Thursday, 2015-08-27

*** SheenaG has quit IRC00:07
*** edtubill has joined #openstack-barbican00:15
*** lisaclark1 has joined #openstack-barbican00:22
*** dave-mcc_ has joined #openstack-barbican00:24
*** dave-mccowan has quit IRC00:27
*** vivek-ebay has quit IRC00:35
*** edtubill has quit IRC00:38
*** lisaclark1 has quit IRC01:17
*** vivek-ebay has joined #openstack-barbican01:29
*** vivek-eb_ has joined #openstack-barbican01:47
*** vivek-ebay has quit IRC01:48
*** lisaclark1 has joined #openstack-barbican02:08
rm_workok, looks like the best way to test this is to do it as an experimental gate temporarily --
rm_workredrobot: if you have a sec to +1 that :P02:08
rm_workor anyone who happens to be around, take a look02:08
pksinghrm_work : Hi, I have a query can you please help me with it02:10
openstackgerritDave McCowan proposed openstack/barbican: Ensure Project and ProjectQuotas Tables are in sync
rm_workwhat's up?02:10
pksinghrm_work: how to setup snake_oil plugin for CA02:11
rm_workI am not sure I can help with that02:11
rm_workI was looking into it (and helping define it) a long time ago but I never actually got it working02:11
rm_workand someone else actually handled the majority of the patch :(02:11
pksinghrm_work: OK02:12
rm_workif you are somewhat far along and just need help with an error or something i can maybe help02:13
pksinghi was just trying to setup snake_oil plugin fir CA02:14
pksinghi changed config file and restarted the services but get api for CAS shows empty list02:14
rm_workis there anything in the logs or in the barbican console output for when it tries to load up that plugin?02:16
pksinghi am running start after modifying the conf file , and there is nothing unusual on the terminal02:18
dave-mcc_pksingh which config file did you change?  depending how you installed and started barbican, there maybe two config files?  " install" makes a copy of the config file.02:19
pksinghdave-mcc_: i changed barbican.conf as in
dave-mcc_in which directory?02:20
pksinghfile is at /etc/barbican/barbican.conf02:20
dave-mcc_" install" will copy ./etc/barbican/barbican.conf to /etc/barbican/barbican.conf  (perhaps overwriting your change).  did that happen?02:22
pksinghi did not executed " install"02:23
pksinghi installed through devstack02:23
*** lisaclark1 has quit IRC02:29
*** vivek-eb_ has quit IRC02:32
rm_workpksingh: so there should still be logs somewhere02:39
rm_workthat shows what happens when barbican starts up02:39
rm_workand it SHOULD spit out *something* about the plugins it loads02:40
rm_workI am pretty sure02:40
pksinghi was checking at /opt/stack/logs/barbican.log and found nothing related to CA02:45
openstackgerritDave McCowan proposed openstack/barbican: Add get_count() method to project resource repositories
*** dave-mcc_ has quit IRC02:54
*** yuanying_ has joined #openstack-barbican03:06
*** yuanying_ has quit IRC03:07
*** yuanying has quit IRC03:10
*** vivek-ebay has joined #openstack-barbican03:13
*** yuanying has joined #openstack-barbican04:01
*** jamielennox is now known as jamielennox|away04:04
*** jamielennox|away is now known as jamielennox04:05
*** xaeth_afk is now known as xaeth04:09
openstackgerritOpenStack Proposal Bot proposed openstack/barbican: Updated from global requirements
*** xaeth is now known as xaeth_afk04:23
*** yuanying has quit IRC04:32
*** vivek-ebay has quit IRC04:33
*** yuanying has joined #openstack-barbican04:37
*** kebray has joined #openstack-barbican04:59
*** elarson has quit IRC05:00
*** elarson has joined #openstack-barbican05:01
*** openstackgerrit has quit IRC05:16
*** openstackgerrit has joined #openstack-barbican05:17
openstackgerritArun Kant proposed openstack/barbican: Barbican audit using audit middleware for pycadf events
*** Nirupama has joined #openstack-barbican05:49
*** yuanying has quit IRC05:58
*** xaeth_afk is now known as xaeth06:22
*** DTadrzak has quit IRC06:35
*** DTadrzak has joined #openstack-barbican06:49
*** xaeth is now known as xaeth_afk07:05
*** insequent has quit IRC07:14
*** insequent has joined #openstack-barbican07:27
*** DTadrzak has quit IRC07:38
*** DTadrzak has joined #openstack-barbican07:39
*** kebray has quit IRC07:45
*** shohel has joined #openstack-barbican07:59
*** jaosorior has joined #openstack-barbican08:02
*** yuanying has joined #openstack-barbican08:03
openstackgerritMerged openstack/barbican: Migrate to using pre-devstack-update
*** shohel has quit IRC09:40
*** shohel has joined #openstack-barbican09:45
*** shohel has quit IRC09:49
*** shohel has joined #openstack-barbican09:59
*** mmdurrant has quit IRC10:09
*** shohel has quit IRC10:12
*** shohel has joined #openstack-barbican10:13
*** Kevin_Zheng_ is now known as Kevin_Zheng10:49
*** insequent has quit IRC10:49
*** insequent has joined #openstack-barbican11:02
*** dave-mccowan has joined #openstack-barbican11:30
*** mmdurrant has joined #openstack-barbican11:58
*** alee has quit IRC12:21
*** openstackgerrit has quit IRC12:31
*** openstackgerrit has joined #openstack-barbican12:32
*** Nirupama has quit IRC12:36
*** zz_dimtruck is now known as dimtruck12:55
*** dimtruck is now known as zz_dimtruck12:58
*** alee has joined #openstack-barbican13:24
*** kfarr has joined #openstack-barbican13:46
*** rellerreller has joined #openstack-barbican13:46
*** shohel has quit IRC13:48
*** shohel has joined #openstack-barbican13:48
*** spotz_zzz is now known as spotz14:05
dave-mccowanrellerreller ping14:07
*** pglass has joined #openstack-barbican14:09
rellerrellerdave-mccowan pong14:11
dave-mccowanrellerreller: were you asking about running only certain tests based on a tag recently?14:11
rellerrellerdave-mccowan I was14:12
rellerrellerdave-mccowan I actually found the source of the bug yesterday afternoon. I was quite excited.14:12
dave-mccowanrellerreller did you find a solution?   i found one, but we're not sure if it's the best one.
*** jlhinson has joined #openstack-barbican14:13
rellerrellerdave-mccowan here's the issue with testing. We are using testtools for unit tests and nosetest for functional tests.14:14
rellerrellerdave-mccowan so for testtools the attributes that are applied to functions are lost if they are a parameterized test. The utils package that builds the functions has a bug that I need to fix.14:15
rellerrellerdave-mccowan I have not found the source of the bug for nosetest attributes, but I believe the issue to be the same.14:16
dave-mccowanrellerreller cool. with my CR I got nosetest attributes to be recognized by the nosetest CLI.  but, the nosetest CLI did not recognize the testtools attributes.  i guess that could be as-designed, or it could be due to the bug you found.14:17
rellerrellerdave-mccowan I just looked at source for nose, and I think it is the same issue.14:17
rellerrellerdave-mccowan no, the two frameworks do not handle each others attributes.14:18
openstackgerritMerged openstack/barbican: Updated from global requirements
rellerrellerdave-mccowan what happens is that the attribute decorators assign attributes to the functions.14:19
rellerrellerdave-mccowan then the filtering capability is applied by looking at the attributes.14:19
dave-mccowanrellerreller what about the "testr" CLI?  do you know which attributes it can read, if any?14:20
rellerrellerdave-mccowan our utils parameterized_test_case class decorator looks for all functions that have our "build_data" attribute assigned to them.14:20
rellerrellerdave-mccowan but our utils for building the new function does not transfer other attributes to the new functions. Hence nosetest and testtools cannot find them.14:21
rellerrellerdave-mccowan I think testr sits above testtools and nosetests. I don't think it knows anything about attributes, but I could be wrong. I'm still learning why testr is used.14:22
dave-mccowanrellerreller: nice find.  i got lucky with my tags, since they were not parameterized_test_cases14:22
rellerrellerdave-mccowan like I said only parameterized tests with attributes have this problem. That was the tricky part for me.14:23
*** lisaclark1 has joined #openstack-barbican14:23
dave-mccowanrellerreller i think using each tool in different places has contributed to some mysterious gate failures in the past.14:23
dave-mccowanrellerreller, so once you fix that bug, how do specify that you want to run a set of tests with a certain tag?14:25
rellerrellerdave-mccowan so for the functional tests that use nose I believe you should be able to pass a '-a' flag to the tox job.14:26
rellerrellerdave-mccowan I do not yet know how to do it for the unit tests. Unfortunately testtools does not have a flag like that.14:27
rellerrellerdave-mccowan It seems like testtools wants you to list all tests by doing ` --list barbican.tests` and then filter on the results.14:27
rellerrellerdave-mccowan The list operation returns a newline delimited list of test methods. Tests that have an attribute are named "barbican.tests.SomeClass.test_foo[some_attrib]".14:29
*** lisaclark1 has quit IRC14:29
*** lisaclark1 has joined #openstack-barbican14:30
dave-mccowanreller Oh cool... so you can regex on [some_attrib].  I know that testr support regex.14:30
rellerrellerdave-mccowan then if you only want to run tests that have an attribute of some_attrib then delete all lines that do not have the regex "\[.*some_attrib.*\]".14:30
rellerrellerdave-mccowan That is good to know because I was disappointed by the implementation.14:30
dave-mccowanrellerreller... this sounds like nose and testr can read each other's attributes.14:31
rellerrellerdave-mccowan I thought it was weird to have a developer do grep commands to some temporary file and then feed that back to
rellerrellerdave_mccowan which part makes you think that?14:31
dave-mccowanrellerreller here's an example of using regex to filter by class name.
rellerrellerdave-mccowan thanks! That is helping me figure out the last pieces of the puzzle here.14:34
dave-mccowanrellerreller currently in the barbican code testtools.testcase.attr() is used in functional tests.  and "tox -e functional" uses nose.  if the -a flag works for functional tests, then nose must read testtools attributes.14:35
rellerrellerdave-mccowan I wonder if we should convert the functional tests to testtools. By that I think that just means changing the attributes to testtool attributes and using testtools to run instead of nose.14:35
rellerrellerdave-mccowan the -a flag does not work.14:35
dave-mccowanrellerreller where do you see nose attributes in functional tests?14:36
rellerrellerdave-mccowan Sorry, I forgot that I added a new test class with nose attributes for testing.14:36
rellerrellerdave-mccowan That was my first step in tracking down this problem.14:37
dave-mccowanrellerreller ah... ok.   me too in the CR i linked above.14:37
dave-mccowanrellerreller... so Barbican should either go nose all the way or testtools all the way.14:38
rellerrellerdave-mccowan I noticed that tests were run with nose but had testtools attributes. I thought if I simply changed the attributes that things would work.14:38
dave-mccowanrellerreller ... and it would have, except there was the bug you found in paramterized14:38
rellerrellerdave-mccowan I think so, but maybe there are some other requirements for why we use both.14:38
rellerrellerdave-mccowan I think you are right that if I can fix the bug and change attributes to nose attributes then things would work.14:39
dave-mccowanrellerreller using both has been a pain for me.  i have a set of functional tests that can't be run concurrently.   i have to tell testr not to do that one way, and nosetests not to do that another way.14:40
*** insequent has quit IRC14:41
rellerrellerdave-mccowan That does not sound pleasant.14:41
rellerrellerdave-mccowan I know that I would prefer to use one. That way I can learn one and stick with it. I do not like have to learn the different command line args when running the tests manually to try and debug.14:42
dave-mccowanrellerereller  it works fine until code review when i have to justify two different tags. :-)14:42
dave-mccowanrellerreller +114:42
*** diazjf has joined #openstack-barbican14:42
rellerrellerdave-mccowan that code review sounds fun :-)14:43
dave-mccowanrellerreller thanks.  this chat helped a bunch.14:44
rellerrellerdave-mccowan Thank you as well.14:44
*** kebray has joined #openstack-barbican14:44
*** xaeth_afk is now known as xaeth14:44
dave-mccowanhockeynut, therve: good read above ^^^  any reason why we can't purge barbican of either testr or nose, and stick with one?14:45
rellerrellerdave-mccowan I need a little time to digest this, but we should bring this up at status meeting and try to fix this.14:45
dave-mccowanrellerreller +1 i'll add it to the agenda.14:45
hockeynutI'm actually taking today off but I will catch up on the discussion.14:45
rellerrellerdave-mccowan sounds good. I need to check on my other project now :(14:46
*** darrenmoffat has quit IRC14:57
*** darrenmoffat has joined #openstack-barbican14:58
*** lisaclark1 has quit IRC15:02
*** edtubill has joined #openstack-barbican15:03
*** ccneill has joined #openstack-barbican15:03
*** kebray has quit IRC15:07
*** silos has joined #openstack-barbican15:09
*** shohel has quit IRC15:15
openstackgerritKaitlin Farr proposed openstack/castellan: Update Barbican wrapper
openstackgerritFernando Diaz proposed openstack/barbican: Allow debug flag in Barbican tox and debugging documentation
*** kebray has joined #openstack-barbican15:34
*** SheenaG has joined #openstack-barbican15:39
*** lisaclark1 has joined #openstack-barbican15:45
*** insequent has joined #openstack-barbican15:45
*** lisaclark1 has quit IRC15:46
*** lisaclark1 has joined #openstack-barbican15:47
*** vivek-ebay has joined #openstack-barbican15:52
openstackgerritElvin Tubillara proposed openstack/barbican: Add request ID to worker-side logs
*** kebray has quit IRC15:58
*** vivek-ebay has quit IRC16:00
openstackgerritAmy Marrich proposed openstack/barbican: Removes uwsgi and pyenv from
*** chadlung has joined #openstack-barbican16:02
*** vivek-ebay has joined #openstack-barbican16:10
*** vivek-ebay has quit IRC16:11
*** woodster_ has joined #openstack-barbican16:20
*** jaosorior has quit IRC16:25
*** arunkant_ has joined #openstack-barbican16:49
*** pglass has quit IRC16:53
*** pglass has joined #openstack-barbican16:54
*** tkelsey has joined #openstack-barbican16:59
dave-mccowanquotas patch review of the day:
*** pglbutt has joined #openstack-barbican17:02
rm_workanyone have a moment to look at and give it a +1? So people on infra know that someone on barbican is actually aware of this? :P17:02
rm_workdave-mccowan: i don't quite understand what is happening in your review -- why project_id vs. external_project_id?17:05
rm_work like in you actually call self.repo.create_or_update_by_project_id(, external_project_id, ...)17:05
*** pglass has quit IRC17:06
rm_workis this function "a user setting a quota for a project that isn't theirs"?17:06
dave-mccowanrm_work: project_id is the id created by the barbican database.  external_project_id is the id created by keystone that is used in the API.17:06
dave-mccowanrm_work  yea.  all the setting/reading of project-quotas is done by a service-admin managing all the customers' projects, not his own project.17:07
openstackgerritKaitlin Farr proposed openstack/castellan: Update Barbican functional tests
openstackgerritMerged openstack/barbican: Remove rpmbuild directory
*** lisaclark1 has quit IRC17:26
*** lisaclark1 has joined #openstack-barbican17:35
*** vivek-ebay has joined #openstack-barbican17:35
openstackgerritAdam Harwell proposed openstack/barbican: Use the new Devstack external plugin method
*** tkelsey has quit IRC17:36
rm_workis this seriously not something we want to do in the client still? Otherwise, this needs to be implemented in every project that takes barbican refs from users, which seems to violate DRY
openstackLaunchpad bug 1383935 in python-barbicanclient "Retrieving secrets from untrusted URIs can leak auth tokens" [High,Won't fix]17:39
*** lisaclark1 has quit IRC17:42
*** ccneill has quit IRC17:44
*** SheenaG has quit IRC17:45
*** vivek-ebay has quit IRC17:51
*** ccneill has joined #openstack-barbican17:56
*** lisaclark1 has joined #openstack-barbican18:00
*** silos has left #openstack-barbican18:15
*** DTadrzak has quit IRC18:17
diazjfhey everyone can you help me get this merged :-D. redrobot, got some time tomorrow to google hangout about the behaviors in the funct tests?18:21
*** lisaclark1 has quit IRC18:33
*** pglbutt has quit IRC18:52
*** silos has joined #openstack-barbican19:05
* kfarr slaps kfarr around a bit with a large fishbot19:10
kfarr^^ Anyone else using ?  FYI, for anyone in a room, you can 'ignore', 'query', 'whois', and now 'slap' someone.  And I just found out that it shows up in the whole room for everyone to see19:17
*** ccneill_ has joined #openstack-barbican19:19
*** ccneill has quit IRC19:22
diazjfthanks for the review kfarr!! :)19:23
*** ccneill_ is now known as ccneill19:26
kfarrdiazjf sure thing!19:29
dave-mccowankfarr haha.  i was wondering what a fishbot was19:33
*** rellerreller has quit IRC19:41
*** mmdurrant has quit IRC19:44
*** jorgem has left #openstack-barbican19:45
*** mmdurrant has joined #openstack-barbican19:46
*** silos has left #openstack-barbican19:50
*** diazjf has quit IRC19:57
*** silos1 has joined #openstack-barbican19:57
*** ccneill_ has joined #openstack-barbican19:57
*** ccneill has quit IRC19:59
*** silos1 has left #openstack-barbican20:00
*** diazjf has joined #openstack-barbican20:01
*** pglass has joined #openstack-barbican20:25
*** diazjf has quit IRC20:30
*** silos has joined #openstack-barbican20:32
*** diazjf has joined #openstack-barbican20:36
*** dave-mccowan has quit IRC20:37
*** alee has quit IRC20:38
*** silos has left #openstack-barbican20:38
rm_workany barbican cores around to +1 this and add some legitimacy? possibly not necessary, but maybe useful:
*** kebray has joined #openstack-barbican20:49
*** gyee has joined #openstack-barbican20:51
*** kebray has quit IRC21:09
*** kfarr has quit IRC21:11
*** insequent has quit IRC21:15
*** mixos has joined #openstack-barbican21:30
*** mixos has quit IRC21:35
*** chadlung has quit IRC21:36
rm_worksoooo close to having this job in to test new gating T_T21:46
rm_workwhere is everyone this week anyway? seems very very quiet21:47
diazjfrm_work, tooooo quiet21:48
*** alee has joined #openstack-barbican21:48
*** diazjf has left #openstack-barbican21:52
*** pglass has quit IRC21:55
*** silos has joined #openstack-barbican21:56
*** mixos has joined #openstack-barbican21:59
rm_workmaybe someone's hunting wabbits22:03
*** edtubill has left #openstack-barbican22:06
*** chadlung has joined #openstack-barbican22:07
openstackgerritChristopher Solis proposed openstack/python-barbicanclient: Adds epilog to parser
*** xaeth is now known as xaeth_afk22:11
*** ccneill_ has quit IRC22:12
*** chadlung has quit IRC22:12
*** insequent has joined #openstack-barbican22:12
*** silos has left #openstack-barbican22:13
*** ccneill_ has joined #openstack-barbican22:16
*** SheenaG has joined #openstack-barbican22:25
*** mixos has quit IRC22:43
*** spotz is now known as spotz_zzz22:59
*** ccneill_ has quit IRC23:07
*** nelsnels_ has joined #openstack-barbican23:10
*** nelsnelson has quit IRC23:12
*** jlhinson has quit IRC23:13
*** chadlung has joined #openstack-barbican23:19
*** arunkant_ has quit IRC23:20
*** chadlung has quit IRC23:24
*** insequent has quit IRC23:25
*** mixos has joined #openstack-barbican23:44
*** SheenaG has quit IRC23:50

Generated by 2.14.0 by Marius Gedminas - find it at!