| *** rods has joined #openstack-dev | 00:07 | |
| *** nati2 has joined #openstack-dev | 00:37 | |
| *** lorin1 has joined #openstack-dev | 00:48 | |
| *** nati2_ has joined #openstack-dev | 00:53 | |
| *** nati2 has quit IRC | 00:54 | |
| *** vladimir3p has quit IRC | 00:57 | |
| *** rods has quit IRC | 01:01 | |
| *** nati2_ has quit IRC | 01:01 | |
| *** jakedahn has quit IRC | 01:02 | |
| *** jdurgin has quit IRC | 01:23 | |
| *** sdake has quit IRC | 01:27 | |
| *** jakedahn has joined #openstack-dev | 01:42 | |
| *** sdake has joined #openstack-dev | 02:05 | |
| *** hugokuo has joined #openstack-dev | 02:38 | |
| *** novas0x2a|laptop has quit IRC | 02:42 | |
| openstackgerrit | Yogeshwar Srikrishnan proposed a change to openstack/keystone: Changes to stored hashed password in backends. https://review.openstack.org/719 | 02:51 |
|---|---|---|
| *** sdake has quit IRC | 02:53 | |
| *** sdake has joined #openstack-dev | 03:05 | |
| *** joonwon has quit IRC | 03:05 | |
| openstackgerrit | Yogeshwar Srikrishnan proposed a change to openstack/keystone: Changes to stored hashed password in backends. Using passlib a password hashing library. Using sha512. https://review.openstack.org/719 | 03:05 |
| *** lorin1 has left #openstack-dev | 03:09 | |
| *** sdake has quit IRC | 03:18 | |
| *** martine has joined #openstack-dev | 03:32 | |
| *** sdake has joined #openstack-dev | 03:33 | |
| *** joonwon has joined #openstack-dev | 03:45 | |
| *** ameade has quit IRC | 03:53 | |
| *** Rajaram has joined #openstack-dev | 04:14 | |
| *** jakedahn has quit IRC | 04:16 | |
| *** markmc has quit IRC | 04:35 | |
| *** zorzar has quit IRC | 04:35 | |
| *** bourke has quit IRC | 04:35 | |
| *** RobertLaptop has quit IRC | 04:35 | |
| *** negronjl has quit IRC | 04:35 | |
| *** ipl31 has quit IRC | 04:35 | |
| *** sandywalsh has quit IRC | 04:35 | |
| *** pvo has quit IRC | 04:35 | |
| *** chmouel has quit IRC | 04:35 | |
| *** zedas has quit IRC | 04:35 | |
| *** bourke has joined #openstack-dev | 04:35 | |
| *** chmouel has joined #openstack-dev | 04:37 | |
| *** pvo has joined #openstack-dev | 04:37 | |
| *** RobertLaptop has joined #openstack-dev | 04:37 | |
| *** ipl31 has joined #openstack-dev | 04:38 | |
| *** sandywalsh has joined #openstack-dev | 04:39 | |
| *** martine has quit IRC | 04:43 | |
| *** markmc has joined #openstack-dev | 04:44 | |
| *** zorzar has joined #openstack-dev | 04:44 | |
| *** negronjl has joined #openstack-dev | 04:44 | |
| *** zedas has joined #openstack-dev | 04:44 | |
| *** hugokuo has quit IRC | 05:23 | |
| *** hugokuo has joined #openstack-dev | 05:24 | |
| *** Rajaram has quit IRC | 05:32 | |
| *** reed has quit IRC | 05:51 | |
| *** zul has quit IRC | 05:56 | |
| *** zul has joined #openstack-dev | 05:58 | |
| openstackgerrit | Yogeshwar Srikrishnan proposed a change to openstack/keystone: Changes to support getTenants to behave differntly for admin users when invoked as a service api or admin api. https://review.openstack.org/725 | 05:59 |
| *** nati2 has joined #openstack-dev | 06:08 | |
| openstackgerrit | Sateesh proposed a change to openstack/nova: bug 861310 https://review.openstack.org/703 | 06:11 |
| uvirtbot | Launchpad bug 861310 in nova "Scheduler hangs if filtered host list is empty." [Undecided,In progress] https://launchpad.net/bugs/861310 | 06:11 |
| *** nati2_ has joined #openstack-dev | 06:40 | |
| *** nati2 has quit IRC | 06:42 | |
| *** popux has joined #openstack-dev | 06:55 | |
| *** jdag has quit IRC | 07:04 | |
| *** Spirilis has quit IRC | 07:04 | |
| *** reidrac has joined #openstack-dev | 07:15 | |
| *** rbasak has joined #openstack-dev | 07:17 | |
| *** Spirilis has joined #openstack-dev | 07:47 | |
| *** nati2 has joined #openstack-dev | 08:10 | |
| *** nati2_ has quit IRC | 08:12 | |
| vishy | mwhooker: I'd love to see this go forward, but it would be really awesome if you could plug it in to the existing effort to get all of the tests in one place. https://github.com/openstack/openstack-integration-tests | 08:29 |
| *** nati2 has quit IRC | 08:33 | |
| ttx | vishy: ping | 08:56 |
| *** darraghb has joined #openstack-dev | 08:58 | |
| ttx | vishy: I'll need the list of nova sessions you don't *need* to attend, and the list of sessions in other tracks you *need* to attend, to start scheduling | 08:59 |
| vishy | ttx: right, I'll get on that now | 08:59 |
| ttx | vishy: By default I assume you attend all nova tracks and nothing else, so just get the diff from that | 09:00 |
| vishy | ttx: did you notice there are two more submissions? | 09:00 |
| ttx | vishy: yeah, I noticed, we need to decide what to do with them | 09:00 |
| vishy | ttx: i think they are both great sessions | 09:01 |
| ttx | vishy: looks like our main contention point will be your availability rather than free slots | 09:01 |
| vishy | it seems like the ipv6 one is especially good for a 25 minute one | 09:01 |
| vishy | the other one maybe we could waitlist? | 09:01 |
| ttx | do you think the other one warrants a full hour ? | 09:02 |
| * ttx accepts the ipv6 one | 09:02 | |
| vishy | probably could do it in a short session | 09:02 |
| ttx | i'll put it in wait list as a 25min session | 09:03 |
| ttx | vishy: i.e. we curently have enough slots in the schedule to fit all the wait list -- but you still can't be in two places at the same time. | 09:06 |
| vishy | right | 09:07 |
| ttx | vishy: so "which sessions can you delegate / afford to miss" is the critical point | 09:08 |
| *** popux has quit IRC | 09:19 | |
| vishy | ttx: i've got a list of the common/other ones i'd really like to be at | 09:26 |
| vishy | the afford to miss is harder, I'm going through it | 09:26 |
| ttx | vishy: maybe prioritize it -- I'll try my best | 09:27 |
| vishy | ttx: easy way to get a text list of all nova sessions that aren't rejected? | 09:29 |
| ttx | vishy: let me extract that from DB and etherpad it | 09:30 |
| vishy | ttx: thanks, copy pasting from the website is not going well :) | 09:30 |
| ttx | vishy: http://etherpad.openstack.org/Az46PUkBuy | 09:36 |
| openstackgerrit | HP Nova Contributors proposed a change to openstack/nova: Enforce snapshot cleanup. https://review.openstack.org/726 | 09:42 |
| *** Rajaram has joined #openstack-dev | 09:43 | |
| *** jeffjapan has quit IRC | 09:54 | |
| *** reidrac has quit IRC | 09:58 | |
| openstackgerrit | HP Nova Contributors proposed a change to openstack/nova: Enforce snapshot cleanup. https://review.openstack.org/726 | 09:58 |
| *** rbasak has quit IRC | 10:04 | |
| *** joonwon has quit IRC | 10:08 | |
| openstackgerrit | Brendan Maguire proposed a change to openstack/nova: Fixes Bug #861293 nova.auth.signer.Signer now honors the SignatureMethod parameter for SHA1 when creating signatures https://review.openstack.org/712 | 10:08 |
| uvirtbot | Launchpad bug 861293 in nova "Client using sha1 won't be able to authenticate if sha256 is available to the server" [Undecided,In progress] https://launchpad.net/bugs/861293 | 10:08 |
| ttx | vishy: some sessions were completely dropped from the list -- I assume that means you can miss them ? | 10:18 |
| vishy | really? | 10:18 |
| ttx | vishy: or is it some copy-paste fumble ? | 10:18 |
| vishy | i may have copy paste removed a few | 10:18 |
| ttx | there was 23 | 10:18 |
| ttx | "Interconnectivity/Bursting between clouds" is missing | 10:19 |
| ttx | as is "Advanced Scheduling" | 10:20 |
| Daviey | Am i being silly, or is adding a comment in gerrit not obvious? | 10:20 |
| vishy | ah darnit | 10:20 |
| Daviey | I'm signed in.. but no "Add Comment" button? | 10:21 |
| Daviey | ahh, "Review" sneaky button | 10:21 |
| vishy | ttx: added them back in to the etherpad in order | 10:21 |
| ttx | vishy: cool thx | 10:21 |
| *** rbasak has joined #openstack-dev | 10:22 | |
| ttx | vishy: does "HA capabilities for guest instances" require 55 min ? | 10:25 |
| vishy | I'm sure we could fill it, but I think it would be fine to do a short one if your trying to save time | 10:26 |
| ttx | ok, let's keep it that way then | 10:26 |
| ttx | vishy: you don't need to be at "Dashboard, Nova, Keystone and RBAC" ? | 10:27 |
| vishy | oh yes i do | 10:27 |
| ttx | ok will add | 10:27 |
| vishy | sorry i missed it in the list of other sessions | 10:27 |
| ttx | so you at least need to miss 3 slot worth of nova sessions | 10:28 |
| ttx | (in the ideal case where I manage to make all the schedule revolve around you) | 10:28 |
| * ttx dinners | 10:28 | |
| *** reidrac has joined #openstack-dev | 10:34 | |
| *** tryggvil_ has joined #openstack-dev | 11:07 | |
| *** rods has joined #openstack-dev | 11:17 | |
| *** Vek has quit IRC | 11:26 | |
| sandywalsh | Daviey, and you can select a line in the code review and press "c". Also "r" is a keyboard shortcut for 'review' | 11:33 |
| Daviey | sandywalsh: sounds too advanced for me :) | 11:37 |
| Daviey | rbasak: Can you follow up on the two questions on your branch? | 11:58 |
| sandywalsh | Daviey, pro-tip for gerrit: 50% frontal lobe removal ... I'm a pro! | 11:58 |
| rbasak | Daviey: I discovered some issues, I've been fixing those up | 11:58 |
| rbasak | Daviey: Also I'll add tests as requested, etc. | 11:59 |
| Daviey | rbasak: rocking! | 12:00 |
| rbasak | Is there a mechanism that I can use to run just one test? "nosetests nova.tests..." doesn't seem to work | 12:18 |
| vishy | rbasak: ./run_tests.sh -N api.ec2.test_cloud:CloudTestCase.test_describe_instances | 12:21 |
| vishy | (for example) | 12:21 |
| rbasak | vishy: aha, that works. Thanks! | 12:23 |
| *** lts has joined #openstack-dev | 12:31 | |
| openstackgerrit | Brendan Maguire proposed a change to openstack/nova: Fixes Bug #861293 nova.auth.signer.Signer now honors the SignatureMethod parameter for SHA1 when creating signatures https://review.openstack.org/712 | 12:40 |
| uvirtbot | Launchpad bug 861293 in nova "Client using sha1 won't be able to authenticate if sha256 is available to the server" [Undecided,In progress] https://launchpad.net/bugs/861293 | 12:40 |
| vishy | ttx: http://summit.openstack.org/sessions/view/109 I also have to be at that one if it makes it past the waiting list, since I'm leading it. | 12:53 |
| ttx | hrm, right | 12:54 |
| openstackgerrit | Brendan Maguire proposed a change to openstack/nova: api.ec2.admin unit tests https://review.openstack.org/684 | 12:57 |
| openstackgerrit | William Wolf proposed a change to openstack/openstack-chef: Make adding glance images use proper id's rather than hard code them https://review.openstack.org/727 | 13:08 |
| *** zul has quit IRC | 13:08 | |
| *** zaitcev has quit IRC | 13:08 | |
| *** pasik has quit IRC | 13:08 | |
| *** johan_-_ has quit IRC | 13:08 | |
| *** zul has joined #openstack-dev | 13:10 | |
| *** zaitcev has joined #openstack-dev | 13:10 | |
| *** pasik has joined #openstack-dev | 13:10 | |
| *** johan_-_ has joined #openstack-dev | 13:10 | |
| openstackgerrit | A change was merged to openstack/openstack-chef: Make adding glance images use proper id's rather than hard code them https://review.openstack.org/727 | 13:12 |
| vishy | woah, we have an openstack-chef repo? | 13:13 |
| vishy | cool | 13:13 |
| *** Vek has joined #openstack-dev | 13:26 | |
| *** Rajaram_ has joined #openstack-dev | 13:27 | |
| *** Rajaram has quit IRC | 13:29 | |
| *** Rajaram_ is now known as Rajaram | 13:29 | |
| *** martine has joined #openstack-dev | 13:30 | |
| *** jdurgin has joined #openstack-dev | 13:42 | |
| *** kbringard has joined #openstack-dev | 13:42 | |
| *** benrodrigue has joined #openstack-dev | 13:56 | |
| *** joesavak has joined #openstack-dev | 14:01 | |
| *** mfer has joined #openstack-dev | 14:04 | |
| *** glenc has joined #openstack-dev | 14:07 | |
| *** Joonwon has joined #openstack-dev | 14:12 | |
| openstackgerrit | Chuck Short proposed a change to openstack/nova: Display LXC container console when using euca-get-console. https://review.openstack.org/728 | 14:13 |
| *** lorin1 has joined #openstack-dev | 14:17 | |
| *** dolphm has joined #openstack-dev | 14:18 | |
| annegentle | is Rick Harris s1rp on here? I have a question about image_cache_enabled in Glance and my grep didn't show a description for it (I think it's a flag). Anyone have info about it? | 14:20 |
| openstackgerrit | Chuck Short proposed a change to openstack/nova: Replace socat with netcat. https://review.openstack.org/729 | 14:24 |
| *** dolphm has quit IRC | 14:26 | |
| *** reed has joined #openstack-dev | 14:38 | |
| *** jdag has joined #openstack-dev | 14:39 | |
| *** jdag has quit IRC | 14:40 | |
| *** jsavak has joined #openstack-dev | 14:42 | |
| *** dolphm has joined #openstack-dev | 14:43 | |
| *** joesavak has quit IRC | 14:45 | |
| *** nati2 has joined #openstack-dev | 14:47 | |
| *** code_franco has joined #openstack-dev | 14:47 | |
| *** rnirmal has joined #openstack-dev | 14:48 | |
| *** lorin1 has quit IRC | 14:50 | |
| *** lorin1 has joined #openstack-dev | 14:50 | |
| openstackgerrit | Brian Waldon proposed a change to openstack/glance: Refactoring registry api controllers https://review.openstack.org/730 | 14:51 |
| *** bcwaldon has joined #openstack-dev | 14:52 | |
| ttx | notmyname: let me know what we should do with "searchable API using keyword to get objects" - http://summit.openstack.org/sessions/view/87 | 14:53 |
| notmyname | ttx: low priority. the proposed session wasn't very clear. I think I know what it means, and I sent feedback. if it's not updated, I'm ok with dropping it | 14:54 |
| ttx | notmyname: not really sure if it was updated -- no proposer comments. | 14:55 |
| ttx | notmyname: Fine with dropping it if you are. | 14:55 |
| openstackgerrit | Anne Gentle proposed a change to openstack/openstack-manuals: Fixes pom file so ant part of the build wont fail https://review.openstack.org/731 | 14:57 |
| *** cmagina has quit IRC | 14:57 | |
| *** cmagina_ has joined #openstack-dev | 14:57 | |
| openstackgerrit | A change was merged to openstack/openstack-manuals: Fixes pom file so ant part of the build wont fail https://review.openstack.org/731 | 14:57 |
| *** cmagina_ has quit IRC | 14:59 | |
| *** cmagina has joined #openstack-dev | 15:00 | |
| notmyname | ttx: if you've got space, keep it. if not, drop it. | 15:00 |
| bcwaldon | jaypipes: do you know phil day? | 15:00 |
| bcwaldon | or ttx ^ | 15:02 |
| ttx | bcwaldon: working for HP, coming to the summit | 15:02 |
| ttx | Author of a couple proposals for the summit. | 15:03 |
| ttx | bcwaldon: that's about it | 15:03 |
| bcwaldon | ttx: I need to talk to him about his session topics, does he get on IRC? | 15:03 |
| ttx | bcwaldon: email sounds like a better bet | 15:03 |
| bcwaldon | kk | 15:03 |
| ttx | bcwaldon: want his email ? | 15:03 |
| *** dragondm has joined #openstack-dev | 15:03 | |
| bcwaldon | yes, pm it to me if you will | 15:03 |
| *** mwhooker has quit IRC | 15:06 | |
| *** mwhooker has joined #openstack-dev | 15:06 | |
| *** reidrac has quit IRC | 15:08 | |
| *** mwhooker has quit IRC | 15:11 | |
| *** nati2 has quit IRC | 15:12 | |
| *** nati2 has joined #openstack-dev | 15:13 | |
| *** rnorwood has joined #openstack-dev | 15:13 | |
| ttx | notmyname: got space, accepted it, scheduled on 3rd day so that if you realize it sucks, we can place something else in there. | 15:14 |
| bcwaldon | vishy: ping | 15:24 |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/nova: install_venv: refactor out pip_install helper https://review.openstack.org/732 | 15:25 |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/nova: install_venv: pass the --upgrade argument to pip install https://review.openstack.org/733 | 15:25 |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/nova: install_venv: don't use --no-site-packages with virtualenv https://review.openstack.org/734 | 15:25 |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/nova: install_venv: remove versioned M2Crypto dependency https://review.openstack.org/735 | 15:25 |
| vishy | bcwaldon: pong | 15:25 |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/nova: install_venv: add support for distro specific code https://review.openstack.org/736 | 15:25 |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/nova: install_venv: pip install M2Crypto doesn't work on Fedora https://review.openstack.org/737 | 15:25 |
| bcwaldon | looks like Mark didn't squash | 15:25 |
| bcwaldon | vishy: is adding a test a blocker for getting this in: https://review.openstack.org/#change,726 | 15:25 |
| markmc | bcwaldon, I only squashed as appropriate | 15:25 |
| vishy | bcwaldon: not to me | 15:26 |
| bcwaldon | markmc: I'm assuming you just git reviewed a branch with multiple commits | 15:26 |
| bcwaldon | markmc: so now there are 6 reviews? | 15:26 |
| bcwaldon | markmc: that all depend on one another | 15:26 |
| bcwaldon | vishy: kk, I'm just going to send it | 15:26 |
| markmc | bcwaldon, yep, that's the way we want to do it, apparently | 15:26 |
| bcwaldon | vishy: i think it is going to be a pain to add a test | 15:26 |
| vishy | bcwaldon: it seems like a nice to have. The complexity of the test outweighs the complexity of the fix by far | 15:26 |
| vishy | bcwaldon: ship it | 15:26 |
| bcwaldon | markmc: I don't think so :) | 15:26 |
| bcwaldon | vishy: kk | 15:26 |
| markmc | bcwaldon, they should be separate commits - whether they should be separate reviews ... | 15:27 |
| bcwaldon | markmc: why should they be separate commits? | 15:27 |
| bcwaldon | markmc: locally, sure, but once they hit nova, no | 15:27 |
| markmc | bcwaldon, https://lists.launchpad.net/openstack/msg04256.html | 15:27 |
| vishy | would be nice if we could review a set of commits together for sure... | 15:27 |
| bcwaldon | my problem is now there are 6 reviews that all have to get a look | 15:27 |
| bcwaldon | that all depend on one another | 15:28 |
| bcwaldon | i would like to send it all in at once | 15:28 |
| markmc | right, gerrit isn't great for this | 15:28 |
| markmc | but the answer isn't to squash everything together IMHO | 15:28 |
| markmc | keeping logical changes separate is valuable | 15:28 |
| vishy | markmc: but all of those changes are touching a couple of files, no? | 15:29 |
| bcwaldon | where's the value in having multiple commits going into nova for a single logical review? I won't want to check out nova halfway through a set of commits that add a feature | 15:29 |
| markmc | vishy, just one file | 15:29 |
| markmc | bcwaldon, ease of review, bisectability | 15:29 |
| markmc | bcwaldon, take a look at any kernel merge request | 15:29 |
| bcwaldon | its more difficult to review | 15:29 |
| markmc | because of the tooling | 15:29 |
| bcwaldon | I purposefully don't look at kernel reviews ;) | 15:29 |
| markmc | fwiw, http://blogs.gnome.org/markmc/2011/09/25/gerrit-patch-review-from-the-command-line/ | 15:30 |
| * markmc spent the time figuring that out because he figures there's going to be lots of reviews | 15:30 | |
| markmc | if there's one per commit | 15:30 |
| bcwaldon | what's the value in accepting commit 6 out of this series then rejecting 5? How do you move forward from that? | 15:31 |
| markmc | you can accept 1-5 and reject 6 | 15:31 |
| bcwaldon | what about the inverse | 15:31 |
| vishy | bcwaldon: I think in this case they are all separate, no? | 15:31 |
| markmc | bcwaldon, in this case, it wouldn't make sense | 15:31 |
| markmc | this is a silly example, anyway | 15:32 |
| markmc | it's just a little helper script | 15:32 |
| markmc | for real code, it's even more important IMHO | 15:32 |
| markmc | for bisectability | 15:32 |
| markmc | and so the reviewer isn't staring at a huge diff with lots of different changes | 15:32 |
| bcwaldon | none of these really depend on eachother, do they? | 15:33 |
| markmc | they do | 15:33 |
| markmc | actually, related commits could do with a "cover letter" | 15:33 |
| bcwaldon | 736 could go in on its own, for example | 15:33 |
| bcwaldon | so why does it have to wait for 4 other reviews ahead of it | 15:34 |
| vishy | markmc: some of them do | 15:34 |
| bcwaldon | I'm honestly interested why this is better | 15:34 |
| vishy | bcwaldon: ^^ | 15:34 |
| bcwaldon | not just flailing here | 15:34 |
| bcwaldon | vishy: what do you mean 'some of them do' | 15:34 |
| vishy | bcwaldon: I think i agree in this case | 15:34 |
| vishy | well the m2crypto rebuild depends on 736 | 15:35 |
| bcwaldon | these are also so small they dont have to depend on one another | 15:35 |
| bcwaldon | they could have been 3 different reviews | 15:35 |
| bcwaldon | independent | 15:35 |
| chmouel | dolphm: hello, did roleRefs changes to roles in latest keystone? (trying on the right chan) | 15:36 |
| bcwaldon | local commits are only useful to the author | 15:36 |
| vishy | bcwaldon: they are all independent reviews? | 15:36 |
| bcwaldon | they are all dependent reviews | 15:36 |
| markmc | what you really want to be able to do is to pull a series of commits like this locally | 15:36 |
| markmc | and approve them all in one go | 15:36 |
| markmc | or approve a subset of them | 15:36 |
| bcwaldon | what I really want is to see the diff in gerrit of a set of commits | 15:36 |
| markmc | but still have the value of bisectability and ease of review of each of the individual changes | 15:36 |
| bcwaldon | and not have them create different reviews | 15:36 |
| bcwaldon | but if there is a series of dependent changes, they then aren't useful individually | 15:37 |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack/keystone: Fix swift middleware with regard to latest changes. https://review.openstack.org/738 | 15:37 |
| markmc | bcwaldon, the typical example is where you have to do a big refactoring before fixing some tiny, subtle one-liner issue | 15:38 |
| markmc | bcwaldon, you want the "no functional changes" patch as one commit, and the one-liner as another commit | 15:38 |
| markmc | bcwaldon, so you can quickly sign off on the first one | 15:38 |
| bcwaldon | markmc: I will definitely agree that gerrit is a little frustrating, and I think you and I *mostly* agree on where we should ultimately be. I'm just trying to make it as useful as possible in its current state | 15:38 |
| markmc | bcwaldon, and then more carefully consider the second one | 15:38 |
| openstackgerrit | A change was merged to openstack/keystone: Revised docstring https://review.openstack.org/722 | 15:38 |
| jsavak | chmouel: yes - called roles now in the auth response (https://github.com/openstack/keystone/blob/master/keystone/content/common/samples/auth.json) | 15:38 |
| markmc | bcwaldon, and test the refactoring independently of the one-liner | 15:38 |
| vishy | bcwaldon: now i see the dependency you are referring to | 15:39 |
| markmc | bcwaldon, this discipline of keeping logical changes separate is really useful when you get into it | 15:39 |
| vishy | so apparently if you propose that way it automatically adds the dependencies | 15:40 |
| * markmc would hate to see gerrit essentially preventing it | 15:40 | |
| bcwaldon | markmc: sure, but why do they all have to *depend* on one another | 15:40 |
| markmc | vishy, right | 15:40 |
| vishy | markmc, bcwaldon: I agree in that case | 15:40 |
| markmc | bcwaldon, I might have been able to split it into two | 15:40 |
| markmc | bcwaldon, but that would have been even more work | 15:40 |
| markmc | bcwaldon, to create two branches | 15:40 |
| bcwaldon | all i do is create a branch, then git commit -a --amend | 15:40 |
| markmc | bcwaldon, make sure they merge together okay | 15:40 |
| vishy | markmc: if you want to squash them down to 2 or 3 separate reviews that are logically separate, this makes sense | 15:40 |
| vishy | markmc, bcwaldon: I think in the current state it makes way more sense to squash them though | 15:41 |
| bcwaldon | current state w.r.t. gerrit itself or this review? | 15:41 |
| vishy | w.r.t. gerrit | 15:41 |
| bcwaldon | ok, that lets you have multiple commits locally then squash before reviewing | 15:42 |
| vishy | it is too much work to separate them into logically separate branches | 15:42 |
| bcwaldon | if they dont depend on one another, its not too hard ;) | 15:42 |
| markmc | vishy, if we had tools to pull a patch series as a branch and review with one command? | 15:42 |
| vishy | markmc: sure that would be fine | 15:42 |
| markmc | s/review with one/approve with one/ | 15:42 |
| markmc | that's doable | 15:42 |
| vishy | markmc: there is some stuff proposed into gerrit to do such a thing, but we don't have it now | 15:42 |
| * markmc is close with the blog above | 15:42 | |
| mtaylor | markmc, vishy: there is almost completed work to support reviewing patch series as a logical unit within gerrit | 15:42 |
| mtaylor | vishy beat me to it | 15:43 |
| vishy | markmc: it isn't just the command line tools. Some people prefer a web interface :) | 15:43 |
| bcwaldon | I like pretty colors | 15:43 |
| markmc | meh :) | 15:43 |
| villep | pretty colors ! | 15:43 |
| vishy | inline commenting is very nice as well | 15:43 |
| bcwaldon | mtaylor: as an aside, is there any way we can get our authors test to check the committers too? | 15:43 |
| mtaylor | uhm. | 15:44 |
| mtaylor | bcwaldon: sure? | 15:44 |
| bcwaldon | mtaylor: the latest review from HP Nova Contributors has a committer that is not in the Authors file, and the test passes | 15:44 |
| mtaylor | ah. then we should do that | 15:44 |
| bcwaldon | mtaylor: but the 'author' of the review *is* in the file | 15:44 |
| openstackgerrit | A change was merged to openstack/keystone: Changes to support getTenants to behave differntly for admin users when invoked as a service api or admin api. https://review.openstack.org/725 | 15:45 |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack/swift: Add support for OS Auth API version 2 https://review.openstack.org/704 | 15:46 |
| dolphm | johan_-_: are you Johannes Erdfelt? | 15:49 |
| mtaylor | markmc: so, I was just reading the full scrollback, and on the topic of not squashing... | 15:51 |
| mtaylor | markmc: this has a good writeup of the squash-before-merging to trunk reasoning: http://sandofsky.com/blog/git-workflow.html | 15:52 |
| comstud | dolphm: it is.. but he's in a mtg ATM | 15:52 |
| markmc | mtaylor, the individual commits aren't "checkpoint commits" in this case | 15:53 |
| markmc | mtaylor, absolutely squash checkpoint commits together | 15:53 |
| mtaylor | markmc: k. great. just making sure | 15:53 |
| mtaylor | markmc: I definitely want to get the patch series review support work in when it lands upstream | 15:54 |
| markmc | mtaylor, honestly, look kernel, qemu, libvirt, etc. mailing lists and you'll see series of 20+ patches | 15:54 |
| markmc | mtaylor, cool | 15:54 |
| mtaylor | markmc: although I _do_ like the typical case being single easy to read patches if it's possible | 15:54 |
| mtaylor | markmc: because we have a hard enough time convincing people to do the work of reviewing single patches :) | 15:54 |
| mtaylor | of course, I _also_ want to get the diff-all-files-on-a-single-page work landed. upstream devs need to code faster! (actually, they need to release faster, I think both features are mostly done) | 15:55 |
| markmc | mtaylor, I'm struggling to understand this, really ... it must just be a POV from using bzr | 15:55 |
| markmc | mtaylor, in at least kernel, qemu, libvirt you have a much higher chance of getting reviewed if you nicely split up your patches | 15:56 |
| markmc | mtaylor, because they're easier to review that way | 15:56 |
| dolphm | comstud: thanks | 15:56 |
| vishy | markmc: that means keeping tests running each commit right? | 15:56 |
| johan_-_ | dolphm: yup | 15:56 |
| markmc | vishy, yep | 15:56 |
| vishy | markmc: how do you do large refactors? | 15:56 |
| markmc | vishy, it depends | 15:57 |
| vishy | markmc: where you're reorganizing a basic piece and you potentially are breaking a lot of tests | 15:57 |
| markmc | vishy, ah | 15:57 |
| markmc | vishy, you refactor in a way that doesn't break tests | 15:57 |
| vishy | markmc: I ask because I recently had to do this in volume-cleanup branch | 15:57 |
| markmc | vishy, even if it means doing stuff that you then remove a few patches down the series | 15:57 |
| markmc | vishy, see, people don't do this out of laziness - they do it to help reviewers and gain bisectability :) | 15:58 |
| mtaylor | markmc: hrm. I don't _think_ it's a bzr pov thing - and I agree, I MUCH prefer that large patches are done in chunks that I can understand | 15:58 |
| markmc | mtaylor, well, with git you should never merge checkpoint commits upstream - that seems standard with bzr? | 15:59 |
| mtaylor | markmc: although most of the time I'd prefer that each chunk is submitted as it's done | 15:59 |
| openstackgerrit | Brian Waldon proposed a change to openstack/glance: Extract image members into new Glance API controller https://review.openstack.org/721 | 15:59 |
| mtaylor | markmc: that's a whole other thing - bzr operates with trees of commits by default, so the way the the mainline commit history looks in bzr is different | 15:59 |
| mtaylor | markmc: but, for instance, on drizzle we used bzr, and required that no merge/review broke tests- and also required that large work be broken up in to reviewable chunks | 16:00 |
| *** ameade has joined #openstack-dev | 16:00 | |
| mtaylor | markmc: so I'm with you on that ... I'm just saying that I'm not sure that getting a sequence of commits all at once is any better than getting each discreet unit of work in the series as it is completed | 16:01 |
| mtaylor | but that may just be a personal thing | 16:01 |
| openstackgerrit | A change was merged to openstack/keystone: Fix swift middleware with regard to latest changes. https://review.openstack.org/738 | 16:03 |
| johan_-_ | mtaylor: do you happen to know what version of paste is on the jenkins server? | 16:03 |
| mtaylor | johan_-_: it's whatever is in the nova-core/trunk ppa | 16:03 |
| *** Rajaram has quit IRC | 16:04 | |
| *** mdomsch has joined #openstack-dev | 16:05 | |
| comstud | mtaylor: it sounds like we need some assistance troubleshoting on jenkins host itself.. if johannes can't reproduce the test failures otherwise | 16:05 |
| johan_-_ | mtaylor: what if there is no paste package in that ppa? | 16:06 |
| comstud | mtaylor: (weird paste failures) | 16:06 |
| mtaylor | comstud: link? | 16:06 |
| comstud | mtaylor: https://jenkins.openstack.org/job/nova/121563/ | 16:07 |
| *** openpercept has joined #openstack-dev | 16:07 | |
| openstackgerrit | Aaron Lee proposed a change to openstack/nova: bug lp845714 https://review.openstack.org/545 | 16:08 |
| johan_-_ | i'm guessing jenkins runs ubuntu and then uses the standard paste packages | 16:09 |
| wwkeyboard | I've rebased 545 onto the current master, could someone please review it? | 16:09 |
| johan_-_ | what version of ubuntu does it run? | 16:09 |
| *** sdake has quit IRC | 16:17 | |
| *** jkoelker_ has quit IRC | 16:17 | |
| *** sdake has joined #openstack-dev | 16:17 | |
| *** jkoelker has joined #openstack-dev | 16:17 | |
| openstackgerrit | A change was merged to openstack/nova: Enforce snapshot cleanup. https://review.openstack.org/726 | 16:18 |
| openstackjenkins | Project nova-docs build #1,439: SUCCESS in 3 min 18 sec: https://jenkins.openstack.org/job/nova-docs/1439/ | 16:24 |
| openstackjenkins | stanislaw.pitucha: Enforce snapshot cleanup. | 16:24 |
| *** zul has quit IRC | 16:26 | |
| mtaylor | johan_-_: the nova slave for jenkins runs natty | 16:26 |
| johan_-_ | ok, so i'm running the same version of paste | 16:26 |
| mtaylor | johan_-_: also seems to be missing lvcreate | 16:27 |
| johan_-_ | jenkins is missing lvcreate? | 16:28 |
| *** code_franco has quit IRC | 16:28 | |
| mtaylor | johan_-_: there are errors further down in that say something about not being able to find lvcreate | 16:29 |
| johan_-_ | i think those are spurious | 16:29 |
| mtaylor | oh, ok | 16:29 |
| johan_-_ | there's a lot of errors | 16:29 |
| mtaylor | johan_-_: is the osapi paste app new in that patch? | 16:30 |
| *** code_franco has joined #openstack-dev | 16:30 | |
| johan_-_ | mtaylor: i change the paste config to use a new handler | 16:30 |
| *** xgmz has joined #openstack-dev | 16:31 | |
| *** openpercept has quit IRC | 16:31 | |
| johan_-_ | the patch certainly touches the paste related parts of nova, so i'm assuming the problem is related to that | 16:31 |
| *** dprince has joined #openstack-dev | 16:37 | |
| mtaylor | johan_-_: hrm. I was mainly asking because I was wondering if the patch might have not included part of something in the commit- thus why it works locally for you? | 16:38 |
| mtaylor | johan_-_: as in, if something didn't get git add'd | 16:38 |
| *** mwhooker has joined #openstack-dev | 16:38 | |
| johan_-_ | mtaylor: i'll double check | 16:40 |
| openstackgerrit | Verification of a change to openstack/keystone failed: Changes to stored hashed password in backends. Using passlib a password hashing library. Using sha512. Setting hashing to be the default behavior. https://review.openstack.org/719 | 16:42 |
| *** zul has joined #openstack-dev | 16:47 | |
| openstackgerrit | A change was merged to openstack/nova: bug lp845714 https://review.openstack.org/545 | 16:47 |
| mtaylor | notmyname, vishy, dolphm: any feedback on the git-review thoughts? if you're not opposed, I'd like to make a plan to move forward... | 16:50 |
| vishy | mtaylor: I'm trying to balance ease of reviews with ease of submission | 16:50 |
| mtaylor | vishy: also, new libvirt push to ppa:mordred/libvirt ... it's broken on lucid and oneirc, which I'll look in to - but built packages for natty and maverick | 16:51 |
| vishy | nice i'm using natty | 16:51 |
| vishy | trying | 16:51 |
| mtaylor | vishy: re: ease of reviews with ease of submission ... is there a thing you'd like for me to make better or worse in there re: replacing the alias with a real command? | 16:52 |
| mtaylor | vishy: (just wanting to make sure I understand what you want/need there) | 16:52 |
| vishy | mtaylor: cat: /var/run/libvirtd.pid: No such file or directory | 16:53 |
| *** zul has quit IRC | 16:53 | |
| openstackjenkins | Project nova-docs build #1,440: SUCCESS in 3 min 14 sec: https://jenkins.openstack.org/job/nova-docs/1440/ | 16:54 |
| openstackjenkins | aaron.lee: bug lp845714 | 16:54 |
| vishy | mtaylor: oh sorry i mixed conversations | 16:54 |
| vishy | mtaylor: i'm ok with what you proposed in the email | 16:54 |
| mtaylor | vishy: goddamit. ok, next time I will do more thorough actual testing before bugging you with it | 16:54 |
| mtaylor | vishy: cool. | 16:54 |
| *** nati2 has quit IRC | 16:54 | |
| vishy | mtaylor: it may just be an issue with the upstart script | 16:55 |
| johan_-_ | mtaylor: i checked out a new tree with my branch and ran the tests locally and it was fine. so i'm definately not missing anything | 16:55 |
| *** nati2 has joined #openstack-dev | 16:55 | |
| mtaylor | johan_-_: bleh | 16:55 |
| *** zul has joined #openstack-dev | 16:56 | |
| openstackgerrit | A change was merged to openstack/nova: Change 'recurse_zones' to 'local_zone_only'. https://review.openstack.org/724 | 16:57 |
| vishy | error : virNetServerMDNSStart:460 : internal error Failed to create mDNS | 16:57 |
| mtaylor | johan_-_: do you have a cloud servers account anywhere? if so, spin up a natty node, do "add-apt-repository ppa:nova-core/ppa; apt-get build-dep nova" and you should have an identical build environment | 16:57 |
| vishy | mtaylor: ^^ that is manually running libvirtd. Missing dep? | 16:57 |
| mtaylor | vishy: I'm going to guess so | 16:58 |
| mtaylor | vishy: doing a three-way merge between debian packaging, ubuntu packaging and new upstream sources is ... fun :) | 16:58 |
| vishy | mtaylor: sounds exciting | 16:58 |
| vishy | mtaylor: don't we have soren for that? :p | 16:58 |
| mtaylor | so exciting I could pee | 16:58 |
| johan_-_ | mtaylor: i'll give that a try | 16:59 |
| mtaylor | vishy: no, we got rid of him :P | 16:59 |
| *** yogirackspace has joined #openstack-dev | 16:59 | |
| *** zul has quit IRC | 17:00 | |
| *** zul has joined #openstack-dev | 17:00 | |
| zykes- | wasn't there a change to glance earlier for dynamic paging ? | 17:01 |
| openstackjenkins | Project nova-docs build #1,441: SUCCESS in 3 min 8 sec: https://jenkins.openstack.org/job/nova-docs/1441/ | 17:05 |
| openstackjenkins | Chris Behrens: Change 'recurse_zones' to 'local_zone_only'. | 17:05 |
| *** zul has quit IRC | 17:10 | |
| *** zul has joined #openstack-dev | 17:10 | |
| *** zul has quit IRC | 17:17 | |
| *** zul has joined #openstack-dev | 17:25 | |
| *** dolphm has quit IRC | 17:27 | |
| *** bcwaldon has quit IRC | 17:29 | |
| *** dolphm has joined #openstack-dev | 17:30 | |
| *** zul has quit IRC | 17:32 | |
| *** dprince has quit IRC | 17:32 | |
| dolphm | mtaylor: just catching up... i'm not opposed! | 17:38 |
| *** zul has joined #openstack-dev | 17:38 | |
| *** nati2 has quit IRC | 17:39 | |
| *** nati2 has joined #openstack-dev | 17:39 | |
| mtaylor | dolphm: great! thanks | 17:40 |
| dolphm | mtaylor: also, where does keystone.conf get installed to? | 17:42 |
| mtaylor | dolphm: I would vote for /etc/keystone/keystone.conf myself (I'm assuming we're talking about the keystone config and not an upstart script) | 17:44 |
| dolphm | mtaylor: there's a apt-get puts a keystone.conf file into /etc/init/ ... but that looks like it's something else? | 17:45 |
| openstackgerrit | A change was merged to openstack/nova: Fixes euca-describe-instances failing or not showing IPs https://review.openstack.org/641 | 17:47 |
| *** dolphm has quit IRC | 17:48 | |
| openstackgerrit | Trey Morris proposed a change to openstack/nova: moved floating ip db access and sanity checking from network api into network manager added floating ip get by fixed address added fixed_ip_get moved floating ip testing from osapi into the network tests where they belong https://review.openstack.org/628 | 17:48 |
| *** dysinger has joined #openstack-dev | 17:50 | |
| *** novas0x2a|laptop has joined #openstack-dev | 17:53 | |
| openstackjenkins | Project nova-docs build #1,442: SUCCESS in 3 min 21 sec: https://jenkins.openstack.org/job/nova-docs/1442/ | 17:54 |
| openstackjenkins | Chris Behrens: Fixes euca-describe-instances failing or not showing IPs | 17:54 |
| *** cp16net has joined #openstack-dev | 17:56 | |
| *** dysinger has quit IRC | 17:59 | |
| *** dysinger has joined #openstack-dev | 17:59 | |
| openstackgerrit | A change was merged to openstack/openstack-ci: Add instructions to push tags on import. https://review.openstack.org/653 | 18:00 |
| zykes- | jaypipes: is there a way to have glance client automatically get credentials instead of always putting exports for AUTH* ? | 18:00 |
| *** xtoddx has joined #openstack-dev | 18:02 | |
| *** mattray has joined #openstack-dev | 18:03 | |
| jaypipes | zykes-: Where should it automatically get creds from? | 18:03 |
| jaypipes | zykes-: some client config file or something? | 18:03 |
| zykes- | i dunno, maybe ? ;) | 18:03 |
| jaypipes | zykes-: open to suggestions :) | 18:03 |
| *** vishy is now known as vishy-afk | 18:04 | |
| zykes- | like a .osauthrc | 18:04 |
| zykes- | or something | 18:04 |
| xtoddx | http://wiki.openstack.org/Nova/ReviewDays still points at launchpad stuff. :( | 18:07 |
| *** mdomsch has quit IRC | 18:07 | |
| soren | xtoddx: *cough* wiki *cough* | 18:09 |
| *** darraghb has quit IRC | 18:11 | |
| xtoddx | how is http://wiki.openstack.org/reviewslist/ built? | 18:19 |
| soren | xtoddx: ttx has a script that scans Launchpad for branches proposed for merge. | 18:20 |
| rmk | Noticing a problem. If rabbit and/or mysql are unavailable when either nova-compute or nova-network startup, they'll never recover without a restart even once the db/bus are available. | 18:20 |
| rmk | I would think the desired behavior is to keep trying to connect and recover without having to nudge them. | 18:20 |
| mtaylor | soren, ttx: we should probably port that list to gerrit... | 18:23 |
| soren | mtaylor: Yes. Yes, we should. | 18:23 |
| mtaylor | although other than score, it seems quite similar to https://review.openstack.org/#q,status:open+project:openstack/nova,n,z | 18:25 |
| mtaylor | soren: do you know how score is calculated? | 18:25 |
| soren | mtaylor: It's based on the importance of the bugs and blueprints linked, the age, the size of the patch and possible other stuff. | 18:26 |
| soren | mtaylor: The script lives on the wiki server, I think. | 18:26 |
| mtaylor | soren: ah, ok | 18:26 |
| mtaylor | soren: I was wondering if it made any sense to add the score just as a built-in gerrit feature, but calculating based on bug/blueprint importance probably puts it over the edge | 18:27 |
| mtaylor | soren: I'll add that to my todo list | 18:27 |
| openstackgerrit | Robie Basak proposed a change to openstack/nova: Move console.log to a ringbuffer https://review.openstack.org/706 | 18:33 |
| zykes- | what's a ringbuffer? | 18:35 |
| wwkeyboard | The oldest stuff gets replaced when newer items come in. | 18:38 |
| notmyname | mtaylor: unfortunately, I haven't looked at it | 18:41 |
| *** nati2_ has joined #openstack-dev | 18:42 | |
| notmyname | jsavak: vishy-afk: I think the auth calls from swift can be reduced to one call. given an auth token, swift can ask for the tenants. if a returned tenant_name matches the swift account in the original request, access should be granted. sound right? | 18:43 |
| *** nati2 has quit IRC | 18:44 | |
| *** dprince has joined #openstack-dev | 18:45 | |
| mwhooker | vishy-afk: thanks for the feedback. I'll look at the extant project and see how that might work | 18:45 |
| dprince | Heads up people reviewing branches! Nova trunk is quickly becoming unusable: | 18:45 |
| dprince | https://bugs.launchpad.net/nova/+bug/862653 | 18:46 |
| uvirtbot | Launchpad bug 862653 in nova "'VDI resize failed' errors on nova-compute (XenServer)" [Critical,Confirmed] | 18:46 |
| dprince | So I can't boot instances on XenServer :( | 18:46 |
| jsavak | notmyname: do you pass the token to other services? or does the user? Your solution is good for swift. In fact https://bugs.launchpad.net/bugs/857671 fixes an issue returning only the scoped tenants for a token. | 18:46 |
| uvirtbot | Launchpad bug 857671 in keystone "Auth protocol doesn't work properly for admin users" [Undecided,Fix committed] | 18:46 |
| dprince | And: https://bugs.launchpad.net/nova/+bug/862658 | 18:46 |
| uvirtbot | Launchpad bug 862658 in nova "fixed_ips.address = ARRAY cast errors with Postgres" [Critical,Confirmed] | 18:46 |
| dprince | Means nova doesn't work w/ PostGres. | 18:46 |
| *** dprince has quit IRC | 18:47 | |
| jsavak | notmyname: but if it's an unscoped token and you allow access based on a get tenants call then pass that token (without getting a new token scoped to that tenant) then the end-user may see access given in swift but not in service x | 18:47 |
| notmyname | jsavak: no, swift does not call out to other services | 18:48 |
| jsavak | notmyname: then your method isn't desirable but should work ok | 18:48 |
| notmyname | jsavak: see my email with updated code that shows only one request | 18:51 |
| mtaylor | notmyname: no worries - have uploaded to pypi so folks can poke at it - will hold off on anything drastic for a bit until we've poked at it for a while | 18:51 |
| mtaylor | zul: ping | 18:52 |
| zul | mtaylor: whats up | 18:52 |
| mtaylor | zul: keystone is adding a depend on passlib for the diablo release | 18:52 |
| mtaylor | zul: which is unpackaged ... I've just made packages of it and uploaded to nova-core/trunk ppa | 18:52 |
| soren | What's passlib | 18:53 |
| soren | :( | 18:53 |
| zul | mtaylor: we havent updated the keystone snapshot in a while btw | 18:53 |
| mtaylor | zul: but I wanted to check in with you about how you wanted to handle that for oneiric | 18:53 |
| mtaylor | zul: ok. | 18:53 |
| mtaylor | soren: it's a password hashing library | 18:53 |
| mtaylor | soren: http://pypi.python.org/pypi/passlib | 18:53 |
| zul | oh the security team is going to have fun with that in oneiric+1 | 18:53 |
| mtaylor | zul: so will it screw you if I add it as a build-dep to lp:~openstack-ubuntu-packagers/keystone/ubuntu ? | 18:54 |
| zul | mtaylor: nope | 18:54 |
| openstackgerrit | Naveed Massjouni proposed a change to openstack/nova: Updating image progress to be more granular. Before, the image progress had only 2 states, 0 and 100. Now it can be 0, 25, 50 or 100. https://review.openstack.org/739 | 18:54 |
| zul | its in the archive thought right? | 18:54 |
| jsavak | notmyname: cool. | 18:54 |
| mtaylor | zul: nope | 18:54 |
| soren | zul: No. 18:52 <+mtaylor> zul: which is unpackaged ... I've just made packages of it and uploaded to nova-core/trunk ppa | 18:54 |
| mtaylor | zul: well, I mean - not yet | 18:54 |
| zul | ah ok | 18:55 |
| notmyname | jsavak: I should clarify. swift doesn't /currently/ call out to any other service, but there is nothing stopping it from doing so in the future, and middleware could certainly be doing so now | 18:56 |
| notmyname | jsavak: so when I pass an auth token to "get tenants", does it return only the tenants currently in scope? | 18:56 |
| jsavak | notmyname: yes - and i'll have to forward to you yogi for more background & fix for bug 857671. It boils down to which URL do you call. One returns all tenants. One returns only the tenants in scope | 18:58 |
| uvirtbot | Launchpad bug 857671 in keystone "Auth protocol doesn't work properly for admin users" [Undecided,Fix committed] https://launchpad.net/bugs/857671 | 18:58 |
| *** mwhooker has quit IRC | 18:58 | |
| *** novas0x2a|laptop has quit IRC | 19:00 | |
| *** mwhooker has joined #openstack-dev | 19:05 | |
| *** bcwaldon has joined #openstack-dev | 19:11 | |
| *** novas0x2a|laptop has joined #openstack-dev | 19:13 | |
| openstackgerrit | Joe Heck proposed a change to openstack/keystone: updating install docs for Mac https://review.openstack.org/740 | 19:14 |
| openstackgerrit | Josh Durgin proposed a change to openstack/glance: Add RBD store backend. https://review.openstack.org/689 | 19:15 |
| *** dysinger has quit IRC | 19:17 | |
| *** bourke_ has joined #openstack-dev | 19:20 | |
| *** bourke has quit IRC | 19:23 | |
| *** blakeyeager has quit IRC | 19:23 | |
| *** blakeyeager has joined #openstack-dev | 19:24 | |
| *** code_franco has quit IRC | 19:25 | |
| *** dysinger has joined #openstack-dev | 19:26 | |
| *** rnorwood has quit IRC | 19:26 | |
| xtoddx | can i see diffs for all files on one page in gerrit? | 19:26 |
| openstackgerrit | Joe Heck proposed a change to openstack/keystone: fixing typo - get -> git https://review.openstack.org/741 | 19:28 |
| *** novas0x2a|laptop has quit IRC | 19:28 | |
| mtaylor | xtoddx: no. it's an open bug - afaik it's been fully fixed by nokia and just hasn't made it in to a release yet | 19:30 |
| xtoddx | mtaylor: thanks, looking forward to that | 19:30 |
| mtaylor | xtoddx: I've been bugging someone for a repo location so I can just pull the patch and apply it to ours so we don't have to wait on the release | 19:30 |
| mtaylor | xtoddx: yes. I think it will be met with much rejoicing | 19:30 |
| *** novas0x2a|laptop has joined #openstack-dev | 19:32 | |
| jaypipes | xtoddx: actually, you can... click the gitweb link next to the patch, then click commitdiff. Only problem is it's readonly (can't do inline comments), but it's all on one page at least ;) | 19:32 |
| xtoddx | thanks jay! | 19:34 |
| *** reed has quit IRC | 19:35 | |
| *** novas0x2a|lapto1 has joined #openstack-dev | 19:35 | |
| *** reed has joined #openstack-dev | 19:35 | |
| *** novas0x2a|laptop has quit IRC | 19:36 | |
| openstackgerrit | Brian Lamar proposed a change to openstack/keystone: Fix bug 861546 https://review.openstack.org/711 | 19:37 |
| uvirtbot | Launchpad bug 861546 in keystone "Keystone tests fail with latest Webob" [Undecided,New] https://launchpad.net/bugs/861546 | 19:37 |
| *** rnorwood has joined #openstack-dev | 19:38 | |
| *** reed has quit IRC | 19:41 | |
| *** novas0x2a|lapto1 has quit IRC | 19:41 | |
| *** novas0x2a|laptop has joined #openstack-dev | 19:44 | |
| *** novas0x2a|laptop has quit IRC | 19:45 | |
| *** novas0x2a|laptop has joined #openstack-dev | 19:48 | |
| *** zaitcev has quit IRC | 19:49 | |
| *** nati2_ has quit IRC | 19:49 | |
| jeblair | dolphm: yogirackspace: if we're going to have to pull in hashlib anyway, why not go ahead and use bcrypt? python-bcrypt is already packaged and in ubuntu | 19:52 |
| *** novas0x2a|laptop has quit IRC | 19:52 | |
| *** reed has joined #openstack-dev | 19:53 | |
| yogirackspace | jeblair: I have used passlib.I liked it as it supports different algorithms.If some one wants to swicth to something different they cud easily do so | 19:54 |
| yogirackspace | jeblair:http://packages.python.org/passlib/ | 19:54 |
| jeblair | yogirackspace: ah, passlib implements bcrypt | 19:55 |
| yogirackspace | it does | 19:55 |
| openstackgerrit | Brian Lamar proposed a change to openstack/nova: Added ==1.0.4 version specifier to kombu in pip-requires to ensure tests pass in a clean venv. https://review.openstack.org/742 | 19:56 |
| yogirackspace | jeblair:http://packages.python.org/passlib/new_app_quickstart.html#recommended-hashes | 19:56 |
| *** novas0x2a|laptop has joined #openstack-dev | 19:56 | |
| jeblair | okay, i gues that doesn't save us anything then. thanks. :) | 19:56 |
| yogirackspace | jeblair: thanks for the suggestions! | 19:57 |
| jeblair | yogirackspace: those docs suggest installing py-bcrypt if you're going to use bcrypt. maybe we should make that a dependency too.. | 19:58 |
| yogirackspace | jeblair: right now I have used sha512_crypt | 19:59 |
| *** Joonwon has quit IRC | 19:59 | |
| jeblair | yogirackspace: oh, so it's not configurable in keystone yet | 20:00 |
| openstackgerrit | Alex Meade proposed a change to openstack/nova: Adds the tenant id to the create server response Location header Fixes bug 862672 https://review.openstack.org/743 | 20:00 |
| uvirtbot | Launchpad bug 862672 in nova "OSAPI 'create image' location header is missing tenant ID " [Medium,Confirmed] https://launchpad.net/bugs/862672 | 20:00 |
| yogirackspace | jeblair: not yet! Wud come pretty soon | 20:00 |
| *** dolphm has joined #openstack-dev | 20:00 | |
| jeblair | yogirackspace: then i guess when that's implemented as an option, we should add that as a depend. since it's already packaged that shouldn't be any more work. | 20:00 |
| yogirackspace | jeblair: sure would create a bug and add as a note | 20:01 |
| jeblair | yogirackspace: I'll do that | 20:01 |
| yogirackspace | jeblair: Thanks | 20:01 |
| *** novas0x2a|laptop has quit IRC | 20:02 | |
| openstackgerrit | Chris Behrens proposed a change to openstack/nova: Fixes bug 862658 -- ec2 metadata issue getting IPs https://review.openstack.org/744 | 20:03 |
| uvirtbot | Launchpad bug 862658 in nova "fixed_ips.address = ARRAY cast errors" [Critical,In progress] https://launchpad.net/bugs/862658 | 20:03 |
| jeblair | bug 862730 | 20:06 |
| uvirtbot | Launchpad bug 862730 in keystone "password hashing should support bcrypt" [Undecided,New] https://launchpad.net/bugs/862730 | 20:06 |
| *** zaitcev has joined #openstack-dev | 20:08 | |
| *** cp16net has quit IRC | 20:13 | |
| openstackgerrit | A change was merged to openstack/nova: Added ==1.0.4 version specifier to kombu in pip-requires to ensure tests pass in a clean venv. https://review.openstack.org/742 | 20:15 |
| *** novas0x2a|laptop has joined #openstack-dev | 20:17 | |
| *** dolphm has quit IRC | 20:17 | |
| openstackgerrit | Alex Meade proposed a change to openstack/nova: Adds the tenant id to the create server response Location header Fixes bug 862672 https://review.openstack.org/743 | 20:17 |
| uvirtbot | Launchpad bug 862672 in nova "OSAPI 'create image' location header is missing tenant ID " [Medium,Confirmed] https://launchpad.net/bugs/862672 | 20:17 |
| openstackjenkins | Project nova-docs build #1,443: SUCCESS in 3 min 19 sec: https://jenkins.openstack.org/job/nova-docs/1443/ | 20:19 |
| openstackjenkins | brian.lamar: Added ==1.0.4 version specifier to kombu in pip-requires to ensure tests pass in a clean venv. | 20:19 |
| *** dolphm has joined #openstack-dev | 20:21 | |
| openstackgerrit | A change was merged to openstack/glance: Extract image members into new Glance API controller https://review.openstack.org/721 | 20:28 |
| *** AhmedSoliman has joined #openstack-dev | 20:29 | |
| jaypipes | comstud: FYI, if in your commit message you mention Bug or LP Bug, the Gerrit system will automatically link (and change status appropriately) the Gerrit review branch with the LP bug... | 20:29 |
| comstud | jaypipes: Yeah, I use that... | 20:30 |
| comstud | jaypipes: which one did I miss? | 20:30 |
| jaypipes | comstud: no, I noticed you just manually added a link to the gerrit review for that latest bug.. | 20:31 |
| jaypipes | comstud: was just pointing it out in case you weren't aware... no worries :) | 20:31 |
| comstud | jaypipes: Oh.. on LP... yeah. For whatever reason, I've seen cases where it's not updated when I go to review | 20:31 |
| comstud | jaypipes: Ie, I'm not sure I've ever seen it update the bug until the commit happens. | 20:32 |
| comstud | jaypipes: Is it supposed to add a comment even at initial review? | 20:32 |
| comstud | jaypipes: because it doesn't seem to | 20:32 |
| *** xtor_ has quit IRC | 20:32 | |
| *** extor has joined #openstack-dev | 20:33 | |
| jaypipes | comstud: ah, I see what you mean... | 20:34 |
| jaypipes | comstud: hmm, no, I don't think it does (that said, easy enough to add that, right jeblair?) | 20:34 |
| comstud | jaypipes: It'd be nice.. I like to point people to an open review when they look up a bug in LP. | 20:36 |
| johan_-_ | i had one not change the status to fix committed: https://bugs.launchpad.net/glance/+bug/850389 | 20:36 |
| uvirtbot | Launchpad bug 850389 in glance "snapshots are not private by default" [Undecided,Fix committed] | 20:36 |
| johan_-_ | i had to change it myself | 20:36 |
| johan_-_ | maybe it was because it was placed on the glance project by mistake? | 20:37 |
| jaypipes | johan_-_: yeah, probably was because of that... | 20:37 |
| johan_-_ | speaking of which, i moved it to the nova project so it gets closed properly | 20:37 |
| openstackgerrit | Verification of a change to openstack/keystone failed: Changes to stored hashed password in backends. Using passlib a password hashing library. Using sha512. Setting hashing to be the default behavior. https://review.openstack.org/719 | 20:39 |
| smoser | in nova, is there anything that will stop the nova-compute node's disk cache from growing indefinitely? | 20:42 |
| openstackgerrit | Jay Pipes proposed a change to openstack/glance: Fixes LP Bug#860862 - Security creds still shown https://review.openstack.org/745 | 20:45 |
| *** rbasak has quit IRC | 20:51 | |
| *** jsavak has quit IRC | 20:54 | |
| *** cp16net has joined #openstack-dev | 20:56 | |
| openstackgerrit | Chris Behrens proposed a change to openstack/nova: Fixes bug 862633 -- OS api consoles create() broken https://review.openstack.org/746 | 20:56 |
| uvirtbot | Launchpad bug 862633 in nova "OS API consoles create() fails because of an extra arg" [Medium,In progress] https://launchpad.net/bugs/862633 | 20:56 |
| *** cp16net has quit IRC | 20:57 | |
| *** cp16net has joined #openstack-dev | 20:58 | |
| openstackgerrit | Brian Waldon proposed a change to openstack/nova: Moving admin actions to extension https://review.openstack.org/621 | 21:04 |
| openstackgerrit | Anne Gentle proposed a change to openstack/openstack-manuals: Adds the Quantum Admin manual and a pom file to build it plus a new disqus ID. https://review.openstack.org/747 | 21:12 |
| *** bcwaldon has quit IRC | 21:13 | |
| openstackgerrit | Josh Durgin proposed a change to openstack/glance: Add RBD store backend. https://review.openstack.org/689 | 21:14 |
| *** hbetts has quit IRC | 21:22 | |
| *** ameade has quit IRC | 21:23 | |
| *** hbetts has joined #openstack-dev | 21:23 | |
| *** mwhooker has quit IRC | 21:25 | |
| *** martine has quit IRC | 21:28 | |
| openstackgerrit | Chris Behrens proposed a change to openstack/nova: Fixes bug 862637 -- make instance_name_template more flexible https://review.openstack.org/748 | 21:35 |
| uvirtbot | Launchpad bug 862637 in nova "FLAGS.instance_name_template needs to be more flexible" [Low,In progress] https://launchpad.net/bugs/862637 | 21:35 |
| *** rbasak has joined #openstack-dev | 21:36 | |
| *** heckj has joined #openstack-dev | 21:36 | |
| *** AhmedSoliman has quit IRC | 21:36 | |
| *** benrodrigue has quit IRC | 21:37 | |
| heckj | I just pulled a fresh checkout of keystone down from github, and I'm not able to get the tests to pass. Anyone else seeing an TypeError from test_extension_overrides_header (http://paste.openstack.org/show/2617/)? | 21:37 |
| devcamcar | anyone from keystone around? | 21:39 |
| devcamcar | hows it going for tomorrow? | 21:39 |
| zykes- | dolphm: probably ;p | 21:39 |
| heckj | dolphm: ping | 21:39 |
| devcamcar | zns: ping | 21:40 |
| yogirackspace | yes | 21:42 |
| yogirackspace | me and dolph r around | 21:42 |
| heckj | yogirackspace: have you see the typeerror paste I dropped in? | 21:42 |
| heckj | http://paste.openstack.org/show/2617/ | 21:42 |
| *** rnirmal has quit IRC | 21:43 | |
| dolphm | heckj: there's an open bug & fix up for review for that | 21:43 |
| *** ohnoimdead has joined #openstack-dev | 21:43 | |
| yogirackspace | saw that just now | 21:43 |
| dolphm | heckj: it's due to a backwards-incompatible update in webob | 21:43 |
| heckj | dolphm: Okay - so I'm not high - wow. | 21:43 |
| yogirackspace | my tests locally do pass | 21:43 |
| heckj | is that dependency coming from PIP or a python package? | 21:44 |
| openstackgerrit | Chris Behrens proposed a change to openstack/nova: Fixes bug 862637 -- make instance_name_template more flexible https://review.openstack.org/748 | 21:44 |
| heckj | I just did a fresh checkout this morning and ran into that - wasn't sure how it got back Jenkins | 21:44 |
| uvirtbot | Launchpad bug 862637 in nova "FLAGS.instance_name_template needs to be more flexible" [Low,In progress] https://launchpad.net/bugs/862637 | 21:44 |
| heckj | (I meant debian package) | 21:44 |
| *** blakeyeager has quit IRC | 21:45 | |
| openstackgerrit | A change was merged to openstack/keystone: Fix bug 861546 https://review.openstack.org/711 | 21:51 |
| uvirtbot | Launchpad bug 861546 in keystone "Keystone tests fail with latest Webob" [Undecided,New] https://launchpad.net/bugs/861546 | 21:51 |
| dolphm | heckj: there's your fix for that ^^ :) | 21:54 |
| heckj | dolphm: Thank you! | 21:55 |
| heckj | dolphm: and thanks for the abandon button clue too - totally missed that in the UI | 21:58 |
| openstackgerrit | razique proposed a change to openstack/openstack-manuals: Fixed typos errors into the nova-volumes section and added clarifications on the setup (iscsitarget VS open-issi). Added notes on KVM device naming https://review.openstack.org/749 | 22:01 |
| *** kbringard has quit IRC | 22:01 | |
| *** rods has quit IRC | 22:03 | |
| dolphm | heckj: np | 22:03 |
| dolphm | i just noticed that the bots triggered each other above... i wonder if we can get them into a loop... | 22:04 |
| dolphm | ^ | 22:05 |
| *** dysinger has quit IRC | 22:06 | |
| openstackgerrit | Yogeshwar Srikrishnan proposed a change to openstack/keystone: Updated guides.Have recompiled to use the latest examples. https://review.openstack.org/750 | 22:06 |
| *** dysinger has joined #openstack-dev | 22:07 | |
| openstackgerrit | A change was merged to openstack/keystone: Updated guides.Have recompiled to use the latest examples. https://review.openstack.org/750 | 22:08 |
| openstackgerrit | Josh Kearney proposed a change to openstack/nova: Adds a script that can automatically delete orphaned VDIs. Also had to move some flags around to avoid circular imports. https://review.openstack.org/693 | 22:08 |
| *** mfer has quit IRC | 22:09 | |
| *** jdurgin has quit IRC | 22:12 | |
| *** ohnoimdead has quit IRC | 22:13 | |
| openstackgerrit | James E. Blair proposed a change to openstack/openstack-ci-puppet: Use deb build depends on keystone slave. https://review.openstack.org/751 | 22:30 |
| openstackgerrit | A change was merged to openstack/openstack-ci-puppet: Use deb build depends on keystone slave. https://review.openstack.org/751 | 22:30 |
| dolphm | jeblair: what OS is gerrit testing on exactly? | 22:31 |
| dolphm | (for keystone) | 22:31 |
| jeblair | dolphm: (you mean jenkins) natty | 22:31 |
| dolphm | ah thanks (and yeah) | 22:32 |
| dolphm | jeblair: what's the strategy for 11.10 and beyond? | 22:32 |
| openstackgerrit | A change was merged to openstack/keystone: Changes to stored hashed password in backends. Using passlib a password hashing library. Using sha512. Setting hashing to be the default behavior. https://review.openstack.org/719 | 22:33 |
| jeblair | dolphm: i'm not positive, but i expect we'll want to move to it soon. it'll be easy to upgrade (actually, replace) the slaves | 22:34 |
| jeblair | dolphm, yogirackspace: ^^ there's your change merged | 22:34 |
| yogirackspace | thanks!! | 22:34 |
| *** yogirackspace has left #openstack-dev | 22:35 | |
| dolphm | i guess it was satisfying enough to /quit | 22:35 |
| *** ohnoimdead has joined #openstack-dev | 22:44 | |
| *** mattray has quit IRC | 22:45 | |
| *** reed has quit IRC | 22:48 | |
| *** reed has joined #openstack-dev | 22:50 | |
| *** hbetts has quit IRC | 22:53 | |
| *** vishy-afk is now known as vishy | 22:58 | |
| *** ohnoimdead has quit IRC | 22:58 | |
| vishy | notmyname: I guess i still don't see the problem with the way the current swift middleware does it. i.e. authorize a token and then match the tenant_id returned to the url. | 23:01 |
| notmyname | vishy: essentially, that's what I think were doing (going to do). some of the confusion earlier was me simply learning more about what keystone is doing | 23:02 |
| vishy | notmyname: ah ok. Coolness. | 23:03 |
| notmyname | vishy: the plan will be to ask keystone for the tenants associated with the auth token and loop through the returned tenants looking for one that matches the swift acount | 23:03 |
| dolphm | notmyname: vishy: i wish it wasn't confusing. | 23:03 |
| vishy | notmyname: so that is still different than the way other services are doing it | 23:04 |
| notmyname | vishy: how are you doing it in nova? | 23:04 |
| vishy | notmyname: the general strategy is authorize token | 23:04 |
| dolphm | do services need/want to share anything in keystone, other than users? (tenants, roles, etc) | 23:04 |
| vishy | that gives you back a single tenant | 23:04 |
| vishy | and you use that tenant for the rest of the request | 23:05 |
| notmyname | vishy: describe "authorize token" | 23:05 |
| openstackgerrit | Dolph Mathews proposed a change to openstack/keystone: POST /tokens: Added tenant id & name to scoped tokens in XML (#862752) https://review.openstack.org/752 | 23:05 |
| notmyname | vishy: do you only make one request to keystone per one request to nova? | 23:05 |
| vishy | notmyname: yes | 23:05 |
| vishy | (discounting ec2 compatibility which does two) | 23:06 |
| notmyname | vishy: and you start with only the auth token? or do you have the tenant id up front too? | 23:06 |
| *** lts has quit IRC | 23:06 | |
| *** rnorwood has quit IRC | 23:06 | |
| vishy | notmyname: we have just the token | 23:06 |
| notmyname | vishy: so what request are you making? | 23:07 |
| vishy | the token should be scoped to a particular tenant id | 23:07 |
| vishy | that is how keystone is supposed to work | 23:07 |
| vishy | https://github.com/openstack/keystone/blob/master/keystone/content/common/samples/validatetoken.json | 23:08 |
| vishy | notmayname: i think that is the one | 23:08 |
| vishy | * notmyname | 23:09 |
| notmyname | that's a response body from a request to generate a token, IIRC | 23:09 |
| notmyname | which is what the client (end-user) does | 23:10 |
| vishy | ah sorry wrong one | 23:11 |
| vishy | basically it is a GET to /v2.0/tokens/<token_id> | 23:11 |
| dolphm | vishy: that's the validate token call | 23:11 |
| dolphm | vishy: which is pretty similar | 23:11 |
| vishy | notmyname: in any case that get returns a tenant id | 23:12 |
| vishy | and we match the tenant_id to the request url and return not-authorized if they don't match | 23:12 |
| notmyname | vishy: I'm currently making a call to GET /v2.0/tenants/ and getting back the list of tenants. I could look at the call to /tokens/<token> and see what happens (tomorrow when I'm back in the office--my test endpoint is internal RAX only) | 23:14 |
| notmyname | vishy: thanks | 23:14 |
| dolphm | notmyname: are you making that call with the client's token? or with an admin token? | 23:15 |
| notmyname | dolphm: the client's | 23:15 |
| *** tryggvil_ has quit IRC | 23:15 | |
| rmk | which iptables rules in nova could be denying multicast? | 23:16 |
| notmyname | vishy: my first priority is to find feature parity to our existing xmlrpc for the rax auth system. we currently call a function with the passed in token from the client and the swift account pulled from the url path. we get back true or falst | 23:16 |
| rmk | Trying to use qugga but it seems like multicast is being blocked | 23:17 |
| rmk | quagga too | 23:17 |
| notmyname | vishy: at least that's where this started :-) | 23:17 |
| *** tryggvil_ has joined #openstack-dev | 23:18 | |
| vishy | notmyname: there is auth middleware in keystone that does all of it for you: https://github.com/openstack/keystone/blob/master/keystone/middleware/auth_token.py | 23:18 |
| vishy | notmyname: underneath you end up with headers that have the user and tenant | 23:19 |
| notmyname | vishy: sort of. the swiftauth is the one in keystone I need to use :-) | 23:19 |
| vishy | notmyname so you can just have an ultra simple middleware underneath that matches the header passed with the url | 23:19 |
| *** jeffjapan has joined #openstack-dev | 23:20 | |
| vishy | so the glance and nova middlewares sit beneath that one in the stack | 23:20 |
| vishy | notmyname: the swift_auth middleware was made to do both, which seems more complicated than necessary | 23:21 |
| vishy | notmyname: if you look at the glance middleware for example, it is ultra simple: https://github.com/openstack/keystone/blob/master/keystone/middleware/glance_auth_token.py | 23:22 |
| vishy | notmyname: it just uses the headers passed down by the auth_token middleware | 23:22 |
| notmyname | dolphm: how similar is keystone to what will be deployed as rack auth? | 23:22 |
| dolphm | notmyname: ... what's rack auth? | 23:23 |
| dolphm | global auth? | 23:23 |
| notmyname | whatever it is your team is deploying at rax | 23:23 |
| notmyname | that cloud files needs to integrate with | 23:24 |
| notmyname | vishy: the swift middleware needs a little more than that. but it looks like I need to spend some time looking more at the middlewares in keystone | 23:25 |
| vishy | notmyname: yes I'm sure it will have to do a little more magic. But the idea in all cases is that you let the auth middleware do the validation for you and you end up with some headers that you can then validate or accept | 23:26 |
| vishy | in your underlying middleware | 23:26 |
| dolphm | notmyname: there's two other teams i sit near, but i don't have much visibility into their work... as far as i know, they're both implementing a subset of the core keystone calls exactly as we are (everything on /tokens at least) | 23:28 |
| notmyname | dolphm: heh. ok :-) | 23:28 |
| dolphm | i'm not even sure what the teams/products are called ... Global Auth and Auth 1.1 i guess? | 23:28 |
| *** lorin1 has quit IRC | 23:30 | |
| openstackgerrit | James E. Blair proposed a change to openstack/openstack-ci: Sync quantum instead of netstack groups https://review.openstack.org/753 | 23:34 |
| *** dolphm has quit IRC | 23:34 | |
| *** heckj has quit IRC | 23:36 | |
| *** nati2 has joined #openstack-dev | 23:41 | |
| *** jeblair has quit IRC | 23:43 | |
| *** mfer has joined #openstack-dev | 23:45 | |
| openstackgerrit | Joe Heck proposed a change to openstack/keystone: updating docs for Mac source install, no docs for mac package install relevant https://review.openstack.org/754 | 23:49 |
| *** xgmz has left #openstack-dev | 23:51 | |
| *** jeblair has joined #openstack-dev | 23:53 | |
| *** anotherjesse has joined #openstack-dev | 23:53 | |
| *** nati2 has quit IRC | 23:54 | |
| rmk | Any ideas which iptables rules being set by nova cause multicasts to be blocked? | 23:56 |
| rmk | I can't see anything but it's definitely being blocked after I start nova | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!