| *** mahadev has quit IRC | 00:00 | |
| *** axelis has quit IRC | 00:00 | |
| *** dendro-afk is now known as dendrobates | 00:04 | |
| *** bkkrw has quit IRC | 00:06 | |
| *** rpedde has joined #openstack | 00:10 | |
| *** joearnol_ has joined #openstack | 00:21 | |
| *** azneita has joined #openstack | 00:21 | |
| *** azneita has joined #openstack | 00:21 | |
| *** pvo has joined #openstack | 00:23 | |
| *** joearnold has quit IRC | 00:23 | |
| *** bkkrw has joined #openstack | 00:25 | |
| *** bkkrw has quit IRC | 00:26 | |
| uvirtbot | New bug: #735836 in nova "euca-get-ajax-console raises unknown exception" [Undecided,New] https://launchpad.net/bugs/735836 | 00:27 |
|---|---|---|
| *** stewart has joined #openstack | 00:27 | |
| *** adiantum has quit IRC | 00:28 | |
| *** grapex has quit IRC | 00:29 | |
| *** pvo has quit IRC | 00:31 | |
| *** adiantum has joined #openstack | 00:33 | |
| *** reldan has joined #openstack | 00:33 | |
| *** maplebed has quit IRC | 00:34 | |
| *** dragondm has quit IRC | 00:35 | |
| *** clauden has quit IRC | 00:35 | |
| *** rpedde has quit IRC | 00:37 | |
| *** patcoll has joined #openstack | 00:48 | |
| *** KenD has quit IRC | 00:49 | |
| *** ewindisch has joined #openstack | 00:52 | |
| *** adiantum has quit IRC | 00:53 | |
| *** aliguori has quit IRC | 00:56 | |
| *** dmshelton has joined #openstack | 00:57 | |
| *** adiantum has joined #openstack | 00:59 | |
| *** reldan has quit IRC | 01:06 | |
| *** drico_ has quit IRC | 01:08 | |
| *** aliguori has joined #openstack | 01:11 | |
| *** adiantum has quit IRC | 01:13 | |
| *** Jbain has joined #openstack | 01:13 | |
| *** dendrobates is now known as dendro-afk | 01:15 | |
| *** adiantum has joined #openstack | 01:18 | |
| *** grapex has joined #openstack | 01:19 | |
| *** dendro-afk is now known as dendrobates | 01:21 | |
| *** pvo has joined #openstack | 01:28 | |
| *** adiantum has quit IRC | 01:29 | |
| *** adiantum has joined #openstack | 01:31 | |
| *** patcoll has quit IRC | 01:39 | |
| *** joearnol_ has quit IRC | 01:41 | |
| *** MotoMilind has quit IRC | 01:44 | |
| *** rchavik has joined #openstack | 01:48 | |
| *** nijaba has quit IRC | 01:52 | |
| *** jfluhmann has quit IRC | 01:52 | |
| *** littleidea has quit IRC | 01:54 | |
| *** nijaba has joined #openstack | 01:58 | |
| *** nijaba has joined #openstack | 01:58 | |
| *** kapil has joined #openstack | 02:36 | |
| *** aliguori has quit IRC | 02:39 | |
| *** rpedde has joined #openstack | 02:41 | |
| *** adiantum has quit IRC | 02:46 | |
| *** adiantum has joined #openstack | 02:52 | |
| *** paltman has joined #openstack | 03:04 | |
| *** j05h has quit IRC | 03:06 | |
| HugoKuo__ | morning | 03:08 |
| *** adiantum has quit IRC | 03:09 | |
| *** kashyap has joined #openstack | 03:11 | |
| *** winston-d has quit IRC | 03:12 | |
| *** adiantum has joined #openstack | 03:15 | |
| *** rpedde has quit IRC | 03:15 | |
| *** herki has quit IRC | 03:26 | |
| *** adiantum has quit IRC | 03:31 | |
| *** adjohn has quit IRC | 03:33 | |
| *** adjohn has joined #openstack | 03:34 | |
| *** adiantum has joined #openstack | 03:36 | |
| *** thielmann_ has joined #openstack | 03:46 | |
| *** adiantum has quit IRC | 03:48 | |
| *** thielmann__ has quit IRC | 03:49 | |
| *** adjohn has quit IRC | 03:51 | |
| *** adiantum has joined #openstack | 03:54 | |
| uvirtbot | New bug: #735898 in nova "get_user in adminclient is broken" [Medium,Confirmed] https://launchpad.net/bugs/735898 | 04:02 |
| *** littleidea has joined #openstack | 04:04 | |
| *** pvo has quit IRC | 04:05 | |
| *** pvo has joined #openstack | 04:08 | |
| *** burris has quit IRC | 04:11 | |
| *** koolhead17|afk is now known as koolhead17 | 04:19 | |
| *** burris has joined #openstack | 04:22 | |
| *** adiantum has quit IRC | 04:23 | |
| *** adiantum has joined #openstack | 04:30 | |
| koolhead17 | hey all | 04:30 |
| uvirtbot | New bug: #735907 in nova "Openstack API /flavors resource does not respect the specification" [Undecided,New] https://launchpad.net/bugs/735907 | 04:31 |
| *** grapex has quit IRC | 04:41 | |
| *** grapex has joined #openstack | 04:43 | |
| *** bcwaldon has quit IRC | 04:45 | |
| *** littleidea has quit IRC | 04:57 | |
| *** littleidea has joined #openstack | 04:58 | |
| *** hadrian has quit IRC | 05:06 | |
| *** jfluhmann has joined #openstack | 05:11 | |
| *** shrewduh has joined #openstack | 05:13 | |
| *** thielmann_ has quit IRC | 05:14 | |
| *** littleidea has quit IRC | 05:23 | |
| *** guynaor has joined #openstack | 05:24 | |
| *** guynaor has left #openstack | 05:24 | |
| *** shrewduh has quit IRC | 05:26 | |
| *** shrewduh has joined #openstack | 05:27 | |
| *** pvo has quit IRC | 05:30 | |
| *** adiantum has quit IRC | 05:32 | |
| *** littleidea has joined #openstack | 05:33 | |
| *** adiantum has joined #openstack | 05:36 | |
| *** f4m8_ is now known as f4m8 | 05:41 | |
| *** kapil has quit IRC | 06:01 | |
| *** adiantum has quit IRC | 06:03 | |
| *** kapil has joined #openstack | 06:05 | |
| *** shentonfreude has quit IRC | 06:07 | |
| *** noguchi has joined #openstack | 06:08 | |
| *** adiantum has joined #openstack | 06:09 | |
| *** cjx has joined #openstack | 06:13 | |
| HugoKuo__ | If I use a single box to install nova-network , I think dnsmasq is no need been install on that box . Am I right ? | 06:27 |
| *** adiantum has quit IRC | 06:31 | |
| *** jk0 has quit IRC | 06:33 | |
| *** rdw has quit IRC | 06:35 | |
| *** littleidea has quit IRC | 06:35 | |
| *** glenc has quit IRC | 06:35 | |
| *** colinnich has quit IRC | 06:35 | |
| *** sleepson- has quit IRC | 06:35 | |
| *** rdw has joined #openstack | 06:36 | |
| *** glenc has joined #openstack | 06:36 | |
| *** colinnich has joined #openstack | 06:36 | |
| *** sleepsonthefloor has joined #openstack | 06:36 | |
| *** jk0 has joined #openstack | 06:36 | |
| *** ChanServ sets mode: +v jk0 | 06:36 | |
| *** adiantum has joined #openstack | 06:37 | |
| *** tr3buchet has quit IRC | 06:37 | |
| *** tr3buchet has joined #openstack | 06:41 | |
| *** mgoldmann has joined #openstack | 06:42 | |
| *** paltman has quit IRC | 06:56 | |
| *** allsystemsarego has joined #openstack | 07:06 | |
| *** allsystemsarego has joined #openstack | 07:06 | |
| *** gaveen has joined #openstack | 07:07 | |
| HugoKuo__ | Current Status(2011.2.25): When you use Bexar release, nova-network should be installed into the same machine with nova-compute if network-manager is FlatDHCPManager. | 07:09 |
| HugoKuo__ | I saw this line in a doc | 07:09 |
| *** paltman has joined #openstack | 07:15 | |
| *** kashyap_ has joined #openstack | 07:15 | |
| *** mikhail_ has joined #openstack | 07:17 | |
| *** kashyap has quit IRC | 07:18 | |
| *** miclorb_ has quit IRC | 07:20 | |
| *** adjohn has joined #openstack | 07:21 | |
| *** mikhail_ has quit IRC | 07:32 | |
| *** huihui has joined #openstack | 07:42 | |
| *** CloudChris has joined #openstack | 07:42 | |
| *** grapex has quit IRC | 07:42 | |
| *** miclorb has joined #openstack | 07:44 | |
| *** ovidwu has quit IRC | 07:53 | |
| *** ovidwu has joined #openstack | 07:54 | |
| *** adiantum has quit IRC | 07:56 | |
| ttx | eday: ah! good catch. | 07:57 |
| ttx | willfix | 07:58 |
| *** adiantum has joined #openstack | 07:59 | |
| *** azneita has quit IRC | 08:06 | |
| HugoKuo__ | Current Status(2011.2.25): When you use Bexar release, nova-network should be installed into the same machine with nova-compute if network-manager is FlatDHCPManager | 08:10 |
| HugoKuo__ | is this bug still exist ? | 08:10 |
| HugoKuo__ | if I install nova by package (ubuntu 10.04) | 08:11 |
| ttx | HugoKuo__: last time I tried I followed https://lists.launchpad.net/openstack/msg00363.html | 08:14 |
| *** mikhail has joined #openstack | 08:14 | |
| *** mgoldmann_ has joined #openstack | 08:16 | |
| *** mgoldmann has quit IRC | 08:16 | |
| HugoKuo__ | ttx : thanks | 08:18 |
| *** mikhail has quit IRC | 08:23 | |
| *** adiantum has quit IRC | 08:24 | |
| HugoKuo__ | ttx: could you plz help me http://i217.photobucket.com/albums/cc280/tonytkdk/archi1.jpg | 08:26 |
| HugoKuo__ | is my architechture correct ? | 08:27 |
| HugoKuo__ | Jack : nova-api , mysql , nova-objectstore | 08:27 |
| HugoKuo__ | Absolut : nova-network nova-scheduler | 08:28 |
| HugoKuo__ | Dewars : nova-compute | 08:28 |
| HugoKuo__ | does it work ? actually it's a basic archi which i use to deploy Eucalyptus.... | 08:28 |
| HugoKuo__ | before | 08:28 |
| *** bkkrw has joined #openstack | 08:32 | |
| ttx | HugoKuo__: i'm lacking hands-on experience about nova-network position vs. the network model you choose. | 08:35 |
| HugoKuo__ | so what's the common archi everyone using ? | 08:36 |
| HugoKuo__ | thanks | 08:36 |
| HugoKuo__ | maybe I use wrong architecture :< | 08:37 |
| *** noguchi has left #openstack | 08:37 | |
| ttx | I'll let "everyone" answer. | 08:38 |
| HugoKuo__ | XD thanks | 08:38 |
| HugoKuo__ | I'm really confusing about how to design multiple host | 08:39 |
| *** Nacx has joined #openstack | 08:40 | |
| *** miclorb has quit IRC | 08:40 | |
| *** kalonike has joined #openstack | 08:48 | |
| *** herki has joined #openstack | 08:58 | |
| *** skiold has joined #openstack | 08:58 | |
| uvirtbot | New bug: #735974 in nova "nova-network logs exception when starting up" [Undecided,New] https://launchpad.net/bugs/735974 | 09:01 |
| *** huihui has quit IRC | 09:04 | |
| *** daveiw has joined #openstack | 09:08 | |
| *** drico has joined #openstack | 09:09 | |
| soren | HugoKuo__: What seems to be the problem? | 09:11 |
| HugoKuo__ | I don't know if my hardware archi is correct .... | 09:12 |
| HugoKuo__ | I try to deploy openstack like eucalyptus CLC+CC+NC | 09:12 |
| HugoKuo__ | but I'm not sure ....the standard archi of openstack-nova........ | 09:13 |
| soren | There is not standard architecture. | 09:13 |
| soren | s/not/no/ | 09:13 |
| soren | Look, you're asking this questino, because something is not working. What is it? | 09:14 |
| HugoKuo__ | my purpose is to seperate nova-network into a single box | 09:14 |
| soren | ok | 09:14 |
| HugoKuo__ | bcz I think all instances's traffic is go out from nova-network .... | 09:14 |
| HugoKuo__ | but I always got some problem ................ so I wounder know is there anyone success to deploy multi hosts | 09:16 |
| soren | Whatever problems you have, we should fix. | 09:17 |
| soren | So if you can explain what they are, we can try to understand them and fix them. | 09:17 |
| HugoKuo__ | ok..... | 09:17 |
| HugoKuo__ | I'll ask several questions ....thx | 09:17 |
| HugoKuo__ | 1. when I install nova-network into a single box | 09:18 |
| HugoKuo__ | should I install dnsmasq on it? | 09:18 |
| soren | Using FlatDHCPManager? | 09:19 |
| HugoKuo__ | assume I have 3 machines A: nova-api nova-objectstore mysql B:nova-network nova-scheduler C:nova-compute | 09:19 |
| HugoKuo__ | yup ..... | 09:19 |
| HugoKuo__ | FlatDHCPManager | 09:19 |
| soren | then yes, | 09:19 |
| soren | How are you installing nova? | 09:19 |
| *** irahgel has joined #openstack | 09:20 | |
| HugoKuo__ | on A : using https://github.com/dubsquared/OpenStack-NOVA-Installer-Script/ nova-CC-install-v1.1.sh but delete nova-networl and nova-scheduler and nova-compute packages | 09:21 |
| soren | Oh. | 09:21 |
| soren | That thing. | 09:21 |
| HugoKuo__ | on B : apt-get install python-mysqldb nova-netwok nova-scheduler dnsmasq | 09:21 |
| soren | Man,I don't know. That script does sooo many things. | 09:22 |
| HugoKuo__ | bcz .....I don't know what should be installed on each nodes .... | 09:24 |
| HugoKuo__ | any suggestion for my install methods? | 09:24 |
| soren | I always recommend using the Ubuntu packages. | 09:25 |
| ttx | soren: did you look into bug 735390 yet ? | 09:27 |
| uvirtbot | Launchpad bug 735390 in nova "Live migration breaks UML (and will break LXC)." [High,Confirmed] https://launchpad.net/bugs/735390 | 09:27 |
| ttx | Trying to see if we'll only get to pick one | 09:27 |
| soren | ttx: I did not. I'll take a look now. | 09:28 |
| ttx | soeren: cool, thanks | 09:28 |
| soren | ttx: Wrong soren. | 09:28 |
| ttx | autocomplete fail | 09:29 |
| soren | ttx: soeren is an imposter. Possibly of the golf playing kind. | 09:29 |
| ttx | that's the worse kind. | 09:29 |
| HugoKuo__ | http://pastebin.com/1gFf9YXr | 09:30 |
| HugoKuo__ | plz correct me .... if any error for my package | 09:30 |
| HugoKuo__ | I install from package :> | 09:30 |
| soren | HugoKuo__: You'll need python-mysqldb on host C, too. | 09:31 |
| HugoKuo__ | Ubuntu 10.04 | 09:31 |
| soren | HugoKuo__: ...and you'll need nova-objectstore on host A. | 09:31 |
| HugoKuo__ | got it........... | 09:31 |
| *** cjx has quit IRC | 09:32 | |
| HugoKuo__ | and then | 09:32 |
| HugoKuo__ | Current Status(2011.2.25): When you use Bexar release, nova-network should be installed into the same machine with nova-compute if network-manager is FlatDHCPManager. | 09:32 |
| soren | What is that? | 09:32 |
| HugoKuo__ | I saw this description in a doc | 09:33 |
| soren | Where do you see that? | 09:33 |
| HugoKuo__ | http://wiki.openstack.org/NovaInstall/NovaDeploymentTool | 09:33 |
| HugoKuo__ | overhere | 09:33 |
| soren | Don't know. | 09:33 |
| HugoKuo__ | and ttx shows me this https://lists.launchpad.net/openstack/msg00363.html | 09:33 |
| HugoKuo__ | I'm going to design a test for my 3 hosts archi..... | 09:34 |
| soren | I've not really used the flat network style much. | 09:35 |
| *** ramkrsna has joined #openstack | 09:36 | |
| *** ramkrsna has joined #openstack | 09:36 | |
| HugoKuo__ | so I try to draw a diagram http://s217.photobucket.com/albums/cc280/tonytkdk/?action=view¤t=archi1.jpg , I'm not sure if this approach correct | 09:36 |
| soren | I'm a black belt debugger, though. | 09:36 |
| HugoKuo__ | ok, I'll do some test | 09:36 |
| HugoKuo__ | then write some docs. | 09:36 |
| HugoKuo__ | all the examples or docs only talk about CC + many nova-compute nodes | 09:38 |
| HugoKuo__ | thanks soren ....forgive me always bother you | 09:38 |
| ttx | soren: does "bzr lp-propose" also push the branch, or assumes it's already been pushed ? | 09:40 |
| soren | ttx: It assumes it exists on launchpad, but it does make sure that it's up to date. | 09:41 |
| soren | ttx: So it starts by pushing, iirc. | 09:41 |
| ttx | let's try. | 09:41 |
| ttx | right. you must push it once first. | 09:42 |
| soren | Yeah, that defines the push branch for your tree. Without it, it won't know what to propose for merging. | 09:44 |
| *** reldan has joined #openstack | 09:55 | |
| bkkrw | HugoKuo__, what diagram drawing software are you using ? | 10:13 |
| HugoKuo__ | bkkrw : SmartDraw | 10:13 |
| bkkrw | HugoKuo__, windows :) | 10:14 |
| HugoKuo__ | bkkrw : yup my desktop is windows | 10:15 |
| bkkrw | HugoKuo__, have a hard time finding something with decent stencils in linux | 10:16 |
| *** adjohn_ has joined #openstack | 10:17 | |
| HugoKuo__ | bkkrw : I do think so :> | 10:17 |
| *** adjohn has quit IRC | 10:19 | |
| *** miclorb has joined #openstack | 10:23 | |
| ttx | soren: should I just close bug 712481 ? | 10:28 |
| uvirtbot | Launchpad bug 712481 in nova "nova testsuite fails in the buildds." [High,Incomplete] https://launchpad.net/bugs/712481 | 10:28 |
| *** reldan has quit IRC | 10:29 | |
| soren | ttx: No. | 10:30 |
| soren | ttx: Let me check up on something. | 10:30 |
| ttx | phew. Triaged all "New" bugs | 10:32 |
| soren | awesome! | 10:32 |
| ttx | Now onto checking the "Incomplete" ones. Which are usually a bit lame. | 10:32 |
| soren | ttx: The problem is just a single test case. It fails if the timing resolution on the host is too coarse. | 10:32 |
| soren | ttx: Hmm.. | 10:33 |
| soren | ttx: /me hacks something | 10:33 |
| *** beens is now known as ck | 10:39 | |
| *** ck is now known as Beens | 10:39 | |
| *** adjohn_ has quit IRC | 10:41 | |
| *** MarcMorata has joined #openstack | 10:53 | |
| soren | ttx: Ok, proposed a fix. | 11:04 |
| soren | ttx: It'll need a packaging adjustment once it lands. | 11:04 |
| ttx | ok, will update bug status accordingly | 11:04 |
| soren | ttx: I basically will just skip the timing sensitive test if run as part of package builds. | 11:04 |
| * ttx finished the update pass on "Incomplete" bugs | 11:04 | |
| *** reldan has joined #openstack | 11:13 | |
| *** azneita has joined #openstack | 11:20 | |
| soren | ttx: Awesome. | 11:20 |
| *** cloud0 has quit IRC | 11:30 | |
| *** cloud0 has joined #openstack | 11:30 | |
| *** uksysadmin has joined #openstack | 11:40 | |
| uksysadmin | hi guys | 11:40 |
| uksysadmin | hit a bit of a snag yesterday when I got my hands on some physical kit and went to fire up my first instance... | 11:40 |
| uksysadmin | it brought down both our datacentres | 11:40 |
| uksysadmin | pretty impressive if you ask me | 11:40 |
| uksysadmin | so... could do with some explanation of what happens when interfaces are set up, bridging, vlans etc/// | 11:41 |
| uksysadmin | essentially a STP storm happened | 11:42 |
| uksysadmin | the default vlan100 - what controls that? | 11:43 |
| *** miclorb has quit IRC | 11:49 | |
| *** azneita has quit IRC | 11:53 | |
| soren | uksysadmin: "controls"? | 11:53 |
| soren | uksysadmin: I'm not sure I understand the question. | 11:53 |
| soren | uksysadmin: I am sure, though, that it's way past lunch time. | 11:53 |
| uksysadmin | is there any way of changing this? | 11:53 |
| soren | Changing lunch time? | 11:54 |
| uksysadmin | huh?! lol | 11:54 |
| soren | It'll happen on its own in a few weeks (DST change and all that). | 11:54 |
| uksysadmin | ok - by default, nova creates vlan100 | 11:54 |
| uksysadmin | is there any way of overriding this? | 11:54 |
| *** daveiw has left #openstack | 11:55 | |
| uksysadmin | I think i had two issues - one, specifying my public_interface as an interface that was already vlanned | 11:55 |
| soren | uksysadmin: "overriding" as in "avoiding"? | 11:55 |
| uksysadmin | and using a vlan that is used elsewhere | 11:55 |
| uksysadmin | we use vlan tag 100 for something completely separate on our network | 11:55 |
| soren | uksysadmin: Ah, you want to use different vlan numbers? | 11:55 |
| uksysadmin | yeah | 11:55 |
| soren | uksysadmin: There's a flag called vlan_start. | 11:56 |
| soren | It defaults to 100. | 11:56 |
| uksysadmin | fingers are pointing to one server that i set up - about 1 min after I fired up an instance, we had a catastophic failure | 11:56 |
| soren | I'm sure the rest of what I'm about to say is obvious? | 11:56 |
| uksysadmin | one of the engineers noticed a lot of STP traffic | 11:56 |
| uksysadmin | yeah | 11:56 |
| soren | Cool. | 11:56 |
| * soren heads lunchwards | 11:56 | |
| uksysadmin | ta | 11:56 |
| soren | np | 11:56 |
| soren | Sorry we made your DC's blow up :) | 11:56 |
| uksysadmin | I'm 50/50 on it being a bad thing | 11:57 |
| soren | :) | 11:57 |
| uksysadmin | sure it was bad, but come on, one small 2 cpu box brought $$$$ of loss of revenue... surely that should be avoided | 11:57 |
| uksysadmin | (I place blame squarely on the network engineers to allow this kind of thing to happen) | 11:58 |
| kalonike | if i would like to install latest bexar release, should i use sudo add-apt-repository ppa:nova-core/release or sudo add-apt-repository ppa:nova-core/bexar ??? | 12:03 |
| kalonike | or something else? | 12:03 |
| soren | ppa:nova-core/release is meant to be the one. | 12:04 |
| kalonike | thanks | 12:04 |
| soren | It's not quite been updated to latest bexar yet. That's entirely my fault. I'll go and fix it now. | 12:04 |
| soren | ttx: The theory was that the package from your ppa should be good, right? | 12:04 |
| kalonike | and is there any extra package or patch wich i should use not described on wiki single machine package install page? | 12:05 |
| soren | Everything should be in the ppa. | 12:06 |
| soren | kalonike: I'll put the newest bexar release into the ppa after lunch. | 12:07 |
| * soren really, REALLY heads lunchwards | 12:07 | |
| kalonike | ok, then i wait | 12:07 |
| kalonike | have a good lunch dear :) | 12:07 |
| *** skiold has quit IRC | 12:15 | |
| *** dmshelton has quit IRC | 12:17 | |
| ttx | soren: that's the theory, yes. | 12:19 |
| *** h0cin has joined #openstack | 12:22 | |
| *** h0cin has joined #openstack | 12:22 | |
| *** adiantum has joined #openstack | 12:23 | |
| *** dmshelton has joined #openstack | 12:26 | |
| *** vish1 has joined #openstack | 12:27 | |
| *** kpepple_ has joined #openstack | 12:27 | |
| *** holoway_ has joined #openstack | 12:27 | |
| *** _morfeas has joined #openstack | 12:28 | |
| *** deshantm_ has joined #openstack | 12:28 | |
| *** zedas_ has joined #openstack | 12:29 | |
| *** kpepple has quit IRC | 12:29 | |
| *** holoway has quit IRC | 12:29 | |
| *** vishy has quit IRC | 12:29 | |
| *** kim0 has quit IRC | 12:29 | |
| *** zedas has quit IRC | 12:29 | |
| *** deshantm has quit IRC | 12:29 | |
| *** morfeas has quit IRC | 12:29 | |
| *** kim0 has joined #openstack | 12:29 | |
| *** kim0 has joined #openstack | 12:29 | |
| *** shentonfreude has joined #openstack | 12:30 | |
| soren | ttx: There's only a maverick package. | 12:33 |
| *** shentonfreude has quit IRC | 12:34 | |
| ttx | soren: indeed... | 12:35 |
| *** dmshelton has quit IRC | 12:35 | |
| ttx | soren: was the packaging for lucid/natty different ? i.e. could you use pocket source copy ? | 12:35 |
| ttx | or should I port the changes to lucid/natty as well | 12:36 |
| soren | ttx: I think I'll create a release-staging ppa under nova-core. | 12:39 |
| soren | ttx: ...and set it up like the trunk one, and upload the packages there. | 12:40 |
| soren | ttx: Using the same stuff as I use for the "real" PPA. | 12:40 |
| ttx | soren: I wouldn't be able to help with that one. Why not push it under ~openstack-ubuntu-packagers ? | 12:40 |
| soren | ttx: That would work, too. | 12:42 |
| soren | ttx: Just seems a bit odd, but technically it doesn't matter. | 12:42 |
| *** bcwaldon has joined #openstack | 12:43 | |
| soren | ttx: Maybe you could do it? I'm on the phone. | 12:43 |
| *** georgy has joined #openstack | 12:43 | |
| *** reldan has quit IRC | 12:44 | |
| ttx | soren: do the staging PPA and populate it ? Sure | 12:44 |
| ttx | in a bit | 12:44 |
| *** CloudChris has quit IRC | 12:45 | |
| soren | ttx: Not populate it, just set it up with the rigth dependencies and stuff. | 12:45 |
| *** lionel_ has quit IRC | 12:45 | |
| *** lionel_ has joined #openstack | 12:46 | |
| *** aliguori has joined #openstack | 12:47 | |
| ttx | soren: ok | 12:47 |
| zul | is there a way i can just test the LibvirtConnTestCase test cases | 12:55 |
| *** shentonfreude has joined #openstack | 12:56 | |
| *** shentonfreude has left #openstack | 12:56 | |
| *** daveiw has joined #openstack | 12:59 | |
| *** CloudChris has joined #openstack | 13:01 | |
| ttx | soren: looks like I can't, -ENOTADMIN | 13:05 |
| *** skiold has joined #openstack | 13:05 | |
| soren | ttx: Oh. | 13:06 |
| soren | zul: You can pass arguments to run_tests.sh exactly like nose itself. | 13:06 |
| *** georgy has quit IRC | 13:08 | |
| *** ivan has quit IRC | 13:08 | |
| *** ggy has joined #openstack | 13:08 | |
| *** ivan has joined #openstack | 13:09 | |
| *** ivan has joined #openstack | 13:09 | |
| ggy | Hi I am facing some issue publishing image in using uec-publish-tarball. I get unknown error. If I try using euca-upload-bundle, I get Permission denied while writing: kernel.manifest.xml | 13:11 |
| *** pvo has joined #openstack | 13:11 | |
| *** ppetraki has joined #openstack | 13:11 | |
| *** gondoi has joined #openstack | 13:15 | |
| *** piken has quit IRC | 13:21 | |
| *** adiantum has quit IRC | 13:24 | |
| *** adiantum has joined #openstack | 13:24 | |
| *** matclayton has joined #openstack | 13:33 | |
| *** matclayton has joined #openstack | 13:33 | |
| *** bcwaldon has quit IRC | 13:34 | |
| *** bcwaldon has joined #openstack | 13:34 | |
| *** hadrian has joined #openstack | 13:35 | |
| *** zykes- has joined #openstack | 13:37 | |
| zykes- | has the block been changed ? | 13:37 |
| *** herki has quit IRC | 13:37 | |
| *** dendrobates is now known as dendro-afk | 13:39 | |
| *** edconzel has joined #openstack | 13:42 | |
| *** reldan has joined #openstack | 13:43 | |
| *** ggy has quit IRC | 13:44 | |
| *** BK_man has joined #openstack | 13:44 | |
| ttx | eday: i meant "openstack-devel". Fixing wiki. | 13:45 |
| zykes- | ttx: would it be an idea to make a openstack iso ? | 13:46 |
| ttx | zykes-: I think that was done at one point | 13:46 |
| ttx | zul: ^? | 13:46 |
| ttx | annegentle: ^? | 13:47 |
| zul | hmm? | 13:47 |
| ttx | openstack livecd | 13:47 |
| ttx | zul: didn't kirkland produce one ? | 13:47 |
| zul | a what? | 13:47 |
| kalonike | shall i use libvirt 0.8.8 for nova? | 13:48 |
| ttx | a nova liveCD demo iso | 13:48 |
| zul | ttx: i dont think he got very far | 13:48 |
| *** deshantm_ has quit IRC | 13:48 | |
| zykes- | ttx: more like a install cd | 13:48 |
| kalonike | i have problems with libvirtd restart | 13:48 |
| *** deshantm has joined #openstack | 13:48 | |
| zul | ttx: should be easy to do with lxc containers though ;) | 13:48 |
| kalonike | first it says pid already exist, after i remove, i got errors | 13:48 |
| *** shrewduh has quit IRC | 13:49 | |
| kalonike | internal error Child process /sbin/iptables | 13:49 |
| kalonike | could anyone help? | 13:49 |
| zul | ttx: its something me might want to consider at the ubuntu uds | 13:49 |
| *** ggy has joined #openstack | 13:51 | |
| ggy | Hi I am facing some issue publishing image in using uec-publish-tarball. I get unknown error. If I try using euca-upload-bundle, I get Permission denied while writing: kernel.manifest.xml. If I sudo the same , I get "EC2_ACCESS_KEY environment variable must be set." even though it is already set | 13:53 |
| *** koolhead17 has quit IRC | 13:53 | |
| uksysadmin | zul: I was mulling over how to do an install/demo cd of sorts | 13:56 |
| uksysadmin | got to the stage of whether its worth the effort | 13:56 |
| *** uksysadmin has quit IRC | 13:58 | |
| soren | ttx: I *think* we have automatically built packages of bexar now. They land in ppa:nova-core/proposed. | 13:58 |
| soren | ttx: end-to-end automatic. tarmac->tests->tarball->source packages->launchpad. | 13:59 |
| soren | We'll know in a couple of minutes | 14:00 |
| *** uksysadmin has joined #openstack | 14:04 | |
| soren | ttx: Oh, craptastic. | 14:06 |
| soren | ttx: I extended the ppa build script to allow for a specific packaging revision to be passed in. | 14:06 |
| soren | ttx: ..so that I could make sure we used the same packaging code as we did for the original bexar release. | 14:07 |
| *** rpedde has joined #openstack | 14:07 | |
| soren | ttx: ... but we grew a dependency since then (python-distutils-extra). | 14:07 |
| *** Ryan_Lane has joined #openstack | 14:07 | |
| * soren headdesks and stops being an idiot. | 14:07 | |
| *** deshantm has quit IRC | 14:08 | |
| *** patcoll has joined #openstack | 14:09 | |
| *** rnirmal has joined #openstack | 14:10 | |
| *** ewindisch_ has joined #openstack | 14:14 | |
| *** Ryan_Lane has quit IRC | 14:15 | |
| *** dendro-afk is now known as dendrobates | 14:17 | |
| kalonike | i am getting confused of many versions, descriptions | 14:18 |
| *** ewindisch has quit IRC | 14:18 | |
| *** ewindisch_ is now known as ewindisch | 14:18 | |
| *** edconzel_ has joined #openstack | 14:18 | |
| kalonike | i would like to install nova so soo much | 14:18 |
| *** viirya has quit IRC | 14:19 | |
| *** viirya has joined #openstack | 14:20 | |
| *** drico has quit IRC | 14:20 | |
| *** edconzel has quit IRC | 14:20 | |
| *** edconzel_ is now known as edconzel | 14:20 | |
| *** omidhdl has joined #openstack | 14:20 | |
| *** drico has joined #openstack | 14:21 | |
| *** omidhdl1 has joined #openstack | 14:22 | |
| *** omidhdl has quit IRC | 14:23 | |
| *** msassak has joined #openstack | 14:23 | |
| kalonike | could anyone help? I cant restart libvirtd :( | 14:26 |
| uksysadmin | what's up kalonike? you can't really expect much help from the terse info provided... | 14:27 |
| ggy | kalonike: what error did you get? | 14:27 |
| annegentle | ttx: zul: yep, kirkland has one in progress, and the latest work (for me) was to add swift to it, it's in lp:openstack-live | 14:27 |
| *** edconzel has quit IRC | 14:28 | |
| annegentle | uksysadmin: I've been using your instructions to improve the virtualbox image, yesterday I was trying to add the openstack-dashboard project to the image as well, but I'm not sure how the networking would work. As in, what IP address represents the dashboard? The eth0 one or the virbr0? | 14:29 |
| uksysadmin | aweseome annegentle - might just check that out | 14:30 |
| uksysadmin | you need to point the dashboard to the address of the api server... | 14:30 |
| annegentle | uksysadmin: and is it normal to have to keep restarting networking? | 14:30 |
| kalonike | i have an error message that the pid file already exist | 14:31 |
| uksysadmin | shouldn't have to do that | 14:31 |
| kalonike | after i delete | 14:31 |
| kalonike | it gives error message like this: | 14:31 |
| annegentle | uksysadmin: ok, so is the nova-api running on the eth0 or virbr0 NAT? | 14:31 |
| kalonike | internal error Child process /sbin/iptables | 14:31 |
| kalonike | i am using version 0.8.8 | 14:31 |
| *** gaveen has quit IRC | 14:31 | |
| annegentle | uksysadmin: (and I'm terrible at networking concepts, so please forgive my ignorance, I'm learning though) :) | 14:31 |
| uksysadmin | annegentle: from what I remember the instructions (although a little back and forth) worked, where the django dashboard just ran on port 8000 by default | 14:32 |
| uksysadmin | annegentle: don't remember anything about interfaces | 14:32 |
| uksysadmin | annegentle, I'm staying away from openstack due to a glitch yesterday in work that doesn't help with the whole evangelising of the stuff lol :-S | 14:32 |
| annegentle | uksysadmin: that's true. but I can't put http://localhost:8000 in my browser or I get my mac localhost, right? So... I'm feeling over virtualized :) | 14:33 |
| annegentle | uksysadmin: aw bummer! | 14:33 |
| uksysadmin | annegentle, indeed. I put blame squarely on the network engineers to allow a STP issue to take down two remote datacentres just because I launched an instance on nova... | 14:34 |
| *** lwollney1 has joined #openstack | 14:35 | |
| uksysadmin | annegentle, I presume you followed these instructions for the dashboard: http://wiki.openstack.org/OpenStackDashboard | 14:35 |
| uksysadmin | essentially you configure that to point to your API (so all your credentials go in a local_settings.py that you have in your environment)... then when you spin it up, say listening on port 8000 you should be able to point your web browser to the ip/address:8000 and you can log in | 14:37 |
| *** ewindisch_ has joined #openstack | 14:38 | |
| *** ewindisch has quit IRC | 14:38 | |
| *** ewindisch_ is now known as ewindisch | 14:38 | |
| *** KevinC has joined #openstack | 14:39 | |
| *** daveiw has left #openstack | 14:39 | |
| annegentle | uksysadmin: yep. what I'm not sure of is "what's the ip/address" - neither the IP on the eth0 (when I run ifconfig) or the virbr0 pulls up the dashboard. | 14:42 |
| annegentle | uksysadmin: as in, I put in http://192.168.122.1:8000 and the browser just sits there. | 14:42 |
| annegentle | uksysadmin: ok, and a big "eek" for your network engineers. | 14:43 |
| *** KevinC has quit IRC | 14:43 | |
| uksysadmin | and you have run: tools/with_venv.sh dashboard/manage.py runserver 0.0.0.0:8000 | 14:43 |
| *** jano_ has joined #openstack | 14:43 | |
| *** gaveen has joined #openstack | 14:44 | |
| *** mark has quit IRC | 14:45 | |
| *** omidhdl1 has quit IRC | 14:45 | |
| *** omidhdl has joined #openstack | 14:45 | |
| *** mark has joined #openstack | 14:46 | |
| *** f4m8 is now known as f4m8_ | 14:46 | |
| *** drico has quit IRC | 14:50 | |
| *** paltman has quit IRC | 14:50 | |
| *** paltman has joined #openstack | 14:50 | |
| *** drico has joined #openstack | 14:50 | |
| *** gondoi has quit IRC | 14:53 | |
| *** ramkrsna has quit IRC | 14:53 | |
| *** paltman has quit IRC | 14:53 | |
| *** paltman has joined #openstack | 14:53 | |
| *** dragondm has joined #openstack | 14:54 | |
| *** reldan has quit IRC | 14:55 | |
| *** dfg has joined #openstack | 15:00 | |
| *** gondoi has joined #openstack | 15:01 | |
| *** reldan has joined #openstack | 15:03 | |
| jaypipes | is it a problem when I can't tell the difference between my *actual* assigned review day and any other day? | 15:06 |
| blamar | jaypipes, your reviews are extremely appreciated by the community! </suckup> :) | 15:09 |
| blamar | jaypipes, actually, now that I think about it we can't get rid of XML because it's hard to do things like </sarcasm> in JSON :( | 15:10 |
| jaypipes | blamar: lol :) | 15:11 |
| creiht | blamar: } //sarcasm | 15:11 |
| creiht | :) | 15:11 |
| jaypipes | creiht: lol2. | 15:11 |
| *** lvaughn_ has quit IRC | 15:13 | |
| *** ewindisch_ has joined #openstack | 15:14 | |
| *** lvaughn has joined #openstack | 15:15 | |
| *** ewindisch has quit IRC | 15:16 | |
| *** ewindisch_ is now known as ewindisch | 15:16 | |
| *** clauden_ has joined #openstack | 15:17 | |
| *** nelson has joined #openstack | 15:17 | |
| *** kashyap_ has quit IRC | 15:18 | |
| *** sparkycollier has joined #openstack | 15:20 | |
| markwash | jaypipes: if you get a chance, can you let me know if my fixes based on your review are sufficient? https://code.launchpad.net/~markwash/nova/lp727225/+merge/52754 | 15:20 |
| *** jimc has joined #openstack | 15:21 | |
| jaypipes | markwash: yes sir! got a couple ahead of you, but should be done in less than an hour. (other two are littler bug fixes) | 15:22 |
| *** kapil has quit IRC | 15:22 | |
| *** imsplitbit has joined #openstack | 15:26 | |
| *** lvaughn_ has joined #openstack | 15:29 | |
| *** lvaughn has quit IRC | 15:32 | |
| *** kashyap_ has joined #openstack | 15:35 | |
| *** MarcMorata has quit IRC | 15:39 | |
| *** MarcMorata has joined #openstack | 15:41 | |
| *** omidhdl has left #openstack | 15:44 | |
| creiht | eday, jaypipes, mtaylor: congrats on the drizzle release! | 15:49 |
| jaypipes | creiht: nothing to do with me. :) | 15:50 |
| creiht | hehe | 15:50 |
| creiht | jaypipes: you don't work on it anymore? | 15:50 |
| *** jimc has left #openstack | 15:50 | |
| jaypipes | creiht: heh, not since last April. | 15:51 |
| creiht | ahh | 15:51 |
| ggy | Hi I am facing some issue publishing image in using uec-publish-tarball. I get unknown error. If I try using euca-upload-bundle, I get Permission denied while writing: kernel.manifest.xml. If I sudo the same , I get "EC2_ACCESS_KEY environment variable must be set." even though it is already set | 15:52 |
| markwash | ggy: did you try with sudo -E ? just a guess. . :-) | 15:53 |
| ggy | then I get : Warning: failed to parse error message from AWS: http://www.w3.org/TR/html4/strict.dtd:81:4: error in processing external entity reference Permission denied while writing: kernel.manifest.xml | 15:54 |
| jaypipes | creiht: heya, got a sec to talk swift auth and glance? Looks like, even though I *thought* I was fixing two bugs with the latest Swift store code in Glance, I may not have fixed this bug: https://bugs.launchpad.net/glance/+bug/717431 | 15:55 |
| uvirtbot | Launchpad bug 717431 in glance "Swift backend not taking into account new Swift account:user:key in URI" [High,In progress] | 15:55 |
| *** rchavik has quit IRC | 15:55 | |
| creiht | jaypipes: sure | 15:56 |
| jaypipes | creiht: talking to marcelo in a private channel (btorch) and he's having some issues when he sets his swift_store_user to "external:marcelo" | 15:56 |
| creiht | hrm | 15:56 |
| ggy | markwash : I tried it then I got :Warning: failed to parse error message from AWS: http://www.w3.org/TR/html4/strict.dtd:81:4: error in processing external entity reference Permission denied while writing: kernel.manifest.xml | 15:56 |
| *** clauden_ has quit IRC | 15:56 | |
| jaypipes | creiht: he's mentioning that "with devauth or swauth the account must be part of the username" | 15:56 |
| creiht | right | 15:57 |
| jaypipes | creiht: does that mean that the auth_url is no longer like this: http://localhost:9090/v1.0/account ? | 15:57 |
| creiht | I don't think that was ever the case? | 15:58 |
| markwash | ggy: sorry, I was just guessing based on troubles I've had with sudo and environment variables in the past | 15:58 |
| creiht | the auth url was http://localhost:9090/v1.0/ with the account and key sent in headers | 15:58 |
| creiht | or /auth/v1.0 if using swauth | 15:58 |
| jaypipes | creiht: the *user* and *key*, not the account, right? | 15:58 |
| ggy | <markwash> thanks anyway :) | 15:58 |
| creiht | jaypipes: where user is account:user | 15:59 |
| creiht | curl -v -H 'X-Storage-User: test:tester' -H 'X-Storage-Pass: testing' http://127.0.0.1:8080/auth/v1.0 | 16:00 |
| creiht | (from the saio page) | 16:00 |
| creiht | is an example auth request | 16:00 |
| jaypipes | creiht: ah, ok. | 16:00 |
| kalonike | i have installed nova finally, thanks all who helped | 16:01 |
| jaypipes | creiht: sorry, yet again, I was confusing the concept of the auth URL with the storage management URl returned by the auth service... sorry :( | 16:01 |
| creiht | heh... no worries | 16:01 |
| *** Pentheus has quit IRC | 16:03 | |
| *** paltman has quit IRC | 16:03 | |
| ggy | clear | 16:06 |
| *** shrewduh has joined #openstack | 16:06 | |
| kalonike | my instances remain in scheduling state | 16:06 |
| kalonike | instead of launching | 16:06 |
| kalonike | does anyone know the reason?? | 16:06 |
| *** shrewduh has quit IRC | 16:07 | |
| creiht | jaypipes: something that we also discussed here is that a better solution down the road would be to urlencode the login information to prevent parsing issues (like having a : in the password) | 16:09 |
| *** bcwaldon has quit IRC | 16:09 | |
| jaypipes | creiht: sure. though in the meantime, I'll fix parse_swift_tokens() to take into account this new user format. pun intended ;) | 16:10 |
| creiht | heh | 16:10 |
| *** zedas_ is now known as zedas | 16:10 | |
| btorch | kalonike: I have had that happened to me a couple of times in the first trials of getting nova (trunk) setup up, have you checked the /var/log/nova logs ? | 16:14 |
| *** MarcMorata has joined #openstack | 16:14 | |
| *** headoff has quit IRC | 16:15 | |
| *** jfluhmann has quit IRC | 16:16 | |
| kalonike | ops, not yet, thanks :) | 16:17 |
| kalonike | i installed nova-core/release | 16:17 |
| *** franck has joined #openstack | 16:17 | |
| uksysadmin | kalonike, assuming ubuntu, you running under vbox or something? def check those logs first, but also run kvm-ok command to see if hw virt supported... then check your libvirt_type in nova.conf | 16:17 |
| *** kapil has joined #openstack | 16:19 | |
| kalonike | my cpu supports hw virt | 16:20 |
| kalonike | i did not choose the libvirt type in nova conf | 16:20 |
| kalonike | i am running in normal hw, no virtual | 16:20 |
| kalonike | ubuntu 10.10 server x64 | 16:21 |
| tr3buchet | anyone know what the "device" parameter passed to a xenserver VIF.create does? | 16:21 |
| uksysadmin | what do those logs say? | 16:22 |
| *** daveiw has joined #openstack | 16:23 | |
| franck | hi, anyone know how can i bind storage manager and swift proxy storage with xcp? | 16:23 |
| kalonike | log : /usr/lib/pymodules/python2.6/MySQLdb/connection.py, line 35, in defaulterrorhandler raise errorclass, errorvalue | 16:24 |
| kalonike | when i describe my images, both of my uploaded images listed twice | 16:26 |
| *** MotoMilind has joined #openstack | 16:27 | |
| kalonike | which one should i choose? two different emi, which one should i use to run? | 16:27 |
| kalonike | shall i confirm the libvirt type in nova.conf?? | 16:30 |
| kalonike | and restart services? | 16:30 |
| *** kbringard has joined #openstack | 16:31 | |
| *** jimc has joined #openstack | 16:31 | |
| *** ggy has quit IRC | 16:32 | |
| franck | I'd like to use swift as a storage manager of xcp. I heard glance maybe help me... Can anyone tell me more | 16:33 |
| *** kashyap_ has quit IRC | 16:34 | |
| * ttx grumbles. | 16:35 | |
| kbringard | I recall seeing it at some point, btu can't seem to find it again... does anyone know where the documentation for creating an EBS out of a raw disk image is? | 16:35 |
| mtaylor | creiht: thanks! | 16:36 |
| kalonike | btorch, how u solved it finally to get ur instances out of scheduling state ? | 16:36 |
| uksysadmin | kalonike, did you just grep for "error" in /var/log/nova/* as that one line doesn't say much | 16:37 |
| uksysadmin | have a look in /var/log/nova/nova-scheduler.log /var/log/nova/nova-compute.log etc | 16:37 |
| btorch | kalonike: I was trying to go over my logs to see if I could find the error the log have already rotated | 16:38 |
| uksysadmin | reckon it could be a number of things - but the logs should help | 16:38 |
| MotoMilind | I created an RHEL image using directions on this WiKi: http://wiki.openstack.org/CreatingRHELImages The image shows up on Dashboard just fine and Dashboard allows me to create an instance just fine. But the instance creation fails with nova/virt/images.py:_fetch_s3_image:utils.execute(cmd_out), because the curl command fails. Any pointers on how I should debug this? Seems to me that the registration of the image to object store is somehow incomple | 16:39 |
| kalonike | ProcessExecutionError: Unexpected error while running commanf | 16:39 |
| btorch | kalonike: in regards to the ami . you should choose the one for the machine image http://paste.openstack.org/show/900/ | 16:39 |
| *** uksysadmin has quit IRC | 16:40 | |
| kalonike | thanks | 16:40 |
| btorch | kalonike: can you paste the full error for us on http://paste.openstack.org | 16:40 |
| kalonike | yes | 16:41 |
| kalonike | but i have this 2 lines very very many times | 16:41 |
| btorch | kalonike: also this is just a suggestion ... when I was trouble shooting nova logs I would always stop everything, clear my nova logs and then run 1-3 commands to minimize stuff in the logs | 16:42 |
| btorch | kalonike: so in your case I would just try to describe the images available and then run euca-run-instances to start the image .. and pay attention to what log files are changing | 16:43 |
| jaypipes | btorch: pls bzr branch lp:~jaypipes/glance/bug717431 and let me know if the Swift account:user bug is fixed with that branch. Thanks! | 16:46 |
| kalonike | paste #901 | 16:46 |
| kalonike | ok, i try, thanks btorch | 16:46 |
| jaypipes | soren, vish1: if you have some free time, I'd appreciate a review on specifically the networking code in the Guest Tool in this patch: https://code.launchpad.net/~citrix-openstack/nova/vmware-vsphere-support/+merge/53060. Much appreciated! | 16:47 |
| jaypipes | markwash: ok, onto your review now :) | 16:48 |
| *** kashyap_ has joined #openstack | 16:50 | |
| kbringard | has anyone had any luck getting a nova-volume to work using just an lo mounted disk image? | 16:51 |
| *** joearnold has joined #openstack | 17:03 | |
| btorch | jaypipes: hmm I'm trying to build it but it's failing http://paste.openstack.org/show/903/ | 17:04 |
| btorch | jaypipes: it works fine for the bzr87 that I built ... ignore the bzr89 I've added ... that was just to make my debbuilding easier for me | 17:05 |
| *** irahgel has quit IRC | 17:05 | |
| annegentle | anyone in nova-core want an easy karma grab? Merge in my change to the front page of the nova.openstack.org site: https://code.launchpad.net/~annegentle/nova/versiontalkbubble/+merge/53105 | 17:06 |
| btorch | kalonike: u sure you have the mysql portion of nova setup correctly and the proper credentials in the nova.conf ? | 17:07 |
| btorch | kalonike: does something like nova-manage user list work ? | 17:07 |
| *** bcwaldon has joined #openstack | 17:08 | |
| *** irahgel has joined #openstack | 17:08 | |
| *** paltman has joined #openstack | 17:09 | |
| *** sparkycollier has quit IRC | 17:10 | |
| *** skiold has quit IRC | 17:11 | |
| btorch | jaypipes: nm | 17:11 |
| franck | how can i use glance as a sm with xcp and swift | 17:12 |
| openstackjenkins | Project nova build #646: SUCCESS in 2 min 2 sec: http://hudson.openstack.org/job/nova/646/ | 17:13 |
| openstackjenkins | Tarmac: added new class Instances for managaging instances | 17:13 |
| openstackjenkins | added new method list in class Instances: | 17:13 |
| openstackjenkins | # nova-manage instance list | 17:13 |
| openstackjenkins | instance node type state launched image kernel ramdisk project user zone index | 17:13 |
| openstackjenkins | i-00000547 XXXXXXX m1.small running 2011-02-18 08:36:37 ami-a03ndz0q ami-0isqekvw testing berendt None 0 | 17:13 |
| openstackjenkins | i-00000548 XXXXXXX m1.small running 2011-02-18 08:37:17 ami-a03ndz0q ami-0isqekvw testing berendt None 1 | 17:13 |
| openstackjenkins | i-00000549 XXXXXXX m1.small running 2011-02-18 08:37:52 ami-a03ndz0q ami-0isqekvw testing berendt None 2 | 17:13 |
| openstackjenkins | # nova-manage instance list ares | 17:13 |
| openstackjenkins | instance node type state launched image kernel ramdisk project user zone index | 17:13 |
| openstackjenkins | i-00000c1c ares m1.tiny running 2011-02-26 22:51:40 ami-pus9dj84 ami-zhcv0yyx ami-av96fu30 testing berendt None 1 | 17:13 |
| openstackjenkins | extended the method list in class FixedIps to lookup ip addresses assigned to a specified node: | 17:13 |
| openstackjenkins | # nova-manage fixed list ares | 17:13 |
| openstackjenkins | network IP address MAC address hostname host | 17:13 |
| openstackjenkins | 192.168.3.0/24 192.168.3.6 02:16:3e:75:d7:9ai-00000c1c ares | 17:13 |
| *** daveiw has left #openstack | 17:16 | |
| jaypipes | btorch: figured it out? working? | 17:22 |
| blamar | i18n question: _("error: %s" % error_msg) or _("error: %s") % error_msg ... I see it both ways in the code, does it matter? | 17:23 |
| jaypipes | franck: sorry, what do you mean by "an sm with xcp and swift"? | 17:23 |
| jaypipes | blamar: the latter is correct. the former is incorrect and should be corrected. | 17:23 |
| btorch | jaypipes: yeah go the debs built and testing out now | 17:24 |
| blamar | jaypipes: ty | 17:24 |
| jaypipes | blamar: np :) | 17:24 |
| jaypipes | btorch: cool, lemme know the results | 17:24 |
| *** paltman has quit IRC | 17:24 | |
| *** paltman has joined #openstack | 17:25 | |
| *** jimc has quit IRC | 17:25 | |
| *** kapil has quit IRC | 17:26 | |
| *** bkkrw has quit IRC | 17:29 | |
| btorch | jaypipes: http://paste.openstack.org/show/906/ | 17:29 |
| *** smeier00 has joined #openstack | 17:31 | |
| *** axelis has joined #openstack | 17:32 | |
| *** reldan has quit IRC | 17:32 | |
| *** joearnold has quit IRC | 17:33 | |
| jaypipes | btorch: use is_public=True and see if that updates the image #1 properly... | 17:34 |
| *** smeier00 has quit IRC | 17:35 | |
| *** zenmatt has quit IRC | 17:35 | |
| *** smeier001 has joined #openstack | 17:35 | |
| *** joearnold has joined #openstack | 17:35 | |
| btorch | jaypipes: nope | 17:35 |
| btorch | jaypipes: gonna go grab a sub and brb | 17:37 |
| jaypipes | btorch: ok. | 17:37 |
| markwash | it looks like the local image service does not return create_time and update_time metadata--does that sound right to folks? | 17:38 |
| jaypipes | markwash: wouldn't be surprised. | 17:38 |
| *** ggy has joined #openstack | 17:41 | |
| *** eday has quit IRC | 17:41 | |
| *** eday has joined #openstack | 17:41 | |
| *** ChanServ sets mode: +v eday | 17:41 | |
| *** MarkAtwood has quit IRC | 17:41 | |
| ggy | Hi I am facing a small problem. I was behind a proxy when I setup my controller node. But I am facing issues when I try to publish an image because of proxy problems. So I tried changin my proxy connection to direct. But still somehow the request gets sent to the proxy... I am confused | 17:46 |
| *** shentonfreude has joined #openstack | 17:48 | |
| *** MarcMorata has quit IRC | 17:48 | |
| *** dprince has joined #openstack | 17:49 | |
| jk0 | soren: mind taking a quick looksee at https://code.launchpad.net/~jk0/nova/lp733232/+merge/53522 ? | 17:53 |
| btorch | jaypipes: ok back | 17:53 |
| *** axelis has quit IRC | 17:54 | |
| jaypipes | btorch: OK, so that looks like a bug with glance update command... is_public=true should make the image public, but it's not doing that. pls file a bug? | 17:55 |
| jaypipes | btorch: but regarding the other bug (account:user:key one), can you test that is working properly? It *looks* like it is.. | 17:55 |
| creiht | jaypipes: you know it isn't that difficult to set up your own saio :) | 17:56 |
| btorch | jaypipes: yep , I haven't changed anything on my glance.conf file and just installed the new deb pkgs built from the glance-bug717431 code. I no longer get the clear error due to (account:user:key ) issue .. I think we are good :) | 17:59 |
| jaypipes | creiht: yeah? last I checked there's more than 20 steps. | 17:59 |
| jaypipes | creiht: not to be poop, but yes, it is difficult. | 18:00 |
| *** reldan has joined #openstack | 18:00 | |
| *** Nacx has quit IRC | 18:01 | |
| jaypipes | creiht: not that I can't do it, I'm just being lazy and relying on btorch since he already has it setup ;) | 18:01 |
| *** gaveen has quit IRC | 18:03 | |
| *** clauden has joined #openstack | 18:05 | |
| *** ggy has quit IRC | 18:06 | |
| *** smeier001 has quit IRC | 18:07 | |
| jaypipes | eday: https://code.launchpad.net/~markwash/nova/lp727225/+merge/52754. pls re-review and set to approved if good. thx. | 18:07 |
| jaypipes | markwash: excellent work on additional tests and fixes. nice job. | 18:08 |
| *** ramd has quit IRC | 18:08 | |
| *** ggy has joined #openstack | 18:08 | |
| *** ccustine has joined #openstack | 18:09 | |
| eday | jaypipes: approved | 18:10 |
| markwash | jaypipes: thanks guys | 18:11 |
| markwash | eday: ^^^ | 18:11 |
| eday | markwash: thank you! :) | 18:11 |
| *** lwollney has joined #openstack | 18:12 | |
| jaypipes | sirp_: around? | 18:12 |
| *** lwollney1 has quit IRC | 18:13 | |
| btorch | jaypipes: ok I'll create the ticket .. the delete also seems to be having issues | 18:13 |
| *** dsockwell has quit IRC | 18:13 | |
| *** dsockwell has joined #openstack | 18:14 | |
| jaypipes | btorch: please do! | 18:15 |
| *** irahgel has quit IRC | 18:17 | |
| *** oldbam has joined #openstack | 18:22 | |
| BK_man | jaypipes: quick question on glance: how to use --uid & --gid options? | 18:22 |
| jaypipes | BK_man: those options are no longer valid. | 18:23 |
| BK_man | jaypipes: ok. what is a way to run glance-* as non-root? | 18:23 |
| BK_man | jaypipes: need a way to run glance under non-root user & able to query services status - TODO for glance-control | 18:26 |
| jaypipes | BK_man: are you using glance-control to run the servers? | 18:26 |
| jaypipes | BK_man: yeah, it's a TODO... :( | 18:26 |
| BK_man | jaypipes: now trying to decide - will it be better to use glance-control inside initscript or to fire services (two at the moment) by myself | 18:26 |
| jaypipes | BK_man: the preferred usage is glance-control all start | 18:27 |
| BK_man | jaypipes: also didn't found a way to specify log-dir, log-file ... | 18:27 |
| jaypipes | BK_man: logging config (and docs) are queued up: https://code.launchpad.net/~jaypipes/glance/bug720816/+merge/52617 | 18:27 |
| btorch | jaypipes: https://bugs.launchpad.net/glance/+bug/736295 | 18:31 |
| uvirtbot | Launchpad bug 736295 in glance "Unable to set image metadata to Public=Trueing image " [Undecided,New] | 18:31 |
| jaypipes | btorch: cheers | 18:31 |
| *** kashyap_ has quit IRC | 18:32 | |
| btorch | jaypipes: nevermind about the delete .. I think my glance db was messed up so I recreated everything from scratch | 18:32 |
| jaypipes | btorch: k | 18:32 |
| btorch | jaypipes: it delete the image from swift just fine | 18:32 |
| openstackjenkins | Project nova build #647: SUCCESS in 2 min 5 sec: http://hudson.openstack.org/job/nova/647/ | 18:35 |
| openstackjenkins | Tarmac: Fixes euca-get-ajax-console returning Unknown Error, by using the correct | 18:35 |
| openstackjenkins | exception in get_open_port() logic. Patch from Tushar Patil. | 18:35 |
| jaypipes | btorch: yeah, it looked like the object remaining in Swift after you glance clear'd was the non-public image #1. (glance clear only deletes public images) | 18:35 |
| kbringard | hey guys, I have a question about floating ips | 18:35 |
| kbringard | I'm not entirely sure what I'm supposed to put for the public_interface | 18:36 |
| kbringard | it looks like it defaults to vlan1 | 18:36 |
| kbringard | but, is it supposed to be the physical interface on my machines that has external connectivity? | 18:36 |
| kbringard | or, rather, on my network controller | 18:37 |
| kbringard | the way I assumed it to work is | 18:37 |
| jaypipes | BK_man: log config now in trunk. | 18:37 |
| kbringard | the IP is allocated to the interface on the network controller, and iptables rules are setup to forward traffic to the internal IP of the instance | 18:37 |
| BK_man | jaypipes: thanks, will try it now | 18:38 |
| jaypipes | justinsb: ping. | 18:39 |
| *** jiboumans has joined #openstack | 18:40 | |
| jiboumans | soren, ttx, are you guys likely to head the celery way for messaging? http://wiki.openstack.org/UseCeleryAsQueueManager | 18:40 |
| pvo | jiboumans: sandywalsh was looking at that a while ago as well. | 18:40 |
| jiboumans | pvo: remember any strong pros or cons? or have an rtfm pointer to it? | 18:41 |
| pvo | would have to ask sandy. sandywalsh: poke | 18:41 |
| justinsb | jaypipes: ack | 18:41 |
| uvirtbot | New bug: #736295 in glance "Unable to set image metadata to Public=True" [Undecided,New] https://launchpad.net/bugs/736295 | 18:42 |
| jaypipes | justinsb: not sure how you managed to commit with a bad LP username... but because http://bazaar.launchpad.net/~justin-fathomdb/nova/bug732204/revision/805 was committed as justinsb instead of justinsb-fathomdb, we're having issues passing the authors test. | 18:42 |
| jaypipes | justinsb: I'm going to reject that merge prop and clean it up locally and push it. | 18:42 |
| justinsb | jaypipes: Thanks | 18:42 |
| jaypipes | justinsb: np :) | 18:42 |
| justinsb | jaypipes: I was trying to commit from inside my IDE | 18:43 |
| justinsb | jaypipes: I decided that was a bad idea | 18:43 |
| jaypipes | justinsb: heh :) | 18:43 |
| justinsb | jaypipes: No idea how it did it though... it was just calling the standard bazaar tools | 18:43 |
| justinsb | jaypipes: So it must have _actively_ screwed things up :-) | 18:43 |
| jaypipes | justinsb: nm :) one of these days, we'll get you sorted. | 18:44 |
| justinsb | jaypipes: You'll have to pry my IDE out of my cold dead hands :-) | 18:44 |
| ironcamel | eday: are you around? | 18:45 |
| eday | ironcamel: I am. | 18:45 |
| *** j05h has joined #openstack | 18:45 | |
| ironcamel | eday: i am working on a branch (based off of the one you reviewed) which implements versioned Controllers | 18:46 |
| BK_man | jaypipes: glance-control: error: no such option: --log-config | 18:46 |
| BK_man | jaypipes: this is on bzr88 | 18:46 |
| ironcamel | it is for supporting v1.1 OS api create server operations | 18:46 |
| BK_man | jaypipes: is my cmdline correct? glance-control --log-config=/etc/glance/logging.cnf api start | 18:46 |
| jaypipes | BK_man: goes in the config file. | 18:47 |
| *** smeier00 has joined #openstack | 18:47 | |
| eday | ironcamel: cool | 18:47 |
| ironcamel | eday: should i merge that into the original branch? or would you like to look at them separately? | 18:47 |
| jaypipes | BK_man: put log_config in your glance.conf... | 18:47 |
| jaypipes | BK_man: damn you for testing code paths we don't really use ;) | 18:48 |
| eday | ironcamel: I'd probably prefer seeing them both, as I'd like to sww how views look once controllers are versioned, but whatever works for you | 18:48 |
| openstackjenkins | Project nova build #648: SUCCESS in 2 min 3 sec: http://hudson.openstack.org/job/nova/648/ | 18:48 |
| openstackjenkins | Tarmac: Don't generate insecure passwords where it's easy to use urandom instead | 18:48 |
| ironcamel | eday: we would like the branch you reviewed to be approved asap, so probably it would be better if i left the 2 branches separate, and you can look at the second one after you are done with the first one | 18:49 |
| sandywalsh | pvo, jiboumans yes, celery would definitely be a great change in nova. I think jaypipes put a bp in somewhere for it? | 18:50 |
| MotoMilind | Turns out that the info.json for the image I created states that "imageState": "decrypting", | 18:50 |
| MotoMilind | And that is preventing me from running a curl for the image when images.py tries to fetch it from s3 storage! | 18:50 |
| sandywalsh | pvo, jiboumans https://blueprints.launchpad.net/nova/+spec/bexar-message-queue-celery | 18:50 |
| *** MarkAtwood has joined #openstack | 18:51 | |
| BK_man | jaypipes: /etc/glance/glance.conf - is this paste.deploy file? | 18:51 |
| pvo | sandywalsh: I know you looked at a few weeks back. jiboumans was asking what your thoughts were on it. | 18:52 |
| jaypipes | BK_man: yes, but its a regular conf file, too... | 18:52 |
| bcwaldon | jaypipes: is the api versioning merge prop on your todo list today? | 18:52 |
| *** joearnold has quit IRC | 18:52 | |
| *** joearnold has joined #openstack | 18:53 | |
| sandywalsh | jiboumans, was there something specific you were wondering about celery? | 18:53 |
| eday | ironcamel: I approved, just need another nova-core | 18:54 |
| eday | jaypipes: you on duty today, right? :) https://code.edge.launchpad.net/~rackspace-titan/nova/openstack-api-version-split/+merge/53510 | 18:54 |
| jaypipes | bcwaldon, eday: yes. | 18:54 |
| ironcamel | eday: thanks! | 18:54 |
| bcwaldon | jaypipes: great. Just checking :) | 18:54 |
| jaypipes | bcwaldon: already done a crap-ton (technical term) of reviews today... pushing through the backlog. | 18:55 |
| bcwaldon | jaypipes: Understandable. Just trying to plan what to work on over here | 18:55 |
| *** MotoMilind has quit IRC | 18:57 | |
| *** iammartian has joined #openstack | 18:59 | |
| *** MarkAtwood has quit IRC | 19:00 | |
| *** MarcMorata has joined #openstack | 19:01 | |
| jiboumans | sandywalsh: i'm evaluating a queueing solution for internal use at my company. Celery is looking promising and as i saw it mentioned as a possible openstack choice I was wondering about your thoughts on maturity, support, and pros/cons compared to others you've looked at. | 19:01 |
| openstackjenkins | Project nova build #649: SUCCESS in 2 min 3 sec: http://hudson.openstack.org/job/nova/649/ | 19:04 |
| openstackjenkins | * Tarmac: Keypairs are not required in the OpenStack API; don't require them! | 19:04 |
| openstackjenkins | This cleans up some weirdness from the original branch (https://code.launchpad.net/~justin-fathomdb/nova/bug732204/+merge/53359) with justin's LP username in r804. | 19:04 |
| openstackjenkins | * Tarmac: Add missing fallback chain for ipv6. | 19:04 |
| justinsb | eday, dabo, sandywalsh : Let's try talking here instead of on the ML. How many zones are we targeting? | 19:10 |
| kim0 | Hi folks, I'd like to announce Ubuntu Cloud Days IRC event in a week. If anyone would like to host an openstack session, that would be great | 19:15 |
| dabo | I prefer the ML for design discussions that change previously agreed-upon designs. That way things aren't known only to those logged in at a particular moment. | 19:15 |
| dabo | That said, there are no limits to zones - they are simply logical groupings | 19:16 |
| openstackjenkins | Project swift build #222: SUCCESS in 28 sec: http://hudson.openstack.org/job/swift/222/ | 19:16 |
| openstackjenkins | Tarmac: Fix for incorrect use of urlparse results in swauth bins | 19:16 |
| dabo | They can be nested to any degree | 19:17 |
| eday | justinsb: if we start thinking hybrid and peering with other providers, potentially quite a fe :) | 19:17 |
| markwash | eday: I had to fix some conflicts in my branch from the "keypairs optional" merge | 19:17 |
| markwash | jaypipes: ^^^ | 19:17 |
| jaypipes | markwash: k, did you push again? | 19:19 |
| markwash | jaypipes: I did | 19:19 |
| justinsb | dabo: Sure, they're abstract, but if they're abstract we shouldn't impose our caching mindset upon them :-) | 19:19 |
| dabo | they're not abstract | 19:20 |
| justinsb | dabo: So what are they? | 19:20 |
| justinsb | eday: Hybrid & peering are v interesting | 19:21 |
| dabo | As I said: logical groupings. The logic behind the grouping is up to the one doing the deployment | 19:21 |
| justinsb | eday: I think if we're going to design that, we should acknowledge that up-front | 19:21 |
| dabo | But once deployed, they are actual devices with their own APIs. | 19:21 |
| justinsb | eday: I'm all for those ideas, but I don't know if sandy's work is actually compatible | 19:21 |
| justinsb | dabo: If it's a logical grouping, that doesn't sound very non-abstract to me | 19:22 |
| justinsb | dabo: Non-Abstract is: "a rack" or "a datacenter" | 19:22 |
| justinsb | dabo: Otherwise how do we know that zones are the right boundaries at which to cache | 19:22 |
| dabo | I'm not going to waste time debating semantics | 19:22 |
| eday | justinsb: It's not all the way there, but I think what he has done so far is good. | 19:23 |
| justinsb | eday: I think it's great | 19:23 |
| justinsb | dabo: It's not semantics | 19:23 |
| justinsb | eday: I just mean that if we're talking hybrid / peering then that means e.g. no message queues | 19:24 |
| dabo | A zone *can* be a rack or a datacenter or a group of racks or a group of datacenters or ... | 19:24 |
| justinsb | eday: At least not as I currently think of message queues | 19:24 |
| eday | justinsb: that's why his zone work is http based | 19:24 |
| dabo | Saying that one type of zone is abstract while another type isn't doesn't add to the discussion | 19:24 |
| justinsb | eday: Great ... I thought he'd started to move away from that, but I'll read his latest code | 19:24 |
| justinsb | I guess the real question: why do we have multiple zones at all? What is the problem that the complication of multiple zones is addressing? | 19:25 |
| *** ggy has quit IRC | 19:25 | |
| justinsb | Then we can avoid arguing about what a zone is | 19:25 |
| eday | justinsb: partitions of resources (availability, organizations, ...) | 19:26 |
| justinsb | eday: OK, but for what purpose? | 19:26 |
| jaypipes | markwash: k, give me a few to get to it | 19:26 |
| jaypipes | dprince: interesting name of your demo extension :) | 19:27 |
| justinsb | I see it in the hybrid / peering sense | 19:27 |
| justinsb | But I didn't see that in the blueprint | 19:27 |
| eday | justinsb: so I can run my own zone, you cna run yours, and they can talk to each other? (peering). or run a zone per dc for availability separation (don't share a db/queue/... between dcs, ...) | 19:27 |
| justinsb | (Off to dig up blueprint) | 19:27 |
| dprince | jaypipes: Yeah. I was trying to slip that in there somewhere. :) | 19:27 |
| jaypipes | dprince: :) | 19:28 |
| *** nelson has quit IRC | 19:31 | |
| *** nelson has joined #openstack | 19:31 | |
| jaypipes | sirp_: hey, could you check out http://launchpadlibrarian.net/66520847/buildlog_ubuntu-lucid-i386.glance_2011.2~bzr89-0ubuntu0ppa1~lucid1_FAILEDTOBUILD.txt.gz please? Ever since the change to make Swift lazy-loaded, Glance hasn't been able to be built in the PPA... | 19:32 |
| justinsb | eday, dabo: Ah ... here's the money shot from the blueprint: "A Huddle Zone is limited in size to 200 Hosts due to networking restrictions." | 19:33 |
| *** ggy has joined #openstack | 19:33 | |
| justinsb | eday, dabo: So that says we have a huge number of zones | 19:33 |
| openstackjenkins | Project nova build #650: SUCCESS in 2 min 3 sec: http://hudson.openstack.org/job/nova/650/ | 19:33 |
| openstackjenkins | Tarmac: fixes bug 735298: start of nova-compute not possible because of wrong xml paths to the //host/cpu section in "virsh capabilities", used in nova/virt/libvirt_conn.py | 19:33 |
| uvirtbot | Launchpad bug 735298 in nova "can't start nova-compute: Invalid: Invalid xml. '<cpu>' must be 1,but 5" [High,In progress] https://launchpad.net/bugs/735298 | 19:33 |
| sirp_ | jaypipes: k | 19:33 |
| dabo | justinsb: That's also a Rackspace-specific limitation | 19:33 |
| justinsb | But why is a zone limited to 200 hosts? | 19:33 |
| jaypipes | sirp_: thx mate | 19:33 |
| eday | justinsb: that's for one deployment, it's not for all | 19:33 |
| justinsb | eday, dabo: Well, that's what is motivating the zones blueprint, as far as I can tell | 19:34 |
| dabo | IIRC, switch limitations on MAC addresses | 19:34 |
| eday | justinsb: that's not the only thing though, I think other deployments will use it for resource separation as well | 19:35 |
| dabo | justinsb: not every example that was discussed at the summits made it into the blueprints | 19:35 |
| eday | justinsb: RS needs to for hardware limitations, others my for hardware, political, geographic, security, ... | 19:35 |
| justinsb | eday, dabo: Never mind, the point is moot. The agreed upon blueprint says caching is out of scope | 19:35 |
| uvirtbot | New bug: #736343 in nova "Incorrect mapping of instance type id to flavor id in Openstack API" [Undecided,New] https://launchpad.net/bugs/736343 | 19:36 |
| eday | justinsb: well, I would disagree with that. we need to make sure the zone work will support the type of caching we want to do :) | 19:36 |
| justinsb | eday: I agree that the blueprint system has its limitations. | 19:37 |
| justinsb | eday, dabo: Oh well, I'll think about this over lunch! | 19:39 |
| eday | justinsb: I don't think it's a blueprint issue, more of a people issue. :) We just need to be holistic in our discussion and planning. | 19:40 |
| eday | a blueprint should never be expected to be comprehensive or be a limit for discussion | 19:41 |
| *** reldan has quit IRC | 19:41 | |
| justinsb | Another topic for the design summit, I think | 19:41 |
| justinsb | I hope you'll get behind me when I suggest that view | 19:41 |
| *** ggy has quit IRC | 19:42 | |
| justinsb | As I think others believe that they are proscriptive | 19:42 |
| *** MotoMilind has joined #openstack | 19:42 | |
| eday | I'll be behind any view I think is correct at the time, and I don't see that one changing :) (taking holistic approaches) | 19:42 |
| justinsb | eday: :-) I'd like to see more of that pragmatism generally! | 19:43 |
| eday | it's a balance of course | 19:43 |
| openstackjenkins | Project nova build #651: SUCCESS in 2 min 2 sec: http://hudson.openstack.org/job/nova/651/ | 19:43 |
| openstackjenkins | Tarmac: Fix lp727225 by adding support for personality files to the openstack api. | 19:43 |
| openstackjenkins | Description: | 19:43 |
| openstackjenkins | This merge adds support for personality files to the openstack api. It leverages previous work which added this functionality to the compute api, compute manager, and xen api. | 19:43 |
| justinsb | eday: OK, lunch for real this time! | 19:43 |
| *** ggy has joined #openstack | 19:44 | |
| *** adiantum has quit IRC | 19:46 | |
| *** MotoMilind has quit IRC | 19:56 | |
| *** axelis has joined #openstack | 19:57 | |
| *** ggy has quit IRC | 19:59 | |
| *** nelson has quit IRC | 20:00 | |
| *** matclayton has left #openstack | 20:01 | |
| dabo | Updated glance, and am getting the error "TypeError: add_log_options() takes exactly 1 argument (2 given)" (full traceback: http://paste.openstack.org/show/907/) | 20:02 |
| dabo | anyone know what that could be? | 20:02 |
| devcamcar | jaypipes: are you around? i have a pypi question for you | 20:02 |
| jaypipes | devcamcar: shoot. | 20:06 |
| *** MotoMilind has joined #openstack | 20:06 | |
| devcamcar | jaypipes: hey, so i'm putting the adminclient from nova in a separate project so it can be packaged properly | 20:07 |
| devcamcar | jaypipes: i did a setup.py sdist upload but when i pip install it something is going wrong somewhere. it's not including the actual .py files | 20:08 |
| devcamcar | jaypipes: http://bazaar.launchpad.net/~devcamcar/nova-adminclient/trunk/files | 20:08 |
| *** mahadev has joined #openstack | 20:08 | |
| devcamcar | jaypipes: so if you have a sec to look at my setup.py i'd appreciate it! i compared it to glance's and it's very similar. not sure where i'm going wrong | 20:08 |
| *** MotoMilind has joined #openstack | 20:08 | |
| eday | devcamcar: just curious, is the rest of nova too much of a dependency for adminclient? | 20:09 |
| *** kbringard has quit IRC | 20:09 | |
| MotoMilind | Could someone possibly help with imagestate = decrypting issue? The images I create and register seem to be stuck in that state forever. Thanks. | 20:09 |
| jaypipes | devcamcar: did you try doing: python setup.py sdist upload -s | 20:10 |
| devcamcar | eday: i proposed a branch to reorganize the adminclient within nova but had a lot of people say that having multiple setup.py's was bad for packaging and recommended i move it to its own project | 20:11 |
| devcamcar | eday: only things in nova dependent on it are the smoketests, and i can fix that with a pip-requires if i get my pypi issues resolved | 20:12 |
| devcamcar | jaypipes: what does -s do? | 20:12 |
| jaypipes | devcamcar: not sure, but that's what mtaylor always told me to do ;) | 20:12 |
| devcamcar | jaypipes: --sign (-s) sign files to upload using gpg | 20:13 |
| *** mgoldmann_ has quit IRC | 20:14 | |
| devcamcar | jaypipes: doesn't look like that would affect the structure of the package | 20:14 |
| jaypipes | devcamcar: ah, guess not. :) to be honest, mtaylor is the guy to ask... I don't know much about it at all. | 20:14 |
| MotoMilind | Found an old issue where vishy suggested to manually copy the image to the image store and update the info.json | 20:14 |
| MotoMilind | I am trying that now | 20:15 |
| mtaylor | devcamcar, jaypipes: what's the question? | 20:15 |
| devcamcar | mtaylor: hey, i'm having a prob registering a project with pypi. i can register it fine (or so it seems) but when i pip install it, it only installs the .egg_info stuff and not the actual package | 20:16 |
| devcamcar | mtaylor: the bits are here: http://bazaar.launchpad.net/~devcamcar/nova-adminclient/trunk/files | 20:16 |
| *** fabiand_ has joined #openstack | 20:17 | |
| devcamcar | mtaylor: first time registering a project on pypi so i'm sure i'm just missing something | 20:17 |
| mtaylor | devcamcar: looking | 20:19 |
| devcamcar | thanks | 20:19 |
| *** ggy has joined #openstack | 20:20 | |
| MotoMilind | Hmm, just changing the value imagestate = available in info.json made it worse | 20:21 |
| MotoMilind | Ok, be right back | 20:22 |
| mtaylor | devcamcar: pip is still installing version 0.1.2 | 20:22 |
| eday | devcamcar: yeah, I was talking with mtaylor about it (and in general) yesterday. curious why you didn't want to just keep it in nova? was installing all of nova too much for places where you only wanted adminclient? | 20:23 |
| mtaylor | devcamcar: oh - you've only done an sdist upload I thinkg | 20:23 |
| mtaylor | devcamcar: do: | 20:23 |
| jaypipes | bcwaldon: are you familiar with how to do a merge prop where there is a branch that is a pre-requisite? | 20:23 |
| mtaylor | devcamcar: python setup.py sdist bdist bdist_egg upload | 20:24 |
| jaypipes | bcwaldon: I saw your comment: "And the reason I haven't merge prop'd it yet is that it depends on another merge prop: | 20:24 |
| jaypipes | https://code.launchpad.net/~rackspace-titan/nova/openstack-api-version-split/+merge/53510" | 20:24 |
| bcwaldon | jaypipes: I know there is a way, but I couldn't find the button when I went to merge prop it | 20:26 |
| bcwaldon | jaypipes: point me in the right direction? | 20:26 |
| jk0 | can I get two core devs to look at https://code.launchpad.net/~jk0/nova/lp733232/+merge/53522 ? it's just adding logging to a function | 20:26 |
| jaypipes | bcwaldon: sure thing. when you do the merge prop, just click the "More options" thingie on the merge prop page. There is a place to put in a branch under "This branch depends on another branch"... | 20:27 |
| jaypipes | bcwaldon: once you fill in the pre-requisite branch, then the diff generated by the merge prop will not contain any changes in the dependent branch. | 20:28 |
| jaypipes | jk0: looking. | 20:28 |
| bcwaldon | jaypipes: thanks. I'm writing the merge prop right now | 20:29 |
| creiht | mtaylor: it seems the ppa auto builds are failing, but not sure why | 20:29 |
| mtaylor | creiht: I blame jaypipes | 20:30 |
| creiht | for swift | 20:30 |
| creiht | heh | 20:30 |
| creiht | http://launchpadlibrarian.net/66522900/buildlog_ubuntu-lucid-i386.swift_1.3-dev%2Bbzr249-0ubuntu0ppa1~lucid1_FAILEDTOBUILD.txt.gz | 20:30 |
| *** Ryan_Lane has joined #openstack | 20:30 | |
| creiht | We thought we had removed that bin from everywhere in the debian branch | 20:31 |
| dabo | jaypipes: rev 90 of glance isn't working for me. sirp_ suggested reverting to r86, and it's working now. See http://paste.openstack.org/show/907/ for the traceback I was getting | 20:31 |
| jaypipes | dabo: we're removing glance-combined... | 20:32 |
| jaypipes | dabo: just use glance-control all start | 20:33 |
| dabo | jaypipes: oh, that's right... a new week, a new way to run glance... :) | 20:33 |
| devcamcar | mtaylor: thanks! i'm not sure how i would have ever just figured out that i needed to do that :) | 20:33 |
| jaypipes | dabo: a new week, a new complaint from dabo. ;P | 20:33 |
| dabo | jaypipes: nah, that's every few minutes | 20:34 |
| pvo | get off my lawn! | 20:34 |
| jaypipes | dabo: FYI, I respond best to bug reports ;) | 20:34 |
| dabo | jaypipes: so that's why you create so many bugs! :-P | 20:35 |
| jaypipes | dabo: indeed :) | 20:35 |
| jaypipes | dabo: feel free to file one on that, btw | 20:35 |
| *** dprince has quit IRC | 20:35 | |
| jaypipes | dabo: just instead of "glance doesn't work" as the description, put "glance-combined doesn't work. remove it or make it work." | 20:36 |
| jk0 | jaypipes / pvo thanks for the reviewz | 20:36 |
| jaypipes | jk0: course. | 20:36 |
| dabo | jaypipes: So you prefer something more specific than "jaypipes hosed things again!" | 20:36 |
| * jaypipes slaps sandywalsh for starting a long ML thread on datastores during my review day! | 20:36 | |
| jaypipes | dabo: no, that's fine too :) | 20:37 |
| devcamcar | mtaylor: so i did a python setup.py sdist bdist bdist_egg upload | 20:37 |
| devcamcar | mtaylor: and then pip install'd but it's even stranger now: http://paste.openstack.org/show/908/ | 20:37 |
| ggy | Hi I need some help ; I was able to run an instance but I am unable to ping or ssh onto it; I had used euca-authorize to authorize icmp and tcp; | 20:37 |
| devcamcar | mtaylor: only thing i see in the build dir from pip install is | 20:38 |
| devcamcar | $ ls | 20:38 |
| devcamcar | Users pip-egg-info | 20:38 |
| devcamcar | mtaylor: does pypi use the MANIFEST.in or is that only for deb packages? | 20:38 |
| *** fabiand_ has left #openstack | 20:39 | |
| justinsb | jaypipes: I don't need that __name__=='main' line at the bottom of my unit tests? Should I do a separate MP where I delete them from every unit test? Otherwise they'll keep being copied-and-pasted back... | 20:40 |
| creiht | justinsb: the 'main' part can still be useful if you want to run a single set of tests without a runner | 20:42 |
| *** kapil has joined #openstack | 20:43 | |
| openstackjenkins | Project nova build #652: SUCCESS in 2 min 2 sec: http://hudson.openstack.org/job/nova/652/ | 20:43 |
| openstackjenkins | Tarmac: Log the use of utils.synchronized. | 20:43 |
| *** reldan has joined #openstack | 20:44 | |
| markwash | creiht: is there a reason to not just use ./run_tests.sh module.path.to.test ? | 20:44 |
| *** bkkrw has joined #openstack | 20:44 | |
| creiht | markwash: I'm just saying that it is the standard python unit test way | 20:44 |
| creiht | it gives you both worlds | 20:44 |
| *** MarkAtwood has joined #openstack | 20:45 | |
| creiht | markwash: but yeah that works as well | 20:45 |
| markwash | creiht: agreed, I just was thinking maybe it is better if we're all testing in the same way, or if maybe some feature needed to be added to run_tests.sh | 20:45 |
| dabo | jaypipes: just for you: https://bugs.launchpad.net/glance/+bug/736395 | 20:46 |
| uvirtbot | Launchpad bug 736395 in glance "glance-combined throws a TypeError when run" [Undecided,New] | 20:46 |
| creiht | I'm not really passionate either way, just giving about the only reason that I can think of keeping them there | 20:46 |
| justinsb | creiht, markwash, jaypipes: I'm going to leave it as-is then. Thanks! | 20:48 |
| *** ggy_ has joined #openstack | 20:52 | |
| *** ggy has quit IRC | 20:53 | |
| *** georgy has joined #openstack | 20:54 | |
| *** ggy_ has quit IRC | 20:57 | |
| justinsb | Who's in charge of Jenkins? I'd like to get the PyLint job so that e.g. it doesn't complain about every "_" method | 20:57 |
| *** nelson has joined #openstack | 20:57 | |
| eday | justinsb: you probably just want to modify .pylintrc | 21:01 |
| eday | justinsb: and propose for merge | 21:01 |
| mtaylor | devcamcar: MANIFEST.in is used to help build the source tarball | 21:01 |
| justinsb | eday: Cool... I'll check whether that works on my own Jenkins install and then propose it | 21:02 |
| *** h0cin has quit IRC | 21:02 | |
| uvirtbot | New bug: #736395 in glance "glance-combined throws a TypeError when run" [Undecided,New] https://launchpad.net/bugs/736395 | 21:02 |
| eday | justinsb: don't need jenkins, just run pulint with the .pylintrc in lp:nova | 21:02 |
| eday | err, pylint | 21:02 |
| *** georgy has quit IRC | 21:03 | |
| justinsb | eday: OK... though _I_ need jenkins. Otherwise I'd have to run tests by hand! | 21:03 |
| justinsb | eday: I love my tools :-) | 21:04 |
| *** Ryan_Lane has quit IRC | 21:04 | |
| *** mahadev has quit IRC | 21:05 | |
| *** mahadev has joined #openstack | 21:06 | |
| devcamcar | mtaylor: when you have a sec can you pip install nova-adminclient and see if it obvious what is wrong with the package? | 21:08 |
| mtaylor | devcamcar: yup. poking at it | 21:08 |
| devcamcar | mtaylor: awesome, thanks | 21:08 |
| *** sparkycollier has joined #openstack | 21:10 | |
| dabo | In the Instance class in sqlachemy, does the 'user_id' represent the instance's "owner"? Is this what would be tied to an account for a provider like Rackspace, or would that be the project_id? | 21:17 |
| *** ghort has joined #openstack | 21:21 | |
| *** sparkycollier has quit IRC | 21:21 | |
| *** ghort2 has joined #openstack | 21:21 | |
| *** ghort has quit IRC | 21:21 | |
| *** ghort2 is now known as ghort | 21:22 | |
| eday | dabo: choose one :) those are going away with new auth and will just have a single 'owner' (probably diablo) | 21:22 |
| dabo | eday: yeah, I know that it's going away - just wondering which would transition better when that happens | 21:23 |
| justinsb | blamar: I'm going to look at fixing up pylintrc so that it doesn't report quite so much stuff. | 21:24 |
| eday | dabo: well, whatever is consuming it will also need to change to use 'owner', so probably doesn't matter which. | 21:24 |
| blamar | justinsb, much appreciated. looking forward to it :) | 21:24 |
| markwash | anyone else not in love with passing "context" variables to image services? | 21:24 |
| dabo | eday: thanks for the sanity check | 21:25 |
| *** ghort has left #openstack | 21:25 | |
| eday | dabo: I know dragondm was using 'project' as 'account' as his last patch | 21:25 |
| *** jdevesa has joined #openstack | 21:26 | |
| *** Tushar has joined #openstack | 21:28 | |
| Tushar | I am facing strange problem, I couldn't see console.log of an vm instance using euca-get-console-output command. What could be the cause? | 21:30 |
| Tushar | also in the syslog, I see "nbd15: unknown partition table" error | 21:30 |
| nelson | from over in #wikimedia-tech: neilk_: thank you! very excited to use Swift | 21:30 |
| creiht | :) | 21:31 |
| creiht | nelson: how goes it? | 21:31 |
| nelson | Going well. We already have an abstracted file system in the mediawiki php code. I need to write a Swift version of it. | 21:31 |
| Tushar | and libvirtd give warning "warning : lxcStartup:2128 : Unable to create cgroup for driver: No such device or address" | 21:31 |
| nelson | Will go looking for existing PHP Swift code to see if it can be reused (technically and legally). | 21:32 |
| nelson | BTW, as of April 1st, I'm off the Open Source Initiative board (term limits). | 21:32 |
| *** bkkrw has quit IRC | 21:33 | |
| *** rnirmal has quit IRC | 21:36 | |
| *** ggy has joined #openstack | 21:36 | |
| ggy | hi I was able to get an instance running but I am unable to ping or ssh on to it | 21:38 |
| *** imsplitbit has quit IRC | 21:39 | |
| ggy | I am using FlatManager | 21:39 |
| *** littleidea has joined #openstack | 21:39 | |
| *** irahgel has joined #openstack | 21:40 | |
| Tushar | ggy: I am also facing same issue. I see the instance is running but could ping, ssh and even couldn't see the console output. | 21:40 |
| Tushar | ggy: couldn't ping, ssh and see cnosole output | 21:41 |
| ggy | TUshar: seems we have the same problem. | 21:42 |
| ggy | Tushar: seems we have the same problem. | 21:42 |
| *** dfg_ has joined #openstack | 21:43 | |
| *** dfg has quit IRC | 21:43 | |
| Tushar | I am struggling to solve this problem since 2 days. | 21:43 |
| ggy | Tushar: what console output do you get? | 21:43 |
| Tushar | ggy: Do you see same errors in the syslog? | 21:44 |
| Tushar | root@ubuntu-openstack-single-server-02:/home/tpatil# euca-get-console-output i-0000000f | 21:44 |
| *** dfg__ has joined #openstack | 21:44 | |
| Tushar | ggy: just the instance id along with the timestamp. Also I tried to use virsh console <domain id>, but it give me error " internal error character device (null) is not using a PTY" | 21:45 |
| ggy | Tushar : same here... And seems I dont have any errors in the syslog. | 21:46 |
| *** dfg_ has quit IRC | 21:48 | |
| *** bkkrw has joined #openstack | 21:48 | |
| *** shentonfreude has quit IRC | 21:49 | |
| ggy | Tushar: Posted a question at. https://answers.launchpad.net/nova/+question/149347 Could you update it if you find a solution? | 21:51 |
| Tushar | ggy: sure. Can you tell me how did you created bridge? | 21:52 |
| *** ppetraki has quit IRC | 21:52 | |
| ggy | I followed the instructions in http://wiki.openstack.org/NovaInstall/MultipleServer | 21:53 |
| Tushar | brctl addbr br100 | 21:53 |
| Tushar | brctl addif br100 eth0 | 21:53 |
| Tushar | ifconfig br100 10.0.0.1 broadcast 10.0.0.0 netmask 255.255.255.0 up | 21:53 |
| Tushar | ifconfig eth0 0.0.0.0 up | 21:53 |
| *** patcoll has quit IRC | 21:53 | |
| ggy | thats pretty much the same I guess! | 21:54 |
| *** ggy has quit IRC | 21:56 | |
| *** smeier00 has quit IRC | 21:56 | |
| *** smeier00 has joined #openstack | 22:00 | |
| *** pvo has quit IRC | 22:05 | |
| *** bcwaldon has quit IRC | 22:07 | |
| *** reldan has quit IRC | 22:08 | |
| *** joearnol_ has joined #openstack | 22:09 | |
| *** reldan has joined #openstack | 22:10 | |
| *** joearnold has quit IRC | 22:10 | |
| *** miclorb has joined #openstack | 22:13 | |
| *** lionel_ has quit IRC | 22:16 | |
| *** lionel_ has joined #openstack | 22:16 | |
| *** ewindisch has quit IRC | 22:22 | |
| *** ewindisch has joined #openstack | 22:23 | |
| *** mahadev has quit IRC | 22:27 | |
| *** jdevesa has quit IRC | 22:30 | |
| justinsb | Can I get another PyLint job added to Jenkins? (One with --errors-only?) | 22:31 |
| *** gondoi has quit IRC | 22:32 | |
| eday | http://hudson.openstack.org/job/nova-pylint/ that's an awesome graph. does it mean we're getting bigger and better? :) | 22:39 |
| justinsb | eday: I hope not, because I'm about to take a huge swing at that number... https://code.launchpad.net/~justin-fathomdb/nova/more-focused-pylint/+merge/53711 | 22:40 |
| justinsb | eday: It should go down to about 3300 right off the bat | 22:41 |
| eday | justinsb: http://hudson.openstack.org/job/nova-pylint-errors/ | 22:41 |
| eday | just created, trying a run now | 22:41 |
| justinsb | eday: Thank you very, very much!! | 22:41 |
| justinsb | eday: What's the pylint command it's running, BTW? I can't see it... | 22:42 |
| eday | 1001 violations | 22:43 |
| eday | pylint --errors-only --rcfile=pylintrc -f parseable bin/* nova > pylint.txt | 22:43 |
| justinsb | eday: Thanks... I'll run the same on my Jenkins | 22:43 |
| justinsb | eday: I hope to get that down to maybe 100 or 200 that are all highly suspicious within a hour or two | 22:43 |
| eday | justinsb: I also added it to the main nova job trigger so it's run with every commit | 22:44 |
| justinsb | eday: Wonderful ... thank you | 22:45 |
| eday | np | 22:45 |
| *** smeier00 has quit IRC | 22:46 | |
| *** iammartian has quit IRC | 22:48 | |
| *** smeier00 has joined #openstack | 22:49 | |
| *** msassak has quit IRC | 22:52 | |
| *** smeier00 has quit IRC | 22:56 | |
| uvirtbot | New bug: #736482 in swift "make use of new eventlet debug flag" [Undecided,New] https://launchpad.net/bugs/736482 | 22:56 |
| uvirtbot | New bug: #736481 in swift "don't let jobs talk to different version servers" [Undecided,New] https://launchpad.net/bugs/736481 | 22:57 |
| *** smeier00 has joined #openstack | 22:57 | |
| *** Ryan_Lane has joined #openstack | 22:58 | |
| *** MarcMorata has quit IRC | 23:00 | |
| uvirtbot | New bug: #736485 in swift "detect and raise exception if request can't be fulfilled" [Undecided,New] https://launchpad.net/bugs/736485 | 23:06 |
| *** elasticdog has quit IRC | 23:08 | |
| *** paltman has quit IRC | 23:12 | |
| *** axelis has quit IRC | 23:12 | |
| openstackjenkins | Project swift build #223: SUCCESS in 27 sec: http://hudson.openstack.org/job/swift/223/ | 23:17 |
| openstackjenkins | Tarmac: New index code for the dbs; should improve performance for containers with a lot of recent deletes. | 23:17 |
| *** ramd has joined #openstack | 23:24 | |
| *** shentonfreude has joined #openstack | 23:25 | |
| ramd | is there image file for OpenVPN - How would I create a ami-cloudpipe image | 23:25 |
| *** beepboop has joined #openstack | 23:29 | |
| beepboop | hi all | 23:29 |
| *** allsystemsarego has quit IRC | 23:30 | |
| beepboop | thought I'd ask before I blew things up; if I want to serve images from a different interface(eth1, etc.), would I just modify the s3_dmz flag in the nova.conf file? | 23:30 |
| *** stewart has quit IRC | 23:31 | |
| *** hggdh has quit IRC | 23:32 | |
| *** clauden has quit IRC | 23:33 | |
| *** hggdh has joined #openstack | 23:33 | |
| *** hggdh has joined #openstack | 23:33 | |
| uvirtbot | New bug: #736505 in openstack-dashboard "modify project user roles from admin UI is broken" [High,In progress] https://launchpad.net/bugs/736505 | 23:35 |
| *** kashyap has joined #openstack | 23:37 | |
| *** elasticdog has joined #openstack | 23:39 | |
| *** Ryan_Lane has quit IRC | 23:39 | |
| justinsb | Hooray! PyLint is now giving useful results (for me, and for everyone once my pylintrc tweaks land) | 23:42 |
| dragondm | heya... is jaypipes still around ? I am having a glance issue | 23:42 |
| justinsb | I also figured out how to correlate the problems to the committer :-) | 23:43 |
| dragondm | has anyone seen this issue w/ glance before: TypeError: add_log_options() takes exactly 1 argument (2 given) | 23:43 |
| *** stewart has joined #openstack | 23:44 | |
| dragondm | on attempting to run glance-manage ? | 23:44 |
| *** mahadev has joined #openstack | 23:44 | |
| justinsb | dragondm: Do you have a line #? | 23:45 |
| dragondm | File "/usr/local/lib/python2.6/dist-packages/glance-2011.2-py2.6.egg/EGG-INFO/scripts/glance-manage", line 57, in create_options | 23:45 |
| justinsb | Looks like a bug introduced here: | 23:48 |
| justinsb | http://bazaar.launchpad.net/~glance-core/glance/cactus-trunk/revision/82.4.1 | 23:48 |
| *** azneita has joined #openstack | 23:48 | |
| *** azneita has joined #openstack | 23:48 | |
| justinsb | add_log_options was changed to take 1 arg from taking 2 | 23:48 |
| justinsb | glance-manage doesn't look to have been changed | 23:48 |
| justinsb | Though e.g. glance-api was | 23:49 |
| justinsb | Bet you a beer that pylint would have caught this :-) | 23:49 |
| dragondm | <signh> | 23:50 |
| creiht | justinsb: or even better, unit tests :) | 23:53 |
| justinsb | creiht: Or both :-) | 23:54 |
| *** dfg__ has quit IRC | 23:54 | |
| justinsb | dragondm: I think you can probably just edit out the first arg in line 57 | 23:54 |
| dragondm | yah, but then line 123 breaks because *that* method nedds an added arg. | 23:55 |
| dragondm | er needs | 23:55 |
| justinsb | Line 123 of which file? | 23:56 |
| *** dfg__ has joined #openstack | 23:56 | |
| justinsb | dragondm: Oh, of config.py I'm guessing | 23:57 |
| justinsb | dragondm: For some reason they've pulled out the calls to setup_logging in that same patch | 23:57 |
| dragondm | no, of glance-manage | 23:57 |
| justinsb | dragondm: I'm going to stop sticking my nose where it doesn't belong | 23:57 |
| *** pvo has joined #openstack | 23:57 | |
| justinsb | dragondm: OK ... it's the call to setup_logging, anyway | 23:58 |
| justinsb | dragondm: Looks like you need jaypipes! | 23:59 |
| *** smeier00 has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
