[00:07] <pmatulis> i notice on 12.04 there is no prompt during install time when setting up raid whether to boot into a degraded mode.  is that right?  the default is also to _not_ boot in such a state.  is that what we want these days?
[00:08] <patdk-lap> the default is only an option when the root is on raid
[00:08] <patdk-lap> if it isn't, the option is never presented
[00:09] <patdk-lap> bug has been filed a long time ago
[00:23] <Gallomimia> briggz i'd like you to make a piece of software that will allow networked systems to share symbolic links and treat them as if they were hard links. maybe that will fill up 8 weeks of time. file integrity checks, hash checks, and in the case of text files versioning
[00:24] <Gallomimia> ah he's not here :(
[00:30] <vicTROLLA> Having trouble with a PXE setup. My preseed file is failing because the OS I'm installing over has LVM data that spans both physical drives. For some reason the installer doesn't want to delete this data. Any ideas?
[00:31] <vicTROLLA> https://answers.launchpad.net/ubuntu/+source/ubiquity/+question/200964 :(
[00:38] <briggzer> I'm tring to install bind9 and samba, I keep getting errors and it says failed to start service, am I in the right place?
[00:38] <briggzer> i'm using ubuntu-server
[01:11] <bfortified> Finally setup NFS at home, My wifes account is able to manipulate files she doesn't have permissions for with out using root. Is this because she has an administrator account or because of a weak NFS? setup?
[06:47] <lerian> hi guys
[06:48] <lerian> i'm looking for someone for this http://pastebin.com/zpCTvK64 (could be remote or on site, it's flexible)
[08:05] <stgraber> hallyn, utlemming: ubuntu-cloud still downloads 20121017 for quantal instead of the new build, so I doubt I'll get a /dev (still downloading)
[08:07] <lerian> looking for a system engineer > http://pastebin.com/zpCTvK64
[08:14] <hallyn> stgraber: gah
[08:14] <hallyn> stgraber: no wait, quantal as fine.  it's precise that had no /dev
[08:14] <hallyn> which is downloading release-20121026.1 fo rme
[08:15] <stgraber> hallyn: gah, right
[08:16] <lerian> i'm looking for a system engineer http://pastebin.com/zpCTvK64
[08:34] <lerian> the refferer will get 200 euros http://pastebin.com/zpCTvK64
[10:37] <xsl> can someone help i get this error wen starting lxc container manually type=1400 audit(1351593348.376:10): apparmor="DENIED" operation="mount" info="failed flags match" error=-13 parent=17347 profile="lxc-container-default" name="/var/lib/ureadahead/debugfs/" pid=17362 comm="ureadahead" fstype="debugfs" srcname="none" flags="rw"
[10:38] <xsl> and the same happens at startup
[10:38] <xsl> tough manually the container is up
[10:38] <xsl> but at startup the container does not start
[10:48] <xsl> stgraber your there? can you tell me if this bug is sorted?
[10:54] <stgraber> xsl: it's not a bug, it's intentionally blocked
[11:06] <bsdman> everyone super happy with ubuntu server 12.10?
[11:06] <bsdman> I am. :-)
[11:17] <xsl> stgraber, but apparmor is preventing me from starting containers at host powerup wen i link the container in /etc/lxc/auto
[11:17] <xsl> manually the container starts up ok , even tough the error on syslog
[11:18] <xsl> but wen the host server reboots the container does not come up
[11:24] <xsl> stgraber, http://pastebin.com/H3EhX6Ds
[11:24] <xsl> sorry if im bothering you
[11:25] <xsl> but i dont see information on the web about this problem, only a bug on launchpad
[11:26] <allegrem> Hi everyone !
[11:26] <allegrem> I've a little problem with udev which fails to mount a key when i plug it
[11:26] <allegrem> here are all the informations : http://pastebin.com/nPyFk8KP
[11:26] <allegrem> thanks in advance for your help !
[11:35] <xsl> the info changes from "failed flags match" wen i manually start it to "failed type match" wen i try it at linking in /etc/lxc/auto
[11:36] <xsl> another thing i see is that the container tries to come up before eth0 is up at boot... can that be a problem ?
[11:42] <xsl> i'm so sorry stgraber to bother you ... i found the issue ... i was doing the symlink wrong
[11:45] <Aison> why do I have got such strange network interface names with 12.10? p10p1, p9p1
[15:47] <roaksoax> jamespage: ping
[15:47] <roaksoax> jamespage: why do relation hooks get executed twice, or so it seems?
[18:05] <hallyn> stgraber: success!  I'll have to re-test once everything is in ppa.  there are some oddnesses, but basically not only does it work, even sudo works :)
[18:10] <highvoltage> whohoo
[18:30] <stgraber> hallyn: wow!
[18:31] <hallyn> stgraber: nice talk :)
[18:33] <stgraber> hehe, thanks. The usual, please use LXC and talk to us if you need anything ;)
[18:34] <highvoltage> you gave a lightning talk stgraber?
[18:34] <stgraber> highvoltage: yeah, at the testing in ubuntu event on using lxc for development
[18:35] <stgraber> with a good 30s preparation time, so was pretty basic ;)
[18:35] <highvoltage> yeah i got here very late :-/
[18:36] <stgraber> I saw Adam and a few others arriving just after I was done, I guess you arrived around that time too
[18:36] <highvoltage> i arrived around 3 mins ago
[18:37] <hallyn> drat, there is at least one complication from the userns stuff (thoug i'll ignore that when i blog about it :) - i have to move the cgroup setting stuff around.  actuall yi'll probably move that to the parent of lxc-init.  no reason for init to do it itself as it currently is!
[19:18] <_jmedina> hi there
[19:18] <roaksoax> jamespage: ping
[19:18] <_jmedina> Im testing packemaker in 12.04
[19:19] <_jmedina> im reading the document pacemaker from scratch, and Im in the stage where I requiere to install the pcs tool for management, I already installed pacemaker but dont see any pcs package, anyone knows if is supported?
[19:22] <mjampala> hi guys, need some help on process D state
[19:22] <mjampala> snmpd gets stuck in in seq_read() of the seq operation in seq_read.c
[19:23] <mjampala> hile reading /proc/net/dev
[19:23] <mjampala> his happens once in every 15 days
[19:23] <mjampala> repro'ing this is not easy and I am basically panic'd with this
[19:23] <mjampala> Does anyone run into this
[19:24] <_jmedina> roaksoax: afaik you were working in the cluster stack, am I right?
[19:32] <jmedina> about my problem with pacemaker in 12.04 it is because Im reading a recent version of the document clusters from scratch
[19:33] <jmedina> I think Im going to search for a document that works with the version of pacemaker and corosync that ships 12.04
[19:34] <roaksoax>  jmedina that document is based on pacemaker 1.1.8 in preise we ship 1.1.7 which uses crm shell
[19:34] <jmedina> roaksoax: thanks, that is what I just relised
[19:35] <roaksoax>  jmedina http://www.clusterlabs.org/doc/en-US/Pacemaker/1.1-crmsh/html/Clusters_from_Scratch/index.html
[19:35] <jmedina> 1.1.7? I just checked and is 1.1.6
[19:35] <jmedina> roaksoax: now I reading the third edition
[19:36] <roaksoax> yeah sory quantal has 1.1.7
[19:43] <jmedina> roaksoax: is there documentation specific for ubuntu server?
[19:44] <jmedina> roaksoax: I few years ago I read you tesis, but now everything changed
[19:48] <roaksoax>  jmedina https://wiki.ubuntu.com/ClusterStack/Natty  https://wiki.ubuntu.com/ClusterStack/Precise
[19:50] <jmedina> thank you, I see it still uses cman
[21:14] <jmedina> roaksoax: Im following the document for precise, I think it lacks a few details, I hope I can contribute
[21:19] <jmedina> roaksoax: did you ever worked with CLVM?
[21:36] <semiosis> hi all... is there an easy way to prevent post-install/upgrade scripts from running when i update a package?  specifically, prevent a service from being restarted on upgrade?
[21:55] <maxb> semiosis: There could be various ways to achieve it - which package?
[21:57] <semiosis> hm, it was a question from a friend... i think in his case it was mysql-server.  is there a general solution?
[21:57] <maxb> Not fully general, no
[21:58] <semiosis> maxb: ok, i found something about policy-rc.d but didnt really understand it & didnt find any easy examples in a few mins of googling
[21:58] <maxb> The closest you get to a general solution would be to use the policy-rc.d control over invoke-rc.d, but as some things move away from rc.d-based control to things like upstart jobs, that's not really a guarantee
[21:58] <maxb> Yeah, policy-rc.d is not well documented
[21:58] <semiosis> ok that makes sense
[22:00] <maxb> The most common / basic use of policy-rc.d is in chroots, when you want nothing daemonish to run
[22:00] <maxb> In that case, it's common to create /usr/sbin/policy-rc.d as a trivial shell script which just runs 'exit 101'
[22:01] <semiosis> ah, but would that prevent one from starting daemons manually also?
[22:01] <semiosis> in addition to preventing package scripts from doing it?
[22:01] <maxb> It prevents anything going through the invoke-rc.d layer
[22:01] <semiosis> ok so calling /etc/init.d/... scripts directly could still work
[22:02] <semiosis> cool, thanks!
[22:04] <maxb> In the case of the mysql server packages currently in Ubuntu, they do use initscripts via the invoke-rc.d layer, so policy-rc.d is a potential solution
[22:04] <maxb> The other being to simply alter /etc/init.d/mysql to enforce policy
[22:08] <maxb> Documentation for policy-rc.d is at /usr/share/doc/sysv-rc/README.policy-rc.d.gz if you want to look into writing a small script to use there to do the job
[22:08] <semiosis> maxb: right, the tell apt-get to keep the installed script.  that would work too
[22:08] <semiosis> thanks!
[22:09] <semiosis> s/the tell/then tell/
[23:02] <stgraber> hallyn: fixed lxc published to quantal-updates
[23:16] <niel> hello, can anyone help me port forward incoming connections on port 54201 to 10.8.0.6
[23:31] <gletob> Hey is bind9 installed when you do apt-get install lamp-server^ ?
[23:32] <gletob> Or more importantly, does it serve any other function than being a DNS server?
[23:56] <maxb> gletob: I'd be very surprised and call it a bug if bind9 was included in lamp-server task