I forgot to say that I will be out tomorrow morning. Please have our usual morning call. I will talk to Graham and Francesco about when we can reschedule our weekly calls.
Talk to you all later, Gary On Mar 14, 2012, at 9:50 PM, Gary Poster <[email protected]> wrote: > - Clint agreed to take over packaging of our charm helpers stuff, so gmb is > off the hook. > > - Benji is pretty confident that he's reduced our failing bug count from > ~2800 to ~200, with https://code.launchpad.net/~benji/launchpad/bug-954319 . > Yay! I started an ec2 instance to land this. It should have landed by 0800 > UTC Thursday, or failed. > > - When trying to verify Benji's fix and frankban's testr fix, I went down an > unpleasantly long rabbit hole that I eventually discovered was because testr > in precise and testr in head are significantly different. You can still > apply his basic idea, but copying over cli.py from his branch is not advised. > > It looks like the idea needs some tweaking. I got this traceback when I ran > the tests. > > Warning: Permanently added '10.0.3.224' (RSA) to the list of known hosts. > Exception in thread Thread-1: > Traceback (most recent call last): > File "/usr/lib/python2.7/threading.py", line 551, in __bootstrap_inner > self.run() > File "/usr/lib/python2.7/threading.py", line 504, in run > self.__target(*self.__args, **self.__kwargs) > File "/usr/lib/python2.7/dist-packages/testtools/testsuite.py", line 85, in > _run_test > test.run(process_result) > File "/usr/lib/python2.7/dist-packages/subunit/__init__.py", line 1149, in > run > protocol.lineReceived(line) > File "/usr/lib/python2.7/dist-packages/subunit/__init__.py", line 548, in > lineReceived > self._state.lineReceived(line) > File "/usr/lib/python2.7/dist-packages/subunit/__init__.py", line 385, in > lineReceived > self.details_parser.lineReceived(line) > File "/usr/lib/python2.7/dist-packages/subunit/details.py", line 119, in > lineReceived > self._parse_state(line) > File "/usr/lib/python2.7/dist-packages/subunit/details.py", line 84, in > _look_for_content > self._state.endDetails() > File "/usr/lib/python2.7/dist-packages/subunit/__init__.py", line 380, in > endDetails > self._report_outcome() > File "/usr/lib/python2.7/dist-packages/subunit/__init__.py", line 411, in > _report_outcome > details=self.details_parser.get_details()) > File "/usr/lib/python2.7/dist-packages/testtools/testresult/real.py", line > 517, in addFailure > return self.decorated.addFailure(test, err) > File "/usr/lib/python2.7/dist-packages/testtools/testresult/real.py", line > 405, in addFailure > test, err, details=details) > File "/usr/lib/python2.7/dist-packages/testtools/testresult/real.py", line > 387, in _add_result_with_semaphore > self.result.stopTest(test) > File "/usr/lib/python2.7/dist-packages/testtools/testresult/real.py", line > 603, in stopTest > return self.decorated.stopTest(test) > File "/usr/lib/python2.7/dist-packages/testtools/testresult/real.py", line > 237, in stopTest > return self._dispatch('stopTest', test) > File "/usr/lib/python2.7/dist-packages/testtools/testresult/real.py", line > 231, in _dispatch > for result in self._results) > File "/usr/lib/python2.7/dist-packages/testtools/testresult/real.py", line > 231, in <genexpr> > for result in self._results) > File "/usr/lib/python2.7/dist-packages/testtools/testresult/real.py", line > 603, in stopTest > return self.decorated.stopTest(test) > File "/usr/lib/python2.7/dist-packages/subunit/test_results.py", line 417, > in stopTest > getattr(self.decorated, method)(*args, **kwargs) > File "/usr/lib/python2.7/dist-packages/subunit/test_results.py", line 125, > in addFailure > return self.super.addFailure(test, err, details=details) > File "/usr/lib/python2.7/dist-packages/subunit/test_results.py", line 58, > in addFailure > return self.decorated.addFailure(test, err, details=details) > File "/usr/lib/python2.7/dist-packages/testtools/testresult/real.py", line > 517, in addFailure > return self.decorated.addFailure(test, err) > File "/usr/lib/python2.7/dist-packages/subunit/test_results.py", line 58, > in addFailure > return self.decorated.addFailure(test, err, details=details) > File "/usr/lib/python2.7/dist-packages/testtools/testresult/real.py", line > 517, in addFailure > return self.decorated.addFailure(test, err) > File "/usr/lib/python2.7/dist-packages/testtools/testresult/real.py", line > 517, in addFailure > return self.decorated.addFailure(test, err) > File "/usr/lib/python2.7/dist-packages/testrepository/ui/cli.py", line 50, > in addFailure > self.stream.write(self._format_error(u'FAIL', *(self.failures[-1]))) > File "/usr/lib/python2.7/dist-packages/testrepository/ui/cli.py", line 42, > in _format_error > ]).encode(encoding, 'replace') > TypeError: encode() argument 1 must be string, not None > > String or Integer object expected for key, unicode found > That suggests I got the patch wrong...here's what _format_error looks like in > the version I ran: > > def _format_error(self, label, test, error_text): > encoding = getattr(self.stream, 'encoding', 'utf-8') > return u''.join([ > self.sep1, > u'%s: %s\n' % (label, test.id()), > self.sep2, > error_text, > ]).encode(encoding, 'replace') > > So...I guess encoding was explicitly None, and the code needs to be on the > order of "getattr(self.stream, 'encoding', None) or 'utf-8'"? > > I don't know what that "String or Integer object expected for key, unicode > found" means--or, more accurately, what raised it. Probably unrelated. > > After that stderr, I found this stdout text, which appeared well after the > tests started: > running=sudo /usr/local/bin/launchpad-lxc-test > Setting up ephemeral container... > Starting up the container... > No handlers could be found for logger "bzr" > No handlers could be found for logger "amqplib" > No handlers could be found for logger "amqplib" > Did something restart? :-/ Not sure what that means. When I looked at ps, > there was only one test process running. > > - All of the above means we *still* need to get a buildbot running again, > ideally with a fix for that unicode issue. (OTOH, the unicode issue may no > longer be triggered, once we have Benji's branch? The odds are good....) > > - I suspect this is entirely in line with what Francesco told us about the > resolv.conf issues, but in case it helps, locally I got this warning when I > started an lxc container (as set up by setuplxc): > > /etc/resolvconf/update.d/libc: Warning: /etc/resolv.conf is not a symbolic > link to /etc/resolvconf/run/resolv.conf > > - I added a new card for missing tests--the Javascript layer tests. They > failed for me when I ran tests using --load-list. Layer problems don't show > up in testr's output for this, but I don't yet have confidence in testr as a > tool in this regard. I worry that in fact the tests are not being run. > > OK, I need to stop. Talk to you all tomorrow > > Gary > -- > Mailing list: https://launchpad.net/~yellow > Post to : [email protected] > Unsubscribe : https://launchpad.net/~yellow > More help : https://help.launchpad.net/ListHelp
-- Mailing list: https://launchpad.net/~yellow Post to : [email protected] Unsubscribe : https://launchpad.net/~yellow More help : https://help.launchpad.net/ListHelp
