Where some of these are only in all.js (auth.uri) and some only in
> firefox.js (the other 4). Which doesn't make any sense to me. rnewman, do
> you know why?

I don't. They should probably all be in firefox.js until FxA is used in other products.

Conceivably the difference is that one is for the programmatic API, and the others are for content-based UI.
Flags: needinfo?(rnewman)
Recent stars in bug 1027181 make it look like this isn't really fixed, at least not when the test before it happens to fail.
You need to log in before you can comment on or make changes to this bug.