| *** anotherjesse has quit IRC | 00:02 | |
| *** vladimir3p_ has quit IRC | 00:05 | |
| *** patelna has quit IRC | 00:12 | |
| *** fyang has joined #openstack-dev | 00:27 | |
| *** rkukura has joined #openstack-dev | 00:42 | |
| *** dtroyer has quit IRC | 00:55 | |
| *** fyang has quit IRC | 01:34 | |
| *** adjohn has quit IRC | 01:39 | |
| *** zykes- has joined #openstack-dev | 01:39 | |
| *** jdurgin has quit IRC | 01:50 | |
| openstackgerrit | Yogeshwar Srikrishnan proposed a change to openstack/keystone: Fixing issue 854425.ie chaning token table name to tokens. Fixing issue 863667.Changes to support updation of user/tenant name as well using api calls. Fixing LDAP backend to have id independent of name.Fixing getuser call to also return name. https://review.openstack.org/884 | 02:03 |
|---|---|---|
| Kiall | Someone doesnt get the whole 1 review = 1 change idea :) | 02:08 |
| *** rkukura has left #openstack-dev | 02:18 | |
| *** neogenix has joined #openstack-dev | 02:29 | |
| *** neogenix has left #openstack-dev | 02:31 | |
| openstackgerrit | John Tran proposed a change to openstack/nova: bug 817872 https://review.openstack.org/862 | 02:35 |
| uvirtbot | Launchpad bug 817872 in nova "Delete security group that contains instances should not be allowed" [Medium,Confirmed] https://launchpad.net/bugs/817872 | 02:35 |
| *** rkukura has joined #openstack-dev | 02:42 | |
| *** nati2 has joined #openstack-dev | 02:56 | |
| *** KiwiMartin has joined #openstack-dev | 02:56 | |
| *** KiwiMartin has quit IRC | 03:01 | |
| *** rkukura has quit IRC | 03:01 | |
| *** anotherjesse has joined #openstack-dev | 03:06 | |
| *** anotherjesse has quit IRC | 03:07 | |
| *** martine has joined #openstack-dev | 03:10 | |
| openstackgerrit | Brian Waldon proposed a change to openstack/nova: Convert instancetype.flavorid to string https://review.openstack.org/859 | 03:39 |
| *** novas0x2a|laptop has quit IRC | 03:46 | |
| *** zul has quit IRC | 03:58 | |
| *** zul has joined #openstack-dev | 03:59 | |
| *** zul has quit IRC | 04:04 | |
| *** nati2 has quit IRC | 04:04 | |
| *** zul has joined #openstack-dev | 04:05 | |
| *** zedas has quit IRC | 04:05 | |
| *** blakeyeager has quit IRC | 04:08 | |
| *** jakedahn has joined #openstack-dev | 04:15 | |
| *** misheska has joined #openstack-dev | 04:21 | |
| *** blakeyeager has joined #openstack-dev | 04:21 | |
| *** blakeyeager has quit IRC | 04:28 | |
| *** novas0x2a|laptop has joined #openstack-dev | 04:28 | |
| *** tsuzuki_ has joined #openstack-dev | 04:44 | |
| *** jakedahn has quit IRC | 04:53 | |
| *** Vek has quit IRC | 05:00 | |
| *** martine has quit IRC | 05:09 | |
| *** nati2 has joined #openstack-dev | 05:10 | |
| *** zul has quit IRC | 05:13 | |
| *** martine has joined #openstack-dev | 05:13 | |
| *** zul has joined #openstack-dev | 05:13 | |
| *** nati2 has quit IRC | 05:16 | |
| *** novas0x2a|laptop has quit IRC | 05:24 | |
| *** jakedahn has joined #openstack-dev | 05:28 | |
| *** reed has quit IRC | 05:36 | |
| *** zul has quit IRC | 05:43 | |
| *** zul has joined #openstack-dev | 05:44 | |
| *** unabonger has joined #openstack-dev | 05:51 | |
| *** nati2 has joined #openstack-dev | 05:56 | |
| *** zaitcev has quit IRC | 06:09 | |
| *** martine has quit IRC | 06:09 | |
| *** unabonger is now known as backburner | 06:13 | |
| *** Vek has joined #openstack-dev | 06:55 | |
| *** nati2 has quit IRC | 07:29 | |
| *** reidrac has joined #openstack-dev | 08:15 | |
| *** pyhole has quit IRC | 08:55 | |
| *** darraghb has joined #openstack-dev | 09:07 | |
| *** rbasak has joined #openstack-dev | 09:29 | |
| *** tryggvil_ has joined #openstack-dev | 10:14 | |
| *** tsuzuki_ has quit IRC | 10:33 | |
| *** hugokuo has quit IRC | 10:41 | |
| *** rods has joined #openstack-dev | 10:46 | |
| *** mancdaz has quit IRC | 10:59 | |
| *** mancdaz has joined #openstack-dev | 11:00 | |
| *** lorin1 has joined #openstack-dev | 11:37 | |
| *** markmc has quit IRC | 11:52 | |
| *** rkukura has joined #openstack-dev | 12:04 | |
| *** Oneiroi has joined #openstack-dev | 12:23 | |
| *** Oneiroi has joined #openstack-dev | 12:25 | |
| *** med_out is now known as medberry | 12:51 | |
| *** dtroyer has joined #openstack-dev | 12:58 | |
| *** dtroyer has quit IRC | 13:02 | |
| *** xtor_ has joined #openstack-dev | 13:14 | |
| *** extor has quit IRC | 13:14 | |
| *** xtor_ has quit IRC | 13:16 | |
| *** extor has joined #openstack-dev | 13:16 | |
| *** misheska has joined #openstack-dev | 13:19 | |
| *** misheska has quit IRC | 13:19 | |
| *** markmc has joined #openstack-dev | 13:28 | |
| *** Oneiroi has quit IRC | 13:30 | |
| *** misheska has joined #openstack-dev | 13:36 | |
| *** kbringard has joined #openstack-dev | 13:57 | |
| *** martine has joined #openstack-dev | 13:58 | |
| openstackgerrit | Chuck Short proposed a change to openstack/nova: Fix pep8 errors https://review.openstack.org/885 | 14:00 |
| openstackgerrit | Chuck Short proposed a change to openstack/nova: Display LXC container console when using euca-get-console. https://review.openstack.org/728 | 14:00 |
| *** fyang has joined #openstack-dev | 14:01 | |
| *** wwkeyboard has joined #openstack-dev | 14:08 | |
| *** rnorwood has joined #openstack-dev | 14:09 | |
| *** dubsquared has joined #openstack-dev | 14:14 | |
| openstackgerrit | James E. Blair proposed a change to openstack-ci/git-review: Add --setup command for a pro-active repo setup. https://review.openstack.org/871 | 14:20 |
| openstackgerrit | A change was merged to openstack-ci/git-review: Add --setup command for a pro-active repo setup. https://review.openstack.org/871 | 14:21 |
| openstackgerrit | Verification of a change to openstack-ci/git-review failed: Replace git fetch with git remote update. https://review.openstack.org/875 | 14:21 |
| openstackgerrit | Alex Meade proposed a change to openstack/nova: Redirects requests to /v#.# to /v#.#/ Added an APIMapper to catch routemaps when the url is '' because routes.Mapper returns an error if this is the case but we want a redirect Fixes bug 865585 Related to Routes issue: https://bitbucket.org/bbangert/route https://review.openstack.org/886 | 14:22 |
| uvirtbot | Launchpad bug 865585 in nova "/v1.1 returns a 500, /v1.1/ gives desired respons" [Undecided,In progress] https://launchpad.net/bugs/865585 | 14:22 |
| openstackgerrit | Alex Meade proposed a change to openstack/nova: Redirects requests to /v#.# to /v#.#/ Added an APIMapper to catch routemaps when the url is '' because routes.Mapper returns an error if this is the case but we want a redirect Fixes bug 865585 Related to Routes issue: https://bitbucket.org/bbangert/route https://review.openstack.org/886 | 14:23 |
| openstackgerrit | Alex Meade proposed a change to openstack/nova: Redirects requests to /v#.# to /v#.#/ https://review.openstack.org/886 | 14:24 |
| openstackgerrit | James E. Blair proposed a change to openstack-ci/git-review: Replace git fetch with git remote update. https://review.openstack.org/875 | 14:25 |
| openstackgerrit | A change was merged to openstack-ci/git-review: Replace git fetch with git remote update. https://review.openstack.org/875 | 14:25 |
| openstackgerrit | James E. Blair proposed a change to openstack-ci/git-review: fixed problem downloading changes https://review.openstack.org/887 | 14:25 |
| openstackgerrit | Chuck Short proposed a change to openstack/nova: Allow user to decide whether or not to inject ssh keys into the image. Fixes LP: 833499 https://review.openstack.org/888 | 14:39 |
| openstackgerrit | A change was merged to openstack/keystone: Fixing issue 854425.ie chaning token table name to tokens. Fixing issue 863667.Changes to support updation of user/tenant name as well using api calls. Fixing LDAP backend to have id independent of name.Fixing getuser call to also return name. https://review.openstack.org/884 | 14:46 |
| *** reed has joined #openstack-dev | 14:47 | |
| *** rnorwood has quit IRC | 14:48 | |
| *** dragondm has joined #openstack-dev | 14:50 | |
| lorin1 | Does Gerrit let you view an entire file in the web interface, or can you only see the parts that have changed? | 14:50 |
| *** code_franco has joined #openstack-dev | 14:51 | |
| *** adjohn has joined #openstack-dev | 14:51 | |
| wwkeyboard | lorin1: In the side-by-side view you can change the context to "whole file" | 14:55 |
| *** pyhole has joined #openstack-dev | 14:55 | |
| *** martine has quit IRC | 14:56 | |
| wwkeyboard | Does anyone with core access to openstack-integration-tests have time to do some reviewing this morning? https://review.openstack.org/#q,status:open+project:openstack/openstack-integration-tests,n,z | 14:56 |
| lorin1 | wwkeyboard: thanks, I hadn't notice that drop-down before. | 14:57 |
| westmaas | jeblair: howdy sir, you mentioned you were going to make a new core group for openstack-integration-tests, did that happen? And who is in the group? thanks! | 14:59 |
| openstackgerrit | Carlos Marin proposed a change to openstack/keystone: First commit for Secret Question and Answer Extension: RAX-KSQA https://review.openstack.org/861 | 15:00 |
| jeblair | westmass: hi! that's in progress now actually. I created the group a while ago, it's synced to gerrit, and i'm currently updating gerrit to give that group +2 access. | 15:00 |
| westmaas | jeblair: cool. just the same group as nova core? | 15:00 |
| jeblair | westmaas: i only added jaypipes so far. is there a consensus on who sholud be in it? | 15:01 |
| westmaas | nope :) | 15:01 |
| *** markmc has quit IRC | 15:02 | |
| jeblair | okay, i'll nudge him to add some folks next time i see him online, or send email. | 15:02 |
| *** rnorwood has joined #openstack-dev | 15:05 | |
| *** xtoddx has left #openstack-dev | 15:06 | |
| *** reidrac has quit IRC | 15:12 | |
| *** cp16net has joined #openstack-dev | 15:14 | |
| openstackgerrit | A change was merged to openstack-ci/git-review: fixed problem downloading changes https://review.openstack.org/887 | 15:16 |
| kbringard | hey guys, a couple quick ?'s (sorry, I couldn't find it elsewhere) | 15:17 |
| kbringard | when is the maintenance release of diablo being targeted? | 15:17 |
| kbringard | and do we know if https://bugs.launchpad.net/nova/+bug/855660 is in it? | 15:17 |
| uvirtbot | Launchpad bug 855660 in nova "DescribeInstances fails sporadically" [Undecided,Fix committed] | 15:17 |
| wwkeyboard | jeblair: In the meantime how do we go about getting branches merged into the openstack-integration-tests? | 15:18 |
| kbringard | it doesn't look like it's in diablo-backport :-/ | 15:23 |
| *** throughnothing has left #openstack-dev | 15:25 | |
| jeblair | wwkeyboard: i'd prefer to wait for jay to approve or nominate others | 15:26 |
| jeblair | kbringard: does this thread help? https://lists.launchpad.net/openstack/msg04748.html | 15:27 |
| kbringard | jeblair: probably, let me read it :-) | 15:27 |
| kbringard | thanks | 15:27 |
| *** xtoddx has joined #openstack-dev | 15:27 | |
| *** martine has joined #openstack-dev | 15:29 | |
| *** misheska has quit IRC | 15:34 | |
| *** misheska has joined #openstack-dev | 15:35 | |
| *** vladimir3p has joined #openstack-dev | 15:35 | |
| kbringard | one last question, I don't see anything in here about packages being built… are there plans for that or will I need to apply patches manually? | 15:36 |
| *** misheska has quit IRC | 15:36 | |
| jeblair | kbringard: openstack may not | 15:37 |
| jeblair | ack | 15:37 |
| kbringard | bleh, stuff like, euca-describe-instances failing 50% of the time in a "released" version doesn't seem like a good thing | 15:38 |
| jeblair | kbringard: openstack may not produce a release based on the backported fixes, and separately, we're going to stop producing packages as a product. instead the proposals are for the distros to handle packaging, and the maintenance branches are to facilitate distros patching their released versions to provide stable release updates | 15:38 |
| kbringard | ah, makes sense | 15:39 |
| jeblair | kbringard: https://lists.launchpad.net/openstack/msg04789.html | 15:39 |
| jeblair | more info about packaging ^ | 15:39 |
| kbringard | man, I need to be better about reading the mailing list mail, heh | 15:39 |
| kbringard | sorry I suck | 15:39 |
| jeblair | wwkeyboard: based on the wed meeting, it looks like westmaas and dwalleck should be in qa-core | 15:42 |
| jeblair | westmaas: ping | 15:42 |
| *** Oneiroi has joined #openstack-dev | 15:43 | |
| *** Oneiroi has joined #openstack-dev | 15:43 | |
| lorin1 | Anybody else having trouble getting to paste.openstack.org? | 15:43 |
| jeblair | mtaylor: ^ | 15:43 |
| *** markmc has joined #openstack-dev | 15:44 | |
| mtaylor | ooh. that's no good | 15:46 |
| *** rnirmal has joined #openstack-dev | 15:46 | |
| westmaas | jeblair: back | 15:48 |
| jeblair | westmaas: wwkeyboard was asking about getting some patches merged to openstack-integration-tests. based on the transcript of the meeting on wed, i added you to qa-core, so you can +2 reviews in gerrit now. | 15:51 |
| mtaylor | lorin1: fixed | 15:51 |
| lorin1 | mtaylor: thx | 15:52 |
| *** darraghb has quit IRC | 15:54 | |
| *** martine_ has joined #openstack-dev | 15:55 | |
| *** lorin1 has quit IRC | 15:59 | |
| *** adjohn has quit IRC | 16:03 | |
| openstackgerrit | A change was merged to openstack/swift: Fix documentation example. https://review.openstack.org/820 | 16:04 |
| *** medberry is now known as med_out | 16:07 | |
| openstackgerrit | A change was merged to openstack/swift: Query only specific zone via swift-recon https://review.openstack.org/702 | 16:13 |
| openstackgerrit | Carlos Marin proposed a change to openstack/keystone: First commit for Secret Question and Answer Extension: RAX-KSQA https://review.openstack.org/861 | 16:20 |
| openstackgerrit | A change was merged to openstack/swift: Adding what acc/cont/obj into the ratelimit error messages https://review.openstack.org/838 | 16:22 |
| *** lts has joined #openstack-dev | 16:26 | |
| *** Oneiroi has quit IRC | 16:37 | |
| *** Oneiroi has joined #openstack-dev | 16:37 | |
| *** Oneiroi has joined #openstack-dev | 16:37 | |
| *** jdurgin has joined #openstack-dev | 16:45 | |
| *** Oneiroi has quit IRC | 16:47 | |
| openstackgerrit | A change was merged to openstack/keystone: First commit for Secret Question and Answer Extension: RAX-KSQA https://review.openstack.org/861 | 16:55 |
| *** lorin1 has joined #openstack-dev | 16:56 | |
| *** cp16net_ has joined #openstack-dev | 17:01 | |
| *** cp16net has quit IRC | 17:04 | |
| *** cp16net_ is now known as cp16net | 17:04 | |
| openstackgerrit | Vish Ishaya proposed a change to openstack/nova: Makes snapshots work for amis. Fixes bug 873156 https://review.openstack.org/889 | 17:07 |
| uvirtbot | Launchpad bug 873156 in nova "Unhandled exception during 'nova image-create' caused by wrong destination format in qemu-img create" [Medium,New] https://launchpad.net/bugs/873156 | 17:07 |
| *** rbasak has quit IRC | 17:11 | |
| *** corrigac has quit IRC | 17:13 | |
| *** jdg has joined #openstack-dev | 17:23 | |
| openstackgerrit | James E. Blair proposed a change to openstack-ci/git-review: Check .gitreview file in repo for location of gerrit. https://review.openstack.org/890 | 17:24 |
| *** adjohn has joined #openstack-dev | 17:27 | |
| openstackgerrit | Vish Ishaya proposed a change to openstack/nova: Makes snapshots work for amis. Fixes bug 873156 https://review.openstack.org/889 | 17:30 |
| uvirtbot | Launchpad bug 873156 in nova "Unhandled exception during 'nova image-create' caused by wrong destination format in qemu-img create" [Medium,New] https://launchpad.net/bugs/873156 | 17:30 |
| *** jhtran has joined #openstack-dev | 17:31 | |
| westmaas | jeblair: cool thanks | 17:35 |
| jhtran | <comstud> i'm actually re-working that whole thing right now again | 17:36 |
| jhtran | <jhtran> oh snap | 17:36 |
| jhtran | <jhtran> nice! | 17:36 |
| jhtran | <comstud> it's related to some other work being done to separate network stuff out | 17:36 |
| jhtran | <comstud> but | 17:36 |
| jhtran | <comstud> this stuff doesn't perform like it needs to | 17:36 |
| jhtran | <jhtran> u think for my env i can get around it for now by just re-adding the prior lazy load relationship to fixed_ips? | 17:36 |
| jhtran | <comstud> so I'm restoring some of the old code for now | 17:36 |
| jhtran | <comstud> i thought I caught all of the cases where instance['fixed_ips'] was referenced | 17:36 |
| jhtran | <comstud> did I miss one? | 17:36 |
| jhtran | <comstud> Soon... instance['fixed_ips'] will be restored | 17:36 |
| jhtran | <comstud> with the work I'm doing now | 17:36 |
| jhtran | <jhtran> ok cool. thanks | 17:36 |
| jhtran | <jhtran> i'll just hold off then and wait | 17:36 |
| jhtran | <jhtran> thanks again for your time | 17:36 |
| jhtran | <comstud> yeah.. it'll be added back to the joins when getting instances | 17:36 |
| jhtran | oh shit | 17:36 |
| jhtran | my bad | 17:36 |
| jhtran | ignore the above everyone | 17:36 |
| kbringard | ignored :-) | 17:37 |
| comstud | haha | 17:37 |
| kbringard | good thing you weren't talking smack ;-) | 17:39 |
| comstud | true | 17:39 |
| kbringard | that's what I call coding in the opc | 17:40 |
| kbringard | open | 17:40 |
| openstackgerrit | A change was merged to openstack/nova: Added vcpu_weight to models. https://review.openstack.org/874 | 17:41 |
| *** vladimir3p has quit IRC | 17:41 | |
| openstackjenkins | Project nova-docs build #1,492: SUCCESS in 4 min 51 sec: https://jenkins.openstack.org/job/nova-docs/1492/ | 17:50 |
| openstackjenkins | Josh Kearney: Added vcpu_weight to models. | 17:50 |
| openstackgerrit | Verification of a change to openstack/nova failed: Adds the ability to automatically issue a hard reboot to instances that have been stuck in a 'rebooting' state for longer than a specified window. https://review.openstack.org/872 | 17:51 |
| *** vladimir3p has joined #openstack-dev | 17:55 | |
| vladimir3p | vishy: ping | 17:55 |
| *** cdub has quit IRC | 17:58 | |
| *** dtroyer has joined #openstack-dev | 18:03 | |
| *** bcwaldon has joined #openstack-dev | 18:03 | |
| openstackgerrit | A change was merged to openstack-ci/git-review: Check .gitreview file in repo for location of gerrit. https://review.openstack.org/890 | 18:04 |
| *** cp16net has quit IRC | 18:04 | |
| *** cp16net has joined #openstack-dev | 18:05 | |
| *** dtroyer has quit IRC | 18:05 | |
| *** donaldngo_hp has joined #openstack-dev | 18:06 | |
| *** novas0x2a|laptop has joined #openstack-dev | 18:07 | |
| openstackgerrit | Chuck Short proposed a change to openstack/nova: Check to see LXC container can actually run properly on a host before starting. Fixes LP: 83351 https://review.openstack.org/891 | 18:07 |
| *** dak has joined #openstack-dev | 18:12 | |
| openstackgerrit | Brian Waldon proposed a change to openstack/nova: Refactoring nova.tests.api.openstack.test_images https://review.openstack.org/892 | 18:23 |
| openstackgerrit | Brian Waldon proposed a change to openstack/nova: Remove OSAPI v1.0 https://review.openstack.org/840 | 18:23 |
| *** martine has quit IRC | 18:25 | |
| *** martine has joined #openstack-dev | 18:25 | |
| openstackgerrit | Brian Waldon proposed a change to openstack/nova: Refactoring nova.tests.api.openstack.test_images https://review.openstack.org/892 | 18:26 |
| *** zaitcev has joined #openstack-dev | 18:30 | |
| *** rnirmal_ has joined #openstack-dev | 18:32 | |
| *** dragondm has quit IRC | 18:32 | |
| *** martine has quit IRC | 18:34 | |
| *** code_franco_ has joined #openstack-dev | 18:34 | |
| *** rnirmal has quit IRC | 18:34 | |
| *** rnirmal_ is now known as rnirmal | 18:34 | |
| *** code_franco_ has quit IRC | 18:34 | |
| *** code_franco has quit IRC | 18:36 | |
| *** openstackjenkins has quit IRC | 18:37 | |
| *** openstackjenkins has joined #openstack-dev | 18:38 | |
| openstackgerrit | Josh Kearney proposed a change to openstack/nova: Adds the ability to automatically issue a hard reboot to instances that have been stuck in a 'rebooting' state for longer than a specified window. https://review.openstack.org/872 | 18:48 |
| *** kbringard has quit IRC | 18:49 | |
| *** nati2 has joined #openstack-dev | 18:50 | |
| openstackgerrit | A change was merged to openstack/nova: Xenapi driver can now generate swap from instance_type https://review.openstack.org/830 | 18:54 |
| *** kbringard has joined #openstack-dev | 18:56 | |
| openstackjenkins | Project nova-docs build #1,493: SUCCESS in 5 min 10 sec: https://jenkins.openstack.org/job/nova-docs/1493/ | 19:01 |
| openstackjenkins | Rick Harris: Xenapi driver can now generate swap from instance_type | 19:01 |
| openstackgerrit | A change was merged to openstack/nova: Adds the ability to automatically issue a hard reboot to instances that have been stuck in a 'rebooting' state for longer than a specified window. https://review.openstack.org/872 | 19:04 |
| *** jdurgin has quit IRC | 19:04 | |
| *** jdurgin has joined #openstack-dev | 19:04 | |
| *** dragondm has joined #openstack-dev | 19:07 | |
| openstackgerrit | A change was merged to openstack/swift: make drive audit regexes detect 4-letter drives https://review.openstack.org/755 | 19:09 |
| openstackgerrit | Brian Waldon proposed a change to openstack/nova: Refactoring nova.tests.api.openstack.test_images https://review.openstack.org/892 | 19:13 |
| *** tryggvil_ has quit IRC | 19:16 | |
| openstackgerrit | A change was merged to openstack/nova: Ensure non-default FLAGS.logfile_mode is properly converted to an octet. https://review.openstack.org/758 | 19:19 |
| *** mdomsch has joined #openstack-dev | 19:23 | |
| openstackjenkins | Project nova-docs build #1,494: SUCCESS in 4 min 46 sec: https://jenkins.openstack.org/job/nova-docs/1494/ | 19:24 |
| openstackjenkins | Josh Kearney: Adds the ability to automatically issue a hard reboot to instances that have been stuck in a 'rebooting' state for longer than a specified window. | 19:24 |
| openstackgerrit | A change was merged to openstack/nova: Use latest version of SQLAlchemy https://review.openstack.org/708 | 19:30 |
| openstackjenkins | Project nova-docs build #1,495: SUCCESS in 4 min 56 sec: https://jenkins.openstack.org/job/nova-docs/1495/ | 19:31 |
| openstackjenkins | adamg: Ensure non-default FLAGS.logfile_mode is properly converted to an octet. | 19:31 |
| *** rnirmal has quit IRC | 19:34 | |
| *** rnirmal has joined #openstack-dev | 19:34 | |
| *** adjohn has quit IRC | 19:34 | |
| openstackgerrit | Brian Waldon proposed a change to openstack/openstack-chef: Remove OSAPI v1.0 from Nova's api-pate.ini https://review.openstack.org/893 | 19:37 |
| openstackgerrit | Brian Waldon proposed a change to openstack/openstack-chef: Remove OSAPI v1.0 from Nova's api-pate.ini https://review.openstack.org/893 | 19:38 |
| *** code_franco has joined #openstack-dev | 19:38 | |
| *** Vek has quit IRC | 19:39 | |
| openstackgerrit | Brian Waldon proposed a change to openstack/openstack-chef: Remove OSAPI v1.0 from Nova's api-pate.ini https://review.openstack.org/893 | 19:42 |
| openstackjenkins | Project nova-docs build #1,496: SUCCESS in 4 min 50 sec: https://jenkins.openstack.org/job/nova-docs/1496/ | 19:43 |
| openstackjenkins | Mark McLoughlin: Use latest version of SQLAlchemy | 19:43 |
| *** adjohn has joined #openstack-dev | 19:45 | |
| *** lts has quit IRC | 20:11 | |
| *** med_out is now known as medbrry | 20:14 | |
| *** medbrry is now known as medberry | 20:14 | |
| *** rkukura has quit IRC | 20:23 | |
| openstackgerrit | A change to openstack/nova has been rejected: Fix pep8 errors https://review.openstack.org/885 | 20:25 |
| *** rkukura has joined #openstack-dev | 20:25 | |
| *** kbringard has quit IRC | 20:29 | |
| *** kbringard has joined #openstack-dev | 20:30 | |
| *** nhubbard has joined #openstack-dev | 20:37 | |
| kbringard | so, to make sure I'm understanding this correctly, there are no plans to release new PPA versions of the diablo branch to fix some of those issues discovered after the release, right? | 20:53 |
| *** rkukura has quit IRC | 20:53 | |
| kbringard | new stable/milestone packages, I mean | 20:53 |
| kbringard | and if so, is there a recommended path to get those patches onto a system that is running packages from the PPA milestone repo? | 20:55 |
| kbringard | I'm trying to get this documented, since I have a feeling people are going to start asking | 20:55 |
| *** toluene has joined #openstack-dev | 20:57 | |
| *** rkukura has joined #openstack-dev | 20:58 | |
| *** bcwaldon has quit IRC | 20:58 | |
| rmk | mtaylor: Don't suppose there are diablo keystone packages available yet? | 20:59 |
| rmk | kbringard: Actually doing a build of the stable branch now heh. | 21:00 |
| mtaylor | rmk: crap. sorry, no - I've been buried in a few other things. still on my list | 21:00 |
| kbringard | rmk: you mean for the launchpad ppas? or an internal build? | 21:00 |
| rmk | cool np | 21:00 |
| rmk | kbringard: Internal since there isn't an official stable build yet | 21:00 |
| rmk | But I'll gladly publish, nothing special about the build | 21:01 |
| *** lorin1 has quit IRC | 21:01 | |
| kbringard | ah, right, yes… so does that mean there are plans for a new stable? I guess that's what I'm confused about | 21:01 |
| rmk | http://wiki.openstack.org/StableBranch | 21:01 |
| rmk | Sounds like it based on that | 21:01 |
| kbringard | going forward it sounds like we don't want to build packages anymore | 21:01 |
| kbringard | but just provide source and backported patches for the vendors to slipstream in upstream | 21:02 |
| *** nhubbard is now known as nhubbard_away | 21:02 | |
| kbringard | then everyone will just get their packages from Fedora/Ubuntu/Whomever | 21:02 |
| rmk | That's a bit scary | 21:02 |
| kbringard | that's what I thought, but either way, I'm just trying to understand the plan… I don't entirely get it from the stable build wiki page :-) | 21:02 |
| kbringard | https://lists.launchpad.net/openstack/msg04748.html | 21:04 |
| kbringard | err, wait | 21:05 |
| kbringard | https://lists.launchpad.net/openstack/msg04789.html | 21:05 |
| kbringard | specifically: | 21:05 |
| kbringard | - We will not be uploading nova/swift/glance/keystone to PPAs. | 21:05 |
| kbringard | - We will still maintain a PPA with backport packages of depends for dev | 21:05 |
| kbringard | purposes (things like libvirt) This will be a new PPA though, as there | 21:05 |
| kbringard | is really no point in having separate PPAs for each project any more. | 21:05 |
| *** Oneiroi has joined #openstack-dev | 21:06 | |
| *** toluene has quit IRC | 21:09 | |
| notmyname | kbringard: mtaylor: w r t a single PPA for dependencies--this needs to be addressed carefully. although we do have some common dependencies across projects, they, at times, have been dependent on different versions. for example, recently glance started using a newer version of webob that broke something in swift. we resolved it (by going to yet a newer version), but these sorts of things require good communication | 21:10 |
| *** martine_ has quit IRC | 21:11 | |
| mtaylor | notmyname: agree re: communication | 21:14 |
| mtaylor | notmyname: I think the main intent of that PPA is for developers for backports and not for deployments - so hopefully breakage there will cause someone to notice - OR, if it doesn't, we should figure out tests to put in to place to catch that stuff. | 21:15 |
| mtaylor | notmyname: but YES to making sure that we pay decent attention to the depend issues cross project | 21:15 |
| rmk | How many people actually pushed for removing auto generated packages? | 21:15 |
| rmk | The PPAs have been super helpful. | 21:15 |
| rmk | Were there actual users representing the group proposing their elimination, or just developers? | 21:16 |
| openstackgerrit | Josh Kearney proposed a change to openstack/nova: Cancel any clean_reboot tasks before issuing the hard_reboot. https://review.openstack.org/894 | 21:16 |
| rmk | It's just going to make it that much harder to get up and running. As it stands now, add a few PPAs, few apt commands and you can run whichever version of OS you need. | 21:17 |
| notmyname | rmk: we (swift deployers at rackspace) don't care because we don't use them at all. so why spend effort on something people don't use? the "we don't care" also means we don't care if they stay either | 21:17 |
| notmyname | we maintain our own set of packages | 21:18 |
| rmk | notmyname: Your own packages because you use a different distro internally, or other reasons? | 21:19 |
| kbringard | I would argue that, assuming we do decide to get rid of the PPAs, we should still use them for the backports of diablo | 21:19 |
| kbringard | and use essex as a transitin | 21:19 |
| kbringard | transition | 21:19 |
| kbringard | because conventional wisdom for diablo has been "Install from the milestone PPA repo" | 21:19 |
| kbringard | so you're going to have a lot of installs using that repo | 21:19 |
| rmk | I think the conclusion to eliminate packages was drawn by developers who stated they weren't using them, with users not being present to represent usage at the ODS. | 21:19 |
| kbringard | and if the required backport fixes never make it to those people, we're going to have a lot of pissed off users | 21:20 |
| notmyname | mtaylor: however, the dependencies are important. webob for example isn't updated in LTS so we have our own newer version of the code. similarly if we submit patches to a dependency, I'd hate to have to wait for ubuntu to include them before we can fix a bug or add a feature in an openstack project. repos for dependencies are important for deployments | 21:20 |
| notmyname | rmk: other reasons. we use ubuntu internally for deployment. the openstack packages have gotten better, but they started out not meeting our needs at all. so we made our own (since we have a product to maintain) | 21:21 |
| *** cp16net has quit IRC | 21:22 | |
| mtaylor | notmyname: yeah - it's a tricky thing ... because once we start publishing package backport ppas, then we need to also maintain those for security fixes and what not ... and we wind up becomming like a mini distribution | 21:23 |
| mtaylor | rmk: ^^ | 21:24 |
| notmyname | mtaylor: what I've always said we are anyway ;-) | 21:24 |
| mtaylor | notmyname: perhaps so :) | 21:24 |
| openstackgerrit | Aaron Lee proposed a change to openstack/openstack-integration-tests: Adding generic run_tests.sh https://review.openstack.org/870 | 21:24 |
| kbringard | 'cause I can tell you, I'm running diablo from the milestone release, and 2 things in particular are a pain, one the source groups not working and 2 is the lazy load errors every other time I run euca-describe-instances | 21:24 |
| kbringard | but the lazy load thing is like, 650 lines spread across like 12 files… that's a pain in the butt to try to monkeypatch | 21:25 |
| villep | now thats fuckd. | 21:25 |
| kbringard | I'm not suggesting we maintain backport PPAs long term, necessarily, but I think we should at least do a maintenance release for diablo that fies some of the major pain points | 21:26 |
| notmyname | mtaylor: one other thing to keep in mind is that nova and swift are different. for example, the "stable branch" conversations may apply to nova, but they don't seem to fit for swift. the "Stable branch" in swift is master. if you ned an older version, pull it from the tag. but master is always stable (or it doesn't get merged) | 21:26 |
| mtaylor | notmyname: indeed. | 21:27 |
| kbringard | fies == fixed | 21:27 |
| kbringard | dammit, fixes | 21:27 |
| kbringard | typing is hard | 21:27 |
| mtaylor | kbringard: I believe the intent is to maintain a backport branch at the moment - no consensus has been reached on how releases should come out of that from the project | 21:27 |
| *** yogirackspace has joined #openstack-dev | 21:28 | |
| mtaylor | kbringard: although the _distros_ certainly want to maintain it because once they ship something with diablo in it they kind of have to support that | 21:28 |
| villep | I noticed that there were roles defined for nova in: http://bazaar.launchpad.net/~nova-core/nova/github/view/head:/nova/api/ec2/__init__.py#L263 | 21:28 |
| mtaylor | it's a weird little intersection here | 21:28 |
| kbringard | mtaylor: yea, I saw the mailing list stuff about it, my concern is that there isn't really wide spread support from the distros, even fro diablo | 21:28 |
| villep | any clues when swift roles are in ? | 21:28 |
| kbringard | and I know from helping people in #openstack all day that a good lot of people have diablo installed form the milestone PPA repo | 21:28 |
| kbringard | since that was the conventional "how to install diablo" wisdom at the time | 21:29 |
| notmyname | kbringard: true, but that may be quite a selection bias because that's pretty much been the only way to get packages... | 21:29 |
| mtaylor | kbringard: agree... and I think that's actually one of the reasons to move to not making the packages ... I think us providing those packages implies that we're going to maintain and continue to update them in some way | 21:29 |
| annegentle | oh but it's such a pain point right now... so. very. painful. I can't tell people how to install Diablo successfully. | 21:30 |
| mtaylor | and apparently, (which was news to me) there was never really any project intent to continue to apply patches to and support the diablo release after it was released, other than via the distros | 21:30 |
| notmyname | annegentle: what is "diablo"? nova? swift? all of it? | 21:30 |
| kbringard | right, I'm not arguing that we should maintain packages long term, I just think that if we don't do at least some transition | 21:30 |
| vladimir3p | mtaylor: sorry guys for my ignorance prior to that, but are there any scripts for auto-package generation for ubuntu? | 21:30 |
| kbringard | and I'm talking specifically about nova | 21:30 |
| rmk | The problem with relying on distro packages is that they're not going to be updated nearly often enough. | 21:30 |
| kbringard | we're going to have a lot of people who are like "I can't even describe my instances reliably, this isn't stable software" | 21:31 |
| annegentle | notmyname: I have a swift/keystone q on twitter for example | 21:31 |
| mtaylor | vladimir3p: not 100% what you mean? we have packaging branches, and can use those to generate packages | 21:31 |
| annegentle | notmyname: so yes I mean all of it | 21:31 |
| rmk | I like that I can use whichever version of Ubuntu I want (for the most part), and get Diablo installed just by pointing at the PPAs. | 21:31 |
| rmk | If you rely on the distros, you're going to get Ubuntu release = Diablo release. | 21:31 |
| vladimir3p | mtaylor: yep, that is what we need. Just never done it before for nova | 21:31 |
| rmk | And that's the end of that. | 21:31 |
| rmk | So, one would need Ubuntu 11.10 to get Diablo. | 21:31 |
| kbringard | this ^^ is my concern about diablo | 21:32 |
| mtaylor | rmk: yes. the problem is, maintaining that packaging and the backports meant we were working at cross-purposes from the ubuntu devs | 21:32 |
| kbringard | with essex perhaps there will be wider adoption | 21:32 |
| rmk | And that's only true because 11.10 came out after Diablo. | 21:32 |
| rmk | What happens if Essex comes out in May? | 21:32 |
| mtaylor | because they were trying to make oneiric packages, many of the things of which broke packages for previous versions of ubuntu | 21:32 |
| kbringard | but I would argue that there are very few people who are using the Ubuntu packages for Diablo | 21:32 |
| rmk | users wait until October for Essex to make it into Ubuntu? | 21:32 |
| mtaylor | rmk: our release dates are tied to ubuntu release dates | 21:32 |
| rmk | I see | 21:33 |
| mtaylor | rmk: so we will always release before final freeze of the upcoming ubuntu | 21:33 |
| kbringard | yea, soren works on an ongoing basis to coordinate the openstack release with the ubuntu release | 21:33 |
| kbringard | at least that's been my understanding | 21:33 |
| notmyname | which doesn't address people using rhel :-) | 21:33 |
| rmk | mtaylor: That's cool then. I suppose my only concern then would be the lack of the latest OS available for the latest LTS. | 21:33 |
| mtaylor | so, although I agree with rmk and kbringard in theory (I was one of the ones suggesting that we ensure backport packages exist for prior versions of ubuntu) | 21:34 |
| kbringard | anyway, my only point here, is that there are enough very visible issues with the current diablo milestone PPA packages, that we should do at least one more | 21:34 |
| rmk | mtaylor: I really like what you guys are doing right now. | 21:34 |
| rmk | mtaylor: I actually thought you guys were ahead of other projects in this regard. | 21:34 |
| mtaylor | well - it got REALLY painful with the debian extremeley-poorly-managed-transition to dh_python2 | 21:34 |
| kbringard | one more PPA release that incorporates those issues | 21:34 |
| mtaylor | don't even get me started on the complete shit-show they made of that | 21:34 |
| rmk | Yes, the stable PPA should reflect a stable release branch which is always being updated. | 21:35 |
| rmk | That's the only piece missing in my opinion right now. | 21:35 |
| rmk | Which is what kbringard has been saying. | 21:35 |
| mtaylor | but that made is exceptionally difficult to have a single set of packaging that built on oneiric/sid as well as the versions that came before | 21:35 |
| *** Vek has joined #openstack-dev | 21:35 | |
| mtaylor | and yes - I personally agree... | 21:35 |
| kbringard | if we're going to get rid of the PPA's, then it is what it is, but we should probably not do it with the broken version of diablo sitting on there | 21:35 |
| rmk | I think you guys are going to hear more "Openstack is hard to install" comments if the PPAs go away. | 21:36 |
| mtaylor | from a project perspective, long-term maintenance has been the real sticking point | 21:36 |
| rmk | You already get a lot of that and it's insanely easy using the PPAs. | 21:36 |
| mtaylor | :) | 21:36 |
| mtaylor | fair | 21:36 |
| rmk | Maybe reduce the overhead by choosing fewer distros to backport. | 21:37 |
| kbringard | this is the last time I'll say it, I promise :-) but my only beef at the moment is that the version of diablo that most people seem to be installing, is the milestone ppa, which is currently fairly broken, so I think it would be in our best interests to at least do one more push of packages there that incorporate the backports fixes | 21:39 |
| rmk | kbringard: I agree with you 100%. | 21:39 |
| rmk | I'm having to hand-apply patches to Diablo as well right now, specifically for live block migration which doesn't work "out of the box". | 21:39 |
| kbringard | after that if we want to get rid of the PPAs, then lets do it, but we can't tell people in the openstack docs that the way to install is to use the milestone ppas, and then be like "Yea, sorry, those are fucked, but we stopped supporting it" | 21:40 |
| kbringard | good luck! | 21:40 |
| kbringard | the diablo openstack docs, I should say | 21:40 |
| notmyname | kbringard: +1 | 21:41 |
| rmk | I mentioned this earlier but I think the negative responses to eliminating the PPAs weren't represented accurately at the design summit. | 21:41 |
| rmk | You mainly had developers there. | 21:41 |
| rmk | If the next release is going to be about the users, which is the theme I've been hearing, I think the last thing you want to do is take away a convenience service which already exists. | 21:42 |
| openstackgerrit | Kevin L. Mitchell proposed a change to openstack/nova: Add XML templates. https://review.openstack.org/818 | 21:42 |
| annegentle | kbringard: yes. | 21:43 |
| rmk | Enhancing it to include builds out of a "stable" branch for each project, pleased in the stable PPA would be the cherry on top for me. | 21:43 |
| annegentle | rmk: also yes. | 21:43 |
| rmk | That's the only thing missing. | 21:43 |
| rmk | placed* | 21:43 |
| annegentle | mtaylor: will it help for me to try to plead this case on the ML? | 21:44 |
| rmk | Anyway, that's my two cents. I'll reply to the thread if there's still time to change this decision. | 21:44 |
| notmyname | rmk: does this meet your needs (imagine the same for nova if you need that): http://crashsite.github.com/swift_debian/ | 21:44 |
| notmyname | rmk: ^ has all of the dependencies and also a separate (stable!) repo for past versions | 21:44 |
| notmyname | rmk: that's the repo we deploy from at rax for swift (cloud files) | 21:45 |
| rmk | notmyname: Yes that's great. Basically I don't care where the repos are, as long as they exist for each project which satisfy all the deps. | 21:45 |
| mtaylor | annegentle: well, it would be great if you or rmk would say something on the ml | 21:45 |
| rmk | If there's one for swift, another for nova, etc, not on launchpad, doesn't matter as long as the location is documented somewhere. | 21:45 |
| rmk | mtaylor: Will do. | 21:45 |
| annegentle | notmyname: yes, I can doc packages that are known to work | 21:45 |
| mtaylor | so - the problem is for the other things that there is nobody currently tasked with maintaining packaging produced by the project | 21:46 |
| mtaylor | I'm CERTAINLY not saying it's a bad idea to make them - I've been championing them for a while now | 21:46 |
| notmyname | mtaylor: isn't that #43 on your list of stuff to do? ;-) | 21:46 |
| rmk | There's a lack of consistency in that regard across packages. In fact, I think that spans across the release process too. | 21:46 |
| rmk | err projects not packages | 21:46 |
| mtaylor | but with no policy for post-release updates for maintenance, it's a bit of a nightmare | 21:46 |
| mtaylor | I mean, I can cut trunk packages just fine | 21:47 |
| mtaylor | but what happens post diablo release? | 21:47 |
| rmk | Yeah that's a huge, huge issue. | 21:47 |
| mtaylor | who owns the package contents? who fixes the bugs in them? | 21:47 |
| rmk | Though honestly it's worthwhile to tackle this within the OS project. | 21:47 |
| mtaylor | with distro-generated packages, at least the distros are on the hook for fixing your bugs | 21:47 |
| mtaylor | it IS worthwhile | 21:47 |
| notmyname | mtaylor: rmk: since nobody on the openstack side is maintaining packages, we will continue to maintain the crashsite ones for swift | 21:47 |
| mtaylor | it's just that no one has stepped up and said they want to do it | 21:47 |
| mtaylor | notmyname: great! | 21:47 |
| notmyname | mtaylor: I definitely think someone should do it for openstack :-) (although I expect pretty much every major deployer to maintain their own deploy repos) | 21:48 |
| notmyname | where major == large scale | 21:48 |
| mtaylor | notmyname: see - that's just the problem | 21:48 |
| mtaylor | the people who are actually using this stuff at scale are maintaining their own repos (understandably) | 21:49 |
| openstackgerrit | A change was merged to openstack/nova: Cancel any clean_reboot tasks before issuing the hard_reboot. https://review.openstack.org/894 | 21:49 |
| mtaylor | so we're not really getting that feedback into the project produced packages | 21:49 |
| rmk | mtaylor: You know a lot of that has to do with the lack of an updated stable package. | 21:49 |
| *** adjohn has quit IRC | 21:49 | |
| rmk | I have to cut my own packages too now because I need several patches just to fix critical issues in Diablo. | 21:49 |
| rmk | Granted having the packaging bzr makes this easy. | 21:50 |
| *** rkukura has left #openstack-dev | 21:50 | |
| mtaylor | my original discussions on packaging at ODS were on trying to make it easier for major deployers to contribute to the main packaging with the intent that perhaps we could produce packages that a major deployer wanted to install from | 21:50 |
| mtaylor | but as we got in to it, it seemes as if we were trying to solve a problem for a use case that was never going to happen ... that is, none of the major deployers were going to use them, ever | 21:50 |
| mtaylor | which means we're producing packages which are clearly not good | 21:50 |
| rmk | Maybe not, but users do/are. | 21:51 |
| notmyname | mtaylor: but major deployers are more likely to have their own internal stuff too. at rax we also have a rackswift repo that includes proprietary auth and billing stuff | 21:51 |
| mtaylor | notmyname: there's a difference though - if you were making packages directly using upstream packaging source and just cutting them in to a repo, that would be one thing | 21:51 |
| rmk | If automated trunk packages weren't being produced leading up to Diablo, I wouldn't have been testing Diablo actively. | 21:51 |
| mtaylor | because then the best way to fix a packaging issue would be to get the commit into packaging trunk | 21:51 |
| mtaylor | (which is what I was trying to get rolled out for essex for openstack) | 21:52 |
| *** rnorwood has quit IRC | 21:52 | |
| mtaylor | I'm happy to revisit the idea ... but I honestly need some sort of buy in from someone that having those packages is a goal, and that we want them to meet some set of criteria | 21:53 |
| notmyname | mtaylor: I think that smaller scale deployers will still use whatever package repo we provide (and I think rmk agrees) | 21:53 |
| kbringard | I personally don't mind installing from source to track trunk, but I can tell you, I help a shitton of people in #openstack and the reason I'm able to is because I was able to track trunk on a daily basis, so I knew most of what needed to be changed migrsting from cactus to diablo | 21:53 |
| mtaylor | which means, much as notmyname (and I) contend - that we are making a micro-distribution - and therefore have to figure out how to support ourselves as one | 21:53 |
| rmk | yep | 21:53 |
| notmyname | mtaylor: project autonomy! ;-) | 21:53 |
| kbringard | so there is definitely utility in tracking trunk | 21:54 |
| mtaylor | notmyname: no. that's the exact OPPOSITE of what we need here :) | 21:54 |
| mtaylor | notmyname: you providing swift packages is great and all - but how exactly do we determine that those work with nova/glance/keystone for a diablo thing? also, your project just releases again in 2 weeks... how does that work with ongoing maintenance? | 21:54 |
| notmyname | mtaylor: well, I certainly don't want to get into that debate again (at least now) ;-) | 21:54 |
| mtaylor | notmyname: aw... | 21:54 |
| mtaylor | notmyname: but we skipped it totally at the conference! :) | 21:55 |
| notmyname | we've got a long plane ride to brazil :-) | 21:55 |
| mtaylor | yes we do | 21:55 |
| mtaylor | although I don't think we're on the same plane... | 21:55 |
| mtaylor | unless you are taking a terrible route | 21:55 |
| notmyname | lol | 21:55 |
| mtaylor | we DO, on the other hand, have many hours of drinking in brazil ahead of us | 21:56 |
| *** rkukura has joined #openstack-dev | 21:56 | |
| openstackjenkins | Project nova-docs build #1,497: SUCCESS in 5 min 2 sec: https://jenkins.openstack.org/job/nova-docs/1497/ | 21:56 |
| *** rnirmal has quit IRC | 21:56 | |
| openstackjenkins | Josh Kearney: Cancel any clean_reboot tasks before issuing the hard_reboot. | 21:56 |
| kbringard | drinking debates are the best debates | 21:57 |
| mtaylor | hell yes | 21:57 |
| notmyname | we determine that the releases work together 2 ways: 1) talk about it 2) the guarantee that each 6-month release works with the other projects (or else you wouldn't be a core project) | 21:57 |
| mtaylor | totally. | 21:57 |
| mtaylor | I'm talking about stable release updates | 21:57 |
| mtaylor | if we release this 6 month thing called diablo ... and then there are bugfixes which need to be applied - let's call them security fixes | 21:57 |
| mtaylor | how does that thing called diablo get managed | 21:58 |
| notmyname | but honestly, I have it easier on the swift side because we are required to be stable and backwards compatible all the time. nova/keystone/etc are in more of a state of flux and will be changing contracts more rapidly | 21:58 |
| mtaylor | ++ | 21:59 |
| mtaylor | this is an easier problem when things aren't quite in as much rapid flux | 21:59 |
| notmyname | indeed. which lets me get up on my ivory tower and look down on everyone ;-) | 22:00 |
| kbringard | isn't that what the backports branch is all about? managing the patches for upstream distros to put in their packages? | 22:00 |
| notmyname | (that was totally kidding, if it wasn't clear) | 22:00 |
| kbringard | and again, I'm not advocating that we do this long term, I just think we need to make sure the diablo milestone packages are less bjorked | 22:01 |
| kbringard | even if it's just one more push | 22:01 |
| notmyname | kbringard: agreed. broken packages are much worse than no packages at all | 22:01 |
| notmyname | (but on that note I'm mostly a spectator. I don't have much skin in the game for nova) | 22:01 |
| kbringard | yea, and for me, nova is the main show | 22:03 |
| kbringard | and I think for the majority of people | 22:03 |
| notmyname | I think you're right ;-) | 22:03 |
| kbringard | so anyway, sorry to stir up that hornet's nest | 22:03 |
| kbringard | lol, I was just trying to get clarification on what the plan was | 22:04 |
| notmyname | mtaylor: so there is noone on the openstack side who's job it is is to manage the packages for the projects? | 22:04 |
| mtaylor | notmyname: me. | 22:04 |
| notmyname | mtaylor: do you have minions? | 22:04 |
| mtaylor | notmyname: except that the unanswered policy stuff I mentioned is the main blocker | 22:04 |
| *** dragondm has quit IRC | 22:05 | |
| mtaylor | notmyname: i.e. - I can make packages - the big question is "what content goes in to them" ... this is especially tricky if the project itself is not producing post-release updates | 22:05 |
| notmyname | indeed | 22:05 |
| mtaylor | notmyname: but ... I'm working on gathering some minions | 22:05 |
| kbringard | mtaylor: yes, that is where what I'm proposing (more more package push) hits a bump | 22:05 |
| kbringard | more more == one more | 22:06 |
| notmyname | medberry: and so the "what content goes in to them" question leads to the stable branch discussions, right? | 22:06 |
| notmyname | mtaylor: ^ | 22:06 |
| mtaylor | notmyname: yes. and we got _somewhere_ on those at least | 22:06 |
| mtaylor | in that we now have stable branches and teams tasked with maintaining them | 22:07 |
| mtaylor | turning the contents of those into releases is still without a person at the help | 22:07 |
| mtaylor | helm | 22:07 |
| notmyname | for everything but swift? | 22:07 |
| mtaylor | nope. swift too | 22:07 |
| notmyname | ah? | 22:07 |
| notmyname | who? | 22:07 |
| mtaylor | the stable branches are owned by the distros and deployers | 22:07 |
| medberry | notmyname, nod. | 22:08 |
| mtaylor | so that any patches they're putting in to their distro releases have a place to land so that they can be sharead by others | 22:08 |
| notmyname | oh ok. so eg someone at canonical for ubuntu | 22:08 |
| mtaylor | yup | 22:08 |
| notmyname | medberry: sorry. autocomplete fail on my part :-) | 22:08 |
| medberry | yup | 22:08 |
| mtaylor | someone at piston, and at nebula, and at rax and at hp and redhat | 22:08 |
| mtaylor | at least, that's the list of stakeholders for those branches off the top of my head | 22:08 |
| notmyname | mtaylor: who at rax for swift? (If feel like I should know this person) | 22:09 |
| notmyname | ah ok | 22:09 |
| mtaylor | notmyname: it's not a person at rax per-project at the moment | 22:09 |
| mtaylor | notmyname: at least, it's not specifically set up that way - I imagine given how swift is organized, someone from your group would likely want to be involved | 22:09 |
| notmyname | s/want to/should/ :-) | 22:10 |
| mtaylor | notmyname: for nova, I think someone from westmaas's or pvo's group is the person we'd be thinking of, ish. | 22:10 |
| mtaylor | notmyname: s,s/want to/should/,s/should/want to/, | 22:10 |
| notmyname | mtaylor: heh. I'm not trying to build fiefdoms. just figuring that the swift devs should be involved in maintaining swift :-) | 22:13 |
| mtaylor | notmyname: god no - I completely agree | 22:13 |
| mtaylor | notmyname: it's more of a "hey, we're gonna set up stuff for folks to maintain this stuff ... now somebody should probably maintain this stuff" | 22:14 |
| notmyname | I do think we (swift) can and should do a better job, long term, of supporting some older versions of swift. currently our team isn't really set up to do this, but it is something we should acknowledge | 22:14 |
| mtaylor | ++ | 22:15 |
| mtaylor | and I think that's actually across the board | 22:15 |
| mtaylor | nobody is really set up to support older versions of stuff | 22:15 |
| mtaylor | you're in a decent place beacause most of the time just upgrading to the latest swift is unlikely to completely hose you | 22:16 |
| notmyname | it's an advantage, but "just upgrade to a newer version" is rarely helpful advice to other deployers | 22:16 |
| notmyname | I can be pretty sure it won't break anything, but the other guys will have their own policies and red tape to go through | 22:17 |
| mtaylor | ++ | 22:18 |
| * notmyname wonders about auto-upgrades in the project...nah, so many people would hunt me with pitchforks | 22:18 | |
| * mtaylor is always amused that people are fine upgrading from 1.0-1 to 1.0-2 but not from 1.0 to 1.1 ... even though the debian maintainers could have added god-knows-what sort of patches in -2 | 22:18 | |
| kbringard | notmyname: use sparkle! | 22:18 |
| kbringard | :-p | 22:18 |
| notmyname | perfect! | 22:19 |
| kbringard | mtaylor: I agree 100% | 22:19 |
| mtaylor | jesus. is that another one of those hippy web 2.0 ruby bullshits? | 22:19 |
| mtaylor | yup | 22:19 |
| kbringard | it's even more entertaining trying to explain debian patches to a PCI auditor | 22:19 |
| mtaylor | software updates for hipster-coder-macs | 22:19 |
| * mtaylor pukes on everyone | 22:20 | |
| mtaylor | kbringard: I'll bet! | 22:20 |
| kbringard | when your entire external scan is red because the "version" is from like 10 years ago | 22:20 |
| kbringard | even though every security patch has been slipstreamed in | 22:20 |
| * kbringard shakes his fist in the air | 22:21 | |
| * mtaylor loved the redhat kernel versions from rhel4 | 22:21 | |
| kbringard | anyway, good discussion guys… so what are we going to do about it? :-p | 22:22 |
| mtaylor | that were labeled 2.4 | 22:22 |
| mtaylor | except that they had backported most of 2.6 in | 22:22 |
| mtaylor | but that way it didn't count as a new kernel version | 22:22 |
| kbringard | oh, that's terrible | 22:22 |
| mtaylor | kbringard: I'll send a follow up email to the list ... let me see what kind of resources for supporting something I can come up with | 22:22 |
| mtaylor | along with some concrete suggestions | 22:22 |
| kbringard | for PCI I had to search for every single vulnerability ID on the debian security site and reference it | 22:23 |
| mtaylor | and then maybe notmyname and I can bum-rush the ppb when they aren't looking | 22:23 |
| kbringard | hehe | 22:23 |
| kbringard | if you're in austin I'll send my boss down | 22:23 |
| kbringard | he's good at bum rushin' | 22:23 |
| mtaylor | it's all virtual - I'm in New York ... but I'll take the help anyway | 22:23 |
| kbringard | hehe, yea, I'm in Colorado | 22:24 |
| kbringard | either way, well, the only concrete thing I think is essential is that we do at least one more diablo milestone PPA push with packages that incorporate fixes for some of the high profile things we've seen in the diablo release | 22:24 |
| kbringard | so there you have it | 22:25 |
| kbringard | :-) | 22:25 |
| kbringard | thanks for taking the time to talk about it | 22:25 |
| kbringard | I need to call it a day, have a good one guys | 22:25 |
| *** kbringard has quit IRC | 22:26 | |
| *** dubsquared has quit IRC | 22:26 | |
| *** wwkeyboard has left #openstack-dev | 22:28 | |
| *** code_franco has quit IRC | 22:31 | |
| *** misheska has joined #openstack-dev | 22:33 | |
| openstackgerrit | Yogeshwar Srikrishnan proposed a change to openstack/keystone: Fixing issue #843226. https://review.openstack.org/895 | 22:35 |
| openstackgerrit | Yogeshwar Srikrishnan proposed a change to openstack/keystone: Fixing issue #843226. Changes to throw appropriate faults during token validation. Fixing issue #854929. Changing incorect 'text/json' references to 'application/json'. https://review.openstack.org/896 | 22:35 |
| *** yogirackspace has left #openstack-dev | 22:41 | |
| *** rkukura has quit IRC | 22:49 | |
| *** medberry is now known as med_out | 22:50 | |
| *** jdg has quit IRC | 23:03 | |
| *** jhtran has quit IRC | 23:04 | |
| *** fyang has quit IRC | 23:18 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!