 |
 |
That's probably your problem. Internally, the decorator is performing isinstance(exc, excType) basically where excType is sqlalchemyExc and that's actually the python module sqlalchemy.exc, which, is not an exception type. If you want to catch all SQLAlchemy errors, you actually want to do: from sqlalchemy.exc import SQLAlchemyError ...with the following decorator: @dbService.errorhandler(SQLAlchemyError) Brian > On Sep 9, 2015, at 11:29 AM, Becla, Jacek <[log in to unmask]> wrote: > > from sqlalchemy import exc as sqlalchemyExc > > see > > https://github.com/lsst/qserv/blob/tickets/DM-2558/core/modules/wmgr/python/dbMgr.py > > > > On 09/09/2015 10:27 AM, Van Klaveren, Brian N. wrote: >> What is sqlalchemyExc? >> >> Brian >> >> >>> On Sep 9, 2015, at 9:49 AM, Becla, Jacek <[log in to unmask]> wrote: >>> >>> Andy (and Brian?) >>> >>> I don't know if that can easily be improved... Here is what I typically get when I introduce a problem in wmgr: >>> >>> Traceback (most recent call last): >>> File "/home/becla/stack/Linux64/flask/0.10.1+11/lib/python/Flask-0.10.1-py2.7.egg/flask/app.py", line 1817, in wsgi_app >>> response = self.full_dispatch_request() >>> File "/home/becla/stack/Linux64/flask/0.10.1+11/lib/python/Flask-0.10.1-py2.7.egg/flask/app.py", line 1477, in full_dispatch_request >>> rv = self.handle_user_exception(e) >>> File "/home/becla/stack/Linux64/flask/0.10.1+11/lib/python/Flask-0.10.1-py2.7.egg/flask/app.py", line 1378, in handle_user_exception >>> if isinstance(e, typecheck): >>> TypeError: isinstance() arg 2 must be a class, type, or tuple of classes and types >>> 2015-09-09 11:37:43,574 [PID:4849] [INFO] (_log() at _internal.py:87) werkzeug: 127.0.0.1 - - [09/Sep/2015 11:37:43] "POST /dbs HTTP/1.1" 500 - >>> >>> >>> to get a sensible stack trace that is actually useful, I end up >>> commenting out: >>> >>> [log in to unmask](DbException) >>> [log in to unmask](sqlalchemyExc) >>> [log in to unmask](MySQLdb.Error) >>> [log in to unmask](MySQLdb.Warning) >>> #def dbExceptionHandler(error): >>> # """ All leaked database-related exceptions generate 500 error """ >>> # return errorResponse(500, error.__class__.__name__, str(error)) >>> >>> >>> This is the error I would get if errorhandler code would be >>> commented out: >>> >>> >>> Traceback (most recent call last): >>> File "/home/becla/stack/Linux64/flask/0.10.1+11/lib/python/Flask-0.10.1-py2.7.egg/flask/app.py", line 1817, in wsgi_app >>> response = self.full_dispatch_request() >>> File "/home/becla/stack/Linux64/flask/0.10.1+11/lib/python/Flask-0.10.1-py2.7.egg/flask/app.py", line 1477, in full_dispatch_request >>> rv = self.handle_user_exception(e) >>> File "/home/becla/stack/Linux64/flask/0.10.1+11/lib/python/Flask-0.10.1-py2.7.egg/flask/app.py", line 1381, in handle_user_exception >>> reraise(exc_type, exc_value, tb) >>> File "/home/becla/stack/Linux64/flask/0.10.1+11/lib/python/Flask-0.10.1-py2.7.egg/flask/app.py", line 1475, in full_dispatch_request >>> rv = self.dispatch_request() >>> File "/home/becla/stack/Linux64/flask/0.10.1+11/lib/python/Flask-0.10.1-py2.7.egg/flask/app.py", line 1461, in dispatch_request >>> return self.view_functions[rule.endpoint](**req.view_args) >>> File "/usr/local/home/becla/stack/repos/qserv/lib/python/lsst/qserv/wmgr/dbMgr.py", line 194, in createDb >>> dbConn = Config.instance().privDbEngine().connect() >>> File "/usr/local/home/becla/stack/repos/qserv/lib/python/lsst/qserv/wmgr/config.py", line 120, in privDbEngine >>> inst = engineFactory().getEngineFromArgs(**kwargs) >>> NameError: global name 'engineFactory' is not defined >>> >>> >>> See? That is so much better. I entered engineFactory instead of >>> EngineFactory. There is no way I could have figured it out from the >>> previous stack trace... >>> >>> How can we improve that? >>> >>> Jacek >> > ######################################################################## Use REPLY-ALL to reply to list To unsubscribe from the QSERV-L list, click the following link: https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=QSERV-L&A=1