[03:43] <fang0654> anyone in here have any experience with grub2, raid10, and gpt?
[03:44] <fang0654> I keep getting an error from grub2 saying "error: device not found" (or something similar)
[03:44] <fang0654> if I do an ls in the grub-rescue, I don't see the mdraid drives
[03:50] <Zanzacar> I was wondering why does no one host a web applet that allows you to send magic/wol packets?
[03:58] <Zanzacar> I would think someone would host or having something out there and I have a feeling I just havent found it. Basically I dont want my friends to have to isntall wake-on-lan software to wake up my server
[04:45] <Psi-Jack> Curious.
[04:45] <Psi-Jack> Does anyone here use Ubuntu 10.04 under VMWare vSphere? I'm curious exactly about the vmware-tools repos for ubuntu and if version matters so much.
[04:45] <Psi-Jack> As in, using latest.
[05:21] <macnix> I ran an upgrade after adding a dotdeb source, can't figure out how to solve this mysql conflict https://gist.github.com/1598925
[05:21] <macnix> google wasn't of much help...
[05:24] <twb> What is dotdeb?
[05:24] <twb> trying to overwrite '/usr/bin/mysql', which is also in package mysql-client-core-5.1 0:5.1.41-3ubuntu12.10
[05:25] <twb> macnix: you need to remove mysql-client-core-5.1 before yo uinstall mysql-client-5.1 because the two repos have broken the mysql source package into different binary packages
[05:25] <macnix> ok, managed to delete the new apt sources, apt-get update, then finished the install -f
[05:25] <twb> http://www.dotdeb.org/ ugh, what a crock
[05:25] <macnix> now removing the problematic packages manually
[05:26] <macnix> well, he has some newer packages
[05:26] <macnix> couldn't find them anywhere else
[05:26] <twb> Use the flipping packages Ubuntu provides in your release instead of grabbing random third-party shite like some RHEL weenie
[05:26] <twb> macnix: and why do you think you need a newer release?
[05:27] <twb> Further, dotdeb packaging appears to be for Debian systems, and results on Ubuntu will be undefined.
[05:28] <macnix> some stuff I compile, but other stuff just takes too long. dotdeb is a good interim solution.
[05:28] <macnix> don't have time to build all the latest packages myself.
[05:28] <twb> New release = new bugs
[05:28] <twb> Especially if you compile it yourself or take some random third-party package.
[05:29] <macnix> yes, I agree, unless you go for specific, newer versions which have proven to be good
[05:29] <twb> Just because upstream says it works doesn't mean it actually does work.
[05:30] <twb> Unless you have specific feature or bugfix that you absolutely must have, DO NOT upgrade away from the packages provided by your release.
[05:31] <macnix> cheers for the wise words ;). I will re-consider my approach.
[05:31] <twb> At this point your system is probably buggered enough that you should consider reinstalling from scratch.
[05:43] <thermionix> hello folks, anyone know what the future holds for NFS with an alternative authentication system that isn't kerberos?
[05:50] <twb> thermionix: zero
[05:50] <twb> Kerberos is the only authentication mechanism that's actually secure
[05:51] <twb> (Well, except for single-host auth infrastructure, which is irrelevant for NFS)
[08:00] <koolhead11> hi all
[08:00] <koolhead11> .
[08:02] <smoser> adam_g, http://paste.ubuntu.com/801476/ that is what i use for thinkpad battery stuff. see usage.
[08:23] <smoser> zul, bug 904532
[08:31] <rbasak> zul: https://bugs.launchpad.net/ubuntu/+source/nova/+bug/873243
[08:38] <rbasak> hallyn: the test still breaks :(
[08:38] <rbasak> hallyn: lxcguest does not exist in lucid
[08:39] <hallyn> rbasak: ah, that's right - there's an open bug on that.  we get it from ppa, but ppa's don't compile for arm.
[08:39] <hallyn> rbasak: can you just edit the test and take out lucid?
[08:40] <rbasak> hallyn: ok, rerunning
[08:41] <hallyn> maybe this should server as motivation for me to work harder on that bug (lxcguest backport)
[09:01] <stgraber> Daviey: hey there. Is https://blueprints.launchpad.net/ubuntu/+spec/foundations-p-ipv6 on your todo? Would be great to have isc-dhcp-server updated (probably moving from one sysvinit job to two upstart jobs, one for ipv4 and one for ipv6) soonish so we can get extensive testing.
[09:02] <stgraber> Daviey: if your team is too busy, I can probably move that to my own todo and try to find some time to do it (instead of copy/pasting init scripts on my servers :))
[09:10] <rbasak> hallyn: failure with set -x: http://paste.ubuntu.com/801518/
[09:11] <rbasak> hallyn: lxc-ps -n oneiric-test -- -ef returns nothing
[09:12] <hallyn> rbasak: all right all right - what does lxc-ls show?
[09:12] <rbasak> root@panda-test:~# lxc-ls
[09:12] <rbasak> oneiric-test  test_container
[09:12] <rbasak> oneiric-test
[09:12] <rbasak> hallyn, I can give you a shell if that's easier?
[09:12] <hallyn> rbasak: yes pls
[09:48] <rbasak> smoser: sed '$s/^foo$/bar/'
[09:49] <rbasak> smoser: sed -i '$s/^exit 0$/d'
[09:49] <rbasak> smoser: sed -i '$s/^exit 0$/d' /etc/rc.local
[09:49] <smoser> { echo 1; echo 2; echo 3; } | sed '$/3/d'
[09:49] <rbasak> smoser: sed '$s/^exit 0$//'
[09:49] <rbasak> smoser: sed -i '$s/^exit 0$//' /etc/rc.local
[09:53] <soren> smoser: Have you used cloud-init with the Rackspace cloud?
[09:53] <smoser> soren, no. i've never tried.
[09:53] <smoser> but you're going to fix that fo rme.
[09:53] <smoser> thanks.
[09:53] <smoser> :)
[09:54] <soren> ngh..
[09:54] <soren> I walked right into that one, didn't I?
[09:54] <soren> It's going to be rather amputated, seeing as I can't upload custom images there and cloud-init isn't pre-installed.
[09:55] <smoser> yeah.
[09:55] <soren> I can inject files, though.
[09:55] <rbasak> smoser: sed -i '${/^exit 0$/d}' /etc/rc.loca
[09:55] <soren> So I can inject an upstart job that installs cloud-init.
[09:55] <smoser> but if we got it working... and ti worked well, it seemed like rackspace might be open to that.
[09:56] <soren> Once they switch to OpenStack, I imagine we'd be able to provide images ourselves.
[09:56] <soren> ...with the right stuff in there.
[09:57] <soren> So, assuming cloud-init was there, and the only thing I could do was to inject files.. Is there any way I can feed data to cloud-init?
[09:57] <smoser> soren, you would know better than i.
[09:57] <soren> smoser: Only marginally so.
[09:57] <smoser> but my impression was always that their openstack solution would be very tight and likely not even allow uploading images.
[09:57] <smoser> i dont knwo why i got that feeling.
[09:57] <soren> smoser: Maybe. Who knows?
[09:58] <soren> (rhetorical question alert)
[09:58] <smoser> soren, yeah, you can inject metadata and user-data into
[09:58] <soren> Cool. where do I put it?
[09:58] <smoser>  see http://bazaar.launchpad.net/~cloud-init-dev/cloud-init/trunk/files/head:/doc/examples/seed/
[09:59] <soren> smoser: It's not entirely obvious to me what the difference is between DataSourceNoCloud and DataSourceNoCloudNet?
[10:00] <smoser> nocloudnet runs afternetwork is up.
[10:00] <smoser> so you can do a #include of a network resource
[10:00] <smoser> (or do apt-get install in a boothook)
[10:01] <soren> I see.
[10:01] <soren> smoser: Cool, thanks.
[10:21] <hallyn> jodh: stgraber: lp:~serge-hallyn/ubuntu/precise/upstart/upstart-containers is what I'll eventualy be proposing for moving upstart bits from lxcguest into upstart.  Won't be until the container-reboot patch hits kernel, but if you see anything wrong, I"d appreciate feedback.
[10:21] <jodh> hallyn: thanks. I'll take a look...
[10:23] <hallyn> \0/ thx
[10:53] <dax_roc> Morning all
[10:57] <dax_roc> I'm trying to setup a network install, I've got the boot process working but it's still looking for the cd after the keyboard detection i've passed in the following http://pastebin.com/CntCMMgW
[11:10] <smoser> rbasak, http://paste.ubuntu.com/801610/
[12:33] <cloudy37> hello, i hope that someone can steer me in the right direction here... i have access to an installation of ubuntu on a server which is accessible over the internet with an adress like   asdf.dhcp.university.com running an apache2 listening on port 80 and i'm wondering if it's possible to "add" another prefix like test.asdf.dhcp.university.com so that the apache can route traffic to that adress to e.g. a tomcat running on the sam
[12:33] <cloudy37> e machine
[12:35] <_ruben> you'd need to reverse proxy those requests towards tomcat then, and it'd require university.com to make sure your "test" subdomain points to the same ip
[12:40] <cloudy37> so my understanding is that university.com has a dns server running which resolves all the subdomains to the corresponding ip adresses but it's not possible to run another dns server on one of the subdomains to create more subdomains originating from e.g. asdf.dhcp.university.com, that correct?
[12:41] <ClaudiuT> hello
[12:41] <ClaudiuT> I tried a remote dist-upgrade and it hanged here http://imgur.com/dCsTK
[12:41] <ClaudiuT> what would be a wise next step now?
[12:42] <_ruben> cloudy37: not without university.com doing some delegation to that "another dns server"
[12:43] <_ruben> ClaudiuT: can you stil open a new ssh session?
[12:43] <ClaudiuT> yes
[12:43] <_ruben> ClaudiuT: then take a look with ps/top/etc what's currently going on and go from there
[12:43] <onre> what happens if you press "enter" in the terminal?
[12:44] <ClaudiuT> onre: nothing, goes one row lower
[12:44] <onre> (just making sure that you haven't accidentally pressed ctrl+S for whatever reason :)
[12:44] <onre> allright, not then
[12:44] <ClaudiuT> top looks clean
[12:45] <ClaudiuT> but I think this is the problem 22140 pts/2    Ss+    0:01 /usr/bin/dpkg --force-overwrite --status-fd 27 --conf
[12:45] <_ruben> i tend look at the top say 10 lines of ps auxOT
[12:45] <_ruben> err
[12:45] <_ruben> bottom
[12:45] <_ruben> not top
[12:47] <ClaudiuT> like this? http://pastebin.com/jvFL5DMM
[12:47] <ClaudiuT> I have a zombie there
[12:47] <ClaudiuT> should I kill -9 it?
[12:48] <onre> killing zombies usually does nothing
[12:50] <ClaudiuT> ctrl-c and apt-get upgrade?
[12:53] <onre> ClaudiuT, that's what i would do.
[12:54] <ClaudiuT> allright, thanks
[12:56] <ClaudiuT> how can I solve this? debconf: DbDriver "config": /var/cache/debconf/config.dat is locked by another process: Resource temporarily unavailable
[12:58] <_ruben> probably there's still a debconf process running
[13:00] <ClaudiuT> yup, that was it
[13:00] <ClaudiuT> apt-get upgrade said 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
[13:01] <ClaudiuT> everything seems to be fine now
[13:31] <stgraber> hallyn: do you know the status of https://bugs.launchpad.net/ubuntu/+source/ifupdown/+bug/632764 ?
[13:54] <eagles051387|> hey ikonia didn't manage to find anything relating to URI :( still trying to look but not finding anything
[14:04] <hallyn> stgraber: hm, no...
[14:04] <hallyn> stgraber: i don't think we're patching around that in lxcguest any more, so maybe it's fixed
[14:05] <stgraber> hallyn: yeah, I don't remember seeing this, so I'm quite tempted to marking it fix released even if I can't point to a specific fix
[14:10] <hallyn> stgraber: i just stuck an upstart job in my container that starts on net-device-up IFACE=lo, and it did start on boot.
[14:10] <hallyn> so yeah, it's fixed, somehow
[14:12] <smoser> utlemming, lp:~smoser/+junk/check-archive/
[14:12] <smoser> https://code.launchpad.net/~smoser/+junk/check-archive/
[14:14] <hallyn> stgraber: would you like to take a look at our current lxc patch before I dput?
[14:19] <hallyn> rbasak: i hosed the panda installing lxc from archive.  a little ridiculous.  i'm going to commit the current proposed version from the archive, then we can bugfix from there as needed
[14:33] <stgraber> hallyn: I'm a bit busy with ifupdown that I'd like to upload today, so if you think it's good and reasonably tested, just upload it. Otherwise I can have a look later tonight (likely after dinner)
[14:33] <hallyn> stgraber: np, uploaded it, thanks.
[16:04] <smoser> roaksoax, did you make it any furthre on odev ?
[16:05] <smoser> lynxman, https://code.launchpad.net/~orchestra/orchestra/odev/
[16:05] <smoser> thats "odev", which guides you through building a orchestra environment on your local system
[16:05] <lynxman> smoser: hmm interesting :)
[16:06] <smoser> you should try walking through it. i'm begging people to, so someone other than i has done it.
[16:27] <roaksoax> smoser: yeah works like a charm now
[16:28] <roaksoax> smoser: for some reason the time kvm ... etc etc was not fully completing
[16:28] <smoser> roaksoax, good.
[16:28] <smoser> thank you for that
[16:29] <roaksoax> smoser: btw.. are you planning to make this a bit more automated?
[16:29] <smoser> yeah, i have some stuff to do on it
[16:29] <smoser> the goal is to have the zimmer-build automated
[16:29] <smoser> and have that available for download
[16:29] <smoser> the stock one
[16:30] <smoser> so that step would be gone
[16:30] <smoser> and then i want to have the 'setup.py' renamed and have it basically set up everything for you and allow you to do things like: 'add node' 'add-node-to-cobbler' and stuff like that
[16:30] <smoser> then...
[16:30] <smoser> (you're more htan welcome to hepl with this if you find it useful)
[16:30] <smoser> then the bug thing i want to get to is the ability to use a cloud-image
[16:31] <smoser> instead of doing a deploy
[16:31] <smoser> (to fast path juju)
[16:32] <roaksoax> smoser: right, I think we could have both, so we can decide whether we wanna use images or do installations. Cause that way we can test "physical" deployments with orchestra
[16:38] <smoser> right
[16:38] <smoser> roaksoax, yeah. i just wanted the shortcut becuase its painful slow otherwise
[16:39] <roaksoax> smoser: agreed. but I still think we need a regular installation to test everything else
[16:40] <LivTyler> hello, I'm trying to set up ipsec in Ubuntu, but not working, when the client send ISAKMP, it receives from Ubuntu Server ICMP destination unreachable, any ideas?
[16:44] <smoser> roaksoax, yeah. we're going to have to be tricky somehow to do it the fast path
[16:44] <LivTyler> never mind, I think I've got it
[16:44] <smoser> as it will require cobbler talking to your system and saying "cheat and clone this node for me"
[16:44] <smoser> adam_g,  http://paste.ubuntu.com/801910/
[16:49] <roaksoax> smoser: probably writing a power management template
[16:50] <roaksoax> smoser: that clones the imagine and then starts it with virsh
[16:51] <roaksoax> smoser: so we would only have to modify the current virsh template
[16:52] <smoser> maybe, yeah
[16:52] <smoser> can a power management template get at ks vars ?
[16:52] <smoser> then we could have a var on the system that we just toggle
[16:55] <roaksoax> smoser: idk actually, but I don't see why it couldn't as it still gets vars from cobbler either wya
[18:38] <adam_g> smoser: # echo "c 1:7 rwm" > /sys/fs/cgroup/devices/lxc/ubuntu-jstack-nova-compute-0/devices.allow
[18:53] <nonotza> anyone know why my public IP address remains the same after connecting to an openvpn server?
[19:21] <JanC> nonotza: I assume the traffic isn't routed through your VPN...
[19:22] <nonotza> if I'm connected - isn't that a default behavior of the vpn?
[19:31] <maswan> depends on the vpn
[19:37] <mtaylor> SpamapS: ola amigo
[19:38] <JanC_> nonotza: what does "ip route list" say?  (put it on a pastebin)
[20:10] <doommaster> hi there
[20:12] <doommaster> I did a plain Ubuntu install on a system which contains a RAID, the OS is on a non RAID HDD
[20:12] <doommaster> now the system does not boot when any of the raid hdds are installed
[20:12] <doommaster> it marks the raid degraded and breaks the init
[20:13] <doommaster> nevertheless there is nothing needed on there
[20:13] <doommaster> and the system boots just fine when the RAID hdds are disconnected
[20:13] <doommaster> wtf is wrong?
[20:13] <doommaster> natty booted just fine
[20:22] <doommaster> I did a plain Ubuntu install on a system which contains a RAID, the OS is on a non RAID HDD now the system does not boot when any of the raid hdds are installed it marks the raid degraded and breaks the init nevertheless there is nothing needed on there and the system boots just fine when the RAID hdds are disconnected wtf is wrong? natty booted just fine
[22:24] <druonysuse> I am working on an Ubuntu server that is starting up with / as read-only
[22:24] <druonysuse> why is this happening?
[22:25] <druonysuse> is there a common cause for this... like a drive dying?
[22:31] <EvilResistance> what's /usr/sbin/named do?
[22:33] <JanC> EvilResistance: resolve DNS addresses?
[22:33] <EvilResistance> JanC:  required in server installs?
[22:33] <JanC> it's part of BIND normally
[22:33] <EvilResistance> (its taking up 18% of a low-mem VPS)
[22:33] <EvilResistance> JanC:  how would i go about killing it?
[22:34] <EvilResistance> killall <program> ?
[22:34] <JanC> EvilResistance: if you don't need yourown DNS service, and you can use a DNS relay from your VPS provider (or elsewhere), it's not strictly required, no
[22:35] <EvilResistance> ack, i found the big mem eaters...
[22:35] <EvilResistance> php-fpm is taking up 48% memory o.O
[22:35] <JanC> I'd suggest you make sure your name lookups don't use a local service if there isn't one  ;)
[22:37] <EvilResistance> right...
[22:37] <EvilResistance> not sure why its running though
[22:37] <EvilResistance> i cant find any package associated with it
[22:37] <EvilResistance> and its got 5 processes running (named)
[22:38] <shauno> $ dpkg -S `which named`
[22:38] <shauno> bind9: /usr/sbin/named
[22:38] <shauno> assuming yours is isc too, I'd assume the same package, but that's a handy one to know
[22:39] <JanC> and if you really want a local DNS relay/cache (for performance reasons) you can also try something like dnsmasq
[22:39] <patdk-lap> hmm, trimming down memory of php-fpm, normally isn't easy
[22:39] <patdk-lap> atleast if you actually needed php for anything
[23:18] <lx-berlin> hello everyone. i just setup my first ubuntu server. but i am having problems with the software raid
[23:19] <lx-berlin> even if i rebuild the raid, when i restart the server the next day, the raid1 is no more working again
[23:20] <lx-berlin> sdb is always getting removed from the raid
[23:22] <lx-berlin> can someone take a look at the smart values of my sdb and tell me if this disc is really faulty ? http://pastebin.com/1Txby5WG
[23:24] <lx-berlin> or is it possible that i didnt configure the raid1 correctly?
[23:29] <lx-berlin> here are
[23:30] <lx-berlin> ups. here are the mdadm details: http://pastebin.com/S0PBPZxd
[23:31] <lx-berlin> is anyone there?