Monday, 2011-09-19

« Sunday, 2011-09-18 Index Tuesday, 2011-09-20 »
*** nati has joined #openstack-dev00:04
*** hbetts has joined #openstack-dev00:18
*** nati has quit IRC00:21
*** nati has joined #openstack-dev00:33
*** RobertLaptop has joined #openstack-dev01:00
*** mandela123 has joined #openstack-dev01:40
*** relateable has quit IRC01:56
*** mandela123_ has joined #openstack-dev02:00
*** mandela123 has quit IRC02:01
*** mandela123_ is now known as mandela12302:01
mandela123fd02:02
*** tryggvil_ has quit IRC02:16
*** nati has quit IRC02:18
*** novas0x2a|laptop has quit IRC02:36
*** dragondm has joined #openstack-dev02:37
*** nati has joined #openstack-dev03:31
rmkCan anyone provide a bit of detail as to what the credentials and services tables are for with Keystone?03:58
*** Rajaram has joined #openstack-dev04:09
*** nati has quit IRC04:15
rmkSome odd stuff with the Dashboard.  I think in a lot of places where the "name" field is intended, "id" is being used.04:15
*** nati has joined #openstack-dev04:33
*** nati has quit IRC05:10
*** nati has joined #openstack-dev05:25
*** HugoKuo has joined #openstack-dev05:47
*** hbetts has quit IRC05:52
*** HugoKuo_ has joined #openstack-dev05:53
*** HugoKuo has quit IRC05:57
*** dragondm has quit IRC06:09
rmkCan anyone help with this nova-api/dashboard error?  http://pastebin.com/kVwLBSyg06:31
*** aryan has quit IRC06:44
vishyrmk: that is because id/name was just added07:26
vishyrmk: it used to be just id07:26
rmkah07:27
*** winston-d has joined #openstack-dev07:28
rmkvishy: dont suppose you have thoughts on what's causing the dashboard error I just pasted?  It's exactly what's described here (https://lists.launchpad.net/openstack/msg04002.html) but I already have the mentioned fix.07:29
*** nati has quit IRC07:29
vishyrmk: not sure about that one07:30
winston-dhi gurus, i'd like to know if there's OSAPI to manipulate with volumes. such as: create/delete/list_volume.07:30
rmkvishy: ok, thanks for looking07:30
vishywinston-d: there is an extension07:32
winston-dvishy : where can i find it?07:32
*** nati has joined #openstack-dev07:33
*** reidrac has joined #openstack-dev07:38
*** AhmedSoliman has joined #openstack-dev07:39
vishywinston-d: nova/api/openstack/contrib/volumes.py07:43
winston-dvishy : thx. i'll take a look07:45
*** chemikadze has joined #openstack-dev07:53
*** nati has quit IRC07:58
*** mandela123_ has joined #openstack-dev09:15
*** mandela123 has quit IRC09:15
*** mandela123_ is now known as mandela12309:15
*** jeffjapan has quit IRC09:25
*** reidrac has quit IRC09:25
*** reidrac has joined #openstack-dev09:25
*** reidrac has quit IRC09:35
*** reidrac has joined #openstack-dev09:37
*** joonwon has joined #openstack-dev09:38
*** Vek has quit IRC09:45
vishymtaylor, ttx, jeblair: or anyone that feels like messing with the packaging09:52
vishyhttps://code.launchpad.net/~vishvananda/nova/ubuntu/+merge/7587509:52
Davieyvishy: must be late in the day for you there?10:00
*** joonwon has left #openstack-dev10:05
*** AhmedSoliman has quit IRC10:45
*** AhmedSoliman has joined #openstack-dev10:45
*** lts has joined #openstack-dev10:59
sorenDaviey: Also remember that mergeing stuff into that branch also gets it built that way in our PPA for lucid/maverick/natty.11:10
sorenDaviey: So it's not just a problem that it's not yet in Oneiric.11:10
Davieysoren: I assumed that it would be carried in the openstack PPA?11:16
sorenIt would.11:16
sorenThat's my point.11:16
sorenSo the dnsmasq needs to also be updated in the PPA.11:17
*** comstud has quit IRC11:20
*** comstud has joined #openstack-dev11:21
*** openstack` has joined #openstack-dev11:28
*** openstack has quit IRC11:28
vishyDaviey: nope, I'm in paris11:29
*** openstack` is now known as openstack11:29
vishysoren: it is updated in the ppa11:30
vishysoren: mtaylor put it in.  He said it was accepted into ubuntu as well, but perhaps it didn't go int oneric?11:30
vishyDaviey: ^^11:30
Davieysoren: I can haz no access to PPA.11:33
*** bsza has joined #openstack-dev11:33
Davieyvishy: it's not in Ubuntu at all.11:33
Davieymtaylor was telling fibs. :)11:33
vishyhe's such a fibber11:34
*** markvoelker has quit IRC11:34
vishywelll it is in the ppa here: https://launchpad.net/~nova-core/+archive/trunk11:35
vishy  * Backported for OpenStack for oneiric11:35
vishy -- Monty Taylor <mordred@inaugust.com>   Fri, 16 Sep 2011 11:00:26 -040011:35
vishyare you sure it isn't anywhere?11:35
vishyDaviey: ^^11:36
Davieyvishy: it's not in /ubuntu/, looks like it is in the nova PPA.11:39
Davieyvishy: I'm trying to get it in Ubuntu Oneiric today.11:39
*** sandywalsh has quit IRC11:41
vishycool11:42
*** markvoelker has joined #openstack-dev11:43
vishymaybe I'm fooled by the 'backport'11:43
sorenDaviey: I can give you PPA privs if you want to help out.11:45
*** Vek has joined #openstack-dev11:45
sorenDaviey: Or you can off your lazy arse and become nova-core and get it for free.11:45
comstudugh11:47
comstudvishy, you're up late.11:47
comstudor maybe you're not on the west coast right now :)11:48
soren11:29 < vishy> Daviey: nope, I'm in paris11:48
comstudahh11:48
comstudthat explains it11:48
comstud:)11:48
comstudwell...11:48
comstudwhy the hell am I awake then?11:48
comstudit's nearly 5am11:48
comstud:)11:48
comstudTime for bed.11:48
Davieysoren: nova is just too broken for me to fix it all :)11:52
Davieylatest news is that the metadata service seems brittle to high throughput. :(11:52
Davieystarting 100 instances concurrently, seemed to make it cry at me.11:53
sorenDaviey: Some day, when someone will send me hardware capable of running 100 instances, I'll fix that.11:53
Davieysoren: So you'll only fix it if you get a server that can run 100 instances.  You do know that nova can be used on multiple servers to pool out, right? :)11:55
sorenDaviey: That's fascinating. Tell me more.11:56
sorenDaviey: Do you have a newsletter I can subscribe to?11:56
Davieysoren: Funny you say that!11:57
DavieyThere is a https://launchpad.net/~sorens-target-audience meetup planned soon.11:58
sorenDaviey: Oh, you heard about it? There must be a leak somewhere.12:01
Davieysoren: bah, ~sorens-target-audience was my idea!12:06
*** bsza has quit IRC12:07
*** bsza has joined #openstack-dev12:12
* ttx tries to understand why we need dhcp_release12:24
*** HugoKuo_ has quit IRC12:29
*** markmc has joined #openstack-dev12:36
smoseri feel so left out by not being in the suggested target audience.12:38
vishyDaviey: run a bunch of api's and load balance12:40
vishyDaviey: or even easier, install a nova-api on every compute host12:40
vishyand use the local nova-api for metadata12:40
*** sandywalsh has joined #openstack-dev12:44
*** AhmedSoliman has quit IRC12:45
*** hbetts has joined #openstack-dev12:47
Davieyvishy: That probably makes sense.. I don't think our suggest topology supports that, but it sounds like it should.12:51
DavieyI also noticed that starting and stopping too many too quickly, zombied qemu-nbd and a kvm process.. not quite sure that is nova's fault tho.12:51
*** zul has quit IRC13:01
*** zul has joined #openstack-dev13:02
*** mfer has joined #openstack-dev13:04
*** jaypipes has joined #openstack-dev13:11
*** bcwaldon has joined #openstack-dev13:12
bcwaldonmtaylor: marco13:12
jaypipesbcwaldon: morning sunshine.13:15
bcwaldongood morning, jay13:18
bcwaldonwas going to ask monty about gitpython on the build server13:19
bcwaldonjaypipes: you got any info on that?13:19
*** kbringard has joined #openstack-dev13:20
jaypipesbcwaldon: I asked him last Thursday...13:20
jaypipesbcwaldon: just ping him again on it when he's up (about 2-3 hours...)13:20
bcwaldonjaypipes: ok, I saw you commented on the bug, just didn't see anything further. I was going to ask if it had gotten installed yet13:20
bcwaldonjaypipes: good point, didn't think about that13:21
*** amccabe has joined #openstack-dev13:21
*** lorin1 has joined #openstack-dev13:23
*** lorin1 has left #openstack-dev13:23
jaypipesbcwaldon: https://review.openstack.org/#change,458 please see my review comments... interested to hear your thoughts on that.13:28
bcwaldonjaypipes: looking13:29
vishyDaviey: I think we're going to change our default arch to an api on the compute hosts for metadata13:30
*** joesavak has joined #openstack-dev13:31
vishyDaviey: I'm going to split out the metadata server soon, so we can actually optimize it and run it alone instead of pulling the entire api onto the comput hosts, but for now, i don't think it is too nasty.13:31
*** mandela123 has quit IRC13:33
ttxjaypipes: What's Glance Diablo status at this point in your opinion ?13:35
ttxI can see only one RC bug fix in the pipe -- can we live without the others ?13:36
jaypipesttx: bcwaldon and I are doing some final reviews on some of the bugs. I believe that the only *real* bug in danger might be the Location security bug. The Attribute context and the update bug I'm going to push to Essex, since I'm not actually sure those are bugs. Monty must fix his one assigned bug.13:37
ttxjaypipes: ok, when do you expect to get all the known and reasonable ones fixed ? Would tomorrow night still work ?13:38
ttx(levaes 1day and a half of stasis before release)13:38
jaypipesttx: all the fix committed ones except for the Swift DELETE have already been merged into milestone-proposed13:38
Davieyadam_g: What do you think about co-locating all nova-api onto nova-computes?13:39
jaypipesttx: no, I'm aiming for EOB *today*. If not done by then, they will be moved to Essex.13:39
ttxcool.13:40
jaypipesbcwaldon: ping13:42
bcwaldonyessir13:42
jaypipesbcwaldon: pls read the above conversation with ttx. you think we can knock em all out today?13:42
bcwaldonjaypipes: yeah, I think I did most of the reviews yesterday13:43
jaypipesbcwaldon: k, coolio.13:43
bcwaldonjaypipes: look through them and ping me on any I haven't reviewed yet13:43
jaypipesbcwaldon: will od13:43
jaypipesdo13:43
zulDaviey: gah?13:46
zulDaviey: oh...nm13:47
*** AhmedSoliman has joined #openstack-dev14:00
*** rcc1 has joined #openstack-dev14:02
*** ahmed_ has joined #openstack-dev14:02
*** AhmedSoliman has quit IRC14:03
*** wwkeyboard has joined #openstack-dev14:03
zykes-anyone know why tokens in keystone that are gotten via the webinterface shows as integers in keystone-manager token list whilst tokens created in the logic when doing auth via web api shows as integers in users and tenant id ?14:13
*** rcc has joined #openstack-dev14:14
*** cp16net has joined #openstack-dev14:14
*** rcc1 has quit IRC14:16
*** zul has quit IRC14:18
*** zul has joined #openstack-dev14:18
*** bcwaldon has quit IRC14:21
zykes-jaypipes: around maybe ?14:21
*** ahmed_ has quit IRC14:22
*** AhmedSoliman has joined #openstack-dev14:23
*** wwkeyboard has quit IRC14:24
*** wwkeyboard has joined #openstack-dev14:24
*** jsavak has joined #openstack-dev14:26
*** code_franco has joined #openstack-dev14:26
*** joesavak has quit IRC14:28
*** wwkeyboard has quit IRC14:28
vishyfyi, nova on maverick with that new dnsmasq is broken in some case14:29
vishyhaven't isolated what is causing it yet14:29
jaypipeszykes-: heya14:30
openstackjenkinsProject nova build #1,388: SUCCESS in 3 min 56 sec: https://jenkins.openstack.org/job/nova/1388/14:30
openstackjenkinsTarmac: Fixes the handling of snapshotting in libvirt driver to actually use the proper image type instead of using raw for everything. Also cleans up an unneeded flag.  Based on doude's initial work.14:30
*** wwkeyboard has joined #openstack-dev14:43
*** rnirmal has joined #openstack-dev14:53
*** nati has joined #openstack-dev14:57
*** dragondm has joined #openstack-dev15:03
openstackgerritVerification of a change to openstack/keystone failed: Get Service Catalog from token  https://review.openstack.org/35915:05
openstackgerritA change was merged to openstack/glance: Fixes LP Bug #804429  https://review.openstack.org/47515:06
uvirtbotLaunchpad bug 804429 in glance "Cannot stream http image" [Critical,In progress] https://launchpad.net/bugs/80442915:06
*** heckj has joined #openstack-dev15:10
*** dprince has joined #openstack-dev15:12
mtaylorvishy: dd you see bug 853892 ... the backport of the package that has dhcp_release in it seems to be causing some issues...15:14
uvirtbotLaunchpad bug 853892 in nova "dnsmasq PPA 2.58-3~openstack0~maverick1 causes 'failed to create listening socket errors'" [Undecided,New] https://launchpad.net/bugs/85389215:14
*** nati has quit IRC15:14
vishymtaylor: i just ran into it15:17
mtaylorvishy: yay!15:17
*** bcwaldon has joined #openstack-dev15:17
dprincemtaylor: Yeah. The dnsmasq issue had me scratching my head a bit. I mistakenly blamed one of Soren's branch for the dnsmasq issue on Friday. On closer look I saw his branch had nothing to do with it. Anyway it does cause me issues and I expect anyone using a DHCP network manager would hit it.15:17
vishyi can beat that15:18
vishyI'm doing a training session using nova.sh15:18
heckjuh oh… that sounds bad15:18
vishyand no-one could launch vms15:18
vishy:(15:18
mtaylorAWESOME15:18
dprinceoh dear.15:18
vishytook me about 2o min to figure out it was the package15:19
vishy* 2015:19
vishydprince: any idea what the bug is?15:20
dprincevishy: didn't chase it that far yet.15:20
vishydprince: does it happen on all versions of ubuntu?15:21
dprincevishy: could be ipv6 related. It was failing to find the bridge's IPv6 addr I think.15:21
dprincevishy: also I just hit it on maverick.15:21
vishyit happened to me on a maverick install in cloudservers15:21
vishybut on my local version of maverick it didn't happen15:21
dprincevishy: ditto15:21
vishywell hold on15:22
dprincevishy: ditto the cloudservers (not the local workstation).15:22
vishyactually15:22
vishymy dnsmasq-base is still the old version15:23
vishylet me try upgrading15:23
dprincevishy: yeah. I'd be surprised if it doesn't break your local version too.15:23
vishydprince: but you haven't tried on natty/oneiric?15:24
dprincevishy: not yet.15:24
vishyi'm wondering if there is a config change15:25
dprincevishy: I'd like to be using natty as my base setup but haven't had time to test it yet.15:25
vishyyup fails locally15:26
heckjnatty is still reporting 2.57-1 as the latest version - no local PPA's installed15:26
*** rcc has quit IRC15:27
*** rcc has joined #openstack-dev15:27
openstackgerritJay Pipes proposed a change to openstack/glance: Fixes LP Bug #804429  https://review.openstack.org/47615:27
uvirtbotLaunchpad bug 804429 in glance "Cannot stream http image" [Critical,Fix committed] https://launchpad.net/bugs/80442915:27
mtaylorvishy, dprince: looking throught the debdiff between 2.55 and 2.58 - I don't see any obvious config changes15:27
vishyhttps://bugzilla.redhat.com/show_bug.cgi?id=73701515:28
uvirtbotbugzilla.redhat.com bug 737015 in openstack-nova "IPv6/dnsmasq issue" [Unspecified,New]15:28
vishyinteresting15:28
openstackgerritJay Pipes proposed a change to openstack/glance: Fixes LP Bug #833285  https://review.openstack.org/47715:28
uvirtbotLaunchpad bug 833285 in glance "swift store delete only removes manifest" [High,Fix committed] https://launchpad.net/bugs/83328515:28
vishyapparently if you delete the ipv6 address from the bridge it works?15:28
mtayloroh lovely15:28
*** rcc1 has joined #openstack-dev15:29
vishyi've had luck with the maintainer of dnsmasq before15:30
vishyI could send him an email15:30
openstackgerritA change was merged to openstack/glance: Fixes LP Bug #804429  https://review.openstack.org/47615:31
uvirtbotLaunchpad bug 804429 in glance "Cannot stream http image" [Critical,Fix committed] https://launchpad.net/bugs/80442915:31
*** rcc has quit IRC15:31
openstackgerritA change was merged to openstack/glance: Fixes LP Bug #833285  https://review.openstack.org/47715:32
uvirtbotLaunchpad bug 833285 in glance "swift store delete only removes manifest" [High,Fix committed] https://launchpad.net/bugs/83328515:32
dprinceUntil its fixed should we pull the package out of the PPA? I suppose we need the newer dnsmasq-utils don't we?15:32
*** bcwaldon has quit IRC15:32
*** reidrac has quit IRC15:34
mtaylorvishy: is there a patch we can apply to dnsmasq that fixes it?15:38
vishytrying to find out15:38
mtaylorvishy: and/or what dprince said ... I can pull 2.58 from our ppa if need be15:38
mtaylorcool. I can either patch it or pull it - neither is a problem15:38
vishyi just sent a message to simon15:39
dprince If it takes more than a day to get a patch I'd say pull it to avoid causing issues for others.15:40
*** nati has joined #openstack-dev15:41
dprinceI can dowgrade for now.15:41
dprincedowngrade15:41
*** bcwaldon has joined #openstack-dev15:41
dprinceapt-get install dnsmasq-base=2.55-115:41
dprincethat gets it working for me.15:41
vishyyeah me 215:45
*** joesavak has joined #openstack-dev15:48
vishyi has a fix15:49
vishyshall i propose it?15:49
dprinceSo it updates our flags in nova to work around the issue?15:50
*** rcc has joined #openstack-dev15:50
vishyyup15:50
*** jsavak has quit IRC15:50
dprinceI'd be Okay with that I guess. Would it work with the old and new versions?15:51
*** rcc1 has quit IRC15:51
vishyworkaround: https://code.launchpad.net/~vishvananda/nova/new-dnsmasq/+merge/7605215:54
dprincelooking...15:54
vishyi also emailed the maintainer to ask him if it is some sort of bug that is causing it to fail15:54
vishybasically just depend on --listen_address to bind to the right place, instead of --bind-interface15:54
dprincevishy: ah. so that avoids that ipv6 lookup.15:55
vishyi think you aren't supposed to use both flags anyway15:55
vishyso it may have been our bug in the first place15:56
vishyalthough I think it is odd that dnsmasq fails when trying to hit ipv615:56
*** Gordonz has joined #openstack-dev15:56
*** markmc has quit IRC15:59
ttxvishy: I know Simon Kelley, if that helps16:00
*** Rajaram has quit IRC16:02
*** wwkeyboard_ has joined #openstack-dev16:03
*** wwkeyboard has quit IRC16:03
*** wwkeyboard_ has quit IRC16:03
blamarchemikadze: Hey :) what is an association? normally I think of it being like a fixed ip address associated with a network16:04
*** wwkeyboard has joined #openstack-dev16:05
*** nati has quit IRC16:10
jaypipesmtaylor: could we install GitPython on the glance builder pretty please? thx!16:10
*** martine has joined #openstack-dev16:11
*** AhmedSoliman has quit IRC16:12
openstackgerritDavid Cramer proposed a change to openstack/keystone: Changing the format of markup to pull in examples in wadl:doc elements  https://review.openstack.org/47816:15
*** AhmedSoliman has joined #openstack-dev16:22
adam_gww/win 2016:23
adam_gDaviey: i suppose we could16:24
heckjls16:27
*** deshantm_laptop has joined #openstack-dev16:27
heckjdamnit16:27
jaypipesglenc: ping16:29
*** vladimir3p has joined #openstack-dev16:34
*** mjs has joined #openstack-dev16:37
*** nati has joined #openstack-dev16:38
mtaylorjaypipes: yes. is it a depend of glance itself?16:38
bcwaldonmtaylor: just tests16:38
bcwaldonmtaylor: I wrote a new Authors file test for git projects16:38
*** mjs has quit IRC16:41
*** mjs has joined #openstack-dev16:41
*** mjs has quit IRC16:45
smoserttx, around ?16:47
smoseror vishy ?16:48
jaypipesmtaylor: tests only.16:50
mtaylorjaypipes: got it16:50
dprincebwaldon: http://paste.openstack.org/show/2472/16:52
*** nati has quit IRC16:54
*** jdurgin has joined #openstack-dev16:57
*** martines_ has joined #openstack-dev16:59
*** martines_ has quit IRC16:59
*** martines has joined #openstack-dev17:00
*** chemikadze has quit IRC17:01
kbringardare there old packages in the trunk repo?17:03
kbringardfor instance, if I want to install a release that's close to the diablo release without installing Essex packages17:03
tjoyuse /milestone, not /trunk17:04
kbringardye,a but that's cactus right now17:04
tjoyno it's not, i installed D4 from it17:04
kbringardahhhh, nice17:04
tjoyyou're thinking of /release17:04
kbringardooooo, I'm retarded17:04
kbringardthank you sir17:05
openstackgerritYogeshwar Srikrishnan proposed a change to openstack/keystone: Initial Changes to move service operations to extensions.  https://review.openstack.org/47917:08
mtaylordprince, jaypipes, bcwaldon: https://code.launchpad.net/~mordred/glance/fix-deb-packaging-diablo/+merge/76070 and https://code.launchpad.net/~mordred/glance/fix-deb-packaging-trunk/+merge/7606817:08
*** novas0x2a|laptop has joined #openstack-dev17:09
smosercould someone able please propose bug 853330 for milestone ?17:10
smoseri would suspect that mtaylor or bcwaldon could17:11
rmkCan anyone shed light on this stack trace (http://pastebin.com/RBbbeTKU)?  It's being triggered when I attempt to launch an instance via Dashboard.17:11
rmkRunning the latest dashboard, keystone and milestone-proposed nova.17:12
*** bengrue has joined #openstack-dev17:18
openstackgerritJames E. Blair proposed a change to openstack/openstack-ci: Add section on access policies.  https://review.openstack.org/48017:18
openstackgerritA change was merged to openstack/keystone: Initial Changes to move service operations to extensions.  https://review.openstack.org/47917:22
*** misheska has joined #openstack-dev17:27
*** misheska has quit IRC17:28
*** misheska has joined #openstack-dev17:30
*** martines has quit IRC17:44
*** martines has joined #openstack-dev17:44
*** markmc has joined #openstack-dev17:45
*** jdag has joined #openstack-dev17:58
zykes-jaypipes: how should the glance endpoint in keystone look url wise ?17:58
*** mgius has joined #openstack-dev17:58
jaypipeszykes-: not with a 1.1 in it... ;)17:58
zykes-and %tenant then ?17:58
jaypipeszykes-: nope.17:58
jaypipeszykes-: should look like this:17:59
zykes-then the keystone samples are wrong or ?17:59
jaypipeszykes-: yes, they are wrong.17:59
jaypipeszykes-: someurl.com/v1/images17:59
zykes-kk17:59
*** markwash has joined #openstack-dev17:59
openstackgerritBrian Waldon proposed a change to openstack/glance: Fixing bug 794582 - Now able to stream http(s) images  https://review.openstack.org/45518:04
uvirtbotLaunchpad bug 794582 in glance "glance sets image size to 0 for http images -> yields connection reset error on GET" [Medium,In progress] https://launchpad.net/bugs/79458218:04
annegentleDoc Blitz in one hour in #openstack, at http://docs.openstack.org/docblitz, and using the Twitter hashtag #docblitz. Log doc bugs at http://bugs.launchpad.net/openstack-manuals.18:06
*** rnorwood has joined #openstack-dev18:09
*** zul has quit IRC18:09
*** mfer has quit IRC18:10
*** mfer has joined #openstack-dev18:11
zykes-jay what about stuff like internal / admin url then jaypipes ?18:17
zykes-annegentle: docblitz meaning ?18:18
annegentlehey zykes- it's an hour-long test and review session for the docs18:18
zykes-ah18:18
annegentlemore details at http://www.openstack.org/blog/2011/09/openstack-documentation-blitz/18:18
zykes-is there any puppet stuff for setting up stuff like glance, switft etc ?18:19
*** zul has joined #openstack-dev18:20
mtaylorzykes-: have you looked at github.com/openstack/openstack-puppet ?18:29
*** AhmedSoliman has quit IRC18:31
zykes-nop18:32
*** AhmedSoliman has joined #openstack-dev18:32
zykes-but i will now :)18:34
*** code_franco has quit IRC18:34
openstackgerritDolph Mathews proposed a change to openstack/keystone: Backend-managed service ID's (#834683)  https://review.openstack.org/48118:34
openstackgerritDolph Mathews proposed a change to openstack/keystone: Backend-managed role ID's (bug #834683)  https://review.openstack.org/43118:34
uvirtbotLaunchpad bug 834683 in keystone "RoleID and RoleName" [Undecided,In progress] https://launchpad.net/bugs/83468318:34
*** code_franco has joined #openstack-dev18:36
openstackgerritDolph Mathews proposed a change to openstack/keystone: Backend-managed role ID's (bug #834683)  https://review.openstack.org/43118:36
uvirtbotLaunchpad bug 834683 in keystone "RoleID and RoleName" [Undecided,In progress] https://launchpad.net/bugs/83468318:36
openstackgerritDolph Mathews proposed a change to openstack/keystone: Backend-managed role & service ID's (bug #834683)  https://review.openstack.org/43118:37
uvirtbotLaunchpad bug 834683 in keystone "RoleID and RoleName" [Undecided,In progress] https://launchpad.net/bugs/83468318:37
openstackgerritJay Pipes proposed a change to openstack/glance: Fixes LP Bug#755916 - Location field shows creds  https://review.openstack.org/48218:43
jaypipeszykes-: there is no internal or admin-specific URI.18:47
jaypipeszykes-: puppet stuff: https://github.com/openstack/openstack-puppet18:48
rmkkeystone-manage is missing any/all help text18:48
*** yogirackspace has joined #openstack-dev18:53
openstackgerritVerification of a change to openstack/glance failed: Adding git-specific Authors file test  https://review.openstack.org/45118:53
jaypipesmtaylor: https://jenkins.openstack.org/job/glance/89231/console18:55
*** markwash has quit IRC18:56
zykes-hi yogirackspace, do you know what OS-KSADM:service is ?18:58
yogirackspacezykes: let me check18:59
zykes-service operations to extentions, what does that mean ?18:59
annegentlermk: hm, there's a man page for it in /doc/source/man, but maybe it doesn't work? That's a doc bug. Want to log it?19:00
yogirackspacezykes:OS-KSADM:service Where did u find it?19:02
*** bcwaldon_ has joined #openstack-dev19:03
zykes-logic/types/services.py in keystone19:04
*** bcwaldon has quit IRC19:04
zykes-service.py sorry19:04
yogirackspace zykes:All the service operations are moved to an extension called OS-KSADM:Service19:06
zykes-ah19:06
yogirackspacezykes:The Service is now part of the name space http://docs.openstack.org/identity/api/ext/OS-KSADM/v1.019:06
*** anotherjesse has joined #openstack-dev19:07
zykes-those pages are not existant yogirackspace19:07
*** jakedahn has joined #openstack-dev19:07
yogirackspacezykes: Suggested JSON equivalent  way of handling type thats part of extension is to add the extension short name as prefix19:07
yogirackspaceZykes: Check keystone/keystone/contrib/extensions/admin/osksadm/extensionHandler.py19:08
zykes-ok : )19:08
yogirackspaceZykes:Earlier calls have been moved here19:08
yogirackspaceZykes:And this is the initial commit.We shall keep refactoring code to move to extension location19:09
mtaylorjaypipes: yes. it's been added to the glance packaging - but we need to get a new glance package built in the ppa before it'll take effect on the builder19:09
*** notmyname has quit IRC19:24
*** notmyname has joined #openstack-dev19:25
*** ChanServ sets mode: +v notmyname19:25
*** markmc has quit IRC19:31
dprincejoesavak: got a keystone question for you.19:37
joesavakdprince: shoot19:38
dprincejoesavak: so I made a commit that updated keystone-manage so that it converted names to ids. I'm getting some feedback that people don't like this behaviour:19:38
dprincehttps://github.com/openstack/keystone/commit/a0d2519c783774762e7b9efbcbdb75c55ad69faa#commitcomment-59785019:38
dprincejoesavak: Sorry to link you the whole thread but it isn't long...19:39
ttxsmoser: pong19:39
smoserttx, see pm19:39
sandywalshjoesavak ping19:39
dprincejoesavak: I actually think the issue is our keystone examples are now wrong.19:39
joesavaksandy: hi19:39
sandywalshjoesavak hey! available to chat about authz at some point?19:40
joesavakwhat's wrong about thim?19:40
zykes-dprince: not only that, but when the CRUD code is updated, some times the list stuff doesn't list properly...19:40
dprincejoesavak: So for example. instead of sending:19:40
dprincecurl -d '{"passwordCredentials": {"username": "joeuser", "password": "secrete", "tenantId": "1"}}' -H "Content-type: application/json"19:40
joesavaksandy: sure - have 20 mins til my next meeting. authz essex?19:40
dprinceSorry. That is what we should send now.19:41
dprinceInstead of:19:41
dprincecurl -d '{"passwordCredentials": {"username": "joeuser", "password": "secrete", "tenantId": "1234"}}' -H "Content-type: application/json"19:41
sandywalshjoesavak, cool, skype?19:41
joesavaksandy: ok - i'll call you after I finish up with dprince19:41
dprinceHere is the confusing part. 1234 is the 'name' of the tenant in the examples. But I think we now need to send IDs.19:41
dprinceAnd this is confusing everyone.19:41
sandywalshjoesavak, take your time!19:41
dprinceI mean the name of the JSON parameter is tenantId.19:42
joesavakdprince: yeah - we should be sending in IDs now.19:42
joesavakping yogi19:42
joesavakyogi: is this right? It should be IDs for tenantId and not name, right?19:42
dprincejoesavak: So if I merge prop something that cleans up the examples in keystone (in the README and the sample data too) would you buy that.19:42
openstackgerritDolph Mathews proposed a change to openstack/keystone: Backend-managed role & service ID's (bug #834683)  https://review.openstack.org/43119:43
uvirtbotLaunchpad bug 834683 in keystone "RoleID and RoleName" [Undecided,In progress] https://launchpad.net/bugs/83468319:43
dprincejoesavak: Again. The confusing thing is that one of our sample tenants is named '1234' (the name) which is likely getting created with ID 1.19:43
joesavakdprince: yes, seems logical to me19:43
yogirackspaceids would soon be replaced with autogenerated  uuids19:43
yogirackspacethats a story thats coming soon19:44
dprinceSUre. I'm fine with that too.19:44
joesavakyogi: "soon" though isn't defined19:44
yogirackspacejoesavak: u r the decider!19:44
dprinceIn the meantime though I think cleaning up the examples to clearly display that we are using IDs accross the board is really needed.19:44
yogirackspaceIll change examples to some random strings19:45
joesavakyup - need to clean the samples to make it clear then when we do uuids and accept string instead of integer for that field, the samples will be reworked then19:45
dprinceIts hard for people to see that "1234" is really the name of the tenant and not the ID.19:45
yogirackspacewhatever thats there the value shud be treated as ID19:45
openstackgerritDolph Mathews proposed a change to openstack/keystone: Backend-managed role & service ID's (bug #834683)  https://review.openstack.org/43119:45
uvirtbotLaunchpad bug 834683 in keystone "RoleID and RoleName" [Undecided,In progress] https://launchpad.net/bugs/83468319:45
yogirackspaceprobably it confuses as id value is also the same as name19:46
*** bcwaldon_ has quit IRC19:47
openstackgerritA change was merged to openstack/keystone: Backend-managed role & service ID's (bug #834683)  https://review.openstack.org/43119:49
uvirtbotLaunchpad bug 834683 in keystone "RoleID and RoleName" [Undecided,In progress] https://launchpad.net/bugs/83468319:49
*** heckj has quit IRC19:52
*** bcwaldon has joined #openstack-dev19:53
openstackgerritA change was merged to openstack/glance: Fixes LP Bug#755916 - Location field shows creds  https://review.openstack.org/48219:54
jaypipesmtaylor: now that a new commit has landed in glance trunk, packages have been built, right? I can throw the Authors patch into approval now?19:55
rmkI'd also like to suggest that the sampledata for keystone be a bit more functional19:56
openstackgerritBrian Waldon proposed a change to openstack/glance: Fixing bug 794582 - Now able to stream http(s) images  https://review.openstack.org/45519:56
uvirtbotLaunchpad bug 794582 in glance "glance sets image size to 0 for http images -> yields connection reset error on GET" [Medium,In progress] https://launchpad.net/bugs/79458219:56
rmkI'll have to trace my steps to remember exactly what it was, but I believe some of the tokens had no tenant association19:56
*** ipl31 has quit IRC19:58
*** ipl31 has joined #openstack-dev19:59
*** AhmedSoliman has quit IRC20:00
rmkdprince: I'm glad you guys are looking at the sample data (just scrolled back and read the conversation).  It was confusing to go through at first.20:03
*** anotherjesse has quit IRC20:04
*** anotherjesse has joined #openstack-dev20:04
rmkI still need to file a bug for the keystone ubuntu package also.  It was missing a default configuration file and content directories.20:04
dprincermk: Sure. I'll try and fix it up here shortly. Sounds like it is confusing lots of people.20:05
*** deshantm_laptop has quit IRC20:06
rmkdprince: I think it would be useful to have sampledata to contain a fully functional initial set of users/tenants/etc.20:06
mtaylorjaypipes: lemme doublecheck20:07
zykes-yeah rmk instead of each commit fucking it up ;p20:08
rmkI guess my point is to only have what's necessary for a user rather than a few senarios for developers20:08
mtaylorjaypipes: currently building - give it a few more minutes20:11
*** cp16net has quit IRC20:12
zykes-yogirackspace: just as a quicky, shouldn't excentionHandler.py be extension_handler.py instead ?20:14
yogirackspaceyes Ill rename my bad20:14
vishymtaylor: any idea why the tests are running twice? https://jenkins.openstack.org/job/nova-tarmac/120785/consoleText20:14
mtaylorvishy: no. I've been wanting to look in to that whole thing, but have also been avoiding too much thought on it pre-gerrit20:15
mtaylorvishy: removing the moving piece of tarmac from the equation gets us to a place where I believe we'll be much better able to actually grok what's happening.20:16
rmkdprince: I think the endpoint templates also need to be updated20:16
mtaylorjeblair: you did a patch to one of the run_tests.sh recently, didn't you?20:16
dprincermk: Okay.20:17
* mtaylor would really love to be able to run the test suites from jenkins without running pep8 - since we've got a pep8 job as well20:17
vishyttx: ready20:18
vishymtaylor: gotcha20:18
ttxvishy: what's current Nova/Diablo status from your perspective20:19
vishywe're pretty darn close20:19
ttxvishy: cool20:20
rmkYeah so here's what threw me off.  The default "token" rows reference use tenant_id of 1234, rather than the actual id.20:20
vishyDaviey: here?20:21
ttxvishy: do you think we can stop backporting bugfixes (unless critical) tomorrow eod ?20:21
vishyttx: two things are critical atm:20:22
vishy85333020:22
vishybug 85333020:22
vishythat one20:22
vishyand20:22
vishyhttps://bugs.launchpad.net/nova/+bug/83355220:22
rmkSame problem with endpoints, tenant_id=1234 rather than the id.20:22
uvirtbotLaunchpad bug 833552 in nova "No easy migration from old auth to keystone" [Critical,In progress]20:22
dprincermk: roger that. Things just need to line up.20:23
dprincermk: I'll just grep the tree....20:23
ttxvishy: how is the fix for that last one looking ?20:23
rmkcool, I just recreated my db is why I am mentioning it here but it sounds like you're on it already so I'll stop :)20:23
vishyttx: I thought it was good, but bwaldon was having issues20:24
vishypvo: status of bug833552 ?20:24
vishybug 833552 even20:24
uvirtbotLaunchpad bug 833552 in nova "No easy migration from old auth to keystone" [Critical,In progress] https://launchpad.net/bugs/83355220:24
bcwaldonvishy: it may just be me, but it didn't functionally work20:25
bcwaldonvishy: dprince said it did before a few more commits went in20:25
vishybcwaldon: yeah we definitely need to verify it20:26
openstackgerritA change was merged to openstack/glance: Fixing bug 794582 - Now able to stream http(s) images  https://review.openstack.org/45520:27
uvirtbotLaunchpad bug 794582 in glance "glance sets image size to 0 for http images -> yields connection reset error on GET" [Medium,In progress] https://launchpad.net/bugs/79458220:27
ttxvishy: what about bug 833331 ?20:28
uvirtbotLaunchpad bug 833331 in nova "OSAPI v1.1 needs to document config-drive as an extension" [Medium,Confirmed] https://launchpad.net/bugs/83333120:28
openstackgerritVerification of a change to openstack/glance failed: Adding git-specific Authors file test  https://review.openstack.org/45120:28
vishyttx: not critical20:28
ttxAlso looks like we'll pass on bug 838386 and bug 845866, as nobody fixed them20:28
uvirtbotLaunchpad bug 838386 in nova "Test suite requires kombu and carrot to be installed" [Low,Triaged] https://launchpad.net/bugs/83838620:28
uvirtbotLaunchpad bug 845866 in nova "exception tracebacks are lost when using raise" [Low,Triaged] https://launchpad.net/bugs/84586620:28
vishyttx: plus doc will be in the api repo i think20:28
ttxvishy: ok, please untarget them if they are no longer in the critical path20:29
annegentleI did get a contribution for 83333120:29
ttxvishy: ideally the list at https://launchpad.net/nova/+milestone/2011.3 should be what still needs to be done before we can release20:29
ttxso we should add what's missing and remove what's no longer a target20:30
mtaylorvishy: so - someone (I forget who) added python-tempo as a depend for nova in trunk (or rather requested that I do)20:30
annegentleCan we target the rest of the missing extension docs? https://bugs.launchpad.net/nova/+bug/85283020:30
vishyuntargeted20:30
uvirtbotLaunchpad bug 852830 in nova "Need documentation for Compute API 1.1 extensions" [Undecided,New]20:30
zulmtaylor: hey can you generate tarballs for quantum and dashboard?20:30
mtaylorvishy: do you happen to know who it was that was the tempo person?20:30
mtaylorzul: sure20:30
vishymtaylor: i would guess s1rp20:30
mtaylors1rp: ping20:30
vishyit was his branch that requires it afaik20:31
ttxvishy: hyper-V stuff will be a MISS too, I think.20:31
mtaylorvishy: getting it to fully backport and work on lucid is turning into a bit of a shitshow ... a few depends down it needs a new/updated version of jinja220:31
vishyttx: yes20:31
mtaylorwhich seems a bit excessive for something that's providing cron services20:31
openstackjenkinsProject nova-milestone build #53: SUCCESS in 20 sec: https://jenkins.openstack.org/job/nova-milestone/53/20:32
openstackjenkinsTarmac: Adds an optional flag to force dhcp releases on instance termination. This allows ips to be reused without having to wait for the lease to timeout.20:32
vishyttx: untarget?20:32
vishymtaylor: agreed20:32
vishymtaylor: we might have to rethink that dependency20:32
*** rcc has quit IRC20:34
ttxvishy: yep, willdo20:34
s1rpmtaylor: yo20:40
mtaylors1rp: python-tempo is your thing, yeha?20:40
s1rpmtaylor: it's a Rackspace/Ozone project, but yeah, right now I own the PyPI release20:41
mtaylors1rp: trying to get a version of it into the ppa for lucid is ... tricky ... because it depends on flask, which in turn depends on a newer version of jinja2 than exists for lucid at the moment20:41
mtaylors1rp: and unfortunately, jinja2 is proving to be a wee bit of a bitch to backport20:41
openstackjenkinsProject nova-milestone build #54: SUCCESS in 17 sec: https://jenkins.openstack.org/job/nova-milestone/54/20:42
openstackjenkinsTarmac: Fixes the handling of snapshotting in libvirt driver to actually use the proper image type instead of using raw for everything. Also cleans up an unneeded flag. Based on doude's initial work.20:42
mtaylors1rp: any way to be able to use tempo without needing newer jinja2?20:42
zykes-ozone ?20:42
s1rpmtaylor: offhand i'm not positive, but my guess would be that we could20:43
vishytr3buchet: ping20:43
zykes-what's that20:43
s1rpmtaylor: ill drop an issue into the github tracker for tempo20:43
mtaylors1rp: cool. thanks!20:44
vishymtaylor: is jenkins broke?20:46
vishyhttps://code.launchpad.net/nova/+activereviews20:46
vishythere are 5 branches stuck waiting to land20:46
* ttx looks20:48
tr3buchetvishy: sup?20:49
mtaylorvishy: looking20:50
vishytr3buchet: is pvo nearby?  Trying to find out what is up with the keystone migration20:50
vishytr3buchet: it is the last patch for milestone-proposed afaik20:50
tr3buchetvishy: yeah he's around20:50
ttxjenkins takes ages to run the tarmac jobs20:50
rmkSince pulling keystone from git an hour ago I can't get the service admin tab to show up in the dashboard..20:51
vishytr3buchet: can you kick him for me?20:51
tr3buchetvishy: i'll go ask him20:51
openstackjenkinsProject nova-milestone build #55: SUCCESS in 17 sec: https://jenkins.openstack.org/job/nova-milestone/55/20:52
openstackjenkinsTarmac: create disk.local the same way ephemerals are created (LP: #851145)20:52
openstackjenkinsIf the user did not specify '--block-device-mapping /dev/vdb=ephemeral0', then20:52
openstackjenkinsthen the first non-root device would end up being created differently then20:52
openstackjenkinsif they had.  It would not have a filesystem on it.20:52
openstackjenkinsThis makes 'local_gb' in the libvirt connection created the same way that20:52
openstackjenkinsit would be if it were named ephemeral0.20:52
ttxmtaylor: I re-enabled nova-1, but last time it was enabled, the nova job failed there.20:52
ttxmtaylor: should help with the bottleneck, but maybe watch them20:52
mtaylorttx: k20:53
openstackjenkinsProject nova build #1,389: SUCCESS in 3 min 53 sec: https://jenkins.openstack.org/job/nova/1389/20:53
openstackjenkins* Tarmac: * Remove the foreign key and backrefs tying vif<->instance20:53
openstackjenkins* Update instance filtering to pass ip related filters to the network manager20:53
openstackjenkins* move/update tests20:53
openstackjenkins* Tarmac: Adds an optional flag to force dhcp releases on instance termination.  This allows ips to be reused without having to wait for the lease to timeout.20:53
openstackjenkins* Tarmac: Fixing case where OSAPI server create would return 500 on malformed body.20:53
openstackjenkins* Tarmac: Fix the issue with the new dnsmasq where it tries and fails to bind to ipv6 addresses.20:53
tr3buchetvishy: kicked. he's on the phone atm20:54
vishymtaylor: any idea: https://code.launchpad.net/~smoser/nova/trunk.lp851145/+merge/7580020:55
vishy?20:55
vishymtaylor: is it the /r in the commit message again?20:55
mtaylorvishy: probably20:55
smoser /r?20:55
*** markvoelker has quit IRC20:57
vishyapparently tarmac pukes if there is /r/n as line beaks20:59
vishyum i mean \r\n20:59
* vishy is a string escaping amateur21:00
tr3buchetand a paintball amateur!21:00
ttxtr3buchet: if you kick hard enough, he could not be on the phone anymore21:01
tr3buchethaha21:01
tr3bucheti didn't know i was supposed to be roundhousing21:01
ttxvishy: about to disappear.. anything you need from me ?21:01
vishyttx: nopers21:02
* smoser wonders how his commit message got /r21:02
*** mdomsch has joined #openstack-dev21:02
tr3buchetvishy: i have a small merge prop21:03
pvovishy:just saw Waldon's note. Will work on it now.21:03
tr3buchetsomething that slipped my mind in D421:03
* vishy has been made very sad by tr3buchet 21:03
tr3buchetit's VERY SMALL!21:03
vishytr3buchet: is it removing the deprecation warning for nova-manage?21:03
tr3buchetyes21:03
vishyI still see that ACHTUNG!21:03
tr3buchetyeah i just pulled it in a branch i'm making sure i didn't do something stupid right quick before i propose21:04
tr3bucheti loved that ACHTUNG! while it was there21:04
pvovishy: should admin user go as part of the export? I was only looking for users and projects.21:08
pvoit doesn't add it if it isn't part of a project21:10
tr3buchetsee!?21:10
tr3buchettr3buchet used kick, it's super effective!21:11
villepI'm having trouble with getting keystone echo-example working...21:11
villepkeystone is running, echod is running21:11
villepkeystone.db contains all tables and sampledata, but I still get 401 from echo_client21:11
openstackjenkinsProject nova-milestone build #56: SUCCESS in 17 sec: https://jenkins.openstack.org/job/nova-milestone/56/21:12
openstackjenkinsTarmac: Fix the issue with the new dnsmasq where it tries and fails to bind to ipv6 addresses.21:12
villep{"unauthorized": {"message": "Unauthorized on this tenant", "code": "401"}}21:12
villepjust can't understand why21:12
openstackgerritJay Pipes proposed a change to openstack/glance: Fixes LP Bug#755916 - Location field shows creds  https://review.openstack.org/48321:12
*** martine has quit IRC21:16
vishypvo: interesting.  It should probably get the admin role21:16
vishy* Admin21:16
openstackgerritA change was merged to openstack/glance: Fixes LP Bug#755916 - Location field shows creds  https://review.openstack.org/48321:17
bcwaldonvishy, pvo: I'm cool on keystone-export21:18
bcwaldonvishy: feel free to send it in21:18
pvovishy: let me check on thing. I might be in an old branch.21:19
tr3buchetvishy: alright they look good proposing21:20
pvoer, one thing.21:20
openstackgerritYogeshwar Srikrishnan proposed a change to openstack/keystone: Changing/introducing actual extension json/xml snippets.  https://review.openstack.org/48421:21
*** mfer has quit IRC21:23
pvook, I'm good. Vishy, the behavior is the user isn't exported if they are not apart of a project.21:27
vishyok21:27
pvothat work for you21:27
pvo?21:27
vishypvo: sure21:28
pvook21:28
vishyapproving21:30
openstackjenkinsProject nova build #1,390: SUCCESS in 3 min 52 sec: https://jenkins.openstack.org/job/nova/1390/21:30
openstackjenkinsTarmac: Adds the ability to automatically confirm resizes after the `resize_confirm_window` (0/disabled by default).21:30
*** lts has quit IRC21:38
vishymtaylor: https://code.launchpad.net/~yamahata/nova/milestone-proposed-lp851218/+merge/75841 ??21:44
vishywhy isn't that merging?21:44
vishypvo: https://code.launchpad.net/~pvo/nova/keystone-export/+merge/7584421:45
vishyneed a backport of that on to milestone-proposed21:46
*** amccabe has quit IRC21:47
vishymtaylor: console output; https://jenkins.openstack.org/view/Milestone-proposed/job/nova-milestone-tarmac/23378/console21:49
pvohow do I do that?21:49
openstackjenkinsProject nova build #1,391: SUCCESS in 2 min 56 sec: https://jenkins.openstack.org/job/nova/1391/21:49
openstackjenkins* Tarmac: Reworked the export command to be21:49
openstackjenkinsnova-manage shell export --filename=somefile21:49
openstackjenkinsThis will output the somefile file in to the keystone-manage syntax that will then be used by a shellscript that is being proposed to keystone.21:49
openstackjenkins* Tarmac: Adds the ability to automatically confirm resizes after the `resize_confirm_window` (0/disabled by default).21:49
openstackjenkins* Tarmac: * Remove the foreign key and backrefs tying vif<->instance21:49
openstackjenkins* Update instance filtering to pass ip related filters to the network manager21:49
openstackjenkins* move/update tests21:49
openstackjenkins* Tarmac: Adds an optional flag to force dhcp releases on instance termination.  This allows ips to be reused without having to wait for the lease to timeout.21:49
openstackjenkins* Tarmac: Fixing case where OSAPI server create would return 500 on malformed body.21:49
openstackjenkins* Tarmac: Fix the issue with the new dnsmasq where it tries and fails to bind to ipv6 addresses.21:49
openstackjenkins* Tarmac: Fixes the handling of snapshotting in libvirt driver to actually use the proper image type instead of using raw for everything. Also cleans up an unneeded flag.  Based on doude's initial work.21:49
openstackjenkins* Tarmac: Remove vestigial db call for fixed_ips.21:49
openstackjenkins* Tarmac: Fixes the user credentials for installing a config-drive from imageRef21:49
openstackjenkins* Tarmac: Some Linux systems can also be slow to start the guest agent. This branch extends the windows agent timeout to apply to all systems.21:49
openstackjenkins* Tarmac: Fix a bug that would make spawning new instances fail if no port/protocol is given (for rules granting access for other security groups).21:49
openstackjenkins* Tarmac: When swap is specified as block device mapping, its size becomes 0 wrongly.21:49
openstackjenkinsThis patch make it set to correct size according to instance_type.21:49
openstackjenkins* Tarmac: This changes the interpretation of 'swap' for an instance-type to be in MB rather than GB.21:49
*** joesavak has quit IRC21:49
openstackjenkins* Tarmac: Authorize to start a LXC instance withour, key, network file to inject or metadata.21:49
bcwaldoninteresting...tell me more21:49
openstackjenkins* Tarmac: Fixes an issue where 'invalid literal for int' would occur when listing images after making a v1.1 server snapshot (with a UUID).21:49
openstackjenkinsv1.1 image id's are now treated as strings (not integer ID's). The v1.0 API still tries to treat image id's as integers but doesn't fail miserably if they are uuid's either.21:49
openstackjenkinsThis should pave the way for image ID's as uuids and more closely matches the v1.1 spec with regards to images and the server refs they contain.21:49
openstackjenkins* Tarmac: Update test_libvirt so that flags and fakes are used instead of mocks for utils.import_class and utils.import_object. Fixes #lp849329.21:49
openstackjenkins* Tarmac: Multi-NIC support for vmwareapi virt driver in nova.21:50
openstackjenkinsDoes injection of Multi-NIC information to instances with Operating system flavors Ubuntu, Windows and RHEL.21:50
openstackjenkinsvmwareapi virt driver now relies on calls to network manager instead of nova db calls for network configuration information of instance.21:50
openstackjenkinsRe-oranized VMWareVlanBridgeDriver and added session parmeter to methods to use existing session. Also removed session creation code as session comes as argument.21:50
openstackjenkinsAdded check for flat_inject flag before attempting an inject operation.21:50
openstackjenkinsThis branch resolves the following bugs:21:50
pvowoah21:50
openstackjenkinsBug #831497 in OpenStack Compute (nova): "Instance spawn operation fails on ESXi compute node"21:50
openstackjenkinshttps://bugs.launchpad.net/nova/+bug/83149721:50
uvirtbotLaunchpad bug 831497 in nova "Instance spawn operation fails on ESXi compute node" [High,Fix committed] https://launchpad.net/bugs/83149721:50
openstackjenkinsBug #839383 in OpenStack Compute (nova): "ESX(i) VIFs and mac addresses"21:50
openstackjenkinshttps://bugs.launchpad.net/nova/+bug/83938321:50
uvirtbotLaunchpad bug 831497 in nova "Instance spawn operation fails on ESXi compute node" [High,Fix committed]21:50
openstackjenkins* Tarmac: Deprecate aoe in preperation for removal in essex.21:50
openstackjenkins* Tarmac: Only allow up to 15 chars for a Windows hostname.21:50
uvirtbotLaunchpad bug 839383 in nova "ESX(i) VIFs and mac addresses" [High,Fix committed] https://launchpad.net/bugs/83938321:50
openstackjenkins* Tarmac: Fix instance rebooting (lp847604) by correcting a malformed cast in compute.api and an incorrect method signature in the libvirt driver.21:50
openstackjenkins* Tarmac: This patch teaches virt/libvirt how to format filesystem on ephemeral device depending on os_type so that the behaviour matches with EC2's.21:50
uvirtbotLaunchpad bug 839383 in nova "ESX(i) VIFs and mac addresses" [High,Fix committed]21:50
openstackjenkinsSuch behaviour isn't explicitly described in the documentation, but it is confirmed by checking realy EC2 instances.  This patch introduces options virt_mkfs as multistring.21:50
openstackjenkinsIts format is --virt_mkfs=<os_type>=<mkfs command> When creating ephemeral device, format it according to the option depending on os_type.  This addresses the bugs,21:50
bcwaldonmmhmm21:50
openstackjenkins* Tarmac: pep8 fixes in nova/db/sqlalchemy/api.py and nova/virt/disk.py21:50
openstackjenkins* Tarmac: Update GlanceClient, GlanceImageService, and Glance Xen plugin to work with Glance keystone.21:50
openstackjenkins* Tarmac: Some arches dont have dmidecode, check to see if libvirt is capable of running rather getInfo of the arch its running on.21:50
openstackjenkins* Tarmac: Fixes rogue usage of sudo that crept in21:50
openstackjenkins* Tarmac: Fixes shutdown of lxc containers.21:50
openstackjenkins* Tarmac: use 'qemu-image resize' rather than 'truncate' to grow image files21:50
openstackjenkinsqcow-image is capable of growing qcow formated disks in addition to21:50
openstackjenkinsbeing able to grow a qcow formated one. (LP: #836759)21:50
openstackjenkinsI've verified that after this change both a full disk uploaded as 'raw' and one that was uploaded as 'qcow2' are correctly resized, and that in oneiric, the instance has its root filesystem resized for /dev/vda1 correctly.21:50
openstackjenkins* Tarmac: When vpn=true in allocate ip, it attempts to allocate the ip that is reserved in the network. Unfortunately fixed_ip_associate attempts to ignore reserved ips.21:50
openstackjenkinsThis fix allows to filter reserved ip address only when vpn=True.21:50
openstackjenkins* Tarmac: Do not require --bridge_interface for FlatDHCPManager (lp:844944)21:50
openstackjenkinsUnlike VlanManager, FlatDHCPManager actually works fine without a bridge21:50
openstackjenkinsinterface on single host deployments.21:50
openstackjenkins* Tarmac: Makes nova-vncproxy listen for requests on the queue like it did before the bin files were refactored.21:50
openstackjenkins* Tarmac: Stock zones follows a fill-first methodology���the current zone is filled with instances before other zones are considered.  This adds a flag to nova to select a spread-first methodology.  The implementation is simply adding a random.shuffle() prior to sorting the list of potential compute hosts by weights.21:50
openstackjenkins* Tarmac: Pass reboot_type (either HARD or SOFT) to the virt layers from the API.21:50
openstackjenkinsThis only works with XenAPI at the moment. The other hypervisors will need to be implemented as their maintainers see fit.21:50
openstackjenkins* Tarmac: Remove the unnecessary sudo from qemu-img as it is unneeded and doesn't work with our current packaging.21:50
openstackjenkins* Tarmac: Remove chanes_since and key_name from basic server entity.21:50
openstackjenkins* Tarmac: adds a fake_network module to tests to generate sensible network info for tests. It does not require using the db.21:50
openstackjenkinsThis is the first round of a few, it creates the fake_get_instance_nw_info function and updates the libvirt tests to use it.21:50
pvomore. more!21:50
zykes-what the hell ?21:50
bcwaldonI think it's done...21:50
annegentletee hee21:51
zykes-1 month of changes or something ? :p21:51
*** jakedahn has quit IRC21:53
*** jakedahn has joined #openstack-dev21:54
*** anotherjesse has quit IRC21:55
Davieyvishy: hey!21:56
*** bcwaldon has quit IRC21:56
vishyDaviey: so merge to ubuntu of dnsmasq-utils?21:57
vishyDaviey: just wondering on status21:57
*** misheska has quit IRC21:59
*** wwkeyboard has quit IRC22:01
*** wwkeyboard_ has joined #openstack-dev22:01
pvovishy: before the flood, were you saying we need to backport that to the last milestone?22:01
Davieyvishy: hold fire.. will check22:04
Daviey(i uploaded it earlier, but we are in a freeze - making it harder)22:04
*** Gordonz has quit IRC22:04
*** misheska has joined #openstack-dev22:05
*** mdomsch has quit IRC22:05
vishypvo: yes please22:06
vishyshould be easy, just diff and patch on -r 1541 or milestone-proposed22:06
pvook. will give it a go.22:06
*** dprince has quit IRC22:08
*** misheska has quit IRC22:09
vishymtaylor: I backported 3 branches to milestone-proposed but they aren't merging22:10
rmkAny thoughts as to why the dashboard refuses to show me the service panel since pulling the latest keystone?22:10
*** cole has joined #openstack-dev22:11
vishyrmk: probably lost the Admin role somehow22:11
rmkYeah, been going through my db and don't see the issue22:12
rmkAdmin role seems to be there and associated22:12
vishyrmk: keystone db?22:12
rmkyeah22:12
vishyrmk: maybe put some logging in the middleware in nova22:12
vishymake sure it is getting the role properly?22:13
vishyrmk: actually, hmm, I don't really know how dashboard checks for admin22:13
vishyrmk: to enable the panel22:13
rmkI'll have a look through the code22:13
rmkIt's odd, it was working prior to updating keystone22:13
*** code_franco has quit IRC22:18
vishymtaylor, jeblair: if you can figure out the magic, I'm trying to get all of the branches here to go in: https://code.launchpad.net/~hudson-openstack/nova/milestone-proposed/+activereviews .   Tarmac is rejecting everything right now22:18
vishymtaylor: jeblair: this one got kicked back with a similar error: https://code.launchpad.net/~smoser/nova/trunk.lp851145/+merge/7580022:20
vishyany help on those would be great22:20
rmkSo it's looking at the tokens it gets back from Keystone for a roleRef named Admin22:24
Davieyvishy: it's not yet published, but i expect it to be tomorrow.  Is it blocking you?22:29
vishyDaviey: no, just wondering when i can merge the req into the packaging22:30
vishyDaviey: I don't really know what to do with your suggestions to propose into ubuntu proper22:30
openstackgerritYogeshwar Srikrishnan proposed a change to openstack/identity-api: Changes to support extensions document.Also bringing cramers wadltodocbook changes.  https://review.openstack.org/46622:31
Davieyvishy: there is a trunk packaging branch, and now a diablo branch under ~openstack-ubuntu-packagers.22:31
Davieyyou only targetted towards one.22:32
vishyDaviey: gotcha k I'll prop to both with changesets22:32
vishylet me know when it is in22:32
rmkDoing a GET on http://localhost:5000/v2.0/tokens/8a5cdc19-c796-4617-bb99-cc217364dae0 (it's a valid token), should generate a response right?22:33
Davieyvishy: awesome, i'll merge it when it lands tommorrow, ok?22:33
*** stanchan has joined #openstack-dev22:34
vishyyup22:36
Davieycool22:36
*** jeffjapan has joined #openstack-dev22:41
*** misheska has joined #openstack-dev22:42
*** yogirackspace has left #openstack-dev22:49
*** rnirmal has quit IRC23:00
*** rnorwood has quit IRC23:03
*** hbetts has quit IRC23:04
rmkSeems like something broke with Keystone and the latest commits23:06
rmkroleRefs are always empty23:06
rmkor not23:09
rmkHowever, here's the json response -- "roleRefs": [{"roleId": "1", "id": "1"}], and here's what Dashboard is expecting ... if role['roleId'] == 'Admin'23:11
rmkDoesn't seem right23:12
*** kbringard has quit IRC23:12
openstackgerritJames E. Blair proposed a change to openstack/openstack-manuals: Add rfc.sh to help with gerrit workflow.  https://review.openstack.org/48523:37
*** wwkeyboard_ has quit IRC23:39
*** lorin1 has joined #openstack-dev23:39
*** lorin1 has left #openstack-dev23:39
rmkI can confirm this was a bug introduced with the most recent keystone commit23:41
*** joonwon has joined #openstack-dev23:41
*** bsza has quit IRC23:41
*** wwkeyboard has joined #openstack-dev23:49
*** wwkeyboard has quit IRC23:53
*** jakedahn_ has joined #openstack-dev23:55
*** jakedahn__ has joined #openstack-dev23:56
*** jakedahn_ has quit IRC23:56
*** jakedahn has quit IRC23:58
*** jakedahn__ is now known as jakedahn23:58
*** rnorwood has joined #openstack-dev23:59
« Sunday, 2011-09-18 Index Tuesday, 2011-09-20 »

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