Re: [Pulp-dev] Changing behavior of the pclean alias

2019-03-20 Thread Daniel Alley
Neither python manage.py test (which is only meant to be used with unit tests) nor pytest (which is only meant to be used with functional tests) should be impacted by this change. On Wed, Mar 20, 2019 at 2:59 PM Dana Walker wrote: > What if you're not using django-admin test but instead pytest?

Re: [Pulp-dev] Changing behavior of the pclean alias

2019-03-20 Thread Dana Walker
What if you're not using django-admin test but instead pytest? I noticed that when I'd been running a lot of tests, and then manually went to create a distribution, etc, the numbers indicated that way more had been created than I had done manually (and this was more than a month ago, before the ch

Re: [Pulp-dev] Changing behavior of the pclean alias

2019-03-20 Thread Pavel Picka
+1 I think it is not issue it was just I exited tests in dirty way - that will be always issue. On Wed, Mar 20, 2019 at 5:44 PM Dennis Kliban wrote: > Austin, this happens when you use the django-admin test runner. It creates > a test database when running tests. If you interrupt the tests, the

Re: [Pulp-dev] Changing behavior of the pclean alias

2019-03-20 Thread Dennis Kliban
Austin, this happens when you use the django-admin test runner. It creates a test database when running tests. If you interrupt the tests, the database does not get removed. The next time your run django-admin test, you get prompted about deleting the existing test db. You can only run the tests ag

Re: [Pulp-dev] Changing behavior of the pclean alias

2019-03-20 Thread Austin Macdonald
That's interesting! Pavel, next time you see this, could you file an issue that includes what is in the DB? On Wed, Mar 20, 2019, 12:01 Pavel Picka wrote: > I noticed when I run test sometimes it left something in DB... but I found > why, as tests didn't finish completely it didn't destroy db af

Re: [Pulp-dev] Changing behavior of the pclean alias

2019-03-20 Thread Pavel Picka
I noticed when I run test sometimes it left something in DB... but I found why, as tests didn't finish completely it didn't destroy db after itself. so take only +1 and skip everything after On Wed, Mar 20, 2019 at 3:56 PM Austin Macdonald wrote: > Pavel, can you elaborate? > > On Wed, Mar 20, 2

Re: [Pulp-dev] Changing behavior of the pclean alias

2019-03-20 Thread Austin Macdonald
Pavel, can you elaborate? On Wed, Mar 20, 2019 at 9:43 AM Pavel Picka wrote: > +1 and what do you think about idea of dropping all DBs as sometimes test > let there some (it doesn't affect pulp but when clean db so clean it > whole)? > > On Wed, Mar 20, 2019 at 12:12 PM Dennis Kliban wrote: > >

Re: [Pulp-dev] Changing behavior of the pclean alias

2019-03-20 Thread Pavel Picka
+1 and what do you think about idea of dropping all DBs as sometimes test let there some (it doesn't affect pulp but when clean db so clean it whole)? On Wed, Mar 20, 2019 at 12:12 PM Dennis Kliban wrote: > +1 > > On Tue, Mar 19, 2019 at 5:39 PM Daniel Alley wrote: > >> I created a new PR here

Re: [Pulp-dev] Changing behavior of the pclean alias

2019-03-20 Thread Dennis Kliban
+1 On Tue, Mar 19, 2019 at 5:39 PM Daniel Alley wrote: > I created a new PR here [0] which changes the behavior of the pclean alias > so that it also wipes out /var/lib/pulp/ in addition to dropping and > recreating the database. > > Unless anyone objects, I plan to merge it tomorrow afternoon (

[Pulp-dev] Changing behavior of the pclean alias

2019-03-19 Thread Daniel Alley
I created a new PR here [0] which changes the behavior of the pclean alias so that it also wipes out /var/lib/pulp/ in addition to dropping and recreating the database. Unless anyone objects, I plan to merge it tomorrow afternoon (Wednesday the 20th) [0] https://github.com/pulp/ansible-pulp3/pull