[00:42] <kieppie1> howdy
[00:42] <kieppie1> anyone online familiar with LXC?
[00:46] <kieppie1> howdy - anyone familiar with LXC? I'm following the official Debian (vanilla) docco's, but LXC as implemented seems FUBAR - does not work "out of the box", and the docco's acknowledge this (going on close to 2 years now) with no intent for resolution in sight. I'm wondering if the state for Ubuntu is much better
[00:51] <blkperl> kieppie1: lxc works fine in ubunut
[00:53] <kieppie1> blkperl: 100% as per docco on wiki? you familiar with the issues @ debian re busted templates?
[00:54] <blkperl> idk about the debain stuff, but you can easily make ubuntu lxcs on ubuntu
[01:24] <kieppie1> blkperl: sweet, thanks
[04:43] <tubaguy50035> any reason why the "install" option wouldn't show up on a live USB?
[04:43] <tubaguy50035> For 13.10 *
[06:44] <andygraybeal> so in general if i want to use openvpn to route only specific data ... is it easy enough to do that?
[06:46] <andygraybeal> the vpn would only be access HR/task management/ldap stuff over vpn, nothing else... all else can go on regular network/internet.
[08:39] <monokrome> Does anyone know how to give options to something for an automated install? I need to set the mail name for postfix
[08:41] <monokrome> google wins
[13:27] <Ravimandal> hi
[13:29] <cfhowlett> Ravimandal, greetings
[14:29] <strixUK> so, i've just set up mysql replication on a new box, and i (deliberately) let it overwrite and update the 'mysql' database
[14:29] <strixUK> other than changing the password in /etc/mysql/debian.cnf, are there any other implications of doing this?
[16:04] <ramadhan> ping !
[16:19] <auronthas> question:  /etc/network/interfaces     where eth0 are net and eth1 are local net,  what to put in on gateway in setup of eth1  when  eth0 are iface eth0 inet dhcp where IP might change now and then ?
[16:20] <auronthas> can one put name of  eth0 connection as gateway or it has to be specified IP address ?
[16:20] <strixUK> you can only have one gateway on the system, and if eth0 is being configured by dhcp, then dhclient will set that automatically.
[16:21] <strixUK> you don't need to do anything special with eth1, since there will be a net route for that automatically
[16:21] <auronthas> ok so I have to comment out gateway on eth1 then
[16:22] <strixUK> if there is a router directly attached to eth1 (the network, not the port) for a network that is not eth1's net, you'll need to add specific net routes for those other networks, or the kernel will try to get there via eth0 - but this situation almost certainly doesn't apply to you.
[16:22] <auronthas> ubuntu server are beeing configured to act as router itself
[16:22] <auronthas> but gets dynamic IP from ISP
[16:23] <strixUK> the only thing relevant to that is whether this machine is configured to forward packets
[16:23] <strixUK> all packets received (by whatever interface) will be routed according to the routing table
[16:23] <auronthas> going to install  isc-dhcp-server to listen on eth1 and shorewall as firewall
[16:24] <strixUK> right, so dhcpd needs to hand its eth1 ip to its clients
[16:24] <auronthas> nods
[16:24] <strixUK> there's no need for an explicit gateway spec in this machine's interfaces
[16:24] <auronthas> so in basic I only need to set  static IP and netmask in eth1
[16:25] <strixUK> right
[16:25] <auronthas> goodie, thx for help  :)  have a nice day/night whereever u are in world :)
[16:26] <auronthas> ah dinner time in UK now
[17:15] <jmedina> HI there
[17:15] <jmedina> I hope you can help me with a libvirt/kvm problem
[17:15] <jmedina> Im using ubuntu 13.04
[17:15] <jmedina> whenever I try to start a virtual machine I get a message like this:
[17:16] <jmedina> qemu-system-x86_64: -drive file=/labs/kvmimages/neti01.example.com-disk1.img,if=none,id=drive-virtio-disk0,format=qcow2,cache=none: could not open disk image /labs/kvmimages/neti01.example.com-disk1.img: Permission denied
[17:17] <jmedina> I checked the permissions/owner and was reseted to root:root and 660
[17:17] <jmedina> I change them to root:libvirtd and 660 and again, when I try to start the VM they are reset to root:root
[17:19] <jmedina> I have been looking in some mailing lists but I have not get a solution
[17:19] <jmedina> I already  disabled the libvirtd apparmor profile
[17:19] <jmedina> It looks it is libvirt who changes the owner
[17:21] <jmedina> my user is member of kvm and libvirtd
[17:21] <jmedina> It worked in 12.04
[17:22] <jmedina> this happens when I try to start a vm from virt-manager or virsh
[17:44] <andygraybeal> jmedina, i recall something similar...
[17:44] <andygraybeal> but i don't remember any of it
[17:45] <jmedina> :S
[17:46] <andygraybeal> sorry, i'm not helpful
[17:53] <andygraybeal> currently my email server is run by gandi.net because i bought my domain with them.  i would like to use mailman for maillists.  is it possible to configure mailman on my own server.. without having to move MX records over to that server?
[17:54] <andygraybeal> i don't want to run a full blown mail server on my own server... .but i want to use mailman ... maybe this is not possible.
[18:45] <jmedina> well, I modified /etc/libvirt/qemu.conf the parameter group from root to libvirtd and restarted libvirtd-bin and everything is working as expected, thanks :)
[19:03] <andygraybeal> jmedina, rock on
[20:03] <auronthas> just installed  isc-dhcp-server    question about  option router  that gives gateway to clients, when router IP on ethernet connection are dynamic  one can't put static IP to option router   how to put dynamic ip in dhcpd.conf file ?
[20:08] <auronthas> should I use option router-discovery ?
[22:12] <pmatulis> auronthas: a router should not have a dynamic address
[22:52] <auronthas> pmatulis: Talked to friend of me,  localnet's router ip is eth1's IP   ofcourse "facepalm"
[22:54] <auronthas> pmatulis:  where eth0 goes to net
[23:15] <gdi2k> hi all, I have a server with high latency on a LAN causing me headaches. Other machines can ping each other in the 0.01ms range, but anything to / from this machine takes around 50ms (avg). It's a RTL8111/8168B card which has given me driver headaches before. Should I just replace it or is there something else I can do?
[23:34] <patdk-lap> gdi2k, heh, that is really odd
[23:35] <patdk-lap> normally you don't get >30ms with gigabit, and that doesn't include any os stack latency/interrupts
[23:35] <patdk-lap> for me, 80-140ms is *normal* for gigabit
[23:46] <gdi2k> yes, this is gigabit. it is capable of delivering - ping summary shows Best less than 0.1ms but worst 141 with avg around 50ms
[23:46] <gdi2k> it's all over the place
[23:47] <gdi2k> load is not too high - 1.6 on a quad core machine
[23:47] <gdi2k> (no HT)
[23:52] <gdi2k> load has no impact - I whacked it with 4 instances of cpuburn to saturate the cores, and ping is unchanged
[23:52] <andygraybeal> i followed these instructions at this website: http://burn.co.nz/blog/?p=245  well.. just the line:  sudo aptitude install request-tracker4 \
[23:52] <andygraybeal>  rt4-apache2 rt4-clients rt4-db-postgresql apache2-doc lynx \
[23:52] <andygraybeal>  postgresql apache2 libapache-dbi-perl fetchmail
[23:53] <andygraybeal> and i didn't get any chances to put in a password for postgres or answer any of the questions that request tracker asks....
[23:53] <andygraybeal> how do i do this over?
[23:53] <andygraybeal> i will try purge, but i'm afraid it won't work
[23:54] <andygraybeal> i'm so good at messing this kind of stuff up