Thursday, 2018-03-22

*** salv-orlando has joined #openstack-sdks00:27
*** salv-orlando has quit IRC00:32
*** chenyb4 has joined #openstack-sdks01:04
adriantmordred: I don't know if we can make it nicer, but having to pass in the service itself when calling actions on the resources is a pain:01:07
adrianttest_project.assign_role_to_user(conn.identity, test_user, member_role)01:07
adriantnot to mention, having those kind of calls directly on the proxy might be nicer, since I don't always want to have to query for the project first if I know what the ids are of all the resources I want to act on are.01:08
adriant*also on the proxy01:08
*** salv-orlando has joined #openstack-sdks01:27
*** harlowja has quit IRC01:31
*** salv-orlando has quit IRC01:32
*** dayou has quit IRC02:18
*** salv-orlando has joined #openstack-sdks02:28
*** ldl has joined #openstack-sdks02:31
*** salv-orlando has quit IRC02:34
*** edmondsw has joined #openstack-sdks02:42
adriantmordred: an error this time. Trying to see what response the sdk gives me when trying to create a keystone user that already exists: HttpException: HttpException: Unknown error02:42
adriantmordred: looks like an easy one. We aren't catching 409s in the raise from exception function02:44
adriantI'll throw up a patch for that02:45
*** edmondsw has quit IRC02:49
*** shu-mutow has joined #openstack-sdks03:05
* adriant is very very confused03:05
adriantmordred: is evaluating as false... expect I KNOW there is a response with status 409...03:06
adriantif I print response.text in that very place I get: "{"error": {"message": "Conflict occurred attempting to store user - Duplicate Entry", "code": 409, "title": "Conflict"}}"03:07
adriantso I have no clue what magic is making a response of type 409 eval to false...03:08
*** dave-mccowan has quit IRC03:11
*** annp has joined #openstack-sdks03:14
adriantwait... wtf response always evals to false03:16
adriantit is false when it isn't a safe status code... I'm willing to bet that's it03:16
* adriant goes to read requests code03:16
adriantmordred: we're using it wrong:
* adriant makes another patch03:17
*** salv-orlando has joined #openstack-sdks03:30
openstackgerritAdrian Turjak proposed openstack/python-openstacksdk master: Add 409 ConflictException
*** harlowja has joined #openstack-sdks03:32
*** salv-orlando has quit IRC03:35
openstackgerritAdrian Turjak proposed openstack/python-openstacksdk master: Fix response always being False
*** d0ugal_ has joined #openstack-sdks03:39
*** d0ugal has quit IRC03:39
* adriant finds all the best bugs03:53
*** harlowja has quit IRC03:54
*** edmondsw has joined #openstack-sdks04:05
*** edmondsw has quit IRC04:10
openstackgerritAdrian Turjak proposed openstack/python-openstacksdk master: Fix response always being False
*** salv-orlando has joined #openstack-sdks04:31
*** salv-orlando has quit IRC04:36
*** gildub has quit IRC04:51
*** gildub has joined #openstack-sdks05:05
*** salv-orlando has joined #openstack-sdks05:29
*** ldl has quit IRC05:33
*** edmondsw has joined #openstack-sdks05:54
*** edmondsw has quit IRC05:59
*** e0ne has joined #openstack-sdks06:15
*** e0ne has quit IRC06:33
*** e0ne has joined #openstack-sdks06:46
*** e0ne has quit IRC06:52
*** d0ugal_ has quit IRC07:14
*** d0ugal has joined #openstack-sdks07:14
*** d0ugal has quit IRC07:14
*** d0ugal has joined #openstack-sdks07:14
*** zacksh has quit IRC07:25
*** Hazelesque has quit IRC07:25
*** slaweq has quit IRC07:25
*** Anticimex has quit IRC07:25
*** tobixen has quit IRC07:25
*** RuiChen has quit IRC07:25
*** Kvisle has quit IRC07:25
*** timss has quit IRC07:25
*** slaweq has joined #openstack-sdks07:26
*** Hazelesque has joined #openstack-sdks07:26
*** Kvisle has joined #openstack-sdks07:26
*** Hazelesque has quit IRC07:26
*** Hazelesque has joined #openstack-sdks07:26
*** RuiChen has joined #openstack-sdks07:27
*** zacksh has joined #openstack-sdks07:27
*** timss has joined #openstack-sdks07:27
*** tobixen has joined #openstack-sdks07:28
*** salv-orlando has quit IRC07:33
*** markvoelker has quit IRC07:35
*** ralonsoh has joined #openstack-sdks07:39
*** edmondsw has joined #openstack-sdks07:42
*** edmondsw has quit IRC07:46
*** GreenBlood has joined #openstack-sdks08:23
openstackgerritAndreas Jaeger proposed openstack/keystoneauth master: Remove and align with constraints consumption
*** GreenBlood has quit IRC08:31
*** salv-orlando has joined #openstack-sdks08:34
*** markvoelker has joined #openstack-sdks08:34
*** yolanda_ is now known as yolanda08:35
*** salv-orlando has quit IRC08:40
*** GreenBlood has joined #openstack-sdks08:47
*** jpena|off is now known as jpena08:48
*** jpich has joined #openstack-sdks08:54
*** e0ne has joined #openstack-sdks08:56
*** e0ne has quit IRC09:07
*** e0ne has joined #openstack-sdks09:11
*** e0ne has quit IRC09:15
*** gkadam has joined #openstack-sdks09:19
*** gildub has quit IRC09:29
*** edmondsw has joined #openstack-sdks09:30
*** shu-mutow is now known as shu-mutow-AWAY09:32
*** edmondsw has quit IRC09:35
*** salv-orlando has joined #openstack-sdks09:35
*** salv-orlando has quit IRC09:38
*** salv-orlando has joined #openstack-sdks09:38
*** dayou has joined #openstack-sdks09:38
openstackgerritAdrian Turjak proposed openstack/python-openstacksdk master: Add 409 ConflictException
*** dayou has quit IRC09:44
*** dayou has joined #openstack-sdks09:53
*** ioggstream has joined #openstack-sdks10:01
openstackgerritchenyb4 proposed openstack/python-openstacksdk master: Fix api request in header miss microversions
openstackgerritOpenStack Proposal Bot proposed openstack/python-openstackclient master: Updated from global requirements
openstackgerritchenyb4 proposed openstack/python-openstacksdk master: Fix api request in header miss microversions
*** e0ne has joined #openstack-sdks10:10
*** chenyb4 has quit IRC10:14
*** dtantsur|afk is now known as dtantsur10:36
mordredadriant: you do find all the best bugs!10:47
*** kmy has quit IRC10:51
*** kmy has joined #openstack-sdks10:52
openstackgerritMonty Taylor proposed openstack/keystoneauth master: Remove and align with constraints consumption
openstackgerritMerged openstack-infra/shade master: Updated from global requirements
*** chenyb4 has joined #openstack-sdks11:44
openstackgerritMerged openstack/python-openstacksdk master: Fix TypeError in case of FloatingIP add and remove
*** chenyb4 has quit IRC12:03
openstackgerritMonty Taylor proposed openstack/python-openstacksdk master: Fix response always being False
mordredrods, slaweq: if you have a sec, and are both nice bugfixes12:08
slaweqmordred: both +W12:09
mordreddoes anyone know if chenyb4  / Yuanbin.Chen is on IRC?12:10
adriantmordred: that requests response is not none fix is gonna fail zuul after that rebase12:11
adriantsince... the second is not none... doesn't catch the empty dict12:12
mordredadriant: oh yeah?12:12
mordredoh - fun12:12
adriantand somewhere someone is passing in a empty dict12:12
adriantso we may have to be silly and do if type(response) is Response:12:12
mordredadriant: like, somewhere something is passing an empty dict instead of an actual Response object?12:12
adriantmordred: looks like12:13
adriantthat's what the test failure looked like12:13
adriantI'm just about to run it locally and check again12:13
mordredadriant: have I mentioned recently that I really miss strong typing?12:13
adriantyeah... normally I don't hit these issues, but every now and again you wish we had it12:14
mordredadriant: cool - I'd love to find/fix the calling location if we can - but yeah, I can see maybe needing to do isinstance(type, Response) ... but that just seems so sad12:14
*** edmondsw has joined #openstack-sdks12:15
adriantisinstance! of course, silly me I use that one but always forget it12:15
mordredI have to remind myself of it all the time12:15
adriantis this and instance to use it? ;)12:15
* adriant fails at his own terrible pun12:15
adriantmordred: self.request_id = response.headers.get('x-openstack-request-id')"12:18
adriant    b"AttributeError: 'dict' object has no attribute 'headers'"12:18
adriantyeah, somewhere it's a dict12:18
openstackgerritMerged openstack/os-client-config master: pypy is not checked at gate
openstackgerritAdrian Turjak proposed openstack/python-openstacksdk master: Fix response always being False
openstackgerritAdrian Turjak proposed openstack/python-openstacksdk master: Fix response always being False
*** chenyb4 has joined #openstack-sdks12:27
adriantmordred: a better fix was found so (to try that pun again), this isn't an instance where we type check. :P12:28
*** e0ne has quit IRC12:31
mordredadriant: \o/12:33
*** cdent has joined #openstack-sdks12:34
*** markvoelker has quit IRC12:34
*** markvoelker has joined #openstack-sdks12:34
*** chenyb4 has quit IRC12:34
*** thrash|g0ne is now known as thrash12:43
*** e0ne has joined #openstack-sdks12:45
*** bobh has joined #openstack-sdks13:06
openstackgerritDoug Hellmann proposed openstack-infra/requestsexceptions master: add lower-constraints job
openstackgerritDoug Hellmann proposed openstack-infra/shade master: add lower-constraints job
openstackgerritMerged openstack/python-openstacksdk master: Add 409 ConflictException
*** dtantsur is now known as dtantsur|brb13:21
amotokidtroyer: regarding storyboard migration along with SDK in the dev list, do you have any specific concern? In general, I can take care of the migration if it happens.13:26
dhellmannmordred : what's up with requestsexceptions? I don't see any of the unit test jobs running against
dtroyeramotoki: the only thing I thought we should do would be a pass through the bugs and clean up obsolete ones, but IIRc all get migrated anyway so I'm not sure that is important13:28
dtroyeramotoki: if you want to handle the migration, that would be awesome, thanks!13:28
amotokidtroyer: honestly I don't know what kind of tools we actually have. On obsolete bugs, I think we can look thru them even after the migration.13:32
amotokidtroyer: tags in storyboard looks like global (from my limited experience), but it needs to be clarified.13:33
amotokiI only have experiences on storyboard around octavia migration.13:34
dtroyeramotoki: sure, that is more than I have with Storyboard :)13:34
amotokidtroyer: from my impressions so far, it seems we have no big concern on osc migration to storyboard.13:40
amotokidtroyer:  I can care things during migrations and discuss broader topics with mordred and/or infra team.13:40
amotokidtroyer: i will send a follow-up mail to the list13:40
dtroyeramotoki: sounds good, thanks13:42
mnaseri think the openstackclient gate is broken?13:42
mnaserosc-functional-devstack seems to always fail13:42
amotokimordred: in my understanding so far, we have a tool to support migration from launchpad to storyboard, right?13:43
mnaserit looks like creating a network segment now does not have a description of 'None' but an empty string13:44
amotokimordred: based on chats with dtroyer here, there is no big blocking issue to block OSC storyboard migration. the remaining thing is how we can automate it.13:45
dtroyermnaser: frickler chased that down yesterday to a neutron commit that is being reverted13:45
fricklerdtroyer: mnaser: yes and the revert is failing due to some stackviz issue, see current discussion in -qa13:47
dtroyermnaser: revert:
mnaserokay, awesome, thanks frickler and dtroyer :)13:49
amotokislaweq and me need to understand how we can prevent this kind of breakage outside of neutron. I haven't figured out the whole picture13:49
amotokislaweq: sorry for pinging you :(13:50
slaweqamotoki: no problem :)13:50
mnaseramotoki: have neutron run that devstack osc job in gate?13:52
mnaserassuming that job is fairly stable13:52
amotokimnaser: no, it is not part of neutron gate jobs.13:54
amotokimnaser: I am checking the relationship of osc-functional-devstack and neutron repo13:55
amotokibasically they are not related directly13:55
mnaserthis type of thing can help catch things .. aka "hi im neutron and is my change breaking osc" seems like a reasonable thing13:56
dtroyerfor a Neutron change like that to break OSC means it is an API breakage,   Using osc's functional tests as API tests is one way to catch that, it could also be done with stronger API testing in Neutron13:57
dtroyerosc's functional tests do not exercise the APIs as thoroughly as I would hope dedicated API tests would do.13:58
amotokias my neutron hat, I believe neutron and tempest teams expect this should be caught in tempest/neutron-tempest-plugin should catch it.14:06
amotokii will investigate more tomorrow14:06
*** gkadam has quit IRC14:09
SamYaplemordred: would a "think about using openstacksdk" warning be a good thing for shade?14:14
SamYapleran into someone who thought the shade/openstacksdk merger was going into shade instead of openstacksdk being the place for future dev14:14
*** dtantsur|brb is now known as dtantsur14:15
mordredamotoki, dtroyer: oh, yes - migration to storyboard basically involves just deciding to migrate - then infra can arrange actually doing the migration14:16
mordredamotoki, dtroyer: so if we want to migrate python-openstackclient, we can just add it to the list of repos to migrate when we migrate openstacksdk14:17
amotokimordred: sounds nice.14:17
mordredSamYaple: it might not be a bad idea14:17
amotokimordred: as osc team, I can follow up minor things on migration of OSC14:17
mordredamotoki: sweet!14:18
mordreddhellmann: requestsexceptions doesn't actually have any tests (it's a very small librar that just squelches 4 useless warnings)14:20
dhellmannmordred : ok, so we don't need a lower-constraints job for it?14:20
mordreddhellmann: that said - looking at that patch to see what's up14:20
dhellmannit has requirements, so...14:20
mordreddhellmann: it only has test-requirements - and that only includes hacking14:20
dhellmannok. I'll abandon that patch14:21
mordreddhellmann: cool. I love it when problems are solved by abandoning patches14:21
mordreddhellmann: I should remove the py35,py27 envs from tox.ini ...14:21
dhellmannmy script isn't fancy enough to figure out whether tests are needed; it just makes the change to the repo14:22
mordreddhellmann: well - I think that's the right choice for your script - things should in general have tests14:22
dhellmannhmm, pbr is failing the py27 tests14:22
dhellmannand shade is failing the new lower-constraints job, so that's fun
mordreddhellmann: you're giving me all sorts of fun and joy today14:23
mordreddhellmann: those are fun failures14:24
dhellmannNoMockAddress exceptions14:24
dhellmanndoes shade use a special test library for some of that? something that is maybe set too low?14:25
dhellmannoh, those run under python 3, fwiw14:25
mordreddhellmann: it uses requests_mock - and python3 should be fine14:25
mordreddhellmann: I'll pull locally and see if I can figure out which thing is breaking it14:25
dhellmannthank you14:25
mordreddhellmann: the lower-constraints.txt is intended to be managed locally, yeah?14:25
dhellmannthere is not currently a test to verify that the values there are not higher than what is listed in requirements txt14:26
mordreddhellmann: btw - your how-many-summits-have-I-been-to plugin is awesome14:26
dhellmannI'm waiting for someone to notice the off-by-one error in the math14:27
mordreddhellmann: I didn't look deeply enough to know if it allows expressing missed summits somewhere14:27
dhellmannfeature requests will be answered with commit privileges14:28
mordreddhellmann: :)14:28
dhellmannin fact I need to give cmurphy commit rights, since she sent a patch14:28
dhellmannI'm trying this new thing where I instigate something and let other people finish it14:31
*** toker_ has joined #openstack-sdks14:31
mordreddhellmann: that strategy has worked pretty well for me over the years14:32
mordreddhellmann: in fact, this last cycle I accomplished it by _saying_ I was going to do something and then doing no work on it so that cmurphy 'fixed' my buggy code - where the bug was "there is no code"14:32
toker_Hi guys, Im trying to debug a problem with octavia-dashboard, which makes a call to openstacksdk to find the public endpoint of octavia. I get an exception from the sdk like this 'SDKException: public endpoint for load_balancer service in regionOne region not found'... And I need to figure out why the sdk is giving me that answer,14:33
toker_totally new at this so ..14:33
mordredso while I used to consider it the clarkb strategy (implement something poorly so that it annoys clark and he fixes it) - I think it's the cmurphy strategy now :)14:33
dhellmannmordred : that's pretty darn zen14:33
mordredtoker_: I have a hunch it's going to be a _ vs - thing - I was just looking at an issue related to that this morning14:34
toker_Well, I tried changing the endpoint to _14:34
toker_bit then I got another error instead :/14:35
toker_service_type)   File "/usr/lib/python2.7/site-packages/openstack/", line 277, in _get_version_match     for link in version["links"]: KeyError: 'links'14:35
mordredtoker_: are these errors in a test job or just locally for you?14:35
toker_no these are errrors in our POC of our openstack setup14:36
toker_so, openstack loadbalancer list works. But when octavia-dashboard tries to get the endpoint through the sdk, it fails.14:36
cmurphymordred: it was a fairly severe bug14:37
toker_And Ive been stuck here for hours trying to figure out why :(14:37
mordredtoker_: kk. I'm making a quick test script to hand you ... I bet we can figure it out14:38
toker_Oh, thanks! That's really nice14:38
openstackgerritPavlo Shchelokovskyy proposed openstack/os-service-types master: Fix docs link in the README
mordredtoker_: <-- if you run that and paste the output into we can see what's going wrong.14:42
mordredtoker_: (it assumes you're using env vars - if you have a named cloud in a clouds.yaml file, you'll want to change cloud='envvars')14:43
johnsomtoker_ note that you are on an older version of OpenStack, not queens14:43
mordredfor reference - this is me running that script against the vexxhost public cloud:
toker_AttributeError: 'module' object has no attribute 'enable_logging' <--14:48
openstackgerritMonty Taylor proposed openstack-infra/shade master: add lower-constraints job
mordredtoker_: ok. that's an older version of sdk then -I'm guessing from the 0.9 range14:49
toker_we are using OSP 12 (Pike)14:49
mordrednod. and you're installing the octavia-dashboard from OSP 12 as well?14:50
toker_There is no official octavia-dashboard for OSP12. This is me trying to get the "master" working for OSP 12.14:52
toker_There is the connection to the SDK14:53
*** bobh has quit IRC14:57
mordredGOTCHA. so - this is a fun14:59
toker_Haha yes, very much so. 12 hours ago it was fun. Now its more like "god damn I just want this shit working"14:59
mordredtoker_: :)15:00
mordredtoker_: how did you install the octavia-dashboard code? pip? globally or into a virtualenv?15:00
mordredasking because I think running master octavia-dashboard with pike openstacksdk is going to be hard - but also wanting to keep troubleshooting paths to a minimum15:01
mordreddhellmann: should fix the shade l-c job15:01
mordreddhellmann: it was requests-mock - 1.1.0 is not sufficient. I'm guessing I should go send in a patch to bump the min on it15:01
toker_mordred: well, I *think* it should work, if just could get the integration to SDK working.15:02
toker_its installed by cloning the repo "python install" inside the container running horizon..15:03
toker_So basically, /usr/lib/python2.7/site-packages/octavia_dashboard/api/rest/ <- making changes to this file and restarting the container everytime I try a change.15:03
mordredtoker_: gotcha. SO - could you try using the python in that container to run that script I pasted?15:05
toker_Yes, same error as I posted before. Hm, the guys in #openstack-lbaas channel actually says I need the 0.9.19 SDK15:06
toker_they got it working with that version...15:06
mordredfor master of octavia-dashboard? that seems weird to me and like something we should track down - but if they can get you unbroken go for it :)15:07
toker_Well, I don't know.. Thats what one of the guys trying to help me said.15:08
toker_I would rather try to get it working with the current version of the SDK (then I dont need to try to upgrade the SDK that horizon may be depending on inside the container)15:09
toker_Is there something obvious with this code that makes it not return the endpoint for the loadbalancer ( I mean the error is pretty verbose, right ? "SDKException: public endpoint for load_balancer service in regionOne region not found")15:10
toker_I'm not sure updating the SDK would help with that error ?15:11
toker_I want to understand why the SDK cant find the public endpoint - however getting there seems hard15:13
*** dave-mccowan has joined #openstack-sdks15:15
toker_I also tried to revert one specific commit, , which supposedly could make it work with "an older sdk". But I still get the same error from the sdk that it cant find the public endpoint..15:18
mordredtoker_: right - so - there's nothing in that code that would cause an error - the heavy-lifting of discovery is done in keystoneauth15:20
mordredtoker_: which is why I was curious about the output of that script - basically hoping something jumps out at me in the rest log15:21
toker_Can one do a similar script but for the older sDK ?15:21
mordredso I think the things to verify are - which version of openstacksdk is installed in that container with octavia-dashboard15:22
toker_Ok, that is 0.9.1715:23
mordredcool. so - now we get to figure out how to debug the old sdk ... one sec15:24
toker_ok ok15:24
mordredtoker_: ok. well - if it makes you feel any better, I can confirm that sdk 0.9.19 does not find the load balancer service in the vexxhost public cloud either - with the same error message15:26
toker_oh :/15:27
toker_Hm, thats interesting15:27
*** bobh has joined #openstack-sdks15:28
*** bobh has quit IRC15:30
*** bobh has joined #openstack-sdks15:30
mordredtoker_: yah. well - all of the code to handle this was basically completely replaced in the 0.10 release15:31
mordredtoker_: but - I'm working on a patch for you ... since I can reproduce your issue here, it's a little easier15:31
toker_Oh, for the better then you mean ? It wasnt working properly before ?15:31
toker_Ok thats very nice of you15:31
mordredtoker_: ok. don't laugh ... <-- if you apply that to the files in your container - it should fix you15:39
mordredtoker_: it's obviously not a GENERAL fix, but the general fix involves upgrading more things15:40
mordredtoker_: part of the issue is that pike octavia doesn't have a working version discovery document ... I believe johnsom got that fixed this last cycle15:40
toker_laugh ? How could I laugh, you just fixed a problem I've been struggling with for over 10 hours !!15:42
toker_This is a very hacky setup at the moment, so adding more hacks to it doesn't really matter.15:42
toker_We are going to reinstall everything when OSP 13 is out anyway.15:43
toker_Omg, so simple fix!15:43
toker_Thank you15:43
toker_took you ten minutes to find, I've never would have found it :)15:43
toker_AttributeError: 'Proxy' object has no attribute 'listeners' <- stumbled on this now when trying to list listeners ...15:45
toker_But thats fine, dont really matter.15:46
toker_ <- maybe I need this15:49
toker_one guy said that in the lbaas-channel15:49
toker_hm dont think thats related -.15:51
*** fbaca has quit IRC15:53
mordredno - that shouldn't be related to the listeners issue - although in general that patch should be fine and safe to backport if you do need to get the list of pools on a health_monitor16:01
*** harlowja has joined #openstack-sdks16:04
*** harlowja has quit IRC16:09
openstackgerritDoug Hellmann proposed openstack-infra/shade master: add lower-constraints job
*** thrash is now known as thrash|biab16:15
dhellmannmordred : I updated your update to also change the test-requirements.txt file ^^16:25
*** bobh has quit IRC16:29
*** cdent has left #openstack-sdks16:46
*** thrash|biab is now known as thrash16:49
elmikoedleafe dtantsur , ready for review
* elmiko notices that cdent left before review, makes note on tally sheet17:04
dtantsurelmiko: one comment there, everything else LGTM17:07
dtantsurmaybe I just misunderstood the phrase though17:07
elmikono no, good spot17:07
elmikoi missed a word lol17:07
edleafeelmiko: LGTM17:12
mordreddhellmann: sweet17:18
*** dtantsur is now known as dtantsur|afk17:23
*** ioggstream has quit IRC17:31
*** jpich has quit IRC17:31
*** jroll has joined #openstack-sdks17:39
*** e0ne has quit IRC17:39
*** Anticimex has joined #openstack-sdks17:50
*** jpena is now known as jpena|off18:00
*** boris_42_ has joined #openstack-sdks18:13
*** harlowja has joined #openstack-sdks18:15
*** harlowja_ has joined #openstack-sdks18:17
*** harlowja has quit IRC18:19
*** dave-mccowan has quit IRC18:20
openstackgerritDoug Hellmann proposed openstack-infra/shade master: add lower-constraints job
openstackgerritDoug Hellmann proposed openstack/cliff master: add lower-constraints job
*** ralonsoh has quit IRC18:31
*** cdent has joined #openstack-sdks18:53
jrolldoes anyone have experience running nodepool or shade against clouds that put services on subpaths, e.g. https://mycloud/nova/v2? we're playing around with that and shade seems to follow the links returned from / and not seeing a path prefix of sorts in most project's config (nova has one, but it's host and all, not just the path)19:13
jrollor experience running a cloud like that, and could point to something I'm missing19:14
jrollconsidering starting a thread on the mailing list about making a consistent config between projects for that, if there isn't something obvious I'm missing here19:14
*** zer0c00l has joined #openstack-sdks19:18
cdentjroll: is the service catalog returning reasonable entries?19:19
cdentalso you might be interested in
mordredjroll: shade should totally work with suburls..19:22
mordredjroll: as long as the service-catalog and version discovery documents are correct19:22
mordredjroll: I'm not sure how to configure the services to return the correct thing in their version discovery docs...19:22
mordredjroll: but devstack is running with suburl layout these days and shade's functional tests work against it -so it's gotta be possible _somehow_19:23
jrollmordred: right, that last part is the catch19:23
jrolloh, devstack does that now? I can look at that19:23
mordrediirc it's the default deployment now19:23
mordredjroll: but the tl;dr is that shade uses version discovery (except for neutron and swift) - and treats the values in the version discovery documents as being *mostly* the source of truth19:24
jrollmordred: yeah, I was curious if we were doing something wrong with the service or with shade19:25
mordredjroll: we do discard whatever host is returned in the version docs - because that is FREQUENTLY wrong in many configs (services return their rfc1918 ip in the url they make - so we replace the version discovery host, scheme and port with the host, scheme and port from the catalog)19:25
mordredjroll: it's always possible we've got bugs, but shade is a pretty good litmus test for "is this service configured correctly" :)19:26
jrollmordred: indeed :)19:27
mordredjroll: (if you ever want WAY more information on this topic I'm happy to assault you - but I'd like consent first :) )19:27
jrollmordred: ha, I'll pass for now. I know I'm missing something obvious, I just don't know what19:29
openstackgerritBailey Johnson proposed openstack-infra/shade master: Adds Senlin support to shade
jrollto follow up, I dug in the code harder and turns out it comes from req.application_url here (at least in nova):
jrollso probably an issue with our apache config19:44
openstackgerritBailey Johnson proposed openstack-infra/shade master: Adds Senlin support to shade
*** boris_42_ has quit IRC20:41
*** cdent has quit IRC20:45
*** edmondsw has quit IRC20:56
*** edmondsw has joined #openstack-sdks20:56
*** edmondsw has quit IRC21:01
*** salv-orlando has quit IRC21:19
*** salv-orlando has joined #openstack-sdks21:19
*** salv-orlando has quit IRC21:24
*** salv-orlando has joined #openstack-sdks21:26
*** dayou has quit IRC22:42
*** dayou has joined #openstack-sdks22:43
*** gildub has joined #openstack-sdks22:43
*** edmondsw has joined #openstack-sdks22:46
*** dayou has quit IRC22:52
*** dayou has joined #openstack-sdks22:55
*** ccha has quit IRC23:21
*** e0ne has joined #openstack-sdks23:22
*** halbbalda has quit IRC23:44
*** halbbalda has joined #openstack-sdks23:45
*** calebb has quit IRC23:50
*** halbbalda has quit IRC23:51
*** calebb has joined #openstack-sdks23:52
*** e0ne_ has joined #openstack-sdks23:54
*** e0ne has quit IRC23:57

Generated by 2.15.3 by Marius Gedminas - find it at!