| *** VW has joined #craton | 00:03 | |
| *** VW has quit IRC | 00:05 | |
| *** VW has joined #craton | 00:06 | |
| *** cloudnull has quit IRC | 00:14 | |
| *** _d34dh0r53_ has quit IRC | 00:14 | |
| *** cloudnull has joined #craton | 00:14 | |
| *** d34dh0r53 has joined #craton | 00:15 | |
| *** Syed__ has quit IRC | 00:35 | |
| *** harlowja has joined #craton | 00:42 | |
| *** klindgren has joined #craton | 03:06 | |
| *** harlowja has quit IRC | 03:35 | |
| *** VW has quit IRC | 03:50 | |
| openstackgerrit | Michael Porras proposed openstack/craton master: Docker and dev Documentation update https://review.openstack.org/448248 | 06:08 |
|---|---|---|
| *** Guest48021 has joined #craton | 06:16 | |
| *** Guest48021 has quit IRC | 06:37 | |
| openstackgerrit | Sulochan Acharya proposed openstack/craton master: Create user under project_id passed in request https://review.openstack.org/446972 | 11:33 |
| sigmavirus | morning all | 12:12 |
| sigmavirus | jimbaker: the -1 on that patch from git-harry is two fold: 1. The standardization seems to be a separate issue worthy of a separate bug (although it's already in that review, so not necessarily a separate review). 2. Validation of request arguments can fail on multiple items rather than just one. Having a single message to represent all of that via unstructured text is a disservice to users. We already have an array of errors, we should | 12:14 |
| sigmavirus | use that. Granted each item in the array is unstructured text, but it's better than a single giant message of all the errors at once. I'll comment on the review as well since it seems the comment I left is either not understandable or too obtuse | 12:14 |
| thomasem | o/ | 12:18 |
| thomasem | sigmavirus: was able to consistently re-record and such with that betamax patch. Left a +1 and some thoughts on how we'll want to use this going forward regarding ensuring these run against an actual API somewhat frequently. | 12:19 |
| sigmavirus | thomasem: yeah the other part of the spec is to have a functional gate much like craton's functional gate | 12:20 |
| sigmavirus | This just provides some good assurances that things don't break in a fast and reproducible manner | 12:21 |
| thomasem | Yep | 12:21 |
| sigmavirus | The functional gate being something that we only ever need to run on OpenStack CI | 12:21 |
| thomasem | +1 | 12:21 |
| thomasem | Cool, thanks for the extra context. | 12:22 |
| sigmavirus | betamax is cruise-control for speed =P | 12:23 |
| thomasem | Lol | 12:24 |
| thomasem | Btw, added link to the testing spec in the BP for those patches. | 12:24 |
| sigmavirus | I'm not surprised that failed to happen automagically | 12:29 |
| thomasem | :( Yeah | 12:31 |
| thomasem | Doesn't look like we're participating here: https://specs.openstack.org/openstack/ | 12:32 |
| sigmavirus | those are all the official big-tent projects | 12:33 |
| thomasem | Ohhhoho | 12:33 |
| thomasem | I see | 12:33 |
| sigmavirus | (Yes, fuel and murano somehow got into the tent) | 12:34 |
| thomasem | Lol | 12:34 |
| thomasem | Gotcha | 12:34 |
| sigmavirus | Looks like jimbaker needs to configure https://readthedocs.org/projects/craton/ better so that it we can publish docs for craton and add a project for cratonclient | 12:38 |
| sigmavirus | Also, should probably add other people as owners/maintainers | 12:38 |
| thomasem | https://python-cratonclient.readthedocs.io/en/latest/readme.html | 12:38 |
| thomasem | Gotcha | 12:39 |
| thomasem | chrisspencer is an owner of python-cratonclient's? | 12:40 |
| sigmavirus | thomasem: I'd prefer just to call it cratonclient | 12:40 |
| thomasem | I agree | 12:40 |
| sigmavirus | Let me create that and make jimbaker an owner | 12:40 |
| sigmavirus | thomasem: do you have an RTD account? | 12:43 |
| thomasem | sigmavirus: I'm about to | 12:43 |
| thomasem | sigmavirus: I do now | 12:44 |
| thomasem | thomasem | 12:44 |
| sigmavirus | added to cratonclient.readthedocs.io | 12:45 |
| sigmavirus | https://cratonclient.readthedocs.io/en/latest/ looks pretty darn good | 12:45 |
| sigmavirus | will look much better after my docs patches land | 12:46 |
| thomasem | Yep! | 12:46 |
| thomasem | sigmavirus: thanks for setting that up | 12:46 |
| thomasem | I see the specs there now, too. :) | 12:47 |
| sigmavirus | hth | 12:47 |
| thomasem | https://cratonclient.readthedocs.io/en/latest/specs/approved/testing-plan.html | 12:47 |
| sigmavirus | yarp | 12:47 |
| *** VW has joined #craton | 12:48 | |
| sigmavirus | I'm setting up the infra jobs | 12:48 |
| *** VW has quit IRC | 12:49 | |
| thomasem | :thumbsup: | 12:49 |
| *** VW has joined #craton | 12:49 | |
| sigmavirus | I included the deets in the message in case infra spots something off with the plan: https://review.openstack.org/448554 | 12:52 |
| thomasem | Excellent. Thanks, sigmavirus! | 12:57 |
| sigmavirus | always happy to get weird shit done | 12:59 |
| fsaad | morning all | 13:25 |
| sulo | o/ | 13:45 |
| sigmavirus | morning fsaad | 14:00 |
| openstackgerrit | Merged openstack/craton master: Create user under project_id passed in request https://review.openstack.org/446972 | 14:08 |
| thomasem | Lol sigmavirus. Well it's appreciated. :P | 14:29 |
| thomasem | Also, think I'm seeing race conditions in the betamax stuff | 14:29 |
| thomasem | Especially with the pagination stuff where large amounts of resources are created. I think it runs these test in parallel. | 14:29 |
| openstackgerrit | Thomas Maddox proposed openstack/craton master: JSON Path-like querying for variables https://review.openstack.org/443941 | 14:50 |
| sigmavirus | thomasem: ah, yeah, testr does parallel testing | 15:02 |
| sigmavirus | not sure there's a great way to do that besides switching to a different test runner | 15:02 |
| thomasem | What about using filters to be sure you're only getting the resources related to that test? | 15:03 |
| sigmavirus | so that feels like a poor solution | 15:04 |
| sigmavirus | by which I mean, we should have tests that use filters, yes, but for the tests already written, those are testing important functionality | 15:04 |
| thomasem | So, what do you propose? Different test runner? | 15:05 |
| sigmavirus | I'm frankly not sure | 15:05 |
| sigmavirus | i bet we could adjust the concurrency of testr/testtools | 15:05 |
| thomasem | I'm pretty certain we can, yep. | 15:06 |
| sigmavirus | might just be the better/simpler option | 15:06 |
| sigmavirus | might also just be a good option to use in a separate tox env to record cassettes with? | 15:07 |
| * sigmavirus shrugs | 15:07 | |
| thomasem | tox -e record | 15:07 |
| thomasem | wipes out cassettes and creates new ones in a serial fashion | 15:08 |
| sigmavirus | that could work | 15:08 |
| thomasem | Maybe not in this patch, but perhaps later on we could add specificity, too. So you could pass in the path to the test(s) you wish to run and it'll target those cassettes specifically | 15:10 |
| sigmavirus | so I'm going to write this as a new patch in the chain | 15:11 |
| sigmavirus | because it's separate functionality that makes sense | 15:11 |
| thomasem | Mmmkay | 15:14 |
| thomasem | I'll lift my -1s, then. I think we understand the problem adequately. | 15:14 |
| anonymike | Do we still have daily IRC standups? | 15:31 |
| anonymike | (when we don't have vidyo meetings) | 15:31 |
| thomasem | anonymike: we do not. It's just Monday (IRC #openstack-meeting-4), Tuesday (Vidyo), and Thursday (IRC #craton). | 15:32 |
| thomasem | The times at the top of this etherpad https://etherpad.openstack.org/p/craton-meetings | 15:32 |
| anonymike | ohh cool | 15:32 |
| anonymike | thanks thomasem | 15:32 |
| thomasem | anonymike: you bet. Lemme be sure you've got the invites. | 15:33 |
| openstackgerrit | Ian Cordasco proposed openstack/python-cratonclient master: Add convenience environment to re-record tests https://review.openstack.org/448646 | 15:33 |
| sigmavirus | thomasem: ^ | 15:33 |
| anonymike | thomasem: i got Mon and Tues but not Thurs | 15:33 |
| sigmavirus | it's also easier to test in that position in the change chain because you have cassettes to fidget with | 15:33 |
| thomasem | anonymike: Got it. I don't have Thursday either. | 15:34 |
| sigmavirus | anonymike: you're looking for meetings? | 15:34 |
| anonymike | ha | 15:34 |
| sigmavirus | anonymike: if you want more meetings, PrisonMike can help you ;) | 15:34 |
| anonymike | haha no no, I'm good | 15:34 |
| sigmavirus | (contrary to what will likely be popular belief, that was not a threat) | 15:34 |
| thomasem | sigmavirus: awesome. Thanks for that. Going to pull it down and test it. :) | 15:34 |
| anonymike | sigmavirus: that's from the office correct? | 15:35 |
| sigmavirus | anonymike: absotootly | 15:35 |
| anonymike | ha it's been a while | 15:36 |
| sigmavirus | https://media3.giphy.com/media/Q4r9RDlu5Si0o/giphy-facebook_s.jpg?t=1 | 15:36 |
| sigmavirus | https://i.imgur.com/DiCdy1e.jpg | 15:36 |
| thomasem | https://i.imgur.com/JGYSRXf.gif | 15:37 |
| anonymike | on mondays I change fron anonymike to prisonmike | 15:37 |
| thomasem | Literally my favorite scene. | 15:37 |
| jimbaker | git-harry, sigmavirus, makes sense re error reporting. i'm going to look at that patch again with that perspective, because the current error reporting is quite nice when it is json | 15:39 |
| jimbaker | the problem is when it isn't, and tooling that expects json then gets a response body that isn't | 15:40 |
| thomasem | anonymike: I see. | 15:41 |
| jimbaker | thomasem, i will send that invite to you for thurs | 15:42 |
| thomasem | jimbaker: and anonymike needs it :) Thank you!! | 15:43 |
| jimbaker | anonymike, thomasem, forwarded the meeting invite to you both | 15:43 |
| thomasem | Danke schön | 15:43 |
| anonymike | thanks! | 15:43 |
| thomasem | brb | 15:44 |
| anonymike | same, relocating to a coffee shop | 15:44 |
| sigmavirus | thomasem: I use that jif a lot | 15:45 |
| anonymike | lol you triggered me with that spelling sigmavirus | 15:46 |
| anonymike | you forced me to say jif in my head | 15:46 |
| sigmavirus | jimbaker: I fully understand the bug that git-harry is fixing (in a quite elegant way) I just disagree with the direction of message over errors | 15:46 |
| sigmavirus | anonymike: you suffer legit ptsd from the great gif v gif wars of 2015? | 15:46 |
| anonymike | I do | 15:47 |
| jimbaker | sigmavirus, got it | 15:47 |
| sigmavirus | https://twitter.com/sigmavirus24/status/629275117709852672 | 15:47 |
| sigmavirus | ^ me right now w/r/t gif v gif | 15:47 |
| anonymike | I like it | 15:49 |
| sigmavirus | and yes, I typed jif to make people say that withthe hard g purposefully | 15:49 |
| sigmavirus | i'm /that/ kind of horrible perosn | 15:49 |
| anonymike | You got in my head man | 15:49 |
| anonymike | lol | 15:49 |
| anonymike | ok srsly running to the coffee shop now brb | 15:49 |
| sigmavirus | btw, if y'all could slightly prioritize the stuff i have in flight pre-friday that'd be great | 15:50 |
| jimbaker | sigmavirus, definitely have this in mind! | 15:51 |
| sigmavirus | thanks | 15:53 |
| thomasem | FYI, these already have a +2: https://review.openstack.org/#/c/447084/, https://review.openstack.org/#/c/447103/, https://review.openstack.org/#/c/447110/ | 15:54 |
| thomasem | sigmavirus: is this still relevant https://review.openstack.org/#/c/396678/, looks like it hasn't been updated in 4 months. | 15:55 |
| sigmavirus | fixed =) | 15:57 |
| thomasem | Thanks! | 15:57 |
| sigmavirus | sorry about that | 16:01 |
| sigmavirus | we wanted to add caching client side to make life nicer for users like OSA but then that stalled out and other things became more important | 16:01 |
| sigmavirus | pre-0.1.0 milestone, priorities in craton were a shifting sands that swallowed everyone up | 16:02 |
| thomasem | No problem at all. Just wanted you to have a clean queue if you could by Friday. It's always nice to make a clean break. :D | 16:02 |
| thomasem | That feeling of completion. | 16:03 |
| thomasem | Like when you get the last point of interest in Witcher 3. | 16:03 |
| * sigmavirus hasn't played witcher 3 =P | 16:03 | |
| thomasem | Are you into RPGs much? | 16:03 |
| jimbaker | thomasem, thanks for raising these. also sigmavirus thanks for the cleanup of the caching. still worthwhile pursuing i think | 16:03 |
| jimbaker | but in good time | 16:04 |
| thomasem | sigmavirus: just imagine a huge map covered in hundreds and hundreds of question marks. | 16:04 |
| sigmavirus | lol | 16:04 |
| thomasem | jimbaker: you bet | 16:04 |
| *** harlowja has joined #craton | 16:06 | |
| openstackgerrit | Antony Messerli proposed openstack/craton master: Modify docs to use python3 -m pip instead https://review.openstack.org/448662 | 16:14 |
| thomasem | So, which do we want it to be: {"network-devices": [...]} or {"network_devices": [...]}? Currently GET /v1/network-devices returns the latter, and GET /v1/devices returns the former. I'm imagining it's the former that we actually want? | 16:15 |
| openstackgerrit | Antony Messerli proposed openstack/craton master: Modify docs to use python3 -m pip instead https://review.openstack.org/448662 | 16:16 |
| thomasem | Well, I think the latter adheres to API WG guidelines... lemme go review that. | 16:16 |
| antonym | sigmavirus: ^ updated for your nits | 16:16 |
| thomasem | I'm wrong. "Resource names may include hyphens." | 16:17 |
| thomasem | "Field names should use the snake_case style, not CamelCase or StUdLyCaPs style." | 16:18 |
| thomasem | https://specs.openstack.org/openstack/api-wg/guidelines/naming.html#fields-in-an-api-request-or-response-body | 16:18 |
| anonymike | interesting | 16:19 |
| anonymike | also I didn't know studlycaps was a thing | 16:19 |
| thomasem | Lol | 16:19 |
| thomasem | https://en.wikipedia.org/wiki/Studly_caps | 16:19 |
| sigmavirus | Didn't know that about HoTMaiL | 16:21 |
| thomasem | https://bugs.launchpad.net/craton/+bug/1675085 | 16:26 |
| openstack | Launchpad bug 1675085 in craton "Devices endpoint returns hyphenated field instead of snake_case" [Undecided,New] | 16:26 |
| openstackgerrit | Michael Porras proposed openstack/craton master: Docker and dev Documentation update https://review.openstack.org/448248 | 16:28 |
| openstackgerrit | Thomas Maddox proposed openstack/craton master: JSON Path-like querying for variables https://review.openstack.org/443941 | 16:30 |
| thomasem | https://bugs.launchpad.net/craton/+bug/1675087 | 16:32 |
| openstack | Launchpad bug 1675087 in craton "NetworkDevice create does not return variables in response" [Critical,New] - Assigned to Thomas Maddox (thomas-maddox) | 16:32 |
| openstackgerrit | Thomas Maddox proposed openstack/craton master: JSON Path-like querying for variables https://review.openstack.org/443941 | 16:52 |
| openstackgerrit | Thomas Maddox proposed openstack/craton master: Return variables in response on Network Device create https://review.openstack.org/448678 | 16:52 |
| openstackgerrit | Thomas Maddox proposed openstack/craton master: JSON Path-like querying for variables https://review.openstack.org/443941 | 17:01 |
| thomasem | Sorry for the spam. | 17:01 |
| thomasem | Can we still be friends? | 17:01 |
| anonymike | sure | 17:02 |
| thomasem | Okay, good. :) | 17:02 |
| sigmavirus | thomasem: rather presumptive of you | 17:06 |
| sigmavirus | =P | 17:06 |
| thomasem | sigmavirus: awwwww | 17:06 |
| * sigmavirus kids | 17:06 | |
| thomasem | https://2.bp.blogspot.com/-EnIbBGDNyXQ/UPXf15tzNLI/AAAAAAAAHTE/3MK6E1poP1c/s1600/sad-panda.jpg | 17:07 |
| thomasem | :P | 17:07 |
| jimbaker | antonym, anonymike - re parsing dbsync bootstrap - https://gist.github.com/jimbaker/9089e381af38d9db6bc4af59baed6d47 | 17:22 |
| jimbaker | anonymike, this is the gist that should get moved into the tools directory, just requires some minor extra work, such as removing such as hard coded dependencies in https://gist.github.com/jimbaker/9089e381af38d9db6bc4af59baed6d47#file-craton-wrapper-functions-sh-L78 | 17:24 |
| anonymike | jimbaker: cool I'll get that in | 17:25 |
| jimbaker | anonymike, sounds great if you can take that tooling and put it in there | 17:26 |
| jimbaker | and add corresponding docs | 17:26 |
| anonymike | jimbaker: so how do I track that? would this be considered a bug? | 17:26 |
| jimbaker | anonymike, yeah | 17:26 |
| anonymike | cool | 17:26 |
| jimbaker | once we go to storyboard, it changes anyway to a story, https://storyboard-blog.io/mapping-launchpad-to-storyboard.html | 17:27 |
| openstackgerrit | Merged openstack/craton master: Return variables in response on Network Device create https://review.openstack.org/448678 | 17:27 |
| jimbaker | launchpad technically uses blueprints for features, but they are too big for small ones like this one | 17:27 |
| anonymike | I heard you guys talk about that yesterday, when are we moving to storyboard? | 17:28 |
| jimbaker | we are in line for migration | 17:28 |
| jimbaker | but in a week or so. i'm going to check on our status. just need to write email to the coordinator! | 17:28 |
| jimbaker | (kendall nelson) | 17:28 |
| anonymike | I'll be sure to read up on that | 17:30 |
| jimbaker | looks like monasca migration is complete; and the one in front of us, refstack is in progress | 17:30 |
| jimbaker | see https://storyboard.openstack.org/#!/project/list | 17:30 |
| thomasem | antonym: Why python3 -m pip over pip3? I didn't catch that, it seems. | 17:31 |
| sigmavirus | thomasem: it's the more durable way to use pip | 17:33 |
| sigmavirus | pip3 may go away in the future | 17:33 |
| thomasem | Ah, gotcha. Okey doke! | 17:34 |
| * sigmavirus has his toes in pip development | 17:34 | |
| thomasem | Handy knowledge. :) | 17:34 |
| jimbaker | and important work too! | 17:35 |
| thomasem | Thanks | 17:35 |
| sulo | thomasem: jimbaker: no functional tests for https://review.openstack.org/#/c/448678/ ?? | 17:41 |
| jimbaker | sulo, i suppose we could do that. i know thomasem plans to forklift these tests into functional, so assume that's the plan | 17:43 |
| thomasem | sulo: We can add some. FYI it's getting tested by my nested variable search functional tests. | 17:43 |
| thomasem | That's how I found the bugs in the first place, since it uses the variables from the create response to check when it queries. | 17:44 |
| fsaad | non-UK cratoneers: hope the demo scheduled time works for y'all, I think this time around we'll focus on data loading using zz_pwnall138 's script and he'll drive that but would love for some (or all) to be around if questions come up | 17:44 |
| fsaad | in particular to back up if there are any issues searching with the new code etc. | 17:44 |
| thomasem | I should be able to make it. | 17:45 |
| fsaad | thanks thomasem | 17:45 |
| jimbaker | anonymike, re that gist, definitely try out craton-start-direct() | 17:45 |
| fsaad | hopefully will be a short/sweet demo and we have more things to look forward to. | 17:45 |
| anonymike | jimbaker: okay, i just set it up. I was running some basic stuff I'll give it a shot | 17:45 |
| thomasem | jimbaker: That's going to be the DB unit tests I'm forklifting. sulo: I certainly don't mind adding some functional tests to assert that variables are returned from all of these resources. It's more a side effect of my nested variable search tests, so we could write ones specifically for that. | 17:46 |
| jimbaker | anonymike, i find being able to run everything directly is highly productive | 17:46 |
| jimbaker | vs having to use a docker container etc | 17:46 |
| anonymike | I agree | 17:46 |
| anonymike | i did that for my dev setup | 17:46 |
| anonymike | and was hitting the api with postman :) | 17:47 |
| anonymike | oof bitten by the pip3 stuff | 17:47 |
| jimbaker | anonymike, also re docker container - if you don't know it already - this can be highly useful: $ docker exec -it craton-api mysql -ucraton -pcraton craton | 17:47 |
| jimbaker | etc with docker exec | 17:47 |
| jimbaker | never want the db to be opaque when debugging stuff on it :) | 17:48 |
| anonymike | awesome!! | 17:48 |
| jimbaker | but again, running craton direct, with instant reloads - highly productive when doing backend stuff | 17:48 |
| thomasem | sulo: this test: https://review.openstack.org/#/c/443941/19/craton/tests/functional/test_jsonpath_search.py@136 | 17:49 |
| jimbaker | btw, what happened to log traces with 500s? sulo, git-harry, thomasem - any ideas? | 17:49 |
| jimbaker | log stacktrace | 17:49 |
| jimbaker | is this a config option now? | 17:49 |
| thomasem | jimbaker: I do not know what happened to them, I'm afraid. | 17:49 |
| thomasem | Maybe something along the error handling path is swallowing them and just assembling the JSON response and not logging? | 17:50 |
| jimbaker | thomasem, no worries. just want them back :) we may want them silenced by default (although maybe not) | 17:50 |
| jimbaker | thomasem, that would be my assumption | 17:50 |
| sigmavirus | craton could stand to improve it's logging overall | 17:51 |
| thomasem | +1 | 17:51 |
| jimbaker | we made great strides there, plus git-harry's recent work. but there's a reason for https://docs.python.org/3.5/library/logging.html#logging.exception | 17:51 |
| sigmavirus | jimbaker: agreed | 17:52 |
| sigmavirus | also, i'd urge y'all to look at structlog | 17:52 |
| sigmavirus | makes for parsing logfiles (by operators) much much nicer | 17:52 |
| openstackgerrit | Thomas Maddox proposed openstack/craton master: JSON Path-like querying for variables https://review.openstack.org/443941 | 17:52 |
| sigmavirus | just an aside | 17:52 |
| sigmavirus | =) | 17:52 |
| jimbaker | sigmavirus, always in the know | 17:52 |
| thomasem | interestingly zz_pwnall138 found that this works yesterday: https://review.openstack.org/#/c/443941/20/craton/tests/functional/test_jsonpath_search.py@181 | 17:53 |
| jimbaker | yes. i know how painful badly formatted logs can be. i could tell stories... but not now | 17:53 |
| thomasem | vars="foo.*:zoo" | 17:53 |
| jimbaker | structlog - https://structlog.readthedocs.io/en/stable/ - not in global requirements, but presumably easy to get in | 17:54 |
| sigmavirus | yeah, I wonder if d34dh0r53 has ever looked at structlog before | 17:55 |
| sigmavirus | it's good stuff, can be human readable if that's desirable | 17:55 |
| sigmavirus | also should be easy to shim into standard library logging module (which oslo.log builds atop) | 17:55 |
| jimbaker | got it | 17:56 |
| jimbaker | so a possible contribution into oslo.log, vs something craton would directly do (except maybe config) | 17:56 |
| sigmavirus | so far from urgent or a priority | 17:56 |
| jimbaker | right | 17:56 |
| sigmavirus | just something that might be friendly to operators in the future =) | 17:56 |
| fsaad | thomasem / jimbaker / anonymike / sigmavirus if any want to play witht he box zz_pwnall138 will use during the demo here's the info for it https://gist.github.com/pwnall1337/5f4add7f776cdba3c273dfbb149a9031 | 17:56 |
| jimbaker | and thus highly desirable | 17:56 |
| fsaad | just in case someone asks about how to query a particular thing or something like that | 17:57 |
| fsaad | don't think it'll be necessary since that has been demoed already, but just in case | 17:57 |
| anonymike | thanks fsaad | 17:57 |
| fsaad | yep thanks you | 17:57 |
| thomasem | It sounded like he was having some success with the nested searching? | 17:57 |
| thomasem | Is that true? | 17:57 |
| jimbaker | certainly we try to do this friendliness, especially not just copy & paste from an openstack project, but be thoughtful | 17:57 |
| thomasem | fsaad: ^^ (my question(s)) | 17:58 |
| antonym | fsaad: yeah, interested in playing with the population script, last i tried it i wasn't having much success in my env | 17:58 |
| thomasem | functional tests are growing around that patch, I think we're getting super close here now. | 17:58 |
| jimbaker | thomasem, good to know | 17:59 |
| d34dh0r53 | would definitely +1 structlog in oslo.log, anything to make logs more parseable | 17:59 |
| antonym | sigmavirus: good to know, figured there was a good reason :D | 17:59 |
| thomasem | A few more scenarios I'd like to test for each resource (int, float, bool, boolstr, conjunctive), then resolved tests . | 17:59 |
| fsaad | thomasem: I believe so yes | 17:59 |
| thomasem | fsaad: good! | 17:59 |
| fsaad | antonym: let me get that for you | 18:00 |
| antonym | was it the one in the osic repo? | 18:00 |
| jimbaker | d34dh0r53, cool about structlog | 18:00 |
| thomasem | Alright, I need to step away for some lunch | 18:00 |
| sigmavirus | d34dh0r53: antonym yeah, I never tried to make the case for structlog to openstack because I never had the energy | 18:00 |
| sigmavirus | could probably be hooked into existing openstack projects too in some way, just never thought about how to inject it appropriately | 18:01 |
| jimbaker | biab myself | 18:03 |
| *** harlowja has quit IRC | 18:59 | |
| *** harlowja has joined #craton | 19:00 | |
| *** harlowja has quit IRC | 19:01 | |
| fsaad | reminder to take a break and judge others for a little bit https://goo.gl/OhEVUp | 19:16 |
| fsaad | be gentle | 19:17 |
| jimbaker | fsaad, thanks for posting that link | 19:19 |
| fsaad | sure thing | 19:19 |
| openstackgerrit | Merged openstack/craton master: Modify docs to use python3 -m pip instead https://review.openstack.org/448662 | 19:26 |
| openstackgerrit | Merged openstack/craton master: Docker and dev Documentation update https://review.openstack.org/448248 | 19:35 |
| *** harlowja has joined #craton | 19:45 | |
| anonymike | relocating | 20:01 |
| *** harlowja has quit IRC | 20:18 | |
| jimbaker | so as i just mentioned in a demo we had, we have the ability to write custom formatters. this should include being able to show just a subset of variables; see https://github.com/openstack/python-cratonclient/blob/master/cratonclient/formatters/json_format.py as one example | 20:26 |
| jimbaker | the other relevant piece of data is it would be nice in the future (hopefully near) to demonstrate elasticsearch integration | 20:26 |
| jimbaker | in particular, it would be great to do this with kibana as well | 20:27 |
| jimbaker | so kibana could be an alternative frontend to craton... | 20:27 |
| jimbaker | but that's not immediate. just want to put it out there however | 20:27 |
| thomasem | zz_pwnall138: I can help you get a ZNC bouncer set up for IRC and help with the nickname problems. You might just need to register with nickserv to stop losing your nick. | 20:28 |
| thomasem | jimbaker: yeah | 20:29 |
| thomasem | craton --format rax-custom ... | 20:30 |
| openstackgerrit | Thomas Maddox proposed openstack/craton master: JSON Path-like querying for variables https://review.openstack.org/443941 | 21:12 |
| fsaad | don't have a meeting invite for monday meeting, do y'all have one ? | 21:13 |
| jimbaker | thomasem, thanks for helping pwnall138 on that | 21:15 |
| thomasem | fsaad: sent | 21:15 |
| thomasem | jimbaker: no problemo! | 21:15 |
| fsaad | thanks thomasem ! | 21:15 |
| thomasem | fsaad: you bet! | 21:15 |
| jimbaker | re craton package on pypi, git-harry, sulo, thomasem - please tell me your pypi usernames so i can add you as admins of this package - https://pypi.python.org/pypi/craton | 21:20 |
| *** harlowja has joined #craton | 21:33 | |
| thomasem | jimbaker: thomasem | 21:40 |
| thomasem | :) | 21:40 |
| thomasem | So, I think it's getting to a point where we do need to add some validation for variable keys if we're expecting them to be python identifiers... I'm running into weirdness in the DB with keys that have hyphens. I can work around it, but it's going to make this more complex than it has to be when we're already wanting folks to only use valid python identifiers as keys in variables. | 21:48 |
| thomasem | Thoughts? | 21:48 |
| thomasem | This also means the keys in the value, for nested ones. | 21:50 |
| jimbaker | thomasem, thanks | 21:59 |
| jimbaker | thomasem, added you to pypi | 22:03 |
| thomasem | Awww yiss | 22:05 |
| thomasem | Really moving up in the world. | 22:05 |
| *** turvey has quit IRC | 22:08 | |
| thomasem | I'm off to see the wizard. Have a lovely evening/day! | 22:12 |
| jimbaker | thomasem, take care, and have a great rest of your day! | 22:13 |
| anonymike | jimbaker: I'm trying to figure out the best approach for being able to run craton both directly and in docker. Would it be okay if we specify in the docs for setting up a dev environment that a user creates a craton-api-conf.dev, and we point the craton-direct tools there while craton-start-docker maintains craton-api-conf.sample | 22:58 |
| jimbaker | anonymike, clearly one of the advantages of craton-start-docker (should be renamed craton-direct-start to be consistent...) is that the user can specify their own config | 22:59 |
| jimbaker | having said that, i think defaulting to the existing etc/ config files could be a nice thing to do | 23:00 |
| jimbaker | but as you may have noticed, the config file entry in https://github.com/openstack/craton/blob/master/etc/craton-api-conf.sample#L13 has to be specified with an abs path, and is hardcoded for docker | 23:01 |
| anonymike | right | 23:01 |
| jimbaker | so that's the only thing - it would be nice to avoid barriers to entry. but maybe that's a little too much of a worry | 23:02 |
| anonymike | hmm okay, I'm gonna submit what I have a little later tonight (after a run and some dinner) we can discuss if anyone is around then or tmrw morning | 23:09 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
