Hide Forgot
Description of problem: If a user with an invalid email address submits a job when the job completes beah will die with the following traceback: 2011-10-20 12:02:28,587 async_proc: INFO Task 63587 done. Submitting logs... 2011-10-20 12:02:28,726 _send_cmd: INFO Command ['Command', 'forward', '78e075fd-61a3-4db5-84c2-a6c53bc506e0', {'event': ['Event', 'extend_watchdog', '6b76ea7a-fe92-4a8d-95ed-65c1c143fbcd', {'source': 'beah_beaker_backend', 'id': '17b94718-0e24-4288-84ff-bd1dcffc58b8'}, None, {'timeout': 14394}]}] sent. 2011-10-20 12:02:29,683 simple_recipe: ERROR Encoutnered problem while running task '63587'. Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/beah/backends/beakerlc.py", line 538, in simple_recipe result = task.getResult() File "/usr/lib64/python2.6/site-packages/twisted/internet/defer.py", line 584, in getResult self.result.raiseException() File "/usr/lib64/python2.6/site-packages/twisted/internet/defer.py", line 605, in _deferGenerator result = g.next() File "/usr/lib/python2.6/site-packages/beah/backends/beakerlc.py", line 608, in _run_task thingy.getResult() File "/usr/lib64/python2.6/site-packages/twisted/internet/defer.py", line 584, in getResult self.result.raiseException() File "/usr/lib64/python2.6/site-packages/twisted/web/xmlrpc.py", line 307, in parseResponse response = xmlrpclib.loads(contents)[0][0] File "/usr/lib64/python2.6/xmlrpclib.py", line 1184, in loads return u.close(), u.getmethodname() File "/usr/lib64/python2.6/xmlrpclib.py", line 838, in close raise Fault(**self._stack[0]) Fault: <Fault 1: 'exceptions.ValueError:Invalid e-mail address.'> Version-Release number of selected component (if applicable): 0.7.3
mcsontos, beah should keep re-trying here too? We could have other failures which are transient. Could this explain some other bugs we've seen?
This looks like the same bug as bug 784237. However should we leave this bug open for beah to retry the failed call?
(In reply to comment #2) > This looks like the same bug as bug 784237. However should we leave this bug > open for beah to retry the failed call? agreed. beah should retry.
Bulk reassignment of issues as Bill has moved to another team.
Beah does already have retry logic, although it doesn't apply for this type of error which is not considered transient (and indeed, it *wasn't* a transient error, at least in the sense that it would've required the user to notice and fix their email address before the call could succeed). Anyway, this kind of failure is not really possible anymore. Nowadays the XMLRPC calls to start/stop recipes and record results are very lightweight on the Beaker side, all the activity like powering off systems and sending email notifications is done separately by the update_dirty_jobs() routine. So I don't think there is really anything else that needs fixing here.