| jamielennox | i like the idea of having release notes in the source though | 00:00 |
|---|---|---|
| bknudson | me too, but I'd rather it was in docstrings. | 00:00 |
| jamielennox | though i would prefer it at python-keystoneclient/ChangeLog | 00:00 |
| *** sarob_ has joined #openstack-dev | 00:01 | |
| bknudson | I've already got one: 135018 Jun 19 11:20 ChangeLog | 00:01 |
| *** nati_uen_ has joined #openstack-dev | 00:02 | |
| *** kui has joined #openstack-dev | 00:03 | |
| jamielennox | oh, thats right, openstack autogenerates a changelog for everything | 00:03 |
| bknudson | jamielennox: docs get generated for keystone here... http://docs.openstack.org/developer/keystone/py-modindex.html | 00:03 |
| bknudson | obviously it's something we use lot. | 00:04 |
| jamielennox | there isn't even any reference to the v3 api | 00:04 |
| *** sarob has quit IRC | 00:04 | |
| bknudson | https://github.com/openstack/python-keystoneclient/blob/master/doc/source/using-api.rst | 00:04 |
| bknudson | has some v2 docs | 00:04 |
| *** nati_ueno has quit IRC | 00:05 | |
| jamielennox | well i guess: http://docs.openstack.org/developer/python-keystoneclient/py-modindex.html is the answer to what is public on keystoneclient | 00:05 |
| *** michchap has joined #openstack-dev | 00:05 | |
| bknudson | lots of undocumented methods | 00:06 |
| *** hemna is now known as hemnafk | 00:06 | |
| *** kui has quit IRC | 00:09 | |
| jamielennox | bknudson: i see your reference to python's warning library. I've never heard of that before - it seems like it'd be really useful. any idea why it's not more adopted? | 00:14 |
| bknudson | jamielennox: I haven't used it myself. Just came across it when looking for "python deprecated" | 00:15 |
| bknudson | looks like it's in 2.6, so should be safe to use. | 00:16 |
| *** senk has joined #openstack-dev | 00:16 | |
| jamielennox | i like the idea that you could run gate jobs at a different warning level so that we don't reference deprecations internally | 00:16 |
| bknudson | "New in version 2.1." -- that was easy to find! | 00:16 |
| ayoung | jamielennox, I know I had some client questions for you... | 00:22 |
| jamielennox | i +1ed the client trusts | 00:23 |
| *** vuil has quit IRC | 00:23 | |
| *** alexpilotti has quit IRC | 00:23 | |
| ayoung | jamielennox, cool. | 00:24 |
| jamielennox | ayoung: there's one place i'm not sure about - but it's more a general problem that a trusts one i think | 00:24 |
| ayoung | jamielennox, and the general problem with long parameter list... | 00:24 |
| ayoung | jamielennox, I want to make an object out of that huge recurreing param list | 00:24 |
| *** stanchan has joined #openstack-dev | 00:24 | |
| jamielennox | in process_token() on httpclient a number of things get set back to the client from the token - i don't think that trusts should be in there, but i'm not sure as to why the rest are done either | 00:24 |
| bknudson | ayoung: won't the object constructor be the same size as the dict constructor? | 00:25 |
| jamielennox | ayoung: so a lot of that will go away with auth plugins | 00:25 |
| ayoung | bknudson, yeah. I think we need to leave the constructor, but lets set all the values on a value object we pass around | 00:25 |
| *** jjmb has joined #openstack-dev | 00:25 | |
| *** gimps_ has joined #openstack-dev | 00:26 | |
| ayoung | jamielennox, OK, I'll hold off until auth plugins are done. My fingers are itching to excise that list, though.... | 00:26 |
| jamielennox | i was having an attempt at some of that, but there is the issue that the caching uses the same list | 00:26 |
| bknudson | maybe it would allow use of **kwargs and then the list of keyword args would only be in the __init__ | 00:26 |
| ayoung | so cache the whole thing | 00:26 |
| jamielennox | no the kwargs in that list get used for the KEY for the cache | 00:26 |
| ayoung | jamielennox, then that value object would be the key | 00:27 |
| jamielennox | i really like the ideas he has in apiclient - his approach scares me | 00:27 |
| ayoung | jamielennox, the other thing was that you should ruthlessly excise any references to uuids you see in the client | 00:27 |
| *** dina_belova has joined #openstack-dev | 00:28 | |
| ayoung | he's young and enthusiastic....just needs some wisdom to guide him | 00:28 |
| *** changbl has joined #openstack-dev | 00:28 | |
| jamielennox | i thought i was young and entusiastic? | 00:28 |
| *** stanchan has quit IRC | 00:28 | |
| ayoung | jamielennox, you are salt of the earth, steady and reliable | 00:29 |
| ayoung | um...did I just insult you? | 00:30 |
| ayoung | heh | 00:30 |
| jamielennox | this is getting oddly personal in such a public forum.... | 00:30 |
| ayoung | jamielennox, it would probably be inappropriate to mention then that I know where you work. | 00:31 |
| jamielennox | and yet... | 00:31 |
| ayoung | jamielennox, so he got the review broken down | 00:31 |
| ayoung | that is a promising start | 00:31 |
| *** shri has quit IRC | 00:32 | |
| jamielennox | the last week or so have made me really paranoid about compatibility in the client - i'm still reviewing it because i can't see anywhere he's broken but it's still a major change in one go | 00:32 |
| *** dina_belova has quit IRC | 00:33 | |
| bknudson | ayoung: good with this? https://review.openstack.org/#/c/41332/ | 00:33 |
| *** otherwiseguy has joined #openstack-dev | 00:34 | |
| ayoung | jamielennox, yeah. No real way to tell. People could have used anything. Still we should know from the other projects in OS what is most likely to get used. | 00:34 |
| ayoung | bknudson, looking | 00:34 |
| morganfainberg | bknudson: just an FYI you probably want to hold of on approving things until zuul is situated | 00:34 |
| ayoung | bknudson, yeah that looks ok. | 00:35 |
| morganfainberg | bknudson: they're working hard over in infra to get it all working | 00:35 |
| bknudson | morganfainberg: it takes a long time to restart zuul! | 00:35 |
| morganfainberg | bknudson: they've had to re-start it like 4 times or something similar today | 00:35 |
| morganfainberg | and they're still chasing a bug down. | 00:35 |
| morganfainberg | (last i saw) | 00:35 |
| bknudson | they must have crossed the streams. | 00:35 |
| morganfainberg | hehe | 00:36 |
| *** chmouel has quit IRC | 00:36 | |
| ayoung | I think Zuul ran into the gatekeeper | 00:36 |
| morganfainberg | ^^ | 00:36 |
| uvirtbot | morganfainberg: Error: "^" is not a valid command. | 00:36 |
| morganfainberg | oh damn it, i forgot the bot does that | 00:36 |
| *** iccha has quit IRC | 00:36 | |
| *** westmaas has quit IRC | 00:36 | |
| gyee | ^^ | 00:36 |
| ayoung | or was Zuul the gatekeeper, and Zuul ran into the Keymaster? | 00:36 |
| *** iccha has joined #openstack-dev | 00:36 | |
| *** westmaas has joined #openstack-dev | 00:36 | |
| ayoung | ^^ | 00:36 |
| morganfainberg | hm, i think that second one is right ayoung | 00:36 |
| uvirtbot | ayoung: Error: "^" is not a valid command. | 00:36 |
| *** chmouel has joined #openstack-dev | 00:37 | |
| *** GheRivero has quit IRC | 00:37 | |
| *** dtroyer has quit IRC | 00:37 | |
| *** juice has quit IRC | 00:37 | |
| *** GheRivero has joined #openstack-dev | 00:37 | |
| ayoung | That would explain why they call is Zuul....all is clear | 00:37 |
| *** dtroyer has joined #openstack-dev | 00:37 | |
| *** jpeeler has quit IRC | 00:37 | |
| morganfainberg | yes yes | 00:37 |
| *** jpeeler has joined #openstack-dev | 00:38 | |
| *** juice has joined #openstack-dev | 00:38 | |
| gyee | ayoung, https://review.openstack.org/#/c/33118/24/etc/keystone-paste.ini | 00:39 |
| *** jjmb1 has joined #openstack-dev | 00:39 | |
| gyee | question is should extension be part of the pipeline | 00:39 |
| gyee | don't think we ever have a clear guideline on that | 00:39 |
| *** jecarey has joined #openstack-dev | 00:39 | |
| ayoung | I really need to get some more eyes on https://review.openstack.org/#/c/29059/ | 00:39 |
| bknudson | should we have tests/contrib/ | 00:40 |
| *** jjmb has quit IRC | 00:40 | |
| *** Mandell has quit IRC | 00:40 | |
| ayoung | gyee, dolphm made a good point: the extension is an extension to the API. So it might extend the v3 or the v2 api, and that is cool. So, yeah, you would add the extension to a specific pipeline/ | 00:40 |
| fabio | morganfainberg: hi | 00:40 |
| morganfainberg | ayoung: i'd offer, but my puppet skillset is limited. | 00:40 |
| *** melwitt has quit IRC | 00:40 | |
| bknudson | ayoung: why copying stuff out of keystone? https://review.openstack.org/#/c/29059/12/files/httpd/keystone.py | 00:41 |
| morganfainberg | fabio: hey there | 00:41 |
| bknudson | you mention that it's not shipped? | 00:41 |
| ayoung | bknudson, cuz not all of the packagers picked up that file yet | 00:41 |
| fabio | morganfainberg: I did all the changes for EP Filter | 00:41 |
| ayoung | we'll drop it once they do | 00:41 |
| fabio | unfortunately there are issues with Jenkins | 00:41 |
| gyee | bknudson, good point! | 00:41 |
| morganfainberg | fabio: yeah i know. | 00:41 |
| fabio | it did not report anything since this morning | 00:41 |
| morganfainberg | fabio: infra team is working on it | 00:41 |
| ayoung | bknudson, we've fixed it in Fedora, and now waiting for the fix to be blessed for EPEL/RDO/RHOS | 00:42 |
| fabio | morganfainberg: can you please review the changes nevertheless? | 00:42 |
| jamielennox | ayoung: whoa, recommended? | 00:42 |
| fabio | it will speed-up things | 00:42 |
| ayoung | >_< | 00:42 |
| ayoung | (>_<) | 00:42 |
| ayoung | Looks like Cartman! | 00:42 |
| morganfainberg | fabio: i will do so, i've been out a good chunk of the day, and still catching up | 00:43 |
| gyee | ayoung, what's that? | 00:43 |
| morganfainberg | ayoung: I totally agree. | 00:43 |
| jamielennox | i think i know 3 emoticons, happy, sad, cheeky - for everything else words are usefule | 00:43 |
| ayoung | gyee, see the above review | 00:43 |
| fabio | morganfainberg: thanks, I appreciate it | 00:43 |
| ayoung | \m/ (>_<) \m/ | 00:44 |
| *** herndon has quit IRC | 00:45 | |
| *** pmathews has quit IRC | 00:45 | |
| gyee | ah, I see dolphm's comment now :) | 00:46 |
| * dolphm is stalking IRC | 00:46 | |
| gyee | dolphm, I am back to review mode | 00:46 |
| ayoung | jamielennox, https://en.wikipedia.org/wiki/List_of_emoticons | 00:47 |
| ayoung | you shall not lack for emoticons | 00:47 |
| dolphm | fabio: is this achieving anything beyond security by obscurity? | 00:47 |
| gyee | (_?_) | 00:47 |
| dolphm | "if don't advertise the service to the naughty user then he won't be able to exploit it" | 00:48 |
| *** maheshp has joined #openstack-dev | 00:48 | |
| gyee | security by obscurity is no security | 00:48 |
| ayoung | fabio, dolphm which review is this? | 00:49 |
| dolphm | ayoung: OS-EP-FILTER in general | 00:49 |
| ayoung | dolphm, I never thought of that as a security patch. Is it being spun as one? | 00:50 |
| *** spzala has joined #openstack-dev | 00:50 | |
| *** sarob_ has quit IRC | 00:51 | |
| *** alexb_ has quit IRC | 00:51 | |
| *** jhesketh__ has joined #openstack-dev | 00:52 | |
| fabio | 33118 | 00:52 |
| fabio | #link https://review.openstack.org/#/c/31118 | 00:52 |
| *** gimps_ has quit IRC | 00:52 | |
| *** schwicht has quit IRC | 00:52 | |
| fabio | #link https://review.openstack.org/#/c/33118 | 00:52 |
| dolphm | ayoung: i think it's just being spun as a solution without a stated problem | 00:53 |
| *** jhesketh has quit IRC | 00:53 | |
| *** jhesketh has joined #openstack-dev | 00:53 | |
| dolphm | "Currently Keystone returns all endpoints in the service catalog, regardless whether users have access to them or not. This is neither necessary nor efficient." | 00:53 |
| gyee | dolphm, its not a security patch. Service themselves still have to authorized on the operation | 00:54 |
| *** nati_uen_ has quit IRC | 00:55 | |
| dolphm | gyee: right, it's redundant with rbac in other services and ?nocatalog already solves for both "necessity" and "efficiency" with extreme prejudice | 00:55 |
| ayoung | dolphm, so my take on this was that the endpoint list filtering was to allow us to move toward more tailored endpoints, but the security will be enforced using the auth_token_middleware. Ideally, we would enforce that the endpoint consuming the token is listed in the token | 00:55 |
| gyee | its just no point of returning the endpoints user have no business of access them in the first place | 00:55 |
| fabio | dolphm, this is to avoid to expose endpoints that are also irrelevant | 00:55 |
| ayoung | so, if the endpoint is not in the token, auth_token middleware can start saying "nope, not going to accept it" | 00:55 |
| ayoung | that is beyond the scope of this change here, though | 00:56 |
| fabio | ayoung, I think it can enable that, but out of scope | 00:56 |
| *** gimps_ has joined #openstack-dev | 00:56 | |
| gyee | this is also a starting point for endpoint scoping | 00:56 |
| ayoung | fabio, yeah, Icehouse | 00:56 |
| morganfainberg | as a nice side benefit, it reduces the size of the token if you filter some endpoints (PKI) | 00:57 |
| ayoung | yep | 00:57 |
| dolphm | morganfainberg: but ?nocatalog does it better | 00:57 |
| gyee | morganfainberg, damn straight! | 00:57 |
| gyee | dolphm, it saves us an extra call | 00:57 |
| *** nati_ueno has joined #openstack-dev | 00:57 | |
| morganfainberg | dolphm: i think both have a usecase. | 00:57 |
| *** noslzzp has joined #openstack-dev | 00:58 | |
| fabio | dolphm, I think with EP filter you have more granularity control, is an improvement over on/off | 00:58 |
| dolphm | gyee: if GET /v3/catalog doesn't vary per user, and i don't think it should, then clients can actually cache it | 00:58 |
| dolphm | fabio: but you're not controlling anything but a superficiality, is all | 00:58 |
| gyee | dolphm, GET /v3/catalog will vary per user | 00:59 |
| fabio | dolphm, it is a filter :-) | 00:59 |
| dolphm | gyee: if by "will" you mean "should not", then sure ;) | 00:59 |
| gyee | I mean it vary per user based on token scope | 01:00 |
| ayoung | dolphm, so, there is the question of: if I get back multiple endpoints, which one do I use. I think that saying a user will only ever get back one endpoint per service based on some server side filter is OK, so long as we don't depend on that for security. | 01:00 |
| fabio | dolphm, I think this is for domain admins that want to control the endpoints that are presented during token request | 01:00 |
| ayoung | It is a "handling scale" mechanism | 01:00 |
| fabio | so then only the relevant endpoints are presented | 01:01 |
| dolphm | ayoung: what does this have to do with scale? | 01:01 |
| gyee | ayoung, that's what this is for, if user only have access to region-a, there | 01:01 |
| *** gimps_ has quit IRC | 01:01 | |
| ayoung | dolphm, in large deployments, people are starting to direct people toward specific endpoints | 01:01 |
| ayoung | quality of service and such | 01:01 |
| gyee | s no point of returning endpoints for region-b | 01:01 |
| gyee | ayoung, precisely | 01:01 |
| morganfainberg | gyee: yeah, that was kind of my thought | 01:01 |
| ayoung | so reducing the number of endpoints returned will affect the load on the various servers | 01:01 |
| openstackstatus | NOTICE: Zuul is offline for troubleshooting | 01:02 |
| *** ChanServ changes topic to "Zuul is offline for troubleshooting" | 01:02 | |
| fabio | dolphm, also, for instance, return only internal endpoints | 01:02 |
| ayoung | Damn Ghostbusters.... | 01:02 |
| morganfainberg | ayoung: there is no zuul =/ | 01:02 |
| dolphm | fabio: why not just give the client a cacheable link to a filtered endpoint list? | 01:02 |
| ayoung | No, no, there is no spoon, only zuul | 01:02 |
| morganfainberg | only dana? :P | 01:02 |
| dolphm | fabio: and if your use case is domains, why on earth are you mapping projects to endpoints? | 01:02 |
| ayoung | dolphm, that is an interesting solution.... | 01:03 |
| *** gimps_ has joined #openstack-dev | 01:03 | |
| ayoung | We'd have to have pre-canned catalogs. | 01:03 |
| fabio | dolphm, I was talking about domain admins not domain | 01:03 |
| gyee | use case is for projects | 01:03 |
| dolphm | GET /v3/domains/{domain_id}/catalog?region=north&interface=internal | 01:04 |
| ayoung | gyee, I could see it per domain as well. If a domain maps to a company, then that domain gets a set of endpoints at a certain quality of service | 01:04 |
| *** hailiang has joined #openstack-dev | 01:04 | |
| morganfainberg | actually i think domains fit the argument (scalability) better overall | 01:04 |
| ayoung | really should be both | 01:04 |
| ayoung | morganfainberg, yes and no | 01:04 |
| *** xmltok has quit IRC | 01:04 | |
| morganfainberg | but if you support one, supporting the other isn't a huge step. | 01:05 |
| morganfainberg | and makes it more flexable | 01:05 |
| morganfainberg | ayoung: wait, do domains span regions? i forget | 01:05 |
| ayoung | morganfainberg, for a large company, you might make different projects again for quality of service. All in the same domain | 01:05 |
| ayoung | morganfainberg, yes they do | 01:05 |
| morganfainberg | ok, nix my last statement | 01:05 |
| *** briancurtin has joined #openstack-dev | 01:05 | |
| morganfainberg | i do like the cachable link idea | 01:05 |
| ayoung | ^H^H^H^H^H^H^H^H^H^H^H^H^H | 01:05 |
| uvirtbot | ayoung: Error: "H^H^H^H^H^H^H^H^H^H^H^H^H" is not a valid command. | 01:05 |
| fabio | dolphm, the extension was approved in m2 and now I just made the implemenation | 01:05 |
| morganfainberg | LOL | 01:05 |
| fabio | so the API docs are there | 01:06 |
| *** feleouet has quit IRC | 01:06 | |
| *** reed has quit IRC | 01:06 | |
| dolphm | fabio: i'm not arguing against having the current impl | 01:06 |
| fabio | dolphm, ah, ok. I just thought I lost all this time ... :-) | 01:06 |
| gyee | dolphm, my bad, yeah, domain-endpoint is there too | 01:07 |
| *** nati_ueno has quit IRC | 01:07 | |
| *** markmcclain has quit IRC | 01:07 | |
| dolphm | fabio: i think it's an interesting idea, although i'm tempteted to argue that it's shortsighted, and i'm mostly curious to see either where we take it next or what we replace it with later on | 01:07 |
| *** briancurtin has quit IRC | 01:08 | |
| gyee | dolphm, its all disposable code :) | 01:08 |
| dolphm | gyee: ++ | 01:08 |
| gyee | but I like endpoint-scoping in IceHouse | 01:08 |
| fabio | dolphm, I have some ideas around both endpoint grouping and catalog search | 01:08 |
| ayoung | dolphm, what do you think it will look like down the road>? | 01:08 |
| gyee | reduce security risk, reduce catalog size | 01:08 |
| fabio | maybe we can discuss these at the next summit | 01:08 |
| gyee | helll, reduce catalog altogether | 01:08 |
| fabio | I will put the bp later on | 01:08 |
| dolphm | fabio: someone from rax did a bunch of work on grouped related endpoints as well | 01:09 |
| morganfainberg | i think endpoint scoping would be a great topic. | 01:09 |
| gyee | morganfainberg, see y'all in HK | 01:09 |
| *** feleouet has joined #openstack-dev | 01:09 | |
| gyee | lets jam | 01:09 |
| morganfainberg | gyee: heck ya! | 01:09 |
| fabio | dolphm, I'll have a look at their proposals | 01:09 |
| *** gimps_ has quit IRC | 01:10 | |
| *** Shaan7 has quit IRC | 01:11 | |
| dolphm | #topic keystoneclient v0.3.2 | 01:11 |
| dolphm | we have two changes gating that i think are critical to doing a v0.3.2 release... | 01:11 |
| dolphm | https://review.openstack.org/#/c/39899/ | 01:11 |
| dolphm | https://review.openstack.org/#/c/43020/2 | 01:11 |
| dolphm | are there any others i should wait on before tagging 0.3.2? | 01:11 |
| morganfainberg | hrm. | 01:11 |
| dolphm | (or bugs) | 01:12 |
| gyee | is there a patch for pluggable auth | 01:12 |
| gyee | would like that one as well | 01:12 |
| holms | people need your help cause #openstack is dead | 01:12 |
| dolphm | gyee: yes, but i don't think it's quite ready yet | 01:12 |
| holms | keystone: error: unrecognized arguments: --service-id | 01:12 |
| holms | in here it's fine http://docs.openstack.org/trunk/openstack-compute/install/yum/content/keystone-service-endpoint-create.html | 01:12 |
| morganfainberg | dolphm: i think those are the important two | 01:13 |
| holms | can't find any argument in "usage" options of keystone endpoint-create | 01:13 |
| dolphm | holms: what does `keystone help service-create` show? | 01:13 |
| dolphm | gyee: that might be a good reason to have a 0.4.0 | 01:13 |
| gyee | dolphm, yeah absolutely | 01:14 |
| holms | --service_id <service-id> | 01:14 |
| dolphm | holms: sorry, endpoint-create, not service-create | 01:14 |
| holms | k | 01:14 |
| holms | why there's --service_id in there | 01:14 |
| holms | not dash like in docs | 01:14 |
| *** alexb_ has joined #openstack-dev | 01:14 | |
| dolphm | holms: what's your keystone --version? | 01:14 |
| holms | i'm on grizzly | 01:14 |
| holms | no suck option =/ | 01:15 |
| holms | such* | 01:15 |
| dolphm | holms: then you're on a *very* old client that predates grizzly | 01:15 |
| holms | k.. why this in wheezy apt-get = | 01:15 |
| dolphm | holms: pip install -U python-keystoneclient | 01:15 |
| holms | sorry for bothering my bad | 01:15 |
| morganfainberg | dolphm: i'll also be up late tonight so i can babysit those reviews (make sure they get through gate) if zuul isn't back up soon. | 01:16 |
| dolphm | morganfainberg: danke | 01:17 |
| *** salv-orlando has quit IRC | 01:17 | |
| *** fifieldt_ has joined #openstack-dev | 01:20 | |
| dolphm | someone remind me to do keystoneclient releases more often | 01:21 |
| gyee | now! :) | 01:21 |
| * dolphm Unable to comply, gating in progress. | 01:21 | |
| *** maheshp has quit IRC | 01:21 | |
| dolphm | this is going to be a fat one https://launchpad.net/python-keystoneclient/+milestone/0.3.2 | 01:21 |
| *** freedomhui has joined #openstack-dev | 01:23 | |
| *** rods has quit IRC | 01:23 | |
| *** salv-orlando has joined #openstack-dev | 01:23 | |
| morganfainberg | dolphm: *sets cron to ping dolphm, release keystoneclient* | 01:25 |
| holms | dolphm: yeah my bad.. i've used default packages from wheezy. | 01:26 |
| holms | thanks | 01:26 |
| *** maheshp has joined #openstack-dev | 01:28 | |
| *** comay has quit IRC | 01:28 | |
| *** dina_belova has joined #openstack-dev | 01:28 | |
| *** Tross has quit IRC | 01:29 | |
| anteaya | morganfainberg: zuul is in the process of being restarted | 01:31 |
| *** briancurtin has joined #openstack-dev | 01:31 | |
| anteaya | *crosses fingers* | 01:31 |
| *** maheshp has quit IRC | 01:31 | |
| * roaet cries for zuul | 01:31 | |
| roaet | There is no dana? :( | 01:31 |
| anteaya | not so far | 01:32 |
| morganfainberg | anteaya: heeh | 01:32 |
| morganfainberg | anteaya: thanks | 01:32 |
| anteaya | welcome | 01:32 |
| roaet | I've been trying to get my review to pass crazy jenkins for some time. so I guess everyone is having the same problem | 01:32 |
| anteaya | it will be a bit before the changes are in I think though, we are trying to do a staggered start | 01:32 |
| morganfainberg | roaet: yeah, it's been an issue the last couple days, infra team is working hard on it. | 01:32 |
| anteaya | so it doesn't bring us down | 01:32 |
| *** dina_belova has quit IRC | 01:33 | |
| morganfainberg | anteaya: am i going to need to issue rechecks / reverifies? | 01:33 |
| morganfainberg | anteaya: or should it pick everything back up. (doesn't bother me in either case) | 01:33 |
| anteaya | morganfainberg: hopefully not | 01:33 |
| morganfainberg | ok, i'll keep my eyes on it. | 01:34 |
| anteaya | we tryed to save the queue | 01:34 |
| anteaya | yes, good plan | 01:34 |
| morganfainberg | i think keystone client has 2 or 3 things waiting on gate, and keystone a couple as well. | 01:34 |
| *** dolphm has quit IRC | 01:34 | |
| morganfainberg | anteaya: again, thanks :) | 01:34 |
| anteaya | np | 01:34 |
| *** danwent has quit IRC | 01:34 | |
| anteaya | I understand everyone wants to know and no one wants to ask | 01:35 |
| morganfainberg | anteaya: hehe, i lurk in #openstack-infra, so i don't usually need to ask | 01:35 |
| anteaya | thought I would share what smidgen of good news I had | 01:35 |
| anteaya | cool | 01:35 |
| *** gyee has quit IRC | 01:35 | |
| *** dolphm has joined #openstack-dev | 01:36 | |
| roaet | anteaya: Someone other than me has been putting in recheck no bug in my review at least 3 times (in addition to my two times), will that queue up or be ignored? | 01:36 |
| anteaya | roaet: hopefully it will queue up | 01:37 |
| roaet | will it queue up 5 times? | 01:37 |
| anteaya | no | 01:37 |
| roaet | That sounds terrible if it does. oh ok. | 01:37 |
| anteaya | go here: http://status.openstack.org/zuul/ | 01:37 |
| anteaya | when zuul is restarted the queues will start to form | 01:37 |
| *** gongysh has quit IRC | 01:37 | |
| anteaya | first gate then post then check | 01:37 |
| anteaya | there will be a staggered start | 01:38 |
| *** IniMe has joined #openstack-dev | 01:38 | |
| roaet | anteaya: are you on the infra team? | 01:38 |
| anteaya | if the check queue is longer than 150 and you don't see your patch number, come ask me | 01:38 |
| openstackstatus | NOTICE: Zuul is running again | 01:38 |
| *** ChanServ changes topic to "OpenStack development || Support is in #openstack" | 01:38 | |
| anteaya | yes infra team | 01:38 |
| roaet | anteaya: I really appreciate the work you're putting into it. | 01:38 |
| anteaya | mostly I am answering what questions I can | 01:39 |
| anteaya | to free up jeblair and clarkb who are doing the lion's share | 01:39 |
| anteaya | and thanks | 01:39 |
| roaet | it's good enough. sure you got a lot of freaked out devs around you. | 01:39 |
| anteaya | nice to have your support | 01:39 |
| anteaya | calm devs grinding their teeth | 01:39 |
| roaet | jblair is the guy doing the other rechecks, so I guess I know who it is now. | 01:39 |
| anteaya | yes that would be him | 01:40 |
| *** erkules_ has joined #openstack-dev | 01:40 | |
| anteaya | everyone is really supportive but we know the pressure everyone is under to get their work done | 01:40 |
| *** rwsu has quit IRC | 01:40 | |
| *** martine has joined #openstack-dev | 01:40 | |
| *** dolphm has quit IRC | 01:41 | |
| *** martine is now known as Guest81016 | 01:41 | |
| roaet | anteaya: is most of the infrastructure that zuul runs on donated by companies, or are they org owned? | 01:41 |
| roaet | I wonder if I could get my company to donate more (any if they haven't)... | 01:41 |
| anteaya | gerrit, jenkins, zuul and everything else infra run on donated space | 01:41 |
| anteaya | Rackspace and HP donate the space | 01:41 |
| roaet | Ah. Guess it'd be more. | 01:42 |
| roaet | <- racker | 01:42 |
| anteaya | ah okay | 01:42 |
| *** erkules has quit IRC | 01:42 | |
| roaet | I'm pretty new around here but trying to focus on upstream issues. | 01:42 |
| anteaya | not sure who at rax takes care of things, but from our side, if you want to play too, the more the merrier | 01:42 |
| anteaya | great | 01:42 |
| anteaya | you are asking the good questions | 01:42 |
| anteaya | mordred would know the answers to your questions better than I and he just left for burning man | 01:43 |
| roaet | burning man! always wanted to go to that. | 01:43 |
| roaet | interesting bunch, openstack has. | 01:43 |
| anteaya | if you want to know more about infra, you can start here: http://ci.openstack.org/ | 01:44 |
| anteaya | we are an eclectic bunch that is for sure | 01:44 |
| *** erkules_ has quit IRC | 01:44 | |
| roaet | anteaya: do you know if we have anyone from rax helping with that project directly? | 01:44 |
| *** bknudson has quit IRC | 01:44 | |
| roaet | I'm sure I can just 'google' it. | 01:45 |
| anteaya | roaet: I know there are rax folks answering our tickets when we file them, but I am not sure who they are | 01:45 |
| * roaet does so. | 01:45 | |
| anteaya | but there isn't anyone from rax that I know of contributing to infra code on a regular basis, that has made themselves known to me | 01:45 |
| roaet | Just wondering if there were people on the team so they could maybe add some push for more support. But i'm probably overstepping my bounds. | 01:45 |
| anteaya | there may be folks contributed to infra code from rax, but maybe I just don't know they are from rax | 01:46 |
| *** erkules_ has joined #openstack-dev | 01:46 | |
| roaet | I'm primarily working with neutron, but I'll look into ci project to see if I can help at all. | 01:46 |
| anteaya | overstepping your bounds is a great fit for infra | 01:46 |
| roaet | I really like devops, and infra seems kind of like it. | 01:46 |
| anteaya | roaet: glad to have you | 01:46 |
| anteaya | it is it for sure | 01:46 |
| *** radsy has joined #openstack-dev | 01:47 | |
| morganfainberg | anteaya: is everything happy? if so, i'm going to go issue some re-verifies because i don't see some pending gate jobs in the queue | 01:47 |
| morganfainberg | might have been "approved" when zuul was down. | 01:47 |
| morganfainberg | or some such. | 01:47 |
| anteaya | zuul is back up, queues are being repopulated | 01:47 |
| *** dhellmann is now known as dhellmann_ | 01:48 | |
| anteaya | 19 in gate, 7 in check | 01:48 |
| *** gongysh has joined #openstack-dev | 01:48 | |
| anteaya | they will take a while to get the whole saved queue back in | 01:48 |
| morganfainberg | anteaya: so queues should be in a stable state, not being "restored"? | 01:48 |
| anteaya | s/they/it | 01:48 |
| morganfainberg | oh oh | 01:48 |
| morganfainberg | i get it | 01:48 |
| morganfainberg | derp. | 01:48 |
| anteaya | no worries | 01:48 |
| morganfainberg | brain, lacking coffee | 01:48 |
| anteaya | yup | 01:48 |
| morganfainberg | i'll check iagain in ~30-40 mins. | 01:48 |
| *** vkmc has quit IRC | 01:48 | |
| anteaya | sounds like a good plan | 01:48 |
| anteaya | hopefully jeblair squished the zuul bug and it will stay up | 01:49 |
| morganfainberg | seriously, that one sounded nasty | 01:49 |
| morganfainberg | and squirrely on top of it | 01:49 |
| anteaya | yes | 01:49 |
| *** Daisy has joined #openstack-dev | 01:49 | |
| morganfainberg | add in git isssues and it makes for a sad "near-feature-freeze" | 01:49 |
| anteaya | yes plus it wasn't our only issue | 01:49 |
| anteaya | and a java somewhere that segfaults | 01:49 |
| anteaya | we haven't found that one yet | 01:50 |
| morganfainberg | when it rains it pours? | 01:50 |
| anteaya | that's about it | 01:50 |
| *** Tross has joined #openstack-dev | 01:51 | |
| *** sthaha has joined #openstack-dev | 01:51 | |
| *** sthaha has quit IRC | 01:51 | |
| *** sthaha has joined #openstack-dev | 01:51 | |
| morganfainberg | java segfault.. that's a specail one | 01:51 |
| lifeless | morganfainberg: heh, sadly not | 01:52 |
| Daisy | Hello. Who knows how to generate PoT in Horizon ? I cannot find any guidance. | 01:52 |
| morganfainberg | lifeless: =/ | 01:52 |
| *** xchu has joined #openstack-dev | 01:52 | |
| morganfainberg | lifeless: at least my python interpreter isn't segfaulting (yet) | 01:52 |
| roaet | anteaya: the check queue maxes at 8 right? Is there a way to know what else is in the queue outside of those 8? | 01:54 |
| roaet | Just checking if my change needs to be 'rechecked'. | 01:54 |
| anteaya | roaet: no it doesn't max at 8 | 01:54 |
| holms | we don't have wiki for hawana do we..? | 01:54 |
| anteaya | the gate queue will populate first, it has priority | 01:55 |
| anteaya | I am expecting to see 60+ patches in the gate once it has finished populating | 01:55 |
| anteaya | then any other nodes will go to the check queue | 01:55 |
| roaet | Is there a way to know that my change is going to be picked up? It should, but I mean, is there a way to know? | 01:55 |
| anteaya | the check queue was sitting around 150+ for most of the day | 01:55 |
| anteaya | yes, wait until the check queue is longer than 150 | 01:56 |
| anteaya | if it is and you don't see your patch, ping me | 01:56 |
| roaet | Got it, thank you. | 01:56 |
| anteaya | it will take some time | 01:56 |
| anteaya | I don't expect it to be fully populated for close to an hour, and it might take longer | 01:57 |
| *** ljjjustin has joined #openstack-dev | 01:57 | |
| anteaya | np | 01:57 |
| roaet | I did a quick search on the ci docs, but couldn't find this particular piece of info... What does it mean when a check (like python26) is LOST? | 01:57 |
| roaet | Is it just the jenkins job went crazy? | 01:57 |
| anteaya | no | 01:57 |
| *** otherwiseguy has quit IRC | 01:57 | |
| anteaya | that is happening right now because when we restart gearman (who works in the background) one of the threads dies | 01:58 |
| anteaya | and slaves that should be deleted are orphaned but not deleted | 01:58 |
| anteaya | so test run on them but the results don't get logged | 01:58 |
| anteaya | hence LOST logs | 01:58 |
| roaet | Alright. So it's an error in the infrastructure of some sort. | 01:58 |
| anteaya | we are working on that one too | 01:59 |
| *** d34dh0r53 has joined #openstack-dev | 01:59 | |
| anteaya | yes, we just noticed it this morning | 01:59 |
| roaet | I've been running into LOST every now and then for the past week or so. | 01:59 |
| anteaya | we are trying to track it down and fix | 01:59 |
| roaet | Should I have notified infra crew? | 01:59 |
| anteaya | after we get past the hump | 02:00 |
| anteaya | when you see a LOST come in and ask about it | 02:00 |
| anteaya | but give us a few days, because we couldn't get to it now anyway | 02:00 |
| roaet | Excellent. I will do so. Good to know that it's something I can help with in the future. | 02:00 |
| anteaya | thank you | 02:00 |
| *** d34dh0r53 has quit IRC | 02:01 | |
| anteaya | if you want to look at gearman: https://git.openstack.org/cgit/openstack-infra/gearman-plugin/tree/ this is where we think some of LOST is coming from | 02:01 |
| *** yaguang has joined #openstack-dev | 02:02 | |
| roaet | Is it just a jenkins plugin? *shot in the dark* | 02:02 |
| *** ljjjustin has quit IRC | 02:03 | |
| anteaya | roaet: take a look at this timestamp: 2013-08-21T14:42:12 in this log: http://eavesdrop.openstack.org/irclogs/%23openstack-infra/%23openstack-infra.2013-08-21.log | 02:03 |
| anteaya | that is what we have so far | 02:03 |
| anteaya | gearman is a jenkins plugin, yes | 02:03 |
| *** egallen has quit IRC | 02:03 | |
| anteaya | was that the answer to your question? | 02:03 |
| anteaya | I might have mis-understood | 02:03 |
| roaet | Yes. got it | 02:03 |
| anteaya | grand | 02:04 |
| roaet | I've been working with the jenkins folk on one of their plugins. I'll see if I can help. | 02:04 |
| anteaya | w00t | 02:04 |
| anteaya | zaro is our gearman guy | 02:04 |
| gongysh | /opt/stack/keystone$ sudo python setup.py develop | 02:04 |
| anteaya | if you have any questions about it, he is your go to man | 02:04 |
| gongysh | error: invalid command 'develop' | 02:04 |
| anteaya | and yes if you have a patch, that would be great, thanks roaet | 02:05 |
| gongysh | Who can tell why the python setup.py develop does not work on my machine? | 02:05 |
| lifeless | gongysh: a better thing is 'pip install -e .' | 02:06 |
| lifeless | gongysh: it replaces python setup.py develop. | 02:06 |
| gongysh | lifeless: sudo python sestup.py develop will install a egg.link into system for my current project, such as keystone | 02:07 |
| gongysh | lifeless: how to run 'pip install -e' to get the same target? | 02:08 |
| lifeless | 'pip install -e .' - the . is important. | 02:11 |
| gongysh | lifeless: it seems it also calls the setup.py develop: | 02:13 |
| gongysh | Installing collected packages: keystone | 02:13 |
| gongysh | Running setup.py develop for keystone | 02:13 |
| gongysh | /usr/local/lib/python2.7/dist-packages/d2to1/core.py:67: UserWarning: Unknown distribution option: 'install_requires' | 02:13 |
| gongysh | warnings.warn(msg) | 02:13 |
| gongysh | /usr/local/lib/python2.7/dist-packages/d2to1/core.py:67: UserWarning: Unknown distribution option: 'include_package_data' | 02:13 |
| gongysh | warnings.warn(msg) | 02:13 |
| gongysh | /usr/local/lib/python2.7/dist-packages/d2to1/core.py:67: UserWarning: Unknown distribution option: 'tests_require' | 02:13 |
| gongysh | warnings.warn(msg) | 02:14 |
| gongysh | usage: -c [global_opts] cmd1 [cmd1_opts] [cmd2 [cmd2_opts] ...] | 02:14 |
| gongysh | or: -c --help [cmd1 cmd2 ...] | 02:14 |
| gongysh | or: -c --help-commands | 02:14 |
| gongysh | or: -c cmd --help | 02:14 |
| gongysh | 02:14 | |
| gongysh | error: invalid command 'develop' | 02:14 |
| gongysh | Complete output from command /usr/bin/python -c "import setuptools; __file__='/opt/stack/keystone/setup.py'; exec(compile(open(__file__).read().replace('\r\n', '\n'), __file__, 'exec'))" develop --no-deps: | 02:14 |
| gongysh | /usr/local/lib/python2.7/dist-packages/d2to1/core.py:67: UserWarning: Unknown distribution option: 'install_requires' | 02:14 |
| gongysh | warnings.warn(msg) | 02:14 |
| gongysh | /usr/local/lib/python2.7/dist-packages/d2to1/core.py:67: UserWarning: Unknown distribution option: 'include_package_data' | 02:14 |
| gongysh | warnings.warn(msg) | 02:14 |
| gongysh | /usr/local/lib/python2.7/dist-packages/d2to1/core.py:67: UserWarning: Unknown distribution option: 'tests_require' | 02:14 |
| gongysh | warnings.warn(msg) | 02:14 |
| lifeless | gongysh: pastebin, please! | 02:14 |
| gongysh | ok | 02:15 |
| lifeless | I'm not sure whats going on there. I thought we'd stopped using d2to1? Is this a grizzly branch? | 02:15 |
| gongysh | http://paste.openstack.org/show/44824/ | 02:16 |
| *** Guest81016 has quit IRC | 02:16 | |
| gongysh | the master branch, | 02:16 |
| gongysh | I am working on a bug and want to have a running to test it before git review to trunk. | 02:17 |
| *** Tross has quit IRC | 02:20 | |
| jamielennox | gongysh: hmm i've seen that | 02:22 |
| jamielennox | i think it's to do with setuptools and distribute | 02:22 |
| jamielennox | try pip install --upgrade distribute | 02:22 |
| gongysh | try... | 02:22 |
| *** martine_ has joined #openstack-dev | 02:23 | |
| jamielennox | lifeless: keystone still uses d2to1, should we not be? | 02:24 |
| *** rwsu has joined #openstack-dev | 02:25 | |
| lifeless | jamielennox: I have likely gotten the wrong end of the stick from Monty :) | 02:25 |
| jamielennox | lifeless: i've no idea, it wouldn't be the first time we've completely missed a trend | 02:26 |
| anteaya | morganfainberg: if your patch was queue more than 4 hours ago, you will have to add the recheck/reverify comment yourself, but please give us maybe another hour to get up to speed? | 02:27 |
| anteaya | if it was less than 4 hours jeblair feels he should either have it saved or is commenting manually on patches | 02:27 |
| morganfainberg | anteaya: there are a couple keystoneclient ones that need to gate before we can cut a release. so i'm just babysitting those | 02:28 |
| anteaya | ah good plan | 02:28 |
| morganfainberg | i am fair certain those have been churning for more than a couple hours. | 02:28 |
| anteaya | okay | 02:28 |
| anteaya | well pop them in then | 02:28 |
| anteaya | let's get them cooking | 02:28 |
| morganfainberg | yeah the rest can wait until tomorrow or late late tonight | 02:28 |
| anteaya | okay thanks | 02:29 |
| anteaya | appreciate holding off on ones you can hold off on | 02:29 |
| morganfainberg | if they can get done by tomorrow sometime, it'll be good. | 02:29 |
| *** dina_belova has joined #openstack-dev | 02:29 | |
| morganfainberg | the important ones. | 02:29 |
| anteaya | let's hope zuul holds up | 02:29 |
| morganfainberg | yep. | 02:29 |
| anteaya | let me know when you see them in the queue | 02:29 |
| anteaya | 66 in the gate queue right now | 02:30 |
| anteaya | do you see them in there? | 02:30 |
| morganfainberg | nope | 02:30 |
| morganfainberg | 43020 | 02:30 |
| morganfainberg | going to issue reverify | 02:30 |
| anteaya | okay, good plan | 02:30 |
| anteaya | two of them? | 02:30 |
| morganfainberg | looking for the other one now. | 02:31 |
| anteaya | k | 02:31 |
| gongysh | jamielennox: I can run python setup.py develop, but I cannot install d2to1 now. I have rm -rf it before since I thought it is the d2tol's problem. | 02:31 |
| gongysh | jamielennox: http://paste.openstack.org/show/44825/ | 02:31 |
| morganfainberg | anteaya: and don't see 39899 either | 02:32 |
| morganfainberg | going to reverify | 02:32 |
| morganfainberg | wait. hrm | 02:32 |
| *** novas0x2a|laptop has quit IRC | 02:32 | |
| morganfainberg | zuul posted 15 minutes ago on that one | 02:32 |
| morganfainberg | weird. | 02:32 |
| morganfainberg | but i don't see it in the queue | 02:32 |
| anteaya | huh | 02:32 |
| anteaya | maybe check with jeblair in -infra | 02:33 |
| morganfainberg | yeah will do. | 02:33 |
| anteaya | don't want to put you to the back of the line if zuul already has it | 02:33 |
| morganfainberg | well, if i am at the back of line because of it | 02:33 |
| morganfainberg | thats fine | 02:33 |
| *** dina_belova has quit IRC | 02:33 | |
| morganfainberg | i jsut want to make sure it's in the line :) | 02:33 |
| anteaya | yes | 02:34 |
| anteaya | don't want to pull it out of the gate though and reset tests for the patches behind it | 02:34 |
| *** ljjjustin has joined #openstack-dev | 02:35 | |
| *** gordc has joined #openstack-dev | 02:35 | |
| *** shang has joined #openstack-dev | 02:35 | |
| *** jasdeepH has quit IRC | 02:36 | |
| *** stevemar has joined #openstack-dev | 02:36 | |
| *** rcleere has joined #openstack-dev | 02:36 | |
| *** ftcjeff has joined #openstack-dev | 02:38 | |
| ayoung | morganfainberg, I was having that same problem earlier. Zuul isn't picking everything up. | 02:40 |
| ayoung | Let's just assume this week is going to suck for merges | 02:40 |
| morganfainberg | ayoung: they are 70 gerrit events behind atm | 02:40 |
| morganfainberg | talking with the guys over in the infra channel | 02:40 |
| morganfainberg | it's slow atm | 02:40 |
| morganfainberg | and git issues abound. | 02:40 |
| morganfainberg | so, it's been rough. | 02:41 |
| morganfainberg | (and likely will be) | 02:41 |
| morganfainberg | now the important question, when is the project named Vinz Clortho going to be made? and what function will it have? | 02:42 |
| morganfainberg | in the openstack world. | 02:42 |
| ayoung | morganfainberg, it should be part of Barbican, of course. | 02:43 |
| morganfainberg | ayoung: you don't think they'd consider renaming the project do you? | 02:43 |
| ayoung | The real deal is that we need to rename OpenStack to Gozer and then it all works. | 02:44 |
| morganfainberg | YES! | 02:44 |
| ayoung | morganfainberg, they like that whole castle thing | 02:44 |
| ayoung | We can call Keystone Vinz if you like. | 02:44 |
| morganfainberg | eh, that works… or maybe we just call the KDS extension vinz? | 02:44 |
| ayoung | Actually, if we spin simo 's KDS off to its own project... | 02:44 |
| ayoung | so glad we did feature freeze early. Next time, feature freeze in Icehouse 1. | 02:45 |
| morganfainberg | oh, i wanted to ask you, rather than just fixing, the deletes in the KDS spec should return 204 right? i wasn't sure if there was a reason to return 200 instead (or i'd have just pushed a fixed changeset up) | 02:46 |
| ayoung | morganfainberg, yeah, 204 makes sense there | 02:47 |
| ayoung | that is the pattern elsewhere as well | 02:47 |
| morganfainberg | ok | 02:48 |
| ayoung | good catch | 02:48 |
| morganfainberg | i'll toss that fix up there so we can merge that in before too long. | 02:48 |
| *** asalkeld is now known as asalkeld_lunch | 02:48 | |
| ayoung | morganfainberg, yeah, that one will go in once we fork for Icehouse development. I actually am on the hook to write some tests for it, the way dolph did for trusts | 02:48 |
| morganfainberg | ayoung: ah. | 02:50 |
| *** jecarey has quit IRC | 02:50 | |
| morganfainberg | ayoung: ok the 2 fixes that need to get in for the keystoneclient release are showing in the queue | 02:52 |
| ayoung | morganfainberg, when you import a keypair into horizon, do you know what format it expects? Won't seem to take id_rsa,pub | 02:54 |
| ayoung | I guess I can use the nova api.... | 02:55 |
| morganfainberg | ayoung: .pem format it looks like | 02:56 |
| morganfainberg | maybe | 02:56 |
| morganfainberg | let me check | 02:56 |
| morganfainberg | ayoung: it seemed to work just fine for me locally, using the whole id_rsa.pub file | 02:57 |
| morganfainberg | oh. hrm this is a grizzly install, maybe different in h? | 02:57 |
| *** markmcclain has joined #openstack-dev | 03:00 | |
| jamielennox | gongysh: try the same thing with d2to1 pip install --upgrade d2to1 | 03:02 |
| jamielennox | there might be a --force or something as well | 03:02 |
| jamielennox | otherwise i recommend just dumping the venv and starrt again | 03:03 |
| gongysh | jamielennox: trying | 03:04 |
| *** yuan has quit IRC | 03:04 | |
| gongysh | jamielennox: no luck to run 'pip install d2to1' http://paste.openstack.org/show/44826/ | 03:07 |
| gongysh | jamielennox: I cannot find the --force for pip install | 03:07 |
| *** shang has quit IRC | 03:08 | |
| jamielennox | gongysh: did you install all this into a virtualevn? | 03:09 |
| gongysh | jamielennox: no, I am not using virtuaenv. | 03:09 |
| jamielennox | gongysh: ah - you definetly should | 03:11 |
| *** gongysh has quit IRC | 03:11 | |
| jamielennox | umm | 03:11 |
| jamielennox | i really don't know enough about the packaging process, i just knew i had that error previously | 03:12 |
| *** pmathews has joined #openstack-dev | 03:12 | |
| jamielennox | can you update setuptools as well? either by pip or easy install? | 03:13 |
| *** blamar has quit IRC | 03:17 | |
| *** gmurphy has quit IRC | 03:23 | |
| *** gmurphy has joined #openstack-dev | 03:25 | |
| *** gmurphy has quit IRC | 03:27 | |
| *** gmurphy has joined #openstack-dev | 03:27 | |
| *** Tross has joined #openstack-dev | 03:28 | |
| *** kushal has joined #openstack-dev | 03:28 | |
| *** kushal has joined #openstack-dev | 03:29 | |
| *** dina_belova has joined #openstack-dev | 03:29 | |
| *** stevemar has quit IRC | 03:30 | |
| *** jfriedly has quit IRC | 03:32 | |
| *** dina_belova has quit IRC | 03:34 | |
| *** noslzzp has quit IRC | 03:37 | |
| *** feleouet has quit IRC | 03:40 | |
| *** yaguang has quit IRC | 03:42 | |
| *** nati_ueno has joined #openstack-dev | 03:42 | |
| *** dims has quit IRC | 03:43 | |
| *** feleouet has joined #openstack-dev | 03:44 | |
| *** nati_ueno has quit IRC | 03:44 | |
| *** Mandell has joined #openstack-dev | 03:45 | |
| *** yaguang has joined #openstack-dev | 03:45 | |
| *** gongysh has joined #openstack-dev | 03:48 | |
| *** Shaan7 has joined #openstack-dev | 03:48 | |
| *** devoid has joined #openstack-dev | 03:48 | |
| *** devoid has quit IRC | 03:48 | |
| *** asalkeld_lunch is now known as asalkeld | 03:52 | |
| *** HenryG_ has joined #openstack-dev | 03:54 | |
| *** dstufft_ has joined #openstack-dev | 03:54 | |
| *** cyeoh has quit IRC | 03:54 | |
| *** lexinator has joined #openstack-dev | 03:54 | |
| *** soren has quit IRC | 03:54 | |
| *** dstufft has quit IRC | 03:55 | |
| *** soren has joined #openstack-dev | 03:55 | |
| *** cyeoh has joined #openstack-dev | 03:55 | |
| *** stevemar has joined #openstack-dev | 03:56 | |
| *** HenryG has quit IRC | 03:57 | |
| keekz | does each openstack review need a blueprint or bug attached to it? | 03:58 |
| lifeless | no | 03:59 |
| lifeless | but if they are working on a blueprint or bug they should include a link | 03:59 |
| keekz | this one i'm looking at is not, but it never got jenkins review | 04:00 |
| *** anteaya has quit IRC | 04:03 | |
| *** eharney has joined #openstack-dev | 04:03 | |
| *** danwent has joined #openstack-dev | 04:07 | |
| *** basha has joined #openstack-dev | 04:10 | |
| lifeless | keekz: do you mean the check jobs? Right now zuul is swamped. | 04:12 |
| keekz | yeah that's probably what was up | 04:13 |
| keekz | actually smokestack just ran and +1'd it | 04:13 |
| keekz | like 5 hours later :P | 04:13 |
| *** alexb_ has quit IRC | 04:13 | |
| *** basha has quit IRC | 04:15 | |
| *** phschwartz has quit IRC | 04:19 | |
| *** phschwartz has joined #openstack-dev | 04:19 | |
| *** basha has joined #openstack-dev | 04:21 | |
| *** vladan has quit IRC | 04:22 | |
| *** basha has quit IRC | 04:22 | |
| *** vladan has joined #openstack-dev | 04:23 | |
| *** dklyle has joined #openstack-dev | 04:23 | |
| *** dtroyer has quit IRC | 04:23 | |
| *** sballe_ has quit IRC | 04:23 | |
| *** dtroyer has joined #openstack-dev | 04:23 | |
| *** retr0h has quit IRC | 04:23 | |
| *** david-lyle has quit IRC | 04:23 | |
| *** _cerberus_ has quit IRC | 04:23 | |
| *** samalba has quit IRC | 04:23 | |
| *** comstud has quit IRC | 04:23 | |
| *** mberwanger has joined #openstack-dev | 04:24 | |
| *** retr0h has joined #openstack-dev | 04:25 | |
| *** retr0h has quit IRC | 04:25 | |
| *** retr0h has joined #openstack-dev | 04:25 | |
| *** comstud has joined #openstack-dev | 04:25 | |
| *** _cerberus_ has joined #openstack-dev | 04:26 | |
| *** samalba has joined #openstack-dev | 04:27 | |
| *** jhesketh__ has quit IRC | 04:28 | |
| *** gordc has quit IRC | 04:29 | |
| *** jhesketh_ has joined #openstack-dev | 04:29 | |
| *** dina_belova has joined #openstack-dev | 04:30 | |
| *** hemna_ has quit IRC | 04:30 | |
| *** rcleere has quit IRC | 04:32 | |
| *** ljjjustin has quit IRC | 04:33 | |
| *** markmcclain has quit IRC | 04:33 | |
| *** markmcclain has joined #openstack-dev | 04:33 | |
| *** ljjjustin has joined #openstack-dev | 04:33 | |
| *** basha has joined #openstack-dev | 04:34 | |
| *** radsy has quit IRC | 04:34 | |
| *** dina_belova has quit IRC | 04:35 | |
| *** boris-42 has joined #openstack-dev | 04:36 | |
| *** eharney has quit IRC | 04:37 | |
| *** spzala has quit IRC | 04:39 | |
| *** jhesketh_ has quit IRC | 04:40 | |
| *** jhesketh_ has joined #openstack-dev | 04:40 | |
| *** briancurtin has quit IRC | 04:45 | |
| *** dstufft_ is now known as dstufft | 04:46 | |
| enikanorov | what did happen with jenkins? | 04:47 |
| *** networkstatic has quit IRC | 04:48 | |
| clarkb | there have been a few issues related to the increased load ahead of the proposal freezes | 04:48 |
| *** ftcjeff has quit IRC | 04:49 | |
| clarkb | zuul fell over a few times due to a bug that was recently fixed, some of the git caching fell behind putting a lot more stress on our git servers which slowed them and the things depending on git down. The fix for this is in progress | 04:50 |
| *** vartom4 has joined #openstack-dev | 04:51 | |
| *** networkstatic has joined #openstack-dev | 04:51 | |
| *** basha has quit IRC | 04:53 | |
| *** SergeyLukjanov has joined #openstack-dev | 04:53 | |
| *** Tross has quit IRC | 04:55 | |
| *** alexb_ has joined #openstack-dev | 04:55 | |
| *** IniMe has quit IRC | 04:55 | |
| *** nati_ueno has joined #openstack-dev | 04:56 | |
| *** mjfork has quit IRC | 04:58 | |
| *** mjfork has joined #openstack-dev | 04:59 | |
| *** martine_ has quit IRC | 05:00 | |
| *** stevemar has quit IRC | 05:02 | |
| *** danwent has quit IRC | 05:05 | |
| *** Tross has joined #openstack-dev | 05:07 | |
| *** Ryan_Lane has quit IRC | 05:10 | |
| *** erkules_ is now known as erkules | 05:14 | |
| *** mberwanger has quit IRC | 05:16 | |
| *** ydacheville has quit IRC | 05:18 | |
| *** bashok has joined #openstack-dev | 05:19 | |
| *** chandankumar has joined #openstack-dev | 05:21 | |
| *** primeministerp has quit IRC | 05:22 | |
| *** garyk has quit IRC | 05:24 | |
| *** primeministerp has joined #openstack-dev | 05:29 | |
| *** gongysh has quit IRC | 05:30 | |
| *** dina_belova has joined #openstack-dev | 05:30 | |
| *** arezadr has joined #openstack-dev | 05:31 | |
| *** jbresnah has quit IRC | 05:31 | |
| *** jbresnah has joined #openstack-dev | 05:32 | |
| *** sridevi has joined #openstack-dev | 05:32 | |
| *** mrunge has joined #openstack-dev | 05:34 | |
| *** dina_belova has quit IRC | 05:35 | |
| *** vuil has joined #openstack-dev | 05:35 | |
| *** gongysh has joined #openstack-dev | 05:35 | |
| *** chandankumar has quit IRC | 05:35 | |
| *** Mandell has quit IRC | 05:37 | |
| *** chandankumar has joined #openstack-dev | 05:39 | |
| gongysh | nati_ueno: ping | 05:40 |
| *** egallen has joined #openstack-dev | 05:40 | |
| holms | everywhere i google how to create image people are using kvm-img and "kvm" commands. | 05:44 |
| holms | for kvm-img under RDO alternative is qemu-img.. and what about kvm.. then? | 05:44 |
| holms | all i have is qemu-io and qemu-img available | 05:44 |
| holms | virt-intall also unavailable.. | 05:44 |
| *** basha has joined #openstack-dev | 05:55 | |
| *** morganfainberg is now known as morganfainberg|a | 05:56 | |
| *** egallen has quit IRC | 05:58 | |
| *** dmakogon_ has joined #openstack-dev | 05:58 | |
| *** feleouet has quit IRC | 05:59 | |
| gongysh | henrynash: ping | 06:01 |
| *** devvesa has joined #openstack-dev | 06:01 | |
| *** feleouet has joined #openstack-dev | 06:02 | |
| *** lexinator has quit IRC | 06:03 | |
| *** salv-orlando has quit IRC | 06:05 | |
| henrynash | gongysh: hi | 06:05 |
| *** Mandell_ has joined #openstack-dev | 06:06 | |
| *** xchu has quit IRC | 06:07 | |
| *** jcoufal has joined #openstack-dev | 06:07 | |
| gongysh | henrynash: Do keystone have thread to clear expired token periodically? | 06:07 |
| *** garyk has joined #openstack-dev | 06:07 | |
| *** avishay has joined #openstack-dev | 06:07 | |
| henrynash | gongysh: you mean to actually delete these records? | 06:08 |
| gongysh | henrynash: Do -> Does | 06:08 |
| gongysh | henrynash: yes | 06:08 |
| henrynash | gongysh: so there isn't a thread, but I believe we added an api that could be called by such a thread to delete the records | 06:09 |
| henrynash | gongysh: let me see if I can find it for you | 06:09 |
| *** SlickNik has quit IRC | 06:09 | |
| *** SlickNik has joined #openstack-dev | 06:10 | |
| henrynash | gongysh: it's called: flush_expired_tokens | 06:11 |
| henrynash | gongysh: and you can use keystone-manage cli to call it | 06:12 |
| *** cfriesen has quit IRC | 06:13 | |
| gongysh | henrynash: so operator should add a cron like job to call keystone-manage token_flush, right? | 06:14 |
| henrynash | gongysh: exactly | 06:14 |
| *** morganfainberg|a is now known as morganfainberg | 06:14 | |
| *** SergeyLukjanov has quit IRC | 06:15 | |
| gongysh | henrynash: how about adding a periodic thread in keystone api service to do it? | 06:15 |
| henrynash | gongysh: we could do that…feel free to propose... | 06:16 |
| gongysh | henrynash: ok, I will try. thanks | 06:16 |
| henrynash | gongysh: propose a blueprint so others can comment first… | 06:17 |
| gongysh | henrynash: sure | 06:17 |
| gongysh | henrynash: thanks | 06:19 |
| henrynash | gongysh: your welcome | 06:19 |
| *** pmathews has quit IRC | 06:21 | |
| *** dguitarbite has quit IRC | 06:21 | |
| *** nshaikh has joined #openstack-dev | 06:21 | |
| *** sridevi has quit IRC | 06:23 | |
| *** noorul is now known as noorul-afk | 06:23 | |
| *** xchu has joined #openstack-dev | 06:24 | |
| *** MaxV has joined #openstack-dev | 06:27 | |
| *** basha has quit IRC | 06:29 | |
| *** HenryG_ has quit IRC | 06:30 | |
| *** mikal has quit IRC | 06:30 | |
| *** ffio has joined #openstack-dev | 06:31 | |
| *** dina_belova has joined #openstack-dev | 06:31 | |
| *** mikal has joined #openstack-dev | 06:32 | |
| *** lexinator has joined #openstack-dev | 06:33 | |
| *** dina_belova has quit IRC | 06:36 | |
| *** corXi has joined #openstack-dev | 06:37 | |
| *** AnilV4 has quit IRC | 06:38 | |
| *** MaxV has quit IRC | 06:40 | |
| *** lexinator has quit IRC | 06:40 | |
| *** zoresvit has quit IRC | 06:40 | |
| *** nayward has joined #openstack-dev | 06:42 | |
| *** zoresvit has joined #openstack-dev | 06:43 | |
| *** jprovazn has joined #openstack-dev | 06:45 | |
| *** vartom4 has quit IRC | 06:45 | |
| *** ndipanov has joined #openstack-dev | 06:47 | |
| *** aeperezt has quit IRC | 06:47 | |
| *** xga has joined #openstack-dev | 06:50 | |
| *** MaxV has joined #openstack-dev | 06:50 | |
| *** gongysh has quit IRC | 06:50 | |
| *** MaxV has quit IRC | 06:52 | |
| *** nati_ueno has quit IRC | 06:58 | |
| *** reidrac has joined #openstack-dev | 06:58 | |
| *** doron_afk has joined #openstack-dev | 07:01 | |
| *** hailiang has quit IRC | 07:03 | |
| *** shang has joined #openstack-dev | 07:04 | |
| *** ljjjustin has quit IRC | 07:05 | |
| *** SergeyLukjanov has joined #openstack-dev | 07:05 | |
| *** ljjjustin has joined #openstack-dev | 07:05 | |
| *** gongysh has joined #openstack-dev | 07:06 | |
| *** egallen has joined #openstack-dev | 07:09 | |
| *** odyssey4me4 has joined #openstack-dev | 07:14 | |
| *** fbo_away is now known as fbo | 07:16 | |
| *** basha has joined #openstack-dev | 07:17 | |
| *** markmcclain has quit IRC | 07:18 | |
| *** xqueralt-afk is now known as xqueralt | 07:19 | |
| *** yolanda has joined #openstack-dev | 07:21 | |
| *** sridevi has joined #openstack-dev | 07:24 | |
| *** hailiang has joined #openstack-dev | 07:24 | |
| yonglihe_ | ping Daniel Berrange | 07:25 |
| *** avishay has quit IRC | 07:26 | |
| *** mmagr has joined #openstack-dev | 07:27 | |
| *** sridevi has quit IRC | 07:28 | |
| *** vuil has quit IRC | 07:30 | |
| *** JordanP has joined #openstack-dev | 07:30 | |
| *** zoresvit has quit IRC | 07:32 | |
| *** dina_belova has joined #openstack-dev | 07:32 | |
| *** tkammer has joined #openstack-dev | 07:33 | |
| *** zoresvit has joined #openstack-dev | 07:35 | |
| *** dina_belova has quit IRC | 07:36 | |
| *** jpich has joined #openstack-dev | 07:37 | |
| *** lexinator has joined #openstack-dev | 07:38 | |
| *** mkerrin has joined #openstack-dev | 07:39 | |
| *** jistr has joined #openstack-dev | 07:40 | |
| *** SergeyLukjanov has quit IRC | 07:40 | |
| *** safchain__ has joined #openstack-dev | 07:41 | |
| *** lexinator has quit IRC | 07:43 | |
| *** ljjjustin has quit IRC | 07:44 | |
| *** ljjjustin has joined #openstack-dev | 07:44 | |
| *** MaxV has joined #openstack-dev | 07:45 | |
| *** johnthetubaguy has joined #openstack-dev | 07:45 | |
| *** Max_ has joined #openstack-dev | 07:45 | |
| *** MaxV has quit IRC | 07:45 | |
| *** vartom4 has joined #openstack-dev | 07:47 | |
| *** johnthetubaguy1 has joined #openstack-dev | 07:48 | |
| *** safchain__ has quit IRC | 07:49 | |
| *** johnthetubaguy has quit IRC | 07:49 | |
| *** safchain has joined #openstack-dev | 07:51 | |
| *** afazekas has joined #openstack-dev | 07:52 | |
| *** egallen has quit IRC | 07:56 | |
| *** giulivo has joined #openstack-dev | 07:57 | |
| *** boris-42 has quit IRC | 07:58 | |
| *** kms has joined #openstack-dev | 07:59 | |
| *** xga has quit IRC | 08:03 | |
| *** xchu has quit IRC | 08:04 | |
| *** vartom4 has quit IRC | 08:04 | |
| *** johnthetubaguy1 is now known as johnthetubaguy | 08:05 | |
| *** JonnyNomad has quit IRC | 08:07 | |
| *** basha has quit IRC | 08:08 | |
| *** JonnyNomad has joined #openstack-dev | 08:08 | |
| *** valgrind has joined #openstack-dev | 08:10 | |
| *** lucasagomes has joined #openstack-dev | 08:11 | |
| *** ljjjustin has quit IRC | 08:12 | |
| *** feleouet has quit IRC | 08:13 | |
| *** fifieldt_ has quit IRC | 08:15 | |
| *** xchu has joined #openstack-dev | 08:16 | |
| *** martyntaylor has joined #openstack-dev | 08:16 | |
| *** jaimegil has joined #openstack-dev | 08:16 | |
| *** feleouet has joined #openstack-dev | 08:17 | |
| *** sushils has quit IRC | 08:17 | |
| *** cthulhup has joined #openstack-dev | 08:21 | |
| *** markmc has joined #openstack-dev | 08:22 | |
| *** michchap_ has joined #openstack-dev | 08:25 | |
| *** cthulhup has quit IRC | 08:25 | |
| *** dmakogon_ has quit IRC | 08:26 | |
| *** henrynash has quit IRC | 08:27 | |
| *** michchap has quit IRC | 08:27 | |
| *** jtomasek has joined #openstack-dev | 08:27 | |
| *** ruhe has joined #openstack-dev | 08:30 | |
| *** romcheg has joined #openstack-dev | 08:32 | |
| *** dina_belova has joined #openstack-dev | 08:32 | |
| *** romcheg has left #openstack-dev | 08:32 | |
| *** Madkiss has quit IRC | 08:34 | |
| *** AnilV4 has joined #openstack-dev | 08:35 | |
| *** prekarat has joined #openstack-dev | 08:36 | |
| *** prekarat has left #openstack-dev | 08:37 | |
| *** dina_belova has quit IRC | 08:37 | |
| *** lexinator has joined #openstack-dev | 08:39 | |
| *** doron_afk is now known as doron | 08:39 | |
| *** lexinator has quit IRC | 08:43 | |
| *** Alexei_987 has joined #openstack-dev | 08:47 | |
| *** doron is now known as doron_afk | 08:47 | |
| *** sridevi has joined #openstack-dev | 08:57 | |
| *** danpb has joined #openstack-dev | 09:00 | |
| *** pschaef has joined #openstack-dev | 09:02 | |
| *** xBsd has joined #openstack-dev | 09:02 | |
| *** ydacheville has joined #openstack-dev | 09:03 | |
| *** sridevi has quit IRC | 09:03 | |
| *** vartom4 has joined #openstack-dev | 09:05 | |
| *** Tross1 has joined #openstack-dev | 09:07 | |
| *** swills has quit IRC | 09:07 | |
| *** swills has joined #openstack-dev | 09:08 | |
| *** swills has joined #openstack-dev | 09:08 | |
| *** Tross has quit IRC | 09:08 | |
| *** Daisy has quit IRC | 09:08 | |
| *** sushils has joined #openstack-dev | 09:09 | |
| *** mkollaro has joined #openstack-dev | 09:11 | |
| *** avishay has joined #openstack-dev | 09:12 | |
| *** cthulhup has joined #openstack-dev | 09:15 | |
| *** AnilV4 has quit IRC | 09:16 | |
| *** alexxu has joined #openstack-dev | 09:17 | |
| *** matiu has quit IRC | 09:17 | |
| *** cthulhup has quit IRC | 09:20 | |
| *** hartsocks1 has joined #openstack-dev | 09:21 | |
| *** jbresnah has quit IRC | 09:22 | |
| *** iartarisi has joined #openstack-dev | 09:24 | |
| *** michchap_ has quit IRC | 09:30 | |
| *** hartsocks1 has quit IRC | 09:31 | |
| *** hartsocks has joined #openstack-dev | 09:32 | |
| *** hartsocks has left #openstack-dev | 09:33 | |
| *** dina_belova has joined #openstack-dev | 09:33 | |
| *** dina_belova has quit IRC | 09:37 | |
| *** boris-42 has joined #openstack-dev | 09:37 | |
| *** sumanthns has joined #openstack-dev | 09:39 | |
| *** lexinator has joined #openstack-dev | 09:39 | |
| *** ema has joined #openstack-dev | 09:43 | |
| *** lexinator has quit IRC | 09:44 | |
| *** hate has joined #openstack-dev | 09:45 | |
| *** hate has quit IRC | 09:46 | |
| *** JordanP has quit IRC | 09:46 | |
| *** enikanorov_ has joined #openstack-dev | 09:47 | |
| *** gtt116 has joined #openstack-dev | 09:48 | |
| *** hailiang has quit IRC | 09:50 | |
| *** nshaikh has left #openstack-dev | 09:52 | |
| *** AnilV4 has joined #openstack-dev | 09:53 | |
| *** gongysh has quit IRC | 09:57 | |
| *** BobBallAway is now known as BobBall | 09:57 | |
| *** xchu has quit IRC | 09:58 | |
| *** JordanP has joined #openstack-dev | 09:58 | |
| *** xga has joined #openstack-dev | 10:01 | |
| *** afazekas has quit IRC | 10:07 | |
| *** dolphm has joined #openstack-dev | 10:10 | |
| *** donaldh has joined #openstack-dev | 10:10 | |
| *** jhesketh_ has quit IRC | 10:12 | |
| *** gongysh has joined #openstack-dev | 10:13 | |
| *** dolphm has quit IRC | 10:14 | |
| *** ruhe has quit IRC | 10:15 | |
| *** feleouet has quit IRC | 10:17 | |
| *** corXi has quit IRC | 10:19 | |
| *** feleouet has joined #openstack-dev | 10:20 | |
| *** morganfainberg is now known as morganfainberg|a | 10:21 | |
| *** ffio has quit IRC | 10:21 | |
| *** ffio_ has joined #openstack-dev | 10:21 | |
| *** Max_ has quit IRC | 10:22 | |
| *** whocares has joined #openstack-dev | 10:24 | |
| *** ruhe has joined #openstack-dev | 10:26 | |
| *** tonix has joined #openstack-dev | 10:26 | |
| whocares | are there any plans to support libvirt netfs pools on openstack? so that we have the cabability to share nfs or glusterfs based directories between multiple instances? like for webclusters e.g. | 10:29 |
| *** ruhe has quit IRC | 10:30 | |
| *** dina_belova has joined #openstack-dev | 10:33 | |
| *** Lingxian has joined #openstack-dev | 10:37 | |
| *** dina_belova has quit IRC | 10:38 | |
| *** openstack has joined #openstack-dev | 15:12 | |
| *** reed has joined #openstack-dev | 15:13 | |
| *** edmund has joined #openstack-dev | 15:14 | |
| *** jruzicka has joined #openstack-dev | 15:15 | |
| *** obondarev has joined #openstack-dev | 15:15 | |
| *** ruhe has quit IRC | 15:17 | |
| *** martyntaylor has joined #openstack-dev | 15:18 | |
| *** jmontemayor has joined #openstack-dev | 15:19 | |
| *** rfolco has joined #openstack-dev | 15:21 | |
| *** drewlander has quit IRC | 15:22 | |
| dolphm | stevemar: like for real i never got an email from you | 15:23 |
| stevemar | dolphm: really? | 15:23 |
| stevemar | dolphm: strange | 15:23 |
| dolphm | stevemar: for really | 15:23 |
| stevemar | dolphm: sent it at 10:14 am EST | 15:23 |
| dolphm | stevemar: received at null | 15:23 |
| stevemar | grrr | 15:24 |
| *** jaybuff has joined #openstack-dev | 15:24 | |
| *** dims has joined #openstack-dev | 15:24 | |
| stevemar | well, anywho | 15:24 |
| *** hemna has joined #openstack-dev | 15:24 | |
| stevemar | dolphm: someone fixed the bug (add forgein keys to oauth tables) | 15:24 |
| stevemar | dolphm: i was wondering if it's bad practise to also fix the nullable=False (should be True), for the consumer description; in the same patch? | 15:25 |
| dolphm | stevemar: fix them all at once | 15:25 |
| dolphm | stevemar: oh you mean have the other dude fix it? | 15:26 |
| stevemar | dolphm: yeah, in the same patch | 15:26 |
| *** matiu has joined #openstack-dev | 15:26 | |
| *** matiu has quit IRC | 15:26 | |
| *** matiu has joined #openstack-dev | 15:26 | |
| stevemar | dolphm: he has a proposed patch for adding forgein keys: https://review.openstack.org/#/c/43237/ | 15:26 |
| *** ifarkas has quit IRC | 15:27 | |
| dolphm | stevemar: up to ya'll | 15:27 |
| stevemar | i'll email him and ask, no idea who he is | 15:27 |
| *** martyntaylor has quit IRC | 15:28 | |
| *** jjmb has quit IRC | 15:28 | |
| stevemar | dolphm: would he start 003... or fix both issues in 002? | 15:28 |
| *** matiu has quit IRC | 15:28 | |
| *** hemna has quit IRC | 15:29 | |
| *** hemna has joined #openstack-dev | 15:29 | |
| dolphm | stevemar: if they go in as two patches, then 003; if it goes in as once patch, 002 | 15:29 |
| stevemar | m'alright | 15:29 |
| *** jdurgin1 has joined #openstack-dev | 15:29 | |
| *** matiu has joined #openstack-dev | 15:29 | |
| *** matiu has quit IRC | 15:30 | |
| *** matiu has joined #openstack-dev | 15:30 | |
| *** ruhe has joined #openstack-dev | 15:30 | |
| *** Vek has quit IRC | 15:30 | |
| *** matiu has quit IRC | 15:31 | |
| *** alop has joined #openstack-dev | 15:33 | |
| *** whocares has quit IRC | 15:33 | |
| *** matiu has joined #openstack-dev | 15:33 | |
| *** matiu has quit IRC | 15:33 | |
| *** matiu has joined #openstack-dev | 15:33 | |
| *** lexinator has quit IRC | 15:33 | |
| *** rods has quit IRC | 15:35 | |
| *** dina_belova has joined #openstack-dev | 15:35 | |
| *** networkstatic has quit IRC | 15:35 | |
| *** markwash has quit IRC | 15:36 | |
| *** CaptTofu has quit IRC | 15:36 | |
| *** tmclaugh[work] has joined #openstack-dev | 15:38 | |
| *** jaybuff has quit IRC | 15:40 | |
| *** Ryan_Lane has joined #openstack-dev | 15:40 | |
| *** gyee has joined #openstack-dev | 15:42 | |
| *** fabio__ has joined #openstack-dev | 15:43 | |
| *** sumanthns has joined #openstack-dev | 15:45 | |
| *** kushal has quit IRC | 15:45 | |
| *** topol has quit IRC | 15:47 | |
| *** kushal has joined #openstack-dev | 15:47 | |
| *** ema_ has quit IRC | 15:49 | |
| *** xqueralt is now known as xqueralt-afk | 15:52 | |
| *** wu_wenxiang has joined #openstack-dev | 15:52 | |
| *** lexinator has joined #openstack-dev | 15:52 | |
| *** jaybuff has joined #openstack-dev | 15:52 | |
| *** Shaan7 has quit IRC | 15:53 | |
| *** Shaan7 has joined #openstack-dev | 15:54 | |
| *** pabelanger_ has joined #openstack-dev | 15:55 | |
| *** pabelanger_ has quit IRC | 15:56 | |
| *** pabelanger_ has joined #openstack-dev | 15:56 | |
| *** CaptTofu has joined #openstack-dev | 15:57 | |
| *** dina_belova has quit IRC | 15:58 | |
| *** drewlander has joined #openstack-dev | 15:58 | |
| *** bdpayne has joined #openstack-dev | 15:58 | |
| *** ema has joined #openstack-dev | 15:59 | |
| *** ema has joined #openstack-dev | 15:59 | |
| *** pabelanger has quit IRC | 16:00 | |
| *** pabelanger_ is now known as pabelanger | 16:00 | |
| *** pabelanger_ has joined #openstack-dev | 16:00 | |
| *** CaptTofu_ has joined #openstack-dev | 16:01 | |
| *** CaptTofu has quit IRC | 16:02 | |
| *** S3Xinfo has joined #openstack-dev | 16:04 | |
| *** martine_ has quit IRC | 16:05 | |
| *** S3Xinfo has quit IRC | 16:05 | |
| *** markmc has quit IRC | 16:05 | |
| *** dkranz has quit IRC | 16:07 | |
| *** cthulhup has joined #openstack-dev | 16:09 | |
| *** datsun180b has quit IRC | 16:09 | |
| *** zbitter has joined #openstack-dev | 16:09 | |
| *** datsun180b has joined #openstack-dev | 16:09 | |
| *** mmagr has quit IRC | 16:11 | |
| *** alexpilotti_ has joined #openstack-dev | 16:11 | |
| *** dklyle is now known as david-lyle | 16:11 | |
| *** ruhe has quit IRC | 16:11 | |
| *** pabelanger has quit IRC | 16:11 | |
| *** zul has quit IRC | 16:12 | |
| *** nachi has joined #openstack-dev | 16:12 | |
| *** jbresnah_ has joined #openstack-dev | 16:12 | |
| *** zaneb has quit IRC | 16:13 | |
| *** zbitter is now known as zaneb | 16:13 | |
| *** alexpilotti has quit IRC | 16:13 | |
| *** alexpilotti_ is now known as alexpilotti | 16:13 | |
| nachi | bknudson, ayoung, gyee, morganfainbergy: Please review the migrating ec2 credentials. I have uploaded the patch implementing all the review comments. https://review.openstack.org/#/c/38367/ | 16:14 |
| ayoung | nachi, will look | 16:14 |
| *** wu_wenxiang has quit IRC | 16:14 | |
| nachi | thanks | 16:14 |
| *** SergeyLukjanov has quit IRC | 16:14 | |
| *** freedomhui has quit IRC | 16:15 | |
| ayoung | shardy, can you take a look at nachi | 16:16 |
| ayoung | 's patch ^^ as I think it impacts on how Heat wants to use EC2 creds to get a token | 16:16 |
| *** MaxV has quit IRC | 16:17 | |
| *** jistr has quit IRC | 16:17 | |
| *** jfriedly has joined #openstack-dev | 16:17 | |
| shardy | ayoung: will do, but we actually want to do the reverse, create an ec2-keypair *from* a token, obtained via a trust | 16:17 |
| *** nunosantos has joined #openstack-dev | 16:18 | |
| ayoung | shardy, right...and I think that this extension is where to do that work | 16:18 |
| shardy | ayoung: ok, sounds good :) | 16:18 |
| *** MaxV has joined #openstack-dev | 16:19 | |
| *** ruhe has joined #openstack-dev | 16:19 | |
| *** ruhe has quit IRC | 16:19 | |
| *** garyk has joined #openstack-dev | 16:19 | |
| *** MaxV has quit IRC | 16:20 | |
| *** arezadr has quit IRC | 16:21 | |
| *** arezadr has joined #openstack-dev | 16:24 | |
| *** dkranz has joined #openstack-dev | 16:25 | |
| gyee | ayoung, shardy, that patch is a significant first step for managing *credentials* | 16:26 |
| gyee | now all credentials are managed in one place | 16:26 |
| ayoung | gyee, including private keys? | 16:27 |
| gyee | I can imagine Barbican will play a role in IceHouse | 16:27 |
| ayoung | gyee, yeah...but PKI, man....why can't we get the hang of :generate a key local, post the public one for signing | 16:28 |
| gyee | ayoung, sure, once we have the access sorted out | 16:28 |
| ayoung | this is just public? | 16:28 |
| gyee | ayoung, yes, credential API can be used for both create and import | 16:28 |
| *** briancurtin has joined #openstack-dev | 16:28 | |
| *** SergeyLukjanov has joined #openstack-dev | 16:28 | |
| *** dina_belova has joined #openstack-dev | 16:29 | |
| gyee | ayoung, both public and private, once we figure out the access control | 16:29 |
| *** datsun180b has quit IRC | 16:29 | |
| *** JordanP has quit IRC | 16:29 | |
| *** danwent has quit IRC | 16:29 | |
| gyee | ec2 secrets are private | 16:29 |
| *** toabctl has quit IRC | 16:30 | |
| ayoung | ok...yeah, I was getting confused with another project that insists on handling private keys in the server | 16:30 |
| ayoung | why are they ec2 secrets? Aren't they just some standard like rsa or something? | 16:30 |
| gyee | ayoung, just some random bits | 16:31 |
| * ayoung has turned a blind eye to this thus far | 16:31 | |
| gyee | for HMAC | 16:31 |
| *** salv-orlando has joined #openstack-dev | 16:31 | |
| ayoung | gyee, I thought they were used for access to the ec2 images, like openssh keys were | 16:32 |
| *** briancurtin has quit IRC | 16:32 | |
| *** xga_ has quit IRC | 16:33 | |
| *** drewlander has quit IRC | 16:33 | |
| *** mrodden has quit IRC | 16:35 | |
| *** toabctl has joined #openstack-dev | 16:35 | |
| *** toabctl has joined #openstack-dev | 16:35 | |
| gyee | ayoung, it can be openssh keys too | 16:35 |
| *** toabctl has quit IRC | 16:35 | |
| ayoung | ec2 can be openssh keys? Or just, credential API supports openssh keys? | 16:36 |
| gyee | same key can be used for both generating HMAC and SSH | 16:36 |
| *** toabctl has joined #openstack-dev | 16:36 | |
| *** feleouet has quit IRC | 16:37 | |
| *** iartarisi has quit IRC | 16:39 | |
| *** giulivo has quit IRC | 16:39 | |
| *** alexb_ has joined #openstack-dev | 16:39 | |
| *** feleouet has joined #openstack-dev | 16:39 | |
| *** datsun180b has joined #openstack-dev | 16:40 | |
| *** markwash has joined #openstack-dev | 16:40 | |
| *** zul has joined #openstack-dev | 16:41 | |
| *** cthulhup has quit IRC | 16:41 | |
| *** jaybuff has quit IRC | 16:41 | |
| *** danpb has quit IRC | 16:42 | |
| *** morganfainberg|a is now known as morganfainberg | 16:44 | |
| *** ema has quit IRC | 16:45 | |
| morganfainberg | stevemar: ping | 16:45 |
| *** maheshp has joined #openstack-dev | 16:46 | |
| *** xqueralt-afk is now known as xqueralt | 16:47 | |
| dolphm | morganfainberg: on the cache impl, why did you create `[cache] verbose` instead of using the existing `[DEFAULT] verbose` ? | 16:48 |
| *** mrodden has joined #openstack-dev | 16:49 | |
| *** Vek has joined #openstack-dev | 16:49 | |
| morganfainberg | dolphm: likely i should couple that with the standard debug instead, i was thinking verbose caching was _very_ verbose and it makes reading the logs hard. | 16:49 |
| *** lucasagomes has quit IRC | 16:49 | |
| morganfainberg | every get of a cache key generates a log line, same with set or deletes. | 16:50 |
| *** jpich has quit IRC | 16:50 | |
| *** krtaylor has quit IRC | 16:50 | |
| *** lexinator has quit IRC | 16:50 | |
| dolphm | morganfainberg: so does sqlalchemy though | 16:50 |
| morganfainberg | dolphm: actually, with debug it is even worse with dogpile, you get a ton of output that doesn't even tell you anything useful, but yeah, i'll couple that with the standard debug/verbose stuff. | 16:51 |
| morganfainberg | dolphm: good call. | 16:51 |
| *** drewlander has joined #openstack-dev | 16:52 | |
| *** bingbu has quit IRC | 16:53 | |
| *** hartsocks has joined #openstack-dev | 16:53 | |
| *** safchain has quit IRC | 16:53 | |
| *** jaybuff has joined #openstack-dev | 16:54 | |
| *** lexinator has joined #openstack-dev | 16:54 | |
| *** fbo is now known as fbo_away | 16:56 | |
| *** ifarkas has joined #openstack-dev | 16:56 | |
| *** dina_belova has quit IRC | 16:57 | |
| *** danwent has joined #openstack-dev | 16:57 | |
| *** bswartz has quit IRC | 16:57 | |
| *** SergeyLukjanov has quit IRC | 17:00 | |
| *** dkranz has quit IRC | 17:01 | |
| *** dkranz has joined #openstack-dev | 17:01 | |
| *** dims has quit IRC | 17:01 | |
| *** nati_ueno has joined #openstack-dev | 17:02 | |
| *** tmclaugh[work] has quit IRC | 17:03 | |
| *** dims has joined #openstack-dev | 17:03 | |
| *** Alexei_987 has quit IRC | 17:04 | |
| *** fabio__ has quit IRC | 17:04 | |
| *** morganfainberg is now known as morganfainberg|a | 17:05 | |
| *** BobBall is now known as BobBallAway | 17:05 | |
| *** otherwiseguy has quit IRC | 17:05 | |
| *** SergeyLukjanov has joined #openstack-dev | 17:05 | |
| *** nayward has quit IRC | 17:06 | |
| *** jruzicka has quit IRC | 17:08 | |
| ayoung | rcrit, on Puppet, ever come across: Could not autoload puppet/type/keystone_config: Could not autoload puppet/provider/keystone_config/ini_setting: undefined method `provider' for nil:NilClass | 17:12 |
| ayoung | I'm doing this all out of the git repos | 17:12 |
| *** alexpilotti has quit IRC | 17:18 | |
| *** amohn9 has joined #openstack-dev | 17:19 | |
| *** SergeyLukjanov has quit IRC | 17:22 | |
| *** svarnau has joined #openstack-dev | 17:22 | |
| *** boris-42 has joined #openstack-dev | 17:25 | |
| *** alunduil has joined #openstack-dev | 17:25 | |
| *** ChanServ changes topic to "OpenStack development || Support is in #openstack" | 17:25 | |
| *** bashok has quit IRC | 17:26 | |
| *** tmclaugh[work] has joined #openstack-dev | 17:27 | |
| *** bashok has joined #openstack-dev | 17:27 | |
| *** pcm_ has quit IRC | 17:28 | |
| nachi | bknudson: thanks for reviewing the changes | 17:29 |
| *** venkatesh has joined #openstack-dev | 17:29 | |
| *** pabelanger has joined #openstack-dev | 17:29 | |
| *** morganfainberg|a is now known as morganfainberg | 17:29 | |
| *** svarnau has quit IRC | 17:29 | |
| *** svarnau has joined #openstack-dev | 17:30 | |
| *** maheshp has quit IRC | 17:30 | |
| *** vartom4 has joined #openstack-dev | 17:31 | |
| *** pcm_ has joined #openstack-dev | 17:32 | |
| gyee | ayoung, bknudson, I am trying to help out fabio | 17:32 |
| gyee | ran into this when running the tests | 17:33 |
| *** markmcclain has quit IRC | 17:33 | |
| gyee | InvalidRequestError: Table 'project_endpoint' is already defined for this MetaData instance. Specify 'extend_existing=True' to redefine options and columns on an existing Table object. | 17:33 |
| gyee | any idea? | 17:33 |
| *** xBsd has quit IRC | 17:34 | |
| *** bashok has quit IRC | 17:35 | |
| *** bashok has joined #openstack-dev | 17:36 | |
| *** venkatesh has quit IRC | 17:36 | |
| bknudson | gyee: this is running the tests using master or some changes? | 17:37 |
| bknudson | is this live sql test? | 17:37 |
| morganfainberg | that sounds like live sql without clearing the schema out. | 17:37 |
| *** mrunge has quit IRC | 17:37 | |
| morganfainberg | but thats just my untrained eye. | 17:37 |
| *** alanmac has quit IRC | 17:38 | |
| gyee | bknudson, no, I am working on fabio's patch | 17:38 |
| gyee | https://review.openstack.org/#/c/33118 | 17:38 |
| *** fbo_away is now known as fbo | 17:38 | |
| bknudson | gyee: mysql? postgres? | 17:39 |
| *** svarnau has quit IRC | 17:39 | |
| *** svarnau has joined #openstack-dev | 17:40 | |
| bknudson | gyee: keystone.tests.test_sql_migrate_extensions ? | 17:40 |
| *** sandeepr_ltp has joined #openstack-dev | 17:41 | |
| *** svarnau has quit IRC | 17:42 | |
| ayoung | gyee, is the table in the database already? | 17:44 |
| ayoung | if it was added by an old migration that was, say in the common set, and then you try running the migration against mysql... | 17:44 |
| *** vkmc has quit IRC | 17:45 | |
| *** dina_belova has joined #openstack-dev | 17:45 | |
| ayoung | but this is against the sqlite db, right gyee ? | 17:45 |
| gyee | bknudson, ayoung, yeah, the error seem to suggest table already exist | 17:46 |
| gyee | I am using sqlite | 17:46 |
| *** avishay has joined #openstack-dev | 17:46 | |
| *** SergeyLukjanov has joined #openstack-dev | 17:46 | |
| morganfainberg | bknudson / ayoung / dolphm: for https://review.openstack.org/#/c/37233 the common-library-extension, will changing out the exceptions.from_url method raise the same concerns as moving HTTPClient? if so, we might need a translation method. i'm not sure where we are drawing that line at the moment. | 17:47 |
| ayoung | gyee, so that test runs against the db file in keystone/tests/tmp | 17:47 |
| ayoung | morganfainberg, you mean on client backwards compat? | 17:47 |
| morganfainberg | yeah | 17:47 |
| *** alunduil has quit IRC | 17:47 | |
| morganfainberg | ayoung: and shuffling all the exceptions around. | 17:48 |
| ayoung | most people just do except: | 17:48 |
| ayoung | pass | 17:48 |
| *** ruhe has joined #openstack-dev | 17:48 | |
| dolphm | morganfainberg: from_response()? | 17:48 |
| *** mars has joined #openstack-dev | 17:50 | |
| dolphm | morganfainberg: our exceptions are definitely public api | 17:50 |
| morganfainberg | dolphm: keystoneclient.exceptions.from_response, it builds an exception from the response, https://github.com/openstack/python-keystoneclient/blob/master/keystoneclient/exceptions.py#L136 | 17:50 |
| *** danwent has quit IRC | 17:50 | |
| dolphm | morganfainberg: i can't think of a reason why anyone would consume from_response() externally to keystoneclient, can you? | 17:50 |
| bknudson | morganfainberg: it's a public api. It's even got a docstring. | 17:50 |
| *** briancurtin has joined #openstack-dev | 17:51 | |
| *** changbl has joined #openstack-dev | 17:51 | |
| dolphm | bknudson: technically | 17:51 |
| morganfainberg | dolphm: exactly, but the exceptions being moved is more problematic. | 17:51 |
| morganfainberg | dolphm: otherwise i think aababilov's review is pretty good. i just wanted to make sure i was on the same page about what we accept as public, and if it's public - i'll comment as much. | 17:53 |
| *** jmontemayor has quit IRC | 17:53 | |
| *** alanmac has joined #openstack-dev | 17:54 | |
| openstackstatus | NOTICE: restarting gerrit to pick up a configuration change | 17:55 |
| dolphm | morganfainberg: from_response is still available after his patch | 17:56 |
| morganfainberg | dolphm: oh i see the import * now. damn more coffee | 17:57 |
| morganfainberg | but the function signature does change. | 17:57 |
| *** vkmc has joined #openstack-dev | 17:57 | |
| morganfainberg | from "def from_response(response, body=None):" to "def from_response(response, method, url):" is that an issue? | 17:58 |
| dolphm | morganfainberg: well, that's not backwards compatible at all | 17:58 |
| morganfainberg | yeah. | 17:59 |
| *** feleouet has quit IRC | 17:59 | |
| *** jmontemayor has joined #openstack-dev | 18:00 | |
| stevemar | morganfainberg: pong | 18:00 |
| dolphm | morganfainberg: i definitely wouldn't want that kind of incompatibility going out in 0.3.2 today/tomorrow... | 18:00 |
| *** alop has quit IRC | 18:01 | |
| morganfainberg | dolphm: makes sense, that was my initial gut reaction, but i wanted to be sure since this is still a bit uncharted until we get more into the definition of public/private | 18:01 |
| *** maheshp has joined #openstack-dev | 18:01 | |
| *** briancurtin has quit IRC | 18:02 | |
| dolphm | morganfainberg: agree ... i'm guessing this would be minor compared to other things we've missed in the past | 18:02 |
| *** feleouet has joined #openstack-dev | 18:02 | |
| dolphm | morganfainberg: s/minor/very low impact/ | 18:02 |
| dolphm | if any | 18:02 |
| morganfainberg | stevemar: patches are up, please be brutal in reviews, cache needs some serious close eyes on it — it could have very negative impact if done wrong | 18:02 |
| morganfainberg | dolphm: yeah, thats why i am asking. | 18:02 |
| *** bswartz has joined #openstack-dev | 18:03 | |
| *** gkotton has joined #openstack-dev | 18:03 | |
| gyee | ayoung, bknudson, guess what the problem was | 18:04 |
| gyee | something really stupid | 18:05 |
| bknudson | gyee: reversed the polarity on something? | 18:05 |
| rcrit | ayoung, sorry, that's a new one to me | 18:05 |
| ayoung | crossed the streams? | 18:05 |
| gyee | no, I have a typo in the code | 18:05 |
| *** donaldh has quit IRC | 18:05 | |
| gyee | handle_conflict instead of handle_conflicts | 18:05 |
| ayoung | rcrit, who was helping with the puppet stuff? | 18:06 |
| gyee | and the error was as confusing as you can imagine | 18:06 |
| ayoung | nice | 18:06 |
| rcrit | ayoung, me myself and I | 18:06 |
| *** garyk has quit IRC | 18:06 | |
| *** gkotton is now known as garyk | 18:06 | |
| ayoung | rcrit, thought you got linked up with someone on the foreman side? | 18:06 |
| rcrit | Dominic in #theforeman pointed me in the right direction | 18:06 |
| rcrit | but it was mostly procedural. | 18:06 |
| *** tkammer has quit IRC | 18:06 | |
| *** joesavak has joined #openstack-dev | 18:07 | |
| *** sumanthns has quit IRC | 18:08 | |
| joesavak | dolphm, chmouel, would it make sense to say "use_contract_binding_code_path" instead for https://review.openstack.org/#/c/43042/2/keystoneclient/httpclient.py? | 18:10 |
| joesavak | trying to figure out next step for that review | 18:10 |
| *** redbo is now known as gholt | 18:12 | |
| *** gholt is now known as redbo | 18:12 | |
| openstackstatus | NOTICE: stopping gerrit to correct a stackforge project rename error | 18:12 |
| *** dmakogon_ has joined #openstack-dev | 18:13 | |
| *** prekarat has quit IRC | 18:13 | |
| *** jbresnah_ has quit IRC | 18:13 | |
| nikhil | https://review.openstack.org seems to be down | 18:13 |
| *** nachi has left #openstack-dev | 18:14 | |
| *** mrodden has quit IRC | 18:14 | |
| *** jrwren has joined #openstack-dev | 18:14 | |
| *** alanmac has quit IRC | 18:14 | |
| joesavak | back up to me | 18:14 |
| *** ffio_ has quit IRC | 18:14 | |
| nikhil | yeah, thanks joesavak | 18:15 |
| *** martine has joined #openstack-dev | 18:16 | |
| *** martine is now known as Guest75025 | 18:16 | |
| *** briancurtin has joined #openstack-dev | 18:16 | |
| *** briancurtin has quit IRC | 18:17 | |
| ohadlevy | rcrit: let me konw if you need any help | 18:18 |
| rcrit | ohadlevy, thanks, I just might :-) | 18:19 |
| *** Mandell has joined #openstack-dev | 18:19 | |
| stevemar | morganfainberg: cool, i will look soon | 18:19 |
| morganfainberg | stevemar: thanks. | 18:19 |
| *** neelashah has quit IRC | 18:19 | |
| *** networkstatic has joined #openstack-dev | 18:20 | |
| *** melwitt has joined #openstack-dev | 18:20 | |
| *** Guest75025 is now known as martine_ | 18:21 | |
| *** zul has quit IRC | 18:21 | |
| *** hemanth_ has quit IRC | 18:22 | |
| *** hemanth_ has joined #openstack-dev | 18:22 | |
| *** mrodden has joined #openstack-dev | 18:22 | |
| *** vuil has joined #openstack-dev | 18:22 | |
| *** sushils has quit IRC | 18:24 | |
| *** erfanian has quit IRC | 18:24 | |
| *** svarnau has joined #openstack-dev | 18:25 | |
| *** zodiak has joined #openstack-dev | 18:27 | |
| *** danwent has joined #openstack-dev | 18:29 | |
| *** svarnau has left #openstack-dev | 18:29 | |
| *** sarob has joined #openstack-dev | 18:29 | |
| *** sarob has quit IRC | 18:31 | |
| *** alexpilotti has joined #openstack-dev | 18:31 | |
| *** alop has joined #openstack-dev | 18:31 | |
| *** sarob has joined #openstack-dev | 18:31 | |
| *** danwent has quit IRC | 18:32 | |
| *** garyk has quit IRC | 18:32 | |
| *** kushal has quit IRC | 18:34 | |
| *** dolphm has quit IRC | 18:34 | |
| *** sarob has quit IRC | 18:35 | |
| thingee | dtroyer: ping | 18:36 |
| *** neelashah has joined #openstack-dev | 18:36 | |
| *** krtaylor has joined #openstack-dev | 18:37 | |
| *** arbrandes has joined #openstack-dev | 18:38 | |
| *** pabelanger has quit IRC | 18:40 | |
| *** neoXsys has quit IRC | 18:45 | |
| *** ms_brown has joined #openstack-dev | 18:46 | |
| *** chandankumar has quit IRC | 18:46 | |
| *** danwent has joined #openstack-dev | 18:47 | |
| *** pixelbeat has joined #openstack-dev | 18:47 | |
| *** Tross1 has quit IRC | 18:48 | |
| *** Tross has joined #openstack-dev | 18:48 | |
| *** novas0x2a|laptop has joined #openstack-dev | 18:48 | |
| *** salv-orlando has quit IRC | 18:49 | |
| *** salv-orlando has joined #openstack-dev | 18:49 | |
| *** nachi has joined #openstack-dev | 18:50 | |
| *** nayward has joined #openstack-dev | 18:51 | |
| dtroyer | thingee: yo | 18:52 |
| *** networkstatic has quit IRC | 18:56 | |
| *** insanidade has joined #openstack-dev | 18:58 | |
| insanidade | Hi all. Is there a guide on how to create pluguins for Openstack? Is that possible or is it something strictly controlled by the Openstack foundation ? | 18:58 |
| *** neoXsys has joined #openstack-dev | 18:58 | |
| insanidade | I mean: what if I need a plugin for a given equipement I have to use in my architecture? Is it possible to develop your own plugin ? | 19:00 |
| *** hartsocks has left #openstack-dev | 19:00 | |
| *** martine_ has quit IRC | 19:01 | |
| *** sarob has joined #openstack-dev | 19:02 | |
| *** lbragstad has left #openstack-dev | 19:02 | |
| thingee | dtroyer: hi. wrt https://review.openstack.org/#/c/43045 according to the console logs, it appears 8776/v2 isn't being registered in the keystone catalog. http://logs.openstack.org/45/43045/1/check/gate-grenade-devstack-vm/1de2851/console.html.gz | 19:02 |
| thingee | dtroyer: and should be since this change was made https://github.com/openstack-dev/devstack/commit/b3862f98718317042dd48871d50da1e5255c0329#lib/cinder | 19:03 |
| thingee | dtroyer: just wanted to see if you have any insight on this front with grenade. | 19:03 |
| *** Tross has quit IRC | 19:08 | |
| dtroyer | The change to lib/cinder to add v2 endpoints was committed after stable/grizzly devstack, which is what grenade uses as its starting point. | 19:08 |
| dtroyer | so that needs to be an upgrade step in grenade if you want to use it in the trunk (aka target) phase in grenade | 19:09 |
| *** CaptTofu_ has quit IRC | 19:09 | |
| *** CaptTofu has joined #openstack-dev | 19:10 | |
| dtroyer | i.e., grenade/upgrade-cinder needs to have all of the steps to get from stable/grizzly to trunk in it | 19:10 |
| *** sarob has quit IRC | 19:10 | |
| *** lbragstad has joined #openstack-dev | 19:10 | |
| *** jmontemayor has quit IRC | 19:11 | |
| ayoung | ohadlevy, I might, too. I'm trying to wrap my head around the Keystone puppet module. | 19:14 |
| *** CaptTofu has quit IRC | 19:14 | |
| *** Tross has joined #openstack-dev | 19:14 | |
| *** jprovazn has quit IRC | 19:17 | |
| *** SergeyLukjanov has quit IRC | 19:23 | |
| *** alexb_ has quit IRC | 19:24 | |
| *** terryh has quit IRC | 19:25 | |
| *** ruhe has quit IRC | 19:27 | |
| *** xBsd has joined #openstack-dev | 19:28 | |
| *** drewlander has quit IRC | 19:29 | |
| *** nunosantos has quit IRC | 19:30 | |
| *** vipul is now known as vipul-away | 19:32 | |
| *** jmontemayor has joined #openstack-dev | 19:34 | |
| *** xBsd has quit IRC | 19:35 | |
| *** HenryG has quit IRC | 19:36 | |
| *** sarob has joined #openstack-dev | 19:37 | |
| *** xqueralt is now known as xqueralt-afk | 19:38 | |
| insanidade | Hi all. Is there a guide on how to create pluguins for Openstack? Is that possible or is it something strictly controlled by the Openstack foundation ? | 19:38 |
| *** yolanda has quit IRC | 19:38 | |
| *** donaldh has joined #openstack-dev | 19:39 | |
| *** sdake_ has quit IRC | 19:39 | |
| thingee | dtroyer: ok my mistake. I thought this was failing on target. | 19:40 |
| *** sarob has quit IRC | 19:41 | |
| *** jbresnah_ has joined #openstack-dev | 19:42 | |
| *** emagana has joined #openstack-dev | 19:43 | |
| *** eharney has quit IRC | 19:43 | |
| emagana | Any Horizon developers? | 19:43 |
| *** feleouet has quit IRC | 19:43 | |
| emagana | File "/usr/local/lib/python2.7/dist-packages/keyring/core.py", line 126, in load_config | 19:45 |
| emagana | [Thu Aug 22 18:46:01 2013] [error] local_path = os.path.join(os.getcwd(), filename) | 19:45 |
| emagana | [Thu Aug 22 18:46:01 2013] [error] OSError: [Errno 2] No such file or directory | 19:45 |
| emagana | I am seeing an error in Horizon using Devstack (clean environment): | 19:45 |
| *** feleouet has joined #openstack-dev | 19:46 | |
| ayoung | rcrit, http://adam.younglogic.com/2013/08/deploying-keystone-via-puppet-on-f19/ | 19:47 |
| david-lyle | emagana: when are you seeing this error and what filename is it not finding? | 19:49 |
| ayoung | ohadlevy, is there some way to sterilize a machine after puppet gets run, to get it back into the pristine "pre" state? | 19:49 |
| rcrit | ayoung, nice | 19:49 |
| thingee | dtroyer: thanks for the help | 19:49 |
| rcrit | thoguh stdlib is probalby in some rpm | 19:50 |
| emagana | david-lyle: when horizon is started: /opt/stack/horizon$ cd /opt/stack/horizon && sudo tail -f /var/log/apache2/horizon_error.log || echo "horizon failed to start" | tee "/opt/stack/status/stack/horizon.failure" | 19:50 |
| ayoung | rcrit, yeah, and the git repos are in etc, so change can be made right there | 19:50 |
| ayoung | rcrit, assumption is that anything in etc puppet modules is fair game for a patch | 19:50 |
| emagana | david-lyle: File "/usr/local/lib/python2.7/dist-packages/keyring/core.py", line 126, in load_config | 19:50 |
| dtroyer | thingee: np, that was just grenade doing its job… | 19:50 |
| *** SergeyLukjanov has joined #openstack-dev | 19:51 | |
| *** pixelbeat has quit IRC | 19:51 | |
| ayoung | rcrit, also, if puppet-openstack is in moduels it will conflict with this. | 19:51 |
| rcrit | what I can't get used to is all this development as root | 19:52 |
| ayoung | rcrit, I suspec that a better approach for development is to have a snap shot vm as a puppet client. You do your dev work on the server, and then push the change to the client to test | 19:53 |
| jswarren | If any glance core reviewers could spare a moment or two reviewing https://review.openstack.org/#/c/40232/ and https://review.openstack.org/#/c/40247/, it would be greatly appreciated. | 19:53 |
| ayoung | rcrit, well, I was doing a chown of modules to the fedora user (added by default, I normally us an ayoung account) and you can do everything as yourself until "puppet apply" | 19:53 |
| rcrit | yeah, but that's all the important stuff anyhow | 19:54 |
| rcrit | I've been doing dev on the Foreman box and applying changes on a controller client | 19:55 |
| *** spzala has joined #openstack-dev | 19:55 | |
| rcrit | the changes on Foreman, once working, I apply locally to my git trees | 19:55 |
| rcrit | which is quite cumbersome, but it's what I've got so far | 19:55 |
| *** jasdeepH has joined #openstack-dev | 19:55 | |
| ayoung | so that piece at least can be cleaned up. | 19:56 |
| *** martine has joined #openstack-dev | 19:57 | |
| *** gyee has quit IRC | 19:57 | |
| *** martine is now known as Guest52448 | 19:57 | |
| *** tmclaugh[work] has quit IRC | 19:57 | |
| *** CaptTofu has joined #openstack-dev | 19:58 | |
| *** alexb_ has joined #openstack-dev | 19:58 | |
| david-lyle | emagana: trying to reproduce, not having any luck, did you modify the devstack config at all? | 19:59 |
| emagana | not at all | 19:59 |
| emagana | I am using my dev server | 20:00 |
| *** Guest52448 has quit IRC | 20:00 | |
| *** sarob has joined #openstack-dev | 20:00 | |
| emagana | I will try to reproduce using a new VM | 20:00 |
| *** pcm_ has quit IRC | 20:00 | |
| *** nayward has quit IRC | 20:01 | |
| david-lyle | emagana: Seems like os.getcwd() is what's failing | 20:02 |
| david-lyle | seems that can happen if the cwd has been deleted by another process | 20:02 |
| emagana | david-lyle: mmhhh, weird! | 20:02 |
| david-lyle | no kidding | 20:02 |
| *** dina_belova has quit IRC | 20:03 | |
| emagana | david-lyle: not sure, why this is happening | 20:03 |
| emagana | I did not use this VM for a while | 20:04 |
| *** dina_belova has joined #openstack-dev | 20:04 | |
| emagana | then I deleted devstacl and /opt/stack, started from a clean environment | 20:04 |
| emagana | and I found the problem | 20:04 |
| david-lyle | good | 20:04 |
| *** networkstatic has joined #openstack-dev | 20:05 | |
| *** sdake_ has joined #openstack-dev | 20:06 | |
| *** sdake_ has quit IRC | 20:06 | |
| *** sdake_ has joined #openstack-dev | 20:06 | |
| *** jbresnah_ is now known as jbresnah | 20:07 | |
| *** nati_uen_ has joined #openstack-dev | 20:09 | |
| insanidade | quick hint, please (I'm new to openstack development): I installed Openstack with packstack. Where should the source code be ? I'd like to take a look at some plugins. | 20:11 |
| russellb | insanidade: if you're going to hack on the code, there is a thing called devstack that's good for setting up a development environment | 20:11 |
| russellb | insanidade: but if you just want to browse the code i can point you to it | 20:11 |
| *** alop has quit IRC | 20:11 | |
| russellb | insanidade: what distro? RHEL/CentOS/Fedora? one of those i assume based on packstack | 20:12 |
| *** nati_ueno has quit IRC | 20:12 | |
| insanidade | russellb: Yeah, for the time being I just want to take a look. I'll start developing some plugins. And yeah: I've tried devstack and I'll probably use it for development. | 20:12 |
| russellb | cool | 20:12 |
| russellb | /usr/lib/python2.6/site-packages/nova/ | 20:12 |
| insanidade | russellb: CentOS. | 20:12 |
| russellb | or s/nova/project_of_your_choice/ | 20:12 |
| russellb | for the most part anyway | 20:13 |
| *** vkmc has quit IRC | 20:13 | |
| insanidade | russellb: so that's the nova folder, right? The quantum one, for instance, would be /usr/lib/python2.6/site-packages/quantum , correct ? | 20:13 |
| russellb | yep | 20:13 |
| russellb | and if you have any other centos/packstack specific questions, if you don't get help here, try the #rdo channel | 20:14 |
| insanidade | russellb: thanks a lot. I was not aware of the rdo channel. I'll try it. | 20:14 |
| russellb | sure, np | 20:14 |
| russellb | it's basically a room to discuss or get support for using the stuff you get from openstack.redhat.com | 20:14 |
| *** ctlaugh_ has quit IRC | 20:14 | |
| *** safchain has joined #openstack-dev | 20:15 | |
| *** ctlaugh has joined #openstack-dev | 20:16 | |
| *** alop has joined #openstack-dev | 20:18 | |
| *** stevemar has quit IRC | 20:23 | |
| *** bashok has quit IRC | 20:24 | |
| *** dina_belova has quit IRC | 20:25 | |
| *** SergeyLukjanov has quit IRC | 20:25 | |
| *** vipul-away is now known as vipul | 20:25 | |
| *** vkmc has joined #openstack-dev | 20:25 | |
| *** nati_uen_ has quit IRC | 20:26 | |
| *** eharney has joined #openstack-dev | 20:26 | |
| *** sarob_ has joined #openstack-dev | 20:29 | |
| *** sushils has joined #openstack-dev | 20:30 | |
| *** sarob_ has quit IRC | 20:32 | |
| *** sarob_ has joined #openstack-dev | 20:32 | |
| *** mkollaro has quit IRC | 20:33 | |
| *** sarob has quit IRC | 20:33 | |
| anteaya | all passengers flying openstack air, please return to your seats and bring your seat backs and tray tables to the upright and locked position, we may be experiencing slight turbulence ahead | 20:34 |
| morganfainberg | anteaya: do we have to disable electronics too? ;) | 20:34 |
| *** sushils has quit IRC | 20:35 | |
| anteaya | nah | 20:35 |
| anteaya | listening you your music | 20:35 |
| anteaya | :D | 20:35 |
| anteaya | s/listening/listen | 20:35 |
| dtroyer | playing Dick Dale loud enough can get you through anything | 20:35 |
| anteaya | ha ha ha, there is a soundtrack for every situation | 20:36 |
| * anteaya heads to youtube for some Dick Dale | 20:36 | |
| anteaya | yeah, that works | 20:36 |
| *** sushils has joined #openstack-dev | 20:39 | |
| *** boden has quit IRC | 20:43 | |
| *** woodspa has quit IRC | 20:44 | |
| *** avishay is now known as avishay_zzz | 20:45 | |
| *** arbrandes has quit IRC | 20:45 | |
| *** amohn9 has quit IRC | 20:45 | |
| ayoung | rcrit, so, it looks like the puppet module for HTTPD might be very out of sync with Fedora. I ran the code from the httpd/keystone patch, and the httpd config referenced a slew of Apache configuration options that were now invalid | 20:45 |
| *** Mandell has quit IRC | 20:46 | |
| ayoung | removing the rpms and files and doing another yum install showed that the httpd server runs fine in default configuration | 20:46 |
| *** Mandell has joined #openstack-dev | 20:46 | |
| *** johnthetubaguy has quit IRC | 20:47 | |
| ayoung | my guess is the puppet module identified the OS as a Red Hat type machine and attempted to configure it the same as RHEL or something | 20:47 |
| *** jjmb has joined #openstack-dev | 20:49 | |
| *** dkranz has quit IRC | 20:51 | |
| *** tonix has quit IRC | 20:52 | |
| *** alop has quit IRC | 20:53 | |
| *** jjmb1 has joined #openstack-dev | 20:53 | |
| rcrit | that could be | 20:54 |
| *** jjmb has quit IRC | 20:54 | |
| anteaya | dtroyer: you around for a question from psedlak? | 20:55 |
| *** ifarkas has quit IRC | 20:55 | |
| anteaya | his Dick Dale might be on kind of loud right now | 20:56 |
| *** devvesa has quit IRC | 20:56 | |
| psedlak | dtroyer: is stable/grizzly branch of devstack used for stable/grizzly gates or is it master branch (as is tempest)? | 20:56 |
| dtroyer | anteaya: you caught me during an acoustic number... | 20:56 |
| *** radez is now known as radez_g0n3 | 20:56 | |
| *** jbresnah has quit IRC | 20:57 | |
| anteaya | ah good | 20:57 |
| dtroyer | psedlak: stable/grizzly devstack is used for stable/grizzly gating akaik | 20:57 |
| *** xmltok has joined #openstack-dev | 20:57 | |
| psedlak | dtroyer: as i've noticed that reinstallation of conflicting versions of (for ex.) python-keystoneclient as there are deps on <0.3.0 and >=0.3.0 | 20:57 |
| psedlak | dtroyer: so it switches between 0.2.5 and 0.3.1 multiple times ... and when i've tried to reproduce stable/grizzly env in my vm nova failed on that so i wonder if there is some workaround for that in stable/grizzly gate ... | 20:58 |
| dtroyer | psedlak: yup, it was a mess and was painful to clean up. however, we're not planning to backport all of that work (ie, fixing all of the requirements across projects, etc) | 20:59 |
| dtroyer | psedlak: the gate is different as it attempts to have all of the repos pre-loaded and nothing should fiddle with them later. | 21:00 |
| psedlak | dtroyer: yeah i understand that, i just wonder what is messed in my place that nova (nova-rootwrap...) fails at the end of devstack | 21:00 |
| psedlak | dtroyer: i've cut out this http://bpaste.net/show/vnYioO66WaD27IC7C1dh/ list of keycli deps from projects (stable/grizzly repos where available, rest is master) | 21:01 |
| *** gyee has joined #openstack-dev | 21:01 | |
| *** mkollaro has joined #openstack-dev | 21:01 | |
| psedlak | dtroyer: it caused me and issue (in the middle of comment) https://bugs.launchpad.net/tempest/+bug/1213912/comments/2 and i'm not sure why it did not happend at gate too | 21:02 |
| uvirtbot | Launchpad bug 1213912 in tempest/grizzly "keystone cli tests fail in grizzly" [Critical,Confirmed] | 21:02 |
| *** henrynash has joined #openstack-dev | 21:02 | |
| dtroyer | psedlak: I don't know specifically about the rootwrap failure, but whatever grenade is installing for stable/grizzly works. Again that is pre-installed by the gate but it might make a good reference point. | 21:03 |
| psedlak | dtroyer: and well based on console log from gate it seems for me that it touched the key-cli package ... :/ | 21:03 |
| *** johnthetubaguy has joined #openstack-dev | 21:03 | |
| *** CaptTofu has quit IRC | 21:04 | |
| *** jyhc has joined #openstack-dev | 21:04 | |
| *** jjmb1 has quit IRC | 21:04 | |
| jyhc | Does --availability-zone require admin? | 21:04 |
| *** CaptTofu has joined #openstack-dev | 21:04 | |
| psedlak | dtroyer: this http://bpaste.net/show/4uxTiwaLuLWhZ8ZoPLxV/ is grep for py-keycli from console log | 21:05 |
| psedlak | it clearly shows how it was switching from 0.3.1 to 0.2.5 and back :( | 21:05 |
| psedlak | dtroyer: so are you the right person for questions about this (and right time)? or should i ask someone else/other time? | 21:06 |
| *** sarob_ has quit IRC | 21:08 | |
| *** sarob has joined #openstack-dev | 21:08 | |
| *** CaptTofu has quit IRC | 21:08 | |
| *** rods has joined #openstack-dev | 21:09 | |
| dtroyer | psedlak: from the standpoint of devstack doing the right/wrong thing, I am. But the details of making current clients work in old stable releases I'll be discovering with everyone else as they happen... | 21:09 |
| dtroyer | psedlak: and this could be noted as one point of not maintaining backward-compatibility with previous releases | 21:09 |
| *** changbl has quit IRC | 21:10 | |
| *** sarob_ has joined #openstack-dev | 21:12 | |
| *** sarob has quit IRC | 21:13 | |
| *** sarob_ has quit IRC | 21:13 | |
| psedlak | dtroyer: this reminds me, i was using master devstack (with checkouts of stable/grizzly of nova, keystone, ...) | 21:13 |
| *** sarob has joined #openstack-dev | 21:13 | |
| psedlak | bcs stable/grizzly devstack failed quickly with 'error: Installed distribution python-keystoneclient 0.2.5 conflicts with requirement python-keystoneclient>=0.3.0' | 21:13 |
| dtroyer | psedlak: oh, don't do that…stable/grizzly devstack is what you want. What has the >=0.3.0 req? | 21:14 |
| psedlak | so maybe that's why/how i get to 'nova-not-working' state | 21:14 |
| psedlak | dtroyer: i've probably lost info for that first try with stable devstack, does it store logs somewhere? | 21:15 |
| dtroyer | look in /opt/stack/logs, stack.log.<timestamp> may still be there | 21:16 |
| *** bswartz has quit IRC | 21:17 | |
| psedlak | dtroyer: :( i don't have that dir at all (i have all repos under /opt/stack/new, but there also no logs dir) | 21:17 |
| psedlak | dtroyer: oh, i've used localrc from gate as base for my own, so i have LOGFILE set there ;) | 21:19 |
| *** vipul is now known as vipul-away | 21:22 | |
| *** nati_ueno has joined #openstack-dev | 21:22 | |
| *** jasdeepH has quit IRC | 21:22 | |
| *** boris-42 has quit IRC | 21:23 | |
| *** henrynash has quit IRC | 21:23 | |
| *** johnthetubaguy has left #openstack-dev | 21:24 | |
| psedlak | dtroyer: oh, ok those deps are same as what i've gathered http://bpaste.net/show/vnYioO66WaD27IC7C1dh/ | 21:25 |
| *** dina_belova has joined #openstack-dev | 21:25 | |
| psedlak | dtroyer: i can see same reasons for who required >=0.3.0 in that devstacklog.... | 21:26 |
| psedlak | dtroyer: basically other clients require new key-cli | 21:26 |
| *** jsavak has joined #openstack-dev | 21:26 | |
| *** dolphm has joined #openstack-dev | 21:28 | |
| *** dina_belova has quit IRC | 21:28 | |
| psedlak | dtroyer: extracted from devstacklog http://bpaste.net/show/OndmU5LbuS6FdrCwmLWp/ | 21:28 |
| *** jasondotstar has quit IRC | 21:29 | |
| *** joesavak has quit IRC | 21:29 | |
| *** dprince has quit IRC | 21:30 | |
| *** henrynash has joined #openstack-dev | 21:30 | |
| *** krtaylor has quit IRC | 21:31 | |
| *** henrynash has quit IRC | 21:31 | |
| *** tong|2 has joined #openstack-dev | 21:32 | |
| *** jtomasek has quit IRC | 21:32 | |
| *** jsavak has quit IRC | 21:32 | |
| *** litong has quit IRC | 21:32 | |
| *** krtaylor has joined #openstack-dev | 21:32 | |
| *** herndon has quit IRC | 21:33 | |
| *** litong has joined #openstack-dev | 21:34 | |
| *** dina_belova has joined #openstack-dev | 21:35 | |
| *** tong|2 has quit IRC | 21:37 | |
| *** donaldh has quit IRC | 21:37 | |
| *** dina_belova has quit IRC | 21:40 | |
| *** jasdeepH has joined #openstack-dev | 21:42 | |
| *** vipul-away is now known as vipul | 21:42 | |
| *** gmurphy has quit IRC | 21:43 | |
| *** ftcjeff has quit IRC | 21:45 | |
| *** Ryan_Lane has quit IRC | 21:47 | |
| *** Ryan_Lane has joined #openstack-dev | 21:47 | |
| cfriesen | I just noticed that in Grizzly regardless of the number of vCPUs the value of /sys/fs/cgroup/cpu/libvirt/qemu/instance-X/cpu.shares seems to be the same. If we were overloaded, this would give all instances the same cpu time regardless of the number of vCPUs in the instance. | 21:49 |
| cfriesen | is this design intent? | 21:49 |
| *** markvoelker has quit IRC | 21:50 | |
| *** jayg is now known as jayg|g0n3 | 21:52 | |
| *** neelashah has quit IRC | 21:52 | |
| *** dmakogon_ has quit IRC | 21:54 | |
| *** maheshp has quit IRC | 21:55 | |
| *** rharwood has quit IRC | 21:55 | |
| gyee | dolphm, ayoung, bknudson, https://review.openstack.org/#/c/33118 | 21:56 |
| *** jjmb has joined #openstack-dev | 21:56 | |
| gyee | I am giving fabio a lift | 21:56 |
| ayoung | gyee, uh oh | 21:56 |
| gyee | just helping the brother out | 21:56 |
| ayoung | should I just -2 it now? | 21:57 |
| gyee | heh | 21:57 |
| ayoung | what's different about it? | 21:57 |
| gyee | ayoung, pretty much cleanup most of the code | 21:57 |
| ayoung | ok | 21:58 |
| gyee | incorporated last round of comments | 21:58 |
| *** mrodden has quit IRC | 21:58 | |
| gyee | ayoung, I am about the push the button on this one https://review.openstack.org/#/c/38367/ | 21:58 |
| gyee | unless you object | 21:59 |
| ayoung | looking | 21:59 |
| ayoung | gyee, all you | 21:59 |
| *** sdake_ has quit IRC | 21:59 | |
| gyee | thank you sir! | 21:59 |
| *** feleouet has quit IRC | 22:00 | |
| *** pixelbeat has joined #openstack-dev | 22:01 | |
| ayoung | gyee, so if there is no filter defined, return all endpoint, right? | 22:01 |
| *** prad_ has quit IRC | 22:02 | |
| *** feleouet has joined #openstack-dev | 22:03 | |
| *** wolfdreamer has quit IRC | 22:06 | |
| gyee | ayoung, correct | 22:09 |
| gyee | if endpoint filter extension is disabled, you'll get the whole motherload | 22:09 |
| *** jecarey has quit IRC | 22:10 | |
| ayoung | gyee, not if it is disabled...if it is enabled, but no filter exists... | 22:10 |
| gyee | ayoung, if enabled and no endpoints assigned, you'll get no endpoints | 22:11 |
| ayoung | gyee, that means that enabling this extension is going to break everyone's client | 22:12 |
| ayoung | default filter should be "no change" | 22:12 |
| ayoung | and then you winnow | 22:12 |
| *** cfriesen has quit IRC | 22:12 | |
| gyee | ayoung, can't, we can't tell if user meant for no endpoint or all endpoints | 22:12 |
| *** cfriesen_ has joined #openstack-dev | 22:13 | |
| *** mkollaro has quit IRC | 22:13 | |
| gyee | if extension is enabled, user meant for endpoint filtering | 22:13 |
| ayoung | gyee, you are an admin. You enable the extension and restart keystone | 22:13 |
| ayoung | now your install is broken | 22:13 |
| ayoung | you have two choices | 22:13 |
| ayoung | disable the extension | 22:13 |
| ayoung | or type furiously | 22:13 |
| ayoung | and add all of the filters | 22:13 |
| gyee | extension is disabled by default | 22:13 |
| ayoung | argh! | 22:13 |
| ayoung | I know that | 22:14 |
| ayoung | you do this deliberately, don't you? | 22:14 |
| ayoung | heh | 22:14 |
| gyee | ayoung, the goal is to establish project-endpoint relationship and filter the endpoints | 22:14 |
| ayoung | gyee, once more from the top, with feeling. you are an admin. You enable the extension and restart keystone, now no one gets any endpoints | 22:14 |
| gyee | the first part has to be done when using this extension | 22:14 |
| ayoung | you need to go through and create all of the filters | 22:15 |
| *** kbringard has quit IRC | 22:15 | |
| gyee | ayoung, that's correct, admin has to plan it out accordingly | 22:15 |
| ayoung | you don't want people doing direct sql | 22:15 |
| ayoung | but...no. | 22:15 |
| ayoung | I can't approve this....it needs to be "if no filter is definied, return all endpoints" | 22:15 |
| ayoung | if and only if a filter is defined to you filter | 22:16 |
| ayoung | I assume that is not too hard to implement | 22:16 |
| *** burt has quit IRC | 22:16 | |
| gyee | ayoung, we can't tell if user mean for filter all by default or do not filter by default | 22:17 |
| gyee | I suppose I can make it a configurable flag | 22:17 |
| ayoung | dolphm, am I crazy here...ok, loaded question, and we know the answer....the real question is, can we allow the endpoint filtering to go in with out a default rule that says "no defined filter means return all endpoints?" | 22:17 |
| ayoung | filter all makes no sense | 22:17 |
| ayoung | like. zero. | 22:17 |
| ayoung | I never want to show anyone my endpoints? | 22:18 |
| dolphm | you could operate by explicitly blacklisting project-endpoint pairs instead of whitelisting them | 22:18 |
| gyee | dolphm, that means change of spec, the whole 9 yard | 22:19 |
| ayoung | dolphm, I think that the whitelist is fine, just we need to provide for rolling in the feature. For large deployments, whitelist is probably more efficient | 22:19 |
| gyee | I though that issue was understood at last summit | 22:19 |
| ayoung | gyee, it is...he is just playing "thought experiement here" | 22:19 |
| gyee | dolphm, ayoung, how about a configurable flag | 22:19 |
| ayoung | gyee, nope | 22:20 |
| dolphm | ayoung: i don't know, it seems inefficient either way to me | 22:20 |
| gyee | dolphm, security dudes would say filter everything :) | 22:20 |
| ayoung | dolphm, you mean "every single project is the same filter except for this little tweak?" Yeah, I can see that, too | 22:20 |
| ayoung | dolphm, blacklist and whitelist? | 22:20 |
| dolphm | gyee: security dudes wouldn't bother recommending that you obscure anything | 22:20 |
| ayoung | allow either filters | 22:20 |
| gyee | ayoung, my magic 8 ball says whitelist | 22:21 |
| ayoung | dolphm, gyee here's what I propose. For now, no filter means existing behavior. Allow everything through. Explicit filter is a whitelist. Follow on work allows blacklist filtering, too | 22:21 |
| dolphm | i really wouldn't bother with both white & black together | 22:22 |
| *** dvarga has quit IRC | 22:22 | |
| ayoung | gyee, assume that your have one end point for everything but...glance. FOr glance, you want to set either glance1 or glance2. Whitelist is enumerate every endpoint. Blacklist is to block out only one endpoint.... | 22:22 |
| ayoung | dolphm, if there are dozens of endpoints of each service, then whitelist makes more sense. If there are a few, then blacklist makes sense. I honestly don't know how people are going to use this | 22:23 |
| dolphm | ayoung: me neither | 22:23 |
| gyee | ayoung, is not about the number of endpoints | 22:24 |
| ayoung | dolphm, since whitelist was designed and approved, lets let that go forward. But I think you make a solid point about blacklists. And I don't see a reason to not discuss it at the summit. | 22:24 |
| gyee | its about controller the endpoints user can do something with | 22:24 |
| ayoung | gyee, it is. It is about keeping two lists in sync | 22:24 |
| gyee | ayoung, there should be only one list | 22:25 |
| ayoung | the total lists of endpoints, and the list of endpoint to project mappings | 22:25 |
| ayoung | if I add an endpoint, I need to go and update each and every service to see that endpoint | 22:25 |
| gyee | ayoung, no | 22:25 |
| ayoung | now, that may make sense in some deployments but not others | 22:25 |
| lbragstad | Hey dolphm quick question on your comment here https://review.openstack.org/#/c/43340/ which would allow me to address the comments here https://review.openstack.org/#/c/41970/11/keystone/identity/core.py | 22:25 |
| gyee | ayoung, then don't enable the extension | 22:26 |
| ayoung | gyee, what if I need the extension for a different reason? | 22:26 |
| gyee | that extension mean for precisely filter the endpoints | 22:26 |
| *** cfriesen_ is now known as cfriesen | 22:26 | |
| ayoung | gyee, yes, it is the whitelist filter. | 22:26 |
| ayoung | And I could see that being useful, but it is not necessarily the right approach for all. | 22:26 |
| ayoung | So, lets leave it in place for now, but discuss the blacklist at the summit, ok? | 22:27 |
| gyee | ayoung, would adding a flag to control that behavior change your outlook? | 22:27 |
| ayoung | But make the default return all endpoints. no filter means do not filter | 22:27 |
| ayoung | flag is not the right level of control | 22:27 |
| fabio | ayoung, but if filter is active and no relationships have been established, why should it return the full list? | 22:28 |
| ayoung | gyee, if no endpoints are assigned to a project, return all endpoints. Anything else is going to be nothing but pain | 22:28 |
| *** vkmc has quit IRC | 22:28 | |
| ayoung | fabio, why would you ever want 0 endpoints returned? | 22:28 |
| ayoung | I mean, we have no service catalog flag for that | 22:28 |
| *** gordc has quit IRC | 22:29 | |
| gyee | ayoung, if the project have no access to any service, why bother return any endpoints | 22:29 |
| *** jhesketh__ has joined #openstack-dev | 22:29 | |
| ayoung | gyee, why bother having a project then? | 22:29 |
| ayoung | gyee, you are not thinking practically. You are holding too tight to the abstraction | 22:30 |
| gyee | ayoung, you can create a project without anything today | 22:30 |
| fabio | ayoung, maybe it should return a link to suggest how to add endpoints to the project instead | 22:30 |
| fabio | since now the extension is active | 22:30 |
| ayoung | fabio, this is not done in a browser | 22:30 |
| ayoung | fabio, so the link would not be automatically processed | 22:30 |
| ayoung | fabio, if there are no endpoints assigned to a project, do not filter. How hard would it be to implement it that way? | 22:31 |
| ayoung | OK...I can think of one reason | 22:31 |
| ayoung | but even that does not really make sense... | 22:31 |
| fabio | ayoung, which would be? | 22:31 |
| ayoung | fabio, to make sure you never allowed anyone to see all of the endpoints | 22:32 |
| ayoung | but that makes no sense | 22:32 |
| *** sushils has quit IRC | 22:32 | |
| ayoung | you would just not add any users to the project before you add the filter | 22:32 |
| gyee | ayoung, there's an upcoming GET /catalog API | 22:32 |
| gyee | ayoung, for large scale deployment, all the projects needs to plan out beforehand | 22:33 |
| ayoung | gyee, ok...tell you what. Add in a flag to allow them to modify the behavior, but make sure the default is "no filter, all endpoints" and I would be satisfied | 22:33 |
| gyee | for mom-n-pop deployment, that filter won't be enabled anyway | 22:33 |
| gyee | ayoung, your satisfaction is guaranteed | 22:34 |
| ayoung | it isn';t even likely | 22:34 |
| ayoung | ok, off I got back to puppet land | 22:34 |
| *** dina_belova has joined #openstack-dev | 22:36 | |
| *** lbragstad has quit IRC | 22:36 | |
| *** fabio has quit IRC | 22:37 | |
| * anteaya quietly passes harlowja a party hat and noisemaker | 22:37 | |
| *** dolphm has quit IRC | 22:37 | |
| anteaya | shhh, don't use them yet though | 22:37 |
| ayoung | dolphm, gyee for your reading pleasure http://adam.younglogic.com/2013/08/deploying-keystone-via-puppet-on-f19/ | 22:37 |
| ayoung | should be the same for ubuntu. I' | 22:38 |
| ayoung | m trying rhel right now, too | 22:38 |
| gyee | ayoung, now you are making me learn puppet | 22:38 |
| *** jaybuff has left #openstack-dev | 22:40 | |
| *** devoid has quit IRC | 22:40 | |
| ayoung | gyee, no. Now I am making it so you don't have to learn puppet. I am learning puppet for you. | 22:41 |
| *** dina_belova has quit IRC | 22:41 | |
| *** fabio has joined #openstack-dev | 22:41 | |
| *** mlavalle has quit IRC | 22:46 | |
| *** gmurphy has joined #openstack-dev | 22:48 | |
| morganfainberg | ayoung: oooh, you going to learn chef for us too? ;) | 22:48 |
| *** sushils has joined #openstack-dev | 22:51 | |
| *** jimfehlig has quit IRC | 22:53 | |
| *** carl_baldwin has quit IRC | 22:54 | |
| *** mrodden has joined #openstack-dev | 22:54 | |
| *** otherwiseguy has joined #openstack-dev | 22:55 | |
| *** aeperezt has quit IRC | 22:57 | |
| *** kbrierly has quit IRC | 22:58 | |
| *** vuil has quit IRC | 23:00 | |
| *** datsun180b has quit IRC | 23:00 | |
| *** sushils has quit IRC | 23:02 | |
| *** rnirmal has quit IRC | 23:03 | |
| *** notmyname has quit IRC | 23:04 | |
| *** notmyname has joined #openstack-dev | 23:04 | |
| *** matiu has quit IRC | 23:05 | |
| *** pabelanger_ has quit IRC | 23:07 | |
| *** pabelanger has joined #openstack-dev | 23:08 | |
| *** notmyname has quit IRC | 23:08 | |
| *** matiu has joined #openstack-dev | 23:09 | |
| *** matiu has quit IRC | 23:09 | |
| *** matiu has joined #openstack-dev | 23:09 | |
| *** notmyname has joined #openstack-dev | 23:09 | |
| *** Tross has quit IRC | 23:10 | |
| *** racktater has joined #openstack-dev | 23:11 | |
| *** jhesketh has quit IRC | 23:11 | |
| *** sdake_ has joined #openstack-dev | 23:12 | |
| *** FunnyLookinHat has quit IRC | 23:12 | |
| insanidade | quick help, please: I just finished running devstack's installer. By the script output, looks like everything was fine. Do I have to execute something so that horizon and the other projects are available ? | 23:13 |
| *** jhesketh has joined #openstack-dev | 23:14 | |
| *** _TheDodd_ has quit IRC | 23:14 | |
| clarkb | insanidade: no they should all be running. The end of the devstack output should tell you what ports they are listening on | 23:17 |
| *** MaxV has joined #openstack-dev | 23:17 | |
| insanidade | clarkb: and it did. but for some reason I can't reach horizon. I think there's no networking problem. | 23:18 |
| insanidade | clarkb: the end of the devstack output presented an address for reaching it. it was the machine's ip. is there something else in the url I should consider? | 23:18 |
| *** markwash has quit IRC | 23:19 | |
| clarkb | insanidade: so, but you may want to make sure that that ip is accessible from wherever you are hitting it | 23:19 |
| insanidade | clarkb: it is. I've tried other services through port 80. | 23:19 |
| *** mrodden1 has joined #openstack-dev | 23:19 | |
| clarkb | does horizon listen on 80 by default? it has been so long since I ran devstack | 23:19 |
| *** nachi has left #openstack-dev | 23:20 | |
| insanidade | clarkb: I think so. The address provided by eh end of the output specifies no port. | 23:20 |
| *** mrodden has quit IRC | 23:20 | |
| clarkb | insanidade: check netstat -lnp ? | 23:20 |
| clarkb | or lsof -i whatever your favorite incantation is | 23:20 |
| *** edmund has quit IRC | 23:21 | |
| insanidade | clarkb: just found the reason. iptables. | 23:22 |
| insanidade | thanks for the support :) | 23:22 |
| clarkb | no problem | 23:22 |
| *** jmontemayor has quit IRC | 23:23 | |
| *** jmontemayor has joined #openstack-dev | 23:24 | |
| *** lexinator has quit IRC | 23:25 | |
| *** Adri2000 has quit IRC | 23:25 | |
| *** noslzzp has quit IRC | 23:25 | |
| *** lexinator has joined #openstack-dev | 23:25 | |
| *** jmontemayor has quit IRC | 23:31 | |
| *** Adri2000 has joined #openstack-dev | 23:33 | |
| *** bswartz has joined #openstack-dev | 23:33 | |
| *** jhesketh has quit IRC | 23:35 | |
| *** MaxV has quit IRC | 23:36 | |
| *** dina_belova has joined #openstack-dev | 23:36 | |
| *** jhesketh has joined #openstack-dev | 23:39 | |
| *** dina_belova has quit IRC | 23:41 | |
| *** jmontemayor has joined #openstack-dev | 23:42 | |
| *** shardy is now known as shardy_afk | 23:42 | |
| *** jmontemayor has quit IRC | 23:42 | |
| *** jmontemayor has joined #openstack-dev | 23:44 | |
| *** sdake_ has quit IRC | 23:44 | |
| *** rods has quit IRC | 23:44 | |
| *** alunduil has joined #openstack-dev | 23:45 | |
| *** fbo is now known as fbo_away | 23:46 | |
| *** sushils has joined #openstack-dev | 23:46 | |
| *** vuil has joined #openstack-dev | 23:48 | |
| *** sthaha has joined #openstack-dev | 23:49 | |
| *** vartom4 has quit IRC | 23:50 | |
| *** michchap has joined #openstack-dev | 23:52 | |
| *** Adri2000 has quit IRC | 23:53 | |
| *** zodiak has quit IRC | 23:55 | |
| *** rcleere has quit IRC | 23:55 | |
| *** insanidade is now known as insanidade-afk | 23:56 | |
| *** jmontemayor has quit IRC | 23:57 | |
| *** danwent has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!