Re: Calendar-related errors during |make mozmill| test of C-C TB on tryserver.

On 2015年11月26日 02:19, Stefan Sitter wrote:
> See Bug 1226884. I suggest to test again tomorrow after the fix was
> merged from mozilla-inbound into mozilla-central. Hopefully Lightning
> will work again afterwards.
> 
> /Stefan
> 

Thank you for the info.
I will try.
(I tried earlier once today, but the tree was closed due to taskcluster
issue for try-comm-central then.)

Any idea, though, why linux build was not affected?
I am puzzled a bit.
I read the bug 1226884 and it seemed as if the bug ought to have happened
for linux build, too.
(Maybe a matter of synchronization of trees, or some differences of binary
utility tools.
I am not that familiar with build tools of mozilla.)

TIA

> 
> On 25.11.2015 16:58, ISHIKAWA, Chiaki wrote:
>> I am testing my patches for fixing some File I/O issues.
>>
>> I noticed I experienced errors in OS X build and Windows build during
>> |make mozmill| tests.
>> These errors are not observed in my local testing under linux and also
>> on linux build on tryserver.
>>
>> I am focusing on Calendar-related issues in my post here.
>> There are two errors I noticed:
>>
>> My tryserver log:
>> https://treeherder.mozilla.org/#/jobs?repo=try-comm-central&revision=c8c1373a5ae6&selectedJob=12322
>>
>> (1) testLocalCS.js
>>
>> In a mozmill test called testLocalCS.js,
>> I am getting the following error on Windows and OS X platforms while
>> linux shows no such error.
>>
>> NS_ERROR_XPC_GS_RETURNED_FAILURE
>>
>> I wonder why.
>>
>> (2) testTodayPane.js
>>
>> This file also produced error result.
>>
>> According to the log, in the early phase, I saw the following lines.
>> It looks as if the backend interface is not registered or recognized.
>>
>> Obviously something is wrong.
>> The following is from windows log, but
>>
>>   22:33:17     INFO -  [2468] WARNING: This method is lossy. Use
>> GetCanonicalPath !: file
>> c:/builds/moz2_slave/tb-try-c-cen-w32-d-00000000000/build/mozilla/xpcom/io/nsLocalFileWin.cpp,
>> line 3459
>>
>>   22:33:17     INFO -  1448433197340	addons.xpi-utils	WARN	addMetadata:
>> Add-on [email protected] is invalid: Error: Invalid addon ID:
>> expected addon ID [email protected], found
>> [email protected] in manifest
>> (resource://gre/modules/addons/XPIProvider.jsm ->
>> resource://gre/modules/addons/XPIProviderUtils.js:1626:15) JS Stack
>> trace: [email protected]:1626:15 <
>> [email protected]:1967:23 <
>> [email protected]:3614:34 <
>> [email protected]:2633:25 <
>> [email protected]:223:12 <
>> [email protected]:831:5 <
>> [email protected]:1011:9 <
>> [email protected]:2788:5 < [email protected]:58:7
>>
>>   22:33:17     INFO -  1448433197342	addons.xpi-utils	WARN	Could not
>> uninstall invalid item from locked install location
>>
>>   22:33:17     INFO -  [2468] ###!!! ASSERTION: must be a file:// url:
>> 'Error', file
>> c:/builds/moz2_slave/tb-try-c-cen-w32-d-00000000000/build/mozilla/netwerk/base/nsURLHelper.cpp,
>> line 188
>>
>>   22:33:17     INFO -  ### Error loading backend: [Exception...
>> "Component returned failure code: 0x8000ffff (NS_ERROR_UNEXPECTED)
>> [nsIFileProtocolHandler.getFileFromURLSpec]"  nsresult: "0x8000ffff
>> (NS_ERROR_UNEXPECTED)"  location: "JS frame ::
>> jar:file:///C:/slave/test/build/application/thunderbird/extensions/%7Be2fda1a4-762b-4020-b5ad-a41df1933103%7D.xpi!/components/calBackendLoader.js
>> :: loadBackend :: line 47"  data: no]
>>
>>   22:33:17     INFO -  JavaScript error:
>> resource://calendar/modules/calUtils.jsm, line 36: ReferenceError:
>> __LOCATION__ is not defined
>>
>>   22:33:17     INFO -  JavaScript error:
>> resource://calendar/modules/calUtils.jsm, line 36: ReferenceError:
>> __LOCATION__ is not defined
>>
>> There is a bugzilla entry filed this past February as follows: but it
>> seems to deal with a much simpler case of errors. It did not mention
>> anything about backend failure.
>>
>> Bug 1130047 - TEST-UNEXPECTED-FAIL |
>> /builds/slave/test/build/mozmill/testTodayPane.js |
>> testTodayPane.js::testTodayPane
>>
>> The latter also saw NS_ERROR_XPC_GS_RETURNED_FAILURE, but it also saw
>> NS_ERROR_XPC_CI_RETURNED_FAILURE
>>
>> Something is wrong with XPCOM registration and/or calling, etc.?
>> But since linux build does not show anything like that, I am a little
>> puzzled.
>>
>> Any pointers are appreciated.
>>
>> TIA
>>
>> PS: I am testing with the latest C-C source tree.
>> My local test environment is linux 64-bit.
>> The local C-C tree was refreshed this morning about an hour ago before
>> the tryserver submission was done, I think. I tested that the local
>> compilation succeeded before submitting a job to tryserver.
>> I don't have OS X or Windows build/test environment locally and so
>> I need to test my build using tryserver.
>>
>