[00:14] dannf: I doubt those are *signed* with the *public* key ;) [00:15] but you probably want the 'ubuntu-cloud-keyring' package [00:16] New bug: #1065511 in cinder "cinder incorrectly creating tgt files in /usr/lib/python2.7/dist-packages/volumes/" [Undecided,New] https://launchpad.net/bugs/1065511 [00:20] undersun: get your mailserver offline, learn how to be a better sysadmin, re-install your VPS from scratch and configure it properly === n0ts_off is now known as n0ts === fleish_ is now known as fleish === n0ts is now known as n0ts_off === shantorn_ is now known as shantorn [01:46] JanC: i checked u-c-k, didn't see that *private* key in the keyring === tarzxvf is now known as tarcvf|rar === n0ts_off is now known as n0ts === n0ts is now known as n0ts_off === n0ts_off is now known as n0ts [05:25] hi === erichammond1 is now known as erichammond === smb` is now known as smb [08:26] New bug: #1030610 in qemu-kvm (main) "qemu-kvm depends on undesirable additional packages" [Low,Triaged] https://launchpad.net/bugs/1030610 === mcclurmc_away is now known as mcclurmc === n0ts is now known as n0ts_off [09:31] New bug: #1071694 in exim4 (main) "CVE-2012-5671: Heap-buffer overflow in DNS decode logic used for DKIM" [Undecided,New] https://launchpad.net/bugs/1071694 === n0ts_off is now known as n0ts [10:11] j #puppet-razor [10:56] New bug: #1069734 in maas "Filestorage is unique to each appserver instance" [Undecided,Confirmed] https://launchpad.net/bugs/1069734 === n0ts is now known as n0ts_off === n0ts_off is now known as n0ts === cpg is now known as cpg|away [12:42] hi there, is there an easy way to use local storage for volumes in openstack folsom? local as in the compute node uese its own HDDs [12:43] hi all [12:45] I just installed ubuntu server 12.10 on our new server, but while I can clearly see that my network devices are detected (em1, em2, p2p1, p2p2), /et/udev/rules.d/70-persistent-net.rule is just plain empty [12:45] meaning that, if I run ifconfig, it only returns the loopback device === dendro-afk is now known as dendrobates [13:32] stgraber: say, does this ring any bells for you? lxc-start 1351200114.339 ERROR lxc_conf - Device or resource busy - error unlinking /usr/lib/x86_64-linux-gnu/lxc/dev/console [13:32] i seem to rcall someone mentioning that in the last few days... [13:32] utlemming: still no /dev in precise ubuntu-cloud [13:33] hallyn: I saw the bug, my guess was that it was related to devtmpfs and /dev/console being open on the host [13:33] but I only saw it quickly in bugmail and didn't investigate, so jsut a guess really === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [13:39] stgraber: d'oh, right, thanks [13:44] Hi, I'm installing Ubuntu 12.04 server via cobbler 2.2.3 using a seed file and I'm noticing, when trying to use a static IP, that it always uses DHCP unless I put "netcfg/disable_autoconfig=true" in kopts. My confusion is that Ubuntu just flat out ignores the "netcfg/disable_autoconfig boolean true" in the seed file. [13:45] New bug: #1071756 in bacula (main) "package bacula-director-mysql 5.2.5-0ubuntu6.2 [modified: usr/share/bacula-director/update_mysql_tables] failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/1071756 [13:51] Hi [13:52] I'm trying to configure 2 static ethernet interfaces, but when I run sudo /etc/init.d/networking restart it gives me : RTNETLINK : File exists [13:52] and fails to bring up either eth0 or eth1 [13:52] s/either/both/ [13:52] *and fails to bring up both eth0 and eth1 [13:53] I googled it a bit and there's no definitive answer [13:53] shadeslayer: I had the same problem on a couple of vm's, what fixed it for me was removing /etc/udev/rules/70-persistent-net.rules [13:54] aha [13:54] shadeslayer: And then a reboot.. [13:54] alright [13:55] if this works I'll be jumping like a maniac [13:58] shadeslayer: Hehehe :) [14:11] lordievader: doesn't work [14:13] shadeslayer: That is too bad. [14:14] * shadeslayer tries something === shantorn_ is now known as shantorn [14:17] so has anyone here installed ubuntu 12.04 server with a seed file from cobbler before and gotten static IPs configured? [14:28] lordievader: any other ideas? [14:33] * shadeslayer is utterly lost with this error [14:33] shadeslayer: Not really, it worked for me, so didn't bother looking for more. [14:33] shadeslayer: Sorry. [14:33] no need to be :) === mcclurmc is now known as mcclurmc_away [14:54] just starting up with a ubuntu 12 vps, used for a webserver... portscanning it reveals port 53 open, should I close it via iptables? it's not needed externally, right? [14:57] New bug: #1071772 in bridge-utils (main) "Precise DHCP bridge with bond doesn't come up correctly" [Undecided,New] https://launchpad.net/bugs/1071772 [15:02] lordievader: fixed it [15:02] stupid virtual networks were causing issues [15:03] probably misconfigured [15:09] does anyone know why my ubuntu server says that iptables -t nat -F says "can't initialize table 'nat'" [15:09] ? [15:15] xorred: are you running a custom kernel or something like that? Check /var/log/kern.log [15:21] rbasak: it's empty [15:21] I just got this vps from my vps provider, guess the distro is custom since it had apache preinstalled [15:22] lol, lsmod returns 0 [15:22] is that bad === dendrobates is now known as dendro-afk [15:27] okay, so, my server has 2 ethernet interfaces, I can ping google from one, but not the other [15:27] I get a Destination not reachable [15:27] *destination host unreachable [15:29] xorred: sounds like your host is providing the kernel and not supplying logs either, and hasn't built the nat module. Is this virtuozzo by any chance? [15:30] xorred: I think the short answer is to ask your host :/ [15:34] rbasak: I just installed another kernel - 3.5 lowlatency [15:34] xorred: just make sure that you can actually run the kernel you have installed. Depending on the hosting environment you may not have control over it. [15:35] well so far it's strange. openvz, and rebooting it - halts it [15:35] weird [15:35] have to boot up from the control panel manually every time [15:37] lol wth, uname -a shows 2.6, initrd and vmlinuz point to 3.5! /boot contains only 3.5 [15:37] what the.... === dendro-afk is now known as dendrobates === erichammond1 is now known as erichammond [16:06] strange, why is add-apt-repository not present in ubuntu 12.10? === matsubara is now known as matsubara-lunch [16:07] xorred: Because you don't have the software-properties-common package installed? [16:19] I've got an upstart task that simply won't start. http://pastebin.ubuntu.com/1307508/ [16:19] hallyn, ^ ? [16:21] ppetraki: anything in /var/log/upstart/ ? [16:22] umm.. yeah, but it doesn't add up as to why it works from shell, and not from upstart [16:22] i'm wondering whether the job started, and the script simply had a failure [16:23] hallyn, http://pastebin.ubuntu.com/1307519/ [16:23] ppetraki: you're aware that upstart scripts run in dash? [16:24] hallyn, no [16:24] so what you're sourcing probably isn't proper dash [16:24] hallyn, so how do I source bash? [16:24] you don't - can you toss a full script in /bin for it to run? [16:25] or convert /var/lib/juju/units/opengrok-2/charm/inc/common to dash [16:26] grr, that just makes more work, the whole idea was to have entry points callable as functions [16:26] I guess I could do the same symlink business with inc/common as is done with the rest of hooks [16:29] ppetraki: you could just switch the /bin/sh symlink [16:30] no big deal for your juju charm imo [16:30] hallyn, sounds like alternatives magic, got an incantation to do that programatically? === n0ts is now known as n0ts_off [16:33] ppetraki: i don't. and i don't see any /etc/alternatives/sh [16:33] hallyn, I'm just going to create some tiny driver scripts, atleast I understand the problem now [16:33] i think a simple 'rm -f /bin/sh; ln -s /bin/bash /bin/sh' in hooks/install is the way to go [16:34] ok [16:34] hallyn, that was sneaky http://pastebin.ubuntu.com/1307536/ [16:56] New bug: #1071817 in tomcat7 (main) "Missing org.apache.tomcat.dbcp.dbcp.BasicDataSourceFactory" [Undecided,New] https://launchpad.net/bugs/1071817 === matsubara-lunch is now known as matsubara [17:15] Anyone have time and expertise to help me figure out why my NFS4 mounts don't share the same usernames? [17:15] I tried following the documentation, but it seems like the id mapping is not mapping correctly [17:27] demersus: using kerberos? [17:28] demersus: and same UIDs and GIDs on both systems? [17:28] RoyK, no I am just using the simple quickstart nfs4 config. No kerberos, /etc/passwd on both machines [17:29] No the UIDs, and GIDs are different. Isn't imapd supposed to support different UIDs? [17:29] imapd? or idmapd? [17:29] RoyK: sorry typo. idmapd [17:30] I'd recommend something of a shared user database [17:30] ldap usually works [17:30] nis too, although a bit old-style [17:31] unix systems generally work *far* better with the same UID/GID on users [17:31] This is just temporary so I can migrate websites to another server while DNS updates. I don't want to go through all the trouble of syninc UIDs [17:31] Maybe I don't understand the use of idmapd. I thought it was supposed to sync user names instead of UIDs [17:32] if it's just moving the data, you could transfer them and run something like "find /path -user -exec chown {} \;" [17:32] I've only used idmapd for connecting windows/unix [17:32] Ok. [17:33] Thanks for your suggestions. I appreciate the help. [17:33] or - change the uid/gid on the new server to the old uid/gid [17:33] probably the easiest [17:34] True. I was just hoping NFS + idmap would be quicker and easier for the time being. (Lots of vhosts & users) [17:34] I should have copied the UIDS in the first place. [17:35] well, just change them [17:35] and chown /home/someuser to that uid [17:36] chowning a bunch of files takes far longer [17:36] New bug: #1071829 in tomcat7 (main) "directories for shared libraries do not match current documentation" [Undecided,New] https://launchpad.net/bugs/1071829 [17:36] Okay. [17:36] make that chown -R ... [17:37] I figured that is what you meant [17:37] Thanks === protoCall7_ is now known as protoCall7 === koolhead17|afk is now known as koolhead17 [18:16] hi, I was wondering if someone could help me a wee bit in setting up a VM on my server [18:17] I can't quite figure out how to get my VM a public IP [18:17] The host has 2 static IP's and I'd like to assign one of them to the VM [18:27] shadeslayer: no sure, but I'd use a bridge [18:27] that's what I'm using :) [18:27] but for some reason the VM doesn't pick the ip [18:27] shadeslayer: that is, configure the primary as a bridge and don't assign that IP to the host [18:27] oh [18:27] I'm not sure I want to do that [18:28] because the primary IP is already for something else [18:28] well [18:28] what I do on my server is: I have eth0 without an IP address [18:28] I have br0 with the host's IP [18:29] and then the guests bridge on that with different IP addresses [18:29] I think this is the "preferred" setup [18:30] RoyK: I have this : http://paste.kde.org/582494/ [18:30] where br0 is supposed to be the bridge [18:31] you shouldn't set an IP address on the bridge interface [18:32] at least not on the same network as the guests [18:32] so drop 18-22? [18:33] ( As you can tell, I'm doing this for the first time and I simply followed the KVM networking guide ) [18:33] set the ip address on br0, not on eth1 [18:33] https://help.ubuntu.com/community/KVM/Networking [18:33] RoyK: but don't I have to set the static IP on eth1? [18:34] you don't need/want an IP address on eth1 [18:34] it'll be bridged to br0 [18:34] okay [18:34] and bleh, how do I get rid of this : RTNETLINK answers: File exists [18:34] Failed to bring up br0. [18:35] eth1 is just a bridge point [18:35] br0 is using that [18:35] ( happens once I run : invoke-rc.d networking restart ) [18:35] okay [18:35] got it [18:35] restart networking or reboot [18:36] and if you're using RFC1918 addresses for the guests, keep in mind that they won't be accessible from the internet unless you do some NAT magick [18:36] uh oh [18:37] RoyK: okay, what if I want a publically accessibly IP for the VM [18:37] absolutely no NAT'ing [18:39] just don't set any IP address other than the primary on br0 [18:39] and none on eth1 [18:39] then tell the vm to use br0 and set the official IP on its interface (eth0) [18:39] it'll be bridged through br0 and sent out [18:40] alrighty [18:40] http://paste.ubuntu.com/1307846/ [18:40] this is the configuration of one of my servers - it runs a few VMs with official IP addresses [18:41] RoyK: http://paste.kde.org/582512/ [18:41] does that look sane? [18:43] no use for netmask or gateway [18:43] well [18:44] netmask, yes, if it has an ip address, but no gateway [18:44] a system has one gateway [18:44] okay [18:44] regardless of interfaces [18:44] if you have more gateways, you need OSPF or BGP or something [18:45] (or RIP if you're from the ninetees) [18:45] hah :D [18:45] no idea what those are but reading up on them :) [18:46] you don't need to [18:46] both OSPF and BGP will require routing deals with the ISP, so you don't need that [18:47] BGP is used between ISPs [18:47] heh, I'm reading to satisfy my curosity as to what this is :P [18:47] : [18:47] :) [18:47] that's good [18:47] *curiosity [18:47] curiousity is always good [18:48] * shadeslayer notes that he should probably sleep within 30 minutes to catch his flight for UDS [18:48] UDS? [18:48] Ubuntu Developer Summit [18:48] starts next week [18:48] where's that? [18:49] Copenhagen [18:49] too bad my employer sticks to redhat... [18:49] :D [18:50] hmph [18:50] I keep getting this interface called vnet0 [18:50] that's usual [18:51] virtual interface created to interface with vr0 [18:51] br0 [18:51] one per vm [18:51] http://paste.kde.org/582518/ [18:51] configure the vm to use the IP you want to use [18:51] okay [18:53] and btw, posting official ip addresses isn't dangerous unless you haven't secured the server, but then, if you haven't secured the server, it's out bad anyway [18:54] :) [18:55] ok, still doesn't connect to the internet [18:56] can't ping the vm, can't ping from the vm [18:56] pastebin the config from both host and guest [18:56] ifconfig from both should do [18:57] hmm .. VM doesn't have pastebinit/internet [18:57] will take a screenshot :) [18:57] ok [18:57] RoyK: VM : http://i.imgur.com/moyEN.png [18:57] I guess the host setup is prominent [18:58] vm http://paste.ubuntu.com/1307888/ [18:58] erm [18:58] host ^ [18:58] same IP on host br0 and guest [18:59] guess the host will reply, then [18:59] try to remove the ip address on br0 [18:59] okay [19:00] Missing required variable: address [19:00] Missing required configuration variables for interface br0/inet. [19:00] sec [19:00] sure [19:01] iface eth1 inet manual [19:01] make that br0 [19:02] * shadeslayer tries [19:04] mmm ... rebooting VM [19:04] nop [19:41] anyone know if it's possible to disable scp within ssh? [19:41] i know how to disable sftp, but would like to nix scp as well === cpg|away is now known as cpg [20:50] Hi, all! I run a small biz and wanting to set up a ubuntu server to handle webhosting, SOGo (and it's dependencies), ftp, samba (for active dir and file sharing), and quickbooks db. any problems with running all of those services from one machine? Or will I need to split any of them up? [21:08] I know it's not exactly a tech question, but does anybody see any issues with running the following services all from one machine? SOGo, postgreSQL, OpenLDAP, postfix, Cyrus IMAP, as well as Samba (for Active Dir and file sharing), web hosting and ftp hosting, and Quickbooks DB? I vaguely recall that there might be a problem with using samba for AD along with sogo and openLDAP on the same box. I'm asking because I'm about to b [21:09] you can run as much stuff on one box as you want [21:09] the only issue you will have is if things require certain ports [21:10] but normally, you should split them up for security, flexability, or just general update/upgrade mainance ease === cpg is now known as cpg|away [21:14] Thanks patdk-lap, I'll keep that in mind [21:14] Hello - it was suggested to me at Linuxcon by an Ubuntu developer that ubuntu has a system called juju that might fit my needs for cluster provisioning. Are there any juju developers/users around? I wasn't able to find the sort of details that I was looking for on the juju website [21:17] I'm looking specifically to deploy minimal Debian or Ubuntu system images to an HPC cluster that I run at my university. The cluster has 7 different classes of compute nodes, and I need to be able to manage 4 different system images. [21:18] I need something that supports PXE booting, but must be able to install the system image to a hard drive as I cannot have the OS image living in memory. [21:18] When installing Server 12.10 I get an error after detecting the cd-rom, this is the error "ERROR MOUNTING CD-Rom" === cpg|away is now known as cpg [21:28] nabbend [21:37] evening [21:39] is ubuntu server release upgraded the same way as the desktop version ? [21:39] it is [21:39] clear and nice. [21:39] thanks :) [21:39] but most of us that run server releases stick to LTS [21:40] LTS! [21:40] !lts [21:40] LTS means Long Term Support. LTS versions of Ubuntu will be supported for 3 years on the desktop, and 5 years on the server; with the exception of 12.04 (Precise Pangolin), which will be supported for 5 years on the desktop. The current LTS version of Ubuntu is !Precise (Precise Pangolin 12.04) [21:40] i see. [21:41] i'll stick around for at bit. [21:44] New bug: #1071910 in lxc (universe) "lxc stop will hang forever" [Undecided,New] https://launchpad.net/bugs/1071910 === TJ- is now known as mooc-bot === _TJ_ is now known as TJ- === kInOzAwA is now known as Guest88746 === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha