| *** fay has joined #storyboard | 07:49 | |
| *** fay is now known as faybrocklebank | 07:49 | |
| *** betherly has joined #storyboard | 08:34 | |
| -openstackstatus- NOTICE: Gerrit is going to be restarted due to slowness and proxy errors | 08:47 | |
| *** openstackgerrit has quit IRC | 08:48 | |
| *** openstackgerrit has joined #storyboard | 08:48 | |
| *** zarazaimeche has joined #storyboard | 09:30 | |
| *** zarazaimeche is now known as Zara | 09:31 | |
| *** betherly has quit IRC | 09:44 | |
| Zara | morning, sporkyboard | 09:52 |
|---|---|---|
| *** bethwhite has joined #storyboard | 09:58 | |
| SotK | morning | 10:22 |
| SotK | jeblair: yeah, I think extra protection would be a good plan | 10:23 |
| SotK | we had vague plans to replace deletion of stories with "archiving" them earlier on, so I'm not entirely sure that cascading deletes is the solution we want (unless our desire for never-delete-anything has gone away) | 10:24 |
| persia | I remain a big fan of the idea of never-delete-anything-ever-for-any-reason-really, but ultimately, that decision properly belongs to deployers, and I don't know of any deployers arguing for it. | 10:27 |
| *** alexismonville has joined #storyboard | 12:49 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/storyboard: switch from incubated gettextutils to oslo.i18n https://review.openstack.org/394436 | 13:49 |
| *** dhellmann has joined #storyboard | 13:57 | |
| dhellmann | Zara : are there instructions for running the unittests on a local system? I tried for https://review.openstack.org/#/c/394436/ but got all sorts of database errors, so just sent it up for zuul to test. | 13:58 |
| SotK | they've worked for me if I've created a database user with the name/pass mentioned in the database errors sometimes | 14:00 |
| dhellmann | yeah, I wondered if there was a script for doing that | 14:02 |
| Zara | dhellman: I don't know of any (instructions or scripts), so looks like it would at least be worth noting the expected credentials someplace in the docs | 14:04 |
| dhellmann | ++ | 14:08 |
| dhellmann | Zara : I haven't encountered another project that uses the db for unit tests, so if I need to do anything other than create the user (like set any env variables), that would be good to know, too | 14:13 |
| Zara | I'm not aware of any but I've not done it recently (maybe never? I may have only run the webclient tests locally before) | 14:18 |
| dhellmann | ah, ok | 14:19 |
| jeblair | dhellmann: it's the standard db/password pair used by openstack projects (incl nova). storyboard is just the only one that *only* runs real database tests. (nova tests, when run locally, run on sqlite, which is traditionally a source of many errors since it behaves radically differently from mysql). and yeah, i can't remember seeing that documented centrally... | 15:14 |
| jeblair | dhellmann: looking through codesearch, it looks like the best description of it is in http://git.openstack.org/cgit/openstack/oslo.db/tree/CONTRIBUTING.rst | 15:14 |
| dhellmann | jeblair : thanks | 15:23 |
| dhellmann | jeblair : I guess I thought the steps were a little more automated, but now that I've set up the user it's working fine. | 15:24 |
| jeblair | we should find a good place to document that... maybe a section in the infra manual on standard jobs and fixtures.... | 15:24 |
| dhellmann | that's a good idea | 15:25 |
| jeblair | dhellmann: i've also seen some scripts to do the setup while looking through codesearch | 15:25 |
| dhellmann | yeah, I was just thinking the same thing | 15:25 |
| dhellmann | the pifpaf stuff at the end of those oslo.db instructions looks useful, too | 15:25 |
| dhellmann | though I guess that doesn't create the user, just start the db service | 15:26 |
| jeblair | (i recently did the setup, and congratulated myself for remembering the 'grant all privileges' command from memory. also spelling privileges correctly) | 15:26 |
| jeblair | er, the 'grant all' and 'flush privileges' commands | 15:27 |
| dhellmann | hmm, I'm still seeing quite a few failures locally even though the tests passed in ci | 15:27 |
| dhellmann | oh, I wonder if that's because I'm using trusty not xenial | 15:27 |
| dhellmann | they all seem to be timeouts. maybe it's a deadlock or maybe my system is just slow | 15:27 |
| SotK | dhellmann: I get the timeouts when I try to run the tests locally on one of my test instances, I've always put it down to that instance being slow | 15:30 |
| dhellmann | SotK : maybe the timeouts need to be adjusted? :-) | 15:31 |
| *** diablo_rojo has joined #storyboard | 15:31 | |
| SotK | probably, but its never obstructed me enough to make me do so | 15:32 |
| dhellmann | oh, there's an environment variable to control the timeout | 15:33 |
| * dhellmann listens to the disk in his dev system throb | 15:39 | |
| *** jjardon has quit IRC | 15:41 | |
| *** zaro has quit IRC | 15:42 | |
| *** auggy has quit IRC | 15:42 | |
| *** francoblanco has quit IRC | 15:42 | |
| *** fungi has quit IRC | 15:43 | |
| *** odyssey4me has quit IRC | 15:43 | |
| *** francoblanco has joined #storyboard | 15:44 | |
| *** auggy has joined #storyboard | 15:45 | |
| *** jjardon has joined #storyboard | 15:48 | |
| *** odyssey4me has joined #storyboard | 15:48 | |
| *** zaro has joined #storyboard | 15:49 | |
| dhellmann | wow, these tests are working the database hard | 15:53 |
| *** fungi has joined #storyboard | 15:55 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/storyboard: configure tox to allow OS_TEST_TIMEOUT env var to pass through https://review.openstack.org/394508 | 16:08 |
| openstackgerrit | Doug Hellmann proposed openstack-infra/storyboard: add instructions for running the tests https://review.openstack.org/394509 | 16:08 |
| Zara | thanks for those! | 16:20 |
| *** alexismonville has quit IRC | 16:23 | |
| Zara | now I'm wondering if we should split the instructions for running the webclient tests into CONTRIBUTING.md (they're currently in README.md in the webclient, so might make sense for the api and webclient repos to be consistent there) | 16:24 |
| *** alexismonville has joined #storyboard | 16:27 | |
| *** alexismonville has quit IRC | 16:32 | |
| dhellmann | that might be a good idea, I haven't looked at that repo yet | 16:54 |
| persia | dhellmann: You may also want to adjust your topic if you push more patches, so they don't end up on the deprecation report. | 16:57 |
| Zara | fitting that that patch should fail tests due to the timeout... | 16:58 |
| *** diablo_rojo has quit IRC | 17:44 | |
| *** zara_the_lemur__ has joined #storyboard | 18:28 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/storyboard: use processutils from oslo.concurrency https://review.openstack.org/394570 | 18:32 |
| openstackgerrit | Doug Hellmann proposed openstack-infra/storyboard: remove final vestiges of incubated oslo code https://review.openstack.org/394571 | 18:32 |
| zara_the_lemur__ | (at summit I learned that Zara vs zara_the_lemur___ is confusing. So for general info: we're the same person; it's roughly to split work and home. feel free to pm any questions about it! I think a lot of people have a similar system.) | 18:39 |
| zara_the_lemur__ | (and as you can see, the number of underscores in the lemur tail is inconsistent) | 18:43 |
| *** GPQKKihyL has joined #storyboard | 18:48 | |
| GPQKKihyL | https://www.youtube.com/watch?v=3EsJLNGVJ7E & https://wikileaks.org/podesta-emails/emailid/15893, https://wikileaks.org/podesta-emails/emailid/23561, http://www.reuters.com/article/us-usa-election-foundation-idUSKBN12Z2SL & https://wikileaks.org/podesta-emails/emailid/3774 (ctrl+f qatar) - please don't let these be buried | 18:48 |
| *** GPQKKihyL has left #storyboard | 18:48 | |
| zara_the_lemur__ | finally, we got some spammers, too! :D | 18:49 |
| zara_the_lemur__ | I thought only #zuul was cool enough | 18:50 |
| jeblair | SotK: i think as long as we can physically delete stories, we should do it correctly. so i think we need 394051. if you want to change from deleting to archiving in the future, it could be dropped. though it still would be correct code regardless. | 19:24 |
| jeblair | SotK: (honestly, the data model suggests that the relationships should be constructed that way regardless of what we choose to permit in the app, so i would suggest correcting it (and possibly some similar things in the future) and keeping it) | 19:25 |
| jeblair | SotK: (then just change the rest api server to alter behavior to archiving if that's desired) | 19:26 |
| *** alexismonville has joined #storyboard | 19:29 | |
| *** diablo_rojo has joined #storyboard | 19:36 | |
| SotK | that makes sense, I will review once I've eaten | 19:37 |
| zara_the_lemur__ | oh, apparently this docs patch has been lounging around waiting for merge: https://review.openstack.org/#/c/384177/ | 20:06 |
| openstackgerrit | Merged openstack-infra/storyboard: switch from incubated gettextutils to oslo.i18n https://review.openstack.org/394436 | 20:08 |
| *** jmccrory has quit IRC | 20:24 | |
| *** jmccrory has joined #storyboard | 20:25 | |
| zara_the_lemur__ | ah, the 'add filters for stories/task search' patches are waiting on a reply to persia's comment in review. my understanding is that a user would not be able to construct the query feared, but that its existence does make things more fragile (since the flag might accidentally end up in the wrong place in the codebase in the future) https://review.openstack.org/#/c/389997/ | 20:29 |
| zara_the_lemur__ | I might be wrong about that | 20:29 |
| zara_the_lemur__ | well, the first bit, I'm fairly sure it does make things more fragile | 20:29 |
| zara_the_lemur__ | it also fixes an annoying bug | 20:30 |
| zara_the_lemur__ | also, yay, boartty project merged! :D I'm very behind, sorry; I went on holiday the day this went up. https://review.openstack.org/#/c/391916/ for those following along at home | 20:33 |
| zara_the_lemur__ | it's jeblair's brand new (give or take a week) commandline client for storyboard, a la gertty. http://git.openstack.org/cgit/openstack/boartty/tree/README.rst | 20:35 |
| zara_the_lemur__ | pedroalvarez ^ this is what I mentioned the other week | 20:36 |
| zara_the_lemur__ | just fyi, I'm henceforth nicknaming it 'Boartty McBoardface' | 20:39 |
| SotK | yes, it does make things slightly more fragile, but I will glare angrily at anyone who merges code that passes no_permissions willy-nilly | 21:28 |
| SotK | :D | 21:28 |
| * SotK is comfortable that story_get_simple is mostly an implementation detail and cannot be used by folk with only API access to determine the existence of a story | 21:37 | |
| SotK | persia: I replied to your comment on https://review.openstack.org/#/c/389997/ | 21:43 |
| * SotK will rework to remove the no_permissions parameter that was added by that patch, but the one for story_get_simple is much tidier than trying to pass a user ID all the way through to the function which calls it when updating stories updated_at after the story receives a timeline event | 21:45 | |
| *** alexismonville has quit IRC | 21:53 | |
| *** alexismonville has joined #storyboard | 22:05 | |
| *** zara_the_lemur__ has quit IRC | 22:05 | |
| *** diablo_rojo has quit IRC | 22:08 | |
| SotK | jeblair: that patch solves the issue of the orphaned tasks being counted in the worklist length as displayed on the list page, but because past-SotK is a total idiot the "worklist_items" aren't removed from the worklist and just always ignored | 22:09 |
| SotK | testing this has also shown/reminded me that items that have been deleted mess up the worklist positions | 22:12 |
| SotK | s/positions/item positions/ | 22:12 |
| SotK | perhaps we should be cleaning and renumbering the contents of each worklist every time we make a change to one, before making the change | 22:13 |
| SotK | rather than making deleting a story also mean updating potentially every item in every worklist in the database | 22:14 |
| * SotK disappears to sleep | 22:14 | |
| *** persia_ has quit IRC | 22:46 | |
| *** persia_ has joined #storyboard | 22:48 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!