Tuesday, 2011-10-18

*** adam_g has quit IRC00:00
*** dragondm has quit IRC00:18
*** dolphm_ has quit IRC00:18
*** adjohn has quit IRC00:20
*** adjohn has joined #openstack-dev00:21
*** mattray has quit IRC00:21
*** dragondm has joined #openstack-dev00:38
*** bourke has quit IRC00:45
*** bourke has joined #openstack-dev00:45
*** rods has quit IRC00:47
*** f-yang1 has quit IRC00:51
*** jdag has quit IRC00:51
*** zul has quit IRC00:51
*** ipl31 has quit IRC00:51
*** misheska has quit IRC00:51
*** loceur has quit IRC00:51
*** colinnich has quit IRC00:51
*** xtoddx has quit IRC00:51
*** nci has quit IRC00:51
*** termie has quit IRC00:51
*** Glacee has quit IRC00:51
*** JordanRinke has quit IRC00:51
*** dabo has quit IRC00:51
*** Daviey has quit IRC00:51
*** openstackgerrit has quit IRC00:51
*** ttx has quit IRC00:51
*** troytoman-away has quit IRC00:51
*** cburgess has quit IRC00:51
*** asalkeld has quit IRC00:51
*** ameade has quit IRC00:51
*** clayg_ has quit IRC00:51
*** novas0x2a|laptop has quit IRC00:51
*** jdurgin has quit IRC00:51
*** Guest66319 has quit IRC00:51
*** creiht has quit IRC00:51
*** negronjl has quit IRC00:51
*** zorzar has quit IRC00:51
*** misheska has joined #openstack-dev00:55
*** f-yang1 has joined #openstack-dev00:55
*** jdag has joined #openstack-dev00:55
*** jdurgin has joined #openstack-dev00:55
*** asalkeld has joined #openstack-dev00:55
*** zul has joined #openstack-dev00:55
*** xtoddx has joined #openstack-dev00:55
*** Guest66319 has joined #openstack-dev00:55
*** ameade has joined #openstack-dev00:55
*** 31NAADY1U has joined #openstack-dev00:55
*** ipl31 has joined #openstack-dev00:55
*** creiht has joined #openstack-dev00:55
*** negronjl has joined #openstack-dev00:55
*** troytoman-away has joined #openstack-dev00:55
*** ttx has joined #openstack-dev00:55
*** openstackgerrit has joined #openstack-dev00:55
*** Daviey has joined #openstack-dev00:55
*** dabo has joined #openstack-dev00:55
*** JordanRinke has joined #openstack-dev00:55
*** cburgess has joined #openstack-dev00:55
*** Glacee has joined #openstack-dev00:55
*** termie has joined #openstack-dev00:55
*** nci has joined #openstack-dev00:55
*** colinnich has joined #openstack-dev00:55
*** loceur has joined #openstack-dev00:55
*** zorzar has joined #openstack-dev00:55
*** hubbard.freenode.net sets mode: +v creiht00:55
*** mtaylor has quit IRC00:55
*** mtaylor_ has joined #openstack-dev00:55
*** ohnoimdead has quit IRC01:07
*** mgius has quit IRC01:13
*** novas0x2a|laptop has joined #openstack-dev01:23
*** vladimir3p has quit IRC01:26
*** hugokuo has joined #openstack-dev01:31
*** _adjohn has joined #openstack-dev01:36
*** adjohn has quit IRC01:39
*** jakedahn has joined #openstack-dev01:53
*** thickskin has joined #openstack-dev01:53
*** jdurgin has quit IRC02:02
*** HugoKuo_ has joined #openstack-dev02:04
*** medberry is now known as med_out02:06
*** hugokuo has quit IRC02:06
openstackgerritLorin Hochstein proposed a change to openstack/nova: Updated RST docs: bzr/launchpad -> git/github  https://review.openstack.org/91302:10
*** rnorwood has joined #openstack-dev02:23
*** toluene has joined #openstack-dev02:25
*** beekhof has joined #openstack-dev02:27
*** novas0x2a|laptop has quit IRC02:35
*** blamar has quit IRC02:41
*** blamar has joined #openstack-dev02:41
*** adjohn has joined #openstack-dev02:47
*** adjohn has quit IRC02:52
*** thickskin has left #openstack-dev03:00
*** joonwon has quit IRC03:01
*** HugoKuo__ has joined #openstack-dev03:02
*** HugoKuo_ has quit IRC03:06
*** asalkeld has quit IRC03:10
*** chomping has quit IRC03:17
*** asalkeld has joined #openstack-dev03:18
*** asalkeld has joined #openstack-dev03:19
*** asalkeld has quit IRC03:23
*** vladimir3p has joined #openstack-dev03:23
*** asalkeld has joined #openstack-dev03:23
*** vladimir3p has quit IRC03:28
*** hugokuo has joined #openstack-dev03:30
*** dragondm has quit IRC03:33
*** mattray has joined #openstack-dev03:33
*** HugoKuo__ has quit IRC03:34
*** f-yang1 has quit IRC03:43
*** chomping has joined #openstack-dev03:48
*** nci has quit IRC03:49
*** nci has joined #openstack-dev03:51
*** chomping has quit IRC03:58
*** adjohn has joined #openstack-dev04:14
*** mattray has quit IRC04:23
*** chomping has joined #openstack-dev04:30
*** adjohn has quit IRC04:31
*** chomping has left #openstack-dev04:34
*** novas0x2a|laptop has joined #openstack-dev04:39
*** adjohn has joined #openstack-dev04:43
*** chomping_ has joined #openstack-dev04:50
*** chomping_ has quit IRC04:57
*** zaitcev has quit IRC04:59
openstackgerritBrad Hall proposed a change to openstack/quantum: Change port/net create calls to take an additional kwargs param  https://review.openstack.org/83505:00
openstackgerritBrad Hall proposed a change to openstack/quantum: Add the ability to specify multiple extension directories  https://review.openstack.org/83905:00
openstackgerritBrad Hall proposed a change to openstack/quantum: ovs plugin: Remove reference to set_external_ids.sh  https://review.openstack.org/66605:00
openstackgerritBrad Hall proposed a change to openstack/quantum: Add code-coverage support to run_tests.sh (lp860160)  https://review.openstack.org/67705:00
openstackgerritBrad Hall proposed a change to openstack/nova: Fix quantum/melange ipam interaction  https://review.openstack.org/91405:09
openstackgerritBrad Hall proposed a change to openstack/nova: Change network delete to delete by uuid instead of cidr  https://review.openstack.org/91505:09
openstackgerritBrad Hall proposed a change to openstack/nova: Add DHCP support to the QuantumManager and break apart dhcp/gateway  https://review.openstack.org/91605:09
openstackgerritBrad Hall proposed a change to openstack/nova: Change floating-snat to float-snat  https://review.openstack.org/91705:09
*** jakedahn has quit IRC05:10
*** adjohn has quit IRC05:12
*** tsuzuki_ has joined #openstack-dev05:18
*** adjohn has joined #openstack-dev05:18
*** mdomsch has quit IRC05:20
openstackgerritMark McLoughlin proposed a change to openstack/nova: Chain up to superclass tearDown in ServerActionsTest  https://review.openstack.org/89805:24
*** zigo has joined #openstack-dev05:25
*** itooon has joined #openstack-dev05:44
*** misheska has quit IRC06:02
*** rnorwood has quit IRC06:18
*** rnorwood has joined #openstack-dev06:21
openstackgerritA change was merged to openstack/quantum: Add code-coverage support to run_tests.sh (lp860160)  https://review.openstack.org/67706:21
openstackgerritA change was merged to openstack/quantum: ovs plugin: Remove reference to set_external_ids.sh  https://review.openstack.org/66606:21
*** itooon has quit IRC06:43
*** itooon has joined #openstack-dev07:04
*** cmagina has quit IRC07:15
*** reidrac has joined #openstack-dev07:21
*** adjohn has quit IRC07:26
*** hugokuo has quit IRC07:30
*** hugokuo has joined #openstack-dev07:30
*** Oneiroi has joined #openstack-dev07:30
*** Oneiroi has joined #openstack-dev07:30
*** rbasak has joined #openstack-dev07:38
*** superbobry has joined #openstack-dev07:40
*** nickon has joined #openstack-dev07:52
*** itooon has quit IRC08:11
zykes-does gance have a bug in it's diablo release towards keystone or am I doing something wrong? keystone requires "auth" to be present in the body of the request and glance doesn't send it ?08:28
zykes-https://github.com/openstack/keystone/blob/stable/diablo/keystone/logic/types/auth.py < i'm looking at this and comparing it with https://github.com/openstack/glance/blob/stable/diablo/glance/common/auth.py08:28
*** zigo has quit IRC08:31
*** zigo has joined #openstack-dev08:33
*** itooon has joined #openstack-dev08:44
*** darraghb has joined #openstack-dev08:58
*** jeffjapan has quit IRC08:59
*** hugokuo has quit IRC09:01
*** naehring has joined #openstack-dev09:10
*** Oneiroi_ has joined #openstack-dev09:13
*** Oneiroi_ has joined #openstack-dev09:13
*** zigo has quit IRC09:15
*** Oneiroi has quit IRC09:15
*** Oneiroi_ is now known as Oneiroi09:15
zykes-jaypipes: or any keystoner her ?09:19
*** reidrac has quit IRC09:20
*** mnour has joined #openstack-dev09:29
*** reidrac has joined #openstack-dev09:38
*** tsuzuki_ has quit IRC10:30
*** tryggvil_ has joined #openstack-dev10:37
*** tryggvil_ has quit IRC10:42
*** toluene has quit IRC10:44
openstackgerritMark McLoughlin proposed a change to openstack/nova: Start switching from gflags to optparse  https://review.openstack.org/90011:10
*** markvoelker has joined #openstack-dev11:11
zykes-why not use argparse instead ? thought that was newer11:17
superbobryyeah, optparse is deprecated11:22
*** toluene has joined #openstack-dev11:31
*** toluene has quit IRC11:39
ttxzykes-: note that a diablo-stable branch should be created soon with post-release fixes going in12:07
ttx(maintained by a stable updates team)12:07
DavieyHow can i search for a bug branch in gerrit?12:14
*** rods has joined #openstack-dev12:17
*** f-yang1 has joined #openstack-dev12:17
jeblairDaviey: https://review.openstack.org/#q,topic:bug/755916,n,z12:24
jeblairDaviey: https://review.openstack.org/#q,message:856564,n,z12:24
jeblairDaviey: based on feedback at ODS, soon we'll modify the bug hook to update bugs when a patch is proposed (not just when it's merged) so there will be a link in the bug as soon as something hits gerrit.12:25
*** hugokuo has joined #openstack-dev12:27
Davieyjeblair: awesome. thanks12:28
*** corrigac has quit IRC12:28
* Daviey screams.12:33
Davieybug 861440, Fix Committed.  Not by searching for the bug number it isn't.12:33
uvirtbotLaunchpad bug 861440 in nova "test_create_multiple_servers fails on master with venv" [Undecided,Fix committed] https://launchpad.net/bugs/86144012:33
*** cmagina has joined #openstack-dev12:35
*** rnorwood has quit IRC12:36
*** lts has joined #openstack-dev12:38
*** lts has joined #openstack-dev12:40
*** rkukura has joined #openstack-dev12:40
*** hugokuo has quit IRC12:58
*** lorin1 has joined #openstack-dev13:06
*** kbringard has joined #openstack-dev13:18
*** naehring has left #openstack-dev13:23
*** bcwaldon has joined #openstack-dev13:23
*** naehring has joined #openstack-dev13:24
*** markmc has joined #openstack-dev13:30
*** naehring has quit IRC13:31
openstackgerritVerification of a change to openstack/nova failed: Chain up to superclass tearDown in ServerActionsTest  https://review.openstack.org/89813:33
*** naehring has joined #openstack-dev13:33
*** mattray has joined #openstack-dev13:33
*** naehring has quit IRC13:33
*** superbobry has left #openstack-dev13:35
*** mdomsch has joined #openstack-dev13:38
*** markvoelker has quit IRC13:39
*** naehring has joined #openstack-dev13:42
*** naehring has quit IRC13:43
*** naehring has joined #openstack-dev13:43
openstackgerritVerification of a change to openstack/nova failed: Chain up to superclass tearDown in ServerActionsTest  https://review.openstack.org/89813:44
*** mattray has quit IRC13:47
*** chemikadze has quit IRC14:07
*** wwkeyboard has joined #openstack-dev14:07
openstackgerritA change was merged to openstack/keystone: Changes to support roles and services calls via extensions. Change-Id: I1316633b30c2be07353dacdffb321791a4e2e231  https://review.openstack.org/91114:10
*** cdub has joined #openstack-dev14:13
*** adjohn has joined #openstack-dev14:16
*** yogirackspace has joined #openstack-dev14:18
*** ewindisch has joined #openstack-dev14:20
ewindischhas anyone updated Vagrant and/or novascript for git?14:20
openstackgerritA change was merged to openstack/nova: Refactoring nova.tests.api.openstack.test_servers  https://review.openstack.org/90914:20
*** adjohn has quit IRC14:22
*** zaitcev has joined #openstack-dev14:25
openstackjenkinsProject nova-docs build #1,505: SUCCESS in 4 min 35 sec: https://jenkins.openstack.org/job/nova-docs/1505/14:26
openstackjenkinsbrian.waldon: Refactoring nova.tests.api.openstack.test_servers14:26
*** ewindisch has quit IRC14:27
*** vladimir3p has joined #openstack-dev14:34
*** nhubbard_away is now known as nhubbard14:35
openstackgerritBrian Waldon proposed a change to openstack/openstack-chef: Removing old glance.conf file  https://review.openstack.org/91914:38
*** dubsquared has joined #openstack-dev14:39
*** mdomsch has quit IRC14:40
*** code_franco has joined #openstack-dev14:40
*** naehring has quit IRC14:41
*** dragondm has joined #openstack-dev14:44
*** chemikadze has joined #openstack-dev14:45
openstackgerritA change was merged to openstack/glance: Adds SSL configuration params to the client  https://review.openstack.org/19014:48
*** rnorwood has joined #openstack-dev14:50
*** armaan has joined #openstack-dev14:57
*** armaan has quit IRC15:00
*** armaan has joined #openstack-dev15:01
*** cp16net has joined #openstack-dev15:05
*** ewindisch has joined #openstack-dev15:07
*** Oneiroi_ has joined #openstack-dev15:19
*** Oneiroi_ has joined #openstack-dev15:19
*** Oneiroi has quit IRC15:23
*** Oneiroi_ is now known as Oneiroi15:23
*** armaan has left #openstack-dev15:24
*** markvoelker has joined #openstack-dev15:26
openstackgerritAaron Lee proposed a change to openstack/openstack-integration-tests: Making run_test.sh python version and directory  agnostic  https://review.openstack.org/92015:27
notmynamevishy: why is the nova volume team listed as openstack-volume? the rest are nova-*15:27
*** misheska has joined #openstack-dev15:30
*** rnorwood has quit IRC15:32
notmynamettx: is there any way to delete blueprints?15:33
ttxnotmyname: I think the closest thing is to mark them superseded15:34
ttxnotmyname: or use it for your next creation, since you can change all fields15:35
openstackgerritA change was merged to openstack/keystone: Simplified README  https://review.openstack.org/90715:35
notmynameheh15:35
ttxI bet the non-deletion is a feature in someone's twisted mind15:35
zulmtaylor_: http://people.canonical.com/~chucks/libvirt15:35
*** rkukura has quit IRC15:37
*** rnorwood has joined #openstack-dev15:38
notmynamechmouel: are you working on https://bugs.launchpad.net/swift/+bug/869235?15:40
uvirtbotLaunchpad bug 869235 in swift "user permissions on directories" [Undecided,New]15:40
*** rnorwood has quit IRC15:43
*** ewindisch has quit IRC15:44
*** ewindisch has joined #openstack-dev15:44
*** rnorwood has joined #openstack-dev15:48
*** bencherian has joined #openstack-dev15:48
*** reed has joined #openstack-dev15:49
*** f-yang has quit IRC15:49
*** corrigac has joined #openstack-dev15:52
*** cp16net_ has joined #openstack-dev16:00
openstackgerritAaron Lee proposed a change to openstack/openstack-integration-tests: Fixing test_images.py to work with v1.1  https://review.openstack.org/92116:00
*** cp16net has quit IRC16:01
*** cp16net_ is now known as cp16net16:01
*** ohnoimdead has joined #openstack-dev16:02
*** mtaylor_ is now known as mtaylor16:03
*** mtaylor has joined #openstack-dev16:03
*** ChanServ sets mode: +v mtaylor16:03
*** reidrac has quit IRC16:08
lorin1How do you search for a string in a commit log in Gerrit?16:14
openstackgerritAlex Meade proposed a change to openstack/nova: Redirects requests from /v#.# to /v#.#/  https://review.openstack.org/88616:17
mtaylorin gerrit?16:19
mtaylorlorin1: go to its gitweb and search16:20
mtaylorlorin1: https://review.openstack.org/gitweb?p=openstack/nova.git;a=summary16:20
mtaylorlorin1: https://review.openstack.org/gitweb?p=openstack%2Fnova.git&a=search&h=HEAD&st=commit&s=some+text  ... for instance16:21
openstackgerritA change was merged to openstack/openstack-ci: Document gerrit ACL change for stable branches.  https://review.openstack.org/91016:22
lorin1mtaylor: thanks.16:24
lorin1mtaylor: Any chance a link to gitweb could be added to the Gerrit front page?  (Or is it already there?)16:24
mtaylorlorin1: not easily - it would be a patch to gerrit - but feel free to file a bug at http://bugs.launchpad.net/openstack-ci16:27
mtaylorlorin1: you can get to gitweb in gerrit from the gitweb link in any of the changes up for review - which is a gitweb link for the diff for that change16:27
*** dtroyer has joined #openstack-dev16:27
*** f-yang has joined #openstack-dev16:29
lorin1mtaylor: Bug filed.16:31
*** mnour has quit IRC16:31
*** hisaharu has joined #openstack-dev16:31
*** f-yang1 has left #openstack-dev16:32
*** cp16net has quit IRC16:32
*** cp16net has joined #openstack-dev16:33
*** novas0x2a|laptop has quit IRC16:33
*** nhubbard is now known as nhubbard_away16:35
*** vladimir3p has quit IRC16:37
*** heckj has joined #openstack-dev16:42
*** bencherian has quit IRC16:43
*** jdurgin has joined #openstack-dev16:43
*** rnorwood has quit IRC16:51
openstackgerritPaul Voccio proposed a change to openstack/keystone: Fixes LP844959, typo in Authors file  https://review.openstack.org/92216:56
openstackgerritAlex Meade proposed a change to openstack/nova: Redirects requests from /v#.# to /v#.#/  https://review.openstack.org/88616:57
zykes-jaypipes: around ? and yogirackspace ? maybe you two can anyswer on why glance and keystone doesn't cooperate?16:57
jaypipeszykes-: can you provide some more details than "doesn't cooperate"?16:58
zykes-i think glance doesn't send the correct credentials ? It's missing "auth" in the req obdy16:58
zykes-body16:59
*** rnorwood has joined #openstack-dev17:02
*** nickon has quit IRC17:05
rmkzykes-: You mean on the command line, right?17:07
*** patelna has joined #openstack-dev17:07
jaypipeszykes-: I think we covered this before... that means you are using a glance client package that is old.17:07
zykes-correct rmk17:07
*** rnorwood_ has joined #openstack-dev17:07
zykes-jaypipes: i tried with the newest trunk as well17:07
rmkzykes-: Yeah the diablo trunk does not work17:07
rmkzykes-: Need to pass an auth token via -A17:07
jaypipesVek: around? hoping you could help zykes- with Glance/Keystone.17:08
*** rnorwood has quit IRC17:08
openstackgerritJay Pipes proposed a change to openstack/glance: Overhauls the image cache to be truly optional  https://review.openstack.org/92317:08
rmkzykes-: The environment variables to pass credentials don't work17:08
*** Oneiroi has quit IRC17:09
rmkNor does specifying username/pass on the command line.  Only way is auth token.17:09
zykes-rmk: why doesn't it work ?17:11
*** adjohn has joined #openstack-dev17:14
jaypipeszykes-: what does env | grep OS_ show?17:14
*** lorin1 has quit IRC17:15
*** misheska has quit IRC17:16
*** adjohn has quit IRC17:17
*** adjohn has joined #openstack-dev17:17
zykes-OS_AUTH_ etc17:18
zykes-i can paste you it17:18
*** vladimir3p has joined #openstack-dev17:19
zykes-http://pastebin.com/PCx8vQnk17:21
*** blamar has quit IRC17:24
*** Oneiroi has joined #openstack-dev17:27
openstackgerritA change was merged to openstack/nova: Redirects requests from /v#.# to /v#.#/  https://review.openstack.org/88617:28
*** blamar has joined #openstack-dev17:28
*** susanne-balle__ has joined #openstack-dev17:28
zykes-jaypipes: ?17:30
*** blamar has quit IRC17:30
zykes-rmk: did you have experience with it ?17:30
*** blamar has joined #openstack-dev17:31
jaypipeszykes-: looks like those are correct... really unsure what is wrong. Vek, can you assist?17:31
zykes-vek's a glance dev or ?17:32
jaypipeszykes-: works on nova, keystone and glance. he did the auth work in glance, along with s1rp_17:33
openstackjenkinsProject nova-docs build #1,506: SUCCESS in 4 min 37 sec: https://jenkins.openstack.org/job/nova-docs/1506/17:36
openstackjenkinsalex.meade: Redirects requests from /v#.# to /v#.#/17:36
*** novas0x2a|laptop has joined #openstack-dev17:37
rmkzykes-: yes, actually several of us have run into this already17:40
rmkSomeone had a patch to fix it, I don't have a reference to it at the moment17:40
*** lorin1 has joined #openstack-dev17:43
openstackgerritJay Pipes proposed a change to openstack/glance: Fixes LP Bug#844618 - SQLAlchemy errors not logged  https://review.openstack.org/86317:43
*** rkukura has joined #openstack-dev17:43
*** mgius has joined #openstack-dev17:45
*** rnorwood_ has quit IRC17:52
*** darraghb has quit IRC17:53
zykes-that's bad rmk :(17:54
*** dtroyer has quit IRC17:55
zykes-rmk: anyway you can dig it up ?17:56
*** cp16net_ has joined #openstack-dev18:00
*** dolphm has joined #openstack-dev18:03
*** cp16net has quit IRC18:03
*** cp16net_ is now known as cp16net18:03
*** nickon has joined #openstack-dev18:03
*** susanne-balle__ has quit IRC18:04
*** rnorwood has joined #openstack-dev18:07
*** dtroyer has joined #openstack-dev18:09
rmkmtaylor: ping for keystone diablo packages? :)18:09
zykes-hmmms, how can i get the full stack i wonder to work together...18:24
*** med_out is now known as medberry18:24
zykes-noone that knows ?18:32
openstackgerritJosh Kearney proposed a change to openstack/nova: Only log instance actions once if instance action logging is enabled (now disabled by dfault).  https://review.openstack.org/92418:34
*** bencherian has joined #openstack-dev18:35
zykes-is there none that has setup the full stack of openstack ?18:42
*** nickon_ has joined #openstack-dev18:43
rmkzykes-: I have it working, what do you need to do with glance?18:44
zykes-rmk: get keystone working..18:45
*** nickon has quit IRC18:45
rmkzykes-: It works you just need to pass -A <token> with your glance commands18:45
openstackgerritJosh Kearney proposed a change to openstack/nova: Only log instance actions once if instance action logging is enabled (now disabled by dfault).  https://review.openstack.org/92418:45
zykes-rmk: yeah and that's a bit contra dictory of what the docs says :p18:45
rmkdocs are completely wrong in regards to all things keystone, sadly18:45
zykes-yeah18:45
rmkin the case of glance, it's simply because the code as released for diablo was broken for keystone auth18:46
zykes-but i mean even "new" versions of glance doesn't work with it ?18:46
rmkIt works in the sense that nova is able to pull images from glance18:46
rmkand you can interact with glance directly via the -A workaround18:46
rmkJust can't use username/password credentials right now as far as I can tell18:46
zykes-hmmm18:47
zykes-what would one need to do to get it patched ? :p18:47
rmkI haven't tried to be honest, you could attempt pointing at the glance trunk ppa and testing it out18:47
rmklet me see if I can dig up the patch18:47
rmkI haven't bothered patching it myself18:47
zykes-rmk: i've tried even with trunk from github, same problems18:47
zykes-;p18:47
rmkJust dealing with it at the moment18:47
zykes-do you have swift as well rmk ?18:49
rmkNegative18:50
zykes-doh18:50
rmkIt's difficult enough to get this combination going right now -- my threshold for pain hasn't allowed me to look at swift yet ;)18:50
*** nati2 has joined #openstack-dev18:50
zykes-heh18:50
zykes-mine hasn't allowed me to go post glance even18:50
rmkThe individual projects are fine really, it's once you start trying to get everything to play nice with keystone/dash that becomes troublesome18:51
zykes-exactly, and for a fully fledged demeo i want it to work ;p18:51
rmkyep18:51
rmkI hear you18:51
rmkWhy are you concerned about glance though?  It does work, just not with username/password auth.18:52
rmkTry passing it your auth token directly like I suggested.18:52
zykes-i just want it to work :)18:52
rmkWork with OPENSTACK environment variables?  Define the definition of work.18:53
zykes-correct ;P18:53
zykes-or blah, i'll just go ./care for now and carry on with other stuff instead18:53
rmkYeah not worth worrying about this since the workaround is fine really.18:54
rmkNot ideal but tolerable for now.18:54
zykes-which order should one do it really ? keystone > swift > glance > quantum > nova > dashboard ?18:54
dolphmhow long has devstack worked on rackspace's Cloud Servers?18:55
rmkI've only done keystone, glance, nova, dash -- no quantum or swift18:55
zykes-ok rmk i got swift going at least18:55
zykes-done nova from source or pkgs ?18:55
rmkdolphm: meant to say hi at the conference, I think you were sitting in front of me for the project update talk :)18:55
dolphmrmk: there were a few people i never got a chance to formally say hi to as well :)18:56
rmkAnd no idea regarding devstack18:56
*** nickon_ has quit IRC18:56
*** crobinso has joined #openstack-dev18:57
*** crobinso has quit IRC18:58
openstackgerritYogeshwar Srikrishnan proposed a change to openstack/keystone: Adding calls to get roles for user as per new format.Cleaning references to old code.  https://review.openstack.org/92518:58
zykes-wonder what redhat is using OpenStack for18:59
*** bencherian has quit IRC19:00
*** jakedahn has joined #openstack-dev19:00
*** donaldngo_hp has quit IRC19:00
*** mdomsch has joined #openstack-dev19:00
*** crobinso has joined #openstack-dev19:01
*** bencherian has joined #openstack-dev19:04
openstackgerritBrian Waldon proposed a change to openstack/nova: Convert instancetype.flavorid to string  https://review.openstack.org/85919:06
openstackgerritVerification of a change to openstack/nova failed: Refactoring image and server metadata api tests  https://review.openstack.org/91219:07
*** bencherian has quit IRC19:07
*** bencherian has joined #openstack-dev19:08
zykes-anyone running quantum yet ?19:16
openstackgerritA change was merged to openstack/nova: Refactoring image and server metadata api tests  https://review.openstack.org/91219:18
zykes-rmk: do oyu know what i need to get nova working with keystone ?19:18
zykes-dolphm: you around for a keystone question ?19:21
zykes-the admin_token in different .conf's does that need to belong to a tenant / user with "Admin" or "ServiceAdmin" ?19:21
dolphmzykes-: sure19:22
*** bcwaldon has quit IRC19:22
zykes-or who can answer that ;p19:22
dolphmthat should be a regular Admin19:23
zykes-what's the difference ?19:23
dolphm'Admin' is basically a *keystone* 'service admin'19:23
zykes-and serviceadmin si ?19:23
dolphmthat's a question for yogi19:24
zykes-h ok ;)19:24
openstackgerritAaron Lee proposed a change to openstack/openstack-integration-tests: Making run_test.sh python version and directory  agnostic  https://review.openstack.org/92019:24
openstackjenkinsProject nova-docs build #1,507: SUCCESS in 4 min 40 sec: https://jenkins.openstack.org/job/nova-docs/1507/19:26
openstackjenkinsbrian.waldon: Refactoring image and server metadata api tests19:26
*** Susanne-Balle has joined #openstack-dev19:27
*** Susanne-Balle has quit IRC19:29
*** Susanne-Balle has joined #openstack-dev19:30
*** dolphm has quit IRC19:40
*** jdg has joined #openstack-dev19:42
*** markmc has quit IRC19:47
*** crobinso has quit IRC19:47
*** markmc has joined #openstack-dev19:48
vishynotmyname: only because it was created first and not by me.  Also there is some potential of it not being part of nova at some point.19:57
notmynamevishy: the same potential as other teams?19:57
*** mancdaz has quit IRC19:58
vishynotmyname: the other teams aren't really things that would split out afaik19:58
vishynotmyname: i don't mind a rename if you think it is best19:59
vishynotmyname: I didn't create the group.19:59
notmynamevishy: just curious. I'm a little sensitive to things that imply openstack == nova ;-)19:59
notmynameI know19:59
*** mancdaz has joined #openstack-dev20:01
*** toluene has joined #openstack-dev20:04
*** mancdaz has quit IRC20:05
*** mancdaz has joined #openstack-dev20:08
*** mancdaz has quit IRC20:12
*** mancdaz has joined #openstack-dev20:13
*** novas0x2a|laptop has quit IRC20:15
*** kbader has joined #openstack-dev20:17
heckjmtaylor: what's the "not working" part with generic packages?20:17
rmkI missed the opening of this debate but it sounds like packaging which has been a near and dear concern.20:17
mtaylorheckj: so, in a generic package - how do you know if you're using vlan maanager or flatdhcp20:17
mtaylorheckj: or mysql vs. postgres20:18
mtaylorheckj: or kvm vs. xen20:18
openstackgerritA change was merged to openstack/keystone: Adding calls to get roles for user as per new format.Cleaning references to old code.  https://review.openstack.org/92520:18
mtaylorunless you start having a whole bunch of meta packages for every architectural issue20:18
mtayloralso, debian packaging is not built to understand cross-machien depends20:18
mtaylor(although juju is attempting to solve that)20:18
Daviey.. which was why juju ws created :)20:18
mtayloryes20:18
Davieygah!20:18
mtaylorand once it's solid and stuff, I find it very exciting20:19
mtaylorof  course, I can't use it, because it doesn't support rackspace cloud :)20:19
mtaylorso it's kinda hard for me to be a cheerleader yet20:19
heckjSo the packages that I used in this writeup: http://www.rhonabwy.com/wp/2011/09/23/installing-openstack-diablo-release-nova-and-glance/, they had an opinion about what to use: MySQL+KVM - I thought that was normal.20:20
zykes-what doesn't support rackspace yet ?20:20
sorenmtaylor: It's hardly impossible for a package (be it "generic" or not (Whatever that means)) to have a default setting.20:20
mtaylorzykes-: juju20:20
*** novas0x2a|laptop has joined #openstack-dev20:20
mtaylorheckj: it's not20:20
heckjOpenstack supports using other databases and Xen, but it's not packaged that way today - that seemed very reasonable to me.20:20
zykes-mtaylor: was ist das ?20:20
mtaylorheckj: that happens to be a choice that Ubuntu made for how they wanted standard UEC openstack deploys to work20:21
mtaylorheckj: but rackspace cloud servers, for instance, disagrees20:21
mtaylorheckj: so we lose the rackspace cloud servers guys and their deployement from our pool of potential package users20:21
mtaylorand that's fine, really20:21
Daviey"nova-compute-xen"20:21
Daviey(already does ship)20:21
zulheckj: yes it is..there is a package for xen20:22
heckjso the racker cloud guys have a sep. set of packages they're using?20:22
kbringardheckj: I'd probably consider changing the bind-address stuff for mysql to be on an internal only management network20:22
sorenYup.20:22
mtayloryes. I'm just trying to come up with an example of why apt-get install nova-compute doesn't produce a fully working system - anybody want to help me out here?20:22
kbringardbut for simplicity sake, that works :-)20:22
*** ohnoimdead has quit IRC20:23
*** clayg has quit IRC20:23
*** comstud has quit IRC20:23
*** cweidenkeller has quit IRC20:23
*** sleepsonthefloor has quit IRC20:23
*** bhall_ has quit IRC20:23
*** 31NAADY1U is now known as clayg20:23
sorenmtaylor: It should. If it doesn't, that's a bug and a regression. I certainly used to.20:23
zulagain...just make a package as generic/simple as possible and people can use that as a base20:23
mtayloressentially, if there were an openstack meta package that apt-get install openstack gave me a fully functional single-machine install that would respond to the API test suites, there would be no need for devstack20:23
heckjmtaylor: Wait - I buy that statement - I don't expect "apt-get install" to make a fully working system today, but I did expect it to lay down the files so that I can finish any config changes and tweaks (i.e. nova.conf, nova-manage commands) to make it a working system20:23
sorenmtaylor: I can make one for you in an hour.20:23
mtaylorheckj: yes. I think that's reasonable20:23
sorenmtaylor: Installing all of nova-{scheduler,api,compute,objectstore} used to get you a working system.20:24
sorenER... nova-network, too.20:24
heckjmtaylor: I might have the "configure from" package script already made actually20:24
*** clayg is now known as 31NAADY1U20:24
*** ohnoimdead has joined #openstack-dev20:24
*** clayg has joined #openstack-dev20:24
*** comstud has joined #openstack-dev20:24
*** cweidenkeller has joined #openstack-dev20:24
*** sleepsonthefloor has joined #openstack-dev20:24
*** bhall_ has joined #openstack-dev20:24
*** hubbard.freenode.net sets mode: +v comstud20:24
ttxmtaylor: from what you're saying, we are dumping a tightly integrated package-driven continuous integration framework just because we don't have a good metapackage20:24
kbringardsoren: of course you'd have to configure stuff with nova-manage for your specific network setup, etc20:25
mtaylorttx: no. sorry20:25
mtaylorI was trying to answer heckj's question about why installing from packages doesn't get you a fully functional openstack system20:25
heckjmtaylor: https://gist.github.com/090c416a59596512e2de20:25
mtaylorwhich may have digressed20:25
sorenkbringard: If devstack can make reasonable default choices, so can the packages.20:25
kbringardtouche soren20:26
mtaylordevstack solves a problem of "I have this code and I want to test to see if it works"20:26
mtaylorthere are other possible ways to solve that problem20:26
openstackgerritsleepsonthefloor proposed a change to openstack/keystone: Fix issues in the ec2 middleware  https://review.openstack.org/92620:26
mtaylorthusfar, none of them have existed in an easily consumable manner20:26
rmkEliminating packages means I have to do a lot more than just configure openstack.  Now I  need to figure out how to actually get the proper bits and dependancies for those bits.20:26
mtaylorobviously the same problem can be resolved in any one of 100 different ways20:26
ttxmtaylor: I though it solved I have time on my hands and I'm better at shell scripts than at debian packaging or puppet recipes20:27
heckjmtaylor: we're doing this internally now - I can make a quick update and see if we can include it in devstack.org if that would make your life easier.20:27
mtaylorttx: we have never had a consistent openstack install that works that wasn't smokestack and driven by ruby on rails and a bunch of chef20:27
Davieyheckj: internally where?20:27
mtaylorheckj: sorry man - don't want to lead you astray20:27
heckjDaviey: Nebula20:27
mtaylorheckj: we're not having a problem with devstack ... although definitely send patches20:28
ttxback to the maintenance issue I guess -- whoever commits to maintain stuff wins. Choosing something maintainable should help, though20:28
heckjmtaylor: already doing that. What I don't have is a "good place" to put this little shell script I just pasted into gist20:28
mtaylorheckj: the problem we're talking about here is "why are we going to use devstack to drive CI instead of just using packages"20:28
openstackgerritChris Behrens proposed a change to openstack/nova: Fix EC2 test_cloud timing issues  https://review.openstack.org/92720:29
mtaylorheckj: I think that script and devstack are very similar20:29
heckjmtaylor: I get wanting to use a CI system from source, which is awesome, but I think we would be MUCH better off as a project building from packages20:29
mtaylorheckj: so, that's a whole other thing20:29
Davieyis it?20:29
mtayloryes20:29
mtaylorbecause as a package we do not release packages20:29
heckjmtaylor: They are - that little gist is devstack with packages and it's defaults rather than source installs. Same intention - create a quick, runnable all-in-one instance20:29
mtaylorbecause we decided as a project that we do not want to maintain them20:30
jeblair"baause as a projcet" ?20:30
jeblairwow.  lame correction.  "because as a project"?20:30
ttxmtaylor: the syable-team might disagree with that20:30
ttxstable20:30
ttx"as a project"20:30
heckjmtaylor: I kinda disagree with that - but maybe I missed the conversation.20:30
mtaylorttx: no. they are collecting packages, not making releases20:30
Davieymtaylor: 'released packages' != 'development packages'20:30
mtaylorcollecting patches20:30
mtaylorDaviey: what is the point of dev packages if they do not lead to release packages?20:30
mtaylorheckj: I actually disagree with it too20:31
Davieylong term, they no doubt will be.20:31
mtaylorheckj: but it's what was decided20:31
openstackgerritAaron Lee proposed a change to openstack/openstack-integration-tests: v1.1 support, again  https://review.openstack.org/92820:31
ttxmtaylor: we said at the summit that we could release point releases based on their work though20:31
Davieyprobably the basis for Ubuntu and Debian packags.20:31
mtaylorttx: could. not would20:31
mtaylorttx: and there is no policy for doing so20:31
Davieyper commit packages make no sense to be 'released'20:31
heckjmtaylor: I'm honestly not trying to be a dick, but when was it decided and can we revisit that decision, because I don't think it's a good one.20:31
mtaylorheckj: the stable release updates session20:31
Davieyoh rly?20:32
DavieyThat was the session i drove.20:32
*** Oneiroi has quit IRC20:32
*** bcwaldon has joined #openstack-dev20:32
mtaylorDaviey: do you not agree that the SRU session decided that we would create a repo that distros could use to collaborate on SRU patches, but that the project made no comittment to actually every releasing anything from that work?20:32
ttxheckj: basically openstack is not a downstream distro of openstack20:32
Davieymtaylor: yes, erm no.. yes.. I'm not sure how to answer that20:33
mtaylorand that it was decided that openstack was not in the business of long term mainenance of its releases, but that the distos had a vested interest in doing so and would thus shoulder the burden?20:33
Daviey"do you not agree" confuses me, i'm tired.20:33
mtaylorDaviey: sorry man20:33
mtaylorDaviey: do you agree with the sentences I just wrote20:33
ttxheckj: maintaining updates over the long run is costly, and something that distributions actually already do, and do well20:33
Davieymtaylor: i believe so, yes.20:33
jeblairthat's what i heard too; i thought that was one of the more clear and explicit sessions as far as outcomes go20:34
ttxheckj: so during that session we said that we should have a stable branch that help distros doing that job20:34
mtaylorDaviey: ok. so that's what I meant by "it was decided at the SRU session" ... based on that, we just decided a few minutes ago that releasing release packages as the project was irresponsible because there was no ongoing support mechanism20:34
mtaylorBUT20:34
mtaylorthat distros do an excellent job of doing such things20:35
mtaylorso I could expect shiny and wonderful packages in oneiric from ubuntu that would be maintained by ubuntu20:35
heckjttx: I think as a project we're getting incredible value out of having latest/up-to-date packages. I don't think we need to take on the long update sequence, but to drop the latest packages would be a hurt to the project in general.20:35
mtaylorheckj: sorry - I'm going to have to ask that you be slightly more specific20:35
claygany xenapi users on now?  _cerberus_, s1rp_ ?20:35
Davieymtaylor: yes... but CI'ing via packages != releasing them20:35
mtaylor"lastest packages" is WAY to vague in an already confusing conversation20:35
openstackgerritBrad Hall proposed a change to openstack/quantum: Swich over to update_{net,port} instead of rename_net and set_port_state  https://review.openstack.org/92920:36
mtaylorDaviey: so, part of CI'ing code is about CI'ing the changes that get sent in20:36
mtaylorDaviey: I should say, the important part20:36
claygI've been using KVM, but I'm trying out Xen - seeing some weird issues on the compute worker - httplib CannotSendRequest Exceptions?20:36
heckjmtaylor: "latest packags" means (to me) whatever the current release of ubuntu or centos/redhat has as a pre-set default package based install.20:36
mtaylorI'm concerned with testing code that the project owns20:36
mtaylorIf I use devstack to run CI tests, both the CI driving script and the code itself are all managed by our CI system20:36
heckjmtaylor: I don't want to assert we'll maintain anything other than the latest release, but that's probably where you or others would rationally assert that as being irresponsible20:37
mtaylorso I'm not testing something external that could break something outside of the control of the devs of the project20:37
ttxheckj: I agree there is value in providing 0-day packages for trunk and milestones and releases.20:37
claygit feels/looks like the XenAPI session object is being shared between greenthreads and the xmlrpc/http stuff is getting interleaved... that can't be right?20:37
ttxheckj: I just know nobody wants to work on stable updates, everyone is after the latest crack (at least in Nova)20:38
rmkttx: Which is horrible imo.20:38
jeblairyes, i wish it were not so20:38
rmkThere's no single place I can go to get the latest version of Diablo with critical patches/fixes.20:38
ttxheckj: so committing "the project" to deliver credible and usable stable updates is not doable, imho20:38
mtaylorttx: ++20:38
rmkI was actually shocked to find out there is no stable release process.20:38
ttxespecially if you no longer want to special-case one distro20:39
jeblairi hope that the stable branch may change that.20:39
sorenmtaylor: I agree that using something that is managed by the project to do the CI stuff is valuable. I just also think that using something that someone will actually use to deploy this stuff is valuable, too. Why can't we find something that meets both criteria?20:39
ttxjeblair: maybe the stable branch will be so successful they become "openstack" in the minds and hearts of everyone20:39
rmkThat's the other issue with relying on the distros.  It forces people to use the specific distro shipping the specific version of openstack which is desired.20:39
jeblairttx: that would make me happy. :)20:39
*** cp16net has quit IRC20:40
ttxrmk: and I was shocked to find out that nobody wanted to work on that.20:40
rmkttx: wow20:40
mtaylorsoren: I am open to discussions about the actual thing, but devstack exists now, and I need to start running integration tests.20:40
rmkI mean let me ask a simple question, what are people running in production environments if there are no stable point releases?20:40
sorenLast I checked, packages also exist.20:40
mtaylorsoren: and I think "something that someone will actually use to deploy" is what we spun wheels on for the last 6 months20:41
ttxrmk: a distribution. Most of the time, their own.20:41
openstackgerritMark McLoughlin proposed a change to openstack/nova: Point tools/rfc.sh at the right branch  https://review.openstack.org/93020:41
rmkttx: That's the conclusion I arrived at as well for my company.20:41
sorenmtaylor: Only because whenever someone suggests something to use, the response is "no, because someone uses something else".20:41
rmkReally unfortunate.20:41
ttxrmk: distributions are where people that care about updates and stability and bugfixes live20:41
heckjrmk: we're running from the "release" packages (0-day packages) that have been created : https://gist.github.com/090c416a59596512e2de20:42
ttxrmk: I'm not sure I see the value in duplicating their work20:42
heckjrmk: that's missing the "add-apt-repo pap:nova-core/release2011.3" in front of it20:42
rmkttx: It's not really duplication of work.  How else would someone get Diablo on LTS for example?20:42
rmkheckj: That's great if you don't need anything in Diablo fixed at all.20:42
ttxrmk: yes, that's the loophole :)20:43
rmkheckj: Which, right now, means you're not using keystone or block migrations.20:43
rmkBecause neither of those work in the final diablo packages.20:43
rmkAmong other issues.20:43
heckjrmk: correct20:43
*** rbasak has quit IRC20:43
ttxrmk: in the discussion in #openstack-meeting we said some startup whould make a distro of openstack for lucid20:43
ttxshould*20:44
rmkttx: Also, I don't necessarily want to upgrade my entire distro because a newer openstack is out.20:44
heckjttx:20:44
rmkCertainly not across large hypervisor/compute pools.20:44
heckjttx: I don't think that's a terribly rational thought20:44
heckjttx: I'd see Canonical doing it, or Citrix, but less likely a startup20:44
rmkI can tolerate the downtime associated with package updates but doing a full on distro update every 6 months to be able to run the latest openstack is just not sustainable.20:44
*** bencherian has quit IRC20:45
ttxheckj, rmk: If someone in this project cares enough to maintain those packages in a branch / PPA / whatever and that becomes popular, it's still "the project"20:45
ttxBasically, it's not a decision that was made -- it's just that it's not happening20:45
rmkI'd take it on myself but I've not been a user long enough to do the work right at this point.20:46
ttxthis is open source - if you care, do it20:46
ttxI know that *I* personally don't have the time to pull it off, in addition to just releasing plain tarballs20:46
ttxso I feel uncomfortable saying "we as a project will do it"20:47
mtaylorand that's the thing ... I think that number one priority for us to release solid plain source tarballs20:47
ttxsaying "someone will do it" doesn't work. Doing it and then saying it exists, works20:48
jeblairthat's feedback we got from distros at ODS: openstack should take care of being a good upstream and let distros make packages.20:48
*** code_franco has quit IRC20:48
mtaylorand that is what we're trying to accomplish20:48
ttxjeblair: apparently there is a need for a lucid stable distro of openstack. A need that could be addressed inside or outside the project20:48
ttxby people who care20:48
ttxand have time20:49
jeblairttx: yes20:49
mtaylorso if we're trying to produce solid source tarballs, we should ensure that we test the code that is in those tarballs20:49
mtaylorother stuff around that is distracting from the thing we're trying to produce20:49
ttxmtaylor: which is different from the branches themselves20:49
ttx(digressing)20:49
heckjjeblair: I hear you, just disgruntled about it - but I suppose if Ubuntu packages from the tar balls, I don't really have anything to bitch about.20:50
mtaylorttx: well yeah. don't get me started on setuptools not having a distcheck equiv :)20:50
mtaylorheckj: in theory, it should be completely possible to take the tarballs we cut and apply ubuntu's packaging to them20:50
mtaylorheckj: in practice, you are going to run in to the dh_python2 hell20:50
mtaylorheckj: and when you do, please send an angry email to debian for being shortsighted in their managnement of that transition :)20:50
rmkYou know what, at this point if dropping distro packages means we have a chance at getting "stable" tarballs, then great.20:51
* mtaylor digresses20:51
rmkIf it means neither than I am sad.20:51
mtaylorrmk: if it means neither I doubt I will have a job in six months20:51
openstackgerritYogeshwar Srikrishnan proposed a change to openstack/keystone: Moving non core users and tenants calls to appropriate extensions.  https://review.openstack.org/93120:51
heckjmtaylor: thanks for taking the time to walk me through it20:51
ttxheckj: I agree we need to discuss all of this more. Jumping to conclusions might not be the best way to handle this20:52
mtaylorheckj: any time! it's a 'fun' situation, because the line between upstream/downstream can get blurry when we're all friends20:52
ttxas is the line between openstack and... not openstack20:52
openstackgerritJason Cannavale proposed a change to openstack/keystone: bug lp:865448 change abspath to dirname in controllers/version.py to correct path problems.  https://review.openstack.org/93220:52
heckjYeah, I don't know anything about dh_python2, time to do some learning.20:52
mtaylorheckj: I can rant on it for a while - it's a _good_ change20:54
mtaylorheckj: but the way it was done makes it hard to do no-change backports of python packages from oneiric to pre-oneiric20:54
ttxmtaylor: how desperate is the current situation ? Can we survive until UDS and get into brainstorming mode there ?20:54
heckjmtaylor: I take it that it screws backward compatibility though20:54
openstackgerritA change was merged to openstack/nova: Point tools/rfc.sh at the right branch  https://review.openstack.org/93020:54
heckjmtaylor: yeah20:54
mtaylorttx: about what?20:55
mtaylorttx: or about which thing?20:55
heckjtax, mtaylor: I'm heading to UDS, but I don't know all that much about packaging. More than happy to learn more and get into though20:55
heckjttx: ^^^ (can20:55
heckj't20:55
mtaylorheckj: awesome! excited that you'll be there20:55
heckj (can't' type)20:55
openstackgerritA change was merged to openstack/keystone: Moving non core users and tenants calls to appropriate extensions.  https://review.openstack.org/93120:55
ttxmtaylor: the decision on what to produce at which point, and what the CI will use for trunk gating20:56
*** dolphm has joined #openstack-dev20:56
*** dolphm has quit IRC20:57
markmcjeblair, hey; any way you can allow me propose commits which I'm not the author of to stable/diablo ?20:57
markmcremote: ERROR:  In commit 1c8fedc82c4b388b0a3a876a54c7f14d3c2a834220:57
markmcremote: ERROR:  author email address vishvananda@gmail.com20:57
markmcremote: ERROR:  does not match your user account.20:57
ttxmeeting starting in 2 min in #openstack-meeting20:57
jeblairmarcmc: absolutely, i should have done that already.  i will do so for the whole openstack-stable-maint group20:58
mtaylorttx: at the moment, I have two mandates20:58
*** bencherian has joined #openstack-dev20:58
Davieymarkmc: You are the proposing author, right?20:58
markmcjeblair, thanks20:58
Davieyah, if it's easy to do - sounds good.20:58
mtaylorttx: a) get some form of minimal API-level functional testing in as a part of the trunk gating as soon as possible20:58
markmcDaviey, for this patch, vishy is the author and I'm the committer - gerrit rejects it atm20:59
mtaylorttx: b) whatever is used for trunk gating needs to be reproducible by devs so that they can debug problems20:59
Davieymarkmc: ah!20:59
*** nati2 has quit IRC21:00
mtaylorttx: in order to implement that at the moment, we will use devstack, because it allows us to get something done very soon to meet (a)21:00
ttxmeeting starts now in #openstack-meeting21:00
mtaylorttx: obviously, it's an open source project, and anything can change at any point in time - so we can ALSo continue to have conversations21:01
jeblairmarkmc: can you try again now?21:01
*** nati2 has joined #openstack-dev21:01
openstackgerritMark McLoughlin proposed a change to openstack/nova: Makes sure to recreate gateway for moved ip  https://review.openstack.org/93321:01
*** lorin1 has quit IRC21:01
openstackgerritMark McLoughlin proposed a change to openstack/nova: put fully qualified domain name in local-hostname  https://review.openstack.org/93421:01
openstackgerritMark McLoughlin proposed a change to openstack/nova: Fix the grantee group loading for source groups  https://review.openstack.org/93521:01
openstackgerritMark McLoughlin proposed a change to openstack/nova: Call endheaders when auth_token is None. Fixes bug 856721.  https://review.openstack.org/93621:01
uvirtbotLaunchpad bug 856721 in nova "image snapshot creation fails" [High,Fix committed] https://launchpad.net/bugs/85672121:01
openstackgerritMark McLoughlin proposed a change to openstack/nova: Removed db_pool complexities from nova.db.sqlalchemy.session. Fixes bug 838581.  https://review.openstack.org/93721:01
uvirtbotLaunchpad bug 838581 in nova "Failures in db_pool code: 'NoneType' object has no attribute '_keymap' or not returning rows." [High,Fix committed] https://launchpad.net/bugs/83858121:01
markmcjeblair, awesomeness, thank you21:01
markmcDaviey, ^^^ some more for you to poke through21:01
*** troytoman-away is now known as troytoman21:01
jeblairmarkmc: no problem, sorry for the inconvenience21:02
mtaylorttx: and if we can agree and what and why, then we can move forward with updates/changes to what the CI system is using21:02
markmcDaviey, I'll push them in small batches21:02
mtaylorttx: it is, of course, an orthogonal issue to publishing dev ppas, which we can also continue to discuss21:02
ttxack21:02
markmcDaviey, if I'm just cherry-picking patches, perhaps it's fine for you to +2 ?21:03
markmcDaviey, anything other than a cherry-pick, we'd want 2 real reviews21:03
openstackgerritA change was merged to openstack/nova: Fix EC2 test_cloud timing issues  https://review.openstack.org/92721:03
*** nati2 has quit IRC21:05
openstackjenkinsProject nova-docs build #1,508: SUCCESS in 4 min 47 sec: https://jenkins.openstack.org/job/nova-docs/1508/21:11
openstackjenkinsChris Behrens: Fix EC2 test_cloud timing issues21:11
beekhofsandywalsh: followup question from yesterday… if you've got multiple schedulers per zone, how are you making sure 2 or more requests dont try to take the the last available slot on a given compute node?21:15
*** jdg has quit IRC21:16
sandywalshbeekhof, right now you don't. But we're working on adding a reservation mechanism during the scheduling process.21:16
sandywalshbeekhof, so currently, we're over estimating and assuming some will fail21:17
beekhofahhhh21:17
beekhofbtw. i have most of a basic scheduler going21:17
sandywalshbeekhof, but that'll have to change and we'll reserve with a timeout21:17
sandywalshnice!21:17
sandywalshlooking forward to it21:18
beekhofhttp://fpaste.org/iipN/21:18
beekhofshould give you a rough idea21:18
sandywalshcool ... I like it. I think the idea of a hierarchy of schedulers has merit.21:19
sandywalshcan you have a bucket of buckets?21:20
beekhofsure21:20
sandywalshcool ... otherwise it's the same problem: pick a zone and by the time you pick a host, the landscape has changed.21:21
*** lts has quit IRC21:21
sandywalshbeekhof, gotta run ... but keep me informed!21:21
beekhofwill do21:21
beekhofi'll push this up to github shortly21:21
openstackgerritFlorian Hines proposed a change to openstack/swift: Add documentation for Swift Recon.  https://review.openstack.org/93821:25
*** markvoelker has quit IRC21:25
zykes-rmk: you got a clue on what i need for nova to work ?21:26
zykes-with keystone21:26
zykes-vishy: could i run a nova controller node as a dedicated node ?21:33
*** cmagina_ has joined #openstack-dev21:33
*** kbader has quit IRC21:33
*** cmagina has quit IRC21:34
zykes-any Quantum leapers around ?21:34
*** ohnoimdead has quit IRC21:35
*** ohnoimdead has joined #openstack-dev21:35
*** mikeyp-3 has quit IRC21:39
*** kbader has joined #openstack-dev21:44
*** Susanne-Balle has quit IRC21:44
rmkzykes-: yes21:47
rmkzykes-: you need the appropriate api-paste and to install the keystone middleware components21:47
rmkWhich, right now, I'm doing in such a disgusting way that I feel dirty just talking about it21:48
markmcbcwaldon, could you prod https://review.openstack.org/#change,898 again now that the test is fixed?21:49
zykes-;p21:49
bcwaldonmarkmc: yessir21:49
markmcbcwaldon, thankie muchely21:49
*** devcamcar has quit IRC21:49
*** sleepsonthefloor has quit IRC21:49
*** xtoddx has quit IRC21:49
*** vishy has quit IRC21:49
rmkSomeone killed znc.  Oops.21:50
zykes-:o21:51
rmkzykes-: So, best way really is to grab keystone from git, use the stable/diablo branch and setup.py install it21:51
zykes-ok : )21:51
rmkThat should place everything in the correct places for nova and glance to see them21:51
openstackgerritVerification of a change to openstack/keystone failed: Fixing issue #843226.  https://review.openstack.org/89521:52
zykes-heh, i'm trying to draft up my idea of starting a dns as a service thing21:53
* markmc scratches his head over https://review.openstack.org/#change,90021:54
*** rkukura has quit IRC21:54
markmcsmokestack failure looks python 2.6 specific21:54
markmcwhy are we supporting 2.6 again?21:55
markmcand how do I get hold of the exact version smokestack is using?21:55
markmc(without installing Ubuntu :-)21:55
* markmc doesn't see anything obvious in cpython hg history which might cause this21:56
markmcyet, it was reported against 2.6 http://bugs.python.org/issue956321:56
wwkeyboardIs anyone here using pythonbrew for testing with different versions of python?21:56
f-yangIs Sandy there for question?21:56
wwkeyboardI'm not sure what the best practice is for switching versions.21:56
* markmc looks up pythonbrew21:56
markmcwwkeyboard, thanks for the tip21:57
f-yangSandywalsh: question?21:57
*** bencherian has quit IRC21:58
zykes-rmk: have you tested out quantum yet ?21:59
rmknope22:01
*** kbringard has quit IRC22:01
rmkquantum doesn't have feature parity with nova-network at this point so its use cases are limited22:01
openstackgerritA change was merged to openstack/nova: Chain up to superclass tearDown in ServerActionsTest  https://review.openstack.org/89822:05
openstackgerritAaron Lee proposed a change to openstack/openstack-integration-tests: test supporting API v1.1  https://review.openstack.org/93922:07
openstackgerritYogeshwar 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/89622:09
openstackjenkinsProject nova-docs build #1,509: SUCCESS in 4 min 52 sec: https://jenkins.openstack.org/job/nova-docs/1509/22:11
openstackjenkinsMark McLoughlin: Chain up to superclass tearDown in ServerActionsTest22:11
openstackgerritA change was merged to openstack/keystone: Resubmitting change. Fixing issue #843226. Changes to throw appropriate faults during token validation.  https://review.openstack.org/89622:11
openstackgerritJohannes Erdfelt proposed a change to openstack/nova: Don't leak exceptions out to users  https://review.openstack.org/94022:13
*** vishy has joined #openstack-dev22:13
vishyyay, finally found my way around stupid hotel firewall22:13
zykes-:p22:14
*** mdomsch has quit IRC22:22
vishyit blocks ssl traffic on non port 443, so lame22:22
*** nati2 has joined #openstack-dev22:27
*** yogirackspace has left #openstack-dev22:31
*** xtor_ has joined #openstack-dev22:32
*** patelna has quit IRC22:32
rmkhttp://paste.openstack.org/show/2763/22:33
rmklooks like a bug yeah?22:33
*** extor has quit IRC22:35
*** jwhonce has joined #openstack-dev22:36
*** heckj has quit IRC22:39
openstackgerritBrian Waldon proposed a change to openstack/glance: Fixing functional tests that require keystone  https://review.openstack.org/94122:40
zykes-does anyone have some slides regarding swift for infrastructure / architectural presentations ?22:41
vishycomstud: just saw those ec2 test fixes22:43
vishycomstud: very nice22:43
*** rnorwood has quit IRC22:53
*** ewindisch has quit IRC22:58
*** ewindisch has joined #openstack-dev22:59
rmkI've noticed something all across Nova.  When any "job" fails, the consumer of said job doesn't report back what happened.23:05
rmkSeems like there should be a results queue, which the scheduler can figure out how to handle at some point.23:05
rmkWhether it's retrying the job, notifying someone, etc.23:06
openstackgerritRick Harris proposed a change to openstack/nova: Repartition and resize disk when marked as managed  https://review.openstack.org/90623:07
*** dtroyer has quit IRC23:07
*** kbader has quit IRC23:11
*** dubsquared has quit IRC23:13
*** reed has quit IRC23:17
*** wwkeyboard has left #openstack-dev23:22
*** bcwaldon has quit IRC23:22
*** xtor_ has quit IRC23:26
*** joonwon has joined #openstack-dev23:28
*** troytoman is now known as troytoman-away23:29
openstackgerritA change was merged to openstack/nova: Don't leak exceptions out to users  https://review.openstack.org/94023:32
*** jeffjapan has joined #openstack-dev23:34
*** dragondm has quit IRC23:36
comstudvishy: Yep.. thanks :)23:39
vishyrmk: yes the main issue is we don't have a concept of "job"23:39
comstudI resurrected an old unmerged branch where I'd started removing sleeps from tests.. when I realized it would fix these random test failures23:40
rmkvishy: I'd say it's probably one of the most significant architectural concerns I have with nova right now23:40
*** ohnoimdead has quit IRC23:41
vishythe orchestration team is supposed to be figuring this out and solving it23:41
vishyrmk: maybe you should join it?23:41
rmkSure.  I'm not a proficient python coder but I've certainly got architectural thoughts on this.23:41
comstudrmk: There's somewhat of a blueprint for that.. although I think it'll be replaced by what the orchestration team comes up with23:42
comstudor modified/enhanced23:42
vishyhttps://launchpad.net/~nova-orchestration23:43
*** ohnoimdead has joined #openstack-dev23:44
rmkGreat, joined.  Thanks.23:44
*** misheska has joined #openstack-dev23:44
openstackjenkinsProject nova-docs build #1,510: SUCCESS in 8 min 12 sec: https://jenkins.openstack.org/job/nova-docs/1510/23:44
openstackjenkinsJohannes Erdfelt: Don't leak exceptions out to users23:44
*** misheska has quit IRC23:45
*** ohnoimdead has quit IRC23:45
*** ohnoimdead has joined #openstack-dev23:45
*** misheska has joined #openstack-dev23:45
*** misheska has quit IRC23:52
*** misheska has joined #openstack-dev23:55
*** novas0x2a|laptop has quit IRC23:57
*** novas0x2a|laptop has joined #openstack-dev23:58

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!