Last modified: 2014-01-30 14:05:56 UTC
Some API calls on beta cluster throw exceptions which eventually lead to a fatal error. Exception: 2014-01-30 01:17:23 deployment-apache33 enwiki: [48af842e] /w/api.php Exception from line 84 of php-master/extensions/Wikidata/extensions/Wikibase/repo/Wikibase.hooks.php: Wikibase: Incomplete configuration: $wgWBRepoSettings["entityNamespaces"] has to be set to an array mapping content model IDs to namespace IDs. See ExampleSettings.php for details and examples. #0 [internal function]: Wikibase\RepoHooks::onSetupAfterCache() #1 php-master/includes/Hooks.php(206): call_user_func_array('Wikibase\RepoHo...', Array) #2 php-master/includes/GlobalFunctions.php(3976): Hooks::run('SetupAfterCache', Array, NULL) #3 php-master/includes/Setup.php(520): wfRunHooks('SetupAfterCache') #4 php-master/includes/WebStart.php(153): require_once('/data/project/a...') #5 php-master/api.php(44): require('/data/project/a...') #6 w/api.php(3): require('/data/project/a...') #7 {main} Which then lead to the fatal error: [30-Jan-2014 01:17:23] Fatal error: Call to a member function disable() on a non-object at php-master/includes/GlobalFunctions.php on line 2140 Server: i-0000031b Method: POST URL: http://en.wikipedia.beta.wmflabs.org/w/api.php Cookie: XXXXXXX Backtrace: #0 php-master/includes/GlobalFunctions.php(2140): wfHttpError() #1 php-master/includes/Exception.php(257): wfHttpError(500, 'Internal Server...', '[48af842e] /w/a...') #2 php-master/includes/Exception.php(612): MWException->report() #3 php-master/includes/Exception.php(686): MWExceptionHandler::report(Object(MWException)) #4 [internal function]: MWExceptionHandler::handle(Object(MWException)) #5 {main} Note how line #1 of the fatal error stacktrace points to an error 500 generated by the exception.
Note the exception is thrown before $wgOut has a chance to be initialized. That leads to a fatal error in wfHttpError() global functions when it attempts to call: $wgOut->disabled(). The configuration check for $wgWBRepoSettings["entityNamespaces"] seems to happen too early during initialization :(
quite sure this is a duplicate and is resolved. if it reoccurs, of course, reopen *** This bug has been marked as a duplicate of bug 60606 ***