[04:43] <d3wy> Howdy all, anyone free to give a quick hand with samba (Ubuntu) host with OSX client?
[07:48] <a5m0> is there a way to auto-login to root on console cli and run a program on boot?
[07:52] <mikal> /etc/rc.local is run at the end of each runlevel as root. You could put something there if you had the right safeguards?
[07:59] <lifeless> a5m0: why do you want to login on the console, why not just run it on boot ?
[08:00] <lifeless> a5m0: @reboot in crontab may be all you need
[08:00] <lifeless> mikal: ^ in case you didn't know ;)
[08:01] <mikal> :P
[08:03] <lifeless> I suppose you could set getty to bash
[08:04] <mikal> Or do something with a bonkers custom init...
[08:11] <a5m0> i just want it to boot up into a cli status program
[08:21] <cwillu_at_work> a5m0, look at /etc/init/tty1.conf
[08:23] <cwillu_at_work> I've done exec /sbin/getty -n -l /usr/local/bin/nobodytop -8 38400 tty6 on one of my boxes
[08:23] <cwillu_at_work> where nobodytop is just a wrapper that starts top as nobody
[16:00] <sliddjur> hello my dns server doesnt resolv domain name. only subdomain. what did i miss?
[16:20] <RoyK> sliddjur: sorry, I don't understand - does 'host google.com' work?
[16:21] <sliddjur> RoyK, nothing happens
[16:21] <sliddjur> i have tried to set up my own bind server. but when i ping my server (student481.linuxkurs.tfe.umu.se) i get no replys. but for example www.student481.linuxkurs.tfe.umu.se) works
[16:28] <RoyK> sliddjur: there isn't an A entry for student481.linuxkurs.tfe.umu.se
[16:28] <RoyK> roy@smilla:~$ host student481.linuxkurs.tfe.umu.se
[16:28] <RoyK> student481.linuxkurs.tfe.umu.se mail is handled by 10 mail.student481.linuxkurs.tfe.umu.se.
[16:54] <cloudman> sliddjur, try putting 8.8.8.8 in your resolve.conf  might be something temporary
[16:56] <cloudman> resolv.conf I mean
[16:57] <tedski> or 4.2.2.1 if you don't like the googles
[16:58] <cloudman> whatever works :)
[16:58] <tedski> true :)
[16:59] <cloudman> I had the problem the other day, it went away ??
[17:07] <i3luefire> http://pastebin.com/EEMDhFzh why does an upgrade want to remove my kernel?
[17:09] <tedski> it's only removing the kernel header packages
[17:10] <i3luefire> ok. good point. but is that not bad?
[17:10] <tedski> i3luefire: uname -r output, please?
[17:11] <tedski> it is not bad... if they're needed, they won't be removed
[17:11] <i3luefire> root@ubuntu-server:~# uname -r output
[17:11] <i3luefire> uname: extra operand `output'
[17:11] <i3luefire> Try `uname --help' for more information.
[17:11] <tedski> skip the output
[17:12] <tedski> i was asking for the output of 'uname -r'
[17:12] <i3luefire> 3.2.0-32-generic
[17:12] <tedski> so, you must have had a package that required the source of the kernel (kernel headers) and you may have removed it
[17:13] <tedski> aptitude realizes this and says it will remove those header packages
[17:13] <i3luefire> ok. thank you very much. it just seemed scary at first
[17:13] <i3luefire> http://arstechnica.com/gadgets/2012/11/how-to-set-up-a-safe-and-secure-web-server/2/
[17:14] <tedski> a scared, reluctant sysadmin is a safe sysadmin :)
[17:14] <i3luefire> that is a guide i am following
[17:14] <i3luefire> :) thanks
[17:15] <i3luefire> i may be back later i guess
[17:15] <tedski> looks like a good guide
[17:15] <tedski> i'm glad they explain the worker model in brief rather than just saying, "type apt-get install blah" and look for the It works! page!
[17:16] <i3luefire> oh. that sounds good
[17:18] <i3luefire> i am currently using apache for webmin and some other minor stuff
[17:20] <i3luefire> http://imgur.com/62crA
[18:11] <i3luefire> i think i am going to stop with the nginx install because i dont want to mess up my currently working setup with apache. and running them side by side seems too difficult.
[18:40] <anepanal1ptos> Hello, how do i change the metric of a dhcp gateway route?
[18:41] <anepanal1ptos> for example, in debian, the dhcp'd default gateway has metric 0, but in ubuntu it has metric 100
[19:03] <maxb> Mine doesn't seem to have any displayed metric at all ... does that just mean metric 1 ?
[19:03] <maxb> Although wouldn't the metric on a default route only be useful if you had multiple default routes?
[19:03] <maxb> Which doesn't seem to make much sense
[19:04] <qman__> it's a failsafe
[19:05] <qman__> if that interface or router goes down, you still have another route to the internet
[19:05] <qman__> the other purpose is, if you want DHCP on one interface and static on another, but you want the static gateway to take precedence
[19:26] <maxmahem> O
[19:27] <maxmahem> I've noticed that the 12.10 install seems to hang when detecting hardware if it has like 128mb of ram, am I correct on this or just not patient enough w/ it?
[19:28] <escott> maxmahem, 128MB is very very little ram, suggesting a very very old system. it might be that it hangs because it tries to access some kind of hardware bus that doesn't exist
[19:29] <maxmahem> escott, well in this case it's a virtual machine.
[19:29] <maxmahem> Hardware detection works fine if I increase it so say 512 or so.
[19:29] <escott> maxmahem, so give it more ram for the install
[19:30] <escott> maxmahem, you will want that extra ram during install just to make the install go faster
[19:30] <maxmahem> escott, I did :). I'm just wondering if this is expected behavior, or maybe I'm not being patient enough w/ the hardware detector when running on such a small amount of rma.
[19:30] <maxmahem> *ram
[19:31] <escott> maxmahem, i dont think anyone cares
[19:32] <escott> maxmahem, there is a javascript implementation of qemu. if you want to see something go slow you could install ubuntu in your web-browser
[19:33] <maxmahem> I'm not particularly interested in watching things go slow, I'm interested in the minimum memory required for the installer.
[19:33] <escott> !requirements
[19:36] <qman__> the requirements aren't hard, though; for example, I've run lubuntu on 128MB and the system runs fine
[19:36] <qman__> can't use youtube with that little RAM, but can't really expect that
[19:48] <anepanal1ptos> maxb: the problem is, i already have a 0.0.0.0 route advertiesed on my network
[19:49] <anepanal1ptos> maxb: and i want this machine to use the one that is local, as it is also connected to the internet.
[19:50] <anepanal1ptos> this is my debain route list
[19:50] <anepanal1ptos> 0.0.0.0         76.239.22.1     0.0.0.0         UG    0      0        0 eth1
[19:50] <anepanal1ptos> 0.0.0.0         10.4.254.1      0.0.0.0         UG    2      0        0 eth0
[19:50] <anepanal1ptos> as you can see, the advertised one has metric 2 (the bottom one) so the machine prefers it's local gateway
[19:51] <anepanal1ptos> but in ubuntu, the advertiesed route gets metric 2, and the local gateway has metirc 100, so it still prefers to go out another 'exit node' (so to say).
[19:51] <anepanal1ptos> idealy, i need to find the script that calls the dhclient and start from there (i think.)
[19:52] <qman__> you just need to configure your static address with a metric of 0
[19:52] <qman__> there are a few ways to do it
[19:52] <anepanal1ptos> no no, the internet address is dynamic
[19:52] <anepanal1ptos> it comes from a cable modem
[19:53] <qman__> so the internal address isn't static either?
[19:53] <anepanal1ptos> (sorry to break you/cut you off, i just didnt want you wasting your time in the wrong direction. i know how to do that via etc network interfaces)
[19:53] <anepanal1ptos> the internal is, and it has metric 0
[19:53] <qman__> then set its metric over 100
[19:54] <anepanal1ptos> and from the internal, via a routing protocal, it learns there is another gateway on the network
[19:54] <anepanal1ptos> so it gives the 'other gateway' metric 2.
[19:54] <anepanal1ptos> however this machine also has a local gateway, and for some reason, dhclient is adding it to the route table with metric 100
[19:55] <qman__> fair enough
[19:55] <qman__> sounds like it would be configured in the dhcp client somewhere
[19:56] <anepanal1ptos> yeap. ill rtfm for the rest of it, i just need a little push in which files deal with calling dhcp
[19:57] <qman__> might look into man dhclient, or /etc/dhcp3/dhclient.conf
[19:58] <qman__> it might also be different in the version you're using, I'm on 10.04
[19:58] <anepanal1ptos> i am also on 10.04
[19:59] <anepanal1ptos> hmm
[19:59] <anepanal1ptos> i think the magic is someplace else, my debian and ubuntu dhclient.conf look identical
[20:28] <maxmahem> qman__, for what I'm doing w/ this server, 128 is more than sufficent, it's just that setting the ram to one thing for the install then changing it once I get it up and running is kind of a hassle.