[00:01] <sunya7a> hi...I have an Xserve xeon (nephalem) server and wish to install Ubuntu server on it. someone suggested to use the 64-bit Mac (AMD64) server install CD image for it..but it is intel based cpu is it not? should I use the PC (Intel x86) server install cd image or the AMD64 Mac image? I tried burning the AMD64 Mac version onto a DVD but was not able to boot from the DVD...I'm not sure if
[00:01] <sunya7a> this is related to the image I chose or if this is un unrelated problem
[00:03] <sunya7a> I believe I created the DVD correctly...I downloaded the iso image. then mounted it on another mac using disk utilities...once I found the disk id I went into terminal and used dd to extract the iso to another location. I am able to mount this extracted image and eventually burned this extracted iso to a DVD
[00:03] <sarnold> sunya7a: the 'amd64' name is because AMD's 64 bit extensions to x86 took hold; intel implemented them
[00:03] <sarnold> sunya7a: intel's 64 bit ("itanium") went nearly no-where for a wide variety of reasons that are fun to discuss over beers :)
[00:04] <sarnold> sunya7a: so when you buy a modern intel 64 bit chip, chances are good it is running the amd64 instruction set
[00:05] <sunya7a> sarnold: i see...so I take it the image is not the issue then...I read much online regarding difficulties booting on mac's due to the EFI only requirement
[00:05] <sarnold> sunya7a: (though I don't have advice on what _you_ need to use on your xserve machines -- the mac amd64 image would be my first starting place, too, fwiw.)
[00:05] <sunya7a> However, the issue is that I don't seem to be able to even see the option to boot from the DVD which is what stumps me
[00:06] <sunya7a> sarnold: well...to be honest, since most macs these days are intel based...I would assume that an image labelled for mac (albeit being labeled with amd64) shouldn't be wrong...so I'm sure you are right on point
[00:07] <sunya7a> sarnold: my next attempt is going to be to boot from a USB instead...maybe there's just something wrong with booting from DVD (which might be why it was so difficult getting the image onto the dvd in the first place)
[00:08] <sarnold> sunya7a: ooh, usb is a good idea. At least it doesnt make another coaster, anyway.. :)
[00:09] <sunya7a> sarnold: I did verify that there is nothing wrong with the DVD device though...the system still have mac os server installed so booting into it allows me to use the DVD just fine...and it looks the way it is supposed to (all icons on the root, instead of a single iso file, like my first 3 dvd burning attempts)
[00:09] <sunya7a> the system still has**
[00:10] <sarnold> sunya7a: hehe, yeah, that's happened to me too :)
[00:11] <sunya7a> sarnold: btw...when I put the DVD into any computer I have...and right-click to view information (within disk utility) it says that the DVD is not bootable...I wonder if this has any bearing though because I'm not aware of how I could make the DVD bootable...it seems that just burning the image with disk utility should have made it bootable (please correct me if this assumption is wrong)
[00:12] <escott> sunya7a, did you burn as image or as file
[00:12] <sarnold> sunya7a: tbh I've never used a bootable DVD before; I've only ever used bootable CDs.
[00:12] <sarnold> escott: first three as file, next one as image :D
[00:13] <shauno> I wouldn't put too much hope behind booting from usb, apple's efi implementation doesn't usually agree with it
[00:14] <sarnold> :(
[00:14] <sunya7a> escott: well...disk utility seems to have no option to pick to "burn as files" or "burn as image". it just says to burn.......I didn't know you had to extract the iso from the iso (using dd to extract it) the first 3 times...so i ended up with a dvd that had just the iso file on the root...the next time I extracted the iso..and when I burned that one I got a DVD with a bunch of files
[00:14] <sunya7a> (there's an image on ubuntu documentation that shows "not good" , "good" )
[00:15] <sarnold> hrm. that probably breaks the dvd-variant of the boot specification thingy (like cd's el torito..)
[00:16] <shauno> very much so (it's no different for dvd; in this regard it's just a bigger CD)
[00:16] <sunya7a> sarnold: the extraction breaks the boot specification you say?
[00:17] <sunya7a> i was working through this suggestion: http://ubuntuforums.org/showthread.php?t=1870261&page=5
[00:17] <sunya7a> on the very bottom someone linked an explaination as to why you need to use dd to extract the iso
[00:18] <sunya7a> the working theory is that ubuntu started to create a hybrid iso file...that can be used for USB as well as DVD...but somehow that makes burning to DVD impossible unless you extract the ISO
[00:20] <adam_g> zul: hows about this  http://people.canonical.com/~agandelman/folsom/keystone_2012.2.1-0ubuntu1.2~cloud0/
[00:20] <zul> adam_g: looks good
[00:33] <adam_g> SpamapS: ping
[00:38] <SpamapS> adam_g: pong, just have 2 minutes, whats up?
[00:39] <adam_g> SpamapS: the keystone packagte you accepted to q-proposed was superceded ~2 hours later by a security update
[00:40] <adam_g> SpamapS: im putting together a new stable/folsom snapshot for q-proposed to supercede what you had accpeted today, that will include those security fixes
[00:42] <ssvss> Hi, this isn't ubuntu specific. I use "du -sh" to see the size of a folder, the folder is getting updated. but I feel du cmd doesn't calculate the size everytime I run the cmd. I see to just display the previously calculated sizes.
[00:43] <SpamapS> adam_g: doh
[00:43] <SpamapS> adam_g: ok, will look tomorrow morning
[00:43] <SpamapS> or later tonight
[00:44] <ssvss> *It seem to just display the previously calculated sizes. any idea on how I can find the updated sixe of the dir.
[00:44] <escott> sunya7a, "extracting" the filesystem from the iso breaks the bootloader
[00:45] <escott> sunya7a, as you no doubt realized
[00:45] <sarnold> ssvss: updated files may not have their metadata updated until the writing programs performs an fsync(2) or sync(2) syscall or closes the file...
[00:46] <sunya7a> escott: I've found another article that states so =(  ya i guess it's not gonna happen this way...it seems that the only way that people have been successfully installed ubuntu on macs is with a dual boot setup...by taking advantage of rEFIt
[00:47] <ssvss> Thanks.  I am using rsync, this might be the reason.
[00:47] <sunya7a> escott: I haven't found a single instance of someone installing Ubuntu cleanly on a mac
[00:48] <histo> sunya7a: don't buy a mac then.
[00:48] <histo> !uefi | sunya7a
[00:49] <shauno> I haven't lately, but I've installed 10.04 with no issues at all  (well, at this level anyway)
[00:49] <histo> ssvss: rsync has a progress option also du should be updating.
[00:54] <sunya7a> ubottu: I've attempted the instructions described in the article you linked but I am not able to get the mac to boot from the DVD I created...holding down 'C' results in booting to the mac osx..and holding down alt/option shows only my primary hard drive, no dvd
[00:54] <sunya7a> ubottu: i see...=) thank you any way =P
[00:56] <shauno> sunya7a: you don't want UEFI instructions, apple's EFI implementation is pre-UEFI (hence the -mac images, they cut out the attempts at UEFI booting which didn't work on apple's implementation)
[00:57] <escott> sunya7a, macs are proper efi so things are a bit messy
[00:57] <shauno> you specifically want https://help.ubuntu.com/community/Xserve2-1  because the xserv doesn't support bootcamp, so trying to boot it the same way as a macbook will get you nowhere
[00:58] <sunya7a> shauno: okay...thank you for that information...I will take a look at that post
[00:59] <shauno> this becomes very relevant because along with the driver packages and helper apps, bootcamp also includes the firmware update that provides bios-emulation for booting from "legacy" methods.  this update never reached the xserve
[01:08] <jdstrand> adam_g: fyi, nova and cinder are coming too
[01:08] <jdstrand> cinder tonight, nova tomorrow
[01:12] <jdstrand> well, nova might be tonight
[01:12] <jdstrand> soon anyway
[01:55] <ekaj> I am having trouble getting ubuntu server 12.04 LTS to 'sudo apt-get update' - every link is an "Err" or "Ign".... us.archive.ubuntu.com is active ( I can go on my lappy) but not the server. I believe the problem is my college's DNS server, because when I ping the us.archive.ubuntu.com I get 10.0.0.112 (college DNS) - I tried putting dns-nameservers 8.8.8.8 in /etc/network/inferfaces, but that didn't work... Could anyone help me?
[01:56] <histo> ekaj: does your /etc/network/interfaces get over written?
[01:57] <ekaj> No... I know that's not the correct way to do that anymore, but it's not gettong over written
[01:57] <sarnold> did the nameservers line make it all the way through to /etc/resolv.conf ?
[01:57] <ekaj> Should I flush the cache?
[01:57] <ekaj> Yep
[01:57] <ekaj> nameserver 10.0.0.112, 8.8.8.8
[01:58] <ekaj> Should I just remove the 10.0.0.112 from resolv.conf?
[01:58] <histo> ekaj: you should add nameservers to /etc/resolv.conf.d/head
[01:59] <ekaj> Just create that file and add 'dns-nameservers 8.8.8.8'?
[01:59] <histo> ekaj: sorry /etc/resolvconf/resolv.conf.d/head
[01:59] <histo> ekaj: just nameserver 8.8.8.8  in that file  When you restart resolvconf it will put that in your /etc/resolv.conf file
[02:00] <ekaj> That one it says not to edit by hand =p so I added that there, I guess now /etc/init.d/networking restart
[02:01] <histo> ekaj: right that's how the do not edit message gets into the /etc/resolv.conf file it comes from the head file.
[02:01] <ekaj> Ahh okay
[02:01] <ekaj> I love you, thank you =p
[02:02] <ekaj> I've been screwing with that for an hour
[02:03] <ekaj> Now, sorry to bug you again, but what would be the proper way to set up a RAID configuration on a server with 4 HDDs? I've never used RAID =/ a reliable guide would be excellent
[02:03] <histo> !raid | ekaj
[02:03] <ekaj> Thankya
[02:08] <ekaj> This is what happens when a freshman sets up server =p
[02:08] <ekaj> *servers
[03:49] <histo> ?
[03:49] <sarnold> !
[03:53] <histo> ^
[03:59] <cfhowlett> #eol
[03:59] <cfhowlett> !eol
[04:03] <luminous> hello, here is an upstart config for logstash, pretty basic: http://dpaste.de/62AvF/
[04:04] <luminous> I can run logstash in the exact way outlined by the config, and logstash will work fine. but starting through upstart eventually results in logstash closing and being respawned
[04:41] <esuave> Can anyone tell me what lacp_rate actually does? I'm going to go nutz! Can't find any solid explanation anywhere on google
[06:05] <goddard> how can i debug a email domain issue
[06:08] <histo> what kind of domain issue?
[06:15] <adam_g> zul: http://people.canonical.com/~agandelman/folsom/cinder_2012.2.1-0ubuntu1.1~cloud0/ && http://people.canonical.com/~agandelman/folsom/euca2ools_2.0.2-1ubuntu2.1~cloud0/
[06:15] <adam_g> jamespage: ^
[06:18] <someperson> Anyone know how to set up s3 keys on an Ubuntu server like Heroku's "heroku config:add S3_KEY=XXXXXXXXX S3_SECRET=XXXXXXXXXX" ?
[06:20] <someperson> #vps
[06:20] <someperson> #linode
[06:28] <someperson> Anyone know how to set up s3 keys on an Ubuntu server like Heroku's "heroku config:add S3_KEY=XXXXXXXXX S3_SECRET=XXXXXXXXXX" ?
[06:44] <goddard> how do you trouble shoot smtp issues with a domain on your network
[07:02] <histo> goddard: What sort of issues?
[07:08] <goddard> histo: for some reason on my virtual domains can only pull mail from postfix but not sent mail via smtp
[07:14] <someperson> #rubyonrails
[07:16] <histo> goddard: can local users send mail?
[07:16] <histo> goddard: you'd have to look at your logs and see what's goign on.
[07:34] <goddard> histo: is there a way i can test smtp connects with verbose logging like ssh -vvvv ?
[07:40] <histo> goddard: Isn't there output in mail.log?
[07:42] <andol> goddard: Play smtp manually? Typing in the HELO, MAIL FROM, RCPT TO, etc yourself?
[07:44] <andol> (Will of course get a bit trickier if it is used in a submission scenarion, with starttls, smtp auth, etc)
[07:50] <goddard> histo: ill look at the log
[08:01] <someperson>  What is the equivalent of heroku config:add S3_KEY=XXXXXXXXX S3_SECRET=XXXXXXXXXX on a remote Ubuntu server?
[08:14] <melmoth> someperson, your question is not clear enough
[08:14] <andol> someperson: The relevant equivlant isn't towards a remote Ubuntu server, but towards a specific piece of software you are running on your Ubuntu server, which speaks to S3.
[08:15] <melmoth> it looks like environment variables.
[08:15] <melmoth> actually, looks like ec2 environment variable (amazon s3 stuff) indeed
[08:16] <someperson> So there's no way I can set in my server's command shell like heroku?
[08:16] <melmoth> i dont know heroku
[08:16] <melmoth> but you can set an environment variable in bash with "export S3_KEY=XXXXXXXXX"
[08:16] <someperson> Aren't environment variables insecure though? Since it shows the keys in the code?
[08:17] <melmoth> good question. I guess only you and root will be allowed to lookfor your environment
[08:17] <someperson> melmoth: What would you do?
[08:18] <andol> someperson: You can't really compare Herouke compared to a Ubuntu server, the later being a much more arbitrary multi purpose environment.
[08:18] <andol> someperson: Again, the relevant part is what daemons/code you will be running which needs to communicate with S3.
[08:18] <melmoth> i dont know what i would do for sure, because i m still not sure what is the problem to solve
[08:20] <melmoth> if you dont trust root, you will be stuck anyway, even if you do not use an env variable.
[08:20] <melmoth> the program that will need your pass need to read it from somewhere
[08:20] <melmoth> if the program can, then root can
[08:20] <melmoth> worst case, root can dig into ram itself (i dont know how, but i bet it can be done)
[08:20] <someperson> I was using the dragonfly gem for images on Heroku and then migrated to Digital Ocean...but I need to reconfigure my s3 keys
[08:20] <someperson> http://markevans.github.com/dragonfly/file.Heroku.html
[08:21] <melmoth> those are environmen variables inded
[08:21] <someperson> I'm assuming I just need to config S3_Key=XXXX and S3_Secret=XXXX in my remote shell somehow
[08:21] <melmoth> yes
[08:21] <melmoth> export S3_Key=XXXX
[08:21] <melmoth> and there you go
[08:22] <someperson> Ah, I tried it without export and nothing worked, but let me try now
[08:22] <melmoth> if you do not do an export, the variable will only exist in the current shell, not in the new shell created for a new application
[08:23] <someperson> now do I need to do that under app@HostName: or app@HostName: ~/myapp/current/ ?
[08:25] <melmoth> export the variable before starting your app
[08:25] <melmoth> if you need to do it automatically, set stuff in your ~/.bashrc (assuming you are using bash)
[08:26] <someperson> I'm using ~/.zshrc locally but ~/.bashrc remotely
[08:27] <someperson> What do you mean by export the variable before starting your app? Do I need to stop nginx?
[08:27] <melmoth> i dont know
[08:27] <melmoth> i dont know what nginx is
[08:27] <someperson> it's my web server
[08:28] <melmoth> well, if this the application that needs the environment, it must be set before it start. right ?
[08:28] <someperson> let me just do this and see if it works:
[08:28] <PadNet_201> So here's a question. Say i wanna report a program crash, which repos do i need to add these necessary programs
[08:28] <someperson> app@HostName: export S3_KEY=XXXXX S3_SECRET=XXXXX
[08:29] <melmoth> i dont understand app@Hostname
[08:29] <melmoth> and i m not sure you can export 2 variable in a single line
[08:29] <melmoth> you need to test it
[08:29] <someperson> app is the user and Hostname is my remote IP address
[08:30] <someperson> I also have root@HostName:
[08:30] <melmoth> ok. sounds ok (except i would do  2 export, one per variable)
[08:30] <melmoth> then launch your app
[08:30] <someperson> ok
[08:30] <someperson> so don't stop nginx
[08:33] <someperson> They're showing any responses in the shell command, they just go to the next line
[08:33] <someperson> Is that normal?
[08:33] <someperson> *They're not
[08:34] <melmoth> you should have a prompt, no message.
[08:34] <someperson> Aren't I supposed to get some feedback, like S3_Key saved or something?
[08:34] <melmoth> no
[08:34] <someperson> It just goes to the next shell line without any message
[08:35] <someperson> Oh well, lets see if it worked
[12:37] <luminous> hello, here is an upstart config for logstash, pretty basic: http://dpaste.de/7phHf/
[12:37] <luminous> I can run logstash in the exact way outlined by the config, and logstash will work fine. but starting through upstart eventually results in logstash closing and being respawned
[12:47] <jacobw> ahoy
[12:47] <jacobw> i want to find out if the snmpd package is built with the --with-mib-modules=smux option, what's the easiest way to do that?
[13:09] <Skaag> Something keeps rewriting my /etc/resolv.conf when the machine boots...
[13:09] <Skaag> that's extremely annoying
[13:09] <Skaag> how do I disable that?
[13:10] <jacobw> Skaag: it's resolvconf
[13:10] <Skaag> ok, uninstalled it
[13:10] <Skaag> I thought it only fetches them upon dhcp connection establishment
[13:10] <jacobw> Skaag: embracing it is easier than fighting it, it's used by network-manager
[13:10] <Skaag> my machine uses a static ip
[13:11] <Skaag> it's a server, not a desktop
[13:12] <Skaag> jacobw: how do I embrace it such that name servers will always be the same after a reboot?
[13:40] <Quest> regarding http://stackoverflow.com/questions/14990973/tomcat7-and-apache2-on-ubuntu-in-integration-with-some-sort-of-proxy , what modes exactly do I need?
[14:06] <Quest> hi, i dont see mod_proxy_ajp nor mod_proxy in mods-available dir of apache 2. any solution?
[14:14] <Pici> Quest: you don't see proxy_ajp.load ?
[14:15] <Quest> i do
[14:15] <Quest> Pici,  thats actually the mod_proxy_ajp?
[14:15] <Quest> what about mod_proxy?
[14:15] <Pici> proxy.load
[14:21] <jdstrand> smoser: hi! I've got a diablo stack deployed and am running into 'Kernel not found for image...'
[14:21] <Quest> Pici,  are you sure?
[14:21] <smoser> jdstrand, i think the first issue was using diablo
[14:21] <smoser> :)
[14:21] <jdstrand> smoser: smoser heh, well, I don't have a choice-- testing a security update
[14:22] <smoser> did something work at first and then start to fail ?
[14:22] <smoser> (an upgrade?)
[14:22] <jdstrand> ok, let me explain
[14:22] <Pici> Quest: 100%, what are you trying to do that isn't working?
[14:22] <smoser> jdstrand, it is possible that you cannot boot a 'disk1.img' (aka "full disk image") on quantal.
[14:22] <jdstrand> smoser: I use these commands to add an image to glance: https://wiki.ubuntu.com/SecurityTeam/TestingOpenStack#Adding_an_image_via_glance
[14:22] <smoser> i dont recall for certain if that is the case, but if it is, then you can use the uec style image.
[14:23] <jdstrand> smoser: this is oneiric
[14:23] <jdstrand> smoser: the image boots fine with euca2ools
[14:23] <jdstrand> smoser: but if I use nova boot, it fails:
[14:23] <jdstrand> $ nova boot --image 5 --flavor 1 testme
[14:23] <jdstrand> Cannot find requested image 5: Kernel not found for image 5. (HTTP 400)
[14:23] <Quest> Pici,  i was trying to be sure before tying anything as it usually leads to messy things you know.
[14:23] <Pici> Quest: using a2enmod should alleviate any mess
[14:23] <Quest> Pici,  + i would need a guide on how to setup the config files for those. tomcat and apache
[14:24] <Quest> Pici,  ok. do i need to be in a specific dir to run a2enmod?
[14:24] <jdstrand> smoser: I tried both ubuntu-12.04-beta1-server-cloudimg-amd64-disk1.img and ubuntu-11.10-server-cloudimg-amd64-disk1.img. it seems like I need to add something else to glance add, but I don't in essex and higher so I am a bit confused
[14:25] <Pici> Quest: no. It automatically creates the symlinks
[14:25] <smoser> jdstrand, can you pastebin 'euca-describe-images <id>' and 'nova show <id>' ?
[14:25] <smoser> i'm surprised that ec2 api would work here but nova not.
[14:25] <Quest> Pici,  so i can be just in /hom and run a2enmod proxy.load?
[14:25] <Pici> Quest: just: sudo a2enmod proxy
[14:26] <jdstrand> smoser: http://paste.ubuntu.com/1699230/
[14:26] <jdstrand> smoser: I think you menat image-show
[14:27]  * jdstrand gets it
[14:27] <smoser> right.
[14:27] <smoser> :)
[14:27] <jdstrand> smoser: http://paste.ubuntu.com/1699237/
[14:27] <Quest> Pici,  ok.
[14:27] <Quest> thanks
[14:27] <Quest> Pici,  + i would need a guide on how to setup the config files for those. tomcat and apache
[14:28] <jdstrand> smoser: fyi, this is all in a snapshotted vm, so I can do anything you wnat
[14:28] <smoser> jdstrand, ok. can you just confirm, you're saying it boots fine using 'euca-run-instances ami-00000005' but *not* using 'nova boot 5' ?
[14:29] <smoser> that just seems odd to me. but if you want to just get something booting via 'nova boot', i'd suggest just registering one of the "uec style" images, which have their own kernel/ramdisk.
[14:31] <jdstrand> smoser: that's right: http://paste.ubuntu.com/1699260/
[14:32] <jdstrand> smoser: that is acceptable to me. I don't really care about the bug, I just need nova boot to work
[14:32] <jdstrand> smoser: let's say I am at http://uec-images.ubuntu.com/releases/oneiric/release/, you know, just for kicks
[14:33] <jdstrand> smoser: which image should I grab and can I continue to use 'glance add'?
[14:34] <smoser> jdstrand, just for kicks, lets try
[14:34] <smoser> hm..
[14:35] <jdstrand> (fyi, using glance is important since it is also supported in 11.10-- though if I had to, I could use something else)
[14:35] <smoser> oh wow. i think you used to be able to specify a kernel with nova boot. but you can't in my client (raring).  if you could i was going to suggest '--kernel=1'
[14:36] <jdstrand> yeah, nova boot doesn't have a kernel option on 11.10 either
[14:37] <smoser> ok. so download ubuntu-11.10-server-cloudimg-amd64.tar.gz
[14:37] <jdstrand> (this problem doesn't exist in essex)
[14:37] <smoser> then, for simplicity i might suggest trying cloud-publish-image
[14:37] <smoser> err..
[14:37] <smoser> cloud-publish-tarball ubuntu-11.10-server-cloudimg-amd64.tar.gz mybucket/amd64
[14:37] <smoser> cloud-publish-tarball ubuntu-11.10-server-cloudimg-amd64.tar.gz mybucket amd64
[14:37] <smoser> and seeing if that boots.
[14:37] <soren> greping through the entire history of python-novaclient, it never mentions "kernel" at all.
[14:37] <jdstrand> I think I tried cloud-publish-image last night, and glance didn't know about it (though euca2ools did)
[14:37] <smoser> soren, don't you find that odd ?
[14:38] <soren> smoser: I do.
[14:38] <smoser> if that doesn't work, then somehow we have to use glance client to upload and specify the kernel image and then kernel-id.
[14:38] <smoser> its just more commands.
[14:38] <smoser> and i'm not familiar with it.
[14:38] <jdstrand> will cloud-publish-tarball use glance?
[14:39] <smoser> well, everything uses glance indirectly.
[14:39] <smoser> cloud-publish-tarball talks to nova api, and nova api talks to glance.
[14:39] <jdstrand> or rather, is glance supposed to know about the image after using cloud-publish-tarball?
[14:39] <smoser> yes.
[14:39] <smoser> it should be listable by 'glance image-list'
[14:39] <Quest> any help for config files?
[14:39] <smoser> or whatever the right glance syntax is.
[14:40] <Quest> any help for config files Pici ?
[14:40] <jdstrand> smoser: with glance, I (personally) don't use buckets
[14:40] <jdstrand> glance index
[14:40] <jdstrand> smoser: what is a bucket?
[14:40] <soren> smoser: Equally odd: grep -ir kernel nova/api/openstack
[14:40] <soren> No output.
[14:41] <soren> WEll, not equally odd. It just sort of takes over the oddness.
[14:41] <smoser> jdstrand, arbitrary string.
[14:42] <smoser> soren, it wouldn't be that odd to me if nova were moving towards "its stupid to know about these things called 'kernels' specifically"
[14:42] <smoser> but i saw a gerrit review recently that made it know *more* about kernels
[14:42] <jdstrand> I think glance may use 'my-glance' as a default
[14:42] <smoser> (so you could boot with a custom kernel command line!)
[14:43] <smoser> glance has no concept of buckets.
[14:43] <smoser> just names.
[14:43] <smoser> that name you give it with 'my-glance' is just a string.
[14:43] <jdstrand> what I mean is that using glance add, I have:
[14:44] <jdstrand> my-glance/ubuntu-12.04-beta1-s...
[14:44] <jdstrand> euca-describe-images has
[14:44] <jdstrand> IMAGEami-00000005None (my-glance/ubuntu-12.04-beta1-server-cloudimg-amd64-disk1.img)availablepublic
[14:45] <smoser> well, you registered the image with 'my-glance/foo'
[14:45] <jdstrand> I'm just more familiar with glance atm, so I am just trying to understand the differences between glance and cloud-publish-...
[14:46] <jdstrand> smoser: while I am downloading the .tar.gz you mentioned, I tried this:
[14:46] <smoser> jdstrand, cloud-publish-tarball just uses the ec2 api for publishing images. and to do that  you ahve to load the image into a s3 bucket, and then register it.
[14:46] <jdstrand> cloud-publish-image -t image x86_64 ./ubuntu-11.10-server-cloudimg-amd64-disk1.img mybucket
[14:46] <smoser> so that is why cloud-publish-tarball wants a 'bucket' .
[14:46] <jdstrand> I see
[14:46] <smoser> so what happens when you do that to openstack is
[14:46] <smoser>  * you upload an image to s3 work-alike
[14:46] <smoser>  * nova pulls it from that location
[14:47] <smoser>  * nova uploads it to glance
[14:47] <smoser> so at this point you should have 3 new images in glance
[14:47] <smoser> a kernel and initrd and image
[14:47]  * jdstrand does not
[14:47] <jdstrand> I only have the old one that was already in there
[14:48] <smoser> that is odd.
[14:48] <jdstrand> euca-describe-images shows two
[14:48] <smoser> maybe i'm wrong, and diablo did something different.
[14:48] <jdstrand> (the old and the new)
[14:48] <smoser> maybe make it public?
[14:48] <smoser> euca-modify-image-attribute -l -a all <image-id>
[14:48] <jdstrand> it is private
[14:48] <smoser> then try looking in glance.
[14:49] <jdstrand> that did it. interestingly, the name in flance is 'None'
[14:49] <jdstrand> s/flance/glance/
[14:50] <jdstrand> http://paste.ubuntu.com/1699383/
[14:50] <jdstrand> I may not care, but it is interesting
[14:51] <jdstrand> $ nova boot --image 6 --flavor 1 testme
[14:51] <jdstrand> Cannot find requested image 6: Kernel not found for image 6. (HTTP 400)
[14:51] <jdstrand> so, cloud-publish-image is out. it'll be a while before I can try cloud-publish-tarball
[14:52] <jdstrand> smoser: thanks for your help
[14:52]  * jdstrand will try the other command later
[14:52] <smoser> jdstrand, you can register a name with cloud-publish-ubuntu if you want
[14:53] <smoser> just pass '--name'
[14:53] <smoser> and newer versions of it (i think) figure out that you probably wanted to do that and just register with the same name as the bucket/image if you dont give it one.
[14:54] <smoser> maybe i'm wrong though.
[14:54] <smoser> anyway.
[14:56]  * jdstrand doesn't have a cloud-publish-ubuntu. tried passing --name to cloud-publish-image, but that didn't work
[14:56] <jdstrand> again, don't think I care
[14:59] <jdstrand> fyi, nova image-meta 7 set "name=foo"
[15:03] <jdstrand> hmm, that only does it in nova, not glance
[15:04] <jdstrand> ok, glance update 7 name=foo
[15:09] <Quest> Pici,  do i need fcgid?
[15:13] <Pici> Quest: I don't know.  You might want to check out the server guide: http://help.ubuntu.com/12.04/serverguide/C/
[15:14] <Quest> ok
[15:18] <Quest> Pici,  whats the difference between documentroot and directory in site file?
[15:19] <Quest> one is with ending / and one without
[15:19] <Pici> Quest: I don't know off the top of my head, if you're looking for apache specific answers, your best bet is to ask in #httpd
[15:32] <zul> adam_g: +1
[15:38] <roaksoax> ScottK: howdy!! I recall you rejected the SRU for python-django (maas related). this was uploaded with ubuntu1.5. Now, since I need to upload a new version dropping GenericIpAddressField, should I do it with 'ubuntu1.6' as a version?
[15:39] <ScottK> roaksoax: No.  If it was rejected, you can reuse the revision number.
[15:39] <roaksoax> ScottK: right, cause I get this: Rejected:
[15:39] <roaksoax> File python-django_1.3.1-4ubuntu1.5.debian.tar.gz already exists in Primary Archive for Ubuntu, but uploaded version has different contents.
[15:40] <ScottK> Odd.
[15:40] <ScottK> OK, then I guess it does need to be 1.6.
[15:40] <ScottK> Make sure someone else didn't do another SRU in the mean time.
[15:40] <roaksoax> ScottK: will do! thanks
[15:40] <ScottK> (maybe a security update)
[15:45] <roaksoax> ScottK: seems the previous upload is still there: https://launchpad.net/ubuntu/+source/python-django/1.3.1-4ubuntu1.5
[15:45] <agu10^> hi
[15:46] <ScottK> Looking
[16:00] <ScottK> roaksoax: It looks like it was accepted and then removed, so you will need to bump the revision.
[16:00] <roaksoax> ScottK: yep, did that already! Thanks!
[16:29] <fahosting> Hello
[16:42] <Quest> virtualhost and proxy for apache with tomcate config aid http://pastebin.ca/2316365   ?
[16:50] <Quest> virtualhost and proxy for apache with tomcate config aid http://pastebin.ca/2316370   ?
[16:51] <RoyK> !patienc
[16:51] <RoyK> !patience
[16:51] <RoyK> !ask | Quest
[16:51] <Quest> RoyK,  thats a question in the paste
[16:53] <Skaag> hey there, tryint to build a new package and getting this:
[16:53] <Skaag> dpkg-source: error: can't build with source format '3.0 (quilt)': no upstream tarball found at ../opentxs_0.88.f.orig.tar.{bz2,gz,lzma,xz}
[16:53] <Skaag> but that file isn't supposed to exist yet
[16:55] <Quest> can anyone answer 3?
[16:55] <Quest> can anyone answer 3. in  http://stackoverflow.com/questions/14990973/tomcat7-and-apache2-on-ubuntu-in-integration-with-some-sort-of-proxy ?
[16:58] <fahosting> using MAAS to build an environment suitable for Hadoop and OSCAR...completely virtualized on KVM with ipmi's enabled to each vm.
[16:59] <fahosting> dev environment and proof of concept using a single machine...I know I know..but as I said it's a proof of concept to microsize the environment.
[17:00] <Skaag> what do you mean ipmi enabled for each vm?
[17:03] <smoser> fahosting, 'virtual maas' https://code.launchpad.net/~virtual-maasers
[17:03] <smoser> is very similar to what (i think) you're describing.
[17:04] <smoser> we use maas to deploy a charm on a hardware system that sets up a maas itself with kvm nodes.
[17:04]  * Quest waits
[17:05] <fahosting> smoser yep exactly
[17:05] <fahosting> I didnt know about this project, very cool
[17:06] <smoser> fahosting, we'd love to have improvements or other users.
[17:07] <smoser> and i personally would'nt mind de-coupling the initial provisioning from a charm.
[17:07] <smoser> ie, make the charm invoke more gneeral purpose stuff than have all the setup in itself.
[17:08] <smoser> fahosting, so the hangup as it is right now is that you have to have maas and juju setup.  then  juju tells maas to deploy a new system running the 'virtual-maas' charm.
[17:09] <fahosting> smoser exactly, the problem I am facing is a very large dev group in a crappy environment. They dont have the 30 physicals to go that route but I can defintly virt the environment for them
[17:09] <fahosting> If this works then great, if not then I will need to explore the RHEL / CentOS clustering and obtain a few low end machines.
[17:10] <fahosting> i.e OSCAR
[17:34] <Skaag> there's an ipmi controller for virtual machines?
[17:34] <Skaag> virtual ipmi?
[18:21] <jdstrand> smoser: well, I used cloud-publish-tarball, made it public then updated the name and still have the same problem: http://paste.ubuntu.com/1700683/
[18:21] <jdstrand> smoser: I can boot the aki kernel though, which is actually enough for my testing
[18:23] <smoser> jdstrand, ok. i'm sorry. i'ms ure there is some way to get what you're wanting.  maybe the issue was 'container_format=ami disk_format=ami'
[18:24] <smoser> container_format=bare disk_format=qcow2
[18:24] <smoser> perhaps.
[18:24] <jdstrand> let me try that
[18:24] <smoser> you might just be getting lucky if it works with 'ami' elsewhere.
[18:24] <jdstrand> smoser: thanks
[18:24] <smoser> jdstrand, i have a question though
[18:24] <jdstrand> k
[18:24] <smoser> given an inline signature
[18:25] <smoser> hold on..
[18:26] <smoser> jdstrand, given http://paste.ubuntu.com/1700723/
[18:27] <smoser> how can i get information about who signed that ?
[18:27] <adam_g> zul: https://code.launchpad.net/~gandelman-a/ubuntu/quantal/cinder/2012.2.3-2/+merge/149749 plz
[18:28] <smoser> i guess ideally, given just the PGP SIGNATURE part, i'd like to know *who* signed it, and what checksum and such is expected.
[18:28] <zul> adam_g: +1
[18:29] <jdstrand> smoser: glance add name="my-glance/$img" is_public=true container_format=bare disk_format=qcow2 < "$img" seemed to work! :)
[18:29]  * jdstrand looks at paste
[18:33] <jdstrand> smoser: otoh (sorta), you can do 'gpg --verify <file>' and if it isn't in your keyring, snag the key ID, then feed that to 'gpg --search-keys'
[18:34] <smoser> jdstrand, right.
[18:34] <smoser> i was just wanting to somehow extract the info without actually attemping to verify.
[18:35] <smoser> ie, i just wanted to know who it came from.
[18:36] <jdstrand> otoh idk, but maybe something like libgpgme11/python-gpgme could help
[18:52] <adam_g> SpamapS: theres also now a rebased cinder package in the quantal queue to include another sec update
[19:00] <Quest> why this config http://pastebin.ca/2316408  for alpinema.com nor alpinema.com/alpinemadotcom is working but only the independant tomcate request as alpinema.com:8080 works?
[19:23] <genii-around> I think you probably need a trailing slash on line 8
[19:42] <agu10^> hi
[19:42] <RoyK> ho
[19:45] <zerick> hi, anybody knows if it is possible - I mean, the normal way, not forcing - to install Varnish 2.1 on Ubuntu 12.04 ?
[20:07] <Angtaga> Hello, somebody successfully installed "OwnCloud"-Server on a Ubuntu server already?
[20:23] <sliddjur> im not sure i understand useradd -p
[20:23] <sliddjur> and man useradd tells me to go to man 3 crypt witch i dont have
[20:24] <adam_g> zul: http://people.canonical.com/~agandelman/folsom/nova_2012.2.1+stable-20121212-a99a802e-0ubuntu1.2~cloud0/
[20:24] <sarnold> sliddjur: crypt(3) is in manpages-dev
[20:24] <zul> adam_g: +1
[20:45] <rockpapergoat> anyone have any tips on mounting kvm guest lvm images on the host? guest won't boot; says there's no device at the UUID, even though it's there. guest/host are ubuntu 10.04lts.
[20:48] <xnox> rockpapergoat: loopmount diskimage, then use kpart -a on it.
[20:49] <xnox> then vgchange -a y
[20:49] <rockpapergoat> xnox: cool. i got it to mount once before but haven't since i unmounted it.
[20:49] <xnox> in my case I use lv volumes as backing store for kvm, so I just run kpartx on the volume.
[20:51] <rockpapergoat> xnox: so i should be able to 'kpartx -av /var/lib/libvirt/images/vm_name.img' before mounting devices under /dev/mapper?
[20:52] <rockpapergoat> xnox: in doing so, i was able to mount the /boot partition but not the guest's lvm volume.
[20:52] <xnox> run $ vgscan
[20:53] <xnox> $ vgchange -a y
[20:53] <rockpapergoat> xnox: after mounting the /boot?
[20:53] <xnox> no need to mount boot, but if you can mount that you should be able to detect lvm2 groups with vgscan
[20:54] <xnox> and then activate them with vgchange -a y
[20:54] <rockpapergoat> xnox: i can mount /dev/mapper/loop1p1 but neither of the other two. vgscan shows no other volume groups.
[20:56] <adam_g> zul:  PITA! https://code.launchpad.net/~gandelman-a/ubuntu/quantal/nova/nova-2012.2.3-2/+merge/149933
[20:56] <xnox> rockpapergoat: how about vgscan /dev/mapper/loop1p*
[20:56] <xnox> ?
[20:56]  * xnox can't quite remember the syntax.
[20:58] <rockpapergoat> xnox: vgscan only picks up the host's group. can't get it to see the guest's group devices under /dev/mapper.
[20:58] <rockpapergoat> xnox: thanks for the help, by the way.
[20:59] <xnox> rockpapergoat: hmm... does it help mounting the partition with the volumegroup (e.g. the p3 i think?)
[20:59] <rockpapergoat> xnox: not sure what you mean there.
[21:00] <rockpapergoat> xnox: i have dev/mapper/loop1p[1,2,5], no 3.
[21:01] <xnox> boot, (vg group), swap
[21:01] <xnox> rockpapergoat: so mount p2 -> try vgscan, mount p5 -> try vgscan.
[21:01] <zul> adam_g: +1
[21:04] <rockpapergoat> xnox: weird. i can't mount p2 or p5. "/dev/mapper/loop1p2 already mounted or /mnt/vm2 busy" even though there's nothing mounted at that mount point.
[21:04] <rockpapergoat> mount /dev/mapper/loop1p5 /mnt/vm3
[21:04] <rockpapergoat> mount: unknown filesystem type 'LVM2_member'
[21:04] <rockpapergoat> xnox:
[21:05] <rockpapergoat> oops… sticky return key
[21:36] <HHT> I'm running 12.04.2 and having problem with netwoking. status always shows "netwoking stop/waiting"
[21:39] <HHT> I have no problem accessing internet or other machines but DLNA devices won't find the server
[22:17] <sunya7a_> hi guys
[22:17] <sunya7a_> is there a live cd for ubuntu server? or is that just crazy speak?
[22:18] <sunya7a_> i'm asking because i'm trying to install ubuntu on a mac os server…and I can only find instructions on how to do it with rEFIt … and those instructions seem to require a Ubuntu Live CD
[22:44] <SpamapS> adam_g: sorry I suck, probably won't be able to re-review those two today. Will try to get to it tomorrow early
[22:46] <adam_g> SpamapS: ah, np. theres also a nova CVE too! so 3 total currently.
[22:46] <SpamapS> bad timing :-P
[22:48] <adam_g> yea
[22:54] <pwarner> how can I get my cloud-precise LXCs to start reliably in precise?
[22:54] <pwarner> about 1/5 of the time networking fails to come up.
[22:54] <pwarner> I get an errors in /var/log/upstart/networking.log "ifup: failed to open statefile /run/network/ifstate: No such file or directory" and similar
[22:54] <pwarner> http://paste.ubuntu.com/1702277/
[22:55] <pwarner> it appears to me like networking is trying to come up in the LXC before /run FS is there, which does not seem to go well