| *** dave-mccowan has joined #openstack-barbican | 00:00 | |
| *** lee_calcote has joined #openstack-barbican | 00:04 | |
| *** lee_calcote has quit IRC | 00:14 | |
| *** xaeth_afk is now known as xaeth | 00:20 | |
| *** everjeje has quit IRC | 00:32 | |
| *** dimtruck is now known as zz_dimtruck | 00:48 | |
| *** zz_dimtruck is now known as dimtruck | 00:49 | |
| *** dimtruck is now known as zz_dimtruck | 01:04 | |
| *** lee_calcote has joined #openstack-barbican | 01:15 | |
| *** jmckind has joined #openstack-barbican | 01:17 | |
| *** jmckind has quit IRC | 01:17 | |
| *** lee_calcote has quit IRC | 01:20 | |
| *** xaeth is now known as xaeth_afk | 01:37 | |
| *** chlong has joined #openstack-barbican | 01:47 | |
| *** vivek-ebay has joined #openstack-barbican | 01:49 | |
| *** chlong_ has joined #openstack-barbican | 01:53 | |
| *** vivek-ebay has quit IRC | 02:11 | |
| *** vivek-ebay has joined #openstack-barbican | 02:13 | |
| *** zz_dimtruck is now known as dimtruck | 02:19 | |
| *** xaeth_afk is now known as xaeth | 02:24 | |
| *** david-lyle has joined #openstack-barbican | 02:34 | |
| *** chlong_ has quit IRC | 02:50 | |
| openstackgerrit | Ade Lee proposed openstack/barbican: Add functional test for project CA https://review.openstack.org/222583 | 02:52 |
|---|---|---|
| openstackgerrit | Ade Lee proposed openstack/barbican: Fix ca related controllers https://review.openstack.org/224126 | 02:54 |
| *** edtubill has joined #openstack-barbican | 03:26 | |
| *** dave-mccowan has quit IRC | 03:33 | |
| *** chlong has quit IRC | 04:16 | |
| *** chlong has joined #openstack-barbican | 04:33 | |
| *** chlong has quit IRC | 04:48 | |
| *** xaeth is now known as xaeth_afk | 05:03 | |
| *** dimtruck is now known as zz_dimtruck | 05:03 | |
| *** vivek-ebay has quit IRC | 05:13 | |
| *** edtubill has quit IRC | 05:18 | |
| *** jaosorior has joined #openstack-barbican | 05:54 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/barbican: Imported Translations from Zanata https://review.openstack.org/224227 | 06:14 |
| *** chlong has joined #openstack-barbican | 06:18 | |
| *** jaosorior has quit IRC | 07:04 | |
| *** lee_calcote has joined #openstack-barbican | 07:09 | |
| *** lee_calcote has quit IRC | 07:14 | |
| *** jaosorior has joined #openstack-barbican | 07:20 | |
| *** shohel has joined #openstack-barbican | 07:39 | |
| openstackgerrit | Elvin Tubillara proposed openstack/python-barbicanclient: barbican help needs authentication https://review.openstack.org/224467 | 07:40 |
| openstackgerrit | Elvin Tubillara proposed openstack/python-barbicanclient: barbican help needs authentication https://review.openstack.org/224467 | 07:44 |
| -openstackstatus- NOTICE: Gate is currently stuck, failing grenade upgrade tests due the release of oslo.utils 1.4.1 for Juno. | 08:00 | |
| *** jaosorior has quit IRC | 08:23 | |
| *** jaosorior has joined #openstack-barbican | 08:24 | |
| *** darrenmoffat has quit IRC | 08:39 | |
| *** darrenmoffat has joined #openstack-barbican | 08:40 | |
| *** jaosorior has quit IRC | 09:05 | |
| *** jaosorior has joined #openstack-barbican | 09:09 | |
| -openstackstatus- NOTICE: Gate back to normal, thanks to the backlisting of the problematic version | 10:14 | |
| *** everjeje has joined #openstack-barbican | 10:22 | |
| *** shohel has quit IRC | 11:01 | |
| *** shohel has joined #openstack-barbican | 11:01 | |
| *** dave-mccowan has joined #openstack-barbican | 11:32 | |
| *** shohel1 has joined #openstack-barbican | 12:03 | |
| *** shohel has quit IRC | 12:03 | |
| dave-mccowan | jaosorior ping | 12:10 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/barbican: Updated from global requirements https://review.openstack.org/224598 | 12:10 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/castellan: Updated from global requirements https://review.openstack.org/224599 | 12:10 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-barbicanclient: Updated from global requirements https://review.openstack.org/217193 | 12:16 |
| jaosorior | dave-mccowan: pong | 12:21 |
| dave-mccowan | jaosorior are you working on subca stuff today? i just opened two bugs on it, if you want to grab one. | 12:22 |
| jaosorior | dave-mccowan: No, this week I'm working on other stuff :/ So I won't be able to do Barbican work, except for reviews, those I can always do | 12:22 |
| jaosorior | which bugs did you open? | 12:22 |
| dave-mccowan | https://bugs.launchpad.net/barbican/+bug/1496821 | 12:23 |
| openstack | Launchpad bug 1496821 in Barbican "HTTP 403 Returned when trying to get preferred CA" [Undecided,New] | 12:23 |
| dave-mccowan | https://bugs.launchpad.net/barbican/+bug/1496819 | 12:23 |
| openstack | Launchpad bug 1496819 in Barbican "HTTP Code 500 Returned on bad CA resource in URL" [Undecided,New] | 12:23 |
| *** alee has quit IRC | 12:24 | |
| *** jaosorior has quit IRC | 12:31 | |
| *** shohel1 has quit IRC | 13:02 | |
| *** shohel has joined #openstack-barbican | 13:03 | |
| *** peter-hamilton has joined #openstack-barbican | 13:10 | |
| *** lee_calc_ has joined #openstack-barbican | 13:24 | |
| *** alee has joined #openstack-barbican | 13:25 | |
| *** openstackgerrit has quit IRC | 13:31 | |
| *** openstackgerrit has joined #openstack-barbican | 13:31 | |
| *** tristanC has left #openstack-barbican | 13:32 | |
| *** shohel1 has joined #openstack-barbican | 13:33 | |
| *** shohel has quit IRC | 13:33 | |
| dave-mccowan | alee ping | 13:42 |
| alee | dave-mccowan, pong | 13:42 |
| alee | dave-mccowan, morning | 13:42 |
| *** rellerreller has joined #openstack-barbican | 13:42 | |
| dave-mccowan | alee good morning | 13:42 |
| dave-mccowan | alee did you see the bugs on cas controller i opened this morning? | 13:43 |
| alee | dave-mccowan, no .. where are they? | 13:43 |
| dave-mccowan | https://bugs.launchpad.net/barbican/+bug/1496821 | 13:43 |
| openstack | Launchpad bug 1496821 in Barbican "HTTP 403 Returned when trying to get preferred CA" [Undecided,New] | 13:43 |
| dave-mccowan | https://bugs.launchpad.net/barbican/+bug/1496819 | 13:43 |
| openstack | Launchpad bug 1496819 in Barbican "HTTP Code 500 Returned on bad CA resource in URL" [Undecided,New] | 13:43 |
| alee | dave-mccowan, for the first one, my guess is that there is a decorator missing for content types or similar | 13:45 |
| alee | dave-mccowan, and will have to look into the second one. | 13:46 |
| alee | unless you happen to have time to look into it | 13:46 |
| dave-mccowan | alee there is some ambiguity between /{CA_ID}/ and other keyword resource names. i think that may be contributing to the weirdness. | 13:46 |
| dave-mccowan | alee i still have some work to do on the refresh_table at init. we should divide and conquer. | 13:47 |
| alee | dave-mccowan, yes -- I'm still working on getting jaosoriors patch to pass the gate. | 13:48 |
| alee | dave-mccowan, https://review.openstack.org/#/c/222583/ | 13:49 |
| alee | dave-mccowan, its a strange failure -- trying to track it down | 13:49 |
| alee | once I do, you can rebase your change | 13:49 |
| alee | dave-mccowan, and there is jaosorior's next patch to restrict the get_cas -- which I have rewritten somewhat | 13:50 |
| dave-mccowan | alee there is something weird with the CA plugins. sometimes, i do a GET /v1/cas twice in a row. the first time I see two (simple and snake), the second time comes back empty list. any idea what might cause a disappearing act like that? | 13:50 |
| *** su_zhang has joined #openstack-barbican | 13:50 | |
| alee | dave-mccowan, like right after each other? | 13:51 |
| redrobot | crap. looks like I missed the deadline for ptl candidacy | 13:51 |
| redrobot | :( | 13:51 |
| alee | dave-mccowan, or within a couple days of each other? | 13:51 |
| *** rellerreller has quit IRC | 13:51 | |
| dave-mccowan | alee immediately after. without an order posted, just refresh_table() called. | 13:52 |
| alee | dave-mccowan, so the CA plugins have the concept of expiration. that is the plugin tells the server when the information becomes stale. But I think thats set to something like 2 days by default. | 13:53 |
| alee | dave-mccowan, so yeah , I've never seen it disappear like that | 13:53 |
| dave-mccowan | redrobot time time buy presents for the technical committee | 13:53 |
| dave-mccowan | alee sounds like refresh table needs to be called before every CAs operation, otherwise it's a crapshoot if the information is current. | 13:54 |
| *** rellerreller has joined #openstack-barbican | 13:54 | |
| dave-mccowan | alee it might only happen with my refresh at init patch | 13:55 |
| dave-mccowan | alee is there anything about placing a cert order that might make a plugin more permanent? | 13:57 |
| *** spotz_zzz is now known as spotz | 14:00 | |
| dave-mccowan | alee the failure in the gate looks like a parallel running problem. the test is checking that a specific number of cas is in the list. but, with multiple tests running at the same there is no predicting how many will be in the list. i had to fix this for quotas. the fix is to added tests like that to a separate class, then update functionaltests/run_tests.sh to skip them during the parallel run. | 14:01 |
| alee | dave-mccowan, yeah - I was wondering whether the tests were running in parallel | 14:02 |
| alee | how do you skip during the parallel run? | 14:02 |
| dave-mccowan | alee https://github.com/openstack/barbican/blob/master/functionaltests/run_tests.sh#L32 | 14:04 |
| dave-mccowan | alee the plan is to do it with an [attribute], but for now regex is the way | 14:05 |
| *** xaeth_afk is now known as xaeth | 14:07 | |
| *** xaeth has left #openstack-barbican | 14:08 | |
| *** su_zhang has quit IRC | 14:09 | |
| alee | dave-mccowan, thanks -- ok - let me get that fixed .. | 14:25 |
| *** lisaclark1 has joined #openstack-barbican | 14:25 | |
| *** shohel1 has quit IRC | 14:26 | |
| *** pglass has joined #openstack-barbican | 14:29 | |
| dave-mccowan | alee there's more weirdness about calling refresh_table() to start. some unit tests are failing with that change. one expecting 0 cas, but getting two. another expecting 11 and getting 13. | 14:31 |
| alee | dave-mccowan, right - you're going to need to revise the tests | 14:31 |
| alee | to account for the assumption that now there are cas defiend. | 14:32 |
| dave-mccowan | alee ah... ok. that makes sense, we should always start with the number of plugins configured. | 14:33 |
| *** xaeth has joined #openstack-barbican | 14:34 | |
| dave-mccowan | alee which leads to another question.... i don't think we should have snakeoil enabled in the default config file. we'll need to update test cases for that too. | 14:34 |
| *** lee_calc_ has quit IRC | 14:35 | |
| alee | dave-mccowan, why not have it configured? | 14:35 |
| alee | dave-mccowan, it needs to be there for sure for the functional tests | 14:36 |
| dave-mccowan | alee hmmmm. we're supposed to discourage its use, since it is "snakeoil". i guess i'd feel better if someone had to purposefully turn it on. but, thinking about it I guess all the other defaults are insecure too (http, static encryption key, etc.). maybe it is ok. | 14:39 |
| alee | dave-mccowan, well we would never expect anyone to use the simple ca plugin either | 14:40 |
| alee | dave-mccowan, think of these as two testing plugins | 14:40 |
| *** lisaclark1 has quit IRC | 14:41 | |
| *** lisaclark1 has joined #openstack-barbican | 14:41 | |
| dave-mccowan | alee i do. i'm just not sure what the barbican adopter expects out of the default config. i guess something that works out of the box is good, even if it is not secure. | 14:41 |
| alee | dave-mccowan, yup | 14:43 |
| alee | dave-mccowan, at this point, the only secure ca plugin we have is dogtag | 14:44 |
| dave-mccowan | alee ok. i'm convinced. | 14:45 |
| dave-mccowan | alee did you know that the gate (with only committed patches) is failing due to the parallel testing of cas? | 14:46 |
| alee | dave-mccowan, sigh .. link? | 14:47 |
| dave-mccowan | https://review.openstack.org/223219 | 14:47 |
| *** jorge_munoz has joined #openstack-barbican | 14:47 | |
| dave-mccowan | alee you just need to get your patch with the change to run_tests.sh up next. | 14:47 |
| *** edtubill has joined #openstack-barbican | 14:48 | |
| *** lee_calcote has joined #openstack-barbican | 14:49 | |
| *** silos has joined #openstack-barbican | 14:49 | |
| dave-mccowan | alee what do you think about call refresh_table before every CA operation, to make sure we never fail due to expiration? | 14:54 |
| alee | dave-mccowan, well the reason I did not do that was that I was worried about the overhead of making a call to the ca plugin and hence to the ca each time we did a ca operation | 14:56 |
| alee | for many ca plugins that would not be a big deal because the data does not change | 14:56 |
| alee | for dogtag, that could be crippling | 14:56 |
| alee | unless we do caching in the plugin | 14:57 |
| dave-mccowan | alee when the plugin expires, what happens? the system is broken. | 14:58 |
| alee | dave-mccowan, well no - actually .. | 14:59 |
| alee | dave-mccowan, if I recall correctly, each time we request a cert - we call refresh_cas() | 15:00 |
| alee | so if a plugin has expired, it will be queried | 15:00 |
| alee | if not expired , it wont | 15:01 |
| alee | dave-mccowan, question is whether to do this for any ca operation .. | 15:01 |
| *** kebray has joined #openstack-barbican | 15:02 | |
| *** everjeje has quit IRC | 15:03 | |
| dave-mccowan | alee if i am an orchestration script, and i need to order a cert, the first thing i need to get is get a list of CAs to find the ID of the one I want to use. | 15:03 |
| *** zz_dimtruck is now known as dimtruck | 15:03 | |
| alee | dave-mccowan, let me revisit what refresh does .. | 15:04 |
| alee | dave-mccowan, just a sec -- being pinged for escalation of the day - back in a sec. | 15:05 |
| *** lee_calcote has quit IRC | 15:19 | |
| *** kfarr has joined #openstack-barbican | 15:19 | |
| *** diazjf has joined #openstack-barbican | 15:20 | |
| *** ccneill has joined #openstack-barbican | 15:26 | |
| *** dave-mccowan has quit IRC | 15:35 | |
| *** vivek-ebay has joined #openstack-barbican | 15:39 | |
| openstackgerrit | Ade Lee proposed openstack/barbican: Add functional test for project CA https://review.openstack.org/222583 | 15:52 |
| peter-hamilton | redrobot: whaaaaaaaaat? | 15:52 |
| redrobot | peter-hamilton hehe, yeah... bad UTC conversion on my part | 15:53 |
| * peter-hamilton sagely nods his head | 15:53 | |
| *** rellerreller has quit IRC | 15:54 | |
| peter-hamilton | redrobot: what dark magics are required to get you back in? | 15:54 |
| redrobot | peter-hamilton Dave recommended buying the TC presents. :) | 15:55 |
| redrobot | peter-hamilton hoping the TC will appoint me, but we could possibly have a surprising new PTL | 15:56 |
| peter-hamilton | redrobot: haha, worth a shot | 15:56 |
| peter-hamilton | redrobot: are there any other candidates? | 15:56 |
| peter-hamilton | redrobot: o.o | 15:56 |
| redrobot | peter-hamilton no, we're in the TC appoitment process because there were no candidates at all | 15:56 |
| peter-hamilton | redrobot: ah, ok | 15:56 |
| *** kfarr has quit IRC | 15:56 | |
| peter-hamilton | redrobot: well good luck! can we hold a write-in election? that'd be fun | 15:57 |
| redrobot | peter-hamilton haha, thanks | 15:59 |
| rm_work | hey0! | 16:00 |
| rm_work | please review because i did a dumb: https://review.openstack.org/#/c/224220/ | 16:00 |
| redrobot | rm_work looking | 16:01 |
| redrobot | rm_work does that mean that anyone who uses the barbican plugin will get python-barbicanclien trunk? | 16:01 |
| *** kebray has quit IRC | 16:01 | |
| rm_work | yes | 16:02 |
| rm_work | well, in devstack | 16:02 |
| *** lisaclark1 has quit IRC | 16:02 | |
| rm_work | none of this applies if you don't install via the devstack mechanism | 16:02 |
| rm_work | the reason the barbican clone wasn't required is that the gate checks out whatever project is being tested already -- and obviously if the script was being run, it was already cloned | 16:03 |
| rm_work | but unless you've linked a client change to it as a dependency, it won't clone that | 16:03 |
| *** diazjf has quit IRC | 16:03 | |
| rm_work | actually I am not sure how this would effect doing a Depends-On: from a client CR | 16:04 |
| *** diazjf has joined #openstack-barbican | 16:04 | |
| rm_work | but, it's required to un-break OTHER project-gates that use barbican | 16:04 |
| rm_work | and it puts us back to parity with the other devstack script | 16:04 |
| redrobot | rm_work +2ed | 16:04 |
| rm_work | so | 16:04 |
| rm_work | thx | 16:04 |
| *** woodster_ has joined #openstack-barbican | 16:04 | |
| diazjf | redrobot, are there any problems with the gate, some of my minor edits are failing. Should I rebase https://review.openstack.org/#/c/223219/ | 16:09 |
| alee | diazjf, its an issue with the parallel tests in one of the ca tests | 16:12 |
| alee | diazjf, I have put up a patch to fix | 16:12 |
| alee | waiting for it to clear gate testing | 16:12 |
| diazjf | awesome, thanks alee. Do I run reverify or recheck after it merges? | 16:13 |
| alee | yeah recheck | 16:13 |
| alee | redrobot, once it clears gate checks, I'll be looking for +2's .. | 16:14 |
| alee | redrobot, https://review.openstack.org/#/c/222583/ | 16:15 |
| diazjf | alee, awesome!! I'll give it a review soon as well. Also I'm gonna need to rebase my stuff as well right? | 16:16 |
| alee | yeah - if you rebase it will probably kick off tests again | 16:17 |
| alee | diazjf, wait till it merges of course. | 16:17 |
| diazjf | alee, thanks! | 16:19 |
| *** vivek-ebay has quit IRC | 16:20 | |
| edtubill | hi, can someone review my patch https://review.openstack.org/#/c/222328/ | 16:21 |
| *** kebray has joined #openstack-barbican | 16:21 | |
| *** lee_calcote has joined #openstack-barbican | 16:24 | |
| *** lee_calcote has quit IRC | 16:28 | |
| *** vivek-ebay has joined #openstack-barbican | 16:33 | |
| *** david-lyle has quit IRC | 16:36 | |
| *** silos has left #openstack-barbican | 16:42 | |
| *** peter-hamilton has quit IRC | 16:43 | |
| *** vivek-ebay has quit IRC | 16:43 | |
| *** david-lyle has joined #openstack-barbican | 16:47 | |
| *** edtubill has quit IRC | 16:55 | |
| *** ccneill has quit IRC | 16:55 | |
| *** alee is now known as alee_lunch | 16:57 | |
| *** lee_calcote has joined #openstack-barbican | 16:57 | |
| *** su_zhang has joined #openstack-barbican | 16:57 | |
| *** su_zhang has quit IRC | 16:58 | |
| *** su_zhang has joined #openstack-barbican | 16:58 | |
| *** dave-mccowan has joined #openstack-barbican | 17:02 | |
| *** david-lyle has quit IRC | 17:03 | |
| *** vivek-ebay has joined #openstack-barbican | 17:15 | |
| *** lee_calcote has quit IRC | 17:23 | |
| *** lee_calcote has joined #openstack-barbican | 17:28 | |
| *** diazjf has quit IRC | 17:31 | |
| *** mragupat has joined #openstack-barbican | 17:37 | |
| *** rm_work has quit IRC | 17:43 | |
| *** rm_work has joined #openstack-barbican | 17:45 | |
| *** su_zhang has quit IRC | 17:50 | |
| *** su_zhang has joined #openstack-barbican | 17:50 | |
| *** lee_calcote has quit IRC | 17:50 | |
| *** diazjf has joined #openstack-barbican | 18:02 | |
| *** edtubill has joined #openstack-barbican | 18:02 | |
| *** su_zhang_ has joined #openstack-barbican | 18:02 | |
| *** ccneill has joined #openstack-barbican | 18:04 | |
| *** jamielennox is now known as jamielennox|away | 18:05 | |
| *** su_zhang has quit IRC | 18:05 | |
| *** rellerreller has joined #openstack-barbican | 18:06 | |
| *** kfarr has joined #openstack-barbican | 18:13 | |
| *** lee_calcote has joined #openstack-barbican | 18:14 | |
| *** everjeje has joined #openstack-barbican | 18:22 | |
| *** lee_calcote has quit IRC | 18:28 | |
| *** su_zhang_ has quit IRC | 18:35 | |
| *** alee_lunch is now known as alee | 18:37 | |
| alee | redrobot, something wrong with the gate -- been waiting for 3 hours for my CR to clear | 18:38 |
| alee | https://review.openstack.org/#/c/222583 | 18:38 |
| alee | redrobot, wheres the site that shows the status of gate jobs? | 18:40 |
| alee | dave-mccowan, ^^ | 18:40 |
| dave-mccowan | alee i think it's just a busy day. everyone submitting last minute fixes for Release Candidates. http://status.openstack.org/zuul/ | 18:40 |
| *** diazjf has quit IRC | 18:41 | |
| dave-mccowan | alee any new thoughts on how to keep GET /v1/cas working without swamping the plugins? | 18:42 |
| *** jaosorior has joined #openstack-barbican | 18:43 | |
| alee | dave-mccowan, sorry - let me look again at the refresh_cas() | 18:43 |
| jaosorior | alee, dave-mccowan: Anything I should be reviewing? | 18:44 |
| alee | jaosorior, just waiting on the damn gate .. | 18:45 |
| alee | https://review.openstack.org/#/c/222583/ | 18:45 |
| jaosorior | alee: haha well, it's gate rush hour | 18:45 |
| alee | once it passes, then would be good to get in +2 from you and dave-mccowan | 18:45 |
| alee | jaosorior, I'm reworking your follow on patch | 18:46 |
| alee | it does not do exactly what is needed | 18:46 |
| jaosorior | alee: oh, sorry about that | 18:47 |
| alee | dave-mccowan, so refresh_cas() basically checks each plugin and sees if there are any cas present | 18:47 |
| alee | jaosorior, no worries -- its close | 18:47 |
| alee | dave-mccowan, for that plugin | 18:47 |
| alee | with the idea being that all cas will expire at the same time | 18:47 |
| alee | for a given plugin | 18:48 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/barbican: Fix ca related controllers https://review.openstack.org/224126 | 18:48 |
| alee | jaosorior, eh? | 18:48 |
| alee | dave-mccowan, so point is - that for the most part this is a no-op for the plugins | 18:48 |
| jaosorior | alee: rebased | 18:49 |
| alee | dave-mccowan, so it should be ok to refresh the table | 18:49 |
| alee | on GET /cas | 18:49 |
| dave-mccowan | alee cool. i think that would be a good solution. | 18:49 |
| alee | because this will likely not do anything most of the time for any given plugin | 18:49 |
| jaosorior | alee: Also remember to add the Co-Authored-By: in git. If you're doing work there it should be reflected ;) | 18:49 |
| alee | jaosorior, will do :) | 18:50 |
| dave-mccowan | alee to make sure i understand, is there a database entry for each CA, or is the only record stored in the plugin? | 18:50 |
| alee | dave-mccowan, there is a db entry for each ca | 18:50 |
| alee | CertificateAuthoritites | 18:51 |
| alee | the plugin/backend keeps its own records | 18:51 |
| jaosorior | clear | 18:51 |
| jaosorior | wrong terminal | 18:51 |
| alee | ok -- now to try and get jaosorior patch working .. | 18:52 |
| dave-mccowan | alee so, i'm really confused why GET /v1/cas can return an empty list. there must be a subtle bug, in addition to the missing "init" | 18:52 |
| alee | dave-mccowan, yeah -- I dont know how that can happen | 18:55 |
| alee | dave-mccowan, the refresh should fix that though - hopefully | 18:56 |
| kfarr | alee, this is the site btw http://status.openstack.org/zuul/ | 19:02 |
| alee | kfarr, got it - thanks | 19:02 |
| *** lee_calcote has joined #openstack-barbican | 19:05 | |
| *** david-lyle has joined #openstack-barbican | 19:11 | |
| *** silos has joined #openstack-barbican | 19:11 | |
| *** rellerreller has quit IRC | 19:14 | |
| *** kfarr has quit IRC | 19:14 | |
| *** lee_calcote has quit IRC | 19:15 | |
| *** lisaclark1 has joined #openstack-barbican | 19:16 | |
| silos | dave-mccowan: ping | 19:19 |
| *** diazjf has joined #openstack-barbican | 19:19 | |
| *** spotz is now known as spotz_zzz | 19:24 | |
| dave-mccowan | silos pong | 19:24 |
| *** david-lyle has quit IRC | 19:25 | |
| silos | dave-mccowan: I know you're really busy today so I'll keep it short. If I'm a user and do a get for my effective quotas, I should get the values in the barbican.conf file if the service-admin user hasn't set them yet correct? | 19:25 |
| dave-mccowan | silos yes | 19:26 |
| * dave-mccowan crossing fingers that's what you're seeing :-) | 19:26 | |
| * silos becomes the bearer of bad news | 19:26 | |
| silos | I'm not seeing those values :( | 19:27 |
| dave-mccowan | what values are you seeing? | 19:27 |
| silos | -1 all the time | 19:27 |
| silos | Do the values need to be set before installing barbican?? | 19:28 |
| dave-mccowan | silos barbican.sh install makes a copy of barbican.conf in /etc/barbican and that's the file that is used. so... edit them in /etc/barbican, or edit them in-tree before installing | 19:29 |
| silos | dave-mccowan: ok. I'll edit it in tree and see what happens. thx | 19:29 |
| dave-mccowan | silos thanks for testing that. there is a chance there is a bug there; the functional tests don't cover changing config files. | 19:31 |
| silos | dave-mccowan: no prob. I'll let you know what happens. | 19:32 |
| *** spotz_zzz is now known as spotz | 19:33 | |
| *** lee_calcote has joined #openstack-barbican | 19:46 | |
| *** kebray has quit IRC | 19:48 | |
| *** spotz is now known as spotz_zzz | 19:51 | |
| silos | dave-mccowan: I installed barbican from scratch twice. Once with keystone middleware enabled and one with it not turned on. And both times I tested a GET request for quotas for a user with admin privileges to their project and both times it gave me -1 for all values :'( | 19:52 |
| dave-mccowan | silos hmm. i'll give it a try too. | 19:54 |
| rm_work | hmmm | 19:54 |
| rm_work | api.v1.functional.test_cas.CertificateAuthoritiesTestCase.test_list_and_get_cas <-- failing on my change in gate | 19:54 |
| edtubill | dave-mccowan: I tried it too and I got a -1 | 19:55 |
| rm_work | can't POSSIBLY be related <_< | 19:55 |
| dave-mccowan | rm_work alee is fixing that. that test works in serial mode, then fails in parallel. | 19:56 |
| rm_work | kk | 19:56 |
| rm_work | i feel like every time i turn around, something else has broken one of my changes in gate today/yesterday T_T | 19:56 |
| diazjf | rm_work its gonna fail until alee, gets her patch merged | 19:56 |
| rm_work | the oslo.utils thing hit me pretty hard yesterday | 19:56 |
| diazjf | his* | 19:56 |
| rm_work | heh | 19:57 |
| rm_work | yeah k | 19:57 |
| rm_work | is a review up? | 19:57 |
| diazjf | yeah, let me pull it up | 19:57 |
| *** su_zhang has joined #openstack-barbican | 19:57 | |
| diazjf | https://review.openstack.org/#/c/222583/ | 19:57 |
| dave-mccowan | rm_work it should just skip that test when using testr in parallel mode, so you know it can't fail. :-) | 19:58 |
| *** lee_calcote has quit IRC | 19:58 | |
| dave-mccowan | silos yep, fails for me too. it should be an easy fit in barbican/common/quota.py, where the config parameters are defined, but the problem isn't jumping out at me. | 20:00 |
| dave-mccowan | silos it seems like it is always taking the defaults, instead of reading the config file. | 20:01 |
| *** su_zhang has quit IRC | 20:02 | |
| silos | dave-mccowan: Hmmm It should read the config file every time bin/barbican.sh start is run. So idk why it's not doing it for those values. | 20:02 |
| silos | dave-mccowan: I'll file a bug and take a look into it. If I come up with the solution I'll assign it to myself. But you know the code better than me so you'll probably find the solution faster. | 20:04 |
| dave-mccowan | silos maybe that config code needs to be moved to (or called from) common/config.py. | 20:04 |
| dave-mccowan | jaosorior, alee ^^ do all config parameters need to be registered under new_config() | 20:05 |
| *** jaosorior has quit IRC | 20:05 | |
| dave-mccowan | silos yea, i think that's it. if you cut and paste out that register config code, and make it fit in config.py, consistent with the other groups, I think you will fix the bug. | 20:08 |
| silos | dave-mccowan: okay. I'll try that. | 20:09 |
| *** kebray has joined #openstack-barbican | 20:13 | |
| *** lee_calcote has joined #openstack-barbican | 20:17 | |
| *** diazjf1 has joined #openstack-barbican | 20:19 | |
| *** diazjf has quit IRC | 20:19 | |
| *** spotz_zzz is now known as spotz | 20:22 | |
| *** vivek-ebay has quit IRC | 20:28 | |
| *** mragupat has quit IRC | 20:29 | |
| *** gyee has joined #openstack-barbican | 20:31 | |
| *** lisaclark1 has quit IRC | 20:41 | |
| *** su_zhang has joined #openstack-barbican | 20:46 | |
| silos | dave-mccowan: that was it :-D. Some tests are failing so those need to be refactored. | 20:50 |
| dave-mccowan | silos cool. as long as the tests pass when the config is reverted to default, then i think you're ok. | 20:51 |
| silos | ah ok. I'll test that to be sure then. thanks. | 20:52 |
| alee | dave-mccowan, have you had to do any controller unit tests where you pass in the external project id? | 20:52 |
| alee | just trying to figure out how to do it .. | 20:53 |
| dave-mccowan | alee yea, i think it have. let me see if i can find an example. | 20:55 |
| dave-mccowan | alee in test_models.py, WhenCreatingNewProjectQuotas() I create a project entry and point to it. | 20:57 |
| alee | dave-mccowan, yeah - thats at the model level .. | 20:57 |
| alee | dave-mccowan, I'm thinking about the api level -- ie controllere | 20:57 |
| alee | controllers | 20:58 |
| dave-mccowan | alee in tests/api/controllers/test_quotas.py, I was able to fake it. the in-memory DB used for unit test doesn't enforce foreign keys. | 20:58 |
| alee | yeah - but how do you pass in the external_project_id in the request? | 20:59 |
| alee | so for instance .. we end up doing something like this ... | 20:59 |
| alee | resp = self.app.get('/cas/', self.params) | 20:59 |
| *** kebray has quit IRC | 21:00 | |
| alee | what is external_project_id set to in this case -- None? | 21:00 |
| *** kebray has joined #openstack-barbican | 21:00 | |
| alee | and if so, how can you change it? | 21:00 |
| rm_work | psst lisaclark | 21:02 |
| *** lee_calcote has quit IRC | 21:02 | |
| alee | dave-mccowan, is it maybe in the extra_environ ? | 21:04 |
| *** kfarr has joined #openstack-barbican | 21:04 | |
| dave-mccowan | alee ah... yea that looks like it. test_acls.py uses that and sets project_id and user_id. | 21:05 |
| alee | dave-mccowan, looking | 21:05 |
| alee | ah coool | 21:06 |
| *** david-lyle has joined #openstack-barbican | 21:06 | |
| *** david-lyle has quit IRC | 21:09 | |
| openstackgerrit | Christopher Solis proposed openstack/barbican: Updates quota values to be read from conf file https://review.openstack.org/224903 | 21:21 |
| *** jamielennox|away is now known as jamielennox | 21:24 | |
| *** vivek-ebay has joined #openstack-barbican | 21:28 | |
| *** silos has left #openstack-barbican | 21:28 | |
| *** diazjf1 has left #openstack-barbican | 21:37 | |
| *** dave-mccowan has quit IRC | 21:39 | |
| *** vivek-ebay has quit IRC | 21:41 | |
| *** edtubill has quit IRC | 21:54 | |
| *** xaeth is now known as xaeth_afk | 21:57 | |
| *** kfarr has quit IRC | 22:02 | |
| *** kebray has quit IRC | 22:04 | |
| *** dave-mccowan has joined #openstack-barbican | 22:08 | |
| *** dimtruck is now known as zz_dimtruck | 22:09 | |
| *** kebray has joined #openstack-barbican | 22:13 | |
| alee | dave-mccowan, ping | 22:15 |
| dave-mccowan | alee pong | 22:16 |
| alee | dave-mccowan, trying to get GET /cas right -- but I think I have the query wrong ... let me post what I have and you can tell me what I need to do instead .. | 22:16 |
| alee | dave-mccowan, http://www.fpaste.org/268694/25282711/ | 22:18 |
| alee | dave-mccowan, the part in lines 18 - 26 is wrong | 22:19 |
| alee | dave-mccowan, specifically what I want is select cas where ca.project_id=project_id OR join(pca) where pca.project_id=project_id .. | 22:21 |
| alee | doe sthat make sense? | 22:21 |
| dave-mccowan | alee i think so. quick check: are you sure all project_id's are internal project id's? | 22:24 |
| alee | yup | 22:24 |
| dave-mccowan | i'm not sure. i need to step away for a couple hours. back later. | 22:26 |
| alee | dave-mccowan, I think I just need a union of two queries | 22:26 |
| alee | yeah me too | 22:26 |
| alee | dave-mccowan, wonder if the change will get through the gate today :/ | 22:27 |
| *** alee has quit IRC | 22:36 | |
| *** spotz is now known as spotz_zzz | 22:37 | |
| *** pglass has quit IRC | 22:40 | |
| *** vivek-ebay has joined #openstack-barbican | 22:41 | |
| *** vivek-ebay has quit IRC | 22:46 | |
| *** kebray has quit IRC | 22:52 | |
| *** david-lyle has joined #openstack-barbican | 22:59 | |
| *** vivek-ebay has joined #openstack-barbican | 23:09 | |
| *** alee has joined #openstack-barbican | 23:29 | |
| dave-mccowan | alee what do you think about doing the join this way? http://paste.openstack.org/show/467024/ | 23:50 |
| *** ccneill has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!