| *** galstrom has joined #openstack-dev | 00:04 | |
| *** halfss has joined #openstack-dev | 00:09 | |
| *** edygarcia has quit IRC | 00:10 | |
| *** sacharya has quit IRC | 00:10 | |
| *** rods_ has quit IRC | 00:14 | |
| *** ncode has joined #openstack-dev | 00:25 | |
| *** halfss has quit IRC | 00:26 | |
| *** ayoung has joined #openstack-dev | 00:28 | |
| *** ayoung has quit IRC | 00:29 | |
| *** ncode has quit IRC | 00:30 | |
| *** edygarcia has joined #openstack-dev | 00:33 | |
| *** winston-d has joined #openstack-dev | 00:49 | |
| *** halfss has joined #openstack-dev | 00:58 | |
| *** galstrom has quit IRC | 01:01 | |
| *** halfss has quit IRC | 01:05 | |
| *** roge has joined #openstack-dev | 01:17 | |
| *** nunosantos has quit IRC | 01:18 | |
| *** littleidea has quit IRC | 01:24 | |
| *** littleidea has joined #openstack-dev | 01:25 | |
| *** sacharya has joined #openstack-dev | 01:28 | |
| *** samkottler has joined #openstack-dev | 01:37 | |
| *** roge has quit IRC | 02:00 | |
| *** amotoki has joined #openstack-dev | 02:01 | |
| *** gongys has joined #openstack-dev | 02:04 | |
| *** anderstj has joined #openstack-dev | 02:07 | |
| *** anderstj has quit IRC | 02:23 | |
| *** bencherian has joined #openstack-dev | 02:28 | |
| *** Ryan_Lane has quit IRC | 02:41 | |
| *** littleidea has quit IRC | 02:42 | |
| *** littleidea has joined #openstack-dev | 02:43 | |
| *** Ryan_Lane has joined #openstack-dev | 02:43 | |
| *** littleidea has joined #openstack-dev | 02:43 | |
| *** littleidea has quit IRC | 02:43 | |
| *** littleidea has joined #openstack-dev | 02:44 | |
| *** Ryan_Lane has quit IRC | 02:45 | |
| *** danwent has quit IRC | 02:54 | |
| *** dtroyer is now known as dtroyer_zzz | 03:00 | |
| *** dtroyer_zzz is now known as dtroyer | 03:04 | |
| *** littleidea has quit IRC | 03:12 | |
| *** samkottler has quit IRC | 03:14 | |
| *** danwent has joined #openstack-dev | 03:15 | |
| *** littleidea has joined #openstack-dev | 03:15 | |
| *** pixelbeat has quit IRC | 03:28 | |
| *** littleidea has quit IRC | 03:33 | |
| *** littleidea has joined #openstack-dev | 03:43 | |
| *** galstrom has joined #openstack-dev | 03:45 | |
| *** pixelbeat has joined #openstack-dev | 03:45 | |
| *** steveb_ has quit IRC | 03:55 | |
| *** gongys has quit IRC | 04:02 | |
| *** shang has quit IRC | 04:19 | |
| *** dtroyer is now known as dtroyer_zzz | 04:34 | |
| *** otter1 has quit IRC | 04:34 | |
| *** galstrom has quit IRC | 04:37 | |
| *** galstrom has joined #openstack-dev | 04:39 | |
| *** dtroyer_zzz is now known as dtroyer | 04:47 | |
| *** garyk has quit IRC | 04:51 | |
| *** asakhnov has joined #openstack-dev | 05:03 | |
| *** galstrom has quit IRC | 05:05 | |
| *** galstrom has joined #openstack-dev | 05:08 | |
| *** jsgotangco has joined #openstack-dev | 05:16 | |
| *** davidha has quit IRC | 05:18 | |
| *** jsgotangco has quit IRC | 05:18 | |
| *** davidha has joined #openstack-dev | 05:18 | |
| *** sacharya has quit IRC | 05:20 | |
| *** davidha has quit IRC | 05:23 | |
| *** edygarcia has quit IRC | 05:25 | |
| *** garyk has joined #openstack-dev | 05:25 | |
| *** asakhnov has quit IRC | 05:26 | |
| *** asakhnov has joined #openstack-dev | 05:49 | |
| *** garyk has quit IRC | 05:53 | |
| *** galstrom has quit IRC | 06:00 | |
| *** hattwick has quit IRC | 06:02 | |
| *** davidha has joined #openstack-dev | 06:02 | |
| *** dtroyer is now known as dtroyer_zzz | 06:11 | |
| *** Mandell has quit IRC | 06:27 | |
| *** Mandell has joined #openstack-dev | 06:32 | |
| *** al-maisan has joined #openstack-dev | 06:37 | |
| *** littleidea has quit IRC | 06:42 | |
| *** dachary has quit IRC | 06:51 | |
| *** mindpixel has joined #openstack-dev | 06:53 | |
| *** eglynn has joined #openstack-dev | 07:00 | |
| *** hugokuo1 has quit IRC | 07:02 | |
| *** hugokuo1 has joined #openstack-dev | 07:03 | |
| *** davidha has quit IRC | 07:03 | |
| *** davidha has joined #openstack-dev | 07:04 | |
| *** davidha has quit IRC | 07:10 | |
| *** reidrac has joined #openstack-dev | 07:11 | |
| *** asalkeld has quit IRC | 07:16 | |
| *** asalkeld has joined #openstack-dev | 07:16 | |
| *** dachary has joined #openstack-dev | 07:18 | |
| *** anderstj has joined #openstack-dev | 07:20 | |
| *** markmc has joined #openstack-dev | 07:30 | |
| *** ghe has joined #openstack-dev | 07:33 | |
| *** ghe is now known as GheRivero | 07:33 | |
| *** bencherian has quit IRC | 07:38 | |
| *** anderstj has quit IRC | 07:39 | |
| *** sulochan has joined #openstack-dev | 07:42 | |
| *** m4xmr has joined #openstack-dev | 07:46 | |
| *** Hien has quit IRC | 07:51 | |
| *** Hien has joined #openstack-dev | 07:53 | |
| *** m4xmr has left #openstack-dev | 07:55 | |
| *** derekh has joined #openstack-dev | 07:56 | |
| *** Hien has quit IRC | 08:04 | |
| *** Hien has joined #openstack-dev | 08:05 | |
| *** erikzaadi has joined #openstack-dev | 08:09 | |
| *** darraghb has joined #openstack-dev | 08:20 | |
| *** danwent has quit IRC | 08:24 | |
| *** apevec has joined #openstack-dev | 08:28 | |
| *** anniec has joined #openstack-dev | 08:39 | |
| *** anniec_ has joined #openstack-dev | 08:40 | |
| *** epim has quit IRC | 08:40 | |
| *** hattwick has joined #openstack-dev | 08:43 | |
| *** anniec has quit IRC | 08:44 | |
| *** anniec_ is now known as anniec | 08:44 | |
| *** steveb_ has joined #openstack-dev | 08:47 | |
| *** danpb has joined #openstack-dev | 08:48 | |
| *** Mandell has quit IRC | 08:54 | |
| *** maploin has joined #openstack-dev | 08:57 | |
| *** maploin has quit IRC | 08:57 | |
| *** maploin has joined #openstack-dev | 08:57 | |
| *** johngarbutt has joined #openstack-dev | 09:02 | |
| eglynn | markmc: would it make sense to split bug #1026600 into two bugs, one for the Location header, another for the 201 status code? (in the light of this discussion on the ML ...) | 09:05 |
|---|---|---|
| uvirtbot | Launchpad bug 1026600 in nova "Volume creation 201 API response does not include a Location header" [Low,In progress] https://launchpad.net/bugs/1026600 | 09:05 |
| eglynn | markmc: http://lists.openstack.org/pipermail/openstack-dev/2012-July/000132.html | 09:05 |
| eglynn | markmc: (not very conventional to return Location with "200 OK" but not explicitly disallowed by the spec either AFAICS) | 09:05 |
| markmc | eglynn, I'm gonna respond later, but I think the backwards compat concerns are a bit BS | 09:06 |
| markmc | eglynn, do we know of any clients which would be affected? do we think they're buggy | 09:06 |
| markmc | eglynn, I think you're suffering such concerns just because you posted to the list :) | 09:06 |
| markmc | eglynn, e.g. 338ef025 | 09:07 |
| markmc | eglynn, merged on friday, switches a response from 200 to 202 | 09:07 |
| eglynn | markmc: yep, my view is maintaining bug compatability is not strictly required, but the ML discussion was heavily skewed to the other opinion | 09:07 |
| * eglynn looking ... | 09:07 | |
| eglynn | makrmc: yep, that is a counter-example all right! | 09:08 |
| markmc | eglynn, just replied | 09:15 |
| eglynn | markmc: just got it, thanks | 09:20 |
| * eglynn v. impressed with the speed-of-delivery on the new ML! | 09:21 | |
| eglynn | lightening-fast compared to the old ... | 09:21 |
| *** m4xmr has joined #openstack-dev | 09:31 | |
| *** rods_ has joined #openstack-dev | 09:39 | |
| *** RoboticCheese has quit IRC | 10:10 | |
| *** steveb_ has quit IRC | 10:10 | |
| *** RoboticCheese has joined #openstack-dev | 10:10 | |
| *** GheRivero has quit IRC | 10:15 | |
| *** ghe has joined #openstack-dev | 10:18 | |
| *** nunosantos has joined #openstack-dev | 10:19 | |
| *** ghe has quit IRC | 10:21 | |
| *** garyk has joined #openstack-dev | 10:41 | |
| *** winston-d has quit IRC | 10:42 | |
| *** ghe has joined #openstack-dev | 10:47 | |
| *** ghe is now known as GheRivero | 10:49 | |
| *** milner has joined #openstack-dev | 11:11 | |
| *** thovden has joined #openstack-dev | 11:11 | |
| eglynn | markmc: can I be a pain and harrass you about https://review.openstack.org/9534? | 11:14 |
| *** samkottler has joined #openstack-dev | 11:16 | |
| markmc | eglynn, thanks for the reminder, +2ed | 11:16 |
| eglynn | markmc: thank you sir! | 11:17 |
| *** dachary has quit IRC | 11:52 | |
| markmc | russellb, there? | 11:57 |
| markmc | russellb, sorry, too early - sent mail | 11:58 |
| *** salgado has joined #openstack-dev | 12:03 | |
| *** salgado has quit IRC | 12:09 | |
| *** darraghb has quit IRC | 12:09 | |
| *** Ruetobas has quit IRC | 12:09 | |
| *** markvoelker has quit IRC | 12:09 | |
| *** uvirtbot has quit IRC | 12:09 | |
| *** jgriffith_away has quit IRC | 12:09 | |
| *** chmouel has quit IRC | 12:09 | |
| *** Xtrapni has quit IRC | 12:09 | |
| *** bhuvan_ has quit IRC | 12:09 | |
| *** justinsb has quit IRC | 12:09 | |
| *** nikhil has quit IRC | 12:09 | |
| *** bcwaldon has quit IRC | 12:09 | |
| *** yazirian has quit IRC | 12:09 | |
| *** ameade has quit IRC | 12:09 | |
| *** pballand has quit IRC | 12:09 | |
| *** salgado has joined #openstack-dev | 12:10 | |
| *** darraghb has joined #openstack-dev | 12:10 | |
| *** Ruetobas has joined #openstack-dev | 12:10 | |
| *** uvirtbot has joined #openstack-dev | 12:10 | |
| *** jgriffith_away has joined #openstack-dev | 12:10 | |
| *** chmouel has joined #openstack-dev | 12:10 | |
| *** Xtrapni has joined #openstack-dev | 12:10 | |
| *** bhuvan_ has joined #openstack-dev | 12:10 | |
| *** justinsb has joined #openstack-dev | 12:10 | |
| *** nikhil has joined #openstack-dev | 12:10 | |
| *** bcwaldon has joined #openstack-dev | 12:10 | |
| *** yazirian has joined #openstack-dev | 12:10 | |
| *** ameade has joined #openstack-dev | 12:10 | |
| *** pballand has joined #openstack-dev | 12:10 | |
| *** Xtrapni has quit IRC | 12:10 | |
| *** markvoelker has joined #openstack-dev | 12:10 | |
| *** m4xmr has left #openstack-dev | 12:10 | |
| *** trapni has joined #openstack-dev | 12:10 | |
| *** galstrom has joined #openstack-dev | 12:11 | |
| *** uvirtbot has quit IRC | 12:12 | |
| *** uvirtbot has joined #openstack-dev | 12:12 | |
| *** dachary has joined #openstack-dev | 12:21 | |
| *** thovden has quit IRC | 12:21 | |
| *** markvoelker has quit IRC | 12:21 | |
| *** salgado has quit IRC | 12:21 | |
| *** darraghb has quit IRC | 12:21 | |
| *** Ruetobas has quit IRC | 12:21 | |
| *** jgriffith_away has quit IRC | 12:21 | |
| *** chmouel has quit IRC | 12:21 | |
| *** bhuvan_ has quit IRC | 12:21 | |
| *** justinsb has quit IRC | 12:21 | |
| *** nikhil has quit IRC | 12:21 | |
| *** bcwaldon has quit IRC | 12:21 | |
| *** yazirian has quit IRC | 12:21 | |
| *** ameade has quit IRC | 12:21 | |
| *** pballand has quit IRC | 12:21 | |
| *** salgado has joined #openstack-dev | 12:22 | |
| *** darraghb has joined #openstack-dev | 12:22 | |
| *** Ruetobas has joined #openstack-dev | 12:22 | |
| *** jgriffith_away has joined #openstack-dev | 12:22 | |
| *** chmouel has joined #openstack-dev | 12:22 | |
| *** bhuvan_ has joined #openstack-dev | 12:22 | |
| *** justinsb has joined #openstack-dev | 12:22 | |
| *** nikhil has joined #openstack-dev | 12:22 | |
| *** bcwaldon has joined #openstack-dev | 12:22 | |
| *** yazirian has joined #openstack-dev | 12:22 | |
| *** ameade has joined #openstack-dev | 12:22 | |
| *** pballand has joined #openstack-dev | 12:22 | |
| *** markvoelker has joined #openstack-dev | 12:22 | |
| *** markvoelker has quit IRC | 12:23 | |
| *** markvoelker has joined #openstack-dev | 12:23 | |
| *** drewlander has joined #openstack-dev | 12:25 | |
| *** littleidea has joined #openstack-dev | 12:31 | |
| *** littleidea has joined #openstack-dev | 12:33 | |
| *** littleidea has quit IRC | 12:33 | |
| *** littleidea has joined #openstack-dev | 12:34 | |
| *** galstrom has quit IRC | 12:38 | |
| *** chrisfer has joined #openstack-dev | 12:39 | |
| *** halfss has joined #openstack-dev | 12:41 | |
| *** GheRivero has quit IRC | 12:51 | |
| *** sacharya has joined #openstack-dev | 12:52 | |
| *** ayoung has joined #openstack-dev | 12:53 | |
| *** littleidea has quit IRC | 12:53 | |
| *** jgriffith_away is now known as jgriffith | 12:57 | |
| *** dprince has joined #openstack-dev | 13:00 | |
| *** sacharya has quit IRC | 13:01 | |
| *** chrisfer has joined #openstack-dev | 13:01 | |
| *** davidkranz_ is now known as davidkranz | 13:02 | |
| *** salv-orlando has joined #openstack-dev | 13:08 | |
| *** kbringard has joined #openstack-dev | 13:10 | |
| jgriffith | jeblair: Ping | 13:11 |
| *** kbringard has quit IRC | 13:11 | |
| *** kbringard has joined #openstack-dev | 13:11 | |
| *** ewindisch has joined #openstack-dev | 13:16 | |
| *** jgriffith has quit IRC | 13:18 | |
| *** GheRivero has joined #openstack-dev | 13:19 | |
| *** jgriffith has joined #openstack-dev | 13:24 | |
| *** timello has quit IRC | 13:25 | |
| davidkranz | markmc: "backwards compat concerns are a bit BS" are pretty harsh words. If that is really the sentiment of the OpenStack dev community, this needs to be taken to a forum that includes users and operators. | 13:35 |
| *** zaneb has joined #openstack-dev | 13:35 | |
| markmc | davidkranz, I was referring to this specific concern, not backwards compat in general | 13:35 |
| markmc | davidkranz, I think compat is very important | 13:35 |
| davidkranz | markmc: But if I am an operator and I take this change, some random user's code could stop working, right? | 13:36 |
| davidkranz | markmc: Why is it so important to make this change to fix a minor design bug? | 13:36 |
| markmc | davidkranz, potentially, but this true of a lot of bug fixes (that user's code relying on the bug stops working) | 13:37 |
| markmc | davidkranz, it's not a design bug and it's important we fix bugs and have a consistent API | 13:37 |
| markmc | davidkranz, no-one would thank us if we added v3 for this | 13:37 |
| davidkranz | markmc: I completely agree with your last point. But not one would curse at us if we didn't make this change either. | 13:38 |
| markmc | davidkranz, broken window syndrome - an inconsistent API will get more inconsistent over time; we should be rectifying bugs like this | 13:39 |
| davidkranz | markmc: Well, I guess whoever is running this show (whoever that is) will have to decide. But any change that impacts operators and users in this way should be published on a non-dev list. | 13:41 |
| markmc | davidkranz, I honestly thing you're blowing this out of proportion | 13:42 |
| markmc | davidkranz, do you know of any clients would be affected? | 13:42 |
| markmc | davidkranz, do you realize these kind of fixes are quite commonly made anyway? | 13:43 |
| markmc | davidkranz, e.g. like the one that was merged on Friday? | 13:43 |
| davidkranz | markmc: I understand that. | 13:43 |
| davidkranz | markmc: I have no idea of all the applicatinos out there using Openstack compute API v2 so how could know who would be affected? | 13:44 |
| *** sacharya has joined #openstack-dev | 13:44 | |
| markmc | davidkranz, right, no concrete data of this causing problems - so all I have is my instinct which says it probably won't | 13:45 |
| *** dtroyer_zzz is now known as dtroyer | 13:45 | |
| markmc | davidkranz, and I don't see the need to consult more widely on this - no more than other similar bug fixes | 13:45 |
| davidkranz | markmc: I really think we are talking past each other. Operators don't need to know that their user's code might stop working? | 13:47 |
| markmc | davidkranz, operators benefit from warnings about changes that might break stuff | 13:47 |
| markmc | davidkranz, they don't need to be notified about every bug fix (well, honestly they want to be - but they can do that themselves) | 13:48 |
| markmc | davidkranz, there's a threshold - I don't think this goes over it | 13:48 |
| *** rpedde_away is now known as rpedde | 13:49 | |
| davidkranz | markmc: I wish I could agree with you. If I am in a tiny minority with this point of view I will shut up. | 13:49 |
| *** zaitcev has joined #openstack-dev | 13:50 | |
| markmc | davidkranz, IMHO it only appears to be significant because you noticed it, which is only because it was discussed on list | 13:50 |
| markmc | davidkranz, fixes like this are quite regular occurrences I think | 13:51 |
| *** Dr_Who has joined #openstack-dev | 13:51 | |
| *** markmcclain has joined #openstack-dev | 13:51 | |
| davidkranz | markmc: That is completely untrue. I noticed it because the tempest gate failed. | 13:51 |
| markmc | davidkranz, wait, what? this caused a tempest test to fail? | 13:52 |
| davidkranz | markmc: The fact that tempest did not see the other changes represents the lack of complete coverage in tempset. | 13:52 |
| davidkranz | markmc: https://review.openstack.org/#/c/10058/ | 13:52 |
| markmc | davidkranz, am I misreading? that failure looks unrelated to the change? | 13:53 |
| davidkranz | markmc: The change was to return 201 instead of 200 right? | 13:54 |
| jgriffith | davidkranz: markmc: The tempest change was intended to handle *both* | 13:54 |
| davidkranz | markmc: Yes, temporarily until it could be really changed. Needed to get around the tempest failure. | 13:55 |
| markmc | davidkranz, nevermind, misread | 13:55 |
| markmc | self.assertEqual(200, resp.status) | 13:55 |
| markmc | meh | 13:55 |
| markmc | davidkranz, could you summarize the above on the mailing list thread? it's good context | 13:55 |
| *** sacharya has quit IRC | 13:55 | |
| davidkranz | markmc: Sure. | 13:55 |
| *** sacharya has joined #openstack-dev | 13:56 | |
| *** edygarcia has joined #openstack-dev | 13:56 | |
| *** sacharya has left #openstack-dev | 13:58 | |
| *** roge has joined #openstack-dev | 14:04 | |
| *** utlemming has joined #openstack-dev | 14:04 | |
| *** timello has joined #openstack-dev | 14:05 | |
| *** AlanClark has joined #openstack-dev | 14:06 | |
| *** al-maisan is now known as almaisan-away | 14:08 | |
| *** galstrom has joined #openstack-dev | 14:11 | |
| *** sacharya has joined #openstack-dev | 14:14 | |
| *** matiu has quit IRC | 14:14 | |
| *** erikzaadi has quit IRC | 14:15 | |
| *** dtroyer is now known as dtroyer_zzz | 14:18 | |
| *** renier has joined #openstack-dev | 14:19 | |
| russellb | jkoelker: hey, i've got a few openstack-common patches pending that are blocking a nova patch series. if you have some time today, could you take a look? | 14:22 |
| *** Gordonz has joined #openstack-dev | 14:23 | |
| *** Gordonz has quit IRC | 14:23 | |
| *** jaypipes has joined #openstack-dev | 14:23 | |
| *** mnewby has joined #openstack-dev | 14:24 | |
| *** Gordonz has joined #openstack-dev | 14:24 | |
| *** timello has quit IRC | 14:26 | |
| jgriffith | ttx: thanks! | 14:26 |
| ttx | jgriffith: np :) need to look into devstack since I suspect it needs to be updated as well to install the new rootwrap configfiles | 14:26 |
| ttx | this should really move to some common project, will look into that | 14:27 |
| jgriffith | ttx: I think my changes from Friday should cover it. I just need to figure out how to get tempest to run so I can submit the final switch over | 14:27 |
| jgriffith | ttx: I basicly just took the nova rootwrap changes and put them into lib/cinder | 14:28 |
| *** matiu has joined #openstack-dev | 14:28 | |
| *** matiu has quit IRC | 14:28 | |
| *** matiu has joined #openstack-dev | 14:28 | |
| ttx | jgriffith: ok, should be ok then, lets see if the tests pass | 14:28 |
| *** maoy has joined #openstack-dev | 14:30 | |
| *** flaviamissi has joined #openstack-dev | 14:30 | |
| dansmith | russellb: oh sure, let *him* change something to a synchronous call! :D | 14:32 |
| russellb | ha | 14:33 |
| russellb | what was it that you changed though? | 14:33 |
| dansmith | I'm kidding, I know the difference | 14:33 |
| russellb | k .. just wanted to make sure i'm not being inconsistent | 14:33 |
| dansmith | I'm annoyed though, because if this slips in before jaypipes comes back and +1s my patch from friday, I'll have to rebase :) | 14:33 |
| russellb | was yours something that takes a while? | 14:34 |
| dansmith | russellb: no, just poking fun | 14:34 |
| *** anderstj has joined #openstack-dev | 14:34 | |
| russellb | i'll have to rebase a bunch of patches if it goes in ... | 14:34 |
| russellb | i have a series of stuff that changes the rpc api and bumps the version every time, and so does this | 14:34 |
| dansmith | mine was start/stop, which I didn't think was much of a blocking operation on the backend, but you and others said it was | 14:34 |
| russellb | so it will cause a conflict in every single patch in the series :) | 14:34 |
| dansmith | nice :) | 14:34 |
| russellb | well feel free to call me out if i'm wrong, heh | 14:35 |
| dansmith | I figured that setting the password required interacting with the guest or opening up the root disk and thus could take a while, but, it's not a critical path thing I guess | 14:35 |
| *** galstrom has left #openstack-dev | 14:35 | |
| jaypipes | danpb: sorry... link for me? | 14:36 |
| jaypipes | dansmith: ^^ | 14:36 |
| dansmith | jaypipes: https://review.openstack.org/#/c/9932/ | 14:36 |
| *** blamar has joined #openstack-dev | 14:37 | |
| *** anderstj has quit IRC | 14:37 | |
| *** blamar has quit IRC | 14:37 | |
| jaypipes | dansmith: done. | 14:38 |
| *** dtroyer_zzz is now known as dtroyer | 14:38 | |
| dansmith | jaypipes: thanks man :) | 14:38 |
| dansmith | comstud: care to finish us off? https://review.openstack.org/#/c/9932/ | 14:38 |
| jgriffith | jaypipes: Trying to get devstack changes in, can't get past gate tests https://jenkins.openstack.org/job/gate-tempest-devstack-vm/4400/console | 14:38 |
| jgriffith | jaypipes: Any tips/docs for how to run this on my own devstack system to debug? | 14:39 |
| *** blamar has joined #openstack-dev | 14:39 | |
| jaypipes | jgriffith: looking now.. | 14:39 |
| *** timello has joined #openstack-dev | 14:39 | |
| jaypipes | jgriffith: search for "MySQL server has gone away" on https://jenkins.openstack.org/job/gate-tempest-devstack-vm/4400/artifact/logs/syslog.txt | 14:41 |
| *** littleidea has joined #openstack-dev | 14:41 | |
| jaypipes | jgriffith: something's up with the MySQL server used in testing it seems ... | 14:41 |
| jgriffith | jaypipes: Ahh... thanks! | 14:42 |
| jaypipes | jgriffith: not sure if it has a lick to do with your patch though... | 14:42 |
| jgriffith | jaypipes: Well, me neither :) | 14:42 |
| jgriffith | jaypipes: Is it possible to run tempest on my own setup to step through? Or is that non-trivial? | 14:43 |
| jaypipes | jgriffith: FYI, I usually go to the test result page: https://jenkins.openstack.org/job/gate-tempest-devstack-vm/4400/ and click the plus sign next to the "artifacts" area and search through those files for errors.. | 14:43 |
| jaypipes | jgriffith: absolutely possible. is your local a devstack setup or non-devstack setup? | 14:43 |
| jgriffith | jaypipes: devstack... used vagrant | 14:43 |
| jgriffith | jaypipes: Have tempest cloned, but now sure on the config | 14:44 |
| *** asakhnov has quit IRC | 14:44 | |
| dansmith | jgriffith: tools/configure_tempest.sh :) | 14:44 |
| dansmith | (in devstack) | 14:44 |
| jaypipes | jgriffith: there's a helpful little script in devstack :) tools/configure_tempest.sh | 14:44 |
| jaypipes | jgriffith: it assumes, however, that tempest is installed to /opt/stack/tempest | 14:44 |
| jgriffith | dansmith: jaypipes: Ahaaa.. nice | 14:44 |
| jaypipes | jgriffith: so clone tempest into there first, then run it. | 14:44 |
| jgriffith | jaypipes: Cool, that's where I put it | 14:44 |
| jgriffith | jaypipes: Trying now | 14:45 |
| jaypipes | jgriffith: http://paste.openstack.org/show/19540/ | 14:45 |
| jaypipes | jgriffith: that's an easy little script I use to reset my tempest env from devstack... | 14:45 |
| jgriffith | jaypipes: Nice.. thanks! | 14:45 |
| jaypipes | jgriffith: easy enough to mod for your needs | 14:46 |
| jeblair | jgriffith: pong | 14:46 |
| jgriffith | jeblair: Hey there... | 14:46 |
| jgriffith | jeblair: Was going to ask you about debugging the tempest failures on that patch from Friday | 14:46 |
| jgriffith | jeblair: Got some help from jaypipes and dansmith | 14:46 |
| jgriffith | jeblair: Running it now | 14:46 |
| jeblair | jgriffith: ah, cool | 14:47 |
| *** datsun180b has joined #openstack-dev | 14:50 | |
| *** tryggvil__ has joined #openstack-dev | 14:51 | |
| *** edygarcia_ has joined #openstack-dev | 14:56 | |
| *** edygarcia has quit IRC | 14:58 | |
| *** edygarcia_ is now known as edygarcia | 14:58 | |
| *** lorin1 has joined #openstack-dev | 14:58 | |
| *** e1mer has quit IRC | 15:00 | |
| *** cmagina-lunch has joined #openstack-dev | 15:02 | |
| *** edygarcia has quit IRC | 15:03 | |
| jkoelker | russellb: doneskis | 15:03 |
| russellb | thanks! | 15:03 |
| *** cmagina has quit IRC | 15:03 | |
| *** reidrac has quit IRC | 15:04 | |
| *** asakhnov has joined #openstack-dev | 15:05 | |
| *** thovden has joined #openstack-dev | 15:11 | |
| *** ecarlin has joined #openstack-dev | 15:12 | |
| *** rnirmal has joined #openstack-dev | 15:12 | |
| *** sstent has quit IRC | 15:13 | |
| *** andrea_ has joined #openstack-dev | 15:14 | |
| *** sstent has joined #openstack-dev | 15:14 | |
| *** ecarlin has quit IRC | 15:15 | |
| *** mindpixel has quit IRC | 15:18 | |
| ttx | vishy: could you revive https://review.openstack.org/#/c/9041/ in order to fix https://bugs.launchpad.net/nova/+bug/1017418 for stable/essex ? | 15:18 |
| uvirtbot | Launchpad bug 1017418 in nova/essex "NoMoreFloatingIps: Zero floating ips available after repeatedly creating and destroying instances over time" [Critical,In progress] | 15:18 |
| *** dachary has quit IRC | 15:20 | |
| *** amotoki_ has joined #openstack-dev | 15:36 | |
| *** amotoki has quit IRC | 15:36 | |
| *** heckj has joined #openstack-dev | 15:38 | |
| *** anniec has quit IRC | 15:40 | |
| *** dachary has joined #openstack-dev | 15:41 | |
| *** matiu has quit IRC | 15:44 | |
| *** PotHix has joined #openstack-dev | 15:46 | |
| *** halfss has quit IRC | 15:47 | |
| andrewbogott | _cerberus_, could I get your comments on https://review.openstack.org/#/c/10035/? (That's the patch that makes the list notifier the standard behavior for the notification driver config.) | 15:47 |
| *** littleidea_ has joined #openstack-dev | 15:47 | |
| _cerberus_ | Yep, I can take a look in a bit | 15:47 |
| andrewbogott | thanks | 15:48 |
| *** cmagina-lunch is now known as cmagina | 15:50 | |
| *** PotHix has quit IRC | 15:51 | |
| *** littleidea has quit IRC | 15:52 | |
| *** littleidea_ is now known as littleidea | 15:52 | |
| *** reed has joined #openstack-dev | 15:54 | |
| *** maploin has quit IRC | 15:55 | |
| jgriffith | jeblair: Odd, what's failing actually is the delete call for the volume it succesfully created | 15:56 |
| *** amotoki_ is now known as amotoki | 15:57 | |
| jgriffith | jeblair: Hmm... It looks like the endpoint is wrong? | 15:59 |
| *** andrea_ has left #openstack-dev | 15:59 | |
| *** PotHix has joined #openstack-dev | 16:01 | |
| *** garyk has quit IRC | 16:01 | |
| *** matiu has joined #openstack-dev | 16:02 | |
| *** matiu has quit IRC | 16:02 | |
| *** matiu has joined #openstack-dev | 16:02 | |
| *** belliott has quit IRC | 16:03 | |
| *** belliott has joined #openstack-dev | 16:03 | |
| *** mnewby has quit IRC | 16:04 | |
| *** edygarcia has joined #openstack-dev | 16:06 | |
| *** garyk has joined #openstack-dev | 16:07 | |
| *** belliott has quit IRC | 16:07 | |
| *** amotoki has quit IRC | 16:08 | |
| *** belliott has joined #openstack-dev | 16:08 | |
| *** Slower has joined #openstack-dev | 16:13 | |
| *** jtran has joined #openstack-dev | 16:15 | |
| *** edygarcia has quit IRC | 16:18 | |
| *** nati_ueno has joined #openstack-dev | 16:20 | |
| jeblair | jgriffith: you probably want jaypipes for that; i'm not familiar with how that's configured or used by tempest | 16:21 |
| *** edygarcia has joined #openstack-dev | 16:21 | |
| jgriffith | jeblair: Yeah, that's likely not the problem anyway... just thinking out loud and talking to you makes me look less insaine than talking to my dog | 16:23 |
| jeblair | jgriffith: whenever you ask me a question, i just repeat it to the bird out my window, so... | 16:23 |
| jgriffith | jeblair: LOL | 16:24 |
| jaypipes | jgriffith: you probably want jeblair for that. I'm just a dumbass. | 16:25 |
| jgriffith | jaypipes: Hmm... I'm noticing a patern here | 16:25 |
| jaypipes | jgriffith: :) | 16:25 |
| *** danpb has quit IRC | 16:25 | |
| jaypipes | jgriffith: so the create call is working but the delete (of the same volume UUID) isn't? | 16:26 |
| jgriffith | jaypipes: So yeah... here's a paste http://paste.openstack.org/show/19545/ | 16:26 |
| jgriffith | jaypipes: At first I noticed the url was xxx:8776 and thought that was the issue but don't think that's true now (since create works fine) | 16:27 |
| *** edygarcia has joined #openstack-dev | 16:27 | |
| jaypipes | looking... | 16:27 |
| jgriffith | jaypipes: Also here's the url for the delete: u'http://192.168.137.6:8774/v2/0307e291310a4f72a8edf2b4887d30fc/os-volumes/b8742911-8a02-467c-bd23-23cdaf4e68c8' | 16:29 |
| jgriffith | jaypipes: I need to run again and catch create | 16:29 |
| jaypipes | jgriffith: wait a minute... | 16:30 |
| jaypipes | jgriffith: that's a different UUID than the ID of the volume in the cinder log above. | 16:30 |
| *** edygarcia has quit IRC | 16:31 | |
| jgriffith | jaypipes: Yeah, ignore that :) Sorry, different run | 16:31 |
| jaypipes | ah, k | 16:32 |
| *** jtran has quit IRC | 16:32 | |
| *** thovden has quit IRC | 16:34 | |
| jaypipes | jgriffith: so I have a suspicion that the 404 is actually referring to something other than the volume itself... | 16:35 |
| *** nati_ueno has quit IRC | 16:36 | |
| jaypipes | jgriffith: perhaps some foreign key is being deleted before the volume record itself? | 16:36 |
| jgriffith | jaypipes: hmmmm | 16:36 |
| jgriffith | jaypipes: That would make a lot more sense | 16:36 |
| jgriffith | jaypipes: okee... tracing through it again | 16:36 |
| *** edygarcia_ has joined #openstack-dev | 16:37 | |
| *** jtran has joined #openstack-dev | 16:38 | |
| jaypipes | jgriffith: hold on a sec :) | 16:38 |
| jaypipes | jgriffith: so looking back at this log... everything is actually kosher. | 16:38 |
| jgriffith | jaypipes: do tell? | 16:38 |
| jaypipes | jgriffith: the DELETE is called, returns a 202, then a GET is called and of course it returns a 404 because the volume is deleted. | 16:39 |
| jgriffith | jaypipes: Ahh, yes.. that's the verification of the delete | 16:39 |
| jaypipes | right! | 16:39 |
| jaypipes | jgriffith: I bet it's just the test case that needs to be refactored a bit. | 16:40 |
| jaypipes | jgriffith: looks to me like there is a loop of GETs until a 404 is returned after the delete. | 16:40 |
| jgriffith | jaypipes: Well I traced it with PDB and the actual delete request did throw the COMPUTE exception | 16:40 |
| *** edygarcia_ has quit IRC | 16:40 | |
| jaypipes | jgriffith: all I see is the cinder log above... | 16:41 |
| jaypipes | jgriffith: and all that seems kosher to me./ | 16:41 |
| *** sniperd has joined #openstack-dev | 16:41 | |
| jgriffith | jaypipes: so in rest_client.py (tempest) I stepped through the delete request and it threw the copute fault | 16:42 |
| *** garyk has quit IRC | 16:42 | |
| jgriffith | jaypipes: I introduced a red herring I think with the cinder log I sent :( | 16:42 |
| jaypipes | jgriffith: do you have a paste of that traceback? | 16:42 |
| jgriffith | jaypipes: Nope but I'm running it again now with a clean config | 16:43 |
| jaypipes | kk | 16:43 |
| jaypipes | on a totally unrelated note, it would be great if someone figured out how to get rid of all the damn "Exception AssertionError: AssertionError() in <module 'threading' from '/usr/lib/python2.7/threading.pyc'> ignored" messages in log output. :( | 16:43 |
| jgriffith | jaypipes: Amen brother!! I tried the patch that's out there without luck | 16:44 |
| *** Mandell has joined #openstack-dev | 16:50 | |
| *** nunosantos has quit IRC | 16:52 | |
| jaypipes | zul: this something you can assign to the appropriate Ubuntu-er? :) https://bugs.launchpad.net/ubuntu/+source/glance/+bug/1027719 | 16:54 |
| uvirtbot | Launchpad bug 1027719 in glance "glance-client package has incorrect description" [Undecided,Confirmed] | 16:54 |
| *** jaypipes has left #openstack-dev | 16:54 | |
| *** jaypipes has joined #openstack-dev | 16:54 | |
| zul | jaypipes: yep | 16:55 |
| *** jdurgin has joined #openstack-dev | 16:55 | |
| zul | jaypipes: the approiate person is probably me though | 16:55 |
| jaypipes | zul: ah, ok then! :) | 16:56 |
| *** johnpur has joined #openstack-dev | 16:59 | |
| *** nati_uen_ has joined #openstack-dev | 16:59 | |
| *** nati_uen_ has quit IRC | 16:59 | |
| *** nati_uen_ has joined #openstack-dev | 17:00 | |
| *** johnpur has quit IRC | 17:01 | |
| eglynn | mtaylor, jeblair: does this look like a breakage in the Jenkins config for gating Tempest changes: https://jenkins.openstack.org/job/gate-tempest-pep8/285/consoleFull | 17:01 |
| eglynn | (i.e. the ERROR: toxini file '/home/jenkins/workspace/gate-tempest-pep8/tox.ini' does not exist) | 17:01 |
| mtaylor | eglynn: I think we were talking to/working with jaypipes on that on friday | 17:01 |
| mtaylor | jaypipes: do you remember what it was that we decided to do abot that? it was me wasn't it. I was supposed to fix it... | 17:02 |
| eglynn | mtaylor: a-ha, cool ... | 17:02 |
| jaypipes | mtaylor: actually, you were going to submit a patch to make tempest toxified :) | 17:02 |
| mtaylor | yup. I was. then I got derailed by thinking I should pbr it too ... but then I went down a rathole | 17:03 |
| mtaylor | I'll have a patch in an hour or so | 17:03 |
| mtaylor | sorry for the delay | 17:03 |
| jaypipes | SLAP! | 17:03 |
| jaypipes | :) no worries | 17:03 |
| * mtaylor punches jaypipes | 17:03 | |
| *** tryggvil__ has quit IRC | 17:03 | |
| * jaypipes strangles mtaylor with the tendon of a dead giraffe. | 17:04 | |
| *** derekh has quit IRC | 17:04 | |
| *** darraghb has quit IRC | 17:04 | |
| mtaylor | dude. NICE! | 17:04 |
| *** mnewby has joined #openstack-dev | 17:04 | |
| dansmith | mtaylor: um, he wins that one. | 17:04 |
| mtaylor | dansmith: yeah, he totally does. I'm going to have to up my game | 17:04 |
| *** tryggvil__ has joined #openstack-dev | 17:05 | |
| dansmith | ...good luck with that | 17:05 |
| * dansmith readies his olympic-judge-esque voting paddle and golf clap | 17:05 | |
| jaypipes | lol\ | 17:06 |
| *** sulochan has quit IRC | 17:10 | |
| *** bencherian has joined #openstack-dev | 17:14 | |
| *** tryggvil__ has quit IRC | 17:14 | |
| *** GheRivero has quit IRC | 17:16 | |
| *** issackel_ has joined #openstack-dev | 17:17 | |
| dansmith | Could some cores help move this along? https://review.openstack.org/#/c/9932/ | 17:18 |
| *** adjohn has joined #openstack-dev | 17:18 | |
| *** salgado is now known as salgado-lunch | 17:19 | |
| *** dtroyer is now known as dtroyer_zzz | 17:20 | |
| *** edygarcia has joined #openstack-dev | 17:28 | |
| *** lorin1 has quit IRC | 17:28 | |
| *** dtroyer_zzz is now known as dtroyer | 17:31 | |
| dansmith | comstud: thanks | 17:32 |
| ewindisch | dansmith: looking at it. | 17:36 |
| *** anniec has joined #openstack-dev | 17:38 | |
| *** lorin1 has joined #openstack-dev | 17:40 | |
| *** anniec has quit IRC | 17:42 | |
| dansmith | ewindisch: thanks.. after working on this for a bit, I think task_state really needs to be pre-success in order to filter out repeat requests at the highest layer, right? | 17:44 |
| *** anniec has joined #openstack-dev | 17:44 | |
| dansmith | I think it was intended to be viewed with vm_state as the post-action state change, and task_state as the pre-action one | 17:44 |
| ewindisch | I see | 17:44 |
| comstud | dansmith: np.. something feels fragile about it.. raising an exception 1 in decorator and catching it in another, but.. | 17:47 |
| dansmith | comstud: I know, it's not great, but I bet jaypipes would testify that he's explored other options (as have I) | 17:47 |
| dansmith | we could go back to how I had it in the first place, where we do the reversal as a function called from the actual failure point, but then you have to keep a mapping of states to revert, which was deemed ugly (rightfully so) | 17:48 |
| *** patelna has joined #openstack-dev | 17:49 | |
| jaypipes | comstud: yeah I spent the day on friday trying unsuccessfully to marry parameterized decorators with multiple nested levels of non-parameterized decorators (like wrapped_exception and catch_lock_instance) | 17:49 |
| jaypipes | comstud: ended up in decorator inception hell. | 17:49 |
| comstud | I'm not sure why the functionality couldn't be added directly in checks_instance_lock | 17:50 |
| comstud | i might have missed that | 17:50 |
| dansmith | comstud: it was, initially | 17:50 |
| *** adjohn has quit IRC | 17:51 | |
| dansmith | the problem was where to keep the list of states we should revert | 17:51 |
| jaypipes | right.. you would need to pass the revert states into the decorator... | 17:51 |
| arosen | Hello, for some reason maybe ~a month ago when running nova commands nova has been prompting me with, "Please input your password for the keyring" Any ideas why? Even with fresh devstack installs this happens to me. | 17:51 |
| dansmith | arosen: add --no_cache | 17:52 |
| jaypipes | and because that decorator is decorating a method of the manager, and that it also decorates the wrapped_exception decorator, it gets VERY ugly. | 17:52 |
| dansmith | jaypipes: good thing time doesn't slow down by an order of magnitude in each level of decorator inception, huh? :) | 17:53 |
| jaypipes | dansmith: no shit! | 17:53 |
| arosen | dansmith: to where? My nova doesn't know about that param. | 17:53 |
| comstud | dansmith: I left you another question inline that I just spotted | 17:53 |
| arosen | To /etc/nova/nova.conf? | 17:53 |
| dansmith | arosen: no, on the command line to nova | 17:53 |
| sdague | jaypipes: did you find the james bond level? ;) | 17:53 |
| dansmith | arosen: maybe you have an old/stale one? | 17:53 |
| *** mnewby has quit IRC | 17:54 | |
| * jaypipes hugs markmc for https://review.openstack.org/#/c/10119/ | 17:54 | |
| jaypipes | sdague: hehe, yep. I wore my white snow suit. | 17:54 |
| dansmith | comstud: cool, because this one didn't have quite enough iterations yet :/ | 17:54 |
| comstud | lol | 17:54 |
| * jaypipes wishes there was a rebase squash command for gerrit reviews :) | 17:55 | |
| comstud | haha | 17:55 |
| dansmith | comstud: ah, oops, missed those I guess.. I guess I have to acquiesce :) | 17:56 |
| *** Ryan_Lane has joined #openstack-dev | 17:56 | |
| comstud | dansmith: :) | 17:56 |
| comstud | dansmith: Note< I didn't go looking for everything | 17:56 |
| comstud | I just spotted those 2 | 17:56 |
| *** Ryan_Lane has joined #openstack-dev | 17:56 | |
| dansmith | okay, I'll recheck | 17:57 |
| comstud | ty! | 17:57 |
| dansmith | good thing I have a productivity-killing conf-call coming up... | 17:57 |
| *** Aaton_off is now known as Aaton | 17:59 | |
| *** thingee has joined #openstack-dev | 18:00 | |
| arosen | dansmith: I don't think so. I'm running master and just pulled down the latest changes | 18:02 |
| *** johnpur has joined #openstack-dev | 18:03 | |
| dansmith | dan@devstack:~/devstack$ nova help | grep cache | 18:03 |
| dansmith | usage: nova [--debug] [--no_cache] [--timings] [--os_username OS_USERNAME] | 18:03 |
| dansmith | arosen: ^ | 18:03 |
| *** ChanServ sets mode: +v johnpur | 18:03 | |
| *** johnpur has quit IRC | 18:04 | |
| *** johnpur has joined #openstack-dev | 18:04 | |
| *** ChanServ sets mode: +v johnpur | 18:04 | |
| *** anderstj has joined #openstack-dev | 18:05 | |
| arosen | dansmith: yup I have the same thing. That option doesn't seem to work though when I specify it. http://pastebin.com/6REZhtdK | 18:05 |
| dansmith | arosen: put it before 'list' | 18:06 |
| arosen | dansmith: doh, thanks! | 18:07 |
| dansmith | arosen: :) | 18:07 |
| *** almaisan-away is now known as al-maisan | 18:10 | |
| dprince | bcwaldon: re: outstanding question on the auth_or_store_url variable... is now a good time to talk? | 18:10 |
| dansmith | comstud: found a few others (oops), but got those fixed up and re-submitted | 18:12 |
| comstud | ok | 18:13 |
| jgriffith | jaypipes: BTW, you were correct... the delete succeeds it's the call to get in the verify loop that throws | 18:13 |
| jgriffith | jaypipes: Need to go to another meeting again but then I'm back for the afternoon (I think( | 18:13 |
| jaypipes | jgriffith: w00t. always good when I'm right. | 18:14 |
| jgriffith | jaypipes: Thought you'd like that! | 18:14 |
| *** thingee has quit IRC | 18:17 | |
| *** mnewby has joined #openstack-dev | 18:18 | |
| *** sulochan has joined #openstack-dev | 18:18 | |
| *** rkukura has quit IRC | 18:19 | |
| *** rkukura has joined #openstack-dev | 18:20 | |
| *** jakedahn_zz is now known as jakedahn | 18:21 | |
| dansmith | comstud: thanks | 18:22 |
| comstud | np | 18:22 |
| * dansmith wonders which core approver he should try to bribe next | 18:22 | |
| *** issackel_ has quit IRC | 18:22 | |
| *** cp16net is now known as cp16net|away | 18:22 | |
| *** mnewby has quit IRC | 18:23 | |
| *** mnewby has joined #openstack-dev | 18:23 | |
| *** matwood has joined #openstack-dev | 18:24 | |
| *** jog0 has joined #openstack-dev | 18:25 | |
| *** epim has joined #openstack-dev | 18:25 | |
| *** asakhnov has quit IRC | 18:28 | |
| *** maoy has quit IRC | 18:29 | |
| *** cp16net|away is now known as cp16net | 18:31 | |
| *** anderstj_ has joined #openstack-dev | 18:32 | |
| *** anderstj has quit IRC | 18:32 | |
| *** anderstj has joined #openstack-dev | 18:33 | |
| *** anderstj_ has quit IRC | 18:33 | |
| *** anderstj_ has joined #openstack-dev | 18:34 | |
| *** thingee has joined #openstack-dev | 18:35 | |
| *** anderstj has quit IRC | 18:38 | |
| *** dachary has quit IRC | 18:44 | |
| *** dachary has joined #openstack-dev | 18:46 | |
| *** novas0x2a|laptop has quit IRC | 18:47 | |
| *** salgado-lunch is now known as salgado | 18:50 | |
| *** heckj has quit IRC | 18:52 | |
| *** apevec has quit IRC | 18:53 | |
| *** dachary has quit IRC | 18:56 | |
| Ryan_Lane | ayoung: I'm looking at ldap for keystone right now… is it possible for keystone to use ldap for the token service? | 18:57 |
| Ryan_Lane | in the old nova code, the access and secret keys were an attribute on the user's object | 18:57 |
| Ryan_Lane | I really want to avoid using password authentication, as it makes users store their passwords on the filesystem | 18:58 |
| *** dachary has joined #openstack-dev | 18:58 | |
| Ryan_Lane | the password auth that almost all openstack cloud providers is using right now seriously makes me cringe | 18:58 |
| *** matwood has quit IRC | 18:59 | |
| dansmith | vishy: are votes counted from everyone or just other core/important people? | 19:00 |
| *** anniec has quit IRC | 19:04 | |
| *** apevec has joined #openstack-dev | 19:06 | |
| *** anniec has joined #openstack-dev | 19:06 | |
| *** nati_ueno has joined #openstack-dev | 19:07 | |
| ayoung | Ryan_Lane, LDAP support is simple->token | 19:07 |
| *** al-maisan is now known as almaisan-away | 19:07 | |
| Ryan_Lane | :( | 19:07 |
| ayoung | what would you like? Kerberos? | 19:07 |
| Ryan_Lane | token, like nova previously had | 19:08 |
| ayoung | Ryan_Lane, how do you get that token? | 19:08 |
| Ryan_Lane | token -> token | 19:08 |
| ayoung | Ryan_Lane, yes. we have that, too | 19:08 |
| Ryan_Lane | I generate the token for the user and stick it onto their user account in LDAP | 19:08 |
| ayoung | no token time out? | 19:08 |
| Ryan_Lane | my web frontend can pull the token, then authenticate using it | 19:08 |
| ayoung | Ryan_Lane, well...you cn still do that | 19:09 |
| Ryan_Lane | well, ideally, you'd auth with the permanent token, then provide a token with an expiration | 19:09 |
| ayoung | Ryan_Lane, no, permanent token bad...here's a better idea... | 19:09 |
| ayoung | Two factor auth to keystone, short lived token | 19:09 |
| ayoung | for example, Kerberos to Keystone, Client cert to keystone | 19:10 |
| ayoung | smart card... | 19:10 |
| *** nati_uen_ has quit IRC | 19:10 | |
| ayoung | then keystone provides a short lived token for OPenstack only operations | 19:10 |
| *** cp16net is now known as cp16net|away | 19:10 | |
| Ryan_Lane | I'm providing access to random people in the world | 19:10 |
| Ryan_Lane | smart cards won't work. we have two-factor authentication using OATH, though | 19:11 |
| Ryan_Lane | I don't want to force everyone to use that, though | 19:11 |
| Ryan_Lane | also, the end-user never needs to see the permanent token | 19:11 |
| Ryan_Lane | the web server can use the permanent token to generate one with an expiration, then provide that to the user | 19:11 |
| jaypipes | dansmith: everyone other than IBMers. ;P | 19:12 |
| ayoung | Ryan_Lane, so userid and password passed through to Keystone from the webfront end is not OK? HOrizon never has to store it | 19:12 |
| *** cp16net|away is now known as cp16net | 19:12 | |
| dansmith | jaypipes: heh, fair enough :D | 19:12 |
| ayoung | Ryan_Lane, you want to use standard auth mechanisms to the web front end. A custom token seems like it would be problematic | 19:12 |
| Ryan_Lane | ayoung: I could do that, but then I need to make sure the web frontend and keystone have the exact same token expiration time | 19:12 |
| ayoung | Ryan_Lane, no.... | 19:12 |
| Ryan_Lane | I don't want to force users to constantly re-authenticate | 19:12 |
| ayoung | web front end used basic auth. takes the uid-password and passed through to keystone | 19:13 |
| Ryan_Lane | it sucks | 19:13 |
| Ryan_Lane | eys | 19:13 |
| ayoung | then token is tored in the users session | 19:13 |
| Ryan_Lane | *yes | 19:13 |
| Ryan_Lane | indeed | 19:13 |
| Ryan_Lane | but, that assumes the user logs in often | 19:13 |
| Ryan_Lane | if users get a long-lived token, and don't re-auth for a week, keystone needs to accept the temporary token for a week | 19:13 |
| ayoung | well, not really. it assumes that either the user remembers their own password, or stores it in their browsers keystone | 19:13 |
| ayoung | er | 19:13 |
| ayoung | key store | 19:13 |
| Ryan_Lane | yes, which means the user needs to re-auth fairly often | 19:14 |
| Ryan_Lane | that sucks | 19:14 |
| Ryan_Lane | I can likely do simple auth, and just increase the expiration time of the keystone token to match mediawiki's long-lived token time | 19:15 |
| *** adjohn has joined #openstack-dev | 19:15 | |
| *** anniec has quit IRC | 19:16 | |
| ayoung | Ryan_Lane, with basic auth, the userid and password are sent on each request. THe webserver can ignore them, but on token time out, make another authentace call to keystone | 19:16 |
| Ryan_Lane | o.O | 19:17 |
| Ryan_Lane | I thought with basic auth, that it authenticates, then gives back a toekn | 19:17 |
| Ryan_Lane | *token | 19:17 |
| Ryan_Lane | then you use the temporary token until it expires | 19:17 |
| Ryan_Lane | then you re-auth | 19:17 |
| Ryan_Lane | (with the username/password) | 19:17 |
| ayoung | Ryan_Lane, only if you talk directly to keystone | 19:17 |
| Ryan_Lane | I definitely can't pass the username and password on every call | 19:18 |
| Ryan_Lane | as the frontend doesn't have it | 19:18 |
| Ryan_Lane | it can only do passthrough on initial login | 19:18 |
| ayoung | OK. If you do long lived tokens, then yes, you can do that, too. | 19:18 |
| ayoung | But then you need to store the tokens, and they are a symmetric secret. | 19:19 |
| Ryan_Lane | I don't mind that | 19:19 |
| ayoung | Ryan_Lane, so, are you good? | 19:19 |
| Ryan_Lane | I think so, yeah | 19:19 |
| Ryan_Lane | thanks | 19:19 |
| Ryan_Lane | I'll let you know when I start writing code ;) | 19:20 |
| *** ayoung is now known as ayoung-afk | 19:20 | |
| *** novas0x2a|laptop has joined #openstack-dev | 19:25 | |
| *** heckj has joined #openstack-dev | 19:26 | |
| *** matwood has joined #openstack-dev | 19:32 | |
| *** thovden has joined #openstack-dev | 19:34 | |
| *** dprince has quit IRC | 19:37 | |
| *** Guest67996 has quit IRC | 19:41 | |
| *** Kiall has joined #openstack-dev | 19:43 | |
| *** Kiall has quit IRC | 19:49 | |
| *** mnewby has quit IRC | 19:55 | |
| *** Kiall has joined #openstack-dev | 19:55 | |
| *** heckj has quit IRC | 19:56 | |
| *** primeministerp has quit IRC | 19:57 | |
| *** drewlander has quit IRC | 19:58 | |
| *** heckj has joined #openstack-dev | 20:01 | |
| *** thingee_zz has joined #openstack-dev | 20:04 | |
| *** GheRivero has joined #openstack-dev | 20:04 | |
| *** issackel_ has joined #openstack-dev | 20:05 | |
| *** thingee has quit IRC | 20:05 | |
| *** anniec has joined #openstack-dev | 20:07 | |
| *** anniec_ has joined #openstack-dev | 20:08 | |
| *** anniec has quit IRC | 20:08 | |
| *** anniec_ is now known as anniec | 20:08 | |
| *** johnpur has left #openstack-dev | 20:09 | |
| *** nati_uen_ has joined #openstack-dev | 20:10 | |
| *** nati_ueno has quit IRC | 20:13 | |
| *** nati_uen_ has quit IRC | 20:18 | |
| *** nati_ueno has joined #openstack-dev | 20:18 | |
| jaypipes | dansmith, russellb: warning... you guys are headed for merge conflicts :) https://review.openstack.org/#/c/10161/ | 20:19 |
| russellb | yeah we were talking about that earlier ... | 20:19 |
| *** dhellmann_ has joined #openstack-dev | 20:19 | |
| russellb | there's yet another patch that conflicts, too | 20:19 |
| russellb | it's a race! | 20:19 |
| jaypipes | :) | 20:19 |
| jaypipes | hmmmm. http://www.marketwatch.com/story/vmware-to-acquire-nicira-2012-07-23 | 20:20 |
| *** dhellmann_ has quit IRC | 20:20 | |
| *** dhellmann has quit IRC | 20:20 | |
| *** dhellmann has joined #openstack-dev | 20:20 | |
| russellb | jaypipes: huh, that's interesting, indeed. | 20:20 |
| dansmith | russellb: any chance you'd fall on your sword and approve mine? | 20:20 |
| dansmith | kinda like letting the person with just a pack of gum go first in the checkout line... :) | 20:21 |
| *** dhellmann has quit IRC | 20:21 | |
| jaypipes | hehe | 20:21 |
| russellb | lol | 20:21 |
| russellb | which one is it again? | 20:21 |
| dansmith | this ill-fated one: https://review.openstack.org/#/c/9932/ | 20:21 |
| russellb | mine isn't going in today ... got a request to refactor these decorators and i'm not even sure how to do it :-) | 20:22 |
| dansmith | I know the feeling :D | 20:22 |
| *** matwood has quit IRC | 20:22 | |
| russellb | these decorators had my banging my head the other day | 20:22 |
| *** GheRivero has quit IRC | 20:22 | |
| jaypipes | lol, yes he does. | 20:22 |
| dansmith | I've successfully worn out jaypipes, comstud, and ewindisch so hopefully the other avenues have been sufficiently checked | 20:23 |
| jaypipes | lmao | 20:23 |
| jaypipes | dansmith: decorators have a funny habit of doing that, eh? :) | 20:24 |
| *** salgado is now known as salgado-afk | 20:24 | |
| dansmith | yeah.. they're uuber cool, but man they can make like unpleasant in a hurry | 20:24 |
| sdague | hmmm... maybe we need some undecorating love at some point? | 20:24 |
| jaypipes | hehe | 20:24 |
| sdague | it seems getting 4 decorators deep is asking for future trouble | 20:24 |
| *** matwood has joined #openstack-dev | 20:24 | |
| dansmith | it depends on how clean they are.. I don't know that we need undecorating specifically, | 20:25 |
| jaypipes | sdague: yep. and mixing parameterized with non-parameterized functools.wraps() ones is asking for trouble! :) | 20:25 |
| dansmith | but some cleanup or refactoring might be good | 20:25 |
| russellb | 4 decorators that make debuggins really fun | 20:26 |
| sdague | sounds like a good summit discussion | 20:26 |
| russellb | when most of them are doing exception magic | 20:26 |
| *** thovden has quit IRC | 20:26 | |
| russellb | i wanted to set that file on fire the other day | 20:26 |
| dansmith | maybe there should be a pep8 rule about five-deep decoration :) | 20:26 |
| sdague | well, you could always add it to hacking.py | 20:27 |
| dansmith | sdague: sssh, let russellb approve first :) | 20:27 |
| sdague | :) | 20:27 |
| Ryan_Lane | ayoung-afk: hm. seems from the commandline, I can't find users with a space in their name | 20:27 |
| dansmith | OH CRAP | 20:27 |
| Ryan_Lane | "string indices must be integers, not str" | 20:27 |
| dansmith | I just hit 'git review' again in my nova tree instead of my tempest tree | 20:28 |
| dansmith | so my review from comstud is gone.. poo! | 20:28 |
| russellb | dansmith: ha, and russellb pulls ahead! | 20:28 |
| dansmith | aaaargh! | 20:28 |
| * Ryan_Lane groans | 20:28 | |
| Ryan_Lane | cn is a hardcoded attribute? | 20:28 |
| *** bencherian has quit IRC | 20:28 | |
| Ryan_Lane | uuuuugggghhhh and name is *sn*???? | 20:29 |
| jaypipes | Ryan_Lane: ah, welcome to LDAP :) | 20:30 |
| *** bencherian has joined #openstack-dev | 20:31 | |
| * jaypipes predicts Ryan_Lane's level of aggravation will soon increase | 20:31 | |
| Ryan_Lane | jaypipes: I re-write nova's implementation because of things exactly like this | 20:31 |
| Ryan_Lane | now I'm going to have to rewrite keystone's too? | 20:31 |
| Ryan_Lane | and of course, I'm going to have to patch my essex version too :( | 20:32 |
| ttx | markmc: Please have a look at https://review.openstack.org/#/c/9041/ when you can | 20:32 |
| Ryan_Lane | jaypipes: it's ok, my aggravation level with every technology sits about about this level at all times | 20:33 |
| * russellb offers the room chill pills | 20:33 | |
| jaypipes | Ryan_Lane: :) | 20:33 |
| *** littleidea_ has joined #openstack-dev | 20:37 | |
| *** sulochan has quit IRC | 20:37 | |
| *** danwent has joined #openstack-dev | 20:37 | |
| *** littleidea_ has joined #openstack-dev | 20:39 | |
| *** littleidea has quit IRC | 20:41 | |
| *** littleidea_ is now known as littleidea | 20:41 | |
| *** danwent has quit IRC | 20:44 | |
| *** utlemming has quit IRC | 20:44 | |
| *** samkottler has joined #openstack-dev | 20:46 | |
| *** thingee_zz has quit IRC | 20:46 | |
| *** markvoelker1 has joined #openstack-dev | 20:47 | |
| *** danwent has joined #openstack-dev | 20:48 | |
| bcwaldon | jaypipes: https://review.openstack.org/#/c/9623/ | 20:52 |
| bcwaldon | jaypipes: sorry :( | 20:52 |
| jaypipes | bcwaldon: you going to do the same for the v2 API? | 20:54 |
| bcwaldon | jaypipes: don't need any magic for the v2 api | 20:54 |
| bcwaldon | jaypipes: thats core code | 20:54 |
| bcwaldon | jaypipes: had to have those two hacks in the v1 side of things for an unrelated reason | 20:55 |
| jaypipes | bcwaldon: we're using httplib in v2 code? | 20:55 |
| bcwaldon | yes | 20:55 |
| jaypipes | kk | 20:55 |
| bcwaldon | jaypipes: client-side, yes | 20:55 |
| jaypipes | right | 20:55 |
| bcwaldon | jaypipes: httplib vs httplib2 only matters in glanceclient/common/http.py | 20:55 |
| bcwaldon | jaypipes: v1 and v2 both consume that code | 20:55 |
| jaypipes | bcwaldon: off to the test pit of despair. | 20:56 |
| bcwaldon | jaypipes: before, both versions were getting the iterated StringIO object | 20:56 |
| *** zhuadl has joined #openstack-dev | 20:57 | |
| *** amotoki has joined #openstack-dev | 20:59 | |
| *** mnaser has joined #openstack-dev | 21:00 | |
| *** garyk has joined #openstack-dev | 21:00 | |
| *** s0mik has joined #openstack-dev | 21:03 | |
| *** heckj has quit IRC | 21:04 | |
| andrewbogott | russellb or dhellmann, can I get one more set of eyes on https://review.openstack.org/#/c/10035/? (btw, dhellmann, this patch eliminates the need for the bit of code that troubled you in my last patchset :) ) | 21:05 |
| *** cp16net is now known as cp16net|away | 21:05 | |
| *** chuck_ has joined #openstack-dev | 21:05 | |
| *** eglynn has quit IRC | 21:06 | |
| *** zul has quit IRC | 21:08 | |
| *** lorin1 has quit IRC | 21:08 | |
| *** chuck_ has quit IRC | 21:12 | |
| *** zul has joined #openstack-dev | 21:12 | |
| *** cp16net|away is now known as cp16net | 21:18 | |
| *** nati_uen_ has joined #openstack-dev | 21:23 | |
| *** edygarcia has quit IRC | 21:25 | |
| *** edygarcia has joined #openstack-dev | 21:26 | |
| *** nati_ueno has quit IRC | 21:27 | |
| *** apevec has quit IRC | 21:27 | |
| *** maoy has joined #openstack-dev | 21:29 | |
| *** reed has quit IRC | 21:40 | |
| *** dachary has quit IRC | 21:44 | |
| Ryan_Lane | ayoung-afk: the ldap backend in keystone makes some really bad hardcoded assumptions | 21:44 |
| Ryan_Lane | the naming attribute is hardcoded. also, the attribute used for identity is hardcoded. also, they are the same thing | 21:45 |
| Ryan_Lane | the backend should search based on id attribute, then fetch the DN | 21:45 |
| Ryan_Lane | it's slower, but it's the only way to ensure the code will actually be usable | 21:46 |
| *** reed has joined #openstack-dev | 21:46 | |
| Ryan_Lane | basically all the fixes I made in nova are gone in keystone :( | 21:49 |
| *** sacharya has quit IRC | 21:49 | |
| *** utlemming has joined #openstack-dev | 21:49 | |
| *** Dr_Who has quit IRC | 21:50 | |
| *** arosen has quit IRC | 21:50 | |
| *** patelna_ has joined #openstack-dev | 21:56 | |
| *** amotoki_ has joined #openstack-dev | 21:57 | |
| Ryan_Lane | well, I take that back. only the subrole is hardcoded, which is likely fine. it still uses DNs directly, though | 21:58 |
| *** amotoki has quit IRC | 21:58 | |
| Ryan_Lane | I can work around that, I guess. I'll do lookups in the frontend before I do keystone auth | 21:59 |
| *** eglynn has joined #openstack-dev | 22:01 | |
| *** mnewby has joined #openstack-dev | 22:01 | |
| *** flaviamissi has quit IRC | 22:02 | |
| *** edygarcia has quit IRC | 22:03 | |
| Ryan_Lane | hm. "An unexpected error prevented the server from fulfilling your request. 'Identity' object has no attribute 'get_tenants' (HTTP 500)". are tenant listing expected to work with the ldap backend in essex? | 22:08 |
| *** amotoki has joined #openstack-dev | 22:08 | |
| jog0 | is anyone having devstack issues? 'error: Installed distribution prettytable 0.6.1 conflicts with requirement prettytable==0.6' | 22:08 |
| *** s34n has joined #openstack-dev | 22:09 | |
| *** amotoki_ has quit IRC | 22:09 | |
| *** arosen has joined #openstack-dev | 22:10 | |
| *** eglynn has quit IRC | 22:12 | |
| *** chrisfer has quit IRC | 22:12 | |
| *** danwent has quit IRC | 22:13 | |
| *** edygarcia has joined #openstack-dev | 22:14 | |
| *** markvoelker1 has left #openstack-dev | 22:17 | |
| *** AlanClark has quit IRC | 22:18 | |
| *** danwent has joined #openstack-dev | 22:20 | |
| Ryan_Lane | ayoung-afk: hm. the attribute mapping is hardcoded, even in trunk. it seems the lookup for a user during authentication is sn | 22:24 |
| Ryan_Lane | that's for sure a bug | 22:24 |
| Ryan_Lane | already a bug, it seems. | 22:25 |
| *** danwent has quit IRC | 22:29 | |
| *** danwent has joined #openstack-dev | 22:31 | |
| *** adjohn has quit IRC | 22:32 | |
| *** markmc has quit IRC | 22:32 | |
| *** heckj has joined #openstack-dev | 22:33 | |
| *** heckj has quit IRC | 22:33 | |
| *** samkottler has quit IRC | 22:33 | |
| Ryan_Lane | ayoung-afk: I really wish you were around so I could discuss this with you.... | 22:33 |
| *** rpedde is now known as rpedde_away | 22:41 | |
| *** e1mer has joined #openstack-dev | 22:42 | |
| dabo | jog0: I'm getting the same prettytable issue | 22:44 |
| *** amotoki_ has joined #openstack-dev | 22:47 | |
| *** epim has quit IRC | 22:48 | |
| vishy | dansmith: just core | 22:48 |
| *** clarkb has quit IRC | 22:48 | |
| *** epim has joined #openstack-dev | 22:48 | |
| *** amotoki has quit IRC | 22:48 | |
| *** sacharya has joined #openstack-dev | 22:49 | |
| *** epim has quit IRC | 22:50 | |
| *** epim has joined #openstack-dev | 22:50 | |
| dansmith | vishy: okay, I figured, but then I saw non-core folks voting so wanted to check.. thanks :) | 22:52 |
| jog0 | dabo: I reset my vagrant config (from git) and that fixed it. | 23:13 |
| *** rnirmal has quit IRC | 23:17 | |
| *** amotoki_ has quit IRC | 23:21 | |
| *** datsun180b has quit IRC | 23:24 | |
| *** samkottler has joined #openstack-dev | 23:25 | |
| *** markmcclain has quit IRC | 23:28 | |
| *** anderstj_ has quit IRC | 23:29 | |
| *** littleidea has quit IRC | 23:32 | |
| *** littleidea has joined #openstack-dev | 23:34 | |
| *** markmcclain has joined #openstack-dev | 23:36 | |
| dabo | jog0: I'm having problems running on a new install - nothing's been touched. What did you do in git to reset vagrant? | 23:38 |
| *** pixelbeat has quit IRC | 23:38 | |
| jk0 | you could delete the newer version of prettytable | 23:38 |
| jk0 | their API changed so we had to pin it in novaclient, if I recall | 23:39 |
| jog0 | dabo: I checked out https://github.com/bcwaldon/vagrant_devstack into a new directory and it worked. (So I just did a full wipe of the repo and tried again) | 23:39 |
| *** PotHix has quit IRC | 23:41 | |
| *** matwood has quit IRC | 23:43 | |
| *** danwent has quit IRC | 23:44 | |
| *** halfss has joined #openstack-dev | 23:46 | |
| *** halfss has quit IRC | 23:49 | |
| *** patelna_ has quit IRC | 23:50 | |
| bcwaldon | dabo: can I offer any help? | 23:50 |
| *** danwent has joined #openstack-dev | 23:51 | |
| jog0 | bcwaldon: slick vagrant script btw | 23:51 |
| bcwaldon | jog0: thanks! | 23:51 |
| bcwaldon | I don't acutally use it anymore :/ | 23:51 |
| bcwaldon | I built something better than vagrant | 23:51 |
| bcwaldon | uses remote openstack instead of local virtualbox | 23:52 |
| bcwaldon | my little air was choking | 23:52 |
| *** patelna has quit IRC | 23:52 | |
| jog0 | bcwaldon: haha yeah I have started doing that too but its nice to have a local option as well | 23:52 |
| bcwaldon | absolutely | 23:52 |
| *** mnaser has quit IRC | 23:55 | |
| *** steveb_ has joined #openstack-dev | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!