| *** adjohn has quit IRC | 00:02 | |
| *** MarkAtwood has quit IRC | 00:02 | |
| *** v0id has joined #openstack | 00:02 | |
| *** Pr0toc0l has quit IRC | 00:03 | |
| *** neogenix has quit IRC | 00:08 | |
| *** llang629 has joined #openstack | 00:09 | |
| *** llang629 has left #openstack | 00:09 | |
| *** bengrue has quit IRC | 00:10 | |
| *** DavorC has joined #openstack | 00:14 | |
| *** DavorC has quit IRC | 00:14 | |
| *** neogenix has joined #openstack | 00:16 | |
| *** huslage has joined #openstack | 00:20 | |
| *** neogenix has quit IRC | 00:21 | |
| *** hezekiah_ has joined #openstack | 00:23 | |
| *** stevegjacobs has quit IRC | 00:27 | |
| *** cdub has joined #openstack | 00:28 | |
| *** rsampaio has joined #openstack | 00:28 | |
| *** cdub has quit IRC | 00:33 | |
| *** bhall has quit IRC | 00:34 | |
| *** doorlock has quit IRC | 00:41 | |
| *** ahasenack has quit IRC | 00:43 | |
| *** MarkAtwood has joined #openstack | 00:44 | |
| *** takarider has joined #openstack | 00:46 | |
| *** rsampaio has quit IRC | 00:48 | |
| *** hezekiah_ has quit IRC | 00:51 | |
| *** rsampaio has joined #openstack | 00:51 | |
| *** maplebed has quit IRC | 00:53 | |
| *** po has quit IRC | 00:55 | |
| *** krow has quit IRC | 00:55 | |
| *** huslage has quit IRC | 01:04 | |
| *** jog0 has quit IRC | 01:11 | |
| *** charless has joined #openstack | 01:15 | |
| *** rnorwood has joined #openstack | 01:15 | |
| *** ohadlevy has quit IRC | 01:19 | |
| *** obino has quit IRC | 01:20 | |
| *** dragondm has quit IRC | 01:24 | |
| *** rnorwood has quit IRC | 01:24 | |
| *** nerdstein has joined #openstack | 01:27 | |
| *** nerdstein has left #openstack | 01:27 | |
| *** mmetheny_ has quit IRC | 01:28 | |
| *** mmetheny has joined #openstack | 01:28 | |
| *** jdurgin has quit IRC | 01:30 | |
| *** gyee has quit IRC | 01:32 | |
| *** charless has left #openstack | 01:41 | |
| *** ton_katsu has joined #openstack | 01:41 | |
| *** dolphm has joined #openstack | 01:42 | |
| *** llang629 has joined #openstack | 01:43 | |
| *** llang629 has left #openstack | 01:43 | |
| *** pixelbeat has quit IRC | 01:48 | |
| *** dolphm has quit IRC | 01:52 | |
| *** Ryan_Lane has quit IRC | 01:57 | |
| *** cereal_bars has quit IRC | 02:02 | |
| *** lionel has quit IRC | 02:10 | |
| *** lionel has joined #openstack | 02:11 | |
| *** obino has joined #openstack | 02:12 | |
| *** dolphm has joined #openstack | 02:17 | |
| *** rnorwood has joined #openstack | 02:27 | |
| *** egant has joined #openstack | 02:28 | |
| *** neogenix has joined #openstack | 02:35 | |
| *** livemoon has joined #openstack | 02:35 | |
| *** osier has joined #openstack | 02:38 | |
| *** rnorwood has quit IRC | 02:44 | |
| *** troya has joined #openstack | 02:48 | |
| *** reed has quit IRC | 02:50 | |
| *** doorlock has joined #openstack | 02:50 | |
| uvirtbot | New bug: #885958 in swift "missing import for HTTPConflict" [Undecided,New] https://launchpad.net/bugs/885958 | 02:51 |
|---|---|---|
| uvirtbot | New bug: #885955 in nova "Non-admin users can't boot instances with simple scheduler" [High,In progress] https://launchpad.net/bugs/885955 | 02:53 |
| *** dolphm has quit IRC | 03:01 | |
| *** dolphm has joined #openstack | 03:02 | |
| *** dolphm has quit IRC | 03:05 | |
| *** dolphm has joined #openstack | 03:06 | |
| *** ttx has quit IRC | 03:07 | |
| *** clauden_ has quit IRC | 03:08 | |
| *** Ryan_Lane has joined #openstack | 03:16 | |
| *** egant has quit IRC | 03:16 | |
| *** egant has joined #openstack | 03:16 | |
| *** rods has quit IRC | 03:17 | |
| *** rsampaio has quit IRC | 03:17 | |
| *** dolphm has quit IRC | 03:18 | |
| *** dolphm has joined #openstack | 03:18 | |
| *** wariola has quit IRC | 03:21 | |
| *** rsampaio has joined #openstack | 03:21 | |
| *** dolphm has quit IRC | 03:23 | |
| *** tux2011 has joined #openstack | 03:33 | |
| *** dgags has quit IRC | 03:33 | |
| tux2011 | hi all | 03:34 |
| tux2011 | i am very close to swift install. | 03:34 |
| tux2011 | proxy-server ERROR with Account server 172.29.200.44:6002/sda3 re: Trying to PUT /AUTH_system: Host unreachable | 03:34 |
| *** misheska has quit IRC | 03:34 | |
| *** wariola has joined #openstack | 03:34 | |
| tux2011 | seeing this error while creating the root account. | 03:34 |
| *** egant has quit IRC | 03:38 | |
| *** dolphm has joined #openstack | 03:46 | |
| *** Lumiere has joined #openstack | 03:48 | |
| *** ttx has joined #openstack | 03:49 | |
| *** ttx has quit IRC | 03:49 | |
| *** ttx has joined #openstack | 03:49 | |
| *** jog0_ has joined #openstack | 03:51 | |
| *** egant has joined #openstack | 03:51 | |
| *** clopez has quit IRC | 03:53 | |
| Lumiere | sandywalsh_: are you here? | 03:54 |
| *** dcthang has joined #openstack | 03:55 | |
| Lumiere | I'm working on the tests for python-novaclient, and I am trying to track down the right way to fix the tests | 03:56 |
| Lumiere | (related to github pull request 144 on python-novaclient) | 03:57 |
| *** syah has joined #openstack | 03:58 | |
| *** jog0_ has quit IRC | 04:00 | |
| *** jog0 has joined #openstack | 04:01 | |
| *** khurram has joined #openstack | 04:02 | |
| khurram | hi every one | 04:02 |
| Lumiere | hi | 04:02 |
| khurram | i am using PHP to query the openstack swift to know the number of GET requests and , Inbounds, Outbounds a particular account has incurred and need help | 04:03 |
| khurram | all i can get from the api is number of storrange bytes | 04:04 |
| khurram | does any one know who can i get more details of activities so that i can bill him ? | 04:04 |
| *** syah has quit IRC | 04:05 | |
| *** oubiwann has joined #openstack | 04:06 | |
| khurram | no response...:( | 04:07 |
| *** rnorwood has joined #openstack | 04:07 | |
| Lumiere | I'm not a swift person | 04:08 |
| khurram | where can i find a swift person ? | 04:10 |
| vidd | khurram, try looking here: http://docs.openstack.org/diablo/openstack-object-storage/admin/content/analyzing-log-files-with-swift-cli.html | 04:11 |
| *** tsuzuki_ has joined #openstack | 04:12 | |
| *** oubiwann has quit IRC | 04:14 | |
| tux2011 | anyone know what this means; proxy-server ERROR with Object server 172.29.200.47:6000/sda3 re: Expect: 100-continue on /AUTH_system/myfiles/file.tmp: Host unreachable | 04:15 |
| *** khurram has quit IRC | 04:15 | |
| tux2011 | i checked permission for /srv/node swift owns the directory. | 04:16 |
| Lumiere | tux2011: the error sounds more like a network issue | 04:16 |
| Lumiere | 'host unreachable' | 04:16 |
| tux2011 | Lumiere: I can ping the storage nodes from the proxy node. | 04:17 |
| Lumiere | from a purely sysadmin point of view, review firewalls and running services | 04:17 |
| *** syah has joined #openstack | 04:17 | |
| tux2011 | Lumiere: swift client is saying Object PUT failed. so that seems that it cannot write. | 04:18 |
| *** ldlework has quit IRC | 04:20 | |
| *** doorlock has quit IRC | 04:22 | |
| Lumiere | not sure what to say, I am not sure what the PUT failure is going to mean | 04:23 |
| *** troya has quit IRC | 04:23 | |
| *** MarkAtwood has quit IRC | 04:27 | |
| *** dolphm has quit IRC | 04:29 | |
| *** dolphm has joined #openstack | 04:30 | |
| *** rnorwood has quit IRC | 04:31 | |
| *** katkee has joined #openstack | 04:33 | |
| *** dolphm has quit IRC | 04:34 | |
| *** hadrian has quit IRC | 04:36 | |
| *** jog0 has quit IRC | 04:40 | |
| *** rnorwood has joined #openstack | 04:41 | |
| *** neogenix has quit IRC | 04:44 | |
| *** jakedahn has quit IRC | 04:45 | |
| *** jakedahn has joined #openstack | 04:46 | |
| *** tux2011 has quit IRC | 04:47 | |
| *** jakedahn has quit IRC | 04:48 | |
| *** MarkAtwood has joined #openstack | 04:48 | |
| *** jakedahn has joined #openstack | 04:49 | |
| *** MarkAtwood has left #openstack | 04:52 | |
| *** v0id has quit IRC | 05:04 | |
| *** egant has quit IRC | 05:07 | |
| *** tsuzuki_ has quit IRC | 05:10 | |
| uvirtbot | New bug: #885986 in nova "Exceeding compute provider metadata limit on create causes a 500 instead of a 413" [Undecided,New] https://launchpad.net/bugs/885986 | 05:11 |
| *** hezekiah_ has joined #openstack | 05:12 | |
| *** hezekiah_ has quit IRC | 05:17 | |
| *** tsuzuki_ has joined #openstack | 05:20 | |
| *** bhall has joined #openstack | 05:23 | |
| *** bhall has quit IRC | 05:23 | |
| *** bhall has joined #openstack | 05:23 | |
| *** sugano has joined #openstack | 05:33 | |
| *** takarider has quit IRC | 05:37 | |
| *** MarkAtwood has joined #openstack | 05:44 | |
| *** MarkAtwood has left #openstack | 05:45 | |
| *** marrusl has joined #openstack | 05:54 | |
| *** jakedahn has quit IRC | 05:55 | |
| *** troya has joined #openstack | 05:59 | |
| *** npmapn has joined #openstack | 06:02 | |
| *** supriya has joined #openstack | 06:05 | |
| troya | hi all | 06:05 |
| *** jedi4ever has joined #openstack | 06:14 | |
| troya | anyone can help me, i have problem on swift | 06:16 |
| *** troya has quit IRC | 06:18 | |
| *** wariola has quit IRC | 06:22 | |
| *** bhall has quit IRC | 06:25 | |
| *** troya has joined #openstack | 06:25 | |
| *** superjudge has joined #openstack | 06:26 | |
| troya | hello | 06:30 |
| *** tsuzuki_ has quit IRC | 06:34 | |
| *** miclorb_ has quit IRC | 06:37 | |
| *** rich_ has joined #openstack | 06:38 | |
| vidd | troya, what is your issue? | 06:39 |
| *** jedi4ever has quit IRC | 06:52 | |
| *** hezekiah_ has joined #openstack | 06:53 | |
| *** Telamon has joined #openstack | 06:59 | |
| *** Ryan_Lane has quit IRC | 07:02 | |
| troya | vidd:i have post issue to forum, check it out, please | 07:05 |
| troya | http://forums.openstack.org/viewtopic.php?f=11&t=414&p=1356#p1356 | 07:05 |
| vidd | root@maple:~# swift -A http://127.0.0.1:8080/auth/v1.0 -U test:tester -K testing stat is WRONG | 07:07 |
| *** datajerk has quit IRC | 07:07 | |
| troya | vidd, what's should ? | 07:08 |
| vidd | your doing curl commands for keystone right? | 07:08 |
| troya | yes | 07:09 |
| vidd | -A is for your keystone auth key | 07:09 |
| vidd | your server is called with the -h flag | 07:09 |
| vidd | * -H | 07:09 |
| vidd | [ the H is capital ] | 07:10 |
| troya | yes,right. then ? | 07:10 |
| vidd | the rest of what you are trying to do | 07:11 |
| vidd | did you set up keystone on your server yet? | 07:12 |
| troya | vidd, may be it'll be clearly if you would like to reply on forum | 07:13 |
| vidd | did you set up keystone yet? | 07:13 |
| troya | hm..how to set up it? | 07:13 |
| vidd | i dont have an account for the forum | 07:13 |
| vidd | so...that is a "no" | 07:14 |
| vidd | then you cannot set up swift with keystone and expect it to work | 07:14 |
| troya | yesa.. i still confuse about keystone vid, would you like to explains about it to me? | 07:14 |
| vidd | set up your swift without keystone...test it...then try to get keystone set up | 07:15 |
| vidd | keystone takes perfectly function stacks....puts all the credentials in one place ... and....if done properly....lets you use dashboard to control it all | 07:16 |
| vidd | does that clear it up for you? | 07:19 |
| *** thickskin has joined #openstack | 07:20 | |
| troya | hm..little bit. i'll learn again from google | 07:20 |
| vidd | i recommend you get your swift to work without keystone first | 07:21 |
| troya | ok vidd, i'll try it | 07:21 |
| vidd | then...when all your parts are working...if you want to...set up keystone....then try to intigrate your apps into it | 07:22 |
| Telamon | Just a tip from another newbie, the default keystone packages (at least the Ubuntu ones) don't seem to work. So you'll have to use the git repo (or devstack) to get keystone + dashboard going. | 07:22 |
| *** thickskin has left #openstack | 07:22 | |
| vidd | Telamon, it is also worth mentioning that when you start to integrate keystone, your apps wont work until they work with keystone =\ | 07:23 |
| vidd | thats why you want a unctional stack before your even think about touching keystone | 07:23 |
| Telamon | Heh, I've been trying to install with the Ubuntu packages. *NOTHING* works for me. ;-) I'm trying to hack together something with devstack at the moment to at least get a basic system running, but I'm not having much more luck. | 07:24 |
| vidd | Telamon, all my stuff worked fine until i typed that fateful command..."apt-get install keystone" | 07:26 |
| *** obuisson has quit IRC | 07:26 | |
| *** yeming has joined #openstack | 07:27 | |
| Telamon | You're lucky. I got tonnes of database errors using the 2011.3 PPA on 11.10 I'm thinking about reinstalling the server and starting from scratch, or starting up some VMWare instances and trying Cloudspace. | 07:27 |
| vidd | here is what i did..... | 07:28 |
| vidd | http://cssoss.files.wordpress.com/2011/10/openstackbookv2-0_csscorp.pdf | 07:28 |
| vidd | follow that guide | 07:28 |
| vidd | if you have 11.10, there is no need to add PPA's | 07:28 |
| Telamon | Yeah, I used the same guide. Lots of problems... Hmm, I wonder if I remove the PPAs and start over if it will work better. | 07:29 |
| vidd | if you did add PPA's you most likely have the devel stuff [the stuff not quite ready for primetime] | 07:29 |
| vidd | id you remove the PPA's properly, then you will NOT need to start from scratch.... | 07:30 |
| vidd | removing the PPA's will get you back to where you were before you added the PPA's | 07:30 |
| vidd | Telamon, http://www.howtogeek.com/howto/29584/safely-remove-ppas-and-roll-back-to-stable-versions-in-ubuntu/ | 07:32 |
| *** katkee has quit IRC | 07:32 | |
| vidd | wait...that is not the right one.... | 07:32 |
| *** beekhof_muc has joined #openstack | 07:33 | |
| *** hezekiah_ has quit IRC | 07:34 | |
| vidd | Telamon, this is the one: http://bigbrovar.aoizora.org/index.php/2010/01/10/how-to-safely-remove-ppa-repository-from-ubuntu/ | 07:37 |
| *** nacx has joined #openstack | 07:39 | |
| Telamon | Thanks, I'll give it a try now. | 07:40 |
| *** datajerk has joined #openstack | 07:41 | |
| Telamon | Hmm, devstack is actually working at the moment. That's interesting... | 07:41 |
| yeming | So far I have my control node setup. I decide to have all-in-one nova so I have no other compute node. I think I'm configuring network now. Is FlatNetworking the most easy one to setup? | 07:41 |
| vidd | yeming, "easy" is a relative term | 07:42 |
| vidd | how many eth cards ya got? | 07:42 |
| yeming | Two | 07:43 |
| vidd | with that, i'd go with flatdhcp | 07:43 |
| yeming | One of the eth cards is currently connected to my office network, where there's already a DHCP server. Does that matter? | 07:45 |
| vidd | no...the "dhcp" is for your NM network | 07:46 |
| vidd | *VM | 07:46 |
| *** jedi4ever has joined #openstack | 07:46 | |
| *** takarider has joined #openstack | 07:48 | |
| yeming | Thanks vidd, I read the manual and decided to give FlatDHCP a shot. | 07:50 |
| *** sugano has quit IRC | 07:50 | |
| vidd | its what i use...and i only have one eth card | 07:51 |
| vidd | [single eth makes networking difficult] | 07:51 |
| *** superjudge has quit IRC | 07:51 | |
| *** guigui1 has joined #openstack | 07:54 | |
| yeming | Yeah the manual said "if you specify an interface that already has an IP it will break" | 07:55 |
| *** takarider has quit IRC | 07:55 | |
| vidd | yeming, what they mean is...."if you are connected to the machine via ssh on eth0 and you set the bridge to eth0" | 07:55 |
| vidd | you just lose your connection | 07:56 |
| vidd | but you do not need to set it up on the controller...only the compute nodes | 07:56 |
| *** beekhof_muc has left #openstack | 07:57 | |
| yeming | In my case, controller and compute are the same node. | 07:58 |
| vidd | yeming, then i would configure the br100 on the eth1 | 07:59 |
| yeming | yes, that's what i'm going to do. | 08:00 |
| yeming | should i create br100 myself? | 08:00 |
| vidd | yes...so it gets tied to the eth1 | 08:01 |
| Telamon | vidd do you know what the benefit is of running on two different interfaces is? Other than less traffic on your main network? | 08:02 |
| *** supriya has quit IRC | 08:02 | |
| vidd | Telamon, less network troubleshooting...noweird glitches causing you to lose connectivity etc | 08:03 |
| vidd | plus...its a must on secondary nodes so outbound traffic goes to the internet and maintainace traffic goes on the private netmask | 08:04 |
| Telamon | Ah, I'm glad I can run it on eth1 then. Our plan is for a two server setup and I didn't want to run DHCP on eth0. | 08:05 |
| vidd | unless your compute node is located in a different geographic location, its best to have the privite network not connected to your "main" router | 08:06 |
| vidd | when i get my new server to do our deployment, i will be migrating stuff from existing servers and then converting those "freed up" servers to new compute nodes =] | 08:07 |
| Telamon | Okay, so private is eth1 and public, with the floating IPs, is eth0. | 08:08 |
| vidd | i might even convert the MS boxes =] | 08:08 |
| vidd | Telamon, sorta...yeah | 08:08 |
| *** livemoon has quit IRC | 08:08 | |
| Telamon | Well you should do that on general principle anyway. Windows is only good for running StarCraft | 08:08 |
| *** naehring has joined #openstack | 08:08 | |
| *** livemoon has joined #openstack | 08:08 | |
| vidd | Telamon, that is so not true....it runs RedAlert3! | 08:09 |
| vidd | as well | 08:09 |
| vidd | *sigh* i need to put an ubur-light X manager on this server without bringing in half the ubuntu repo's | 08:10 |
| Telamon | Heh, haven't played RedAlert3 but I'll take your word for it. | 08:11 |
| Telamon | Hmm, maybe Enlightenment or Openbox. You'll get the hundreds of X server packs, but not all the gnome/kde cruft. | 08:12 |
| *** dobber has joined #openstack | 08:13 | |
| Telamon | vidd: Can you check, what version of nova do you have installed? dpkg -l 'nova*' I'm pulling in 6.2 and I'm worried that's cached from the PPA | 08:14 |
| Telamon | Oops, nevermind. Checked on my workstation and it says 6.2 is regular version | 08:15 |
| *** nerens has joined #openstack | 08:18 | |
| *** superjudge has joined #openstack | 08:18 | |
| vidd | Telamon, Kiall has a PPA out that he claims will apt-get the entire stack [except maybe swift] | 08:18 |
| Telamon | I'm going to have to get OpenStack working so I can test out all these different OpenStack repos/installers to find which one works! That seems very meta... | 08:20 |
| *** reidrac has joined #openstack | 08:22 | |
| vidd | Telamon, it has been recommended to me: glance/swift/nova from base repo's, keystone/dashboard[horizon] from git | 08:22 |
| vidd | i think they call dashboard "horizon" because it full funtionality date is "sometime over the horizon" =] | 08:23 |
| Telamon | vidd: Yeah, that's my current plan. I'm going to get the base system up and then use the devstack.org script to get them going. | 08:23 |
| Telamon | Hah! You may be on to something there... | 08:24 |
| zykes- | horizon + keystone from git | 08:24 |
| zykes- | rest from pkgs | 08:24 |
| zykes- | what worked for me ;p | 08:24 |
| vidd | Telamon, the devstack script isnt permanant | 08:24 |
| vidd | zykes-, your back =] | 08:25 |
| vidd | what time zone are you in? | 08:25 |
| zykes- | gmt+1 | 08:25 |
| Telamon | vidd: As long as you use a custom startup script you are okay. Otherwise, yeah, it blows away your config each time you run it. | 08:25 |
| vidd | gtm-5 here | 08:25 |
| zykes- | working on websecurity and firewalling atm ;) | 08:25 |
| ton_katsu | pypi.python.org is down... | 08:26 |
| vidd | jeez! keystone was update 5 hours ago...this is why i so aggitated about using git | 08:29 |
| *** katkee has joined #openstack | 08:29 | |
| *** livemoon has quit IRC | 08:29 | |
| *** livemoon has joined #openstack | 08:29 | |
| *** jedi4ever has quit IRC | 08:30 | |
| *** mgoldmann has joined #openstack | 08:30 | |
| *** mgoldmann has joined #openstack | 08:30 | |
| vidd | zykes-, you told me b4 how to specify what git brach to git from | 08:30 |
| vidd | can you re-enlighten me? | 08:30 |
| zykes- | vidd: i run latest trunk of horizon, works dandy with diablo | 08:31 |
| Telamon | "git branch" to list them all "git branch <name>" to switch. You might need to use "git branch -a" to list remote branches. | 08:31 |
| vidd | zykes-, does it run in a venv? | 08:31 |
| zykes- | yeah vidd | 08:33 |
| vidd | zykes-, see....thats an issue | 08:33 |
| zykes- | vidd: why ? | 08:33 |
| vidd | because it will be insulated from security updates | 08:34 |
| vidd | and on a production machine....thats a bad thing | 08:34 |
| zykes- | wait for essex then ;) | 08:34 |
| zykes- | or install it system wide | 08:34 |
| vidd | i dont have the time to wait | 08:34 |
| *** Razique has joined #openstack | 08:35 | |
| vidd | zykes-, i dont know how to install it system wide with python scripta | 08:35 |
| zykes- | vidd: make sure you have all deps | 08:36 |
| vidd | Razique, good morning! [or whatever it is where you are] | 08:36 |
| Razique | hi vidd ! | 08:36 |
| Razique | 'sup ? | 08:36 |
| zykes- | use pip to install packages that are in apt or yum | 08:36 |
| Razique | hey zykes- :) | 08:36 |
| vidd | Razique, im ready to lug my system to a 10th floor balcony | 08:36 |
| yeming | looks like i created an instance with euca-run-instances. Where can i find the ip of the instance? | 08:36 |
| vidd | see how far a can "push" it =] | 08:37 |
| Razique | yeming: euca-describe-instances | 08:37 |
| Razique | vidd: hehe, so what's the status ? | 08:37 |
| vidd | Razique, i purged the database and apt-get remove --purged it | 08:37 |
| vidd | like 3 minutes ago | 08:37 |
| Razique | Yesterday I hadn't the time to test the Keystone integration :( I'm openstacking the whole morning | 08:37 |
| zykes- | vidd: you running keystone trunk or ? | 08:38 |
| vidd | Razique, do you hear im getting funding for a single machine? | 08:38 |
| vidd | zykes-, no...stable | 08:38 |
| zykes- | vidd: and it works with nova and swift? | 08:38 |
| zykes- | and glance. | 08:38 |
| Razique | vidd: what getting funding means ? :D | 08:38 |
| *** jwomack has joined #openstack | 08:38 | |
| yeming | Razique: there's no ip address info in `euca-describe-instances` | 08:38 |
| vidd | Razique, i will move as much as I can to the one new machine and convert "freed" machines into additional compute nodes =] | 08:39 |
| Razique | yeming: nova-manage network list | 08:39 |
| vidd | Razique, the bosses are buying me a machine...ink hits checkbook | 08:39 |
| Razique | vidd: oh cool; how many nodes will have u ? | 08:39 |
| Razique | nice | 08:39 |
| vidd | dunno | 08:39 |
| *** supriya has joined #openstack | 08:39 | |
| Razique | here, on production I've 4 compute nodes, almost 6 | 08:40 |
| Razique | and my lab has two small towers | 08:40 |
| Razique | one controller and one node | 08:40 |
| vidd | it depends on how many of the machines are actually ours (we have co-location customers taking up rackspace) | 08:40 |
| *** livemoon has quit IRC | 08:40 | |
| *** livemoon has joined #openstack | 08:41 | |
| vidd | it also depends on how many i need to completely convert the datacenter | 08:41 |
| vidd | i want to use some of the boxes for a swift array | 08:41 |
| Razique | vidd: u work at rackspace ? | 08:42 |
| vidd | no...rack space....as in space on the racks | 08:42 |
| vidd | heh...right now i dont work at all | 08:44 |
| yeming | Razique: `nova-manage network list` says something like IPv4 as 192.168.0.0/24, starting address as 192.168.0.3, VlanID as 100. I remember now that I actually added this network as per the manual during setup. But now I'm using FlatDHCP. | 08:44 |
| Razique | vidd: ok ok | 08:45 |
| Razique | vidd: :D | 08:45 |
| Razique | yeming: ok, show me the output of euca-describe-instances | 08:45 |
| vidd | what is the command to git the stable/diablo version of keystone? | 08:45 |
| *** jwomack has left #openstack | 08:46 | |
| *** livemoon has quit IRC | 08:46 | |
| yeming | Sorry if it looks ugly here. I'm going to past... | 08:46 |
| *** livemoon has joined #openstack | 08:46 | |
| yeming | RESERVATION r-ba4ldwgs dubproject default | 08:46 |
| yeming | INSTANCE i-00000001 ami-00000002 running test (dubproject, server1) 0 m1.tiny 2011-11-04T08:34:15Z nova aki-00000001 ami-00000000 | 08:46 |
| yeming | 08:46 | |
| vidd | yeming, use pastebin | 08:46 |
| Razique | yeming: yup, please :) http://paste.openstack.org/ | 08:46 |
| yeming | ahh, ok. will do that. | 08:47 |
| Razique | vidd: dunno sorry :( | 08:47 |
| vidd | our chat windows are not nesisarrily as wide as yours =] | 08:47 |
| yeming | Razique: euca-describe-instances here - http://paste.openstack.org/show/3070/ | 08:48 |
| vidd | *sigh* the i suppose ill have to download the zip and unpack it | 08:48 |
| Telamon | vidd: One sec, I have it here... | 08:48 |
| foexle | hey guys, i'm using nova+keystone+.... and i have a question about user management. If i create a user in keystone will this user create in nova too ? ... because i need the access and sec-key to the api | 08:48 |
| vidd | foexle, the idea is to add the user in nova and then put it into keystone | 08:50 |
| *** livemoon has quit IRC | 08:50 | |
| Razique | yeming: can you create a new instance and in // tailing nova-network.log ? | 08:50 |
| Razique | foexle: if you create a user via keystone-manage it won't be added into nova database | 08:50 |
| Telamon | vidd: Actually, I guess I don't. Sorry. | 08:51 |
| foexle | vidd: that is the idea ? .... i think keystone are the central auth service .... so you have an inconsistency in your databases | 08:51 |
| Telamon | thought I had it bookmarked. | 08:51 |
| *** livemoon has joined #openstack | 08:51 | |
| foexle | Razique: so i must written a script that pushs a new user in keystone AND nova ? | 08:51 |
| Razique | foexle: nope, since nova checks the user with Keystone DB | 08:52 |
| vidd | foexle, the central auth service for dashboard...and future tools | 08:52 |
| Razique | not it's anymore | 08:52 |
| Telamon | Razique: Hmm, so if I have keystone configured and run "nova-manage user create blah..." will the user get added to keystone? | 08:52 |
| vidd | Razique, is this new? | 08:53 |
| yeming | Razique: I created a new instance and nova-network.log here http://paste.openstack.org/show/3071/ | 08:53 |
| foexle | Razique: yes but you dont have credentials to use all functions in nova ? | 08:53 |
| Razique | yeming: ok, add --debug and --verbose flags into nova.conf ; restart nova-api and nova-network, and recreate an instance please | 08:54 |
| Razique | vidd: I can be wrong on that though | 08:55 |
| Razique | vidd: when u create a user via keystone, you see it into nova db ? | 08:55 |
| foexle | Razique: so i need only to add a user in keystone ok, but then i dont have the keypair to manage the cloud with the api tools | 08:55 |
| vidd | Razique, no | 08:55 |
| *** catbot has quit IRC | 08:56 | |
| Razique | foexle: normally u should, the only thing that require extra steps would be having euca2ools working with keystone | 08:56 |
| Razique | but novaclient works great with keystone | 08:56 |
| foexle | ah ok | 08:56 |
| vidd | foexle, are you using dashboard? | 08:56 |
| foexle | ok i need etra steps :) | 08:56 |
| foexle | yes and euca tools | 08:56 |
| foexle | too | 08:56 |
| foexle | dasboard runs | 08:56 |
| Razique | foexle: for euca2ools u'll need to import the EC2_* stuff into credentials db, but for novaclient, no need | 08:56 |
| Razique | since you have the users into keystone DB only | 08:57 |
| *** pixelbeat has joined #openstack | 08:57 | |
| Razique | same for a Keystone'd glance | 08:57 |
| foexle | ah ok | 08:57 |
| foexle | good i'll see thank you ! | 08:57 |
| vidd | have your script do this: create user in nova ->take eu2 keys as strings -> use eu2 keys in keystone credentials | 08:58 |
| *** jwomack1 has joined #openstack | 08:59 | |
| yeming | Razique: I added --debug (--verbose was there) and here's the tail of nova-network.log - http://paste.openstack.org/show/3072/ | 09:00 |
| yeming | looks like no more information | 09:00 |
| Razique | yeming: have u restarted nova-network ? | 09:01 |
| Razique | yeming: you should have tons of lines here :D | 09:01 |
| *** jwomack1 has left #openstack | 09:04 | |
| *** livemoon has quit IRC | 09:04 | |
| yeming | yeah i did a 'sudo restart nova-network', after a 'sudo restart nova-api'. My nova.conf - http://paste.openstack.org/show/3073/ , in case it helps | 09:04 |
| *** javiF has joined #openstack | 09:04 | |
| *** livemoon has joined #openstack | 09:04 | |
| foexle | vidd: what do you mean with, use this keys as keystone credentials? keystone hase name:pw auth and nova acccess_key udn secret_key .... ah you mean this two keys ? hmmm it's not the cleanest solution or ? :) | 09:06 |
| vidd | there is like 10 steps to adding a single user in keystone....the last step is creating the credentials | 09:08 |
| vidd | the credential require a passkey and a secret key | 09:08 |
| vidd | when you add a user to nova, you get a passkey and a secret key | 09:09 |
| *** jwomack1 has joined #openstack | 09:09 | |
| *** nikolaij has joined #openstack | 09:09 | |
| vidd | so somewhere in your script, throw a bone to nova and get those keys...and use those keys when making the creddentials in keystone | 09:09 |
| *** jwomack1 has left #openstack | 09:10 | |
| foexle | vidd: many thanks perfekt ! | 09:10 |
| *** darraghb has joined #openstack | 09:15 | |
| yeming | vidd: for FlatDHCP, what command should i run to create a network? | 09:16 |
| vidd | yeming, mhat do you mean? | 09:16 |
| vidd | what guide are you following? | 09:16 |
| Razique | yeming: ok, obviously there is a issues regarding the dhcp/ dns server | 09:17 |
| Razique | vidd: +1, yeming which guide ? | 09:17 |
| *** NadeemNayeck has joined #openstack | 09:17 | |
| yeming | vidd, Razique: something like 'nova-manage network create public 192.168.3.0/12 1 256' | 09:18 |
| Razique | won't work since the mask is a /12 imho | 09:18 |
| vidd | well the last number cannot be more than 255 | 09:18 |
| Razique | yeming: try /25 | 09:18 |
| Razique | $ | 09:19 |
| *** livemoon has quit IRC | 09:19 | |
| yeming | I didn't create any network after I changed to FlatDHCP. maybe that's the reason? | 09:19 |
| Razique | try / 24 sorry | 09:19 |
| *** livemoon has joined #openstack | 09:19 | |
| Telamon | yeming: The last number (256) is only for /24 netmasks. If you want to use a different one it's 2^(netmask) | 09:19 |
| vidd | yeming, if you already have private IP's why are you creating more? | 09:20 |
| Telamon | Sorry, 2^(32-netmask) | 09:20 |
| vidd | Telamon, 256 is an invalid number regardless | 09:20 |
| vidd | ip's are 0-255 | 09:20 |
| yeming | I deleted the network. Now i have nothing in 'nova-manage network list' | 09:20 |
| Telamon | So /24 is 2 ^8 = 256, /28 is 2^4 = 16 | 09:20 |
| vidd | yeming, how many private ip's do you expect to use? | 09:21 |
| Telamon | vidd; Ah, but it's zero based. 0-1 is 2 values, not 1. C math. Which is why people started programming in Pascal, the poor bastards... | 09:21 |
| yeming | vidd: 8, 16, 256... any of these works for me | 09:22 |
| vidd | Telamon, the first number is the gateway ip the last number is the broadcast ip | 09:22 |
| Telamon | nova-manage network create public 192.168.3.0/24 1 256 | 09:23 |
| vidd | an ip address CANNOT be a .256 | 09:23 |
| yeming | Telamon: what does 'public' here mean? | 09:23 |
| vidd | yeming, it is the name of the network for this group of IPs | 09:23 |
| Telamon | yeming: I don't think it's actually used. I believe it's just a label. | 09:23 |
| yeming | cool | 09:24 |
| vidd | ther is a concept of "this network is for these people and that network is for those people" | 09:24 |
| vidd | i dont think it is used anymore | 09:25 |
| Razique | vidd: diablo has make it mandatory | 09:25 |
| Telamon | vidd: Sort of. There are only 255 usable IPs, but the first IP (.0) is not usable. It is called the network number, and used for routing only. Then you have the broadcast IP and the gateway IP. So you always lose 3 IPs for every subnet. That's why you don't have /31 subnets. It's a stupid network math thing | 09:26 |
| Razique | without that label, the network cannot be created | 09:26 |
| vidd | s/w anymore/yet | 09:26 |
| yeming | and it works | 09:26 |
| yeming | now i have IP and I have tons of lines in nova-network.log | 09:26 |
| vidd | 256 to be exact =] | 09:27 |
| Razique | hehe | 09:29 |
| Razique | ok great yeming, show us the log now :p | 09:29 |
| *** uncleofthestick has joined #openstack | 09:31 | |
| yeming | Razique: logs here - http://paste.openstack.org/show/3074/ . But seems i cannot ssh into the instance | 09:35 |
| Razique | ok now does euca-describe-instances reports ip ? | 09:36 |
| yeming | Razique: yes | 09:37 |
| *** po has joined #openstack | 09:38 | |
| Razique | yeming: ok, now two things : 1 euca-get-console-output $instance_id | 09:38 |
| Razique | then ssh -vv user@instance-ip | 09:38 |
| vidd | -vv? | 09:39 |
| Telamon | "very verbose" | 09:40 |
| foexle | vidd: question again :) .... tenants in keystone = projects in nova right ? i must create a project in nova with the same name as the tenant in keystone right ? | 09:40 |
| Telamon | Hmm, if I switch my server's IP in /etc/nova/nova.conf (had a type when doing the initial install) do I have to do anything to update the database or the like? | 09:41 |
| vidd | foexle, i have no answer for that....i have yet to unbreak my nova from the clutches of keystone | 09:41 |
| vidd | zykes-, can you answer foexle 's question? | 09:41 |
| vidd | Telamon, if the rest of the stuff is correct, then no...just restart nova-* | 09:42 |
| foexle | vidd: ah k | 09:42 |
| Razique | Telamon: no need, but services need to be restarted | 09:43 |
| Razique | and myabe u'll loose network for running instances | 09:43 |
| Razique | except if you play with iptables i guess | 09:43 |
| Telamon | Yeah, no instances yet. I uploaded an image and now I'm trying to get a list of images and it's hanging. | 09:44 |
| Razique | foexle: no need | 09:44 |
| yeming | Razique: connect to host 192.168.0.5 port 22: No route to host | 09:44 |
| Telamon | Restart fixed it. :) | 09:44 |
| Razique | but yah tenants and projects are pretty the same | 09:44 |
| Razique | cool | 09:44 |
| yeming | and euca-get-console-output prints a lot of url error [[Errno 101] Network is unreachable] | 09:44 |
| Razique | ok, yeming what the get-console gave ? | 09:44 |
| Razique | yeming: ok, could you paste the whole log ? | 09:44 |
| Razique | btw what image are u using ? | 09:45 |
| yeming | ubuntu1010-UEC-localuser-image.tar.gz | 09:45 |
| Razique | ok, so the errors is linked the the metadats fetching | 09:45 |
| vidd | yeming, did you allocate a public(floating) ip and asign it to the instane? | 09:45 |
| Razique | SSH keys cannot be injected | 09:45 |
| Razique | vidd: no need at that stage | 09:45 |
| yeming | vidd: not yet. | 09:45 |
| yeming | Razique: which image should i use then? | 09:46 |
| Razique | yeming: that one is ok | 09:47 |
| *** leitz has joined #openstack | 09:47 | |
| vidd | Razique, i was not aware you could ping a private ip...even from the controller | 09:47 |
| Razique | that trace simply means the instance has not been able to retrieve the meta-datas | 09:47 |
| vidd | i thought you had to assign a public IP first | 09:47 |
| Razique | vidd: sure :) in some configs, instances can reach outside with a private ip | 09:47 |
| Razique | vidd: the private IP belongs to the controller (run ip addr sh u'll see) | 09:48 |
| vidd | this readme says keystone has to be started?!? | 09:48 |
| Razique | vidd: i was talking about floating sorry :p | 09:48 |
| vidd | wtf | 09:48 |
| Razique | vidd: for private the controller acts as a gateway | 09:49 |
| Razique | and it's a route-able range from it | 09:49 |
| yeming | I have to grab some food now and will work on that when i'm back. Thanks guys. | 09:49 |
| Razique | (check ip route sh :) ) | 09:49 |
| Razique | yeming: sure we are here! | 09:49 |
| vidd | Razique, the README for keystone says "Starting both Admin and Service API endpoints: $ ./bin/keystone | 09:50 |
| vidd | " | 09:50 |
| Razique | vidd: yah, you can add & | 09:51 |
| vidd | how do i get keystone to be "always on" | 09:51 |
| Razique | add & | 09:51 |
| Razique | :) | 09:51 |
| Razique | or create a simple script into i'd say rc.local | 09:51 |
| Razique | or the runlevel you wan | 09:51 |
| vidd | let me phase it diferently...how do i get it added to upstart so if the server has a power hiccup i dont have to get woke up at 3 in the afternoon to fix it | 09:52 |
| *** yeming has quit IRC | 09:53 | |
| *** leitz has quit IRC | 09:54 | |
| vidd | "not auto-added to start-up" definantly qualifies as a bug:Critical | 09:54 |
| zykes- | foexle: still wondering or ? | 09:54 |
| Razique | vidd: something like this http://paste.openstack.org/show/3075/ | 09:54 |
| Razique | or dunno | 09:54 |
| zykes- | not autoadded what vidd ? | 09:54 |
| vidd | keystone | 09:54 |
| Razique | then u add the symlink to the run-levels | 09:54 |
| Razique | vidd: you took it from github right ? | 09:55 |
| uncleofthestick | Does anybody knows if the nova stable/diablo branch works with the trunk version of keystone? | 09:55 |
| vidd | yes | 09:55 |
| Razique | vidd: startup script are not linked to keystone | 09:55 |
| Razique | there are part of distros packaging | 09:55 |
| Razique | if you would have taken it from repos | 09:55 |
| Razique | a startup script has been implemented | 09:56 |
| Razique | I think it's soren who wrote it | 09:56 |
| vidd | i still dont understand how you can have any installer script for a critical piece of your product NOT auto-start on boot | 09:57 |
| Razique | vidd: Keystone does start on boot if you use distro' repos | 09:57 |
| vidd | it should not matter where you get it from it should auto boot regardless | 09:58 |
| Razique | vidd: can't agree on that even if I totally get ur point | 09:59 |
| Razique | :) | 09:59 |
| Razique | same goes for all openstack projects | 09:59 |
| *** tryggvil_ has joined #openstack | 09:59 | |
| Razique | if you clone em from github they won't boot at start | 09:59 |
| vidd | yet another reason the ubuntu people need to get on the ball and fix the repo's | 10:00 |
| vidd | Razique, i tried to install keystone from repo's...and we couldnt get it to work | 10:00 |
| Razique | vidd: yah, the thing is keystone from repo is outdated now | 10:01 |
| vidd | now im wondering if half of it was not autostarting =\ | 10:01 |
| Razique | Keystone's moving really fast | 10:01 |
| vidd | it should still work with the nova etc from the same repo's | 10:02 |
| Razique | vidd: I agree with you regarding keystone' version into the repos | 10:02 |
| Razique | vidd: it does, but then novaclient doesn't | 10:02 |
| Razique | neither euca2ools I think | 10:02 |
| Razique | so it's useless | 10:02 |
| Razique | not last version of Dash (need a confirmation here) | 10:02 |
| *** uncleofthestick has quit IRC | 10:05 | |
| *** uncleofthestick has joined #openstack | 10:05 | |
| *** Hakon|mbp has joined #openstack | 10:09 | |
| *** uncleofthestick has quit IRC | 10:14 | |
| *** uncleofthestick has joined #openstack | 10:15 | |
| vidd | Razique, i dontunderstand your script template | 10:16 |
| *** livemoon has quit IRC | 10:19 | |
| Razique | vidd: that would help you https://help.ubuntu.com/community/UbuntuBootupHowto | 10:20 |
| *** joeyang has quit IRC | 10:20 | |
| Razique | you simply create a script that run "/bin/keystone" | 10:20 |
| Razique | and link it to a runlevel | 10:20 |
| Razique | so u tell the system when launch your script | 10:20 |
| Razique | if you won't bother, simply add an entry into /etc/rc.local | 10:21 |
| Razique | but it ain't very pro :p | 10:21 |
| vidd | i have to do it right | 10:21 |
| uncleofthestick | Razique: Do you know if the trunk version of keystone can work with the stable/diablo release of Nova? I can't make it work | 10:22 |
| Razique | yah, cuz u'll remember the hack only a week, one month later, u won't ask urself "why is keystone starting at boot" hehe | 10:22 |
| Razique | uncleofthestick: it does, but recent client don't | 10:23 |
| vidd | or if it locks up and needs to be restarted | 10:23 |
| Razique | ooops misread | 10:23 |
| Razique | the trunk version doesn't | 10:23 |
| Razique | djeez, I can't properly write in english this morning | 10:23 |
| Razique | I must be on crack right now | 10:24 |
| vidd | Razique, what is your native language? | 10:24 |
| Razique | vidd: french | 10:24 |
| uncleofthestick | Razique: ok. I tried the stable/diablo version of keystone and it has a typo, so I dropped it inmmediately. | 10:24 |
| vidd | "a typo" is no reason to drop it | 10:25 |
| Razique | uncleofthestick: Cactus stable works with novaclient and keystone from githu | 10:25 |
| uncleofthestick | vidd: If the typo is the name of a constant of the code and the code does not compile, it is. ec1_url instead of ec2_url. | 10:25 |
| vidd | uncleofthestick, it compiles...i just ran the script | 10:26 |
| Razique | uncleofthestick: also # token_id = result['auth']['token']['id'] | 10:26 |
| Razique | token_id = result['access']['token']['id'] | 10:26 |
| uncleofthestick | Razique: Yeah, but we want to test the dashboard with flaoting ip management in it | 10:27 |
| vidd | and i dont think it was a typo...i think it was an upgrade (or regression) | 10:27 |
| vidd | ubuntu repo version has same issue | 10:27 |
| Razique | uncleofthestick: ok, so Dash/ keystone from github + nova 2011.2 (stable) + euca2ools works | 10:28 |
| Razique | does someone knows how I can choose the network that must be used by an instance | 10:28 |
| Razique | I created two networks for my project, both have a different vlan | 10:28 |
| uncleofthestick | Razique: Dash/Keystone trunk or stable/diablo? | 10:29 |
| Razique | uncleofthestick: trunk | 10:30 |
| zykes- | what you working on vidd ? | 10:30 |
| vidd | keystone | 10:30 |
| vidd | ./bin/keystone & does not give me my cursor back =\ | 10:30 |
| uncleofthestick | Razique: mmmm.... nova 2011.3 ? | 10:31 |
| Razique | uncleofthestick: this is Diablo | 10:31 |
| Razique | Cactus : 2011.2 | 10:31 |
| uncleofthestick | Razique: right, right... I just wanted to move on to Diablo | 10:31 |
| Razique | uncleofthestick: ok we published an upgrade guide here http://docs.openstack.org/trunk/openstack-compute/admin/content/migrating-from-cactus-to-diablo.html | 10:32 |
| *** shang has quit IRC | 10:32 | |
| Razique | vidd here kindly told me there was some minor fixes to do | 10:33 |
| Razique | i'll do that tonight | 10:33 |
| vidd | Razique, i told YOU kindly...my machine didnt think it was kind what i said to it =] | 10:33 |
| uncleofthestick | Razique: thanks, but we are taking a different approach. We have been testing Diablo before releasing our distro and we wanted to know if the Dashboard is functional or not. | 10:34 |
| Razique | haha | 10:34 |
| Razique | uncleofthestick: the dash totally is | 10:34 |
| Razique | but requires patience | 10:34 |
| uncleofthestick | Razique: I know... we have built the dashboard several times before the hard requirement of keystone integration | 10:35 |
| uncleofthestick | Razique: but now, if we want to use the dashboard, keystone must be up and running and integrated with nova and glance. | 10:36 |
| Razique | uncleofthestick: I thnik the dash is great, but somehow useless cuz u end up doing log and cli debugging | 10:36 |
| Razique | CLI's are much more powerfull imho | 10:36 |
| vidd | uncleofthestick, Razique y'all have a way to NOT run the dashboard in a venv? | 10:36 |
| uncleofthestick | Razique: I agree about the CLI | 10:36 |
| uncleofthestick | vidd: +10000000 | 10:36 |
| Razique | vidd: nope :/ | 10:36 |
| *** tryggvil_ has quit IRC | 10:36 | |
| *** NadeemNayeck has quit IRC | 10:37 | |
| vidd | i dont like the idea ofing the most powerful access to my cloud issolated from security updates | 10:38 |
| vidd | *of having* | 10:38 |
| uncleofthestick | Razique: we build a tiny distro for Openstack Nova called Stackops. We are still on Cactus because of its maturity. We tested diablo and we decided to delay the new version until we were confortable with it | 10:39 |
| Razique | uncleofthestick: i'm quite interested about a feedback about Diablo | 10:39 |
| Razique | I myself prepared into my lab a cactus upgrade | 10:40 |
| Razique | I need now to validate the new diablo functionalities, the most important ones to me : billing and boot from volumes | 10:40 |
| uncleofthestick | Razique: we are going to do a massive migrate next week in a customer. We have decided to drop Cactus and use Diablo as the production version, because of the backports | 10:40 |
| vidd | uncleofthestick, i almost went with the Stackops route for deployment...but the people in charge of the purse insist on having a GUI | 10:40 |
| *** troya has quit IRC | 10:41 | |
| uncleofthestick | vidd: StackOps is focused on quick and dirty test drive of Nova | 10:41 |
| Hakon|mbp | it's awesome for that | 10:41 |
| Hakon|mbp | i installed and fell in love in a matter of minutes | 10:42 |
| zykes- | having problems still vidd ? | 10:42 |
| *** stewart has joined #openstack | 10:42 | |
| vidd | zykes-, yeah...still cant force the keystone script into the background | 10:43 |
| Razique | Stackops is a great approach yah | 10:43 |
| Razique | for demos, quick test | 10:43 |
| uncleofthestick | Razique: we are now testing keystone and dashboard, and we think we are going to wait for a while for full integration | 10:43 |
| uncleofthestick | Razique: thanks | 10:43 |
| Razique | you welcome uncleofthestick | 10:43 |
| *** dcthang has quit IRC | 10:43 | |
| uncleofthestick | Razique: So I guess we will announce the new version without keystone and dashboard, but with some other new features | 10:44 |
| *** javiF has quit IRC | 10:44 | |
| Razique | uncleofthestick: yah agree ; I also state into guides that I myself don't consider keystone production-ready totally | 10:45 |
| uncleofthestick | Razique, vidd: guys, should we give a try again to the stable/diablo version of keystone again? | 10:45 |
| uncleofthestick | instead of trunk, I mean | 10:45 |
| Razique | uncleofthestick: u won't be able to make it work with the last tools | 10:46 |
| vidd | uncleofthestick, yes...get the stable | 10:46 |
| Razique | eg novaclient | 10:46 |
| Razique | uncleofthestick: also, the database still uses the old scheme | 10:46 |
| uvirtbot | New bug: #886046 in keystone "Add Quantum Integration Middleware" [Undecided,In progress] https://launchpad.net/bugs/886046 | 10:46 |
| Razique | su I dunno how painfull it'll be to make a keystone upgrade | 10:46 |
| vidd | Razique, stable does not work with novaclient? | 10:47 |
| Razique | vidd: not novaclient from the trunk :D | 10:47 |
| uncleofthestick | Razique: 2.6.6? | 10:47 |
| vidd | uncleofthestick, you are going with stable for everything...right? | 10:47 |
| Razique | uncleofthestick: noob question, but how can I know novaclient version ? | 10:47 |
| vidd | Razique, does --version not work? | 10:48 |
| Razique | nope :s | 10:48 |
| uncleofthestick | I was trying exactly --version, but it's the API version, not the client version | 10:49 |
| uncleofthestick | we bundle version 2.6.6 I guess... let me check... | 10:49 |
| vidd | THIS is the reason these scripts need to be attached somehow to /etc/init.d/ :http://dpaste.com/647188/ | 10:49 |
| vidd | Razique, ^^^^ | 10:50 |
| Razique | vidd: hehe, stop keystone first eh | 10:50 |
| uncleofthestick | vidd: I can give you my start script for keystone. It will be embedded in the new version of the STackops Distro | 10:50 |
| vidd | uncleofthestick, that would be great | 10:51 |
| *** dnjaramba has joined #openstack | 10:51 | |
| vidd | i need to know where to stick it too | 10:51 |
| vidd | uncleofthestick, are you by chance a Jeff Dunnam fan? | 10:52 |
| Razique | uncleofthestick: I'm also interested ; if you want, we could integrate it into doc | 10:52 |
| uncleofthestick | http://dpaste.com/647189/ | 10:53 |
| uncleofthestick | Feel free! | 10:53 |
| Razique | thanks a ton uncleofthestick ! | 10:55 |
| uncleofthestick | vidd: No, I don't know him. Why? | 10:55 |
| vidd | your nick here reminded me of one of his puppets (he is a ventriloquist) | 10:55 |
| *** fungo has joined #openstack | 10:56 | |
| *** NadeemMayeck has joined #openstack | 10:56 | |
| NadeemMayeck | hi all | 10:56 |
| Razique | vidd: I know him, he's great | 10:56 |
| uncleofthestick | Razique: Grab whatever you need from us or our public site. Just say the source | 10:56 |
| NadeemMayeck | i am new to cloud computing and openstack | 10:56 |
| Razique | guys, you know how to colortail nova logs ? | 10:56 |
| NadeemMayeck | i want to try it out | 10:56 |
| vidd | Razique, so do you see the joke? | 10:57 |
| Razique | uncleofthestick: yah sure | 10:57 |
| Razique | vidd : not sure :D | 10:57 |
| NadeemMayeck | i have a question: do I install openstack on the GUEST or the HOST ? | 10:57 |
| Razique | I only saw two of his performances | 10:57 |
| vidd | think of the jalepeno | 10:57 |
| Razique | the one with Archmed the terrorist | 10:57 |
| Razique | vidd: oh yes, I recall that one | 10:57 |
| vidd | uncle of "the stick" | 10:58 |
| vidd | =] | 10:58 |
| Razique | haha, now u say it | 10:59 |
| dosdawg | is citrix XCP worth a look? | 10:59 |
| vidd | NadeemMayeck, openstack IS the host | 10:59 |
| Razique | I wouldn't have think about that :D | 10:59 |
| uncleofthestick | vidd: my nickname is a joke. I'm spanish (I'm in Madrid) and sometimes translations from spanish to english are crazy. It's a nonsense direct translation from spanish to english | 10:59 |
| *** ton_katsu has quit IRC | 11:00 | |
| NadeemMayeck | is it possible to do nested virtualisation? meaning i want to install it in a GUEST virtual machine (Virtualbox on WIndow), and then created a GUEST vms within the VBox VM ? | 11:01 |
| dosdawg | yes nadeem | 11:01 |
| dosdawg | with qeum or lxc | 11:01 |
| vidd | NadeemMayeck, thats where i started my littel adventure =] | 11:01 |
| *** dnjaramba has quit IRC | 11:01 | |
| dosdawg | think openvz too but don't think openstack uses openvz | 11:02 |
| Razique | NadeemMayeck: yup you can | 11:02 |
| dosdawg | qemu | 11:02 |
| Razique | NadeemMayeck: that guide is well written http://uksysadmin.wordpress.com/2011/02/17/running-openstack-under-virtualbox-a-complete-guide/ | 11:02 |
| vidd | uncleofthestick, so you have a nephew named "palo"? | 11:02 |
| *** jchome has joined #openstack | 11:02 | |
| jchome | hej | 11:02 |
| *** shang has joined #openstack | 11:02 | |
| NadeemMayeck | awesome, i am going to read up a bit and probably get back here with ten times more questions :) | 11:03 |
| Razique | NadeemMayeck: sure :) | 11:03 |
| jchome | I kind of having trouble setting up the network for openstack compute | 11:03 |
| Razique | uncleofthestick: vidd any "multi host networking" feedback ? | 11:03 |
| vidd | Razique, didnt we talk to that guy here yesterday? | 11:03 |
| Kiall | vidd / Telamon .. very late response but whatever.. yea I have a PPA with everything packaged for ubuntu oneiric, and yes - swift is in in there, but its untested... https://launchpad.net/~managedit/+archive/openstack | 11:03 |
| jchome | mostly because the docs are a bit shallow | 11:03 |
| Razique | jchome: don't hesitate to open bugs for the docs | 11:04 |
| Telamon | Kiall: Cool, does that include keystone and horizon? I've got everything else seemingly working on one of my boxes now. | 11:04 |
| Telamon | Those two scare me though. Lots of dependancies that aren't in the main repos or well defined. | 11:05 |
| Razique | jchome: We are trying to update docs quickly so it's consistent with the current devs | 11:05 |
| jchome | Razique: it's basically not clear to me how to configure the network interfaces on the openstack nova-network node in order to function properly - I'd help writing docs once I'd know how the setup works | 11:05 |
| Razique | jchome: u are trying to run a flat dhcp mode right ? | 11:06 |
| Kiall | Telamon: yes .. nova/glance/keystone/horizon included + tested .. | 11:06 |
| Kiall | swift is included but, I havent installed it.. | 11:06 |
| jchome | Razique: yes | 11:06 |
| Razique | jchome: I need to create a network schema, WIP :) | 11:06 |
| jchome | Razique: based on a debian system, which is similar to ubuntu of course | 11:06 |
| *** dnjaramba has joined #openstack | 11:06 | |
| *** naehring has quit IRC | 11:06 | |
| jchome | Razique: but basically whats missing is the previous config needed | 11:06 |
| Razique | jchome: how did u installed nova ? | 11:06 |
| jchome | Razique: debian packages | 11:07 |
| uncleofthestick | vidd: Translating 'tio' from spanish to english can be 'uncle' or 'dude'. It's a popular comic show here; http://www.youtube.com/watch?v=NCJN6dimD5I | 11:07 |
| *** ahasenack has joined #openstack | 11:07 | |
| Telamon | Sweet... I might blow my second server away and give that a try. Much easier than installing them from git! | 11:07 |
| jchome | Razique: these ones: http://wiki.openstack.org/Packaging/Debian | 11:07 |
| Razique | jchome: which versions ? | 11:07 |
| vidd | Kiall, i would be interested in your .deb for keystone and dash | 11:07 |
| uncleofthestick | Razique: We don't go with multihosting yet. | 11:07 |
| *** javiF has joined #openstack | 11:07 | |
| uncleofthestick | Razique: Can't help you man! | 11:08 |
| jchome | Razique: the version on there is 2011.2-1 | 11:08 |
| Kiall | vidd: the main reasons the ubuntu packages for keystone+dash dont work is because of a broken openstackx package, and the included keystone being too old.. | 11:08 |
| Razique | uncleofthestick: ok ok I'm testing it, works great =) | 11:08 |
| Razique | but I dunno how's the integration goes with a running cloud | 11:08 |
| Razique | jchome: ok, could you repaste your nova.conf (tons of conf pass here) | 11:09 |
| jchome | Razique: https://gist.github.com/1160675f08e47fc6a359 | 11:09 |
| foexle | Razique: i read you post on launchpad (https://lists.launchpad.net/openstack/msg04936.html) .... if i try with curl and your changes to authenticat with keystone i get bad request (400) ..... why i'm looking i cant use glance-command i get every time from keystone 400 response. | 11:10 |
| foexle | +r | 11:10 |
| jchome | Razique: all the daemons are running, but when it comes to adding a network it fails hard | 11:10 |
| vidd | jchome, did you configure nova-volume? | 11:11 |
| jchome | vidd: how? | 11:11 |
| jchome | vidd: I think it's starting, but right now I don't see it in the process list | 11:12 |
| vidd | jchome, if you have to ask, then the answer would be "no" =] | 11:12 |
| Kiall | lol | 11:12 |
| jchome | vidd: are there docs for it, I'd do it right away :) | 11:12 |
| vidd | did you set the lvm up for nova-volume? | 11:13 |
| Kiall | basically add a LVM volume group called "nova-volumes" is the big one.. | 11:13 |
| vidd | it makes all but 2? of your nova apps die...apinfully | 11:13 |
| *** tim|buntu has left #openstack | 11:13 | |
| Razique | jchome: ok now ifconfig please :) | 11:13 |
| *** dendrobates is now known as dendro-afk | 11:13 | |
| jchome | Razique: https://gist.github.com/3fd64e8bc5405dc7614b | 11:14 |
| vidd | Kiall, i think its the OTHER problem i had with networking =] | 11:14 |
| Razique | foexle: let me help jchome :p | 11:14 |
| Razique | I remember that keystone issue | 11:15 |
| jchome | vidd: harr, didn't do that - wasn't mentioned up until the part where the network happens in the docs | 11:15 |
| foexle | Razique: yeah no problem | 11:15 |
| Razique | jchome: ok the networking part is ok, are you able to run instances (sorry If I already asked the same question yesterday :D ) | 11:16 |
| Razique | time to lunch, be back in an hour tops | 11:17 |
| vidd | jchome, when you do get to that part, make sure you do the viso....if your instructions dont have that....look here http://docs.openstack.org/diablo/openstack-compute/admin/content/managing-volumes.html | 11:17 |
| jchome | Razique: no because the 'nova-manage network create' fails | 11:18 |
| Razique | jchome: ok I remember | 11:18 |
| vidd | Razique, go get lunch...ill lend a hand | 11:18 |
| jchome | Razique: remember? I haven't been here before, but ok :D | 11:18 |
| jchome | https://gist.github.com/3e6c44949fe3317d14a5 | 11:18 |
| jchome | thats the error | 11:18 |
| *** tyska has joined #openstack | 11:18 | |
| jchome | in /var/log/nova/nova-manage.log | 11:18 |
| vidd | jchome, what are you typing for the nova-manage network create? | 11:19 |
| tyska | how guys | 11:19 |
| *** rsampaio has quit IRC | 11:19 | |
| tyska | ops | 11:19 |
| tyska | hi guys | 11:19 |
| tyska | =) | 11:19 |
| jchome | vidd: 'nova-manage network create novanet 10.26.1.0/24 1 256' | 11:19 |
| Razique | jchome: try nova-manage network create private 172.16.1.0/24 1 256 | 11:19 |
| tyska | im having a problem to run a instance | 11:19 |
| tyska | see all the details here: http://pastebin.com/356gurxQ | 11:19 |
| jchome | oh, yea, I changed the fixed_range prefix | 11:19 |
| Razique | jchome: try mine :p | 11:19 |
| tyska | Razique: im still on the battle =) | 11:19 |
| jchome | Razique: https://gist.github.com/6a7de3973170058ddc95 | 11:19 |
| Razique | hi tyska | 11:20 |
| Razique | how r doing ? | 11:20 |
| tyska | im great, u? | 11:20 |
| vidd | jchome, Razique there are lines in his novarc that have http://http..... in them that is messing him up | 11:20 |
| Razique | jchome: tail -100 /var/log/nova/nova-network.log | 11:20 |
| *** rsampaio has joined #openstack | 11:20 | |
| Razique | tyska: (nova.rpc): TRACE: RemoteError: NoMoreNetworks None | 11:20 |
| tyska | Razique: what that means? | 11:21 |
| jchome | Razique: https://gist.github.com/c09a4e1723e138b3f4e5 | 11:21 |
| Razique | tyska: nova-manage network private 172.16.1.0/24 1 256 | 11:21 |
| Razique | jchome: sorry, nova-manage not nova-network | 11:21 |
| vidd | jchome, i had the same issue | 11:22 |
| jchome | Razique: https://gist.github.com/f9fd18e45f7493d316d0 | 11:22 |
| vidd | jchome, there are lines in your novarc that have http://http..... in them that is messing him up | 11:22 |
| tyska | but Razique, the command nova-manage network list returns: | 11:22 |
| tyska | 1 172.16.50.0/24 None 172.16.50.3 None None 100 proj None | 11:22 |
| Razique | ok | 11:22 |
| Razique | then nova-manage network modify --project $project --network $network | 11:22 |
| jchome | vidd: there are only 2 http lines: | 11:23 |
| jchome | --ec2_host=http://10.26.0.12 | 11:23 |
| jchome | --osapi_host=http://10.26.0.12 | 11:23 |
| Razique | tyska: u need to link that network to ur project | 11:23 |
| tyska | Razique: oh yes | 11:23 |
| tyska | $network is the ID? | 11:23 |
| vidd | jchome, that is your nova.conf....the issue is your novarc | 11:23 |
| Razique | nope it's the network with it's mask | 11:23 |
| Razique | eg 172.16.1.0/24 | 11:23 |
| tyska | ok | 11:24 |
| vidd | jchome, you had both those lines in your nova.conf when you created your novarc ant that is what messed you up | 11:25 |
| jchome | vidd: '# find / -name novarc' returns empty | 11:25 |
| vidd | jchome, remember when you created the zipfile? | 11:25 |
| *** dendro-afk is now known as dendrobates | 11:25 | |
| jchome | vidd: I created a zipfile? | 11:25 |
| Razique | jchome: ok, i'll be back later | 11:26 |
| Razique | the network command is definitely good, somewhere, something is weird | 11:26 |
| vidd | you should have...you cant add ip's to your project until you get your credentials | 11:26 |
| Razique | cya guys :) | 11:26 |
| foexle | Razique: tgis is the fix right ? | 11:26 |
| foexle | https://github.com/openstack/glance/commit/bc7aeb4a0a1295f90f7ea8824a1c8bf5313fafb7 | 11:26 |
| jchome | vidd: oh, yea - the 'nova-manage user admin | 11:27 |
| foexle | Razique: but its not in the debian packages released | 11:27 |
| vidd | jchome, right...when you unzip the file it makes a novarc | 11:27 |
| jchome | vidd: it didn't create a zipfile, I'm confused by now | 11:27 |
| *** lelin has joined #openstack | 11:28 | |
| vidd | ok....so you are doing things out of order | 11:28 |
| jchome | vidd: it just returned 'EC2_ACCESS_KEY' and 'EC2_SECRET_KEY' | 11:28 |
| vidd | what doc you following? | 11:28 |
| jchome | http://docs.openstack.org/diablo/openstack-compute/admin/content/setting-up-openstack-compute-environment-on-the-compute-node.html | 11:28 |
| *** takarider has joined #openstack | 11:29 | |
| *** Telamon has quit IRC | 11:29 | |
| vidd | jchome, hold off on making the network for now.... | 11:29 |
| tyska | Razique: instance running! =) | 11:29 |
| lelin | hi, i have a cloud running on 2 systems with one interface each on the same net and using a gateway (which is none of those 2 machines but a third one). if i want to allocate an ip i receive an error. is this because the controller should have 2 network interfaces? | 11:29 |
| vidd | go to the next page | 11:29 |
| tyska | Razique: but now im with troubles in the euca-authorize: see this http://pastebin.com/LSyK19cm | 11:30 |
| tyska | Razique: i guess i need to do a euca-add-group, that is it? | 11:30 |
| vidd | lelin, the compute node should have 2 | 11:30 |
| tyska | euca-add-group c3cloud? | 11:30 |
| vidd | jchome, you see where it tells you about making creds? | 11:31 |
| jchome | vidd: yes | 11:31 |
| jchome | vidd: doing that now | 11:31 |
| vidd | dont go past the unzip part jchome | 11:32 |
| jchome | vidd: and there it has http://http - yes | 11:32 |
| lelin | vidd, thanks. so is not enough to have the controller with 2 nics, and leave all the nodes on the same net, with only one interface. right? | 11:32 |
| jchome | vidd: https://gist.github.com/01b686913df4a597507b | 11:32 |
| vidd | lelin, it CAN be done...just a major PITA | 11:32 |
| *** stewart has quit IRC | 11:33 | |
| vidd | jchome, fix those lines....then cat and source | 11:33 |
| vidd | THEN run the network create command | 11:33 |
| jchome | vidd: https://gist.github.com/529d26bc989ca7ce4756 | 11:34 |
| jchome | https://gist.github.com/1b0a3cb3ffce3ff6f714 | 11:35 |
| jchome | same error | 11:35 |
| tyska | =( | 11:37 |
| tyska | i just cant ssh in to a instance =( | 11:37 |
| tyska | let me pray for some minutes | 11:37 |
| vidd | jchome, you have you database on mysql? | 11:37 |
| *** kaigan_ has quit IRC | 11:37 | |
| jchome | vidd: yep | 11:37 |
| *** fungo has quit IRC | 11:37 | |
| vidd | jchome, you have phpmyadin installed so yo can look at your database? | 11:38 |
| jchome | vidd: I can just use the mysql cli ;) | 11:38 |
| jchome | What should I look for? | 11:39 |
| vidd | the "networks" table | 11:39 |
| jchome | it's empty | 11:39 |
| *** zz_bonzay is now known as bonzay | 11:40 | |
| vidd | it shouldnt be.... | 11:40 |
| dosdawg | citrix xcp sux..... stripped down open core ... pos | 11:40 |
| jchome | vidd: where does the data come from? the network create command fails | 11:40 |
| vidd | what does your nova.conf file say | 11:41 |
| dosdawg | openstack needs a distro... something like stackops but lighter | 11:41 |
| jchome | vidd: https://gist.github.com/effbd9aed81f874d68da | 11:41 |
| vidd | dosdawg, whats heavey with stackops? | 11:42 |
| uncleofthestick | dosdawg: I'm the Stackops guy, can you give me more feedback about that? Sounds interesting... | 11:42 |
| vidd | jchome, is there reason you chose "172" for your class A | 11:43 |
| dosdawg | well it's a little dated is all. | 11:44 |
| vidd | jchome, is 192.x.x.x in use? | 11:44 |
| jchome | vidd: it's a RFC1918, it's free in my test env - that why I used it | 11:44 |
| jchome | vidd: can do 192.168. too | 11:44 |
| jchome | 10/8 is a bit of a trouble | 11:44 |
| *** kaigan_ has joined #openstack | 11:45 | |
| vidd | jchome, i only say because "localhost" is 172.0.0.0 | 11:45 |
| dosdawg | don't really like to register , and all I guess. I know you can local install it. the idea is there , but dosen't seem to be a current "stack" | 11:45 |
| jchome | vidd: localhost is 127/8 :) | 11:45 |
| jchome | vidd: 172.16/12 is a RFC1918 like 10/8 and 192.168./16 | 11:46 |
| vidd | jchome, i didnt say there should be any REASON it shouldnt work =] | 11:46 |
| uncleofthestick | dosdawg: I understand your concerns. This is not the place to talk about business, but we are going on that way. | 11:46 |
| jchome | vidd: yea, it's just a diff network from the localhost network :) | 11:47 |
| vidd | does changing the localnet; restarting nova help any? | 11:48 |
| *** bonzay is now known as zz_bonzay | 11:48 | |
| dosdawg | I'm playing with drbl and stackops .. some interesting things can be done in this way. roll that into an install iso | 11:48 |
| uncleofthestick | dosdawg: to be honest, we never expected to be so succesful. We did the distro to learn about Nova, to force us with a real project to learn how it works. But then it boomed. | 11:48 |
| vidd | dosdawg, you said "something lighter" what is heavey about it? dated != heavey | 11:49 |
| jchome | vidd: nope | 11:49 |
| jchome | same error | 11:50 |
| *** supriya has quit IRC | 11:50 | |
| vidd | jchome, try commenting out the dhcp start line | 11:50 |
| dosdawg | the whole web install process is what I meant. something lighter as an installable distro that is devstack , not a os with devstack loaded on top | 11:50 |
| dosdawg | I mean I like it don't get me wrong , when I started it was awesome to get you going quick | 11:51 |
| jchome | vidd: still the same error persists | 11:51 |
| vidd | dosdawg, devstack is an installer script....a "one&done" demo | 11:51 |
| dosdawg | yea I use devstack now | 11:52 |
| foexle | this resolved bug (https://github.com/openstack/glance/commit/1bb0600efb444fa4d8a27e876a8c002324746c1e) are not in the debian packages ? :( | 11:52 |
| *** PeteDaGuru has joined #openstack | 11:52 | |
| vidd | dosdawg, every time you restart, you have to re-run the script and loose all that dat.stack ops is permanant | 11:52 |
| dosdawg | vidd no you don't | 11:53 |
| dosdawg | just skip to the part of the script that is the startup | 11:53 |
| dosdawg | copy paste into a startup file like I did | 11:53 |
| vidd | dosdawg, trying...prove me wrong | 11:53 |
| uncleofthestick | dosdawg: well, we think nova and swift needs a lot of infrastructure to run, so basically you need to focus in not repeating the configuration process after Node 1. Install Node 1 is hard, but then new nodes must be like a breeze | 11:54 |
| dosdawg | yea thats what I am looking at drbl for | 11:54 |
| dosdawg | pxe boot and provision live | 11:55 |
| vidd | dosdawg, you have a link for drbl? | 11:55 |
| dosdawg | drbl.sourceforge.net | 11:55 |
| vidd | jchome, im not seeing you doing anything wrong =\ | 11:55 |
| *** katkee_ has joined #openstack | 11:55 | |
| jchome | vidd: Yea, thinking about just reinstalling and doing it from the start | 11:55 |
| vidd | jchome, did you install from repo or git? | 11:55 |
| jchome | vidd: from the repo | 11:55 |
| jchome | vidd: the git devstack isn't working on debian | 11:56 |
| vidd | what ? devstack doesnt work for you? | 11:56 |
| n88 | how easy is it to install openstack from git/source? Has anyone done it here? | 11:56 |
| jchome | vidd: on debian? that would be new | 11:56 |
| dosdawg | I wouldn't say it is easy at all :) | 11:56 |
| *** livemoon has joined #openstack | 11:57 | |
| jchome | vidd: ubuntu would work, yea | 11:57 |
| vidd | jchome, your not running ubuntu? | 11:57 |
| jchome | as I'm saying the whole time :) | 11:57 |
| tyska | im back! | 11:57 |
| vidd | where did you get these debs?> | 11:57 |
| jchome | http://wiki.openstack.org/Packaging/Debian | 11:57 |
| *** egant has joined #openstack | 11:58 | |
| *** hggdh has joined #openstack | 11:58 | |
| jchome | I suspect it's just broken. | 11:58 |
| vidd | jchome, i suspect it is.... | 11:58 |
| n88 | dosdawg: I'm encountering this bug https://bugs.launchpad.net/nova/+bug/834633; it looks like it was fixed and merged into the master: https://review.openstack.org/#change,628, but I installed from apt-get on oneiric and the changes have obviously not made it into the packages in the apt-get repos | 11:58 |
| *** katkee has quit IRC | 11:58 | |
| vidd | have you though of grabbing the ubuntu .deb of the software list? | 11:59 |
| vidd | n88, repos dont do "master" they do "stable" | 12:00 |
| *** yeming has joined #openstack | 12:01 | |
| n88 | vidd: ok…how long does it usually take to go from master to stable? months? weeks? … I'd really like my instances to get auto-assigned IPs…I'm somewhat new to using open source software products like this…should I just grab the files they "fixed" and replace the python files I have now on my machine? | 12:01 |
| n88 | vidd: or should I look into building from master since it seems like there's a lot of updates/fixes there... | 12:02 |
| n88 | I had to build a patch to fix this issue as well: https://bugs.launchpad.net/ubuntu/+source/nova/+bug/871278 | 12:03 |
| yeming | Hey Razique, vidd : I killed dnsmasq and restarted nova-network, and then euca-reboot-instances, and now euca-get-console-output looks good and i can ping the instance | 12:04 |
| dosdawg | gnight fellas , checkout drbl and see what you think. if you make everything work on one server , each server you boot has everything in place to turn on or off. | 12:04 |
| vidd | it looks like alot of these changes will be in the 12.04 realease of ubuntu | 12:05 |
| n88 | vidd: yeah I saw that as well…don't really want to wait until next April =P…maybe I'll just grab the source files changed and copy them over to my installs…and then write a patch file or something | 12:05 |
| vidd | n88, Kiall has "works better than ubuntu" PPAs =] | 12:07 |
| n88 | vidd: so a PPA is a not-yet-released repo, correct? so you're saying I should contact Kiall and get the location of his PPAs and then use those instead of the default apt-get repos? | 12:09 |
| *** cereal_bars has joined #openstack | 12:09 | |
| vidd | n88 PPA's are "Private" repo's | 12:10 |
| tyska | ok, lets restart | 12:11 |
| tyska | im having troubles to ssh in the instances | 12:11 |
| vidd | n88, and you dont have to hunt him down...the link is --->https://launchpad.net/~managedit/+archive/openstack/ | 12:11 |
| Kiall | n88: PPA as just somewhere for non-ubuntu folks to put packages.. ie they wont get included in ubuntu directly.. | 12:12 |
| vidd | n88 you MAY have better luck with them | 12:12 |
| vidd | perfect fot jchome | 12:12 |
| n88 | vidd, Kiall: Wow! Thanks! I particularly like this line: WARNING: If these cause your datacenter to spontaneously combust, don't blame us! | 12:13 |
| Kiall | They do all work together, until the current ubuntu packages.. I can guarantee that ;) but .. you beat me to the warning! | 12:13 |
| jchome | vidd: whats perfect for me? ;) | 12:13 |
| Kiall | unlike the current* | 12:13 |
| vidd | jchome, Kiall 's PPA's for keystone | 12:14 |
| vidd | Kiall, you just updated your horizon again | 12:14 |
| vidd | does it work outside a venv? | 12:15 |
| Kiall | vidd: only, I have no idea if combining my packages with ubuntu's will work.. I've only tested using all my packages together, with no trace of manually installed components ;) | 12:15 |
| n88 | vidd, Kiall: sorry for all the n00b questions, but if I add those PPAs, and run an apt-get update and upgrade, will it then pull from the PPAs or would you recommend just starting over and re-installing from the PPAs? | 12:15 |
| Kiall | yea.. i never needed that for the .deb ;) | 12:15 |
| vidd | n88, we were all n))bs once upon ia time..... | 12:15 |
| jchome | vidd: excep they're for ubuntu and not debian, right? | 12:15 |
| vidd | except me...i have yet to acchieve the high rank of noob =] | 12:15 |
| Kiall | n88: if you have only installed from .deb's then you can follow the instructions (there is a extra package to install than ensures my packages are used over ubuntus) | 12:15 |
| *** johnlemon has joined #openstack | 12:16 | |
| Kiall | if you have installed anything manually, eg via devstack, or "python settup.py" commands etc etc | 12:16 |
| *** bsza has joined #openstack | 12:16 | |
| Kiall | then those will override any packages, nothing can be done about that.. you would need to clean your /usr/local/* folders to remove any trace of the manually installed stuff | 12:16 |
| vidd | jchome, that is a Kiall question =] | 12:17 |
| tyska | guys, please help me here. See this: http://pastebin.com/vhLDr4Cc and say why i cant connect in my instance =( | 12:17 |
| Kiall | jchome: yes, they are for ubuntu .. ubuntu 11.10 specifically | 12:17 |
| vidd | tyska, if we knew what to tell you we would | 12:17 |
| Kiall | tyska: have you checked the console output? | 12:17 |
| n88 | Kiall: Ok, so to make sure I'm awake enough to understand everything…if I used apt-get to install my openstack, I should be able to add that PPA, then run an apt-get update and apt-get upgrade on openstack and that **should** update all my openstack components. Correct? | 12:17 |
| jchome | Yea, need Debian packages.. not gonna mixup my architecture by introducing ubuntu.. | 12:18 |
| tyska | Kiall: yeah, im getting the following error: | 12:18 |
| Kiall | tyska: with euca-get-console-output | 12:18 |
| Kiall | jchome: the source for all of them is @ http://github.com/managedit | 12:18 |
| tyska | 2011-11-04 12:00:01,416 - DataSourceEc2.py[WARNING]: 12:00:01 [100/100]: url error [[Errno 111] Connection refused] | 12:18 |
| Kiall | chances are, you can update them for debian by just changing the debian/changelog and debian/control files | 12:19 |
| Kiall | and then re-building them | 12:19 |
| Kiall | (you can rebuild them by cloning + running "git-buildpackage -us -uc") | 12:20 |
| tyska | Kiall: did you see the error? | 12:21 |
| tyska | 2011-11-04 12:00:01,416 - DataSourceEc2.py[WARNING]: 12:00:01 [100/100]: url error [[Errno 111] Connection refused] | 12:21 |
| Kiall | tyska: yea, is nova-api running? | 12:21 |
| tyska | Kiall: yeah | 12:22 |
| vidd | Kiall, does git-buildpackage -us -uc on your package make them autostart? | 12:23 |
| Kiall | and "netstat -lnp | grep 8773" shows a python service listening? | 12:23 |
| Kiall | vidd: that just builds the .deb's.. | 12:24 |
| tyska | tcp 0 0 0.0.0.0:8773 0.0.0.0:* LISTEN 15360/python | 12:24 |
| tyska | yeah | 12:24 |
| tyska | but here there is a thing | 12:24 |
| tyska | when i generate the credentials | 12:24 |
| Kiall | but they are configured to autostart | 12:24 |
| tyska | the novarc comes with the wrong port | 12:24 |
| tyska | it comes with 8443 | 12:24 |
| tyska | i thing this could be causing all this mess | 12:24 |
| Kiall | tyska, and "iptables -t nat -n -L | grep 169" shows? | 12:24 |
| tyska | DNAT tcp -- 0.0.0.0/0 169.254.169.254 tcp dpt:80 to:10.0.0.254:8443 | 12:24 |
| Kiall | thats it | 12:25 |
| tyska | THE WRONG PORT!!!! | 12:25 |
| tyska | =) | 12:25 |
| tyska | i need to change to 8773? | 12:25 |
| Kiall | the metadata url (http://169.254.169.254/) is gettting forwarded to 8443, rather than 8773 | 12:25 |
| Kiall | no idea where/how to change that though ;) | 12:25 |
| tyska | but why? | 12:25 |
| tyska | yeah | 12:25 |
| tyska | iptables i know well | 12:25 |
| Kiall | well - iptables isnt the issue | 12:26 |
| tyska | but i wanna know why the port is 8443? | 12:26 |
| vidd | tyska, look at your nova.conf file...is it correct there? | 12:26 |
| Kiall | try `nova-manage config list | grep ec2_port` | 12:26 |
| tyska | vidd: yeah | 12:26 |
| tyska | vidd: i changed it because it was wrong | 12:26 |
| Kiall | I get --ec2_port=8773 | 12:26 |
| tyska | 8777 | 12:26 |
| tyska | --ec2_port=8443 | 12:27 |
| Kiall | brb | 12:27 |
| tyska | =O | 12:27 |
| vidd | tyska, did you change it befor or after your credentials were built ? | 12:27 |
| tyska | vidd: before | 12:27 |
| n88 | While I'm here…I guess I could ask. On my nova-compute node, my nova-compute is failing to start on reboot because libvirtd hasn't started yet. Can anyone tell me where nova-compute is actually started on boot? I looked through all the rc levels and don't even see nova-compute? I also added libvirt-bin to all the rc levels (default) and nova-compute is still trying to start before libvirt | 12:27 |
| *** hadrian has joined #openstack | 12:28 | |
| tyska | vidd: i changed the port, and generate the creds again | 12:28 |
| tyska | guys, my question is, how can i change the port to 8443? I think it is better, must i change de nova.conf? | 12:28 |
| tyska | *the | 12:28 |
| vidd | tyska, did you wipe the database or just add new credentials? | 12:29 |
| *** yeming has quit IRC | 12:29 | |
| *** shang has quit IRC | 12:29 | |
| tyska | vidd: no, just credentials | 12:29 |
| vidd | yeah...the damage was already done | 12:29 |
| tyska | vidd: i need to change something in the database? | 12:30 |
| vidd | try this: | 12:30 |
| vidd | nova-manage db sync | 12:30 |
| tyska | ok | 12:30 |
| vidd | kill the instances first | 12:30 |
| tyska | done | 12:30 |
| tyska | done | 12:30 |
| vidd | once all instaces are terminated....the database is synced, restart all your nova apps | 12:31 |
| *** sandywalsh_ has quit IRC | 12:31 | |
| vidd | then check your database table ... is there anything in provider_fw_rules? | 12:33 |
| *** msivanes has joined #openstack | 12:34 | |
| *** sandywalsh_ has joined #openstack | 12:35 | |
| *** johnlemon has quit IRC | 12:35 | |
| *** lorin1 has joined #openstack | 12:36 | |
| tyska | what is the default database name of the nova? | 12:36 |
| vidd | it is listed in your nova.conf file | 12:36 |
| *** sandywalsh_ has quit IRC | 12:38 | |
| tyska | vidd: Empty set (0.00 sec) | 12:39 |
| tyska | i think this is not good =( | 12:39 |
| livemoon | hi,all | 12:39 |
| vidd | tyska, no...mine is empty as well | 12:40 |
| vidd | its for port forwarding | 12:40 |
| vidd | try starting another instance and see if it works now | 12:40 |
| vidd | so Kiall your saying git overides apt-get? | 12:42 |
| *** javiF has quit IRC | 12:43 | |
| vidd | tyska, if this does not fix it, im going to suggest wiping your nova database and rebuilding it | 12:43 |
| Kiall | vidd: no | 12:44 |
| *** sandywalsh_ has joined #openstack | 12:44 | |
| Kiall | I have a package called "managedit-openstack-pin" which causes apt to prefer my PPA over stock packages.. | 12:44 |
| Kiall | preventing the ubuntu packages from being used instead.. even if the ubuntu ones version numbers appear higher to apt | 12:45 |
| Kiall | git-buildpackage is just what I use to build the .deb from the sources | 12:45 |
| Kiall | it takes the source, and produces a .deb.. nothing more | 12:45 |
| vidd | so if i run apt-get it will write over what i install from git | 12:45 |
| Kiall | no.. if you install stuff from git, it gets lobbed into /usr/local.. stuff in /usr/local overrides whatever you install from packages into /usr | 12:46 |
| livemoon | hi,vidd | 12:46 |
| vidd | and if i run git it will overwrite what i install from apt-get? | 12:46 |
| livemoon | what are you talking about? | 12:46 |
| vidd | hello livemoon | 12:46 |
| vidd | keystone in ubuntu repos is broken...but they autostart.....keystone in gi [supossedly] work properly but do not autostart | 12:48 |
| *** cereal_bars has quit IRC | 12:48 | |
| vidd | i want to know which i need to install first to get the autostart and the working product =] | 12:48 |
| *** tryggvil has joined #openstack | 12:49 | |
| Kiall | vidd: "if i run git" you mean, if you use git-buildpackage? no. git-buildpackage takes the source, and builds a .deb. nothing gets installed. It just makes the .deb file | 12:49 |
| n88 | Kiall: So I tried to update/upgrade using the managedit PPAs; unfortunately I'm still getting the FloatingIpNotFoundForAddress error; so I'm assuming either something was setup incorrectly on my end OR the PPA packages do not yet include the bug fix found on the master openstack branch. How do I see what revision of the master was used for the PPAs? | 12:50 |
| vidd | Kiall, no...cit clone git xyz...setup.py | 12:50 |
| Kiall | n88: my packages are based on the diablo/stable branches | 12:50 |
| Kiall | ok yea.. that will override the packaged stuff | 12:50 |
| Kiall | at least, mostly.. | 12:50 |
| Kiall | if there are files that have been removed, the packages versions will stay "active" | 12:51 |
| Kiall | so .. bascially you end up with a mess (again) | 12:51 |
| n88 | Kiall: Ah ok…so I still probably need to grab the updated files and create a patch file until those fixes get added into a stable release | 12:51 |
| Kiall | you have to pick 1 method, and roll with it. | 12:51 |
| livemoon | vidd: now I use git in my production | 12:51 |
| Kiall | n88: which bug is this? | 12:51 |
| Kiall | and what commit are you thinking of? | 12:51 |
| vidd | Kiall, does your git setup for keystone autolaunch? | 12:52 |
| Kiall | n88: ah .. its the auto assign floating IP bug? | 12:52 |
| Kiall | vidd: all my packages auto launch | 12:52 |
| vidd | Kiall, sold =] | 12:52 |
| Kiall | lol | 12:52 |
| Kiall | n88: I'm been wanting to backport those fixes myself aswell, but I havent had a chance just yet | 12:53 |
| n88 | Kiall: Yes. I mean I don't think I've got anything configured incorrectly. Have that flag in my nova.conf. Have used the nova-manage floating create; see the floating IPs in my DB | 12:53 |
| *** clopez has joined #openstack | 12:53 | |
| Kiall | n88: If you know how to fix it, I'll happily rebuild the packages with the fix :) | 12:54 |
| vidd | tyska, you good now? | 12:54 |
| tyska | vidd: nova-manage config list still returns the port 8443 to ec2_url | 12:55 |
| n88 | Kiall: I've also noticed that the flag force_dhcp_release doesn't seem to work as well…according to Vish that flag should cause a terminated instance to release it's private IP as soon as the instance gets terminated and the next instance to startup will get that just released IP…but even with that flag set in nova.conf I'm seeing the 2nd instance get the next IP (Vish said otherwise IPs are released after about 10 minutes) | 12:55 |
| tyska | vidd: now im trying to change the port to 8443 | 12:55 |
| tyska | vidd: i changed it on nova.conf | 12:55 |
| tyska | vidd: but netstat -lnp still show 8773 | 12:55 |
| n88 | Kiall: I'm going to work to see if I can patch it and fix it…if it does work, I'll send you my updates (or post them on paste bin or something) | 12:55 |
| vidd | tyska, then you have to delete the database and start it again =\ | 12:55 |
| vidd | Kiall, your git is "master" right? | 12:57 |
| Kiall | its based on the debian/stable branches | 12:57 |
| Razique | baxk | 12:57 |
| Razique | rainy day here | 12:57 |
| Kiall | n88: if you want to have a go at fixing the packages directly ;) http://pastie.org/2809760 | 12:57 |
| vidd | Razique, tyska 's issue is her(?) database is toast | 12:58 |
| *** root has joined #openstack | 12:58 | |
| vidd | it was built with incorrect port # for EU2 | 12:59 |
| *** root is now known as Guest86111 | 12:59 | |
| Razique | vidd: can up sum up ? | 12:59 |
| zykes- | vidd: getting it to run or ? | 12:59 |
| Guest86111 | hi | 12:59 |
| n88 | Kiall: Let me see how far i get on fixing mine…if I can get it working, I'll look at editing directly | 12:59 |
| Guest86111 | vishy? | 12:59 |
| Guest86111 | i have a prblem with the nova-common installation | 12:59 |
| Kiall | n88: just updated that pastie with more detail.. http://pastie.org/2809760 | 13:00 |
| vidd | Razique, i recommend delete the database and rebuild it | 13:00 |
| tyska | Razique: im trying to change my ec2_port to 8443 | 13:00 |
| * vidd has to run...be back in about 1 hour | 13:00 | |
| Guest86111 | because when i type: aptitude reistall nova-commin | 13:00 |
| *** vidd is now known as vidd-away | 13:01 | |
| Guest86111 | the error is: nova.conf no such file or directory | 13:01 |
| n88 | Kiall: So are you mainly working on packages for Debian? Or does your stuff apply to Debian and Ubuntu? | 13:01 |
| Kiall | Just ubuntu.. | 13:01 |
| *** npmapn has quit IRC | 13:01 | |
| Guest86111 | yes for debian | 13:01 |
| Kiall | but ubuntu packages used the word "debian" all over the place | 13:01 |
| tyska | Razique: are u there? | 13:02 |
| n88 | Kiall: Ah ok, my noobishness is exposed again…thanks for the info. | 13:02 |
| Razique | tyska: yup | 13:03 |
| Guest86111 | any ideas? | 13:03 |
| tyska | Razique: how can i change my ec2_port to 8443² | 13:03 |
| tyska | ? | 13:03 |
| Kiall | no problem .. ubuntu packages use the debian format .. hence the confusion ;) | 13:03 |
| tyska | i already changed the nova.conf file | 13:03 |
| tyska | but vidd said my database is toasted | 13:03 |
| Razique | tyska: what do u mean ? | 13:04 |
| *** sandywalsh_ has quit IRC | 13:04 | |
| Razique | ah | 13:04 |
| Razique | here is my keystone scheme | 13:04 |
| Kiall | tyska: i dont think the wrong port would toast your DB | 13:04 |
| Razique | that works | 13:04 |
| Kiall | Its not in the DB anywhere that I know of.. | 13:04 |
| tyska | Razique: ok, then i need to put my port to 8443 | 13:04 |
| tyska | Razique: because iptables are configurated to this port, nova-manage config list is showing this port | 13:05 |
| zykes- | what are you working on ? | 13:05 |
| tyska | Razique: and credentials are being generated to this port | 13:05 |
| tyska | but netstat -lnp | grep 8773 shows a python process listen, and 8443 does not | 13:05 |
| Kiall | tyska: and has the iptables rule been updated aswell? | 13:06 |
| Razique | tyska: vidd-away here http://fs05u.sendspace.com/processupload.html | 13:06 |
| zykes- | what are you helping with today Razique ? | 13:06 |
| tyska | An error occurred while processing the uploaded file. (10000) | 13:07 |
| n88 | Kiall: I also created a patch yesterday to fix an issue I was seeing attaching volumes to images on the compute node; Openstack was expecting the iSCSI drive to be mounted with filename: blah-blah-lun-0, but the iSCSI management software (tgt I believe on oneiric) was mounting it as: blah-blah-lun-1. I did patch that one. Do you want me to attempt to upload that one as well? I'm not sure whether that's an issue on all Ubuntu versions or just oneiric. | 13:07 |
| tyska | Razique: iptables rule was already saying port is 8443 | 13:08 |
| Guest86111 | quit | 13:08 |
| *** Guest86111 has quit IRC | 13:08 | |
| zykes- | tyska: what error are you having ? | 13:09 |
| tyska | zykes-: im trying to change my ec2_port to 8443 | 13:09 |
| tyska | zykes-: changed it on nova.con | 13:09 |
| tyska | zykes-: but netstat -lnp | grep 8773 retuns a process, and 8443 returns nothing | 13:09 |
| *** mrevell has joined #openstack | 13:09 | |
| Razique | tyska: ok,why are u trying to change the port ? | 13:10 |
| Razique | :D | 13:10 |
| zykes- | tyska: please don't bother changing ports on the stuff, it's raping your own brain to be exact | 13:10 |
| Razique | tyska: http://www.sendspace.com/file/uwu7on | 13:11 |
| zykes- | Razique: what version of ks you on ? | 13:11 |
| Razique | 8773 is a cool port to have ec2 listen on :) | 13:11 |
| Razique | zykes-: trunk | 13:11 |
| Kiall | n88: I'm running oneiric with nova-volume without issue.. | 13:11 |
| zykes- | ah ok | 13:11 |
| Razique | the one with updated scheme | 13:11 |
| Kiall | And I dont think I patched that ;) | 13:11 |
| Kiall | It must be included in the stable/diablo branch already | 13:11 |
| zykes- | Razique: yeah, keystone diablo is a pain in the ... | 13:12 |
| *** andyandy has joined #openstack | 13:13 | |
| Razique | zykes-: yup, I tried (u were here) | 13:13 |
| Razique | no success | 13:13 |
| *** tyska has quit IRC | 13:14 | |
| n88 | Kiall: Hmm…that's odd. I installed just using apt-get on oneiric yesterday and I had to fix that issue with iSCSI. I wonder if my nova.conf is causing that or something else…just bizarre. | 13:14 |
| *** rods has joined #openstack | 13:15 | |
| *** mrevell has quit IRC | 13:15 | |
| Kiall | the stock oneiric packages are outta date ;) | 13:15 |
| Kiall | they are based on the 2011.3 tag, not the stable/diablo branch | 13:15 |
| zykes- | Kiall: which ones ? | 13:15 |
| Kiall | all the ubuntu packages are based on the diablo release tag, none are based on the stable bugfix branches | 13:16 |
| n88 | Kiall: Ah ok…I'm getting confused. So there's the Ubuntu-hosted packages which are 2011.3; there's the packages you are providing, which is the 2011.4 stable branch? And then there's the master source which is not available in any packages and must be installed from git….am I sort of on the right track now? | 13:16 |
| Kiall | although, they do include some patches themselves, many of which made it into the stable bugfix branches | 13:16 |
| Kiall | n88: close :) the stock ubuntu packages are based on the 2011.3 tag, with some ubuntu specific patches. (these, and other patches, have been included in the stable/diablo branch) | 13:17 |
| *** mrevell has joined #openstack | 13:18 | |
| Kiall | my packages are based on the stable/diablo branches which are kinda 2011.3.1 | 13:18 |
| zykes- | Kiall: i hope redhat goes on the bandwagon for openstack so we can get OpenShift <3 | 13:18 |
| Kiall | and .. the git master branches are eventually going to be called 2012.1 | 13:18 |
| n88 | Kiall: Ah ok. This is all making sense to me now. So the stable/diablo branch must have that iSCSI issue resolved, which is included in your packages and why you didn't have to make the change; I used apt-get so got the old 2011.3 package from Ubu's repos and had to make the fix but shouldn't going forward on any other nodes if I use the PPA you directed me to.... | 13:19 |
| Kiall | yea.. I'm hoping ubuntu will update their packages soon, so I dont have to ;) But for the moment.. | 13:19 |
| n88 | By the way, this IRC channel is great; I've gotten some really good help and advice and it's nice being able to get insight into how components actually work. | 13:20 |
| Kiall | I'm stuck keeping mine fixed :) | 13:20 |
| Kiall | yea .. people often scoff at IRC and the nerds that use it .. but its surprisingly useful ;) | 13:20 |
| n88 | nerds? what nerds? we all know that the world is secretly ruled by people on IRC channels ;-) | 13:21 |
| zykes- | Razique: you run nova volume right ? | 13:21 |
| Kiall | n88: lol.. | 13:21 |
| Razique | zykes-: yup | 13:22 |
| *** jsavak has joined #openstack | 13:28 | |
| *** mmetheny has quit IRC | 13:28 | |
| *** mmetheny has joined #openstack | 13:28 | |
| *** ChrisAM has quit IRC | 13:32 | |
| Razique | n88: cool then | 13:33 |
| gnu111 | guys..need help with swift. getting Object PUT failed ... 503 Service Unavailable error when I try to upload a file. | 13:34 |
| gnu111 | the container is created but but files is not uploaded | 13:34 |
| *** huslage has joined #openstack | 13:35 | |
| gnu111 | proxy-server ERROR with Object server 172.29.100.12:6000/sda3 re: Expect: 100-continue on /AUTH_system/myfolder/file: Host unreachable | 13:36 |
| gnu111 | all the services are running and /srv/node has swift:swift permission. | 13:37 |
| *** ChrisAM1 has joined #openstack | 13:37 | |
| *** mrevell has quit IRC | 13:38 | |
| *** kieron has joined #openstack | 13:39 | |
| *** ldlework has joined #openstack | 13:40 | |
| *** dendrobates is now known as dendro-afk | 13:41 | |
| *** hggdh has quit IRC | 13:41 | |
| *** jedi4ever has joined #openstack | 13:42 | |
| *** uksysadmin has joined #openstack | 13:44 | |
| *** bcwaldon has joined #openstack | 13:45 | |
| uksysadmin | hello all | 13:46 |
| Razique | hey uksysadmin :) | 13:47 |
| Razique | 'sup ? | 13:47 |
| *** marrusl has quit IRC | 13:47 | |
| uksysadmin | how do I change the VLAN id used for nova? I've installed OpenStack packages from ubuntu and the default is 100 - can I restrospectively change this? | 13:47 |
| n88 | Kiall: Just to make sure…in the pastie you sent me for the export lines, should i just leave them as : export DEBFULLNAME="Name" | 13:47 |
| n88 | export DEBEMAIL="bla@bla.com" | 13:47 |
| n88 | or should I update those to something? | 13:47 |
| uksysadmin | Razique, I've given up on the auto floating for the time being... we're currently porting my dev env to our datacentre | 13:47 |
| *** mrevell has joined #openstack | 13:48 | |
| Razique | ok ok :) | 13:48 |
| uksysadmin | I'll return to it next week, so you don't get off that lightly | 13:50 |
| uksysadmin | ;-) | 13:50 |
| gnu111 | usksysadmin: I think the VLAN id is in the networks table. | 13:50 |
| *** lts has joined #openstack | 13:51 | |
| uksysadmin | I did see it in the networks table - just felt like something I shouldn't need to do directly in a table...? | 13:51 |
| Razique | uksysadmin: u could, I daily dealing with the DB directly | 13:52 |
| Razique | just make sure u keep the consistency accross tables | 13:52 |
| gnu111 | uksysadmin: did you already create the network with nova-manage? I think nova-manage network has a flag for vlan id. | 13:52 |
| *** mrevell has quit IRC | 13:52 | |
| uksysadmin | gnu111, there was a vlan_start flag but my servers still have vlan100 and br100 interfaces | 13:53 |
| uksysadmin | gnu111, I didn't re-do my create network | 13:54 |
| uksysadmin | I'll give that a go | 13:54 |
| NadeemMayeck | guys, should i install qemu or xen or kvm befor einstalling openstack? | 13:55 |
| gnu111 | uksysadmin: According to the doc (http://docs.openstack.org/cactus/openstack-compute/admin/content/configuring-vlan-networking.html) it seems you have to modify the DB. | 13:55 |
| uksysadmin | cheers gnu111 | 13:56 |
| uksysadmin | I'll add in a quick one-liner in my install script to update the table | 13:56 |
| gnu111 | uksysadmin: sounds good! | 13:57 |
| *** jwalcik has joined #openstack | 13:57 | |
| *** aurigus has joined #openstack | 13:58 | |
| *** aurigus has joined #openstack | 13:58 | |
| *** dolphm has joined #openstack | 13:58 | |
| gnu111 | NadeemMayeck: xen or kvm support will depend on the kernel. and I think you will need qemu. | 13:58 |
| NadeemMayeck | i just installed qemu, and enabled KVM in the kernel | 13:59 |
| NadeemMayeck | now goign to install rabbitmq-server | 13:59 |
| NadeemMayeck | is that all i need before installing the openstack tarballs? | 13:59 |
| gnu111 | NadeemMayeck: I installed using yum. so it took care of the dependencies. there are few python stuff you will need to install. | 14:00 |
| *** livemoon has left #openstack | 14:00 | |
| NadeemMayeck | i think i should have all these python stuffs by now | 14:00 |
| *** tryggvil has quit IRC | 14:00 | |
| NadeemMayeck | i am on genntooo | 14:00 |
| *** joesavak has joined #openstack | 14:00 | |
| gnu111 | you will need MySQL also. | 14:01 |
| *** guigui1 has quit IRC | 14:01 | |
| NadeemMayeck | got that as well. | 14:02 |
| *** uncleofthestick has quit IRC | 14:02 | |
| *** dolphm has quit IRC | 14:03 | |
| *** sandywalsh_ has joined #openstack | 14:03 | |
| *** jsavak has quit IRC | 14:03 | |
| *** dolphm has joined #openstack | 14:04 | |
| *** stevegjacobs has joined #openstack | 14:05 | |
| *** cdbs has quit IRC | 14:05 | |
| *** msivanes has quit IRC | 14:06 | |
| *** benrodrigue has joined #openstack | 14:07 | |
| *** mattray has joined #openstack | 14:08 | |
| *** lborda has joined #openstack | 14:08 | |
| *** rsampaio has quit IRC | 14:08 | |
| *** cdbs has joined #openstack | 14:08 | |
| *** msivanes has joined #openstack | 14:10 | |
| *** markvoelker has joined #openstack | 14:11 | |
| *** cereal_bars has joined #openstack | 14:12 | |
| *** PiotrSikora has quit IRC | 14:13 | |
| *** PiotrSikora has joined #openstack | 14:15 | |
| uvirtbot | New bug: #886139 in nova "Extraneous call to list all images in OSAPI Server Create" [Low,In progress] https://launchpad.net/bugs/886139 | 14:16 |
| NadeemMayeck | guys, do I need X to install/run openstack? | 14:18 |
| *** guigui has joined #openstack | 14:18 | |
| *** osier has quit IRC | 14:20 | |
| n88 | Kiall: Are you there? | 14:20 |
| *** benrodrigue has quit IRC | 14:21 | |
| lelin | NadeemMayeck, no you do not need it | 14:21 |
| *** benrodrigue has joined #openstack | 14:21 | |
| *** cp16net has joined #openstack | 14:22 | |
| *** sgd2 has joined #openstack | 14:28 | |
| *** cereal_b_ has joined #openstack | 14:28 | |
| *** sgd2 has quit IRC | 14:28 | |
| *** rnorwood has quit IRC | 14:29 | |
| *** cereal_bars has quit IRC | 14:30 | |
| *** kaigan_ has quit IRC | 14:34 | |
| *** brodrigue has joined #openstack | 14:34 | |
| *** benrodrigue has quit IRC | 14:36 | |
| *** NadeemMayeck has quit IRC | 14:37 | |
| *** dragondm has joined #openstack | 14:37 | |
| *** dendro-afk is now known as dendrobates | 14:38 | |
| *** dragondm has quit IRC | 14:41 | |
| *** dragondm has joined #openstack | 14:42 | |
| *** dnjaramba has quit IRC | 14:43 | |
| *** dnjaramba has joined #openstack | 14:43 | |
| *** j^2 has quit IRC | 14:49 | |
| *** antenagora has joined #openstack | 14:50 | |
| *** j^2 has joined #openstack | 14:50 | |
| *** guigui has quit IRC | 14:53 | |
| *** guigui has joined #openstack | 14:54 | |
| *** dubsquared has joined #openstack | 14:55 | |
| doude | Hi all, I try to set up a multi region cloud but i get an error when I try to start a instance. | 14:57 |
| *** gyee has joined #openstack | 14:57 | |
| doude | The parent zone send a select request to the child API zone and the child scheduler fails with 'AttributeError: select': http://paste.openstack.org/show/3076/ | 14:58 |
| *** sgg has joined #openstack | 14:59 | |
| doude | I check file 'nova/scheduler/multi.py', the method '__get_attr__' try to filter the request to identify which driver must be use | 14:59 |
| *** dgags has joined #openstack | 15:00 | |
| doude | but this method doesn't filter 'select' request and raises an AttributeError exception | 15:00 |
| doude | Can you help me ? | 15:00 |
| *** dtroyer has quit IRC | 15:01 | |
| *** hggdh has joined #openstack | 15:03 | |
| *** rnorwood has joined #openstack | 15:04 | |
| *** alperkanat has joined #openstack | 15:06 | |
| alperkanat | notmyname: do you have time to talk about the low benchmark results that i posted on twitter? | 15:06 |
| notmyname | alperkanat: ah! you are in IRC :-) | 15:07 |
| alperkanat | :) | 15:07 |
| alperkanat | i knew i'd find you here | 15:07 |
| notmyname | alperkanat: sure. what size of files were you using? (-s option) | 15:07 |
| *** blamar has quit IRC | 15:07 | |
| *** robix has joined #openstack | 15:07 | |
| alperkanat | notmyname: i didn't specified any options -- i'll try the same as yours | 15:08 |
| notmyname | ok | 15:08 |
| notmyname | I'm not sure what the default it | 15:08 |
| notmyname | is | 15:08 |
| notmyname | alperkanat: it defaults to 1 byte, so that shouldn't cause a significant difference (I was using zero byte) | 15:09 |
| alperkanat | notmyname: so i don't need to try again? (since i've deployed 1.2.x, bench doesn't delete its files) :) | 15:10 |
| notmyname | the concurrency may have a _little_ to do with it, but not that much. can you describe your testing setup? where are you running swift-bench? what are you running it against? | 15:10 |
| alperkanat | i run swift-bench on the proxy node | 15:10 |
| alperkanat | i have 1 proxy node and 3 storage nodes | 15:10 |
| notmyname | ok | 15:10 |
| alperkanat | this is production btw (unfortunetely) | 15:11 |
| alperkanat | only the storage node has internet access for security reasons and the storage nodes are NAT'd through it | 15:11 |
| notmyname | ok. then your numbers should be much better than mine | 15:11 |
| alperkanat | s/storage/proxy/ | 15:11 |
| alperkanat | these machines are in leaseweb specifially ordered for swift usage | 15:11 |
| alperkanat | none of them have raid | 15:12 |
| notmyname | ok, good | 15:12 |
| alperkanat | they have regular disks however | 15:12 |
| alperkanat | not SSD's | 15:12 |
| notmyname | np | 15:12 |
| alperkanat | i did everything i read on the docs and still this performance | 15:12 |
| alperkanat | i've had very hard time to transfer 5 TB's of data.. still struggling :) | 15:12 |
| alperkanat | i'd appreciate any help about this | 15:12 |
| alperkanat | i could provide more information if you ask | 15:13 |
| *** blamar has joined #openstack | 15:13 | |
| notmyname | a few things to try (just to make sure we're looking at basic numbers): create an empty account and use that. also check the proxy server for load and I/O (network) usage | 15:13 |
| alperkanat | notmyname: 16:16:53 up 60 days, 5:56, 2 users, load average: 0.72, 0.98, 0.99 | 15:14 |
| notmyname | ok, looks ok there, I think | 15:14 |
| alperkanat | any suggestion for network usage info? | 15:14 |
| alperkanat | i mean a cli tool to provide quick info | 15:14 |
| *** rnirmal has joined #openstack | 15:16 | |
| notmyname | alperkanat: ethstatus will give you a basic idea of what's going on | 15:17 |
| alperkanat | ok | 15:17 |
| alperkanat | notmyname: http://cl.ly/BY18 | 15:20 |
| alperkanat | swauth is on proxy as usual | 15:22 |
| alperkanat | 1 memcache | 15:22 |
| alperkanat | auth, container, object servers are on storage nodes | 15:22 |
| notmyname | alperkanat: the network I/O question is simply to make sure you aren't fighting against a full network interface | 15:22 |
| alperkanat | notmyname: full network interface? | 15:23 |
| notmyname | saturated. ie 100MB of traffic on a 100MB connection | 15:23 |
| *** tyska has joined #openstack | 15:24 | |
| *** code_franco has joined #openstack | 15:24 | |
| *** dtroyer has joined #openstack | 15:24 | |
| notmyname | alperkanat: try running swift-bench against an empty account | 15:25 |
| alperkanat | notmyname: i'm on it.. eth0 is Gb btw.. eth1 should also be gigabit and it's unlikely that we reach its boundries which can also be seen in ethstatus i guess | 15:25 |
| alperkanat | notmyname: results look the same so far | 15:26 |
| alperkanat | will paste when it ends | 15:26 |
| doude | I open this question https://answers.launchpad.net/nova/+question/177546 for my problem | 15:28 |
| alperkanat | notmyname: http://paste.openstack.org/show/3079/ | 15:28 |
| uvirtbot | New bug: #886169 in nova "OSAPI duplicates kernel_id/ramdisk_id lookup" [Low,In progress] https://launchpad.net/bugs/886169 | 15:28 |
| *** dnjaramba has quit IRC | 15:29 | |
| notmyname | alperkanat: are you running ssl direct to the proxy server? or are you terminating the ssl at a load balancer? | 15:29 |
| *** dnjaramba has joined #openstack | 15:29 | |
| *** rsampaio has joined #openstack | 15:29 | |
| alperkanat | proxy has SSL.. we have nginx in front of it to hide url's (tokens etc), logging etc.. terminating ssl at the load balancer | 15:30 |
| alperkanat | also for load balancing of course | 15:30 |
| notmyname | ok | 15:30 |
| notmyname | that's good. it's a bad idea to run ssl directly to the proxy servers | 15:31 |
| alperkanat | it's acting as a reverse proxy fyi | 15:31 |
| alperkanat | nginx and proxy speaks through ssl however.. i understand "terminating ssl" from the clients' point of view | 15:31 |
| notmyname | nginx isn't a very good thing to have in front of swift, though. nginx spools the body of PUTs to disk and therefore a client can easily swamp your nginx box (fill up a disk) | 15:32 |
| alperkanat | users never see SSL | 15:32 |
| alperkanat | well it was the best fit for us now.. i understand your concerns but can't think of any other easy setup as nginx | 15:32 |
| alperkanat | you guys have fine cdn setups :) | 15:33 |
| tyska | guys | 15:33 |
| tyska | i need to change the ec2_port | 15:33 |
| tyska | i already changed it in the nova.conf | 15:33 |
| alperkanat | so we're still clueless about the performance i guess notmyname ? | 15:34 |
| tyska | but when i run netstap -lnp | grep 8773, one process still is showed, and with grep 8443 nothing is showed | 15:34 |
| notmyname | I'm a little confused on what you said wrt ssl. you have: client --https--> LB (nginx) --http--> proxy | 15:34 |
| tyska | someone can help me? | 15:34 |
| alperkanat | notmyname: client -> http -> LB (nginx) -> https -> proxy | 15:34 |
| notmyname | ok, help me out then. is the LB -> proxy link untrusted and the client -> LB link is trusted? that seems odd to me. why use ssl at all then? | 15:35 |
| *** bradlis7 has joined #openstack | 15:36 | |
| *** Gordonz has joined #openstack | 15:36 | |
| bradlis7 | Hey all. I've installed devstack, and I can associate an IP address, but it's not forwarding to that IP. | 15:36 |
| alperkanat | notmyname: for future changes we might do but it's meaningless for this setup | 15:37 |
| bradlis7 | If I ssh to the address that I allocated (and then associated), it just takes me to the openstack server | 15:37 |
| alperkanat | notmyname: do you think it can reduce performance this huge? | 15:37 |
| bradlis7 | There's no rules for my floating IP's in iptables at all. | 15:38 |
| notmyname | alperkanat: actually, I've heard reports of ssl to the proxy server causing some big problems. seems that Python itself has some problems with it. something about keeping connections open forever when a client disconnects early. check how many open connections you have on your proxy server | 15:38 |
| alperkanat | notmyname: netstat? | 15:39 |
| notmyname | ya | 15:39 |
| tyska | ITS DONE! =) | 15:40 |
| tyska | there is a param called --ec2_listen_port in the nova.conf =) | 15:40 |
| *** cereal_b_ has quit IRC | 15:40 | |
| alperkanat | notmyname: so far 107 | 15:41 |
| alperkanat | sudo netstat -tapn | grep python | wc -l | 15:41 |
| alperkanat | notmyname: nginx: 609 | 15:41 |
| *** hggdh has quit IRC | 15:43 | |
| *** helfrez has quit IRC | 15:43 | |
| *** nerens has quit IRC | 15:45 | |
| notmyname | alperkanat: just for fun, could you start a new instance of proxy server on a different port and not use ssl. then use swift-bench to hit that proxy instance | 15:45 |
| *** cereal_bars has joined #openstack | 15:46 | |
| alperkanat | notmyname: ok will try.. i believe i should create another proxy.conf and run proxy with that conf, right? | 15:46 |
| Razique | is annegentle around ? | 15:46 |
| *** tryggvil_ has joined #openstack | 15:46 | |
| alperkanat | i was having hard time to understand settings like default_swift_cluster = local#https:........... since it wasn't well documented | 15:46 |
| *** andy-hk has joined #openstack | 15:48 | |
| *** nerens has joined #openstack | 15:49 | |
| bradlis7 | I found nova/tools/setup_iptables.sh... Do I have to create the rules myself? | 15:50 |
| *** masom has joined #openstack | 15:50 | |
| masom | Hello, is there a way to verify an image is really being untarred? Status is untarring but the host has no i/o activity or cpu usage. | 15:51 |
| uvirtbot | New bug: #886189 in nova "vCPU-params doesn't exist" [High,In progress] https://launchpad.net/bugs/886189 | 15:51 |
| *** jsh has left #openstack | 15:52 | |
| *** huslage has quit IRC | 15:52 | |
| notmyname | alperkanat: ya, a separate proxy config would be required to turn off ssl | 15:52 |
| notmyname | and change the port | 15:52 |
| *** dolphm has quit IRC | 15:53 | |
| notmyname | alperkanat: look at the sample config file in swauth to see how that config option works: https://github.com/gholt/swauth/blob/master/etc/proxy-server.conf-sample | 15:53 |
| *** dolphm has joined #openstack | 15:53 | |
| alperkanat | notmyname: i've changed the port, the server seems to run and listens the port but bench says socket.error: [Errno 111] ECONNREFUSED | 15:54 |
| notmyname | alperkanat: any firewalls blocking the new port? | 15:55 |
| alperkanat | there's no firewall afaik | 15:55 |
| tyska | Razique: WORKED OUT!!!! | 15:55 |
| uvirtbot | New bug: #885966 in nova "vnc to windows mouse not sync" [Undecided,New] https://launchpad.net/bugs/885966 | 15:55 |
| tyska | FINALLY! | 15:55 |
| tyska | =) | 15:55 |
| Razique | tyska: how did u do that ? | 15:55 |
| alperkanat | but it listens the local port (which should be fine) | 15:55 |
| *** huslage has joined #openstack | 15:55 | |
| tyska | Razique: changed the port from 8773 to 8443 | 15:56 |
| tyska | Razique: there is a param --ec2_listen_port in th enova.conf | 15:56 |
| alperkanat | hmm it's related with the listened interface i guess | 15:56 |
| tyska | Razique: now, i have some another problems | 15:57 |
| tyska | Razique: =) | 15:57 |
| tyska | Razique: but i will try to figure them out alone | 15:57 |
| Razique | ok, but don't hesitate to ask | 15:57 |
| Razique | maybe someone will have the answer | 15:57 |
| alperkanat | notmyname: is there a way to prevent it to daemonize? | 15:57 |
| *** lelin has quit IRC | 15:57 | |
| *** dolphm has quit IRC | 15:58 | |
| notmyname | alperkanat: try running it directly from the command line (ie not with swift-init) and give it the "-o" option | 15:58 |
| *** dolphm has joined #openstack | 15:58 | |
| alperkanat | -o ? | 15:59 |
| alperkanat | sudo swift-proxy-server proxy.conf -o ? :) | 15:59 |
| notmyname | ya | 15:59 |
| alperkanat | notmyname: swift-proxy-server: error: no such option: -o | 16:00 |
| notmyname | alperkanat: ah. "-v" logs to console | 16:00 |
| notmyname | sorry | 16:00 |
| *** nikolaij has quit IRC | 16:00 | |
| notmyname | alperkanat: unfortunately I have to go to a meeting. I'll be back on later today | 16:00 |
| *** dolphm has quit IRC | 16:00 | |
| alperkanat | ok thanks | 16:01 |
| *** dolphm has joined #openstack | 16:01 | |
| *** obino has quit IRC | 16:01 | |
| *** huslage has quit IRC | 16:02 | |
| *** oubiwann has joined #openstack | 16:03 | |
| *** mihgen has joined #openstack | 16:03 | |
| *** uksysadmin has quit IRC | 16:03 | |
| mihgen | Hi all! Can you guys point me to the devstack revision that is proved to work? | 16:03 |
| *** cereal_bars has quit IRC | 16:05 | |
| *** dolphm has quit IRC | 16:06 | |
| *** hezekiah_ has joined #openstack | 16:06 | |
| *** antenagora has quit IRC | 16:06 | |
| tyska | is there difference between sudo -s and sudo su?? | 16:08 |
| *** cdbs has quit IRC | 16:08 | |
| *** jsavak has joined #openstack | 16:10 | |
| *** freeflyi1g has joined #openstack | 16:10 | |
| *** javiF has joined #openstack | 16:11 | |
| *** cdbs has joined #openstack | 16:12 | |
| tyska | Razique: now im having problem to access instances there are in the server2 | 16:12 |
| tyska | Razique: i think it is a iptables problem | 16:12 |
| tyska | Razique: but i cant find the rrer | 16:12 |
| tyska | error | 16:12 |
| *** joesavak has quit IRC | 16:13 | |
| *** freeflying has quit IRC | 16:14 | |
| *** reidrac has left #openstack | 16:14 | |
| *** guigui has quit IRC | 16:15 | |
| *** vdo has joined #openstack | 16:15 | |
| tyska | Guys, im having problem to access a instance in a different compute node from the others controllers | 16:17 |
| tyska | see it here: http://pastebin.com/8gGGi2Hs | 16:17 |
| tyska | Can anyone help me? | 16:17 |
| Kiall | tyska: what network manager are you using? | 16:18 |
| gnu111 | tyska: can you try euca-get-console-output of that instance. to make sure if it is booting up properly. | 16:18 |
| Kiall | yup, and pinging its private IP from the second server.. just in case | 16:18 |
| bradlis7 | Is the dashboard able to update iptables correctly? At http://docs.openstack.org/cactus/openstack-compute/admin/content/associating-public-ip.html, it says to "Ensure the NAT rules have been added." They have not been added when I use Dashboard. | 16:19 |
| masom | nova-api is stuck at 100% cpu right after registering an image. Doing any query using the API (euca*) locks up. Anyone would know what is happening? (The log files are not showing anything wrong) | 16:19 |
| Kiall | bradlis7: yes, it works for me | 16:20 |
| *** Gordonz has quit IRC | 16:20 | |
| bradlis7 | Kiall: I installed it devstack on my machine. Did you simply allocate, then associate the IP? | 16:20 |
| tyska | Kiall: nova-network | 16:20 |
| Kiall | yea, thats all thats needed | 16:21 |
| tyska | gnu111: good clue: 2011-11-04 16:21:06,022 - DataSourceEc2.py[WARNING]: 16:21:06 [77/100]: url error [[Errno 101] Network is unreachable] | 16:21 |
| tyska | gnu111: but why??? | 16:21 |
| bradlis7 | hm, I don't know what the problem is then... | 16:21 |
| Kiall | tyska, thats actually not the issue, at least. I would imagine that is caused by the issue... | 16:22 |
| tyska | Kiall: yeah, i think so too | 16:22 |
| Kiall | tyska: what network manager are you using? vlan? flatdhcp? | 16:23 |
| *** mihgen has quit IRC | 16:24 | |
| *** Gordonz has joined #openstack | 16:25 | |
| *** mihgen has joined #openstack | 16:25 | |
| *** qazwsx has joined #openstack | 16:25 | |
| *** lborda has quit IRC | 16:27 | |
| *** Hallje has joined #openstack | 16:27 | |
| *** df1 has joined #openstack | 16:27 | |
| *** dolphm has joined #openstack | 16:30 | |
| *** maplebed has joined #openstack | 16:33 | |
| *** nerens has quit IRC | 16:34 | |
| *** qazwsx|2 has joined #openstack | 16:35 | |
| *** DavorC has joined #openstack | 16:35 | |
| *** dobber has quit IRC | 16:36 | |
| *** qazwsx has quit IRC | 16:36 | |
| *** chris_ has joined #openstack | 16:36 | |
| *** dolphm has quit IRC | 16:36 | |
| *** chris_ is now known as Guest91833 | 16:37 | |
| *** dolphm has joined #openstack | 16:37 | |
| Guest91833 | Hi, does anyone know why I am seeing global name 'ec2_utils' is not defined in the nova-api.log when trying to upload an image via euca? | 16:37 |
| DavorC | Is anyone here running Diablo release in VPN network? | 16:37 |
| Kiall | DavorC: I've using diablo+VLAN networking.. | 16:40 |
| Kiall | I'm* | 16:40 |
| DavorC | Kiall: on Ubuntu? | 16:40 |
| Kiall | yea | 16:40 |
| tyska | Kiall: im using vlan | 16:40 |
| Kiall | tyska: then, chances are you issue is being caused by the VLAN's not being bridged correctly. | 16:41 |
| *** jdurgin has joined #openstack | 16:41 | |
| Kiall | Or, If you are using a chep-o home router/switch instead of something decent, it could be stripping off the VLAN tags .. | 16:41 |
| *** dolphm has quit IRC | 16:42 | |
| DavorC | Kiall: what should the correct bridging look like? | 16:42 |
| Kiall | (I've seen that before, and someone else, vishy i think, had the same issue) | 16:42 |
| tyska | Kiall: im connection the server1 directly to server2 | 16:42 |
| tyska | connecting | 16:42 |
| tyska | there is no router switch | 16:42 |
| Kiall | okay, then the bridging is likely setup wrong ;) | 16:42 |
| *** dolphm has joined #openstack | 16:43 | |
| Kiall | DavorC: I dont actually have access to my install to look right now.. but standard bridging... | 16:43 |
| tyska | Kiall: i'm with some kind of problem in the compute node on server2, see this log message http://pastebin.com/hCv2GHQT | 16:43 |
| *** Razique has quit IRC | 16:44 | |
| Kiall | the last line is telling, "ConnectTimeout", but I don't know the OS code (or Python) welll enough to tell what it failed to connect to.. | 16:44 |
| Kiall | my guess would be MySQL | 16:44 |
| Kiall | (a guess based on the first line..) | 16:45 |
| *** dolphm has quit IRC | 16:45 | |
| DavorC | tyska: is your compute node actually joining the cluster? (run 'euca-describe-availability-zones verbose') | 16:45 |
| *** hezekiah_ has left #openstack | 16:45 | |
| *** dolphm has joined #openstack | 16:46 | |
| * Kiall wonders why everyone uses the `euca-*` commands rather than the `nova` and `nova-manage` commands.. | 16:46 | |
| DavorC | Kiall: did you do anything about that vlan that libvirt sets up on 192.168.122.1? | 16:46 |
| Kiall | I find the nova commands much easier to use / remember | 16:47 |
| DavorC | Kiall: it would help if nova and nova-manage were better documented. | 16:47 |
| Kiall | DavorC: I just left it in place :) | 16:47 |
| DavorC | Kiall: I ask because I'm seeing the following in /var/log/syslog: 'dnsmasq-dhcp[6277]: DHCP packet received on br150 which has no address' | 16:48 |
| *** dolphm_ has joined #openstack | 16:48 | |
| tyska | DavorC: yeah, everything looks fine in this sense | 16:49 |
| Kiall | that doesnt seem related, sinec the one libvirt sets up is vibr0 or something like that | 16:49 |
| DavorC | This was added by 'dnsmasq' that's serving 192.168.122.0/8 | 16:49 |
| Kiall | oh .. yea that.. | 16:50 |
| DavorC | Kiall: that's what I thought too, but based on the process that added that log line, it looked like libvirt also messes with br150. | 16:50 |
| *** dolphm has quit IRC | 16:50 | |
| Kiall | edit /etc/default/dnsmasq and disable it .. | 16:50 |
| Kiall | then `killall dnsmasq && service nova-network` restart | 16:50 |
| uvirtbot | New bug: #886224 in nova "Compute Manager makes too many calls to Glance on instance create" [Low,In progress] https://launchpad.net/bugs/886224 | 16:51 |
| Kiall | dnsmasq gets installed and by default, listens on all interfaces.. | 16:51 |
| Kiall | nova starts an instance per interface, and that global one causes issues | 16:51 |
| *** obino has joined #openstack | 16:52 | |
| *** swill has joined #openstack | 16:54 | |
| *** stevegjacobs has quit IRC | 16:54 | |
| DavorC | Kiall: OK, I'll try that. Except that this is on the compute node, which doesn't run nova-network... | 16:55 |
| Kiall | same fix, just dont restart nova-network ;) | 16:56 |
| *** obino has joined #openstack | 16:56 | |
| *** dolphm_ has quit IRC | 16:58 | |
| *** dysinger has joined #openstack | 16:58 | |
| *** dolphm has joined #openstack | 16:58 | |
| DavorC | So you said the way to get rid of that global dnsmasq process is by editing /etc/default/dnsmasq? I don't have that file... | 16:58 |
| *** nacx has quit IRC | 16:59 | |
| Kiall | humm, what version of ubuntu are you running? or something else entirely?] | 17:01 |
| Kiall | never ubuntu versions have this: http://paste.ubuntu.com/728326/ | 17:01 |
| Kiall | with the ENABLED= option | 17:01 |
| *** pixelbeat has quit IRC | 17:01 | |
| *** jtanner has quit IRC | 17:01 | |
| Kiall | you can use update-rc.d to stop it starting aswell.. | 17:01 |
| Kiall | either will do the tick | 17:02 |
| *** dolphm has quit IRC | 17:02 | |
| Kiall | tyska: pastebin the output of `brctl show` maybe something will be obvious from that.. | 17:03 |
| Kiall | unless you already fixed it issue that is ;) | 17:03 |
| *** rich_ has quit IRC | 17:04 | |
| *** andy-hk has quit IRC | 17:05 | |
| *** tyska has quit IRC | 17:05 | |
| *** vdo has quit IRC | 17:06 | |
| *** lborda has joined #openstack | 17:06 | |
| *** tyska has joined #openstack | 17:07 | |
| tyska | Kiall: http://pastebin.com/E7ALds7D | 17:07 |
| tyska | the problem remains | 17:07 |
| Kiall | yea - the bridging looks wrong | 17:08 |
| Kiall | try `brctl addif br100 <eth whatever>` on both servers | 17:08 |
| *** dolphm has joined #openstack | 17:08 | |
| *** dubsquared has quit IRC | 17:09 | |
| *** katkee_ has quit IRC | 17:09 | |
| tyska | what means eth whatever? | 17:09 |
| Kiall | <eth whatever> being the ethernet adaptor you want the VLANs to go over.. | 17:09 |
| tyska | and looks wrong why? | 17:09 |
| *** lorin11 has joined #openstack | 17:09 | |
| Kiall | because they havent been bridged to a physical interface | 17:09 |
| Kiall | they just bridge a bunch of virtual interfaces | 17:10 |
| *** mihgen has quit IRC | 17:10 | |
| tyska | hmm | 17:10 |
| tyska | understand | 17:10 |
| DavorC | Kiall: Ubuntu 10.04.3 LTS | 17:11 |
| Kiall | DavorC: I guess it has been changed since then.. | 17:12 |
| *** lorin11 has quit IRC | 17:12 | |
| *** dendrobates is now known as dendro-afk | 17:12 | |
| *** lorin11 has joined #openstack | 17:12 | |
| *** lorin1 has quit IRC | 17:12 | |
| *** Hakon|mbp has quit IRC | 17:13 | |
| Kiall | tyska: did nova create that br100 interface BTW, or did you create it yourself? | 17:13 |
| tyska | Kiall: connection between server1 and server2 stopped work | 17:13 |
| tyska | Kiall: nova | 17:13 |
| tyska | Kiall: after i run the brctl addif | 17:13 |
| *** Ryan_Lane has joined #openstack | 17:14 | |
| tyska | Kiall: delif solved the problem | 17:14 |
| *** jtanner has joined #openstack | 17:15 | |
| Kiall | tyska: yea, if that interface is also for normal traffic (ie not just for VLAN traffic), you need to move the ip from <eth whatever> to br100 | 17:15 |
| tyska | but im agree there is some problem, in some way the bridge needs to be related with the physical interface | 17:15 |
| Kiall | tyska: did nova create that br100 interface BTW, or did you create it yourself? | 17:15 |
| tyska | Kiall: nova created it | 17:16 |
| tyska | Kiall: this if has the ip adress in the subnet 192.168.1.0/24 | 17:16 |
| *** bengrue has joined #openstack | 17:16 | |
| *** cdub has joined #openstack | 17:16 | |
| tyska | br100 has the subnet 172.16.50.0/24 | 17:16 |
| tyska | public iface of server1 is in the subnet 10.0.0.0/24 | 17:17 |
| tyska | private, eth1, 192.168.1.0/24 | 17:17 |
| tyska | server2 is using just eth0, 192.168.1.0/24 | 17:17 |
| tyska | and br100 is the private ip of the instances, 172.16.50.0/24 | 17:18 |
| *** cereal_bars has joined #openstack | 17:18 | |
| tyska | should i put the eth1 iface (on server1) and eth0 iface (on server2) in the subnet 172.16.50.0/24 ?? | 17:19 |
| *** cereal_bars has joined #openstack | 17:19 | |
| *** javiF has quit IRC | 17:19 | |
| Kiall | No - thats not going to work very well ;) | 17:19 |
| tyska | Kiall: yeah, i think so too | 17:19 |
| Kiall | I've only done the setup with 4x network port servers so ... 2x = internet + nova communication and 2x dedicated for VLAN traffic.. | 17:19 |
| Kiall | I'm not sure how you would get nova to work right with only 1 nic.. I know its possible, but not what needs to be done | 17:20 |
| tyska | but those one for vlan traffic are physically connected where? | 17:20 |
| tyska | probably a swicth | 17:20 |
| Kiall | sure, thats not important though.. | 17:20 |
| tyska | but Kiall my server1 has 2 nics | 17:21 |
| tyska | and server2 has 2 nics too, but im only using one because it's all i need | 17:21 |
| tyska | i think | 17:21 |
| Kiall | the important part was, that those NICs dont handle anything other than the VLANs .. so when they go into the bridge, its not an issue.. | 17:21 |
| tyska | server2 is connected straight to the server1 through a crossover cable | 17:21 |
| Kiall | again - not the issue | 17:21 |
| *** jog0 has joined #openstack | 17:22 | |
| Kiall | With my setup, eth3 and eth4 (well bond1 since they are teamed) do no have IPs. | 17:22 |
| Kiall | so when they go into the bridge, no communication is cut | 17:22 |
| tyska | Kiall: but your compute-node gets Internet connection from where? direct connection? | 17:23 |
| Kiall | from bond0 (eth1 and eth2) | 17:23 |
| tyska | yeah, but eth1 and eth2 provide direct internet connection? | 17:24 |
| tyska | you reach your compute-nodes without pass through the nova-manage controller? | 17:24 |
| Kiall | tyska: it doesnt make a difference ... | 17:25 |
| *** cdub has quit IRC | 17:25 | |
| tyska | Kiall: my question is, on a external network, you can connect directly to a compute node? | 17:25 |
| Kiall | Both servers have bond0, which provides internet access, and this doubles as the interface nova uses to talk to other servers... | 17:25 |
| *** lorin11 has quit IRC | 17:26 | |
| Kiall | Then, both servers have bond1, which only passes VLAN traffic for the instances | 17:26 |
| *** Hakon|mbp has joined #openstack | 17:26 | |
| DavorC | Kiall: I thought VLAN is doable even on machines with only one or two NICs? | 17:26 |
| tyska | Kiall: can you access your compute-node directly from external networks or not? | 17:26 |
| *** df1 has quit IRC | 17:26 | |
| Kiall | DavorC: it is (i think), I'm simply explaining my setup, and why I dont know the fix... | 17:27 |
| Kiall | tyska: all nodes can be accessed directly | 17:27 |
| WormMan | we're running VLAN on one NIC | 17:27 |
| tyska | Kiall: hmm, this is not desirable to me, i wanna just 1 'front controller' | 17:27 |
| Kiall | aha - WormMan might know how to get that going... | 17:27 |
| Kiall | tyska: if you only have 1 nova-network on S1.. you still only have 1 FC... | 17:28 |
| tyska | and i wanna isolate all other servers, unless they are accessed by the front machine | 17:28 |
| *** nerens has joined #openstack | 17:28 | |
| Kiall | iptables ;) | 17:28 |
| Kiall | But .. anyway .. its doable.. I just dont know how, since nova is creating/editing/etc the interfaces | 17:28 |
| gnu111 | I have reconfigured swift twice but still getting this error? Any idea: proxy-server ERROR with Object server 172.29.200.44:6000/sda3 re: Expect: 100-continue on /AUTH_system/myfiles/zero: Host unreachable | 17:28 |
| tyska | Kiall: thanks anyway ;) | 17:28 |
| *** cereal_b_ has joined #openstack | 17:29 | |
| WormMan | the trick I've found is to make sure whatever you name the interfaces match what nova thinks so it doesn't change them | 17:29 |
| *** DavorC has quit IRC | 17:29 | |
| WormMan | we use VLAN, so our vlans are preconfigured in interfaces with a vlan400 name, and then the bridge is brought up with the br400 name and everything pre-assigned | 17:30 |
| gnu111 | noticed this also: account-replicator ERROR reading HTTP response from {'zone': 4, 'weight': 100.0, 'ip': | 17:30 |
| *** cereal_bars has quit IRC | 17:31 | |
| gnu111 | notmyname: ping | 17:32 |
| *** thingee has joined #openstack | 17:38 | |
| *** jog0 has quit IRC | 17:40 | |
| *** jog0 has joined #openstack | 17:40 | |
| n88 | Does anybody know in the glance.images table what the units are for min_disk and min_ram? | 17:41 |
| *** lorin1 has joined #openstack | 17:43 | |
| *** vladimir3p has joined #openstack | 17:44 | |
| *** j^2 has quit IRC | 17:46 | |
| *** j^2 has joined #openstack | 17:47 | |
| *** Razique has joined #openstack | 17:48 | |
| Razique | Hi guys | 17:48 |
| Razique | it's look like the network is unstable with Diablo | 17:48 |
| Razique | instances are unreacheable despite a good network settings | 17:49 |
| *** gnu111 has quit IRC | 17:49 | |
| bradlis7 | are there instructions on how to configure the network for floating interfaces? | 17:50 |
| vishy | n88: force_dhcp_release isn't working? | 17:50 |
| vishy | Razique: when you say diablo, do you mean diablo/stable? | 17:50 |
| Razique | vishy: Hi, yup | 17:50 |
| Razique | I figured :) | 17:50 |
| Razique | explanations | 17:50 |
| vishy | Razique: I haven't had any issues yet | 17:50 |
| Razique | how are you vishy | 17:50 |
| Razique | I have controller eth0 and eth1 | 17:51 |
| Razique | and one node eth0 only | 17:51 |
| Razique | nova.conf on controller : --public_interface=eth0 | 17:51 |
| Razique | --vlan_interface=eth1 | 17:51 |
| *** df1 has joined #openstack | 17:51 | |
| Razique | nova.conf on node : --public_interface=eth0 | 17:51 |
| Razique | --vlan_interface=eth0 | 17:51 |
| Razique | with the Cactus, works fine | 17:51 |
| Razique | with Diablo doesn't | 17:51 |
| Razique | since nova-network is trying to create the bridge with eth1 on the node | 17:52 |
| *** krow has joined #openstack | 17:52 | |
| Razique | (defined under "bridge_interface" under the database) | 17:52 |
| Razique | well it's not working better in fact | 17:53 |
| Razique | :/ | 17:53 |
| tyska | Razique: !!!! | 17:54 |
| n88 | vishy: it doesn't appear to be; I also found a few edits I had to make in manager.py to get floating IPs to be released | 17:54 |
| tyska | Razique: i need your help | 17:54 |
| n88 | vishy: even after applying these patches: https://review.openstack.org/#change,628 | 17:55 |
| vishy | Razique: correct that doesn't work | 17:55 |
| vishy | there is a bug about it | 17:55 |
| vishy | with a plan | 17:55 |
| vishy | just no one has implemented it | 17:55 |
| tyska | i cant reach the instances running on c3compute1 (server2) | 17:55 |
| Razique | tyska: me neither under diablo :/ | 17:55 |
| Razique | working great on my lab but doesn't in pre-production | 17:56 |
| Razique | arf…. /me thinks that the upgrade to diablo won't happen Monday then | 17:56 |
| vishy | Razique: https://bugs.launchpad.net/nova/+bug/833426 | 17:56 |
| *** krow has quit IRC | 17:57 | |
| vishy | vlan interface has to be the same on all nodes at the moment | 17:57 |
| *** n88 has quit IRC | 17:57 | |
| bradlis7 | Ah, I was trying to route to the VM server, but it is bridged I guess... | 17:57 |
| *** n8 has joined #openstack | 17:57 | |
| vishy | n88: you mean auto-released? | 17:57 |
| *** n8 is now known as Guest95550 | 17:58 | |
| vishy | n88: I haven't had any trouble manually releasing ips with diablo | 17:58 |
| *** nati2 has joined #openstack | 17:58 | |
| Razique | vishy: ok thanks, so i'll have to wait, since I've a running cactus cloud that I can't upgrade without that setting | 17:58 |
| Razique | too sad | 17:58 |
| Guest95550 | vishy: I was trying to get auto-public-ip assignment working; I used the patches in the link I pasted; it mostly worked. I was then getting two errors (I can pastebin them if you would like) and editied manager.py to fix them | 17:58 |
| *** jtanner has quit IRC | 17:59 | |
| vishy | Razique: you could try to do the patch, shouldn't be too hard | 17:59 |
| bradlis7 | ugh, nevermind... | 17:59 |
| *** doorlock has joined #openstack | 17:59 | |
| vishy | would be great to see the edits | 17:59 |
| vishy | and the errors | 17:59 |
| vishy | in a bug report! | 17:59 |
| Guest95550 | vishy: as for private IPs…I've got force_dhcp_release set in my config, but it's still giving me a new private IP until the timeout | 17:59 |
| vishy | do you see it trying to release ips? | 18:00 |
| Guest95550 | vishy: I guess I was asking for it, huh? =P | 18:00 |
| Razique | vishy: so I need to remove the entry into the DB for the br_if ? | 18:00 |
| vishy | you should see a log message in nova-network about trying to release the ip | 18:00 |
| Razique | not sure I fully understand Trey suggestion | 18:01 |
| vishy | Razique: no it just needs to use the flag instead of the db entry if it is set | 18:01 |
| vishy | flag should override db | 18:01 |
| Razique | ah ok | 18:01 |
| Razique | well it doesn't work then :D | 18:01 |
| *** sgd2 has joined #openstack | 18:01 | |
| Razique | since I use to have flags into nova.conf | 18:01 |
| Razique | ooops misread sorry | 18:01 |
| vishy | Razique: what doesn't work? | 18:01 |
| vishy | Razique: that is the plan for fixing | 18:01 |
| Razique | I need to add the flag --bridge_interface=eth1 into the node | 18:02 |
| bradlis7 | How do I set up a network as far as floating IPs go? Do I route to the OpenStack bridge address? | 18:02 |
| Guest95550 | vishy: I will check the logs for private_ip released; what I'm seeing in the db, is that instance_id and virtual_interface_id as well as the allocated columns all continue to have vlaues | 18:03 |
| Guest95550 | vishy: I'm assuming the code probably checks for the next free available one of those and something is not clearing or resetting those values during instance termination and with the dhcp_release flag set to true | 18:03 |
| vishy | if force is true | 18:04 |
| vishy | it calls an external utility | 18:04 |
| vishy | that sends a dhcprelease packet to dnsmasq | 18:04 |
| *** Guest91833 has quit IRC | 18:04 | |
| vishy | then dnsmasq calls nova-dhcpbridge which puts a message in the queue for nova-network | 18:04 |
| vishy | then nova-network releases the ip | 18:04 |
| Guest95550 | well that's good to know how it works | 18:05 |
| *** Guest95550 is now known as n88 | 18:05 | |
| *** alperkanat has quit IRC | 18:05 | |
| uvirtbot | New bug: #886260 in openstack-qa "_inject_key_into_fs should check fs is None or not" [Medium,Confirmed] https://launchpad.net/bugs/886260 | 18:06 |
| vishy | so you should see a command in nova-network: 595- utils.execute('dhcp_release', dev, address, mac_address, run_as_root=True) | 18:06 |
| *** tlocalhos has quit IRC | 18:06 | |
| vishy | calling dhcp_release | 18:06 |
| vishy | then you should see a message that the address was released | 18:06 |
| n88 | vishy: ok I'll check the logs….and get back to you; have to step out for about an hour | 18:07 |
| *** tlocalhos has joined #openstack | 18:08 | |
| bradlis7 | Is there a good tutorial on getting openstack with Dashboard set up on a single server? | 18:09 |
| *** dendro-afk is now known as dendrobates | 18:09 | |
| *** bradlis7 has quit IRC | 18:11 | |
| *** cereal_bars has joined #openstack | 18:11 | |
| *** hggdh has joined #openstack | 18:11 | |
| *** bradlis7 has joined #openstack | 18:12 | |
| *** nati2_ has joined #openstack | 18:12 | |
| *** doorlock has quit IRC | 18:13 | |
| *** nati2 has quit IRC | 18:13 | |
| Razique | vishy: the fix seems to work :) | 18:13 |
| Razique | How to make sure bridges have been created on the right interfaces ? | 18:15 |
| *** cereal_b_ has quit IRC | 18:15 | |
| *** sgd2 has quit IRC | 18:15 | |
| uvirtbot | New bug: #886266 in openstack-qa "virt.libvirt.netutils.py should handle UnboundLocalError" [Medium,Confirmed] https://launchpad.net/bugs/886266 | 18:15 |
| Razique | mm nope doesn't work | 18:15 |
| Razique | vlan100 | 100 | eth0 | 18:16 |
| Razique | should have been vlan100 | 100 | eth1 | 18:16 |
| vishy | Razique :( | 18:16 |
| Razique | vishy: here is what I did : controller --public_interface=eth0 | 18:16 |
| Razique | --vlan_interface=eth1 | 18:16 |
| Razique | --bridge_interface=eth1 | 18:16 |
| Razique | node : --public_interface=eth0 | 18:16 |
| Razique | --vlan_interface=eth0 | 18:16 |
| Razique | --bridge_interface=eth0 | 18:16 |
| Razique | I don't have anymore the errors | 18:17 |
| Razique | but the bridge is created on the same interface | 18:17 |
| uvirtbot | New bug: #886263 in openstack-qa "IndexError: pop index out of range on replace_uuid_with_id" [High,Confirmed] https://launchpad.net/bugs/886263 | 18:17 |
| vishy | Razique: did you make a patch? | 18:18 |
| vishy | bradlis7: devstack.org | 18:18 |
| Razique | vishy: I don't know how to make a patch for that bug :( | 18:18 |
| vishy | well it isn't going to work without a patch | 18:19 |
| vishy | that is why it is a bug | 18:19 |
| bradlis7 | vishy: That's what I used, and floating ip's isn't working right | 18:19 |
| vishy | define "right" | 18:19 |
| vishy | unless you give it a range that your router can route to, you will only be able to access them from the host | 18:19 |
| *** reed has joined #openstack | 18:20 | |
| Razique | vishy: but is a patch has been written for that bug ? | 18:20 |
| bradlis7 | I added public IP's that are unused | 18:20 |
| bradlis7 | and I routed my firewall to the openstack box DMZ IP address | 18:20 |
| uvirtbot | New bug: #886268 in openstack-qa "filepath should not be splited by '/'" [Undecided,New] https://launchpad.net/bugs/886268 | 18:21 |
| uvirtbot | New bug: #886272 in openstack-qa "setup_container produces UnboundLocalError " [Medium,Confirmed] https://launchpad.net/bugs/886272 | 18:21 |
| bradlis7 | if I ssh, it goes to the openstack box, not the virtual box | 18:21 |
| *** huslage has joined #openstack | 18:21 | |
| *** krow has joined #openstack | 18:21 | |
| bradlis7 | ...I mean virtual machine | 18:21 |
| *** sgg has quit IRC | 18:22 | |
| *** krow has quit IRC | 18:22 | |
| *** andy-hk has joined #openstack | 18:22 | |
| vishy | bradlis7: interesting. Can you check if ip_forward is on? | 18:23 |
| Razique | vishy: ok in fact after I added the line into my conf's | 18:23 |
| Razique | I can reach the intances | 18:23 |
| Razique | instances* | 18:23 |
| vishy | bradlis7: floating ips set up iptables rules, and that is the only reason that i can think they may not be getting hit. | 18:24 |
| bradlis7 | vishy: `cat /proc/sys/net/ipv4/ip_forward` returns 1 | 18:24 |
| bradlis7 | yeah, I agree. I looked at iptables, and didn't see that address in their | 18:25 |
| bradlis7 | *there | 18:25 |
| vishy | how did you associate the address? | 18:25 |
| vishy | the rule doesn't show up in iptables -t nat ? | 18:25 |
| bradlis7 | through the dashboard (allocate, then associate) | 18:25 |
| vishy | something is wrong in that case | 18:25 |
| bradlis7 | right | 18:25 |
| vishy | did you reboot in between or anything? | 18:25 |
| bradlis7 | I have not rebooted at all | 18:26 |
| *** dendrobates is now known as dendro-afk | 18:27 | |
| bradlis7 | http://paste.openstack.org/show/3090/ | 18:28 |
| Razique | vishy: ok I'm off, thanks for your hint, I'll add a note into the diablo doc then | 18:28 |
| Razique | bye all :) | 18:28 |
| *** jtanner has joined #openstack | 18:28 | |
| Razique | $ | 18:28 |
| *** Razique has left #openstack | 18:28 | |
| bradlis7 | I have no idea where configuration files or log files are on devstack... | 18:29 |
| vishy | bradlis7: do you see any errors in nova-network ? | 18:29 |
| vishy | bradlis7: they all log into screen | 18:29 |
| bradlis7 | ah, let me remote kvm... | 18:30 |
| *** sannes has quit IRC | 18:31 | |
| uvirtbot | New bug: #886275 in openstack-qa "destroy_container should catch exception from utils.execute" [Undecided,Confirmed] https://launchpad.net/bugs/886275 | 18:31 |
| bradlis7 | I don't see anything when I go to screen 7 having to do with nova... | 18:32 |
| *** doorlock has joined #openstack | 18:34 | |
| bradlis7 | oh, it's in apache's error log... | 18:34 |
| *** zaoul1 has joined #openstack | 18:35 | |
| *** krow has joined #openstack | 18:37 | |
| bradlis7 | vishy: Here is the log. I don't see anything fishy: http://paste.openstack.org/show/3091/ | 18:38 |
| vishy | that is the dash log | 18:38 |
| vishy | you actually want the n-net log | 18:38 |
| uvirtbot | New bug: #886281 in nova "zone_manager uses LOG which isn't defined" [Undecided,New] https://launchpad.net/bugs/886281 | 18:41 |
| *** mgoldmann has quit IRC | 18:41 | |
| zaoul1 | what is the recommended storage infrastructure for multiple hypervisors? Just a old fashion DAS/Iscsi array? Does openstack fence properly in this situation? | 18:42 |
| bradlis7 | Is there a way to edit a paste? | 18:44 |
| *** cdub has joined #openstack | 18:45 | |
| *** sannes has joined #openstack | 18:46 | |
| *** stuntmachine has joined #openstack | 18:47 | |
| *** gnu111 has joined #openstack | 18:48 | |
| *** mdomsch has joined #openstack | 18:48 | |
| *** llang629 has joined #openstack | 18:49 | |
| zaoul1 | why is creating images like brain surgery? extract and create compressed files, extract the initrd... overall I have little faith in whatever nasa supports. | 18:51 |
| uvirtbot | New bug: #886289 in nova "schedulers should set instance to error state" [Undecided,New] https://launchpad.net/bugs/886289 | 18:51 |
| *** llang629 has left #openstack | 18:52 | |
| zaoul1 | Especially with kvm, where creating a image block device.. booting an iso is so straight forward. Then just copy the image to duplicate | 18:54 |
| zaoul1 | hey the future! lets add 20 steps to VM creation! | 18:54 |
| bradlis7 | zaoul1: I agree that it's a hassle. They need to let you boot from an install disk | 18:55 |
| *** nati2 has joined #openstack | 18:55 | |
| *** hggdh has quit IRC | 18:56 | |
| bradlis7 | vishy: just noticed that my nova-network starts "sudo arping" every time, but they don't ever break out of that | 18:56 |
| bradlis7 | and I killed the last one, and it added the rules correctly | 18:57 |
| *** stuntmachine has quit IRC | 18:57 | |
| bradlis7 | looks like I found a bug, but I'm not sure where it was at... | 18:57 |
| *** nati2_ has quit IRC | 18:57 | |
| zaoul1 | the docs are one big big | 18:58 |
| zaoul1 | bug | 18:58 |
| vishy | zaoul1: you can do that, as long as you install virtio drivers | 18:58 |
| vishy | bradlis7: interesting the arping is failing? | 18:58 |
| bradlis7 | I guess so | 18:59 |
| zaoul1 | vishy: im talking about the whole image creation process | 18:59 |
| *** zaoul1 has left #openstack | 19:01 | |
| uvirtbot | New bug: #886284 in openstack-qa "NetworkManager should check instance is exist or not" [Undecided,Confirmed] https://launchpad.net/bugs/886284 | 19:02 |
| WormMan | last time I made an image, I booted the CD on my desktop KVM, then uploaded the image | 19:03 |
| vishy | zaoul1: yeah you can boot with kvm iso | 19:03 |
| masom | Anyone had experience with FreeBSD on openstack? The fstab obviously point to the wrong device, but can't figure out which one would be good. Bootloader says disk0s1a is available... | 19:03 |
| vishy | masom: not sure how virtio driver goes with freebsd. Is there one? | 19:04 |
| vishy | masom: you might have to switch your libvirt.xml to attach via scsi | 19:04 |
| *** marrusl has joined #openstack | 19:04 | |
| masom | there's one for 9.x but 8.2 kinda works okay in qemu. | 19:04 |
| *** tyska has quit IRC | 19:05 | |
| *** brodrigue has quit IRC | 19:05 | |
| masom | vishy: i'll take a look at that, thanks :) | 19:06 |
| vishy | masom: we've had plans for a long time to put metadata in glance about whether the image supports virtio, but it hasn't materialized yet | 19:07 |
| *** stuntmachine has joined #openstack | 19:07 | |
| masom | vishy: yeah, well so far i'm happy with the results, 1 day to get everythign setup, starting from scratch 2 times (to make sure i can replicate) and it works. | 19:08 |
| masom | vishy: i'm looking forward to the next releases with a working keystone/dashboard integration | 19:09 |
| *** gnu111 has quit IRC | 19:10 | |
| *** marrusl has quit IRC | 19:10 | |
| masom | vishy: Can you point me in a direction for the scripts generating libvirt.xml, i would like to pre-configure it. | 19:11 |
| *** nati2 has quit IRC | 19:11 | |
| *** hggdh has joined #openstack | 19:12 | |
| uvirtbot | New bug: #886276 in openstack-qa "wrong name match in destroy_container" [High,Confirmed] https://launchpad.net/bugs/886276 | 19:13 |
| uvirtbot | New bug: #886278 in openstack-qa "destroy_container should check instance argument " [Medium,Confirmed] https://launchpad.net/bugs/886278 | 19:13 |
| *** kieron has quit IRC | 19:14 | |
| *** kieron has joined #openstack | 19:15 | |
| *** AhmedSoliman has joined #openstack | 19:17 | |
| *** tryggvil_ has quit IRC | 19:20 | |
| *** mdomsch has quit IRC | 19:21 | |
| vishy | masom: there is a libvirt.xml.template file | 19:21 |
| vishy | masom: not sure where it will be since i'm not sure how you installed | 19:21 |
| masom | vishy: yeah saw the template but can't figure out how it would detect a freebsd guest vs a linux guest | 19:23 |
| masom | vishy: i'm on rhel6.1 (sl) | 19:23 |
| vishy | yeah you can't | 19:23 |
| vishy | that is why we need metadata in the image | 19:24 |
| masom | Ah :/ oh well | 19:24 |
| vishy | masom: you could make all drives attach via iscsi though | 19:24 |
| vishy | masom: the linux ones should be fine with that too i would suspect | 19:24 |
| vishy | * scsi | 19:25 |
| masom | yeah. Still playing with nova and the tools. Primilinary research before implementing it | 19:25 |
| uvirtbot | New bug: #886312 in nova "nova-network stuck on "sudo arping"" [Undecided,New] https://launchpad.net/bugs/886312 | 19:25 |
| *** code_franco has quit IRC | 19:27 | |
| *** gnu111 has joined #openstack | 19:27 | |
| *** AhmedSoliman has quit IRC | 19:28 | |
| *** jblesage has joined #openstack | 19:30 | |
| *** andy-hk has quit IRC | 19:31 | |
| *** Hakon|mbp has quit IRC | 19:33 | |
| *** Hakon|mbp has joined #openstack | 19:34 | |
| *** jdg has joined #openstack | 19:36 | |
| *** hezekiah_ has joined #openstack | 19:37 | |
| *** dendro-afk is now known as dendrobates | 19:37 | |
| *** pixelbeat has joined #openstack | 19:40 | |
| *** hggdh has quit IRC | 19:43 | |
| *** jdg has quit IRC | 19:44 | |
| *** cereal_b_ has joined #openstack | 19:45 | |
| *** jdg has joined #openstack | 19:45 | |
| *** gnu111 has quit IRC | 19:45 | |
| *** foexle has quit IRC | 19:46 | |
| *** Nadeeem has joined #openstack | 19:46 | |
| *** cereal_bars has quit IRC | 19:46 | |
| *** cereal___ has joined #openstack | 19:46 | |
| *** code_franco has joined #openstack | 19:50 | |
| *** cereal_b_ has quit IRC | 19:50 | |
| *** krow has quit IRC | 19:51 | |
| *** dolphm has quit IRC | 19:51 | |
| *** fulanito has joined #openstack | 19:51 | |
| *** dolphm has joined #openstack | 19:51 | |
| *** fulanito_ has joined #openstack | 19:53 | |
| *** dolphm_ has joined #openstack | 19:54 | |
| *** dolphm_ has joined #openstack | 19:55 | |
| *** dolphm has quit IRC | 19:55 | |
| *** thingee has quit IRC | 19:56 | |
| *** nycko is now known as cereal_bars | 19:58 | |
| *** dolphm_ has quit IRC | 19:59 | |
| *** cereal_bars is now known as nycko | 19:59 | |
| *** pixelbeat has quit IRC | 19:59 | |
| *** stuntmac_ has joined #openstack | 19:59 | |
| *** stuntmachine has quit IRC | 20:00 | |
| *** stuntmac_ is now known as stuntmachine | 20:00 | |
| *** cdub has quit IRC | 20:02 | |
| *** df1 has quit IRC | 20:03 | |
| *** dolphm has joined #openstack | 20:03 | |
| *** df1 has joined #openstack | 20:04 | |
| *** jtanner has quit IRC | 20:05 | |
| *** bsza has quit IRC | 20:08 | |
| *** doorlock has quit IRC | 20:11 | |
| *** jblesage has quit IRC | 20:13 | |
| cloudfly | hey ho all. | 20:14 |
| cloudfly | so what will quantum be using by default as a network management utility? | 20:14 |
| *** pothos_ has joined #openstack | 20:17 | |
| *** saju_m has joined #openstack | 20:19 | |
| *** pothos has quit IRC | 20:19 | |
| *** pothos_ is now known as pothos | 20:19 | |
| *** fulanito_ has quit IRC | 20:19 | |
| *** fulanito has quit IRC | 20:19 | |
| *** huslage has quit IRC | 20:20 | |
| *** jtanner has joined #openstack | 20:23 | |
| masom | vishy: actually the only driver that worked on rhel is ide. but now the image boots ! | 20:24 |
| masom | vishy: thanks for your help. | 20:24 |
| vishy | masom: cool | 20:24 |
| vidd-away | cloudfly, most like quantum will integrate with dashboard | 20:25 |
| *** vidd-away is now known as vidd | 20:26 | |
| masom | vishy: is the libvirt.xml per-hosts? If so, I could just setup a specific host for FBSD | 20:26 |
| vishy | yes you could | 20:28 |
| vishy | but how would you ensure that fbsd hosts go there? | 20:28 |
| vishy | * guests | 20:29 |
| swill | hey all... i am starting to work on an auth middleware for swift that will allow for the cloudstack user db to be used as the authentication method. | 20:29 |
| swill | has there been any work on this already? | 20:29 |
| masom | vishy: mmm maybe run it as a separate availability zone? | 20:30 |
| vishy | masom: could work | 20:30 |
| swill | I will be accessing the cloudstack db via their API: http://download.cloud.com/releases/2.2.0/api_2.2.12/TOC_Global_Admin.html | 20:31 |
| *** sgd2 has joined #openstack | 20:31 | |
| *** markvoelker has quit IRC | 20:33 | |
| * WormMan notes he's been asking for custom per-image libvirt settings for a while :) | 20:33 | |
| WormMan | (and a pony) | 20:34 |
| vishy | WormMan: patches accepted? | 20:34 |
| vishy | WormMan: It actually isn't hard, someone just has to define what the metadata settings in glance should be | 20:34 |
| WormMan | vishy: I'm still working on making the base system work right :) | 20:34 |
| vishy | WormMan: :) | 20:35 |
| WormMan | but I can now launch 400 instances and provide them API data almost as fast as they can boot | 20:35 |
| *** lborda has quit IRC | 20:35 | |
| *** lborda has joined #openstack | 20:35 | |
| *** dolphm_ has joined #openstack | 20:36 | |
| WormMan | (and of course, I think the code I changed is now not even in trunk any more as most of the metadata was refactored) | 20:36 |
| *** dolphm__ has joined #openstack | 20:38 | |
| bradlis7 | When I create an instance of my machine, it gives /dev/vda as a partition with the original size of the created device, and /dev/vdb is the partition size created by the flavor... Do I have to manually use /dev/vdb? | 20:38 |
| WormMan | if you want that space, yes | 20:38 |
| *** dolphm has quit IRC | 20:39 | |
| bradlis7 | I thought it would just resize the partition originally created to be the size defined in the instance... | 20:39 |
| bradlis7 | Can I join vda and vdb? | 20:39 |
| bradlis7 | Also, am I modifying future systems by modifying vda? | 20:39 |
| WormMan | no, vda is copy on write, changes will not be preserved | 20:40 |
| *** dolphm_ has quit IRC | 20:40 | |
| WormMan | (you can reboot that specific instance, but new instances with the same image will be fresh) | 20:40 |
| bradlis7 | right | 20:40 |
| bradlis7 | Is this standard practice in cloud computing? | 20:40 |
| WormMan | and I don't have a good answer for how to really use vdb, for our databases we just mkfs it and drop the DB there | 20:41 |
| Kiall | WormMan: you'll like this change: https://blueprints.launchpad.net/nova/+spec/separate-nova-metadata | 20:41 |
| Kiall | ;) | 20:41 |
| WormMan | Kiall: yes, I will, I have to run nova-api on every compute node as it's just too slow :) | 20:41 |
| Kiall | yea .. this one too then ;) https://blueprints.launchpad.net/nova/+spec/nova-metadata-nodb | 20:41 |
| WormMan | the problem was, the db was not the problem. The DB returning 2000 instances for what should be a single row query, and then nova-api doing a regex match, that was the problem :) | 20:42 |
| Kiall | lol | 20:42 |
| Kiall | it seriously does that? | 20:42 |
| WormMan | now it only returns the 400 running instances instead of all(deleted and not) ones | 20:43 |
| Kiall | lol | 20:43 |
| WormMan | in the diablo flavor we're running, yes | 20:43 |
| Kiall | even terminated instances?? | 20:43 |
| WormMan | yep | 20:43 |
| vishy | bradlis7: we are adding 2 fields in the flavor | 20:43 |
| WormMan | I didn't want to actually move the regex search into an actual SQL query as that would have taken many more code changes | 20:43 |
| vishy | bradlis7: so the flavor will have a size for root fs and for second fs | 20:44 |
| bradlis7 | vishy: What do the fields do? | 20:44 |
| bradlis7 | ah | 20:44 |
| *** jsavak has quit IRC | 20:44 | |
| *** lorin1 has quit IRC | 20:45 | |
| *** lorin1 has joined #openstack | 20:45 | |
| bradlis7 | vishy: Sounds good. So I can resize rootfs to be larger? | 20:45 |
| vishy | now? | 20:46 |
| bradlis7 | I was talking when you add the field | 20:46 |
| vishy | if you make an image with a larger rootfs it will use it | 20:46 |
| vishy | bradlis7: yes, after the field you can make them different sizes | 20:46 |
| bradlis7 | Right. I want it to be sized on request though | 20:46 |
| bradlis7 | ok, sounds good | 20:46 |
| bradlis7 | when is the new release coming out? | 20:47 |
| vishy | 5 months :) | 20:47 |
| bradlis7 | cool | 20:47 |
| cloudfly | looking forward to it | 20:47 |
| *** lts has quit IRC | 20:48 | |
| Kiall | vishy: are there any plans to allow an instances security groups to be changed on the fly? (The list of groups, rather than the groups rules) | 20:48 |
| *** mach has joined #openstack | 20:53 | |
| *** msivanes has quit IRC | 20:55 | |
| *** mach has quit IRC | 20:55 | |
| *** stuntmachine has quit IRC | 20:55 | |
| *** swill has left #openstack | 20:55 | |
| *** dholm has joined #openstack | 20:55 | |
| *** rich_ has joined #openstack | 20:56 | |
| *** vidd is now known as vidd-away | 20:57 | |
| *** nati2 has joined #openstack | 21:00 | |
| *** jks has quit IRC | 21:01 | |
| *** doorlock has joined #openstack | 21:01 | |
| *** dolphm__ has quit IRC | 21:02 | |
| *** dolphm has joined #openstack | 21:03 | |
| *** dholm has quit IRC | 21:04 | |
| *** bcwaldon has quit IRC | 21:04 | |
| *** bsza has joined #openstack | 21:04 | |
| *** n88 has quit IRC | 21:05 | |
| *** lorin1 has quit IRC | 21:05 | |
| *** saju_m has quit IRC | 21:06 | |
| *** cereal___ has quit IRC | 21:06 | |
| *** dolphm has quit IRC | 21:07 | |
| *** kieron has quit IRC | 21:08 | |
| uvirtbot | New bug: #886353 in nova "Faults raised by OpenStack API Resource handlers fail to be reported properly" [Undecided,In progress] https://launchpad.net/bugs/886353 | 21:11 |
| *** david__c has joined #openstack | 21:17 | |
| *** nerdstein has joined #openstack | 21:21 | |
| *** nerdstein has left #openstack | 21:21 | |
| *** krow has joined #openstack | 21:22 | |
| *** clopez has quit IRC | 21:22 | |
| *** juddm has quit IRC | 21:25 | |
| *** bsza has quit IRC | 21:26 | |
| *** rods has quit IRC | 21:28 | |
| *** bradlis7 has quit IRC | 21:28 | |
| *** lborda has quit IRC | 21:30 | |
| *** jtanner has quit IRC | 21:30 | |
| *** dolphm has joined #openstack | 21:32 | |
| *** jwalcik has quit IRC | 21:32 | |
| *** masom has quit IRC | 21:34 | |
| *** Turicas has joined #openstack | 21:36 | |
| *** lionel has quit IRC | 21:37 | |
| *** dolphm has quit IRC | 21:37 | |
| *** lionel has joined #openstack | 21:37 | |
| *** dolphm has joined #openstack | 21:38 | |
| *** sgd2 has quit IRC | 21:38 | |
| *** cp16net has quit IRC | 21:39 | |
| *** nati2 has quit IRC | 21:40 | |
| *** nati2 has joined #openstack | 21:40 | |
| *** dolphm has quit IRC | 21:42 | |
| *** dolphm has joined #openstack | 21:46 | |
| *** krow has quit IRC | 21:46 | |
| *** mattray has quit IRC | 21:48 | |
| *** vidd-away is now known as vidd | 21:53 | |
| *** ldlework has quit IRC | 21:56 | |
| *** rnorwood has quit IRC | 22:00 | |
| *** sebastianstadil has quit IRC | 22:02 | |
| *** sebastianstadil has joined #openstack | 22:03 | |
| *** krow has joined #openstack | 22:04 | |
| *** rich_ has quit IRC | 22:05 | |
| *** tlocalhos is now known as localhost | 22:08 | |
| *** bradlis7 has joined #openstack | 22:10 | |
| *** dolphm has quit IRC | 22:12 | |
| *** dendrobates is now known as dendro-afk | 22:12 | |
| *** dolphm has joined #openstack | 22:13 | |
| *** dendro-afk is now known as dendrobates | 22:13 | |
| *** MarkAtwood has joined #openstack | 22:13 | |
| *** egant has quit IRC | 22:13 | |
| *** egant has joined #openstack | 22:13 | |
| *** localhost has quit IRC | 22:13 | |
| *** localhost has joined #openstack | 22:14 | |
| *** misheska has joined #openstack | 22:17 | |
| *** dolphm has quit IRC | 22:17 | |
| *** dendrobates is now known as dendro-afk | 22:18 | |
| *** dendro-afk is now known as dendrobates | 22:18 | |
| *** dragondm has quit IRC | 22:20 | |
| *** rsampaio has quit IRC | 22:20 | |
| *** bradlis7 has quit IRC | 22:21 | |
| *** code_franco has quit IRC | 22:35 | |
| *** doorlock has quit IRC | 22:36 | |
| *** markwash has quit IRC | 22:36 | |
| *** blamar has quit IRC | 22:36 | |
| *** markwash has joined #openstack | 22:38 | |
| *** blamar has joined #openstack | 22:38 | |
| *** doorlock has joined #openstack | 22:39 | |
| vidd | Kiall, you here? | 22:39 |
| *** cloud__ has joined #openstack | 22:41 | |
| *** blamar has quit IRC | 22:42 | |
| *** nati2_ has joined #openstack | 22:42 | |
| *** blamar has joined #openstack | 22:42 | |
| *** nati2 has quit IRC | 22:43 | |
| *** mtucloud has quit IRC | 22:44 | |
| *** nati2 has joined #openstack | 22:48 | |
| *** nati2_ has quit IRC | 22:50 | |
| *** martin has quit IRC | 22:52 | |
| *** doorlock has quit IRC | 22:57 | |
| *** dendrobates is now known as dendro-afk | 22:57 | |
| *** martin has joined #openstack | 22:58 | |
| *** rnirmal has quit IRC | 22:59 | |
| *** martin is now known as Guest37338 | 22:59 | |
| *** Guest37338 is now known as martin | 23:04 | |
| *** aliguori has quit IRC | 23:05 | |
| *** haji has joined #openstack | 23:09 | |
| *** haji is now known as magg | 23:10 | |
| magg | hello | 23:10 |
| magg | i just updated the dashboard from git | 23:10 |
| vidd | magg, how did it go? | 23:10 |
| *** dolphm has joined #openstack | 23:10 | |
| magg | but im getting the passwordCredentials expected error on the login page | 23:10 |
| magg | bad vidd bad :( | 23:11 |
| *** dolphm has joined #openstack | 23:11 | |
| vidd | when you set up your database, did you get prompted for a username and password? | 23:11 |
| magg | keystone database? | 23:12 |
| magg | or nova database? | 23:12 |
| magg | or glance hehe | 23:12 |
| vidd | no...the dashboard database | 23:12 |
| magg | omg | 23:12 |
| *** vladimir3p has quit IRC | 23:12 | |
| vidd | =] | 23:12 |
| vidd | i will take that as a "no" | 23:12 |
| magg | yeah | 23:12 |
| *** egant has quit IRC | 23:12 | |
| magg | is this database new, i never heard of it | 23:13 |
| magg | how do i set it up | 23:13 |
| *** Razique has joined #openstack | 23:13 | |
| Razique | Hi all, what's the point of having multiple networks per projecT ? | 23:14 |
| *** daharon has joined #openstack | 23:14 | |
| vidd | dont know...i cant get past keystone nuking my nova and glance wont do anything but complain about auth keys | 23:14 |
| vidd | =] | 23:14 |
| magg | damn | 23:14 |
| Razique | vidd: still awake N | 23:14 |
| magg | http://wiki.openstack.org/OpenStackDashboard this never talks about a DB | 23:14 |
| Razique | ? :p | 23:14 |
| vidd | Razique, sleep is that aninmal wool comes from right? | 23:15 |
| Razique | sick =D | 23:15 |
| *** dendro-afk is now known as dendrobates | 23:16 | |
| *** dolphm has quit IRC | 23:16 | |
| magg | how come devstack installs all of openstack even the dashboard, but i can't install it manually | 23:18 |
| *** Ryan_Lane has quit IRC | 23:18 | |
| magg | so frustrating | 23:18 |
| magg | i only find errors :( | 23:19 |
| Razique | magg: what are ur issues ? | 23:19 |
| magg | im getting the passwordCredentials expected error on the dashboard login page | 23:20 |
| *** Ryan_Lane has joined #openstack | 23:20 | |
| magg | vidd: is this what you talking about tools/with_venv.sh dashboard/manage.py syncdb | 23:21 |
| magg | i didnt use that command for sure | 23:23 |
| *** nphase has quit IRC | 23:23 | |
| vidd | magg, that is the tool you want | 23:24 |
| magg | thanks mate | 23:25 |
| vidd | magg, ya gatta read the README =] | 23:25 |
| vidd | its not just there to look purty =] | 23:26 |
| magg | i usually follow the docu from the site | 23:26 |
| magg | i guess i will follow the readme then from now on | 23:26 |
| vidd | i do as well....for the items i install from repo's..... | 23:27 |
| *** mtucloud has joined #openstack | 23:27 | |
| vidd | but years of experience have taught me...if its got a README....read it =] | 23:27 |
| magg | hehe | 23:27 |
| *** nerens has quit IRC | 23:28 | |
| vidd | thats one of the reas i like linux so much.....i finally got the extra 2.5 hours to read the EULA =] | 23:28 |
| *** cloud__ has quit IRC | 23:28 | |
| magg | hahaha | 23:30 |
| magg | also im using this to get the dashboard working with apache: http://cssoss.wordpress.com/2011/04/27/openstack-beginners-guide-for-ubuntu-11-04-installation-and-configuration/ | 23:32 |
| magg | but somehow i can't get it to work... | 23:33 |
| vidd | magg, im not surprised =] | 23:33 |
| vidd | zykes-, can you help magg with the dashboard+apache? (are you even here?) | 23:35 |
| *** martin has quit IRC | 23:35 | |
| *** qazwsx|2 has quit IRC | 23:35 | |
| magg | cuz its old or something? | 23:35 |
| vidd | magg, yes | 23:35 |
| magg | oh i see | 23:36 |
| vidd | the git repo isnt even the same =] | 23:36 |
| vidd | that doc says to get it from BZR and it hasnt been there since i started with this =] | 23:37 |
| magg | well | 23:37 |
| magg | yeah | 23:37 |
| vidd | if you spin up devstack in a VM, yo can look at what they do and do likewise on your system | 23:37 |
| magg | installed devstack on vm yesterday | 23:38 |
| magg | i will take a look | 23:38 |
| vidd | =] | 23:38 |
| *** jdg has quit IRC | 23:39 | |
| *** dendrobates is now known as dendro-afk | 23:44 | |
| *** robix has quit IRC | 23:47 | |
| vidd | magg, good luck | 23:48 |
| *** egant has joined #openstack | 23:48 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!