=== eshlox_ is now known as eshlox [01:23] hi guys...any idea how to install Ubuntu desktop on 50 desktop unit on one time only..? any idea..? [01:26] ruben231: investigate 'preseed files' === andol is now known as Guest52726 [01:45] Hi all === TheLordOfTime is now known as teward [05:16] To switch from ubuntu-desktop to ubuntu-server, you'd just uninstall ubuntu-desktop. Right? [05:17] (After you edit /etc/default/grub and change GRUB_CMDLINE_LINUX_DEFAULT="quiet splash" to GRUB_CMDLINE_LINUX_DEFAULT="text" and then run sudo update-grub Which will do away with GUI.) [05:18] sudo service lightdm stop and do ^^^^^ and then sudo apt-get remove ubuntu-desktop [05:18] Does that sound ok? [05:19] usr13: you don't need to remove "quiet splash" but it can help. "splash" is the trigger for the plymouth splash screen, "quiet" reduces the verbosity of the kernel messages [05:20] I removed quiet splash and changed it to text and that worked well... [05:20] TJ-: Ok.. so change "quiet splash" to "quiet text" ? [05:20] TJ-: Is that what you recommend? [05:20] usr13: On servers I replace "quiet" with "debug" since I want to see as much as possible at boot-time [05:21] ok "debug text" ? [05:21] usr13: I don't usually add "text" because if the VGA adapter is capable enough, I have it switch to a high-res graphics mode to get more text on-screen [05:21] Oh, so all you need is "debug" [05:22] usr13: yeah, not need either, its just helpful if/when something goes wrong to see as much as possible [05:23] Ok then, it would be sufficient to just remove "quiet splash" ? [05:23] usr13: :) yes, as I said, its not necessary more a question of personal preference [05:24] Yea, ok. Thanks for the info... [05:24] Crap, when I change it to "debug" the damn login screen comes up.... :( [05:24] If I have it at "text" it goes to the console... [05:25] shreezbot: It will, if you've got lightdm installed and enabled [05:25] shreezbot: Well, you have yet to uninstall ubuntu-desktop [05:25] shreezbot: Right? [05:25] Oh yea, you'd need to dissable lightdm. Or uninstall it... [05:25] usr13, Correct, I haven't uninstalled ubuntu-desktop yet... [05:25] "text" disables lightdm [05:25] Yea that's what I thought. [05:25] AHH! [05:26] TJ-, So I can just uninstall lightdm? [05:26] Do I need to uninstall ubunut-desktop also? [05:27] shreezbot: You don't need to, just override it [05:27] TJ-: Yea, but he no longer needs any of it... right? [05:27] So why not uninstall it.....? [05:29] TJ-: I'm try9ing to think.... what is the scipt that starts all the GUI stuff? [05:29] What's the name of it...? [05:29] usr13: "sudo service lightdm start" [05:31] well yea, sudo service lightdm stop will dissable it, but only for one session. Right? [05:31] yes [05:31] It must be /etc/init/lightdm.conf that I'm thinking of... ? [05:32] no, that's not it... [05:32] to disable services you'd do "sudo update-rc.d lightdm disable" [05:33] TJ-, So when I configure grub to just use debug, when my server starts up, should I eventually get a console prompt? [05:33] TJ-, It currently just sits there at "Stopping Mount filesystems on boot"... [05:33] TJ-, I'm able to SSH into the server just fine though and do everything I need to do and all of my applications have started up correctly. [05:34] shreezbot: Yes, it sounds as it the getty jobs aren't starting the ttys [05:36] TJ-, Weird... It started doing that after I removed lightdm... :( [05:39] shreezbot: check the /etc/init/tty{1..6}.conf scripts exist [05:40] TJ-, Should GRUB_TERMINAL=console be uncommented in my /etc/default/grub? [05:40] shreezbot: the only other thing I can think of, is the kernel command-line's "vt-handoff..." option and plymouth splash screen stay on VT7. In which case pressing Alt+F1 should get you to the login on tty1 [05:41] Seems to me that it would be a lot easier to uninstall some stuff, particularly ubuntu-desktop [05:41] I just found this: http://www.darrinhodges.com/converting-ubuntu-12-04-lts-desktop-to-server/ [05:42] If you are permanently moving to server, uninstalling the packages that are no longer needed just seems like the thing to do. [05:43] SWEET! [05:44] usr13, I've looked everywhere for instructions like that... :) [05:44] Well, there you go.... [05:44] (My google-foo is working...) [05:48] shreezbot: And set to "static IP and remove that pesky network-manager as well" [05:48] usr, this damn thing is going to be a server yet! :) [05:49] Of course, if you have a router giving it a static lease, you only need to uninstall network-manager. [05:50] because dhcp is default. At any rate, from now on, just use the config file /etc/network/interfaces for network configurations [05:51] http://www.cyberciti.biz/faq/setting-up-an-network-interfaces-file/ [05:52] Better one: https://wiki.debian.org/NetworkConfiguration [05:57] I get a message about "login: mounall: Disconnected from Plymouth" and no console.... :( [05:58] I'm able to "CTRL+ALT+F2" to swap to tty2 and so on, but tty1 is hung up... :( [05:59] shreezbot: ahhh, that is interesting [06:00] TJ-, Tell me about it. [06:02] shreezbot: "ps -efly | grep tty1" [06:05] TJ-, It says it is running [06:06] /sbin/getty/ -8 38400 tty1 [06:15] shreezbot: "grep DELAYLOGIN /etc/default/rcS" [06:15] =no [06:16] Shit, I broke my network removing the network-manager... I'm wondering if that could be the problem... [06:19] and of course I can't re-install the network-manager because I can't connect to the internet... :( [06:20] I maybe should have done that... [06:20] shouldn't [06:20] shreezbot: The packages should still be in "/var/cache/apt/archive" so you can simply "sudo apt-get install network-manager" [06:21] TJ-, I'm an idiot, so I purged them... :( [06:23] Looks like I can't get an IPv4 address for some reason... [06:24] shreezbot: You'll have to do it manually then "sudo ifconfig eth0 1.2.3.4 netmask 255.255.255.0" "sudo ip route add default dev eth0" and then add a DNS nameserver entry to /etc/rsolv.conf [06:25] shreezbot: How are you doing it? "sudo ifconfig eth0 up" then "sudo dhclient eth0&" [06:26] I edited /etc/network/interfaces [06:26] lol, I'm so fucking dumb sometimes [06:26] I did a sudo -i not realizing that it changed my working directory to root [06:27] So I created a file called interfaces there, which needless to say did fuck-all when I attempted to bring up the interface... [06:29] Well, that wasn't it though... It still isn't getting an IPv4 address... :( [06:33] Ok, I got it working now... [06:33] Incidentally, that fixed my stupid "Disconnected from Plymouth" as well! :) [06:34] Sometimes I just have to fumble my way through... :) [06:34] shreezbot: if you just want NM back, you can probably dpkg -i /var/cache/apt/something.. [06:35] sarnold, I actually don't mind it being gone now that I know how to properly configure /etc/network/interfaces... :) [06:35] shreezbot: excellent :D [06:35] sarnold, Yea, I'm super stoked now that I've converted this desktop installation to server... [06:35] Been wanting a way to do that for a long time now... :) [06:36] sarnold: if it's in /var/cache/apt/archive/ then you can use apt-get (which will resolve dependencies) [06:36] TJ-: that's nice but if the network is down, fixing the rest of it just to get it back could be a hassle [06:36] of course, in server land, life without NM is probably the best way forward anyway :) [06:37] sarnold, Yep! That's what I'm working toward... This is a learning experience for me, so mission accomplished tonight! :) [06:37] I've learned about 30 new things tonight! :) [06:37] haha, nice :) [06:39] I'll say, you guys are amazingly helpful... :) [06:39] TJ-: nice work :) [06:40] So what do you guys recommend for monitoring Ubuntu? [06:40] Mostly for thinks like CPU utilization, memory utilization and things like that... [06:42] setting up something like that has been on my todo list for ages. there's 'statistics' things like collectd and .. munin? and then there's service alerting things like nagios / icinga, shinken, and I just found out about assimilation project a few weeks back that looks pretty cool === Guest52726 is now known as andol === freeflying is now known as freeflying_away [08:25] hi guys i have 50 PC and we would liek to install desktop ubuntu on it any solutions to install it on network at the same time..? any idea..? === freeflying_away is now known as freeflying === freeflying is now known as freeflying_away === freeflying_away is now known as freeflying === freeflying is now known as freeflying_away [09:08] hi guys need help, just freshly install ubuntu server 12.04 LTS ----> http://pastebin.com/T2wbCMyn <--------------i get this error when i update [09:08] any help..? === freeflying_away is now known as freeflying [09:11] ruben check the contents of the downloaded .gpg files, they're stored in /var/lib/apt/lists/ ... I bet you'll find they contain some HTTP proxy/cache message rather than the GPG signatures they should contain [09:13] TJ- what to do next..? after checking [09:14] this is the content --> http://pastebin.com/TKpBQeZz [09:14] ruben23: if the files contain bad data, fix the proxy/cache that is giving out that data [09:15] ruben23: You'll need to look *inside* those gpg files to see if they have valid contents. The "file" utility will help you there [09:19] http://pastebin.com/Ts9aEpZZ <----------------------my gpg file [09:20] ruben23: So now fix it [09:22] TJ-: i tried putting it into DMZ on my router stil the same..any idea how to fix..? [09:23] Use the information from the server that is saved into those files === freeflying is now known as freeflying_away [09:49] TJ-: it ok now [09:49] thanks [10:09] Hello all. I've got an internal dev server which has been set up with public key authentication for ssh and password authentication disabled. I've edited sshd_config to disable public key authentication and enable password authentication but so far, no luck in logging in with a password. Am I missing something? [10:10] ah ha. it works now [10:10] :D [10:10] zul, hallyn: are you aware of the libvirt FTBFS ATM? I have a trivial patch for libvirt; do you mind uploading it at the same time, please? http://paste.ubuntu.com/6914206/ [11:13] hello there friends [11:13] i've come across a rather unfortunate situation i'm afraid [11:14] i don't know anything about the history of this server, i didn't set it up [11:14] but i'm getting this error: [11:14] https://dpaste.de/O26t [11:31] morning [11:32] I'm not sure if it's the right place to ask, but I have a bug to fix in juju-core https://bugs.launchpad.net/cloud-archive/+bug/1240667, which includes setting a lower priority on the cloud-tools pocket during bootstrap, and then using --target-release with apt-get install to specify the exact archive to use (i.e. cloud-tools, not main) [11:32] Launchpad bug 1240667 in juju-core "Version of django in cloud-tools conflicts with horizon:grizzly" [Critical,In progress] [11:32] so the question is: what's the expected --target-release value format for cloud-tools? "precise-updates/cloud-tools" ? [11:33] i figured someone with more in-depth apt experience might know [11:33] !cloud [11:33] Ubuntu Cloud Infrastructure is a ready to deploy Infrastructure-as-a-Service (IaaS) based on OpenStack. See https://help.ubuntu.com/community/UbuntuCloudInfrastructure for further details. [11:42] is sftp enabled by default with openssh? [11:57] martisj: with openssh-server, yes [11:58] mardraum: I see. [11:58] I can't seem to wrap my head around how to setup access via sftp to the apache document root and then have those files be accessible from www-data [11:59] It should be trivial and I should know how to do this, but I don't :S [11:59] point apache and your user's home at the same place? [11:59] or just use cd commands to move around [12:00] sftp in, cd / [12:00] you are at root. [12:01] there are many ways you could do it, even a symlink [12:02] as for "www-data" sounds like you are struggling with permissions? [12:09] mardraum: yes permissions indeed [12:09] ls -al shows that the user and group is the same as my username and group [12:09] demo:demo [12:09] where demo is my username [12:54] any server guys available? [13:03] just ask the question [13:04] I asked some time ago [13:05] ikonia, the question is: what's the expected apt-get install --target-release value format, so it will pick packages from the cloud-tools pocket, rather than main? [13:05] sorry, I don't know the answer [13:06] ikonia, ok, sorry then [13:06] no need for sorry, I just don't know the answer [13:16] dimitern: it should be automatic, since versions in the cloud-tools pocket are higher than the versions in main. I think. Do you have a specific example where this is not working? [13:18] rbasak, yes, that's exactly the issue with bug 1240667 [13:18] Launchpad bug 1240667 in juju-core "Version of django in cloud-tools conflicts with horizon:grizzly" [Critical,In progress] https://launchpad.net/bugs/1240667 [13:19] rbasak, so I need to add cloud-tools with lower priority and redo all apt-get install commands in cloudinit to use --target-series "precise-updates/cloud-tools" [13:20] rbasak, so we won't mess with charms deployed on the same machine and trying to install packages from main, but now getting them from cloud-tools instead (i.e. django as described) [13:22] dimitern: I was unaware of that bug - thanks. [13:23] dimitern: so --target-series "precise-updates/cloud-tools" doesn't work? [13:24] rbasak, yeah, it's somewhat recent and came up when people started complaining openstack horizon is not working when deployed on machine 0 (on askubuntu.com) [13:24] rbasak, haven't reached that point, still writing the fix, it should work [13:25] * rbasak wonders if lxc and mongodb could specifically be put in precise-backports or something. [13:51] hi guys [13:51] HOw in the name of god do i upload files with sftp to apache docroot [13:52] while still having the files owned by www-data [13:54] I start a service like service "service solr start", which starts solr as daemon. But i want to see it's output. [13:54] Is there a flag or so, to not-start-as-daemon? [13:58] martisj: easiest to focus on groups instead. Look up the directory sgid bit. [13:59] shredding: it's service-dependent. Look in /etc/init/solr.conf for what it does currently, and maybe at the solr manpage for how to change it. For debugging, it's often possible to run a daemon manually without going into the background, but that depends on the daemon. [13:59] rbasak: thx [13:59] rbasak: i'll look into it [14:08] I keep getting permission denied public key,apssword [14:08] it's the most diffuse message [14:08] most frustrating indeed [14:22] rbasak: Hm, it did not work, there is no solr.conf and the solr file itself is weird. [14:23] I'm going to complain to the guy who set up the vagrant box. [14:23] shredding: oh, it must be a SysV script then. Try /etc/init.d/solr === freeflying_away is now known as freeflying [14:59] rbasak: hm, that's not libvirtd, but every kvm vm that will have access to every uvtool backend. [15:00] oh, no, nm :) [15:00] looks good, thx. [15:00] zul: are you fixing up the ftbfs? (comes from the nwfilter patch) === freeflying is now known as freeflying_away [16:00] anyone using ubuntu cloud images ? [16:00] the disk seems to be 4 GB which is pretty small [16:02] with virtualbox, disk are (or can be) COW [16:02] and the documentation states it: http://docs.vagrantup.com/v2/boxes/base.html : "When creating a base box, make sure the user will have enough disk space to do interesting things, without being annoying. For example, in VirtualBox, you should create a dynamically resizing drive with a large maximum size." [16:02] utlemming: would it be possible to raise the disk size of vagrant boxes ? [16:02] utlemming: should I create on issue on launchpad ? [16:06] any server meeting today ? [16:08] caribou, I was thinking the same thing [16:08] arosales: great minds... :) [16:08] adam_g, is listed as the chair [16:08] jamespage, smoser ^ [16:09] per https://wiki.ubuntu.com/ServerTeam/Meeting [16:09] i'm not really here, sorry. [16:09] :-( [16:09] have to run. [16:13] Guess that is a no... [16:13] :) [17:03] can anyone quickly let me know what the backup user does on ubuntu-server [17:05] thesheff17: not much, I don't think. The user is part of the base system. Perhaps something minor uses it; I don't know. [17:06] rbasak: yea its weird my data center just called me...logged onto the server and the backup user was running tons of weird crap like httpd [17:06] the only exposed service is ssh [17:06] thesheff17: sounds like your server might be compromised. [17:07] rbasak: yea :-/ [17:23] hello [17:24] im making a ssh server1 "tail -f /var/log/apache/error.log" >> /tmp/error.log [17:25] but this keeps hanging when network flaps, is there a way to make it die when network flaps, so i can make a watchdog in cron [17:30] jamespage: I just looked at bug 1243076. AFAICT, upstream is dead, Debian is dead, and the code is fundamentally not ported to the Apache 2.4 API. OK to just drop the package (it's in main)? [17:30] Launchpad bug 1243076 in mod-auth-mysql "libapache2-mod-auth-mysql is missing in 13.10 amd64" [High,Confirmed] https://launchpad.net/bugs/1243076 [17:31] rbasak, I would say so yes [17:31] OK thanks. I'll explain on the bug. It seems that many people think that it's required for a LAMP stack because they're cargo culting some instructions that install it for no useful purpose. [17:35] utlemming: just in case you see that here, I just created this : https://bugs.launchpad.net/ubuntu/+bug/1278989 [17:35] Launchpad bug 1278989 in ubuntu "vagrant cloud images disk is too small" [Undecided,New] [17:35] xerxas, you can just grow it. [17:35] ubuntu cloud images are not a package, and I cannot asign it to you ... [17:36] smoser: with ext2resize ? [17:36] smoser: that's not the way other boxes works ... [17:36] do whatever operation you would do in virtual-box to make the disk bigger "physically" (ie, add 10G of zeros to it) [17:36] and cloud-init will resize the root partition to take it. [17:36] thats how things "should" work. [17:36] smoser: these are vagrant boxes, it's supposed to abstract the use of virtualbox [17:36] you'd rather people deliver you gigabytes of zeros ? [17:37] smoser: when you export vagrant base boxes they are supposed to grow === mjohnson15_2 is now known as mjohnson15 [17:37] heres the basic problem [17:37] no matter what size utlemming thinks is "big enough". [17:37] for some work load its not going to be big enough [17:38] smoser: and no, I don't want people to download Gigs of zeroes, virtualbox manages the copy-on-write , it's written in vagrant docs, and commented in my issue [17:38] smoser: yes , but I used a lot of different boxes and all boxes generally built are 10 or 40 gigs [17:38] so the better solution than making a box larger is to fix vagrant [17:39] smoser: I'm not going to fix vagrant, and in the meantime, as we already deliver 4 gb boxes that are files that takes ~ 300 MB , we can easily grow to 10 Gb like most of people do [17:39] vagrant needs to know how to grow disks. [17:39] or, we just can shrink the space de 300 MB [17:39] smoser: no it doesn't, virtualbox does it [17:40] virtualbox certainly can do that. [17:40] xerxas: do you mean resize2fs? [17:40] vagrant needs to learn how to tell virtualbox to do it. [17:40] its the equivalent of: [17:40] RoyK: no , virtualbox does it , you create a 10 Gb filesystem [17:40] qemu-image resize my.img 100G [17:40] then magic is in place the rest of the way to do the right thing. [17:40] but It's just using the needed space on your disk [17:40] vagrant is the thing that is missing function here. [17:40] don't know how virutalbox does it though [17:40] but this is something I saw a lot of time [17:40] EC2, openstack, cloud-X, cloud-Y, .... all have this. [17:41] I have some contribution on vagrant (I think, small ones) and I'm using it for 2 years ... [17:41] you just add zeros to the end of the disk. [17:41] smoser: gotta go , but I'll come back to you, or , can you comment the issue ? [17:41] I'm left, I'm late ! [17:41] thanks! [17:42] xerxas: k [18:07] Hey all, having an issue with NFS, anyone able to help? [18:08] mount.nfs: mount to NFS server 'server:/directory' failed: RPC Error: Program not registered [18:09] Maddeth: iirc, nfs requires an rpcbind daemon or similar on both client and server [18:10] sarnold: they both do, this is after a load of issues over the weekend, they worked before [18:11] Maddeth: hrm. are there any new firewall rules on either machine or routers between? [18:12] on other servers I have just done /etc/init.d/nfs restart and that has cleared the same problems as shown here : http://www.whynix.com/2012/11/06/349/how-to-fix-mount-to-nfs-server-system-failed-rpc-error-program-not-registered/ [18:12] sarnold: nothing has been changed, and I can ping between the servers, no issues with DNS either [18:13] Maddeth: is it exportfs that shows you which filesystems are exported? (sorry, it's been a decade..) [18:15] but, there is no nfs inid.d file, only nfs-common and nfs-kernel-server, sarnold also, the server is showing up in exportfs on the server I am tring to mount from [18:16] Maddeth: how the nfs server is started is different from distribution to distribution [18:17] Maddeth: nfs-kernel-server kicks off the kernel thread that does the nfs server work. maybe stop it and the nfs-common, then restart them both again? [18:18] sarnold: have tried that a few times unfortunately, also bounced the server a number of times today [18:19] Maddeth :( [18:20] sarnold: not been a fun week :) this is just the tail end of a shed load of issues I had over the weekend [18:21] Maddeth: cripes and it's only tuesday. [18:21] Maddeth: is there anything in dmesg or other logs? [18:21] sarnold: replaced some switches, iSCSI dropped and we lost a raid disk, about 100 servers needed thier disks repairing [18:21] sarnold: not anything useful [18:21] Maddeth eek. bad weekend indeed. [18:21] :( [18:22] worked 10 - 3am Sunday, was in again at 7:20 on monday to fix more issues [18:22] as in 10am - 3am [18:24] sarnold: nearing the point where I give up and rebuild from scratch, need to update all out 10.04 servers anyway... [18:25] Maddeth: yeah at some point you've got to do what it takes to move things along. :/ [18:25] Maddeth: before you do, it might be best to file a bug against nfs or linux, it might yet be something that can be found and fixed, even if not on a timeframe that's useful to you [18:26] sarnold: been 5 hours at this already :) would have probably been faster to rebuild from the start [18:26] Maddeth: yeah, if only you'd known it was going to be a fight.. [18:27] sarnold: thanks for the help though [18:28] Maddeth: good luck [19:35] sarnold: I got it working [19:35] Maddeth: oh? what was it?? :) [19:35] nfs on the other server needed a restart -.- [19:35] nfs on a -third- machine needed to be restarted? [19:35] sarnold: spent 6 hours looking at the wrong server :) [19:36] Maddeth: ah! [19:36] sarnold: thought I would let you know :) [19:36] Maddeth: an easy mistake to make after spending 18 hours working under pressure :) [19:36] true [19:36] still a tiny bit annoyed at myself [19:36] Maddeth: thanks for the update :) now time to curl up under the desk and unplug the phone for a while, hehe [19:37] haha [19:37] sarnold: if only, got 20 servers that the backup's are failing now :) [19:37] Maddeth: oh jeeze. what a week. good luck! [19:37] sarnold: it's all ok, last night for 4 days ;) [19:38] sarnold: it's the other guys problem tomorrow! [19:38] Maddeth: woo! === klaas_ is now known as klaas === nxvl_ is now known as nxvl === huats__ is now known as huats === thumper is now known as thumper-afk === nxvl_ is now known as nxvl === justizin_ is now known as justizin === thumper-afk is now known as thumper [23:18] Hi, Is there any easyway to connect ubuntu / machine with gui to openvpn? on the server there is pfsense, and it only export client for windows