[00:26] <uvirtbot> New bug: #1017681 in mysql-5.5 (main) "package mysql-server-5.5 5.5.24-0ubuntu0.12.04.1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Low,Invalid] https://launchpad.net/bugs/1017681
=== Aaton is now known as Aaton_off
=== kermit1 is now known as kermit
[01:42] <smoser> awk is awesome.
[01:42] <smoser> http://bazaar.launchpad.net/~smoser/cirros/trunk/revision/181
[01:43] <smoser> 7566 bytes to get a json parser that i can then utilize in shell.
[03:36] <uvirtbot> New bug: #1005951 in lxc (universe) "lxc-fedora template is broken with fedora 17" [Low,Triaged] https://launchpad.net/bugs/1005951
[04:31] <josePhoenix> I have a VPS that started life as a much older version of ubuntu, and I've been dutifully dist-upgrading to 12.04
[04:32] <josePhoenix> Somewhere along the way, my locale variables seem to have gotten messed up
[04:33] <josePhoenix> And more recently even dpkg has been acting weird, complaining about version numbers for some package I never installed
[04:39] <josePhoenix> hm managed to sort out the locale thing
[05:35] <nocturnal_> if i install openbox on ubuntu-server will openbox start when my server starts?
=== nijaba_ is now known as nijaba
=== segv_ is now known as segv
[07:36] <uvirtbot> New bug: #958312 in juju "Change zk logging configuration" [Critical,Fix committed] https://launchpad.net/bugs/958312
=== smoser` is now known as smoser
[08:11] <uvirtbot> New bug: #988065 in juju "Support smart formatting for relation-ids command" [High,Fix committed] https://launchpad.net/bugs/988065
[08:11] <uvirtbot> New bug: #993034 in juju "lxc deployed units don't support https APT repositories" [Medium,Fix committed] https://launchpad.net/bugs/993034
[08:24] <koopa58> Hi all, quick question on logwatch. Everyday I get the same 404's in httpd section of the logwatch mail. How can I inspect what IP makes these requests? I just see part of the url.
=== ninjak_ is now known as ninjak
[08:38] <Daviey> jamespage / smoser: do you know what the deal with https://jenkins.qa.ubuntu.com/view/Quantal/view/ISO%20Testing%20Dashboard/job/quantal-server-ec2-daily/ is?
[08:38] <smoser> ugh. looking.
[08:46] <smoser> suck.
[08:46] <smb> Daviey, smoser I told yesterday
[08:46] <smb> Its likely to be fixed in the next upload
[08:46]  * smoser apparently missed some ircs yesterday.
[08:46] <smoser> (missed one with daviey too)
[08:47] <smoser> smb, when is next load? as you seem aware, the failures started with 3.5.0-1-generic
[08:47] <smb> I believe a patch in 3.5-rc4 addresses this. Think it is supposed to be uploaded today or yesterday
[08:48] <smb> Let me check
[08:49] <smoser> 3.5.0-1.1. is archive per https://launchpad.net/ubuntu/+source/linux
[08:49] <smb> smoser, Right, so not yesterday at least
[08:49] <smoser> smb but you thought the upload was to happen before alpha2 ?
[08:50] <smb> smoser, Yes, I heard talk about Leann planning to do one
[08:52] <smb> smoser, I will talk to her when she comes up to find out...
[08:52]  * smb wonders about smoser being around...
[08:52] <smoser> yeah. stupid sleep.
[08:53] <Daviey> smb: ah, you told Mr Page
[08:54] <smb> Daviey, Yeah, he said something about possibly disabling i386 until the next upload. Not sure this is worth doing if the upload happens soon
[08:54] <Daviey> smb: Well.. we don't really want to release A2 with a bad cloud experience
[08:55] <smb> Daviey, Sure. Hm, lemme try to test my assumption then... Hopefully one of the amis still has the old kernel to come up
[08:56] <uvirtbot> New bug: #1017862 in lxc (universe) "Migrate to dpkg --add-architecture to track foreign architecture in template lxc-ubuntu" [Undecided,New] https://launchpad.net/bugs/1017862
[08:57] <th0mz> https://bugs.launchpad.net/~ubuntu-server is this the right place to send bug for ubuntu server 12 please ?
=== almaisan-away is now known as al-maisan
=== railsraider_ is now known as railsraider
=== railsraider__ is now known as railsraider
=== railsraider_ is now known as railsraider
[09:53] <smoser> th0mz, if you can, run 'ubuntu-bug <packagename>'
[09:53] <smoser> it will either open a browser, or give you a url
[09:53] <smoser> then paste the url into a browser
[09:58] <th0mz> thanks smoser
[10:04] <smb> Daviey, smoser OK, so at least I can confirm that the next upload will fix the i386 problem on ec2
[10:04] <smoser> and that upload is due today?
[10:04] <smb> smoser, That one I need to get confirmed
[10:05] <Daviey> smb: thanks
[10:11] <smoser> utlemming, ^ it would appear that alpha-2 images are dependent upon the next kernel upload.
[10:29] <lynxman> jamespage: bug triaging mysql-server-5.5 bugs on the incoming queue
[10:36] <Daviey> lynxman: \o/
[10:36] <lynxman> Daviey: most of them are trivial :)
[10:46] <davepigott> Hi Everyone. I'm trying to deploy a private cloud using MAAS and JuJu on Precise, but (a) I only have 5 machines and (b) I don't want the MAAS node to control DHCP just yet. The guides say I need at least 10 machines, and the MAAS guide only documents allowing MAAS to be the DHCP controller, although it says it deals with both scenarios. Is there any documentation I can get which will help me out?
[10:49] <Daviey> davepigott: What is currently providing dhcp?
[10:49] <davepigott> Daviey: A Netgear router
[10:49] <davepigott> Daviey: I'm happy to change it if necessary
[10:49] <Daviey> davepigott: Hmm.. Does that allow you to set a boot server?
[10:49] <davepigott> Daviey: We're getting a leased line soon with a proper Cisco
[10:49] <davepigott> Daviey: Let me check...
[10:49] <Daviey> is it an el-cheapo home thing, or something decent ?
[10:51] <davepigott> Daviey: It's a DGN2200 - i.e. el cheapo
[10:53] <davepigott> Daviey: Looks like it probably doesn't - can't find a config entry. Just checking docs
[10:59] <maedox> If I were to add a line to fstab for an EC2 EBS volume, then detach it and reboot, would the server still boot properly, or would I have to add nofail as an option?
[11:01] <maedox> ah, found what I was looking for; nobootwait
=== matsubara is now known as matsubara-lunch
[11:18] <jamespage> lynxman, much appreciated!
[11:26] <jamespage> rbasak, re bug  1013203 - welcome to the hell that is Java native integration!
[11:26] <uvirtbot> Launchpad bug 1013203 in rampart "Shared library path problem" [Undecided,Invalid] https://launchpad.net/bugs/1013203
[11:27] <rbasak> jamespage: what's the solution? A dependency on librampart-dev? :)
[11:28] <jamespage> rbasak, no
[11:28] <Daviey> oh dear, not rampart
[11:28] <jamespage> rbasak, the issue is that if you use the loadLibrary calls provided by the Java Runtime Environment they require unversioned .so
[11:29] <jamespage> rbasak, there are libraries that help with this - I forget the name but jenkins uses one extensively
[11:29] <rbasak> jamespage: LD_LIBRARY_PATH to somewhere custom for java broken stuff, with a symlink to the right place? :)
[11:29] <jamespage> rbasak, spot on - so this is really a bug in eucalyptus which should be doing that
[11:29] <jamespage> rbasak, I had to implement  a similar hack for the snappy compression in my hadoop packages....
[11:30] <jamespage> rbasak, I'm having trouble believing that no-one has hit that issue before tho
[11:30] <Daviey> rbasak: I'm not saying it's right, but we did maintain path compatible symlinks and hardcoding version numbers in euca.
[11:30] <Daviey> and SONAME versions
[11:32] <rbasak> Daviey: that's fine as long as the brokenness is limited to the broken packages, IMHO :)
=== al-maisan is now known as almaisan-away
[11:38] <Daviey> rbasak: traditionally, it was not :)
[11:44] <rbasak> Daviey: you mean in Ubuntu as opposed to Debian? Or both?
[11:44] <rbasak> Daviey: if we're going to be dirty, surely the cleanest way would be to just depend on librampart-dev?
=== cedr is now known as zombiecedr
[11:54] <Daviey> rbasak: in Ubuntu
[11:55] <Daviey> rbasak: rampart was only added to Ubuntu, to support eucalyptus.
[11:57] <rbasak> It's in debian now, and in universe in ubuntu now
[11:57] <rbasak> Looks like quite a bit of the delta can be dropped next merge
[11:57] <Daviey> rbasak: It was added to Debian for the same reason it was added in Ubuntu.
[11:57] <Azjo> Hi, does anyone have experience with lenovo d20?? it seems to be a bitch to install anything onto when having RAID on.
[11:58] <rbasak> Daviey: so you want to fix this in rampart?
[11:58] <Daviey> rbasak: No, i'm just saying that is what used to happen.
[11:58] <Daviey> rbasak: now, i'm saying i don't care. :)
=== th0mz_ is now known as th0mz
=== matsubara-lunch is now known as matsubara
=== mrmist_ is now known as mrmist
=== zyga is now known as zyga-afk
[12:21] <uvirtbot> New bug: #1017911 in lm-sensors (main) "package libsensors4 1:3.3.1-2ubuntu1 failed to install/upgrade: conffile './etc/sensors.d/.placeholder' is not in sync with other instances of the same package" [Undecided,New] https://launchpad.net/bugs/1017911
=== th0mz_ is now known as th0mz
[12:27] <koolhead11> I want to help with this https://blueprints.launchpad.net/ubuntu/+spec/servercloud-q-openstack-ha
=== th0mz_ is now known as th0mz
[12:38] <smb> Daviey, is there a bug report about the ec2 i386 boot failing in q?
[12:39] <jamespage> smb, I'm not aware of one yet....
[12:41] <koolhead11> Is this https://wiki.ubuntu.com/ClusterStack still in? Looks not updated
[12:41] <zul> Daviey: can you review swiftclient and swift-plugin-s3 please?
[12:43] <jamespage> Ursinha, around? I think that as a team we have generally agreed on the approach to bug triage now
[12:43] <Daviey> zul: yes, i have a call start in 15.. After that, i will.
[12:44] <jamespage> Ursinha, I'd like to get the triage report updated to support the revised process...
=== zyga-afk is now known as zyga
[12:44] <zul> Daviey: cool thanks
[12:44] <smb> jamespage, So I told Leann about the problem and she will try to convince the SRU team to get todays upload accepted. There may be need of speaking in favour (wherever and in whatever form that may be)
[12:45] <zul> smb: whats the number for the xen merge?
[12:45] <smb> zul, https://bugs.launchpad.net/ubuntu/+source/xen/+bug/1013088 though the title is a bit out of date now
[12:46] <uvirtbot> Launchpad bug 1013088 in xen "Please merge xen 4.1.2-7" [Wishlist,In progress]
[12:46] <Daviey> smb: It's not SRU, it's Release Team.. And i'm asking for it :)
[12:46] <smb> Daviey, Meh, did too much stable work in the past. :)
[12:46] <zul> smb: whats the problem with it again?
[12:46] <smb> Daviey, Note it is not yet uploaded. There were some patches promised that arriced today
[12:47] <Daviey> smb: Well, i'll respin all flavours when it lands..
[12:47] <smb> zul, The questions I have/had are in the report and basically most of the delta I would have introduced is now away with the newer Debian merge
[12:48] <zul> smb: okies so besides the  conflict and replace stuff we should just sync
[12:48] <smb> zul, Oh, if you were referring to the things hidden for you to find. Those were referring to two messups in the changelog, the v2 debdiff should be ok
[12:49] <zul> smb: ok cool
[12:49] <smb> zul, not completely
[12:49] <smb> zul, There is the pxe stuff were we deviate a bit more since debian has a qemu-ipxe with differently named roms to depend on
[12:50] <zul> yeah thats fine
[12:51] <smb> zul, I think the part I am mostly unsure is the additional replaces and conflicts for xenlib-3 which my feelign is that it was only needed as long as the supported upgrade paths would have it
[12:51] <smb> And since both precise and precise are xen-4...
[12:51] <zul> smb: ill double check before i upload it but we should be good
[12:57] <smb> zul, Cool, yeah it is rather a cleanup issue and should certainly make no difference in usage. The latest Debian version number is a bit of a pain though. ;)
[12:57] <zul> yah icky poo
=== zyga is now known as zyga-food
=== benji___ is now known as benji
=== zyga-food is now known as zyga
[13:24] <sonne> greetings!
[13:25] <matti> Hi there sonne
[13:26] <sonne> what is the virtualization technology that best suits ubuntu in your opinions?
[13:26] <sonne> that is, the one that i can expect to be most integrated and supported for ubuntu-based infrastructures
[13:27] <RoyK> !kvm
[13:27] <ubottu> kvm is the preferred virtualization approach in Ubuntu. For more information see https://help.ubuntu.com/community/KVM
[13:30] <matti> +1 for KVM
[13:31] <sonne> am i wrong or kvm doesn't support paravirt?
[13:31] <matti> It does.
[13:32] <matti> sonne: http://www.linux-kvm.org/page/FAQ
[13:34] <tash> any exim4 users in here? I don't recall ever having such a problem, but, I just installed exim4 and when trying to send mail I get unroutable address errors
[13:34] <tash> not sure what I should change.
[13:34] <sonne> matti, your link says the opposite actually
[13:35] <matti> sonne: Depends what you are after.
[13:35] <matti> sonne: KVM has PV drivers support.
[13:36] <sonne> but that's a full virtualization that "becomes para" after boot using certain drivers
[13:36] <matti> sonne: If you want to share CPU / CPU time with PV guest, then KVM cannot do it.
[13:36] <sonne> it's not really the same
[13:36] <matti> No, its not. True.
[13:37] <matti> sonne: I personally moved away from Xen and it made me the happiest person ever :)
[13:37] <sonne> anyway this isn't a which-one-is-better discussion nor i want it to be
[13:37] <matti> sonne: Sure, no worries.
[13:37] <matti> sonne: Nobody is having one :)
[13:37] <sonne> i know, it's just so easy to degenerate into one when arguing about these things ;)
[13:37] <matti> sonne: In Debian / Ubuntu with sanely recent kernel KVM just works.
[13:38] <matti> sonne: I use it at home on desktop too, with IO MMU to access my GPU directly, so I can play games :)
[13:38] <matti> sonne: Works like a charm :)
[13:40] <sonne> matti, i'm coming from a full XCP based infrastructure, so there are things i'm used to that i wouldn't like to loose
[13:41] <sonne> such as cpu sharing, live migrations, live disk growing/shrinking, live cpu/memory increasing/decreasing
[13:41] <sonne> last time i tried KVM it was on a redhat, and i could hardly do any of those - but maybe things have changed... do you have experience about these matters?
=== tsimpson_ is now known as tsimpson
[13:45] <RoyK> sonne: live migration works
[13:45] <RoyK> not the others
[13:47] <sonne> RoyK, cheers
[13:53] <FavioEspinoza> How do I tell the version of SASL I have installed?
[13:53] <sonne> FavioEspinoza, i'd go for: dpkg -l | grep sasl
[13:54] <glance> dpkg -l \*sasl\*
[13:55] <FavioEspinoza> sonne, glance : Thanks. :)
[13:55] <glance> mine was shorter =)
[13:56] <sonne> so has anyone tried the xcp toolstack on ubuntu?
[13:56] <sonne> is it any good?
=== manjo` is now known as manjo
[14:01] <uvirtbot> New bug: #1017847 in qemu-linaro (universe) "qemu segfaults when creating an armhf container on an amd64 host" [Undecided,New] https://launchpad.net/bugs/1017847
[14:13] <Daviey> smoser: Are you going to do, "Add support for ports.ubuntu.com to cloud-init: TODO"?
[14:14] <smoser> Daviey, well, one way or another we have to do that.
[14:14] <Daviey> smoser: Yes, but i'm wondering who i need to slap to get it done? :)
[14:16] <uvirtbot> New bug: #1013171 in mythtv (main) "Many package hooks not ported to python3" [Undecided,Confirmed] https://launchpad.net/bugs/1013171
[14:17] <smoser> Daviey, i think you're probably right. the best solution is to get arm to archive.ubuntu.com
[14:18] <jpds> Oh dear.
[14:25] <stgraber> hallyn: I'm preparing a quick fix for bug 1017862, so wait for it before pushing a new lxc with the hooks changes
[14:25] <uvirtbot> Launchpad bug 1017862 in lxc "Migrate to dpkg --add-architecture to track foreign architecture in template lxc-ubuntu" [Undecided,Confirmed] https://launchpad.net/bugs/1017862
[14:27] <hallyn> stgraber: ok
[14:28] <ironhalik> Hello guys
[14:29] <ironhalik> I'm testing Ubuntu server 12.04 in virtualpc as an alternative to my debian server and got a small problem - when I try connecting via ssh to it, it takes a good minute to show me the password prompt
[14:30] <ironhalik> in the exact same conditions in virtual pc and bridget network adapter, debian does it instantaneously
[14:31] <Daviey> ironhalik: probably a dns issue
[14:32] <ironhalik> Im doing it on my local network, ssh'ing to IP, not domain
[14:34] <ironhalik> also, it does it on both openssh server and dropbear
[14:40] <FavioEspinoza> (in ubuntu 12.04) I've installed mail-stack-delivery; (solely); and test that works fine with local domain. From this, I want to configure it with Postgresql for virtual domains. Could you give me some links, besides Ubuntu Server Guide ?
[14:44] <newz2000> Hi, I've got a windows 7 guest in kvm that I use for testing. I've tried to open virsh and run stop win7 but nothing happens
[14:44] <newz2000> is there a trick to get it to shutdown?
[14:44] <newz2000> I know my ubuntu desktop guests sometimes have the same prob unless I enable apmd
[14:45] <glance> stop means: send acpi power button.
[14:45] <glance> the os in the vm needs to understand acpi to be able to do a shutdown
[14:46] <newz2000> glance: cool, that's what I suspected. Anyone know if there's a trick to making Windows understand?
[14:47] <stgraber> hallyn: lp:~ubuntu-lxc/ubuntu/quantal/lxc/bug-1017862
[14:47] <newz2000> ah, may have found it thanks to you helping me get the right terminology
[14:47] <stgraber> hallyn: I'm rebuilding some containers now to test, but I don't see why it wouldn't work
[14:48] <newz2000> hey stgraber, I was doing a lot of juju stuff last week and your blog was an awesome help, thanks!
[14:49] <Daviey> ironhalik: the fact it's local network doesn't matter.. my bet is still a dns issue
[14:50] <ironhalik> Daviey: yeah, google says so too :) Looking into it
[14:51] <stgraber> newz2000: cool! glad it helped
[14:51] <newz2000> me too!
[14:51] <Daviey> ironhalik: your resolvers are probably unreachable.
[14:52] <smoser> stgraber, can you tell me without me having to figure it out myself how i make pastebinit work with paste.openstack.org ?
[14:53] <ikonia> Daviey: is the default 12.04 server setup using dnsmasq like the desktop or is a more regular setup ?
[14:55] <Daviey> smoser: $ echo test | pastebinit
[14:55] <Daviey> http://paste.openstack.org/show/18803
[14:55] <smoser> $ echo test | pastebinit
[14:55] <smoser> http://paste.ubuntu.com/1060869/
[14:55] <Daviey> http://pb.daviey.com/YBqp/
[14:56] <smoser> gracias, Daviey
[14:56] <ironhalik> Daviey: You're right - adding 'UseDNS no' to sshd is a working quick fix - thanks
[14:57] <Daviey> ikonia: same as desktop
[14:57] <ikonia> thanks
[14:57] <stgraber> smoser: you can change the target temporarily by doing "echo test | pastebinit -b http://paste.openstack.org"
[14:58] <smoser> right. /me is embarrased. that was too easy.  i assumed somehow i had to tell pastebinit "paste.openstack.org" is a lodgeit based pastebin
[14:58] <smoser> how did it know that?
[15:26] <hallyn> stgraber: running some tests as well against your tree + my hooks.  i may push the combination to q-proposed later today
[15:28] <stgraber> hallyn: cool. I guess you'll have a changelog conflict when landing both branches into ubuntu:lxc, but it should be easy to sort out
[15:30] <hallyn> you mean just between our two branches, or something more because of your ubuntu:lxc cleanup?
[15:32] <Ursinha> jamespage, okay, I'll finish one other task I have today and will look at the report to update it..
[15:33] <stgraber> hallyn: nope, just our two branches as we both add the same changelog entry, but that's easy to fix :)
[15:43] <hallyn> stgraber: yeah i renamed my patch.  btw do you mind if i add your patch name to your changelog blurb?
[15:43] <stgraber> hallyn: no problem with that
[15:44] <hallyn> (thanks, will help at next merge)
[15:44] <hallyn> ok, everyone, i'm looking at merging the new libcgroup from debian
[15:44] <hallyn> it will drop the automatic running of deamons and setup of mounts at boot
[15:45] <hallyn> how badly will that mess people up?
[15:45] <hallyn> maybe i should send an email to ubuntu-server about it
[15:46] <jamespage> Ursinha, great - I updated the triage process in the wiki with the experimental server-notriage tag information
[15:47] <stgraber> hallyn: I see that as a good thing really ;) it's causing to fix a lot of weird bugs (most of which we haven't even tied to libcgroup yet)
[15:48] <Ursinha> jamespage, do you have that page url so I can bookmark it? :)
[15:49] <hallyn> stgraber: yeah, but i'm worried since even kees was apparently using the libcgroup classification stuff :)
[15:49] <jamespage> Ursinha, https://wiki.ubuntu.com/ServerTeam/KnowledgeBase/BugTriage
[15:49] <hallyn> anyway, all its tools are broken now without manual setup of cgroups, i need to email jbernard and talk more
[15:53] <jamespage> hallyn, pre-empting my non-action from last weeks team meeting
[15:53] <jamespage> the check from the servercloud-q-server-iso-tests-review blueprint was related to checking the functionality of a VM post host upgrade
[15:53] <jamespage> I think
[15:53] <jamespage> i.e. if we upgrade precise->quantal do VM's keep functioning
[15:55] <keyword1> hi guys I have a question I have installed ubuntu server I installed X so when i run export display=:0 && xterm
[15:55] <keyword1> xterm Xt error: Can't open display:
[15:55] <keyword1> xterm:  DISPLAY is not set
[15:58] <hallyn> jamespage: sorry i'm in the middle of an email and can't seem to process your sentence
[15:59] <hallyn> jamespage: oh i see waht your'e talking about (not what you're saying)
[15:59] <hallyn> jamespage: can you just send me an email after the meeting?
[15:59] <jamespage> hallyn, sure
[15:59] <jamespage> I might have actually figures out what I'm trying to say then...
=== pleia2_ is now known as pleia2
=== matsubara is now known as matsubara-afk
[16:11] <uvirtbot> New bug: #1018001 in rabbitmq-server (main) "Please sync rabbitmq-server 2.8.4 from Debian" [Undecided,New] https://launchpad.net/bugs/1018001
[16:21] <imanc> can anyone recommend a decent and easier to use IPtables interface?
=== jamespage is now known as hazmat_mk2
=== hazmat_mk2 is now known as jamespage
[16:30] <hallyn> imanc: ufw?
[16:30] <RoyK> !ufw
[16:30] <ubottu> Ubuntu, like any other Linux distribution, has firewall capabilities built-in. The firewall is managed using the 'ufw' command - see https://help.ubuntu.com/community/UFW | An alternative to ufw is the 'iptables' command - See https://help.ubuntu.com/community/IptablesHowTo | GUI frontends such as Gufw (GNOME) and Guarddog (KDE from Lucid onwards) also exist.
=== oCean_ is now known as oCean
[16:34] <tash> I'm mounting a Windows share and I need my apache2 user, www-data, to be able to mkdir on the share. Any idea how to accomplish this? My mount is currently using the Administrator usernamed on the windows server.
[16:34] <imanc> thanks guys
[17:01] <uvirtbot> New bug: #1018024 in apache2 (main) "Apache Common and Combined definition Vs. documentation disparity" [Undecided,New] https://launchpad.net/bugs/1018024
[17:04] <hallyn> stgraber: all my tests pass...  so what, i just target the changelog at quantal-proposed and dput?
[17:05] <hallyn> oh wait, i wanted to play ab it more with cgroup premounts first anyway
[17:11] <stgraber> hallyn: yep, just change the target to quantal-proposed and dput
[17:13] <hallyn> just making sure :) thx
[17:13] <hallyn> i'm considering putting the mount script for doing cgroup premounting into /usr/share/doc/lxc/...
[17:17] <gary_poster> hallyn, hi.  two quick questions, hopefully.  is there a way to identify that you are running inside an lxc container?  Relatedly, is there a way to get information about your host, such as the IP address?  These are low-priority questions, fwiw.
[17:17] <hallyn> gary_poster: lxc-ps -n $container -- ef
[17:18] <matti> gary_poster: LXC-wise, there is.
[17:18] <matti> gary_poster: Have a look at https://github.com/kwilczynski/facter-facts/blob/master/lxc.rb -- some detection vectors there.
[17:18] <hallyn> gary_poster: what do you mean by 'get information about your host' ?
[17:18] <hallyn> you mean for the container to find out the host's ip address?
[17:19] <hallyn> if so, i think the best bet is to use cloud-init.  whatever ways might exist now, we'd want to patch them up eventually :)
[17:19] <gary_poster> hallyn, yes
[17:19] <hallyn> gary_poster: is the ubuntu-cloud template something you coudl use?
[17:19] <hallyn> (you can pass it user-data files, thanks to utlemming )
[17:20] <gary_poster> matti, interesting reading, thanks; will look more later
[17:21] <hallyn> gary_poster: d'oh, i misread yoru first q
[17:21] <hallyn> gary_poster: see 'running-in-container'
[17:21] <gary_poster> hallyn, ah, nice!
[17:22] <gary_poster> not in lucid :-P :-)
=== Lcawte|Away is now known as Lcawte
[17:22] <hallyn> gary_poster: hm, i thought i had that in the ubuntu-virt ppa as part of lxcguest
[17:23] <hallyn> gary_poster: 'lxc-is-container' ?
[17:26] <gary_poster> hallyn, I've got lxcguest 0.7.5-0ubuntu4.10.04.4 in lucid .  lxc-is-container exists in the lucid container, and exit code is 0.  Does not exist in Precise host.  Doing a google search to try and find expected usage...
[17:27] <gary_poster> hallyn, so it looks like the intent is that if the command exists, lxcguest is installed, which means you are presumably in an lxc container?
[17:28] <ironhalik> wow, installing ubuntu server on 128 megs of ram is painful :>
[17:28] <gary_poster> heh
[17:28] <hallyn> gary_poster: not quite,
[17:29] <hallyn> gary_poster: it doesn't always return 0 just bclxcguest is installed
[17:29] <hallyn> it looks to see if init has the 'container=lxc' env variable
[17:29] <hallyn> unfortunately it got renamed for precise, when it moved from lxcguest into upstart
[17:29] <hallyn> but, the logic would basically be:
[17:30] <hallyn> 1. if /bin/running-in-container is present (precise and above, always), run it and check for 0 return value
[17:30] <hallyn> 2. else, if lxc-is-container is not present, assume lxcguest is not installed and you're not in a container (or are in trimmed container)
[17:30] <hallyn> 3. else, run lxc-is-container, if 0, you're in a container, if 1 you're not
[17:31] <gary_poster> got it, thanks very much hallyn!
[17:31] <hallyn> gary_poster: in particular, the point of lxcguest was that you can boot a system with lxcguest installed as bare metal
[17:31] <gary_poster> ah
[17:31] <gary_poster> cool
[17:31] <hallyn> np, hope it suffices!
[17:31] <gary_poster> absolutely
[17:49] <pdtpatrick> Question - i've installed syslinux - however pxelinux.0 is  0KB - is that normal? if not can someone share theirs ?
[17:50] <ironhalik> is it possible to set up a landscape server of my own?
[17:50] <ironhalik> or is it 100% proprietary?
[17:51] <koolhead17> ironhalik, i think its 100% enterprise
[17:52] <patdk-wk> yep, it's all closed source
[17:53] <ironhalik> shame - it looks like a really nice tool, just not worth the price for my small server
=== zyga is now known as zyga-afk
[18:32] <travisneids> I'm connecting to a VPN with "pon vpnname nodetach" and it looks like its connected fine however I can not ping the shared drive i'm trying to get access to.  I can VPN with my Mac OS and ping the IP.  Am I missing something?  Ubuntu 10.04.4 LTS
[18:34] <hallyn> stgraber: do you think /usr/share/doc/lxc/examples/cgroups-mounthook would be a good place to ship a mount hook for pre-mounting cgroups?
[18:35] <dreamer__> Hello, i'm looking to setup a home network with Ubuntu server 12.04 LTS, I want clients, webserver and a fileserver connected to my main server wich will host routing dhcp and dns. Is this doable
[18:35] <dreamer__> ?
[18:35] <stgraber> hallyn: yep, sounds good
[18:36] <stgraber> hallyn: just wondering, what's your plan for cases where you want to call more than one of these? does the syntax allow duplicate lines or should the user write a script?
[18:36] <uvirtbot> New bug: #1018052 in quagga (main) "quagga security issue CVE-2012-1820" [Undecided,New] https://launchpad.net/bugs/1018052
[18:36] <hallyn> stgraber: http://paste.ubuntu.com/1061263/ looks good
[18:36] <hallyn> stgraber: yeah, you can have as many as you want
[18:36] <hallyn> uh, in theory
[18:36] <stgraber> hallyn: (thinking of a case wher you want an ephemeral container with cgroups for example (once we move ephemeral to using the hooks))
[18:36] <hallyn> perhaps i oughta check :)
[18:37] <stgraber> cool, supporting multiple lxc.hook entries will work fine
[18:37] <TheLordOfTime> hallyn:  wouldnt hurt ;P
[19:01] <travisneids> How about setting the PPTP gateway client side?
[19:57] <K4k> By installing with a ks.cfg file and specifying "url --url http://www.server.com/ubuntu" would the installer be using that url only for the additional packages in %packages or for all files during install?
[20:16] <uvirtbot> New bug: #1018094 in bind9 (main) "failed with nslookup for www.google.com, others domain names are fine" [Undecided,New] https://launchpad.net/bugs/1018094
[20:18] <Pici> odd bug.
[20:26] <Sahua> i have an odd issue - i can ssh into my ubuntu server just fine, but i can't connect to anything from there. ping google.com shows unknown host. running 12.04 LTS - any thoughts?
[20:27] <pretec> Sahua: look at your /etc/resolv.conf maybe the nameserver configuration is wrong
[20:27] <ironhalik> have you tried pinging an IP address? 8.8.8.8 are googles DNS servers
[20:27] <Sahua> oh hey, 8.8.8.8 works just fine
[20:28] <Sahua> so a DNS issue? that'd makes a bit more sense
[20:29] <ironhalik> yeah, youll need to add DNS entries - Googles (8.8.8.8 and 8.8.4.4 as backup) or some of your own choosing
[20:32] <guntbert> ironhalik: Sahua: I recommend not to use 8.8.8.8 - we know they keep everything :-)
[20:32] <ironhalik> Just because you're paranoid does not mean they're not out there to get you.
[20:32] <ironhalik> ;>
[20:34] <Sahua> haha :p i'll at least give it a shot
[20:36] <ironhalik> Sahua: btw, man resolvconf
[20:36] <ironhalik> to change the nameservers
[20:36] <Sahua> yeah, took me a bit
[20:36] <uvirtbot> New bug: #1014657 in python-glanceclient (main) "[MIR] python-glanceclient" [High,Fix released] https://launchpad.net/bugs/1014657
[20:36] <Sahua> thanks - 8.8.8.8 is working
[20:37] <Sahua> appreciate the help!
[20:38] <ironhalik> sure
[20:52] <Daviey> SpamapS: Saw this, thought it might interest you. http://dom.as/2012/06/26/memsql-rage/
[21:09] <glance> hmm... ec2 roles and a modern boto to be able to do #include straight off s3 in cloud-init scripts... that would be nice
=== James__ is now known as JPP
[21:35] <Daviey> hallyn: traditionally LVM was significantly faster image storage for VM's than raw or qcow.. I understand that virtio helped close this gap, but do you have feelings on the current state?
[21:35] <ikonia> Daviey: I can't see how either could be close to lvm's performance at this time from my usage standpoint
[21:36] <Daviey> ikonia: have any metrics ?
[21:37] <ikonia> Daviey: nothing to hand that's recent (at home at the moment) but using lvm devices or file systems on lvm, always produced visable performance improvment, so never bothered actually getting beyond that as it was always visible
[21:37] <uvirtbot> New bug: #1006547 in vlan (main) "vlan doesn't work properly on 12.04 server " [Low,Invalid] https://launchpad.net/bugs/1006547
[21:37] <ikonia> on database the IO access to qcow devices (non-root disk) was always much much lower, I may have some high level stats in the office
[21:45] <Daviey> ikonia: Yeah, i just wanted to check that it was still accurate, that there is a significant gap.
[21:46] <ikonia> Daviey: well I can't go as far as 12.04 sadly as it's not being used as a host, but I'd be happy to bench it against a 10.04 box on identical hardware
[21:46] <Daviey> ikonia: If it's not too much of a pain, i would be interested in 10.04 stats aswell.. thanks
[21:47] <ikonia> Daviey: I can set aside a few hours tomorrow I think, I've got some spare HP hardware.
[21:47] <rbasak> What's with uvirtbot announcing "new" bugs that are days old?
[21:49] <Daviey> rbasak: I noticed that aswell.. uvirtbot uses email as a notifier.. I'm betting that the header parser needs updating or something
[21:49] <rbasak> Daviey: or perhaps there's a massive backlog?
[21:50] <rbasak> Daviey: oh I see. It's seeing bug traffic as new bugs even though the bugs already existed
[21:51] <Daviey> rbasak: My hunch is the X-Launchpad headers have changed or something, and it thinks it is seeing new bugs, yes.
[21:51] <Daviey> Just a hunch.
[21:52]  * rbasak would like to file a new bug against uvirtbot
[21:54] <Daviey> rbasak: bugs should be raised via the irc interface, /msg soren Please fix it.
[21:54] <rbasak> :)
[21:54] <Daviey> it also has an email interface, soren@ubuntu.com
[21:55] <rbasak> Perhaps LP should allow the filing of bugs against users? :-P
[21:55] <Daviey> rbasak: did you just sign up to take ovr maintainership?
[21:55] <rbasak> Daviey: actually I'm fine with that. I already run another private IRC bot anyway
[21:55] <rbasak> (also based on supybot)
[21:56] <Daviey> super!
[21:57] <rbasak> Assuming soren doesn't want to run it any more of course
[22:01] <Daviey> He hasn't commented either way, but i'm sure he wouldn't turn down help.
[22:02] <methods> how does the mysql upstart job know to run as the mysql user ? i don't see a setuid or setgid in the upstart config..
[22:11] <Daviey> utlemming: is the cron job for cloud image building still active ?
[22:11] <Daviey> ie, will there be a new cloud image when i get out of bed?
[22:12] <utlemming> Daviey: yes sir
[22:12] <utlemming> Daviey: there is also a job waiting on the appearance of the new kernel too
[22:12] <utlemming> so you'll likely get two shiny and new images
[22:12] <Daviey> utlemming: ah, that is the one i want..
[22:13] <Daviey> utlemming: new kernel will be published in ~2 hours.
[22:13] <Daviey> utlemming: probably not worth building one, then building another 6 hours later.. might want to flip to manual, if you have a task waiting on that
[22:14] <utlemming> Daviey: right, especially so soon after
[22:14]  * utlemming switches automatic off
=== Lcawte is now known as Lcawte|Away
[22:30] <SpamapS> Daviey: indeed, thats a great post, and domas is basically "the man" :)
[22:46] <uvirtbot> New bug: #1018171 in apache2 (main) "apache2 init script report start success too early" [Undecided,New] https://launchpad.net/bugs/1018171
=== Skaag_ is now known as Skaag
=== Aaton is now known as Aaton_off