| *** jlmjlm has joined #openstack | 00:00 | |
| uvirtbot | New bug: #727501 in openstack-dashboard "Unable to add user to a project" [Undecided,New] https://launchpad.net/bugs/727501 | 00:01 |
|---|---|---|
| uvirtbot | New bug: #727502 in nova "Disallow concurrent snapshots on the same instance" [Undecided,New] https://launchpad.net/bugs/727502 | 00:01 |
| *** daveiw1 has quit IRC | 00:08 | |
| devcamcar | anyone have monty's email? | 00:09 |
| *** troytoman is now known as troytoman-away | 00:10 | |
| *** berendt has quit IRC | 00:13 | |
| *** azneita has quit IRC | 00:18 | |
| *** dubsquared has quit IRC | 00:18 | |
| *** kazu has joined #openstack | 00:21 | |
| *** msassak has quit IRC | 00:24 | |
| *** joearnold has joined #openstack | 00:28 | |
| *** timrc has quit IRC | 00:30 | |
| *** dweimer has joined #openstack | 00:31 | |
| *** CloudChris has quit IRC | 00:34 | |
| *** bcwaldon has quit IRC | 00:37 | |
| *** joearnold has quit IRC | 00:37 | |
| *** joearnold has joined #openstack | 00:37 | |
| *** masumotok_ has joined #openstack | 00:40 | |
| *** kazu has quit IRC | 00:48 | |
| *** masumotok_ has quit IRC | 00:55 | |
| *** slyphon has joined #openstack | 00:57 | |
| *** joearnold has quit IRC | 00:57 | |
| *** clayg_ is now known as clayg | 00:59 | |
| *** cascone has joined #openstack | 01:00 | |
| uvirtbot | New bug: #727522 in nova "test_network unit test cases fails for FlatDHCP/Flat network model" [Undecided,New] https://launchpad.net/bugs/727522 | 01:06 |
| *** dovetail1an has quit IRC | 01:06 | |
| *** clauden_ has quit IRC | 01:07 | |
| *** bcwaldon has joined #openstack | 01:13 | |
| *** dragondm has quit IRC | 01:18 | |
| *** gondoi has quit IRC | 01:20 | |
| *** gondoi has joined #openstack | 01:20 | |
| *** dirakx has joined #openstack | 01:30 | |
| *** gondoi has quit IRC | 01:42 | |
| *** gondoi has joined #openstack | 01:43 | |
| *** ovidwu has joined #openstack | 01:49 | |
| *** littleidea has joined #openstack | 01:54 | |
| *** dendro-afk is now known as dendrobates | 01:55 | |
| *** littleidea has quit IRC | 01:57 | |
| nelson | Can someone take a look at a very small client.put_object program? http://paste.openstack.org/show/786/ | 02:00 |
| *** mahadev has quit IRC | 02:00 | |
| nelson | It's giving me a "499 Client Disconnect" error. | 02:01 |
| nelson | I tried stracing it, and the HTTP client isn't disconnecting, grrrrrrrr. | 02:02 |
| nelson | It opens up the socket, does a PUT with the right URL, and the right headers, a blank line, then the nine bytes it said it was going to send. | 02:05 |
| nelson | Doesn't close the socket. it reads the HTTP response, which is HTTP/1.1 400 Client Disconnect. | 02:06 |
| nelson | The log file also says that it's not sending enough. | 02:07 |
| *** gregp76 has quit IRC | 02:07 | |
| *** nelson has quit IRC | 02:10 | |
| *** nelson has joined #openstack | 02:10 | |
| nelson | Frustrating as heck. | 02:11 |
| *** ovidwu has quit IRC | 02:20 | |
| *** hggdh has quit IRC | 02:23 | |
| *** hggdh has joined #openstack | 02:23 | |
| *** hggdh has joined #openstack | 02:23 | |
| *** dendrobates is now known as dendro-afk | 02:27 | |
| *** westmaas has joined #openstack | 02:31 | |
| *** burris has joined #openstack | 02:32 | |
| *** stewart_ is now known as stewart | 02:35 | |
| *** mahadev has joined #openstack | 02:37 | |
| *** dendro-afk is now known as dendrobates | 02:38 | |
| *** grapex1 has left #openstack | 02:42 | |
| *** miclorb has quit IRC | 02:43 | |
| *** mahadev has quit IRC | 02:44 | |
| *** lvaughn_ has quit IRC | 02:45 | |
| *** arun_ has quit IRC | 02:46 | |
| *** lvaughn has joined #openstack | 02:46 | |
| *** lvaughn_ has joined #openstack | 02:48 | |
| *** mahadev has joined #openstack | 02:50 | |
| *** lvaughn has quit IRC | 02:51 | |
| nelson | oh crap. | 02:54 |
| *** rds__ has quit IRC | 02:55 | |
| *** littleidea has joined #openstack | 03:01 | |
| nelson | double crap. | 03:02 |
| *** miclorb_ has joined #openstack | 03:03 | |
| nelson | If you call put_object like this: | 03:03 |
| nelson | wmf.client.put_object(url, token, container, obj+".txt", "test data", content_type="text/plain") | 03:03 |
| nelson | it fails to read enough. | 03:03 |
| nelson | If you call it like this: | 03:04 |
| nelson | inf = open("test-data") | 03:04 |
| nelson | wmf.client.put_object(url, token, container, obj+".txt", inf, content_type="text/plain") | 03:04 |
| nelson | it reads enough. Fuck. | 03:04 |
| nelson | I'm still running Austin, so there's little point in registering a bug unless somebody running bexar can reproduce it. | 03:05 |
| nelson | BTW, wmf = swift.common | 03:06 |
| *** dirakx has quit IRC | 03:09 | |
| *** mdomsch has joined #openstack | 03:12 | |
| *** dirakx has joined #openstack | 03:23 | |
| *** neckwarmer has quit IRC | 03:26 | |
| *** neckwarmer has joined #openstack | 03:27 | |
| *** bcwaldon has quit IRC | 03:33 | |
| *** westmaas1 has joined #openstack | 03:47 | |
| *** westmaas has quit IRC | 03:49 | |
| *** bcwaldon has joined #openstack | 03:49 | |
| *** westmaas1 has quit IRC | 03:56 | |
| *** kashyap has joined #openstack | 03:58 | |
| *** westmaas has joined #openstack | 03:59 | |
| *** littleidea has quit IRC | 04:08 | |
| *** greg-g has left #openstack | 04:27 | |
| *** arun_ has joined #openstack | 04:31 | |
| *** ccustine has quit IRC | 04:35 | |
| *** arun_ has quit IRC | 04:35 | |
| *** bcwaldon has quit IRC | 04:36 | |
| *** miclorb_ has quit IRC | 04:38 | |
| *** omidhdl has joined #openstack | 04:41 | |
| *** EndEng|Desktop has joined #openstack | 04:53 | |
| *** dendrobates is now known as dendro-afk | 05:09 | |
| *** timrc has joined #openstack | 05:11 | |
| *** kashyap has quit IRC | 05:13 | |
| *** ovidwu has joined #openstack | 05:24 | |
| *** arun_ has joined #openstack | 05:27 | |
| *** arun_ has joined #openstack | 05:27 | |
| *** z0 has joined #openstack | 05:27 | |
| *** Code_Bleu has quit IRC | 05:31 | |
| *** littleidea has joined #openstack | 05:36 | |
| *** littleidea has quit IRC | 05:43 | |
| *** littleidea has joined #openstack | 05:44 | |
| *** kashyap has joined #openstack | 05:46 | |
| *** Code_Bleu has joined #openstack | 05:47 | |
| *** mahadev has quit IRC | 05:52 | |
| *** mahadev has joined #openstack | 05:53 | |
| *** mdomsch has quit IRC | 05:54 | |
| *** ovidwu has quit IRC | 05:56 | |
| *** Code_Bleu has quit IRC | 05:58 | |
| *** adjohn has joined #openstack | 06:05 | |
| *** Manikandan has joined #openstack | 06:15 | |
| Manikandan | Permission denied (publickey). | 06:16 |
| Manikandan | bzr: ERROR: Connection closed: Unexpected end of message. Please check connectivity and permissions, and report a bug if problems persist. | 06:16 |
| tripson_ | Manikandan: it means that permission is denied and connection was closed because of the unexpected end of message. maybe you should check your connectivity and permissions... or maybe its just a bug you need to report | 06:20 |
| *** ovidwu has joined #openstack | 06:21 | |
| *** gregp76 has joined #openstack | 06:21 | |
| *** dirakx has quit IRC | 06:22 | |
| Manikandan | ok | 06:28 |
| Manikandan | ok | 06:28 |
| *** perestre1ka has quit IRC | 06:31 | |
| *** perestrelka has joined #openstack | 06:32 | |
| *** photron has joined #openstack | 06:33 | |
| *** z0 has quit IRC | 06:33 | |
| *** gregp76 has quit IRC | 06:35 | |
| *** dirakx has joined #openstack | 06:36 | |
| *** naehring has joined #openstack | 06:40 | |
| *** 77CAATQDF has joined #openstack | 06:41 | |
| *** ovidwu has quit IRC | 06:49 | |
| *** hazmat has joined #openstack | 06:55 | |
| *** slyphon has quit IRC | 07:00 | |
| *** zenmatt has quit IRC | 07:02 | |
| *** rackerhacker has quit IRC | 07:10 | |
| *** gdusbabek has quit IRC | 07:10 | |
| *** gdusbabek has joined #openstack | 07:10 | |
| *** DigitalFlux has joined #openstack | 07:14 | |
| *** DigitalFlux has joined #openstack | 07:14 | |
| *** rackerhacker has joined #openstack | 07:14 | |
| *** mahadev has quit IRC | 07:21 | |
| *** adjohn has quit IRC | 07:26 | |
| *** mahadev has joined #openstack | 07:28 | |
| *** dirakx has quit IRC | 07:30 | |
| *** shoaib has joined #openstack | 07:32 | |
| shoaib | hi! | 07:34 |
| shoaib | i need help | 07:35 |
| shoaib | how can i start a stopped instance | 07:35 |
| shoaib | ? | 07:35 |
| shoaib | but instance in not found in euca-describe-instances | 07:35 |
| shoaib | i am new to openstack ... maybe i am doing something wrong... | 07:36 |
| shoaib | need help on this | 07:37 |
| shoaib | anyone? | 07:38 |
| soren | Typically, when you terminate an instance, it disappears. | 07:38 |
| shoaib | i didnt terminate them | 07:41 |
| shoaib | the host machine got reboot | 07:41 |
| shoaib | so all instances running are no longer there | 07:42 |
| *** rcc has joined #openstack | 07:42 | |
| soren | Hm... Yeah, I'm not sure instances will survive that at the moment. | 07:43 |
| *** mgoldmann has joined #openstack | 07:43 | |
| shoaib | so how can i check | 07:47 |
| shoaib | i can see instances in /var/lib/nova/instances | 07:47 |
| shoaib | anyway to start them from there? | 07:47 |
| soren | You can start them, but I don't think Nova will adopt them again. | 07:47 |
| soren | This is a problem :( | 07:47 |
| shoaib | oh man | 07:50 |
| *** adjohn has joined #openstack | 07:50 | |
| shoaib | but there muct be a way | 07:50 |
| *** f4m8_ is now known as f4m8 | 07:51 | |
| shoaib | +soren -> is this due to kvm or otherwise? | 07:52 |
| *** allsystemsarego has joined #openstack | 07:53 | |
| *** allsystemsarego has joined #openstack | 07:53 | |
| *** skiold has joined #openstack | 07:56 | |
| *** dirakx has joined #openstack | 07:56 | |
| *** MarkAtwood has left #openstack | 07:57 | |
| *** littleidea has quit IRC | 08:08 | |
| soren | shoaib: No, it's just the way things work in Nova right now. | 08:19 |
| soren | shoaib: When it starts, it checks to see if the VM's that are supposed to be running on it are running. If not, it removes them from its list of instances to worry about. | 08:20 |
| *** hazmat has quit IRC | 08:22 | |
| *** pyros_xx has joined #openstack | 08:26 | |
| pyros_xx | helloo | 08:27 |
| pyros_xx | there is somebody can help me | 08:27 |
| pyros_xx | ?? | 08:27 |
| ttx | pyros_xx: maybe. Depends on your question. | 08:28 |
| pyros_xx | i need answers about nova | 08:28 |
| pyros_xx | hello sorry for my english | 08:28 |
| pyros_xx | i have just installed nova | 08:28 |
| *** ramkrsna has joined #openstack | 08:28 | |
| *** ramkrsna has joined #openstack | 08:28 | |
| ttx | pyros_xx: Ask your questions, maybe somebody will know the answer and reply. | 08:29 |
| pyros_xx | i have a cloud controller and node controller on another machine | 08:29 |
| pyros_xx | when i run mysql -u root -p$MYSQL_PASS nova -e 'select * from services;' | 08:29 |
| shoaib | soren -> how can i make sure VM (instances) are running after a host reboot? is there anyway? | 08:30 |
| pyros_xx | in updated and deleted fields of the table appear NULL NULL NULL | 08:30 |
| ttx | pyros_xx: we have no such thing as a cloud controller and node controller in nova. | 08:30 |
| soren | shoaib: Not that I can think of right now. | 08:31 |
| pyros_xx | where can i look for answerss | 08:31 |
| pyros_xx | ? | 08:31 |
| soren | pyros_xx: Here is fine. We just need to ask questions. | 08:32 |
| soren | shoaib: Typically, when designing applications "for the cloud", you make it so that individual instances don't matter. If one disappears, you should compensate somehow. | 08:32 |
| shoaib | soren -> yes, but the instance should have restarted once we have rebooted the host machine. | 08:33 |
| soren | shoaib: You're missing my point. | 08:34 |
| shoaib | soren -> sorry i am just new pls explain me... | 08:34 |
| pyros_xx | it is possible install nova and make it functional running multiple nodes? | 08:34 |
| soren | shoaib: Whether the instance should have restarted is debatable. Fact of the matter is that right now, that doesn't happen. | 08:34 |
| pyros_xx | sorry for my english | 08:34 |
| pyros_xx | iam spanish | 08:35 |
| soren | pyros_xx: Yes, that is possible. | 08:35 |
| soren | shoaib: ...and then I explain that in typical cloud-enabled applications, this is not a problem, really. | 08:35 |
| pyros_xx | when i try to run an instance appear me unknown error | 08:35 |
| soren | Where? | 08:36 |
| ttx | pyros_xx: I would first start to test it all on one machine though, to start understanding how it works. | 08:36 |
| pyros_xx | where? | 08:36 |
| pyros_xx | whith . novarc with the credentials | 08:37 |
| ttx | pyros_xx: where does that "unknown error" appear ? | 08:38 |
| pyros_xx | i run euca-run-instances ami-r4a7staa -k mykey -t m1.tiny | 08:38 |
| pyros_xx | output UnknownError: Ha sucedido un error desconocido. Por favor repite el intento de nuevo. | 08:38 |
| pyros_xx | ubuntu 10.04 amd64 es-ES | 08:39 |
| soren | How did you install nova? | 08:39 |
| pyros_xx | http://wiki.openstack.org/NovaInstall/MultipleServer | 08:40 |
| pyros_xx | follow Manual installation step by step | 08:40 |
| *** Nacx has joined #openstack | 08:40 | |
| soren | pyros_xx: There are multiple sets of instructions on that page. | 08:41 |
| Manikandan | when i enter the cmd bzr branch lp:nova it runs for a while and it gives bzr branch lp:nova | 08:43 |
| Manikandan | Write failed: Broken pipehing revisions:Inserting stream:Estimate 25960/42003 | 08:43 |
| Manikandan | bzr: ERROR: Connection closed: Unexpected end of message. Please check connectivity and permissions, and report a bug if problems persist. | 08:43 |
| pyros_xx | where APPEAR MANUAL INSTALLATION | 08:43 |
| soren | Manikandan: Retry? Looks like your connection was broken. | 08:44 |
| Manikandan | this is the second time when i use bzr branch lp:nova cmd | 08:44 |
| * soren tries | 08:45 | |
| soren | Manikandan: Works for me. Proxy issues, perhaps? | 08:46 |
| soren | $ bzr branch lp:nova | 08:46 |
| soren | Branched 752 revision(s). | 08:46 |
| *** mahadev has quit IRC | 08:47 | |
| pyros_xx | please could you advice for some manual for a multiple server installation? | 08:47 |
| *** daveiw has joined #openstack | 08:48 | |
| pyros_xx | when run euca-describe-availability-zones verbose | 08:48 |
| pyros_xx | output: AVAILABILITYZONEnovaavailable AVAILABILITYZONE|- CloudController AVAILABILITYZONE| |- nova-schedulerenabled XXX None AVAILABILITYZONE| |- nova-networkenabled XXX None AVAILABILITYZONE| |- nova-computeenabled XXX None AVAILABILITYZONE|- Node1 AVAILABILITYZONE| |- nova-networkenabled XXX None AVAILABILITYZONE| |- nova-computeenabled XXX None AVAILABILITYZONE| |- nova-schedulerenabled XXX None | 08:48 |
| ttx | pyros_xx: the issue is that no manual will precisely match your desired topology. That's why it's simpler to start with a simple topology (all in one) and understand how it works, then move to more complex setups | 08:49 |
| pyros_xx | thank you | 08:50 |
| *** photron has quit IRC | 08:50 | |
| pyros_xx | i try with that | 08:50 |
| *** pyros_xx435 has joined #openstack | 08:52 | |
| *** uksysadmin has joined #openstack | 08:54 | |
| shoaib | soren -> would you know where i could get a centos image to upload on nova? | 08:55 |
| soren | Nope. | 08:55 |
| shoaib | that was quick :) | 08:55 |
| uksysadmin | shoaib - get one from eucalyptus - I pasted a link when you asked that the other day | 08:56 |
| *** rcc has quit IRC | 08:56 | |
| *** MarcMorata has joined #openstack | 08:59 | |
| uksysadmin | we run rhel compat images where I work... so its crucial the guests work great under openstack nova - comparable to what we can do with vsphere | 08:59 |
| soren | annegentle: I see you wrote on http://wiki.openstack.org/NovaInstall/MultipleServer that there can't be comments in flag files. There can. They just need to be on a line of their own. Lines beginning with # or // are ignored. | 09:03 |
| *** hazmat has joined #openstack | 09:14 | |
| *** omidhdl has left #openstack | 09:17 | |
| *** irahgel has joined #openstack | 09:21 | |
| dsockwell | uksysadmin: they hypervisor will be the most important choice then. for the most compatibility, avoid containerization. KVM, kqemu, and xen are probably your best choices. | 09:40 |
| dsockwell | uksysadmin: i'm partial to xen, but all three are fine choices. | 09:41 |
| dsockwell | KVM is, i believe, highly red-hat related | 09:41 |
| dsockwell | kqemu will get you the best OS (windows esp.) compatibility on older hardware, and xen PV will give you good Linux performance on older hardware as well. KVM requires vt-x or amd-v. | 09:42 |
| uksysadmin | actually, its possible that in the short term the hypervisor is vmware - but that's me wondering whether its best to straddle our current large estate or create a new one. hardware isn't an issue - we're at the starting point for a "new environment" on Nehalems | 09:43 |
| soren | kqemu is dead. | 09:43 |
| dsockwell | duly noted | 09:43 |
| soren | Had been for a while. | 09:43 |
| *** reldan has joined #openstack | 09:44 | |
| uksysadmin | part of my role is evangelising openstack for an established vmware estate hosting websites | 09:44 |
| soren | Not sure what being "highly red-hat related" means. Ubuntu was the first major distro to ship it and to make it the default hypervisor. | 09:44 |
| soren | "it" == KVM. | 09:44 |
| dsockwell | i thought kvm was a red-hat project. | 09:44 |
| soren | RedHat bought Qumranet. | 09:44 |
| soren | ..who established the project. | 09:45 |
| uksysadmin | soren types faster than uksysadmin ;-) | 09:45 |
| dsockwell | ok, that's how i thought that | 09:45 |
| soren | RedHat builds a lot of cool stuff. They're just not very good at shipping it. | 09:45 |
| soren | IMO. | 09:45 |
| dsockwell | how is vmware support for openstack coming along? | 09:46 |
| * uksysadmin isn't a RH fan - our a/c manager has seen that over the last few years | 09:46 | |
| dsockwell | website said it was supposed to be in cactus | 09:46 |
| dsockwell | is supposed to be in cactus, which is in the future | 09:46 |
| dsockwell | i should say | 09:46 |
| soren | dsockwell: It's said to be proposed for review next week. | 09:47 |
| soren | 21:20 < ewanmellor> ttx: We are peer reviewing vsphere-support internally at the moment. | 09:48 |
| uksysadmin | I'm currently at the stage where I've got a workable demo (~50% done) under virtualbox as a stop gap for kit being racked to put openstack on... most of the work I'm doing is around provisioning of the instances and deploying the webstack... not too fussed on the hypervisor at the moment [don't have much choice when vbox and hardware vt collide!] | 09:48 |
| soren | 21:21 < ewanmellor> ttx: It's being done on Launchpad, so people can look if they want. We intend to bring it to nova-core for review in a week. | 09:48 |
| uksysadmin | I'm up for looking at that soren | 09:48 |
| dsockwell | fun. | 09:48 |
| dsockwell | so uksysadmin can use openstack to integrate his vsphere stuff with any new hardware running freer hypervisors? | 09:48 |
| uksysadmin | but I'm not 100% convinced I need it despite being 100% vsphere at the mo... my plan of attack is build fresh, scrap the old. | 09:49 |
| uksysadmin | with vsphere integration I could manage the estate whether it was vsphere or kvm would be an integration plan... | 09:49 |
| uksysadmin | but equally I could just let the old environment die as we build up the new environment | 09:50 |
| uksysadmin | integrating vsphere with openstack will help ease things though, especially to the bean counters who see our budgets invest in vmware each year... | 09:51 |
| soren | ttx: What's the status on packacing novatools? | 09:52 |
| ttx | soren: i'm waiting on the latest module name changes | 09:52 |
| uksysadmin | personally all the fun stuff - CBIT environments and spinning up complete environments - will be done on openstack as part of the business already use ec2 and rightscale | 09:52 |
| ttx | soren: looks like it will be python-novaclient | 09:53 |
| soren | ttx: So the packaging is done? | 09:54 |
| soren | ttx: It just needs to settle down? | 09:54 |
| ttx | soren: I tested by copying the cloudservers packaging... trying to find it now | 09:56 |
| ttx | I /may/ have done that in /tmp | 09:56 |
| ttx | ... but I uploaded it in PPA | 09:56 |
| ttx | https://launchpad.net/~ttx/+archive/ppa | 09:56 |
| *** adjohn has quit IRC | 09:57 | |
| *** adjohn has joined #openstack | 09:57 | |
| * soren stares at https://code.launchpad.net/~dan-prince/nova/bug723897/+merge/50987 | 10:02 | |
| ttx | soren: there was a whole thread about it on the ML. | 10:03 |
| uksysadmin | euca-associate-address : when an instance is terminated, what's the reasoning for it not disassociating that address from that instance? intentional or bug? (early-ish morning, coffee not kicked in to be able to think yet) | 10:05 |
| *** miclorb_ has joined #openstack | 10:06 | |
| uksysadmin | the reason I say it is that if I spin up another instance, the public address then gets associated with the new instance (new id, same private address)... but the routing doesn't work | 10:06 |
| soren | uksysadmin: Fixed or floating ip? | 10:08 |
| soren | ttx: Ah, yes, that rings a bell. | 10:09 |
| uksysadmin | floating | 10:10 |
| uksysadmin | I end up disassociating the public IP and associating it with an instance again | 10:15 |
| uksysadmin | I can then connect | 10:15 |
| *** kashyap_ has joined #openstack | 10:20 | |
| *** ramkrsna_ has joined #openstack | 10:20 | |
| *** kashyap has quit IRC | 10:23 | |
| *** ramkrsna has quit IRC | 10:23 | |
| soren | uksysadmin: I don't see any reason why floating ip's shouldn't be disassociated on instance termination. That sounds very odd. | 10:24 |
| uksysadmin | I'll do some more playing to see if its "environmental" then raise a bug if I'm out of ideas | 10:25 |
| soren | uksysadmin: It should happen. the code is there. | 10:25 |
| uksysadmin | v odd - at the moment on my set up after repeating tear downs and builds it does happen consistently where my new instances get the associated public ip from my previous association | 10:27 |
| uksysadmin | but... that could be consistency in the way I'm doing all those steps... | 10:28 |
| soren | What does "tear down and build" mean in your environemnt? | 10:28 |
| uksysadmin | install ubu, install openstack, create instance, delete instance, create instance + voila | 10:29 |
| *** londo_ has quit IRC | 10:29 | |
| soren | You actually install ubuntu all over each time? | 10:30 |
| soren | ..yet it somehow manages to remember something about your floating ip's? | 10:30 |
| uksysadmin | its all a virtual test env on my laptop | 10:30 |
| soren | That sounds rather magical. | 10:30 |
| uksysadmin | oh - no | 10:30 |
| uksysadmin | yes I am | 10:30 |
| uksysadmin | but that's not for this forum | 10:30 |
| uksysadmin | ;-) | 10:30 |
| uksysadmin | what I mean is that I get the problem, assume its some dumbass thing I've done and start again | 10:31 |
| *** pyros_xx_ has joined #openstack | 10:31 | |
| pyros_xx_ | hello | 10:31 |
| uksysadmin | I don't remove openstack/ubu, etc just for creating instances | 10:31 |
| uksysadmin | I do that to try and fix this issue | 10:31 |
| pyros_xx_ | please help when i launch a instance i got failed to spawn mykey | 10:32 |
| pyros_xx_ | "failed to spawn mykey" | 10:32 |
| soren | pyros_xx_: You're missing -k in your command line. | 10:32 |
| *** kashyap__ has joined #openstack | 10:32 | |
| pyros_xx_ | this is the command | 10:32 |
| pyros_xx_ | euca-run-instances ami-o6olt0tm -k mykey -t m1.tiny | 10:32 |
| *** ramkrsna__ has joined #openstack | 10:33 | |
| uksysadmin | for all intents and purposes to this problem assume my env is good - the problem is: create instance, associate address, ssh to new address, log out, terminate instance, create new instance, describe instances and floating ip is associated with instance, can't ping/ssh, disassociate address, associate same public address to new instance, I can ping/ssh. | 10:33 |
| soren | Oh, so it just looks associated, but isn't really? | 10:33 |
| soren | Ah. | 10:33 |
| soren | Yeah, I can sort of see how that would happen. | 10:34 |
| BK_man | hi all. where can I download latest RC for Nova 2011.1.1 version? | 10:34 |
| soren | Not completely, but it doesn't sound like complete voodoo anymore. | 10:34 |
| BK_man | is that file latest? http://nova.openstack.org/tarballs/nova-2011.1.1~bzr653.tar.gz | 10:34 |
| uksysadmin | I'm just happy I'm not going completely mad | 10:34 |
| soren | pyros_xx_: Please post a screenshot of you running that command. | 10:34 |
| soren | uksysadmin: So from the network node's perspective it gets disassociated. | 10:35 |
| uksysadmin | yeah | 10:35 |
| soren | uksysadmin: ...which is why it doesn't work. | 10:35 |
| soren | uksysadmin: ...but the link from the fixed ip to the floating ip isn't properly broken. | 10:35 |
| *** kashyap_ has quit IRC | 10:35 | |
| pyros_xx_ | RESERVATION r-d1fv4gl1 myproject default INSTANCE i-00000005 ami-o6olt0tm 192.168.0.6 192.168.0.6 failed to spawn mykey (myproject, NOVA) 0 m1.tiny 2011-03-02T10 :35:35Z nova | 10:35 |
| soren | uksysadmin: ..and you new instance gets the same fixed ip, right? | 10:35 |
| uksysadmin | yeah | 10:36 |
| uksysadmin | ah - I see what you're getting at | 10:36 |
| soren | pyros_xx_: Ah. | 10:36 |
| soren | pyros_xx_: "failed to spawn" is one column of information, "mykey" is another. | 10:36 |
| *** ramkrsna_ has quit IRC | 10:36 | |
| soren | pyros_xx_: Look in /var/log/nova/nova-compute.log for more info on why it might have failed to spawn. | 10:36 |
| *** miclorb_ has quit IRC | 10:38 | |
| *** miclorb has joined #openstack | 10:40 | |
| soren | uksysadmin: bug 727705 | 10:42 |
| uvirtbot | Launchpad bug 727705 in nova "Floating ip's do not get disassociated from fixed ip's when the fixed ip's are disassociated from instances" [Undecided,New] https://launchpad.net/bugs/727705 | 10:42 |
| uksysadmin | cheers | 10:42 |
| soren | I don't understand why, though. The code seems to be there. | 10:46 |
| soren | nova/network/manager.py calls db.floating_ip_disassociate which should to the right thing. | 10:47 |
| uksysadmin | I'm playing with it right now to see if I can figure it out/confirm it all | 10:48 |
| uksysadmin | I do get this: Disassociating address 172.241.0.3 from (pid=1585) terminate_instance /usr/lib/pymodules/python2.6/nova/compute/manager.py:241 | 10:48 |
| soren | 172.241.0.3 is the fixed ip, right? | 10:50 |
| uksysadmin | yeah | 10:50 |
| soren | Should be, since it's in the compute manager. | 10:50 |
| uksysadmin | my private was like 10.0.0.3 | 10:50 |
| soren | This other thing would be in nova-network's log file. | 10:50 |
| soren | Eh? | 10:50 |
| soren | So 172.241.0.3 is the floating? | 10:50 |
| * soren is confused | 10:51 | |
| soren | and hungry, actually. | 10:51 |
| uksysadmin | sorry - not helping | 10:51 |
| uksysadmin | private is 10.0.0.3, public 172.241.0.3 | 10:51 |
| uvirtbot | New bug: #727705 in nova "Floating ip's do not get disassociated from fixed ip's when the fixed ip's are disassociated from instances" [Undecided,New] https://launchpad.net/bugs/727705 | 10:51 |
| uksysadmin | it seems if my instance gets 10.0.0.3 it gets 172.241.0.3 (because I associated it with an instance that had that private ip) | 10:51 |
| uksysadmin | I'll check the nova. tables - see if there are any duplicates, etc. | 10:52 |
| * soren goes to eat | 10:52 | |
| BK_man | khm.... Should distro install nova-api.conf file to /etc/nova instead of /etc ? | 11:11 |
| BK_man | ttx: around? | 11:11 |
| ttx | BK_man: yes | 11:13 |
| BK_man | ttx: what is default place for config files? I suppose it should be /etc/nova | 11:13 |
| ttx | BK_man: for nova config files, yes it is | 11:13 |
| BK_man | ttx if yes - then we have a bug in 2011.1.1-bzr563 tarball | 11:14 |
| ttx | BK_man: why ? | 11:14 |
| uksysadmin | soren: that bug looks like my fault | 11:14 |
| BK_man | ttx: it install nova-api.conf to /etc/ instead of /etc/nova | 11:14 |
| BK_man | nova.egg-info/SOURCES.txt:etc/nova-api.conf | 11:14 |
| * BK_man trying to prepare packaging for 2011.1.1 to release it as soon as 2011.1.1 will be oficially released for public | 11:15 | |
| ttx | BK_man: that's the file source. ./etc/nova-api.conf | 11:15 |
| ttx | not where it is deployed on the filesystem | 11:15 |
| BK_man | ttx: that's what I got from RPM: | 11:16 |
| BK_man | RPM build errors: | 11:16 |
| BK_man | Installed (but unpackaged) file(s) found: | 11:16 |
| BK_man | /etc/nova-api.conf | 11:16 |
| *** chmouel_ is now known as chmouel | 11:17 | |
| BK_man | $ find /home/abr/rpmbuild/BUILDROOT/openstack-nova-2011.1.1-1.x86_64 -name nova-api.conf | 11:17 |
| ttx | BK_man: also not a regression from 2011.1 | 11:17 |
| BK_man | ttx: I install nova-api.conf by myself in 2011.1 since it was missed in original tarball | 11:18 |
| *** miclorb has quit IRC | 11:18 | |
| ttx | BK_man: I guess that's some automatic behavior in your RPM packaging that needs to be overridden. there is nothing in that tarball that says "nova-api.conf must be deployed in /etc | 11:19 |
| BK_man | ttx: Ok, will investigate. | 11:20 |
| BK_man | ttx: I discovered also a new build dep: python-distutils-extra | 11:20 |
| ttx | BK_man: yes, due to i18n | 11:20 |
| BK_man | ttx: do we have a list of build deps for nova? pip-requires AFAIK for runtime deps | 11:21 |
| *** rds__ has joined #openstack | 11:22 | |
| ttx | BK_man: keep me posted. Maybe the default RPM packaging runs some command that misplaces nova-api.conf, and we can actually do something about it | 11:22 |
| BK_man | %{__python} setup.py install -O1 --skip-build --root %{buildroot} | 11:22 |
| BK_man | ttx: I think that this is done in above command | 11:23 |
| ttx | But that doesn't sound like a regression, just something we can improve to avoid having to override anything | 11:23 |
| ttx | BK_man: i'll try that after lunch. | 11:25 |
| BK_man | ttx: bon appetit! | 11:25 |
| uksysadmin | soren: yeah - my create floating ip address range had wrong host set... | 11:30 |
| uksysadmin | I'll paste all my stuff in the bug | 11:30 |
| *** romain_lenglet_ has joined #openstack | 11:30 | |
| *** h0cin has joined #openstack | 11:31 | |
| *** h0cin has joined #openstack | 11:31 | |
| *** romain_lenglet_ has quit IRC | 11:31 | |
| soren | uksysadmin: I don't think I follow. | 11:36 |
| uksysadmin | ok, simply put: my compute node is "nova1". My floating address space was created with: sudo nova-manage floating create cloud1 172.241.0.0/25 | 11:37 |
| uksysadmin | nova1 != cloud1 | 11:37 |
| uksysadmin | now - I'm not sure if this is a bug or not - its definitely not one as described | 11:38 |
| uksysadmin | I can associate floating addresses, where host isn't part of my environment, so its all getting confused | 11:38 |
| soren | I'm puzzled how it work at all, then. | 11:39 |
| uksysadmin | when I set nova.floating_ips host="nova1" it works as described | 11:39 |
| uksysadmin | (as in it works as intended) | 11:39 |
| uksysadmin | but yeah - that begs the question as to how its all working anyway. | 11:39 |
| uksysadmin | I guess the creation doesn't check host=, but the removal from the table must have something like "where host=blah"... [without looking at the code] | 11:40 |
| soren | Not exactly. | 11:40 |
| uksysadmin | I'm updating that bug to explain all this - step by step | 11:41 |
| soren | Cool, thanks. | 11:41 |
| uksysadmin | but - I think there's something not quite right - how can I associate an address in the first place to an instance when the instance isn't running on the "host" specified in floating_ips table? | 11:45 |
| uksysadmin | it should error/not assign it | 11:46 |
| soren | ah, worked it out. | 11:46 |
| soren | Heh. | 11:46 |
| *** Manikandan has quit IRC | 11:46 | |
| soren | When you call euca-associate-address, it calls the AssociateAddress method call in the EC2 API, which calls the internal compute api's associate_floating_address, which calls the network api's associate_floating_address, which sends an rpc call to the network manager on the host corresponding to the fixed_ip. | 11:48 |
| *** kashyap__ has quit IRC | 11:48 | |
| soren | When you call euca-disassociate-address, it calls the DisssociateAddress method call in the EC2 API, which calls the internal network api's disassociate_floating_ip, which sends an rpc call to the network manager on the host corresponding to the fixed_ip of the floating_ip. | 11:50 |
| soren | However... | 11:50 |
| uksysadmin | I'm following... | 11:52 |
| soren | When you call euca-terminate-instance, it calls the TerminateInstance method call in the EC2 API, which calls the internal compute api's delete method, which sends an "terminate_instance" rpc call to the compute manager, which sends a "disassociate_floating_address" to the network manager on the host corresponding to the floating_ip itself. | 11:53 |
| soren | "the network manager on the host corresponding to the floating_ip itself" is the key. | 11:53 |
| soren | Because that's that "cloud1" thing you set with nova-manage. | 11:54 |
| soren | Everywhere else, we seem to look up floating_ip->fixed_ip->host, which is fine. | 11:54 |
| soren | floating_ip->host is screwed. | 11:54 |
| * soren pastes into the bug | 11:55 | |
| uksysadmin | I get you | 11:55 |
| soren | That should be enough for someone to fix it. | 11:57 |
| uksysadmin | awesome | 11:57 |
| *** dendro-afk is now known as dendrobates | 11:58 | |
| uksysadmin | thanks for explaining. I admit to user error, but nice to know I'm not all to blame. | 11:58 |
| soren | I set it as low priority, since it only happens if you make that particular mistake. | 11:58 |
| * uksysadmin is an edge-case ;-) | 11:58 | |
| soren | uksysadmin: Sure, thanks for raising it. It needs to be fixed. | 11:58 |
| *** zul has quit IRC | 12:00 | |
| soren | sleepsonthefloor, vishy: As nifty as I think the noVNC stuff is, I'd really like if we could somehow expose VNC directly. | 12:21 |
| *** ctennis has quit IRC | 12:24 | |
| *** cascone has quit IRC | 12:25 | |
| BK_man | soren: quick question: how do you packaging ajaxterm for nova? | 12:26 |
| soren | [2011-03-02 13:26:33] soren@lenny:~$ apt-cache search ajaxterm | 12:26 |
| soren | ajaxterm - Web based terminal written in Python | 12:26 |
| soren | Someone else did. Long, long ago. | 12:26 |
| soren | Oh, er.. | 12:27 |
| soren | Bleh. | 12:27 |
| BK_man | soren: Khm.... I thought that nova uses it own version... | 12:27 |
| soren | Um, right. | 12:27 |
| BK_man | Note that ajaxterm was modified in the following ways: | 12:27 |
| BK_man | i. dies after 5 minutes of inactivity | 12:27 |
| BK_man | ii. now requires token authentication. Previously it was trivial to hijack an ajaxterm | 12:27 |
| soren | I actually haven't really looked into that at all yet. | 12:27 |
| BK_man | soren: ok. FYI :) | 12:27 |
| BK_man | soren: do you think that we should start ajaxterm on every compute node to be able to provide console access to the instances? | 12:28 |
| BK_man | s/every/EVERY/ :) | 12:28 |
| *** kainam has joined #openstack | 12:31 | |
| *** omidhdl has joined #openstack | 12:32 | |
| *** kainam- has quit IRC | 12:34 | |
| *** ctennis has joined #openstack | 12:41 | |
| BK_man | ttx: around? | 12:41 |
| BK_man | ttx: another installation glitch: tools dir (with ajaxterm inside) does not installed after calling "python setup.py install" | 12:42 |
| BK_man | ttx: I'm not sure if we need nova-combined in release tarball since it's mostly dev tool AFAIK and not prodution tool | 12:43 |
| *** drico has quit IRC | 12:44 | |
| *** ctennis_ has joined #openstack | 12:48 | |
| ttx | BK_man: I'm trying to understand what "python setup.py install" should do. On my setup it fails, complaining about error: cannot copy tree 'build/scripts-2.6': not a directory | 12:49 |
| ttx | I doubt it was ever supported | 12:49 |
| DigitalFlux | Hi | 12:50 |
| BK_man | DigitalFlux: hi | 12:51 |
| DigitalFlux | Did some body work in snapshotting support for Nova yet ? | 12:51 |
| DigitalFlux | soren: I looked at the branches and can't seem to find anyone who started on doing so ? | 12:51 |
| *** shoaib has quit IRC | 12:53 | |
| DigitalFlux | uksysadmin: keep the blog posts coming man :) | 12:54 |
| *** bcwaldon has joined #openstack | 12:54 | |
| uksysadmin | So you're *the* fan... nice to meet you ;-) | 12:54 |
| DigitalFlux | uksysadmin: I have that desktop wallpaper now as my wallpaper | 12:55 |
| DigitalFlux | uksysadmin: yay you :) | 12:55 |
| uksysadmin | awesome | 12:55 |
| uksysadmin | yeah - I needed something that goes with my openstack sticker on my laptop and my oversized openstack freebie top | 12:56 |
| DigitalFlux | OpenStack sticket, yummy | 12:56 |
| * DigitalFlux HULK WANTS OPENSTACK STICKER | 12:56 | |
| DigitalFlux | :D | 12:56 |
| *** fitzdsl has quit IRC | 12:57 | |
| uksysadmin | lol | 12:57 |
| *** fitzdsl has joined #openstack | 12:57 | |
| DigitalFlux | uksysadmin: did you come across any support for snapshots in OS ? | 12:58 |
| uksysadmin | tbh I've not been looking yet | 12:59 |
| uksysadmin | wallpaper is about my level ;-) | 12:59 |
| DigitalFlux | haha | 12:59 |
| DigitalFlux | But really the posts are pretty cool :), keep pushing | 12:59 |
| ttx | BK_man: ok, I can reproduce now | 12:59 |
| DigitalFlux | annegentle will appreciate those :) | 12:59 |
| BK_man | ttx: nice | 13:00 |
| uksysadmin | I'm working on a centos instance post - taking the euca images and making them work/useful under openstack | 13:00 |
| DigitalFlux | uksysadmin: perfect | 13:00 |
| BK_man | ttx: so, at least two issues with install operation: nova-api.conf installs at /etc instead of /etc/nova and missing tools directory with ajaxterm | 13:00 |
| DigitalFlux | uksysadmin: so what hypervisor are you focusing on ? | 13:01 |
| ttx | BK_man: I still think it never was supported as a way to deploy openstack. Looks like Ubuntnu packaging runs "build" then maps things from there | 13:01 |
| BK_man | ttx: it's a kind of hack, however | 13:02 |
| BK_man | ttx: off to lunch | 13:02 |
| uksysadmin | ha - well I'm rather limited at the mo as I'm waiting on hardware to be commissioned to start more indepth r&d work... so its qemu at the mo... but it will be KVM - don't see any reason why not - its easier than easy to get OpenStack running out of a ubuntu box. | 13:02 |
| ttx | BK_man: I think "install" needs a lot more work than that to produce something you could run from. | 13:02 |
| uksysadmin | but we run a fairly large vmware cluster here so I'm interested in trying openstack with that | 13:03 |
| uksysadmin | just so I can go back to the project and know what direction to take - migrate/complement/pros/cons etc. | 13:03 |
| ttx | BK_man: I agree it should be properly supported... not sure we should hold up 2011.1.1 release until that's implemented though | 13:03 |
| DigitalFlux | uksysadmin: neat | 13:04 |
| DigitalFlux | uksysadmin: I should probably post on KVM+Glance+OpenStack API | 13:04 |
| *** bcwaldon has quit IRC | 13:04 | |
| uksysadmin | yeah started looking at glance so I'd be interested in that | 13:05 |
| uksysadmin | btw - just throwing this out there and not sure of the real nitty gritty details/complexities... but would there be any interest in an openstack "live demo" cd/dvd/boot image? | 13:06 |
| Failbaitr | yes! | 13:06 |
| Failbaitr | even if it where just somewhat like a reference install | 13:06 |
| uksysadmin | :) | 13:06 |
| uksysadmin | I think its feasible - sure it does some copying of files, etc around - but a qemu based openstack nova live image is feasible and could be a good way to play with it | 13:07 |
| ttx | BK_man: I mean, you're not just missing the "tools" directory. | 13:08 |
| DigitalFlux | uksysadmin: I have a puppet module here | 13:08 |
| Failbaitr | I've just setup bunch of virtualbox debian images with storage nodes on top, having those 4 on a live-cd of sorts would have helped understand the setup before installing it | 13:08 |
| * uksysadmin starts to mull over an OpenStack Live CD | 13:09 | |
| DigitalFlux | If it's modular enough it could be something like a 'live demo', but shows you the installer too | 13:09 |
| Failbaitr | and having an understanding of how things work when correctly setup helps with understanding what goed wrong when installing it yourselve | 13:09 |
| DigitalFlux | and configurable too | 13:09 |
| ttx | BK_man: if we are to support "python setup.py install" as a way to deploy a working openstack setup, it needs a lot more work. At this point it supports build and sdist, but that's about it. | 13:11 |
| ttx | soren, mtaylor: opinions ? | 13:11 |
| *** zenmatt has joined #openstack | 13:16 | |
| *** dkocher has joined #openstack | 13:19 | |
| soren | Sorry, was away. | 13:23 |
| * soren catches up | 13:23 | |
| ttx | soren: summary: "python setup.py install" in bexar produces an incomplete setup. We already knew about CA, but tools is also non-deployed | 13:24 |
| ttx | soren: setup.py needs a lot more love to be a supported way of deploying openstack | 13:25 |
| soren | ttx: python setup.py install should install Nova. I know it doesn't, and I consider it a (major) bug. | 13:26 |
| ttx | soren: I agree it's a bug, and should be fixed | 13:26 |
| ttx | I don't think it should hold up 2011.1.1, however. | 13:26 |
| ttx | since we don't even have the start of a fix. | 13:26 |
| DigitalFlux | soren: tyt :) | 13:27 |
| *** wedjat_ has joined #openstack | 13:27 | |
| soren | ttx: Yeah. Let's stop spending more time on Bexar. | 13:27 |
| *** wedjat_ is now known as wedjat | 13:27 | |
| soren | DigitalFlux: "tyt"? | 13:27 |
| ttx | BK_man: if that sounds alright to you, please file a bug on how broken "python setup.py install" currently is. i'll prioritize it and make sure we get it right for Cactus | 13:28 |
| ttx | BK_man: we already have it as a known bug for Bexar in the release notes ("CA" didn't get installed either) | 13:30 |
| ttx | BK_man: patches welcome, btw... since I'm far away from being a setup.py specialist. | 13:31 |
| dkocher | Developer of Cyberduck here. I have an ticket about a different auth context being used for swauth in Bexar at http://trac.cyberduck.ch/ticket/5735 | 13:31 |
| dkocher | Can you point out why a different URI is introduced? | 13:31 |
| dkocher | I have no option to support both without introducing a new graphical configuration option and that is not desired | 13:33 |
| *** omidhdl has quit IRC | 13:34 | |
| DigitalFlux | soren: ttx: no body worked/working on snapshot support, correct ? | 13:36 |
| soren | DigitalFlux: I think you're right. | 13:37 |
| DigitalFlux | soren: But the methods did exist in the API | 13:37 |
| DigitalFlux | soren: so it was just not implemented ? | 13:37 |
| *** bcwaldon has joined #openstack | 13:37 | |
| soren | DigitalFlux: I suppose. | 13:39 |
| *** hggdh has quit IRC | 13:39 | |
| *** westmaas_ has joined #openstack | 13:39 | |
| *** westmaas has quit IRC | 13:41 | |
| *** westmaas_ is now known as westmaas | 13:41 | |
| DigitalFlux | soren: OK, from your own experience, what is the hypervisor with the most focus at this stage ? | 13:42 |
| *** westmaas1 has joined #openstack | 13:42 | |
| *** bcwaldon has quit IRC | 13:42 | |
| *** iammartian_ has joined #openstack | 13:47 | |
| *** dprince has joined #openstack | 13:48 | |
| *** omidhdl has joined #openstack | 13:48 | |
| *** adiantum has joined #openstack | 13:49 | |
| soren | ttx: Grr.... Triggering that rpc thing was incredibly simple under real use. In fact, it was hard to avoid it. It's just rather difficult to make tests for. | 13:50 |
| ttx | soren: if you trivially reproduce it with 2011.1 and can't reproduce it with a reasonable effort in 2011.1.1, I think that's good enough | 13:51 |
| *** iammartian has quit IRC | 13:51 | |
| *** iammartian_ is now known as iammartian | 13:51 | |
| ttx | I understand it's hard to derive a yes/no from a concurrency issue | 13:51 |
| ttx | can prove existence of a bug but not blah blah. | 13:52 |
| *** iammartian_ has joined #openstack | 13:52 | |
| soren | https://bugs.launchpad.net/nova/+bug/716427 last comment actually verifies it in Bexar. | 13:52 |
| uvirtbot | Launchpad bug 716427 in nova/bexar "RPC concurrency problem" [High,Fix committed] | 13:52 |
| soren | But let me try anyway. | 13:53 |
| BK_man | ttx: there actually is a bug #709187 filed on nova. Should I post another bug instead? | 13:54 |
| uvirtbot | Launchpad bug 709187 in nova "AjaxTerm refer to a unexistant folder "tools" with the Ubuntu packaging" [Medium,Confirmed] https://launchpad.net/bugs/709187 | 13:54 |
| ttx | BK_man: it's a bit different. It complains about files missing from ubuntu packaging | 13:55 |
| ttx | BK_man: I'd rather have a "python setup.py install sucks" bug :) | 13:55 |
| BK_man | ttx: ok :) | 13:55 |
| *** iammartian has quit IRC | 13:55 | |
| *** iammartian_ is now known as iammartian | 13:55 | |
| *** ppetraki has joined #openstack | 13:57 | |
| *** hggdh has joined #openstack | 13:58 | |
| *** hggdh has joined #openstack | 13:58 | |
| BK_man | ttx: https://bugs.launchpad.net/nova/+bug/727794 | 14:01 |
| uvirtbot | Launchpad bug 727794 in nova ""python setup.py install" actually does not install a lot stuff from tarball" [Undecided,New] | 14:01 |
| ttx | BK_man: will triage it | 14:01 |
| soren | ttx: Ok, verified. | 14:02 |
| ttx | soren: ok, just say so in the bug and we're all set | 14:03 |
| ttx | thanks | 14:03 |
| soren | ttx: But to be honest, I don't think there was ever any quetion that this fixed the problem. It's whether there are unwanted side-effects, really. And I can't really test for that. | 14:03 |
| ttx | soren: agreed... | 14:04 |
| *** slyphon has joined #openstack | 14:04 | |
| *** ramkrsna__ has quit IRC | 14:10 | |
| *** hazmat has quit IRC | 14:11 | |
| uvirtbot | New bug: #727794 in nova ""python setup.py install" actually does not install a lot stuff from tarball" [High,Confirmed] https://launchpad.net/bugs/727794 | 14:12 |
| *** tom___ has joined #openstack | 14:13 | |
| tom___ | hi | 14:13 |
| *** RobertLaptop has quit IRC | 14:14 | |
| ttx | sandywalsh_: let me know when you finalized the python-novaclient renaming, so that I can start working on the packaging. | 14:17 |
| wedjat | hi guys | 14:18 |
| wedjat | is it possible in swift to read only a part of an object ? (with an offset or something ?) | 14:18 |
| sandywalsh_ | ttx, I promised eday I'd give it a shot to put it in nova ... that's what I'm working through right now. | 14:22 |
| sandywalsh_ | ttx, if it doesn't go well, I'll just rename the external package | 14:22 |
| ttx | sandywalsh_: one problem with adding files is to adjust setup.py in consequence | 14:23 |
| ttx | sandywalsh_: we haven't done that in the past, and now it's a bit broken, as the backlog will show | 14:23 |
| sandywalsh_ | ttx, which setup.py ... the nova one or the novaclient one? | 14:23 |
| * soren pauses | 14:24 | |
| ttx | sandywalsh_: the nova one might pick up files in novaclient | 14:24 |
| ttx | (or not) | 14:24 |
| sandywalsh_ | ttx, wasn't planning on it doing that, yet :) | 14:24 |
| *** hggdh has quit IRC | 14:25 | |
| sandywalsh_ | ttx, think that will float? | 14:25 |
| *** hggdh has joined #openstack | 14:25 | |
| *** hggdh has joined #openstack | 14:25 | |
| *** kashyap has joined #openstack | 14:25 | |
| *** aliguori has joined #openstack | 14:25 | |
| sandywalsh_ | soren.notify() | 14:25 |
| ttx | sandywalsh_: I don't know... just saying it's often overlooked when someone says "it's as simple as adding files to bzr" | 14:25 |
| sandywalsh_ | ttx, agreed. I knew this wasn't going to be trivial "stick it in there" thing | 14:26 |
| sandywalsh_ | ttx I'll push a little longer and bail if it gets yucky | 14:27 |
| sandywalsh_ | ttx so far it's not too bad. | 14:27 |
| tom___ | I've configured nova in VLAN network mode and i started 1 instance but it doesn't running, nova-network log: 2011-03-02 15:09:32,352 DEBUG nova.linux_net [-] Starting VLAN inteface vlanNone from MainProcess (pid=10171) ensure_vlan /usr/lib/pymodules/python2.6/nova/network/linux_net.py:177 2011-03-02 15:09:32,398 CRITICAL nova.root [-] Unexpected error while running command. Command: sudo vconfig add eth0 None Exit code: 3 Stdout: ' | 14:29 |
| *** Ephur has joined #openstack | 14:29 | |
| *** kazu has joined #openstack | 14:29 | |
| *** tom___ has left #openstack | 14:30 | |
| *** aThom has joined #openstack | 14:31 | |
| aThom | hi, i'm Tom just i reconnected another user name | 14:31 |
| *** sandywalsh_ is now known as sandywalsh | 14:31 | |
| aThom | so, can you help me? | 14:32 |
| aThom | nova-network log: http://paste.openstack.org/show/787/ | 14:34 |
| aThom | nova.conf: http://paste.openstack.org/show/788/ | 14:35 |
| sandywalsh | ttx, I'm going to keep it out of nova for now. Getting the novaclient setup.py to work with the new directory structure is going to take a bigger effort. | 14:36 |
| sandywalsh | ttx, I'll ping you when the github repo is updated | 14:36 |
| ttx | sandywalsh: ack | 14:36 |
| *** pvo has joined #openstack | 14:37 | |
| *** EndEng|Desktop has quit IRC | 14:38 | |
| *** rcc has joined #openstack | 14:40 | |
| *** hvaldivia has joined #openstack | 14:40 | |
| *** pvo has quit IRC | 14:41 | |
| hvaldivia | Hi everybody | 14:41 |
| hvaldivia | I have troubles with my network | 14:41 |
| rcc | soren: hi soren, thanks for clarifying, just added a commit message to https://code.launchpad.net/~rcc/nova/lp715424/+merge/51819 , could you please merge when you have time? | 14:41 |
| hvaldivia | When I install openstack on my cluster my nodes lose their ip. | 14:42 |
| BK_man | hvaldivia: ??? what does it mean? | 14:43 |
| *** f4m8 is now known as f4m8_ | 14:44 | |
| hvaldivia | The physical nodes lose their ip, and I need to restart the interface manually | 14:44 |
| *** reldan has quit IRC | 14:44 | |
| *** bcwaldon has joined #openstack | 14:44 | |
| hvaldivia | My nodes use DHCP to get their ip. Now after I install openstack and execute some vm on the cloud, suddenly some nodes lose their ips | 14:45 |
| *** viirya has quit IRC | 14:46 | |
| hvaldivia | I thought that it was my network, But when I unistall the openstack | 14:46 |
| hvaldivia | There is no problems with my ips | 14:46 |
| *** viirya has joined #openstack | 14:47 | |
| BK_man | hvaldivia: is there any reason why you need DHCP on your cluster? | 14:48 |
| hvaldivia | No, my network is connected to DHCP Server. I am not network admin | 14:49 |
| hvaldivia | Should I use static ip in my machines? | 14:50 |
| BK_man | hvaldivia: I have no problems with static adresses. However, I'm on RHEL6 | 14:51 |
| *** deshantm_xenorg has quit IRC | 14:51 | |
| hvaldivia | I'm on ubuntu :( | 14:51 |
| *** pvo has joined #openstack | 14:51 | |
| hvaldivia | I will try to use static ip | 14:52 |
| *** deshantm_xenorg has joined #openstack | 14:52 | |
| *** troytoman-away is now known as troytoman | 14:53 | |
| *** reldan has joined #openstack | 14:53 | |
| *** daveiw has left #openstack | 14:53 | |
| *** EndEng|Desktop has joined #openstack | 14:54 | |
| *** dirakx has quit IRC | 14:54 | |
| *** pvo has quit IRC | 14:55 | |
| *** vvuksan has joined #openstack | 14:55 | |
| *** londo_ has joined #openstack | 14:59 | |
| BK_man | Khm... What is Nova Direct API and how it relates to Nova API? | 14:59 |
| *** troytoman has quit IRC | 15:03 | |
| *** troytoman-away has joined #openstack | 15:04 | |
| *** troytoman-away is now known as troytoman | 15:04 | |
| *** RobertLaptop has joined #openstack | 15:05 | |
| *** zul has joined #openstack | 15:06 | |
| *** olafont has joined #openstack | 15:09 | |
| *** omidhdl has quit IRC | 15:10 | |
| *** omidhdl has joined #openstack | 15:10 | |
| *** herki has quit IRC | 15:12 | |
| *** hvaldivia1 has joined #openstack | 15:13 | |
| *** aThom has quit IRC | 15:16 | |
| *** dkocher has quit IRC | 15:16 | |
| *** hvaldivia has quit IRC | 15:16 | |
| *** pyros_xx_ has quit IRC | 15:17 | |
| *** willtarner has joined #openstack | 15:17 | |
| willtarner | hello | 15:17 |
| zul | hey is it possible to update the version of libvirt in the ppa, it would make things like testing lxc a bit easier | 15:21 |
| zul | ttx: ^^^ | 15:22 |
| *** grapex has joined #openstack | 15:22 | |
| ttx | zul: I think soren did that already | 15:22 |
| zul | ttx: sorry i missed the meeting yesterday i was getting ready for travel today | 15:24 |
| *** m_3 has quit IRC | 15:26 | |
| *** littleidea has joined #openstack | 15:28 | |
| *** m_3 has joined #openstack | 15:33 | |
| *** rnirmal has joined #openstack | 15:34 | |
| *** viirya has quit IRC | 15:35 | |
| zul | ttx: they newer natty images for uec has lxc support now as well | 15:36 |
| *** viirya has joined #openstack | 15:37 | |
| deshantm_xenorg | anybody around? I'm trying to work around this bug manually https://bugs.launchpad.net/nova/+bug/723897 | 15:38 |
| uvirtbot | Launchpad bug 723897 in nova "'No user for access key admin' when using the Openstack API" [Undecided,Fix committed] | 15:38 |
| *** imsplitbit has joined #openstack | 15:38 | |
| deshantm_xenorg | I just need to understand what I need to pass or put in env for it to work | 15:38 |
| deshantm_xenorg | I can get the access key and secret key from mysql | 15:39 |
| deshantm_xenorg | I could alternatively pull from trunk, but I'm not sure what needs to be done after that | 15:40 |
| *** daveiw has joined #openstack | 15:41 | |
| doude | Hi all, I've got a problem with the serial console with libvirt. I copy the Nova XML template of libvirt instances and I set manually the parameters. I create a domain with 'virsh' command but I cannot acces to the serial console through the command 'virsh console mydomain' | 15:44 |
| doude | If I remove the part '<serial type="file">' in ht e XML file, the serial console is available | 15:45 |
| *** omidhdl has quit IRC | 15:46 | |
| *** viirya has quit IRC | 15:50 | |
| *** mahadev has joined #openstack | 15:53 | |
| *** blueadept has joined #openstack | 15:59 | |
| *** 77CAATQDF has quit IRC | 16:01 | |
| *** pvo has joined #openstack | 16:03 | |
| *** DigitalFlux has quit IRC | 16:04 | |
| *** msassak has joined #openstack | 16:04 | |
| *** littleidea has quit IRC | 16:05 | |
| *** mahadev has quit IRC | 16:05 | |
| *** viirya has joined #openstack | 16:05 | |
| *** mahadev has joined #openstack | 16:12 | |
| *** dendrobates is now known as dendro-afk | 16:14 | |
| *** viirya has quit IRC | 16:14 | |
| deshantm_xenorg | anybody around? I just got past the auth problem by pulling the fix | 16:15 |
| jarrod | :) | 16:15 |
| deshantm_xenorg | now I need to figure out how to boot the image | 16:15 |
| deshantm_xenorg | I am running the command: | 16:16 |
| deshantm_xenorg | novatools boot test --flavor 2 --image 3 | 16:16 |
| jarrod | i was doing euca-run-instances -k mykey -t m1.small 3 | 16:16 |
| *** dendro-afk is now known as dendrobates | 16:16 | |
| jarrod | and it would boot from glance, but xen would not find any bootable media | 16:17 |
| *** uksysadmin has quit IRC | 16:17 | |
| deshantm_xenorg | jarrod: what do you mean xen wouldn't find any bootable media? | 16:17 |
| jarrod | i would run that command, openstack would report the instance running | 16:18 |
| sandywalsh | ttx https://github.com/rackspace/python-novaclient | 16:18 |
| jarrod | but when i got on xencenter and pulled up the console | 16:18 |
| jarrod | it said couldnt find any bootable media | 16:18 |
| jarrod | but, it almost worked :) | 16:19 |
| jarrod | instead of using nova-tools | 16:19 |
| deshantm_xenorg | jarrod: so you are using XenServer? or XCP? | 16:19 |
| jarrod | xenserver | 16:19 |
| jarrod | 5.6fp1 | 16:19 |
| ttx | sandywalsh: ack | 16:20 |
| deshantm_xenorg | jarrod: yeah I have the same symptoms on XCP now | 16:22 |
| jarrod | deshantm i think its somethign with it trying to use /dev/xvdX | 16:23 |
| deshantm_xenorg | jarrod: ok so we should be able to modify the file system and re upload the image to glance | 16:24 |
| *** viirya has joined #openstack | 16:26 | |
| *** daveiw has left #openstack | 16:28 | |
| *** mgoldmann has quit IRC | 16:28 | |
| *** dubsquared has joined #openstack | 16:28 | |
| dubsquared | morning, #openstack! anyone familiar with swift stats that can answer some questions for me? | 16:28 |
| *** zul has quit IRC | 16:29 | |
| *** CloudChris has joined #openstack | 16:32 | |
| deshantm_xenorg | jk0: the XenServer image is not booting from http://wiki.openstack.org/XenServerDevelopment | 16:32 |
| *** huismon_ has quit IRC | 16:33 | |
| jk0 | deshantm_xenorg: is there any indictation on why it's not booting in the logs? | 16:33 |
| deshantm_xenorg | jk0: just fyi we had to use euca-run-instances instead of novatools | 16:33 |
| dubsquared | deshantm_xenorg: ive found that image is missing xentools, and thats causing some problems with / mounting | 16:33 |
| deshantm_xenorg | dubsquared: so can we add those in? | 16:34 |
| *** kashyap has quit IRC | 16:34 | |
| dubsquared | but it could be unrelated, but something i ran into... | 16:34 |
| deshantm_xenorg | people are starting to use this image for testing | 16:34 |
| jk0 | there are instructions on the wiki for installing xentools on images | 16:34 |
| deshantm_xenorg | sure, I could do it but it is hosted somewhere | 16:35 |
| deshantm_xenorg | who has the access to change the one that we are pointing people to? | 16:35 |
| deshantm_xenorg | i still need to check the logs | 16:35 |
| *** skiold has quit IRC | 16:40 | |
| deshantm_xenorg | rk0: i see the xentools instructions on that page now, but does that mean they are out of order? | 16:43 |
| deshantm_xenorg | did you get it to boot before installing xentools? | 16:43 |
| jk0 | yep | 16:46 |
| jk0 | it doesn't need xentools to boot | 16:46 |
| jk0 | you may actually be better off using your own image for testing. | 16:46 |
| jk0 | that image is starting to age | 16:46 |
| deshantm_xenorg | jk0: what version of XenServer are you booting it on? | 16:46 |
| jk0 | we're using out own image for testing | 16:46 |
| jk0 | *our | 16:46 |
| jk0 | but we use 5.6fp1 | 16:46 |
| deshantm_xenorg | I'm trying to find anything in the xapi logs | 16:46 |
| mtaylor | soren, ttx: I agree it's a major bug - although it seems you've already agreed with me | 16:47 |
| deshantm_xenorg | jk0: anywhere else I should look? | 16:47 |
| deshantm_xenorg | jk0: is there anyway we could put up a better image? | 16:47 |
| ttx | mtaylor: not really a regression in 2011.1.1, though | 16:47 |
| *** burris has quit IRC | 16:47 | |
| deshantm_xenorg | jk0: it may be booted actually | 16:51 |
| *** littleidea has joined #openstack | 16:51 | |
| deshantm_xenorg | just not showing anything on the vnc console usefule other than boot device hard drive sucess | 16:51 |
| deshantm_xenorg | I'm try to get a command line console | 16:52 |
| dubsquared | you have xencenter installed? | 16:52 |
| dubsquared | and using that for console, as well? | 16:52 |
| deshantm_xenorg | yes | 16:52 |
| deshantm_xenorg | but that is a vnc console | 16:52 |
| deshantm_xenorg | actually i don't see a vncterm for the id of my openstack lucid image | 16:54 |
| dubsquared | oh yeah, you're right.. mixing up kvm/xenserver vnc stuff | 16:54 |
| vishy | soren: are you here? | 16:55 |
| *** dragondm has joined #openstack | 16:55 | |
| soren | vishy: Very, very briefly. | 16:56 |
| vishy | soren: just curious about bug 727705 | 16:56 |
| uvirtbot | Launchpad bug 727705 in nova "Floating ip's do not get disassociated from fixed ip's when the fixed ip's are disassociated from instances" [Low,Triaged] https://launchpad.net/bugs/727705 | 16:56 |
| dubsquared | you should see something like this in compute —> http://paste.openstack.org/show/789/ | 16:56 |
| soren | vishy: Ah. | 16:56 |
| *** ironcamel2 has quit IRC | 16:56 | |
| deshantm_xenorg | rk0: I'm confused at course of action here | 16:57 |
| soren | vishy: He had created the floating ip's using "nova-manage floating create <wrong host name> <subnet info>" | 16:57 |
| *** westmaas has quit IRC | 16:57 | |
| vishy | soren: gotcha, I'm just curious how the hell the ip got associated in the first place | 16:57 |
| soren | vishy: ...which works brilliantly except in this one case. | 16:57 |
| soren | vishy: I explain that,t oo. | 16:57 |
| soren | vishy: In the bug. | 16:58 |
| vishy | ah i see now | 16:58 |
| soren | vishy: Because only in the case where we're disassociating as part.. | 16:58 |
| soren | right. | 16:58 |
| dubsquared | deshantm_xenorg: i would make a new ubuntu image, 10.04, and i would install xentools on it, and then bundle it up | 16:58 |
| vishy | so associate is using fixed_ip host | 16:58 |
| soren | Yeah. | 16:58 |
| *** ironcamel2 has joined #openstack | 16:58 | |
| vishy | i think they both should be using floating_ip host | 16:58 |
| *** westmaas has joined #openstack | 16:58 | |
| soren | And so is disassociate called directly. | 16:58 |
| soren | ...but not when it's being called as part of termination. | 16:58 |
| vishy | hmm, well let me think about that | 16:59 |
| soren | It' probably the others that are wrong, really. | 16:59 |
| deshantm_xenorg | dubsquared is there a wiki page on this? | 16:59 |
| soren | If floating_ip has a host field, it should be what we use to make these rpc calls. | 16:59 |
| *** markwash has joined #openstack | 16:59 | |
| soren | If not, then it shouldn't have a host file.d | 16:59 |
| soren | field, even. | 16:59 |
| * soren wanders off to tinnder | 16:59 | |
| soren | err... | 16:59 |
| vishy | k | 16:59 |
| soren | dinner. | 16:59 |
| * vishy thinks about this issue | 16:59 | |
| * soren types poorly while walking | 16:59 | |
| deshantm_xenorg | dubsquared or just just use ubuntu-vm-builder? | 17:00 |
| dubsquared | i don't use that personally … euca-bundle-*, euca-register, etc | 17:01 |
| dubsquared | looking for a wiki page... | 17:01 |
| *** joearnold has joined #openstack | 17:01 | |
| *** rnirmal has quit IRC | 17:02 | |
| dubsquared | hmm, so this should be useful: http://wiki.openstack.org/CreatingRHELImages | 17:02 |
| dubsquared | the steps at the bottom | 17:02 |
| *** CloudChris has quit IRC | 17:02 | |
| deshantm_xenorg | thanks | 17:02 |
| *** dprince has quit IRC | 17:03 | |
| *** bcwaldon has quit IRC | 17:03 | |
| *** grapex has quit IRC | 17:03 | |
| *** grapex1 has joined #openstack | 17:03 | |
| deshantm_xenorg | dubsquared: the question is still what is the quickest way to get a base image that will be usable for that? | 17:04 |
| *** markwash has quit IRC | 17:06 | |
| *** dubsquared1 has joined #openstack | 17:06 | |
| *** dubsquared has quit IRC | 17:06 | |
| deshantm_xenorg | dubsquared1:i think you missed this since your session died dubsquared: the question is still what is the quickest way to get a base image that will be usable for that? | 17:07 |
| dubsquared1 | you could install ubuntu in virtualbox via iso, probably the quickest way | 17:09 |
| dubsquared1 | if you don't have the install iso, just download the net install and do it that way | 17:09 |
| dubsquared1 | i have an old box that is sitting around, so ive just installed that, and configured how i wanted it | 17:10 |
| *** arun has joined #openstack | 17:10 | |
| dubsquared1 | i have a short script that i have been using that you can drop on a server that you want to be an image | 17:10 |
| dubsquared1 | it's been a while since ive used it, maybe about a month now, but if you want to look at it for ref, or it might still work…i can share | 17:11 |
| *** mahadev has quit IRC | 17:13 | |
| *** dprince has joined #openstack | 17:22 | |
| *** smaresca has joined #openstack | 17:22 | |
| *** littleidea has quit IRC | 17:31 | |
| *** burris has joined #openstack | 17:39 | |
| *** whaley has left #openstack | 17:40 | |
| *** gregp76 has joined #openstack | 17:42 | |
| dabo | can someone explain to me the meaning/intent of 'availability_zone' in nova/scheduler/zone.py? | 17:45 |
| *** Nacx has quit IRC | 17:47 | |
| eday | dabo: that's the old availability zone stuff | 17:47 |
| dabo | ok, but what does it represent? I missed whatever discussions there were about it | 17:48 |
| eday | dabo: there are currently two "zone" mechanisms, sandy's new stuff, and the availability_zone assigned to service, instance, and volume records | 17:48 |
| eday | it's a string | 17:48 |
| dabo | is it still in use? | 17:49 |
| dabo | i.e., still being developed? | 17:49 |
| eday | I think some folks are using it, yes. it's in the simple scheduler (basically launch if zone == or != existing zone) | 17:50 |
| eday | we need to fold that functionality into the new zone work | 17:50 |
| *** gregp76 has quit IRC | 17:51 | |
| dabo | eday: I'll be talking with sandywalsh later today, so I'll mention that to him. | 17:51 |
| *** irahgel has quit IRC | 17:52 | |
| *** MotoMilind has joined #openstack | 17:56 | |
| *** irahgel has joined #openstack | 17:56 | |
| *** littleidea has joined #openstack | 17:59 | |
| *** MotoMilind has quit IRC | 18:00 | |
| *** MotoMilind has joined #openstack | 18:02 | |
| MotoMilind | Q about OpenStack Dashboard on CentOS 5.5. Even though nova-manage lists my project, Dashboard doesn't list it. It may have something to do with Dashboard not being able to locate the nova.conf file (from either location, /etc/nova or /usr/bin/). Can someone point me to a solution, please? Thanks in advance | 18:04 |
| xtoddx | MotoMilind: dashboard talks to the api, so has nothing to do with nova.conf | 18:05 |
| xtoddx | MotoMilind: just check your local/local_settings.py and make sure the endpoint and credentials are correct | 18:05 |
| *** irahgel has left #openstack | 18:05 | |
| kpepple | MotoMilind: make sure you used the correct NOVA_ADMIN_USER | 18:05 |
| *** grapex has joined #openstack | 18:05 | |
| MotoMilind | I see, ok thanks. | 18:05 |
| *** grapex1 has quit IRC | 18:05 | |
| *** Ryan_Lane has joined #openstack | 18:06 | |
| *** dendrobates is now known as dendro-afk | 18:13 | |
| *** kbringard has joined #openstack | 18:13 | |
| *** bwalker7125 has joined #openstack | 18:14 | |
| kbringard | I have a quick ? about glance | 18:16 |
| kbringard | when the functionality gets extended to work with the ec2 api, are the certs still going to be required? | 18:16 |
| *** mahadev has joined #openstack | 18:18 | |
| *** littleidea has quit IRC | 18:18 | |
| *** Ryan_Lane has quit IRC | 18:22 | |
| MotoMilind | Ok. My endpoint and credentials are correct (purposely changed those to wrong values and I got stack trace on the webUI.) The admin user value seems to be unimportant. I can put anything there, but UI still works. In either case, I still don't see my project. Thanks. | 18:24 |
| BK_man | # euca-get-ajax-console --debug i-0000000c | 18:24 |
| *** nelson has quit IRC | 18:24 | |
| BK_man | EC2Connection instance has no attribute 'get_ajax_console' | 18:24 |
| BK_man | what's wrong? | 18:25 |
| *** nelson has joined #openstack | 18:25 | |
| BK_man | I'm on 2011.1.1-rc now | 18:25 |
| kpepple | MotoMilind: hold on. let me look at my install. | 18:27 |
| MotoMilind | Sure, thanks | 18:27 |
| bwalker7125 | QQ: are there any known issues with running swift on a 32-bit Ubuntu OS instead of the 64-bit version? | 18:28 |
| *** dendro-afk is now known as dendrobates | 18:30 | |
| *** dubsquared has joined #openstack | 18:32 | |
| kpepple | MotoMilind: in your nova-api log, what user is it logging with ? you'll find some lines starting with "DEBUG:boto:Path:" when the admin client call nova-api | 18:33 |
| *** dubsquared1 has quit IRC | 18:34 | |
| *** dubsquared has quit IRC | 18:34 | |
| MotoMilind | Actually, I couldn't locate the nova-api.log in /var/log/nova folder, it just has twistd.log | 18:34 |
| *** dubsquared has joined #openstack | 18:34 | |
| kpepple | MotoMilind: are you running nova-* (api, compute, etc.) on the same server as the dashboard ? | 18:35 |
| MotoMilind | yes | 18:35 |
| kpepple | MotoMilind: okay. few questions: what version of nova (trunk, bexar, 2011.1.1) ? are you following the directions on http://wiki.openstack.org/OpenStackDashboard ? | 18:37 |
| MotoMilind | Let me confirm | 18:39 |
| BK_man | has anybody tested ajax console in Bexar? | 18:39 |
| sleepsonthefloor | doude - regarding removing "<serial type="file">" to enable the console - if you do that, it will disable the console log. If you leave that line in the config, you can still access the console, but you will need to find the tty using virsh dumpxml. | 18:41 |
| MotoMilind | Interesting. I followed the doc from your manuals. The manual is very close except for following lines: NOVA_ADMIN_USER: this can be any user with admin privileges in your nova database. The CLOUD_SERVERS_USERNAME from your admin credentials file is fine. | 18:42 |
| MotoMilind | NOVA_PROJECT: this can be any project (defined in your nova database) which the NOVA_ADMIN_USER is defined as project_manager. Refer to http://wiki.openstack.org/RunningNova for assistance if you haven't defined any nova projects. | 18:42 |
| MotoMilind | Let me confirm that the user I have does indeed have admin privileges on the project that I am hoping to see | 18:42 |
| MotoMilind | Or I will try to make him an admin | 18:42 |
| *** zul has joined #openstack | 18:45 | |
| *** lvaughn_ has quit IRC | 18:46 | |
| *** lvaughn has joined #openstack | 18:47 | |
| sleepsonthefloor | BK_man - there are packaging issues with ajaxterm still - but I can help you get it running. You can read the original merge proposal here - https://code.launchpad.net/~anso/nova/ajaxterm/+merge/45068 | 18:49 |
| BK_man | sleepsonthefloor: I already did this. And I have my own packaging on RHEL :) | 18:49 |
| BK_man | sleepsonthefloor: so Ubuntu packaging issues does not bother me - discussed them today's morning with ttx | 18:50 |
| BK_man | sleepsonthefloor: how can I try to get ajax console with OpenStack API CLI? | 18:51 |
| BK_man | # stack compute get_ajax_console i-0000000c | 18:51 |
| BK_man | Traceback (most recent call last): | 18:51 |
| BK_man | File "/usr/bin/stack", line 146, in <module> | 18:51 |
| BK_man | key, value = x.split('=', 1) | 18:51 |
| BK_man | ValueError: need more than 1 value to unpack | 18:51 |
| BK_man | 18:51 | |
| *** dubsquared has quit IRC | 18:52 | |
| *** michael364 has joined #openstack | 18:52 | |
| michael364 | hey | 18:52 |
| sleepsonthefloor | BK_man - so, to avoid confusion, I believe that stack is the "Direct API" cli, despite the name. The OS api cli is presently called "novatools" | 18:56 |
| BK_man | sleepsonthefloor: thanks for clarification. | 18:56 |
| sleepsonthefloor | BK_man: but first, what version of euca2ools are you running? | 18:57 |
| sleepsonthefloor | I'm wondering about you "EC2Connection instance has no attribute 'get_ajax_console'" error | 18:57 |
| BK_man | sleepsonthefloor: euca2ools-1.3.1-2.el6.noarch | 18:57 |
| BK_man | sleepsonthefloor: should I update it? to which version? | 18:57 |
| sleepsonthefloor | BK_man - you probably need to downgrade it to 1.2 - it is possible that they re-architected in 1.3.1. | 18:59 |
| *** hvaldivia1 has quit IRC | 18:59 | |
| BK_man | sleepsonthefloor: thanks for suggestion, will try | 19:00 |
| *** dprince has quit IRC | 19:00 | |
| *** magritte has joined #openstack | 19:03 | |
| *** dprince has joined #openstack | 19:04 | |
| *** gregp76 has joined #openstack | 19:05 | |
| *** ChanServ sets mode: +v _cerberus_ | 19:05 | |
| *** rcc has quit IRC | 19:13 | |
| *** vvuksan has quit IRC | 19:13 | |
| *** photron has joined #openstack | 19:13 | |
| *** lvaughn_ has joined #openstack | 19:14 | |
| *** bwalker7125 has quit IRC | 19:15 | |
| *** mgoldmann has joined #openstack | 19:17 | |
| *** lvaughn has quit IRC | 19:17 | |
| blamar | Can someone give me some insight as to why the OpenStack Direct API is needed? Is it just for quick testing? Or is it meant to be deployed in production? | 19:24 |
| xtoddx | blamar: both/either | 19:27 |
| xtoddx | blamar: i use it for testing, right now it is missing auth | 19:28 |
| xtoddx | blamar: but the stack command is super-easy when doing things locally | 19:28 |
| *** johnpur has joined #openstack | 19:29 | |
| *** ChanServ sets mode: +v johnpur | 19:29 | |
| *** dprince_ has joined #openstack | 19:29 | |
| blamar | xtoddx, Ideally, wouldn't the openstack API be used? It should, once again ideally, have a full set of features and thus negating the need for direct API? | 19:29 |
| xtoddx | blamar: the direct api actually exposes the way openstack works internally, so i like it. | 19:31 |
| blamar | xtoddx, yeah, I can see the appeal of calling things directly, thanks :) | 19:32 |
| *** dprince has quit IRC | 19:32 | |
| *** dprince_ has quit IRC | 19:32 | |
| *** MarcMorata has quit IRC | 19:35 | |
| *** dprince has joined #openstack | 19:40 | |
| *** zul has quit IRC | 19:46 | |
| *** jesse__ has joined #openstack | 19:52 | |
| *** jesse__ is now known as anotherjesse | 19:52 | |
| *** michael364 has left #openstack | 19:53 | |
| *** mgoldmann_ has joined #openstack | 19:55 | |
| *** mgoldmann_ has quit IRC | 19:58 | |
| *** mgoldmann_ has joined #openstack | 19:58 | |
| *** mgoldmann has quit IRC | 19:58 | |
| *** gregp76 has quit IRC | 19:59 | |
| *** dendrobates is now known as dendro-afk | 20:00 | |
| *** zul has joined #openstack | 20:00 | |
| *** dendro-afk is now known as dendrobates | 20:02 | |
| *** grapex1 has joined #openstack | 20:03 | |
| *** grapex has quit IRC | 20:03 | |
| *** anotherjesse has quit IRC | 20:04 | |
| *** joearnold has quit IRC | 20:05 | |
| *** jaypipes has joined #openstack | 20:06 | |
| *** blamar has quit IRC | 20:06 | |
| *** MarkusT has joined #openstack | 20:10 | |
| MarkusT | I'm currently running nova bexar on Ubuntu 10.04. It's working great with a local IP range (192.168.x.x) and now I'd love to connect it to the internet (via euca-associacte-address, etc.). I'm able to use 4 public IPs my hoster provides me, but I can't figure out how to configure networking (bridge, routing). Any pointers to a similar configuration? | 20:11 |
| *** imsplitbit has quit IRC | 20:15 | |
| *** imsplitbit has joined #openstack | 20:15 | |
| MotoMilind | My problem is solved. The user "admin" was not associated to the project, so I inserted a row in the user_project_associations table and it worked. I couldn't find the correct nova-manage command to do that on the CLI. Is there a way? | 20:19 |
| *** h0cin has quit IRC | 20:21 | |
| soren | MotoMilind: "nova-manage role" is the magic trick, I believe. | 20:24 |
| BK_man | sleepsonthefloor: downgraded to euca2ools 1.2 - old error is gone, have a new one! :) | 20:27 |
| BK_man | # euca-get-ajax-console i-0000000c | 20:27 |
| BK_man | No handlers could be found for logger "nova.root" | 20:27 |
| *** naehring has quit IRC | 20:27 | |
| soren | vishy, sleepsonthefloor: I think noVNC is nifty, but I really wish we could provide direct VNC access. | 20:27 |
| BK_man | sleepsonthefloor: that's all. Should I get an URL? | 20:27 |
| sleepsonthefloor | BK_man - havne't seen that one yet - it should return a url | 20:28 |
| sleepsonthefloor | soren - I'll look into it - we gravitated to noVNC because of the dashboard integration, which is great for demos :) | 20:32 |
| soren | sleepsonthefloor: Definitely. I wish gtkvnc-mozilla was more ubiquitous. | 20:33 |
| *** patcoll has joined #openstack | 20:33 | |
| sleepsonthefloor | BK_man - i'll look into that error shortly - on a call | 20:34 |
| *** dragondm has quit IRC | 20:36 | |
| *** clauden_ has joined #openstack | 20:46 | |
| *** dendrobates is now known as dendro-afk | 20:48 | |
| *** dendro-afk is now known as dendrobates | 20:56 | |
| *** reldan has quit IRC | 21:00 | |
| BK_man | sleepsonthefloor: hint for euca-get-ajax-console: | 21:00 |
| BK_man | # euca-get-ajax-console | 21:00 |
| BK_man | 2011-03-03 00:00:03,461 CRITICAL nova.root [-] cannot import name ConnectionFailed | 21:00 |
| BK_man | (nova.root): TRACE: Traceback (most recent call last): | 21:00 |
| BK_man | (nova.root): TRACE: File "/usr/bin/euca-get-ajax-console", line 40, in <module> | 21:00 |
| BK_man | (nova.root): TRACE: from euca2ools import Euca2ool, InstanceValidationError, Util, ConnectionFailed | 21:00 |
| BK_man | (nova.root): TRACE: ImportError: cannot import name ConnectionFailed | 21:00 |
| BK_man | (nova.root): TRACE: | 21:00 |
| BK_man | sleepsonthefloor: I have euca2ools-1.2-2.el6.noarch installed by your advice | 21:01 |
| *** markwash has joined #openstack | 21:02 | |
| sleepsonthefloor | BK_man - 1.2-0ubuntu11.1 is what is on ubuntu. | 21:03 |
| markwash | anybody know why the nova compute quota generally returns "min(actual quota, requested amount)" instead of just "actual quota" | 21:04 |
| *** mahadev has quit IRC | 21:04 | |
| *** adiantum has quit IRC | 21:04 | |
| *** mahadev has joined #openstack | 21:06 | |
| *** gregp76 has joined #openstack | 21:09 | |
| *** dendrobates is now known as dendro-afk | 21:11 | |
| *** dendro-afk is now known as dendrobates | 21:11 | |
| *** jaypipes has quit IRC | 21:12 | |
| *** mahadev_ has joined #openstack | 21:16 | |
| *** mahadev has quit IRC | 21:18 | |
| *** lamar has joined #openstack | 21:18 | |
| *** littleidea has joined #openstack | 21:19 | |
| *** pothos_ has joined #openstack | 21:21 | |
| *** mgoldmann_ has quit IRC | 21:23 | |
| *** pothos has quit IRC | 21:23 | |
| *** pothos_ is now known as pothos | 21:24 | |
| *** smaresca is now known as icarus901 | 21:24 | |
| *** icarus901 is now known as smaresca | 21:24 | |
| *** littleidea has quit IRC | 21:27 | |
| *** joearnold has joined #openstack | 21:27 | |
| *** joearnold has joined #openstack | 21:29 | |
| *** dendrobates is now known as dendro-afk | 21:29 | |
| *** lamar has quit IRC | 21:32 | |
| *** dendro-afk is now known as dendrobates | 21:32 | |
| *** arun_ has quit IRC | 21:33 | |
| *** arun_ has joined #openstack | 21:34 | |
| *** allsystemsarego has quit IRC | 21:42 | |
| *** lamar has joined #openstack | 21:42 | |
| nelson | creiht: I'm in need of an intervention here. I've got just three files stored in my cluster, but sometimes when I ask for one of them, it's giving me a different file. | 21:43 |
| *** dprince has quit IRC | 21:43 | |
| nelson | creiht: the crazy thing is that the log file is always giving me the right file. Somehow it's handing back the wrong file. The problem is in my code, because it only happens when I rewrite. | 21:45 |
| *** mahadev_ has quit IRC | 21:45 | |
| *** mahadev has joined #openstack | 21:46 | |
| nelson | creiht: nahhh, nevermind. I just have to get deep enough into the zone to fix this. | 21:47 |
| *** hazmat has joined #openstack | 21:49 | |
| *** vvuksan has joined #openstack | 21:51 | |
| *** dendrobates is now known as dendro-afk | 21:51 | |
| *** lamar has quit IRC | 21:51 | |
| *** Ryan_Lane has joined #openstack | 21:52 | |
| *** troytoman is now known as troytoman-away | 21:53 | |
| *** ctennis_ has quit IRC | 22:00 | |
| *** dendro-afk is now known as dendrobates | 22:02 | |
| *** joearnold has quit IRC | 22:02 | |
| *** miclorb has joined #openstack | 22:04 | |
| *** dendrobates is now known as dendro-afk | 22:04 | |
| *** reldan has joined #openstack | 22:07 | |
| imsplitbit | I wanted to open a discussion on the behavior of utils.execute() or rather the data that is returned | 22:08 |
| *** brd_from_italy has joined #openstack | 22:11 | |
| imsplitbit | assuming that when you do need to shell out a command that is the proper interface for it, I have run into an edge case that has caused me to need to extend that method | 22:12 |
| imsplitbit | I have a situation where I run a shell command, and it is successful however the command throws a warning to standard error. | 22:12 |
| *** ctennis_ has joined #openstack | 22:13 | |
| soren | It considers that an error? | 22:13 |
| *** blueadept has quit IRC | 22:13 | |
| imsplitbit | This is a problem because I really care both about the return status of the code as well as the error message | 22:13 |
| imsplitbit | a less intrusive solution to nova is to do text analysis on the return of stderr | 22:14 |
| imsplitbit | but seems like that is the wrong way of handling that | 22:14 |
| imsplitbit | right now I check if stderr has contents, if it does I log and raise. thats the simple case | 22:15 |
| *** dendro-afk is now known as dendrobates | 22:15 | |
| *** mray has joined #openstack | 22:15 | |
| *** grapex1 has quit IRC | 22:15 | |
| *** grapex has joined #openstack | 22:15 | |
| *** gregp76 has quit IRC | 22:15 | |
| imsplitbit | and I can write special case code to see if the message is one that I would like to ignore but it seems to me that if you are going to wrap the ability to shell out a command then you should also return the return statuscode as well since that is pertinent information to command being run | 22:16 |
| *** photron has quit IRC | 22:16 | |
| imsplitbit | now adding a third element that is returned is a big thing, I realize this since it doesn't appear that python does lazy assignment | 22:17 |
| soren | If the command returns succesfully (ie calls exit(0)), you just get (stdout, stderr) back. If it returns anythign else, it raises an exception. That exception has information about the exit code and also stdout and stderr. | 22:19 |
| imsplitbit | I guess a precursor to my initial question should have been, if I am writing a virt driver for nova and that driver needs to run a shell command, is util.execute what I am supposed to use? | 22:19 |
| soren | I don't completely understand your use case, so I don't see how this isn't sufficient. :( | 22:19 |
| *** occ has joined #openstack | 22:19 | |
| soren | Yes, utils.execute is how we run things. | 22:20 |
| devcamcar | mtaylor: are you around? | 22:20 |
| imsplitbit | ok so given that the command runs, and returns status code 0, but has contents in stderr how do you handle that? | 22:20 |
| imsplitbit | do you examine the contents of stderr and special case that? | 22:20 |
| imsplitbit | and given the relative small size of the object that is the return code, why not just pass that on since you already have it? | 22:21 |
| soren | imsplitbit: I'm not sure. Can you be more concrete? I think that'll help my understanding. | 22:21 |
| *** mahadev_ has joined #openstack | 22:21 | |
| imsplitbit | soren: I am writing an openvz virt driver | 22:21 |
| *** grapex has quit IRC | 22:22 | |
| imsplitbit | soren: when I run the command "vzctl start CTID" it returns 0 as well as some other info | 22:22 |
| soren | imsplitbit: Gotcha. | 22:22 |
| soren | imsplitbit: Ok. | 22:22 |
| imsplitbit | one line of that info is a warning that may or may not be able to be helped from the end users perspective | 22:22 |
| imsplitbit | regarding the hostname | 22:22 |
| *** Ephur has quit IRC | 22:23 | |
| imsplitbit | if the hostname isn't valid it sends a warning | 22:23 |
| imsplitbit | but it is not indicative of a failure | 22:23 |
| soren | Ok. | 22:23 |
| imsplitbit | but because that is sent to stderr it is treated as an error | 22:23 |
| imsplitbit | right now my code just does an if err: | 22:23 |
| imsplitbit | and the way I am working around it is looking at the text in err | 22:23 |
| soren | What treats it like an error, exactly? Maybe that's what I'm not getting. | 22:24 |
| imsplitbit | well when you are writing a script that shells out a command, how do you judge that the command succeeded? | 22:24 |
| imsplitbit | generally I look at return statuscode | 22:24 |
| *** mahadev has quit IRC | 22:24 | |
| soren | Right. | 22:24 |
| imsplitbit | but in the context of openvz and vz commands in utils.execute() that is handled magically | 22:25 |
| imsplitbit | if return status is > 0 I get an exception | 22:25 |
| imsplitbit | so now I have to assume that the command is always successful unless there is something in err | 22:26 |
| imsplitbit | which is generally true | 22:26 |
| imsplitbit | this is certainly an edge case | 22:26 |
| imsplitbit | which is why I see the point of me writing special code around the contents of stderr | 22:26 |
| imsplitbit | but it brought up the philisophical point that the interface is telling me how to operate rather than exposing the data to me and letting me make decisions | 22:28 |
| imsplitbit | I am not saying it is wrong | 22:28 |
| jk0 | kpepple: how's the flavors branch coming? I'd like to get that merged ASAP :) | 22:28 |
| soren | imsplitbit: I think it may be too late for me.. | 22:28 |
| *** gregp76 has joined #openstack | 22:28 | |
| soren | imsplitbit: What data is it that you're missing from utils.execute? | 22:28 |
| imsplitbit | what I am saying, as concise as possible, is how do I solve the problem of handling running a command and verifying that it worked. | 22:29 |
| soren | Well, typically (as you point out yourself), this is expressed by the command's return code. | 22:31 |
| imsplitbit | ok | 22:31 |
| imsplitbit | and if that is the case then what is the point of getting the contents of stderr? | 22:32 |
| imsplitbit | just in case you care? | 22:32 |
| soren | ...which, as you also point out, determines if execute() returns (stdout, stderr) or raises an exception. | 22:32 |
| *** dubsquared has joined #openstack | 22:32 | |
| imsplitbit | ok | 22:32 |
| imsplitbit | I think I better understand it as a command interface | 22:32 |
| imsplitbit | that actually makes a little more sense to me | 22:32 |
| *** mahadev_ has quit IRC | 22:33 | |
| imsplitbit | so rather than run the commands and examine stderr I should be doing a try / except | 22:33 |
| soren | WEll, yes, you get stderr back in case there might be something interesting there. Not all commands use stderr vs. stdout quite as sanely as one might have wished. | 22:33 |
| imsplitbit | yeah apparently | 22:33 |
| imsplitbit | as I have painfully learned | 22:33 |
| *** mahadev has joined #openstack | 22:33 | |
| imsplitbit | but I can use the interface as designed | 22:34 |
| imsplitbit | it was just a different way of looking at it that I just didn't see | 22:34 |
| soren | In the general case, though, stderr isn't interesting if the command was succesful. | 22:34 |
| imsplitbit | I think I have a better grasp of it now. | 22:34 |
| soren | Cool. | 22:34 |
| imsplitbit | thanks | 22:34 |
| soren | Sorry for being dense. :) | 22:34 |
| *** drico has joined #openstack | 22:34 | |
| imsplitbit | it was most likely my fault but thanks | 22:35 |
| soren | np | 22:35 |
| *** gondoi has quit IRC | 22:35 | |
| *** ppetraki has quit IRC | 22:36 | |
| *** bcwaldon has joined #openstack | 22:37 | |
| *** adiantum has joined #openstack | 22:39 | |
| *** kbringard has quit IRC | 22:39 | |
| *** westmaas1 has quit IRC | 22:41 | |
| soren | vishy: Silly question perhaps, but how do you guys actually run the smoketests? | 22:42 |
| *** paltman has quit IRC | 22:43 | |
| *** hazmat has quit IRC | 22:45 | |
| tr3buchet | any db guru's available? maybe not a guru, just not a db moron (like me) | 22:46 |
| soren | We'll know once your question is asked. :) | 22:47 |
| tr3buchet | i'm trying to figure out when i call fixed_ip_associate_pool it replaces the previous allocation instead of adding a new one | 22:47 |
| tr3buchet | (sorry was phrasing it) | 22:47 |
| soren | :) | 22:47 |
| *** zul has quit IRC | 22:47 | |
| *** mustfg has joined #openstack | 22:48 | |
| soren | vishy: never mind :) | 22:51 |
| *** imsplitbit has quit IRC | 22:53 | |
| * soren goes to bed | 22:53 | |
| mustfg | can someone help me plz, i need some help on how to authenticate to the openstack server. | 22:54 |
| mustfg | can the credentials only be created local by the openstack admin or are they created by the clouduser ? | 22:55 |
| kpepple | mustfg: which openstack server: swift (storage) or nova (compute) ? | 22:55 |
| mustfg | compute | 22:55 |
| mustfg | nova | 22:55 |
| kpepple | mustfg: AFAIK credentials (the zipfile) can only be created through the nova-manage command which requires "admin" role | 22:56 |
| jk0 | devcamcar: you around? | 22:56 |
| mustfg | ok | 22:56 |
| mustfg | then i send the credential to user | 22:56 |
| devcamcar | jk0: i am | 22:57 |
| *** adiantum has quit IRC | 22:57 | |
| kpepple | mustfg: yes, you would usually send them the zip file ... | 22:57 |
| dubsquared | mustfg: yeah, you could share the novarc file with an end-user | 22:57 |
| jk0 | devcamcar: would you mind having another look at https://code.launchpad.net/~ken-pepple/nova/dynamicinstancetypes/+merge/50257 please? I'd like to get it merged :) | 22:57 |
| mustfg | ok | 22:57 |
| *** patcoll has quit IRC | 22:57 | |
| devcamcar | jk0: sure | 22:57 |
| jk0 | thanks | 22:58 |
| mustfg | so the nova-manage command is only used by the admin, not remote by a cloud user. i need to get it straight, sorry for the dumb questions :D | 22:58 |
| kpepple | mustfg: you could automate the nova-manage command through a web app or something, but you'd need to write that ... | 22:59 |
| mustfg | ok | 22:59 |
| mustfg | second question: | 22:59 |
| *** adiantum has joined #openstack | 22:59 | |
| mustfg | can i use the ec2 api on the nova api server for euca-... commands and for ec2-.... commands ? | 23:00 |
| mustfg | which one is better for a cloud user ? | 23:01 |
| kpepple | mustfg: i use the euca-* tools ... | 23:01 |
| mustfg | ok | 23:01 |
| mustfg | thx | 23:03 |
| *** bcwaldon has quit IRC | 23:04 | |
| *** RobertLaptop has quit IRC | 23:05 | |
| *** paltman has joined #openstack | 23:05 | |
| *** grapex has joined #openstack | 23:10 | |
| devcamcar | soren, jaypipes: either of you have access to our hudson environments? its currently broken for openstack-dashboard and needs some attention | 23:12 |
| *** adiantum has quit IRC | 23:12 | |
| *** aliguori has quit IRC | 23:14 | |
| *** nelson has quit IRC | 23:15 | |
| *** nelson has joined #openstack | 23:15 | |
| *** markwash has quit IRC | 23:20 | |
| dweimer | In swift is there a way to set a different number of replicas at the account or container level? | 23:24 |
| *** adiantum has joined #openstack | 23:25 | |
| *** reldan has quit IRC | 23:27 | |
| *** westmaas1 has joined #openstack | 23:29 | |
| *** reldan has joined #openstack | 23:30 | |
| *** hggdh has quit IRC | 23:33 | |
| *** hggdh has joined #openstack | 23:33 | |
| *** hggdh has joined #openstack | 23:33 | |
| jk0 | devcamcar: you get a chance to look at that merge? | 23:37 |
| devcamcar | jk0: looking at it now | 23:37 |
| jk0 | cool | 23:37 |
| openstackhudson | Project nova build #591: SUCCESS in 1 min 46 sec: http://hudson.openstack.org/job/nova/591/ | 23:38 |
| openstackhudson | Tarmac: Enable IPv6 injection for XenServer instances. Added addressV6, netmaskV6 and gatewayV6 columns to the fixed_ips table via migration #007 as per NTT FlatManager IPv6 spec. | 23:38 |
| devcamcar | jk0: is this intended? | 23:38 |
| devcamcar | 697+ def create(self, req): | 23:38 |
| devcamcar | 698+ """Create a flavor.""" | 23:38 |
| devcamcar | 699+ #TODO(jk0): Finish this later | 23:38 |
| *** littleidea has joined #openstack | 23:38 | |
| jk0 | devcamcar: yeah, that's another BP entirely | 23:39 |
| *** reldan has quit IRC | 23:39 | |
| jk0 | devcamcar: just didn't have time to finished it | 23:39 |
| devcamcar | what BP is that? | 23:39 |
| jk0 | doesn't exist yet :) | 23:39 |
| kpepple | jk0: i wondered about that ... | 23:39 |
| jk0 | bascially we eventually want to be able to manage the flavors thru the OS API | 23:39 |
| devcamcar | jk0: we should probably omit that then | 23:40 |
| devcamcar | or at very least throw a notimplemeted | 23:40 |
| jk0 | yeah, I'd forgotten I left that in there | 23:40 |
| jk0 | kpepple: would you mind taking care of that real quick? :D | 23:41 |
| *** Ryan_Lane has quit IRC | 23:41 | |
| kpepple | jk0: is that only in the flavors.py file ? or is also in the api.py files ... | 23:41 |
| jk0 | yeah, just in the flavors.py | 23:42 |
| jk0 | in create() and destroy() | 23:42 |
| openstackhudson | Project nova build #592: SUCCESS in 1 min 45 sec: http://hudson.openstack.org/job/nova/592/ | 23:43 |
| openstackhudson | Tarmac: Provide the ability to rescue and unrescue a XenServer instance. | 23:43 |
| kpepple | jk0: running test and then i'll commit | 23:43 |
| jk0 | thanks man | 23:43 |
| jk0 | (and sorry about the extra hassle) | 23:43 |
| kpepple | jk0: not a worry | 23:43 |
| *** Jbain has quit IRC | 23:44 | |
| *** adiantum has quit IRC | 23:44 | |
| *** Jbain has joined #openstack | 23:45 | |
| kpepple | jk0: it looks like you also have some tests for create and delete ... i'm pulling those too | 23:45 |
| jk0 | oh right, good catch | 23:45 |
| jk0 | I think those just watched for a 200 | 23:45 |
| kpepple | yeah ... should be fine to just delete tehm | 23:46 |
| jk0 | maybe I'll have time to swing back to that next week | 23:47 |
| *** RobertLaptop has joined #openstack | 23:47 | |
| *** msassak has quit IRC | 23:48 | |
| *** joearnold has joined #openstack | 23:48 | |
| kpepple | jk0: pushed to revno 685 | 23:50 |
| kpepple | jk0: also added a test for /flavors/1 (specific flavor details) | 23:50 |
| * jk0 waits for diff update | 23:51 | |
| *** gregp76 has quit IRC | 23:51 | |
| *** openfly has quit IRC | 23:56 | |
| *** openfly has joined #openstack | 23:56 | |
| *** mray has quit IRC | 23:56 | |
| *** adiantum has joined #openstack | 23:56 | |
| *** mray has joined #openstack | 23:57 | |
| jk0 | kpepple / devcamcar that looks great | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!