Thursday, 2015-03-19

openstackgerritJohn Vrbanac proposed openstack/barbican: Starting to rework docs around the secret resource
*** dave-mcc_ has joined #openstack-barbican00:02
*** dave-mccowan has quit IRC00:04
*** jamielennox is now known as jamielennox|away00:09
*** jamielennox|away is now known as jamielennox00:23
*** tkelsey has joined #openstack-barbican00:26
*** gyee has quit IRC00:30
*** tkelsey has quit IRC00:31
openstackgerritAdam Harwell proposed openstack/barbican: Make the default devstack config use the right password
*** dave-mcc_ has quit IRC01:38
*** SheenaG has quit IRC01:46
*** dave-mccowan has joined #openstack-barbican01:46
*** jkf has quit IRC01:58
*** tkelsey has joined #openstack-barbican02:28
*** tkelsey has quit IRC02:32
*** zz_dimtruck is now known as dimtruck02:35
*** alee_afk has quit IRC02:58
*** alee_afk has joined #openstack-barbican03:11
openstackgerritDave McCowan proposed openstack/barbican: Change certificate unit tests to use strong algorithms
alee_afkjvrbanac, redrobot you there?03:31
*** dimtruck is now known as zz_dimtruck03:39
*** greghaynes has quit IRC03:43
*** kebray has joined #openstack-barbican03:48
*** kebray has quit IRC03:48
*** woodster_ has quit IRC03:50
rm_workalee_afk: I am here, but pretty sure redrobot just +A'd something and jvrbanac +2d something03:57
alee_afkrm_work, I think I figured out what was wrong with my setup using redrobots scripts and wanted to see if he knew what was up04:01
alee_afkrm_work, I'm going to try your scripts next (modfied for fedora)04:01
rm_workwhelp, yeah, can't help you with his scripts :P04:01
rm_workalee_afk: ah yeah04:01
rm_workit should be minimal04:01
rm_workjust apt-get -> yum04:01
rm_workI think the package names are… SOMEWHAT similar04:02
alee_afkrm_work, yeah and maybe a couple of package name changes04:02
rm_workand soon this will merge: :P04:02
alee_afkwhat I saw was wrong was that the keystone instance did not have the barbican entries in the service catalog04:03
rm_workthose are supposed to be added by … damnit, he told me what added them like, two days ago04:03
rm_workline 20604:04
rm_workper my scripts, that all gets copied into the devstack project folder prior to running stack04:05
alee_afkah - so maybe I can just run those commands ..04:05
alee_afkfingers crossed ..04:06
*** jamielennox is now known as jamielennox|away04:06
rm_workpossibly just source that script and run create_barbican_accounts :P04:06
rm_workhow did you get your devstack set up with barbican WITHOUT having that file in place04:07
rm_workmy motto for devstack is: when in doubt, nuke the VM from orbit and start over04:08
*** kebray has joined #openstack-barbican04:14
alee_afkrm_work, I'm trying something a little different , but its not working -- starting vm from scratch04:16
openstackgerritMerged openstack/barbican: Make the default devstack config use the right password
rm_workwow that was fast04:19
openstackgerritMerged openstack/python-barbicanclient: Removing assertItemsEqual workaround, fixed upstream
rm_workwoo again04:46
openstackgerritDave McCowan proposed openstack/barbican: Fix string substitution in exception messages missing the s
*** dave-mccowan has quit IRC05:12
*** greghaynes has joined #openstack-barbican05:40
*** tkelsey has joined #openstack-barbican06:29
*** tkelsey has quit IRC06:33
*** kebray has quit IRC06:38
*** gitorres has joined #openstack-barbican07:22
*** gitorres has left #openstack-barbican07:23
*** chlong has quit IRC07:47
*** nickrmc83 has quit IRC08:16
*** nickrmc83 has joined #openstack-barbican08:22
*** tkelsey has joined #openstack-barbican08:28
openstackgerritJuan Antonio Osorio Robles proposed openstack/castellan: Start using oslo.policy
openstackgerritTim Kelsey proposed openstack/castellan: Fixing some warning about oslo namespace
openstackgerritTim Kelsey proposed openstack/castellan: Fixing some warning about oslo namespace
*** woodster_ has joined #openstack-barbican12:00
*** gitorres has joined #openstack-barbican12:00
*** chlong has joined #openstack-barbican12:14
*** gitorres has quit IRC12:24
*** chlong has quit IRC12:26
*** dave-mccowan has joined #openstack-barbican12:37
*** chlong has joined #openstack-barbican12:43
*** chlong has quit IRC13:21
*** SheenaG has joined #openstack-barbican13:28
*** alee has joined #openstack-barbican13:54
aleereaperhulk, rm_work ping13:57
aleereaperhulk, rm_work , jvrbanac , hockeynut , redrobot -- I'm trying to run tox on a fedora 20 machine and am running into problems with pip installing the test requirements -- something about python-cryptography not compiling.14:00
alee {standard input}: Error: open CFI at the end of file; missing .cfi_endproc directive14:00
redrobotalee hmm... weird.14:01
*** zz_dimtruck is now known as dimtruck14:01
redrobotalee I'm sure reaperhulk will be on soon enough...  or maybe tryp #cryptography-dev ?14:02
aleeredrobot, incidentally I tried running your scripts to set up a keystone docker instance and postgres instance14:02
aleeand pointing my current barbican instance there14:02
aleefound that I needed to add barbican to the service catalog14:03
redrobotalee were you using the client?14:03
aleeI ran the commands in contrib/openstack/barbican to do that, but still getting 401s14:03
aleejust running tox -e functional14:04
aleeagainst a running barbican server14:04
aleethat points to keystone14:04
redrobotalee oh I know!  The functional test use "secretservice" as the pw, to match devstack-gate.14:04
redrobotalee I think my script defaults to "orange"14:04
aleeredrobot, yeah I changed that to password14:05
aleelet me try orange14:05
redrobotalee try changing to orange14:05
*** SheenaG has left #openstack-barbican14:06
aleeredrobot, is the user right?14:07
redrobotalee no, that doesn't look right14:08
aleeso username=barbican?14:08
redrobotthat'll give you the service user14:09
redrobotor if you want secret-admin14:09
*** SheenaG has joined #openstack-barbican14:09
aleedaomain_name=Default ?14:10
redrobotalee yeah14:10
aleeredrobot, cool - much better14:13
aleeredrobot, excellent - now the only ones that fail are the ones I added :)14:14
aleeredrobot, hey - is it possible to run just a single functional test or set of functional tests?14:16
*** rellerreller has joined #openstack-barbican14:18
*** dhellmann has quit IRC14:19
*** dhellmann has joined #openstack-barbican14:19
redrobotalee I'm not sure...  we're using nose in the tox command, but I can't find how to run a single test with nose14:20
aleeredrobot, no worries -- I'll just dig through the logs14:20
rellerrellerYou cannot run a single test with nose14:21
rellerrellerI can dig up the command real quick. Is it for functional or unit test?14:22
rellerrelleralee redrobot does this help `python -m barbican/tests/plugin/`14:24
rellerrelleralee redrobot for functional test ` nosetests functionaltests/api/v1/functional/`14:25
aleerellerreller, cool - let me try that ..14:26
rellerrellerredrobot What is the deadline for kilo-3? I thought today was the last day, but I thought someone said yesterday was.14:26
rellerrellerThe kilo release schedule says it is the 19th,
redrobotrellerreller the release should happen today14:26
redrobotI bumped all pending reviews to RC114:27
redrobotrellerreller I'll be meeting with the release-manager in about an hour and a half14:27
redrobotrellerreller then I'll give him whatever the HEAD sha is in master14:27
rellerrellerSo code that is pending now can be in RC1? That makes me feel better.14:27
redrobotrellerreller yeah, since we're still the-status-formerly-known-as-incubated, then we don't need to ask the release manager for Feature Freeze Exemptions14:28
redrobotbut, I think we should try to minimize those next cycle14:28
redrobotwe had 4 this cycle. :-\14:29
elmikoredrobot: got a minute to chat about castellan?14:40
redrobotelmiko yeah, what's up?14:40
elmikoso, i've been following the github repo and see more stuff coming in14:40
elmikoi'm curious about how we might integrate it into our code14:41
elmikodo we need to write our own plugin for it?14:41
redrobotelmiko no, the castellan lib should eventually have a barbicanclient implementation.14:42
redrobotelmiko but it still has to be written. :-\14:43
elmikoi'm trying to rewrite a spec about using an external keystore and i want to use castellan, but i'm not sure how to proceed14:44
elmikowill we evenutally control the key manager backend with a config option, or how does that work?14:44
jvrbanacredrobot, ping when you're done14:46
redrobotelmiko yeah, there'll be a config option that tells Castellan what implementation to load.14:48
elmikoredrobot: ok, now the unpopular question, when do you think we should look towards a barb client for it? (i'm curious if we should use the abc's now and work towards barb when it's available)14:49
*** kebray has joined #openstack-barbican14:49
redrobotelmiko I think we'll have one for Liberty14:49
elmikoredrobot: ok, so sounds like we could use the abc's in the short term?14:50
elmikoor at least start experimenting with them14:50
redrobotelmiko yeah, once we release 0.1.0 you can start using the interface.14:53
elmikoredrobot: awesome!14:53
elmikoredrobot: if i can make time in my schedule, is there work that extra hands could help with?14:54
elmikoi'm trying to think ahead about scheduling here14:54
redrobotelmiko nobody's working on the barbicanclient impl yet, so that'll definitely be a good chunk of work to pick up14:55
elmikoredrobot: yea, i have a feeling i would need to collaborate with others.14:55
elmikobut, i'd rather not implement some pluggable system that only exists in sahara14:56
elmikoredrobot: is there any sort of castellan meeting or sync up that i could attend, or is all happening during barb meetings14:57
redrobotelmiko it's all lumped into barbican right now, so castellan topics would be good for the weekly barbican meeting14:58
elmikoredrobot: ok, many thanks for the information. i'll just keep showing up and pestering people =)14:59
elmikoand probably experiment with what's available14:59
elmikoare you and kfarr the main contact points?15:00
redrobotelmiko yep!  and rellerreller too.15:00
elmikosweet, thanks again. i appreciate it15:00
*** kebray has quit IRC15:00
redrobotelmiko no problem!15:01
redrobotjvrbanac what's up?15:01
jvrbanacredrobot, sooo, I've noticed that people started making comments assertion order. I didn't think the discussion was done on the matter as we didn't vote on it during the meeting. Also, If we are actually gonna change it, I would suggest we hold off until after the Kilo is cut so Kilo doesn't have large number of inconsistencies around this matter.15:05
*** dave-mccowan has quit IRC15:05
jvrbanacs/comments/comments on/15:05
*** dave-mccowan has joined #openstack-barbican15:06
*** dave-mccowan has quit IRC15:06
redrobotjvrbanac I thought we all agreed on following the testtools docs15:06
redrobotjvrbanac you were the only one opposed, so I didn't think we needed a vote.  I can add one to the agenda if you feel it's necessary.15:06
openstackgerritThomas Dinkjian proposed openstack/python-barbicanclient: First set of negative functional test for secrets
jvrbanacredrobot, so, regardless of my personal opinions on this, we really shouldn't be creating inconsistencies in the code-base shortly before release15:08
* redrobot shrugs15:09
redrobotjvrbanac I think there's plenty of incosistencies, and would rather start doing the thing we agreed on now.15:09
redrobotjvrbanac the final Kilo release is still a ways out, so I think we have time to go back and fix stuff that needs to be fixed...15:10
redrobotjvrbanac I guess another way to think of it is I value to start having consisten reviews now, instead of switching how we review when the next cycle starts15:11
openstackgerritMerged openstack/barbican: Removed get_secret_type
*** kebray has joined #openstack-barbican15:14
jvrbanacredrobot, I'm talking about code consistency here15:14
jvrbanacredrobot, Let me make sure I'm hearing you right... You're saying that you don't care about that because there are already inconsistencies within the code?15:15
redrobotjvrbanac It's not that I don't care about code consistency.   We agreed on how we want our code to be consistent, so we should start writing code that way.  I understand your argument to be that we shouldn't start enforcing the convention we agreed on because we're releasing soon.  I disagree that the next release (RC1) is too soon to start enforcing the convention now.15:18
rellerrellerelmiko I saw your comment about Castellan. We have developed a Barbican KeyManager implementation.15:24
jvrbanacredrobot, I'm saying that we have about 3-4 weeks before the RC's. Since we have so many higher priority things to take care of, we're essentially saying that we're gonna live with supporting mixed assertions for the life of Kilo.15:24
rellerrellerelmiko kfarr will be adding it.15:24
jvrbanacredrobot, oh. not that this makes a difference but here was the BDFL's view on this:
*** igueths has joined #openstack-barbican15:28
jvrbanacredrobot, TL;DR for that was to remove the concept of expected, actual and do first, second due to religious wars on the matter. lol15:29
rm_workyeah i don't see EVERYONE ever agreeing15:33
rm_workI came from Java most recently, so as he said, I would expect the Java jUnit ordering15:34
rm_workbut first/second makes sense to me15:34
rm_workit just means you have to actually look at the code of the unit test to make sense of the message15:34
*** gyee has joined #openstack-barbican15:37
woodster_I guess I'm thinking of this not from the canonical unit test side of things, but from the perspective of a library function call I'm making...if its docstring says to pass in data a certain way, that's the way I pass it in.15:39
woodster_so if the docstring says this:15:39
woodster_that's what I do.15:40
rm_workyeah, are THOSE all consistent?15:40
woodster_I would not feel comfortable using that contact differently?15:40
jvrbanacwoodster_, rm_work, redrobot, again... my personal opinions aside. My concern is about code consistency, quality, and readability. I would be much more comfortable for us to get 80-90% of the assertions changed before we started to enforce it in reviews. Considering we have a lot of things to do before the RCs, my feeling is that this should be deferred until Liberty15:40
rm_workif so, that seems like the obvious answer, but… <_<15:40
rm_workwoodster_: yeah if it's been like that since (I assume) forever, then why is this even an argument?15:41
woodster_I've been thinking of this as a diff-cover sort of thing...whereby we just enforce a standard/approach on changed lines in code reviews initially (esp. for the crunch that is Kilo)15:42
rm_workjvrbanac: honestly, i seriously doubt it MATTERS for a release cycle which order these args are in -- not only do very few people who run a deployment care about the output of the unit tests -- the unit tests WILL still pass/fail correctly15:42
rm_workso the actual perceived impact for anyone besides a developer working with a specific failing unit test is … essentially zero15:43
openstackgerritDouglas Mendizábal proposed openstack/castellan: Renames for consistent namespaces
woodster_I do agree with jvrbanac that consistency across the code base is important. But I also agree that do so NOW for Kilo is challenging for sure. That's why a new-lines-in-CR-only approach makes sense to me for now.15:44
jvrbanacrm_work, as developers, we have to support Kilo. the inconsistencies affect us.15:44
woodster_I'd also add that we are already inconsistent in the code base...I've been following (expected, observed) for my commits :)15:44
rm_workjvrbanac: yeah but you can always just LOOK at the argument order in the test you're running :P15:45
rm_workjvrbanac: and since we're developers, we can use our mental capacity to realize what it is telling us based on that ;)15:45
rm_workI would only be worried about non-devs being confused15:45
rm_workyeah, we've done the "follow-this-for-new-CRs-only" approach before, right? just to stop the bleeding on whatever it is, so going in and fixing the existing code isn't an ever-growing task15:46
hockeynutsounds like it should be the new way for new CRs, then post-RC we move on being consistent across the codebase?  I would think its not worth risk of breakage before RC15:47
elmikorellerreller: awesome, any idea when that might be in the repo?15:47
woodster_rm_work we really want all tests to report errors the same way, so we *don't* have to look at the code to determine these things. That said, it would REALLY matter If the output message from testtools was BETTER for determining what is expected vs does a terrible job of this IMHO. :\15:48
rm_workwoodster_: yeah, but we'd ALWAYS have to look for Guido's suggestion15:48
rm_workbut honestly, if you're debugging code using a test, you're going to be looking at the code anyway <_<15:49
woodster_yeah, maybe we should add a Liberty 'paper cut' to LP to convert these things over throughout the code base?15:49
*** bdpayne has joined #openstack-barbican15:49
rm_workI really just feel like this issue is being blown out of proportion15:49
openstackgerritDouglas Mendizábal proposed openstack/castellan: Renames for consistent namespaces
rm_workI wasn't commenting earlier because I thought all the argument was ridiculous T_T15:50
openstackgerritDouglas Mendizábal proposed openstack/castellan: Renames for consistent namespaces
aleeredrobot, can we have a quick workflow on ?15:51
redrobotalee done15:52
aleeredrobot, also
hockeynutstill looking for feedback on ... :-D15:52
rm_workalee: CAID doesn't need one?15:53
rm_workthe last two exceptions15:53
rm_workInvalidCAID and CANotDefinedForProject15:54
rm_workare still missing the s15:54
aleerm_work, good catch15:54
rm_workredrobot: REVERSE COURSE15:54
aleeyeah - those do too ..15:54
rm_workor whatever, fix those in another one T_T15:54
aleewe'll just add another quick one15:55
rm_worksince god, who knows how long the gate will take15:55
redrobotrm_work doh15:55
rm_worksometimes you gotta look at what ISN'T changed, not just what IS :P15:55
aleeI'll add to my next big CR15:55
aleewhich is functional tests for cert issuance15:56
aleethat tests for this in any case15:56
aleeredrobot, dont forget
redrobotalee done15:57
*** rellerreller has quit IRC16:00
*** arunkant_ has joined #openstack-barbican16:00
*** rellerreller has joined #openstack-barbican16:06
*** arunkant_ has quit IRC16:13
openstackgerritMerged openstack/barbican: Fix string substitution in exception messages missing the s
*** SheenaG has quit IRC16:17
*** arunkant_ has joined #openstack-barbican16:20
openstackgerritMerged openstack/barbican: Change certificate unit tests to use strong algorithms
openstackgerritDouglas Mendizábal proposed openstack/castellan: Remove Python 3.3 from setup.cfg and tox.ini
chellygelo/ redrobot question16:47
redrobotchellygel \o answer16:47
chellygelwhat do you mean apply to the 'transfer of the secret'16:47
chellygelin your comment on the docs16:47
chellygeli really like the table idea instead of the long paragraph that we had in comparison in the api.rst16:48
chellygelits much easier to read16:48
chellygelbut~ id ont know what your words mean.16:48
redrobotchellygel there's a slight chance I don't know what I'm talking about :)16:49
redrobotchellygel let me check the api real quick16:49
chellygelalso, i would love a good pointing to other docs that you reference16:53
chellygeli have been googling around for them but i feel liek they are all different16:53
chellygelthank yooou16:53
chellygelwhere can we view the source on thse?16:54
chellygelis the idea that we have a page on this page?16:54
*** gyee has quit IRC16:56
redrobotchellygel I'm not sure where the source is... I think it's a single repo where the page gets generated from a WADL16:57
redrobotchellygel yes, we'll eventually have a section for barbican alongside the rest of the OpenStack services16:58
redrobotKilo-3 is released: :D16:58
chellygelim not familiar with how WADL works so well, will we be able to do this with RST still?16:58
redrobotchellygel yes, I'm not sure if we're allowed to be in the main API repo yet...  may be worth reaching out to openstack-docs16:59
chellygelis that a channel? i can dig into it16:59
chellygeli see a bunch of maven commands T^T17:00
redrobotchellygel would probably involve adding a new xml here
chellygelsounds like we're going down the wrong path with these docs then for this CR :(17:01
redrobotchellygel not necessarily.  It would be good to reach out to the doc team.  The guidance for Incubated projects was to keep everything in-tree in RST format17:03
redrobotchellygel but now that Incubated is no longer a thing, we may be allowed to contribute to api-site directly17:03
chellygelredrobot, i mean to say this probably should've started as a spike17:03
redrobotchellygel maybe Constanze would know? She's listed as our doc liaison
chellygelredrobot, im waiting for jvrbanac to get back, to discuss17:04
*** dave-mccowan has joined #openstack-barbican17:04
*** redrobot changes topic to "Kilo RC1 due April 9"17:05
*** darrenmoffat has quit IRC17:06
*** darrenmoffat has joined #openstack-barbican17:07
*** rellerreller has quit IRC17:11
elmikosahara is going through the wadl dance currently, it's not as much fun as reported ;)17:12
redrobotelmiko weren't you working on some swagger + barbican side project?17:12
chellygelany tips elmiko ?17:12
elmikoredrobot: yea, i actually have a barbican swagger output17:13
elmikochellygel: it's a lot of manual work as there are no good tools that will go from pecan to wadl17:13
elmikochellygel: i would recommend looking at one of the smaller APIs, don't dive into nova for an example17:13
elmikoredrobot: if you look in the output directory, you'll see what it generated for barbican17:14
elmikoi think it's a little out of date now as i still see the performance endpoint in there17:15
*** dimtruck is now known as zz_dimtruck17:16
redrobotelmiko yeah... there's a ton of new endpoints for the CA ID blueprint17:18
redrobotchellygel the irc channel for the docs team is #openstack-doc17:19
elmikoredrobot: i should really run it again, i need to do a little markup to make it work though17:19
elmikoredrobot: if you're curious, contains the changes i needed to make it work17:20
elmikoi also had to put together the pecan-swagger package, but that can hopefully be reused on other projects17:20
jvrbanacredrobot, chellygel back... so from I was given the guidance by our doc people to put our docs in-tree as RST.17:22
jvrbanacredrobot, I'm confused here17:28
redrobotjvrbanac about?17:28
jvrbanacredrobot, sooo we're trying to split out the docs here. The idea of this that secrets.rst is to put everything related to secrets in there for now.17:31
jvrbanacExamples, error docs, api reference, and resource usage17:32
jvrbanacbrb lunch17:32
redrobotjvrbanac I thought the goal was to produce the API Reference only?17:32
rm_workman, i still need to get loonch, anyone in SA that hasn't loonched yet?17:33
chellygelcome over i will make rice and seaweed~ \o/17:33
rm_workis that your lunch today? :P17:34
*** rellerreller has joined #openstack-barbican17:34
rm_workI should PROBABLY just grab fast food and head into the office quickly tho T_T17:37
chellygelstop at freddy's17:37
rm_workyou have a knack today for the restaurants I avoid :P17:38
chellygelwow T~T17:38
rm_workthe amount of grease on their burgers makes my stomach wumbly17:38
rm_workI honestly have a very short list of places i don't go >_>17:38
rm_worknot sure how you're singling most of them out :P17:39
redrobotrm_work /me is hungry at castle17:40
rm_workredrobot: if you want to go somewhere that serves meat in some capacity, i would meat you there :P17:40
redrobotduuuuudte, I go to Freddy's FOR the grease.17:40
chellygelmeat you there17:40
rm_workredrobot: T_T17:41
redrobotrm_work heh.. I'm actually not eating meat right now... lent and all...17:41
elmikoi take it you all are in the same town?17:41
rm_workfreaking… Catholics… ruining everything…. fffffff17:41
redrobotrm_work but I'd be down for some asian food... they always have tofu handy17:41
rm_workno offense intended for Catholics17:41
rm_workbut #($&#()%$17:41
chellygeldont suggest thai chalurn redrobot  o/17:41
chellygelyes elmiko we are all in SA17:41
redrobotelmiko yeah, rm_work chellygel woodster_ and I all work out of SA17:41
rm_workI already heckled redrobot for his Thai Chalurn excursion the other day17:42
elmikoooh, must be nice and warm =)17:42
chellygelyou should get korean again o/17:42
elmikoi'm in detroit, still kinda cold =(17:42
rm_worki guess something Asian would work if you know a good place17:42
redrobotelmiko it's a little rainy this week, but it's definitely warming up17:42
rm_workohh Sawasdee is good17:42
elmikook, i need to stop reading this, you are making me seriously hungry17:43
redrobotrm_work you at home?17:43
rm_workchellygel: what part of SA you live in again? I am sure you could be excused for a QUICK excursion17:43
rm_workredrobot: yeah, was about to be on my way in17:43
chellygel35 and 160417:43
rm_workah so, Bangkok Thai17:43
redrobotrm_work +1 would eat there for sure!17:44
rm_workis the bomb17:44
chellygeli think im gonna drive up to austin for my lunch break tho17:44
rm_workchellygel: T_T_T_T_T17:44
rm_workI can't have enough T_T faces in there17:44
rm_workneed to make them… denser17:44
chellygelim sorry, austin is calling meeee ~17:44
chellygelmy lease isnt up until june or july17:44
redroboti love how chellygel waited to tell you where she lived first, before saying she wasn't interested in lunch trolololol17:44
chellygelwell the thing was17:45
chellygeli was gonna work through lunch17:45
chellygelbut then jvrbanac left17:45
chellygeland now we're at an impasse so im like \o/ wat do17:45
rm_workchellygel: yeah, so, Bangkok Thai17:45
chellygelwhere is that17:45
redrobotrm_work chellygel +2 would eat there for sure17:45
rm_workI-35 and 160417:45
chellygeli am not familiar and ive eaten at a lot of places17:46
chellygelis it off pat booker?17:46
rm_workchellygel: it's like… next to you17:46
redrobotchellygel technically it's 1604 and Pat Booker17:46
chellygelahkay i ahvent been there before17:46
rm_workrofl wat17:46
rm_workit's like… tied for 1st Thai Food in SA17:46
chellygeli got o thai spice and sushi17:46
chellygelwhich was always good17:46
rm_workwell I am going to just leave for the office now17:46
rm_worksince i'm 15m away17:46
rm_workmay as well just meet redrobot there17:46
rm_workif we're doing Bangkok17:47
rm_worksince it's not on the way for me at all :P17:47
redrobotrm_work I'm in for Bangkok17:47
rm_workkk, work on chellygel in the next 15m17:47
chellygeli am so torn.17:47
redrobotchellygel voted #1 thai in SA for like 5 years in a row now.17:47
rm_workchellygel: yeah I don't even17:47
*** crc32 has joined #openstack-barbican17:47
chellygeli _want_ to go but i have to be in austin by 4 pm17:47
rm_workyou live like across the street17:47
chellygelare they fast?17:48
* rm_work leaves for srs17:48
rm_workchellygel: they aren't… SLOW....17:48
chellygelbai. maybe see you maybe not :P17:48
rm_workat least not at 1pm17:48
rm_workrush is over17:48
* rm_work leaves for srs for srs17:48
chellygelok... i'll go eat lunch17:49
chellygeli give in redrobot rm_work  you win17:49
redrobotchellygel VICTORY!17:49
redrobotchellygel plus I need another Street Pass :D17:49
redrobotchellygel ok, I'm heading out17:51
chellygelok ill leave in 5 min17:52
*** zz_dimtruck is now known as dimtruck17:52
openstackgerritDave McCowan proposed openstack/barbican: Fix string substitution in exception messages missing the s
*** gyee has joined #openstack-barbican18:06
*** crc32 has quit IRC18:11
openstackgerritJohn Wood proposed openstack/barbican: Add sub-status logic to worker/task processing
*** rellerreller has quit IRC18:37
*** SheenaG has joined #openstack-barbican18:38
*** jkf has joined #openstack-barbican18:40
aleewoodster_, did you just receive an email about the openstack preso?18:41
jvrbanacredrobot, chellygel, i'm back18:41
aleewoodster_, that cr is timely.  I'm creating functional tests for certs and they are all failing because the status/ sub-status is not set correctly18:42
elmikojvrbanac: i think they all went to lunch18:42
*** dimtruck is now known as zz_dimtruck18:42
jvrbanacredrobot, so technically an "API reference" can include resource usage, information on errors, and examples18:42
jvrbanacelmiko, :(18:42
elmikosorry =(18:43
woodster_alee, I'm sending another update to it now btw18:43
aleewoodster_, I just got an email saying the preso was "open for voting"18:44
*** kebray has quit IRC18:44
aleeand that voting ends February 2318:44
aleesomething is confused18:44
woodster_alee, what?? I didn't get is like you went back in time or something! Back those pecking creatures under your eaves???18:46
woodster_alee, that said I do have a 3pm meeting today that I'm pretty sure was for last month....18:47
aleewoodster_, other folks I know got it too .. need to get back in my time machine18:47
woodster_alee, was that email directly from me?18:47
*** tkelsey has quit IRC18:48
openstackgerritJohn Wood proposed openstack/barbican: Add sub-status logic to worker/task processing
aleewoodster_, no from summit folks -- I think someone is confused18:50
woodster_alee, ok that should do it ^^^  hockeynut, please re-review when you can....18:50
aleewoodster_, ok will look soon18:50
woodster_alee, must be something to do with the finalizing of presentations (we were asked to combine two into one for example)18:50
openstackgerritJohn Wood proposed openstack/barbican: Add retry periodic task and worker-client logic
*** xaeth_afk is now known as xaeth19:11
*** zz_dimtruck is now known as dimtruck19:13
*** rellerreller has joined #openstack-barbican19:17
*** bdpayne has quit IRC19:24
jvrbanacredrobot, ping19:24
openstackgerritJohn Wood proposed openstack/barbican: Allow business logic and plugins to retry tasks
rm_workalee: yeah, the Ooops email just went out :P19:40
redrobotjvrbanac pong19:57
redrobotrellerreller ping20:01
rellerrellerredrobot pong20:01
redrobotrellerreller woodster_ was saying you were concerned about the Asymmetric Key Support?20:02
redrobotrellerreller This is what I have in Feature Freeze Exception for Kilo:
rellerrellerredrobot I was not sure if accepted for feature freeze extension20:02
jvrbanacredrobot, so when you get a sec. can we discuss this docs stuff?20:03
rellerrellerredrobot It looks like it will make it into kilo, so I feel a lot better now :)20:03
redrobotrellerreller yeah, sorry I thought Launchpad would send you an email when I changed the target milestone to RC120:04
*** dave-mccowan has quit IRC20:04
rellerrellerredrobot It probably did, but I get so many emails for code reviews and this and that that it probably got lost in the shuffle.20:04
redrobotrellerreller hehe, makes sense.  Yeah, I figured the 4 outstanding BPs should go into Kilo since the code is already up for review.  I'm hoping we can land all 4 in the next week.20:05
redrobotjvrbanac what's up?20:05
jvrbanacredrobot, soo the purpose of this doc is to just get split out and correct the information around the secrets resource into a different page.20:06
redrobotjvrbanac I thought the goal was to create a General API reference?20:07
aleeredrobot, I can safely say that cert api is not done.20:08
aleeredrobot, given that I'm writing all these functional test cases and they are all failing20:08
redrobotalee uh oh...20:08
redrobotalee is it broken in K-3?20:08
aleeredrobot, absolutely20:08
redrobotalee because that ship has sailed. :-\20:08
aleeredrobot, it wont be fixed in a day or two.  got a weeks worth of work at least20:09
redrobotwell, it's a good thing we're still a few weeks out from RC1 :)20:09
aleeredrobot, yup20:09
jvrbanacredrobot, what do you mean by general api reference?20:09
aleeredrobot, at least now with the functional tests, we have a target to go to.20:09
* alee note to self. next time, write the functional tests first.20:10
redrobotjvrbanac A comprehensive doc that lists all endpoints and sample payloads for each operation.20:10
redrobotjvrbanac although I'm not sure if the goal is to get to something that looks like OpenStack API reference:
redrobotjvrbanac or if it's to make something that looks like a Rackspace reference
*** kebray has joined #openstack-barbican20:11
jvrbanacredrobot, soo... those are all wadl generated things20:11
jvrbanacredrobot, so I started a similar reference section in that page20:12
redrobotjvrbanac similar to which? ...  I think it would be easier to make a bare-bones API ref like the OpenStack ones.20:14
jvrbanacredrobot, that bare-bones API ref is generated from their wadls.
redrobotjvrbanac right, but I'm not sure we're allowed in the party...  I meant content-wise, bare bones API20:15
jvrbanacredrobot, but that barebones thing isn't anything more than just telling you the call. That's not really sufficient as our only source of documentation on our resources.20:17
jvrbanacalso woodster_ ^^20:18
*** dave-mccowan has joined #openstack-barbican20:18
redrobotjvrbanac that's true, and if you want to make a Quickstart section as well, then that'll be awesome.  My comment in the CR was that Quickstart and the comprehensive API should be separate20:18
redrobotjvrbanac so that someone who is just looking for the call reference can find it quickly20:19
redrobotjvrbanac and someone who needs hand-holding and long explanations can look at the Quickstart20:19
jvrbanacredrobot, ok... how would you like this? You want this in a separate page or directory?20:19
jvrbanacYou just want a very visible section20:20
jvrbanacredrobot, you mentioned it should be moved out of the page20:20
redrobotjvrbanac so the client, for example, has a Client Usage page, where it talks about workflows, and then a separate Reference page, where it just gives the barebones details of the different classes/methods etc20:21
redrobotjvrbanac I think somehting similar would be good for the API.  A QuickStart section with workflows, and a API Reference section with just the call details.20:22
redrobotjvrbanac uses that same organization.  A section for Quickstart workflows, and a section with the call listing.20:23
redrobotjvrbanac with the admin-only calls possibly in a separate API section20:24
*** igueths has quit IRC20:26
*** tkelsey has joined #openstack-barbican20:27
jvrbanacredrobot, ok. I'll do that. Yeah, I was thinking you wanted us to try do some sort of wadls thing to generate that stuff.20:28
woodster_so I didn't think we were concerned about look and feel at this  point for the APIs, just content and *maybe* structure similar to existing projects20:28
woodster_It seems like the primary goal is to move away from cloudkeep's wiki correct?20:29
woodster_jvrbanac the wadls are eventually required for the 'formal' docs20:30
woodster_jvrbanac well, unless that is all changing now :)20:30
jvrbanacwoodster_, yeah, not sure that's true anymore20:30
*** rellerreller has quit IRC20:31
woodster_maybe if we procrastinate enough docs won't be required anymore? 8^)20:31
*** tkelsey has quit IRC20:31
redrobotwoodster_ lol20:31
woodster_good thing I only write self-documenting code20:31
redrobotwe do HATEOS right?  Can't the client just figure everything out on its own?20:33
jvrbanacredrobot, woodster_, lol. soooo eventually, I would like to tie in some of these docs with docstrings around our routes/controllers20:33
redrobotjvrbanac :meth:`` would work, I think.20:33
jvrbanacredrobot, yeah... that is more of a long-term goal20:34
woodster_jvrbanac, elmiko is looking into that already...scroll up to 12:14 or so today20:34
woodster_redrobot, where would that meth tag go?20:35
*** gitorres has joined #openstack-barbican20:36
*** gitorres has quit IRC20:36
redrobotwoodster_ in an .rst file inside doc/source.  It'll pull in whatever docstring the method has.20:36
woodster_jvrbanac, well I think he is trying to generate things from the pecan structure, not from docstring markup20:36
woodster_redrobot, oh I see20:36
redrobotYeah, elmiko's work was around being able to automagically generate the wadl20:36
redrobotjvrbanac eventually we'll have to produce the WADLs...  I thought maybe we could kill two birds with one stone since y'all are compiling all the API information right now, but I don't have a strong opinion either way...   I don't even know if the doc folks would merge a patch with a wadl for barbican.20:38
redrobotelmiko ears burning?20:40
jvrbanacredrobot, OpenStack is going to require a WADL? I thought they were moving away from them.20:40
elmikohehe, just saw the name highlight20:40
elmikojvrbanac: if you're curious, i started putting this together for pecan projects to help markup the controller paths20:41
redrobotjvrbanac afaik, this is the new standard way to document the API reference
*** gitorres has joined #openstack-barbican20:41
elmikojvrbanac: it could use more work to fill out all the gory details, but its a start for gathering the basic info20:42
jvrbanacelmiko, interesting20:42
elmikoi was actually thinking about writing a tool to go from swagger to wadl to help out the sahara efforts, but i'm not sure it will be needed20:42
elmikocurrently, pecan-swagger will attempt to analyze controller classes and pull all the routes out20:43
elmikobut, because of object-dispatch, you need to markup the classes and inform about the hierarchy20:44
*** SheenaG has quit IRC20:47
*** alee has quit IRC20:55
jvrbanacelmiko, just curious. I don't know much about swagger. What's so special about it? I've been seeing people talking about api-blueprint and swagger both lately. They look interesting, but I really haven't spent time with them20:56
*** SheenaG has joined #openstack-barbican20:56
elmikojvrbanac: swagger is a format for describing a rest api, the basic layer is a json file that has all the routes, methods, and can even contain extended information about accepted schemas, example info, etc.20:58
elmikojvrbanac: i find it interesting from the perspective of some projects that have attempted to do things where they define the swagger first and then autogenerate docs and even code skeletons from that definition20:59
*** bdpayne has joined #openstack-barbican20:59
elmikojvrbanac: so, you could in theory centralize the api to a document and then have all the bits descend from the master doc20:59
elmikoi don't see us using all that functionality in openstack, but it's nice as a neutral format to describe an api20:59
elmikothere are also some cool server apps that create the type of output you might see at the api-ref site for example21:00
elmikojvrbanac: is an example of something you could produce from a swagger definition21:01
elmikoplus, imo, the json is way more human readable than the xml of wadl21:02
elmikoi only know a little about api-blueprint, but i imagine it's similar21:02
jvrbanacelmiko, interesting. I'll need to look more into this stuff.21:02
jvrbanacI know I looked into api-blueprint a bit, but I haven't sat down and worked with any of this stuff21:03
jvrbanacelmiko, the markdown-based idea of api-blueprint sounded kind of appealing.21:04
elmikojvrbanac: here's the spec i've mainly been working from
elmikojvrbanac: that does sound intersting, markdown++21:05
elmikoi put together the barbican impl of swagger back in december because the api-wg was curious about it21:06
elmikoi also did one for sahara, but flask is waaaaay easier to analyze than pecan21:07
elmikojvrbanac: if you're curious, is what i put together for openstack stuff21:07
elmikomy thought was it would be way easier to maintain a site like api-ref if we used swagger and could dump all the local html/css/js that is carried with the current wadl stuff21:08
jvrbanacelmiko, yeah21:08
elmikoi've talked with the doc folks though, and i think they like the idea of not completely autogenerated stuff. which i can get behind, you still want a set of eyes making sure everything is kosher21:09
elmikobut there's no need for those eyes to bleed from xml overdose ;)21:09
jvrbanacelmiko, exactly21:11
*** jorge_munoz has quit IRC21:12
*** gitorres has joined #openstack-barbican21:12
*** gitorres has quit IRC21:12
rm_work[16:06:13]  <elmiko>i also did one for sahara, but flask is waaaaay easier to analyze than pecan21:15
* rm_work disappears21:16
openstackgerritJohn Vrbanac proposed openstack/barbican: Starting to rework docs around the secret resource
jvrbanacredrobot, ^^21:20
jvrbanacredrobot, does that look better?21:20
redrobotjvrbanac yes.  I like it! :)21:20
jvrbanacredrobot, awesome ok! thx21:21
*** gitorres has joined #openstack-barbican21:24
*** igueths has joined #openstack-barbican21:26
*** jorge_munoz has joined #openstack-barbican21:27
*** gitorres has joined #openstack-barbican21:31
*** gitorres has quit IRC21:32
openstackgerritThomas Dinkjian proposed openstack/python-barbicanclient: Second set of negative secrets tests.
*** SheenaG has quit IRC22:08
*** jamielennox|away is now known as jamielennox22:09
*** xaeth is now known as xaeth_afk22:12
openstackgerritMerged openstack/barbican: Fix string substitution in exception messages missing the s
*** igueths has quit IRC22:28
*** jkf has quit IRC22:44
*** arunkant_ has quit IRC23:04
*** dimtruck is now known as zz_dimtruck23:05
*** chlong has joined #openstack-barbican23:27
*** bdpayne has quit IRC23:30
*** kebray has quit IRC23:43

Generated by 2.14.0 by Marius Gedminas - find it at!