=== arosales_ is now known as arosales [01:54] Hello folks! [01:55] When I try to check for ruby at my server the following error message happens : https://gist.github.com/anonymous/9283668, what can I do? [01:56] webfox: interesting [01:56] sarnold: yes [01:56] webfox: command-not-found is a little helper program that will tell you if a command that isn't found is a small typo or if there is a specific package that you should install [01:57] webfox: please do file the bug report that it suggests -- but the short answer is you do'nt currently have ruby installed [01:57] yes, I already restarted the machine and the error still there. [01:57] sarnold: how can I check which is the latest ruby package available at apt-get repository? [01:59] webfox: apt-cache search ruby | grep ^ruby [01:59] webfox: (the grep is just to make the list more manageble :) [02:01] sarnold: \o/ thank you! [02:03] sarnold: man, it shows several versions. Do you use Ruby by any chance? [02:04] webfox: not much these days.. [02:05] webfox: which release are you on? we've changed which version is in main over time.. 1.9.3 is in main in precise, quantal, saucy, and trusty (maybe we'll drop trusty before release?) -- and ruby2.0 is in main in trusty [02:06] Not sure what version is running at this server. [02:06] lsb_release -a should say [02:06] I think those errors are Python related, gem -v command generates the same error. [02:06] (that should say something useful on any linux system) [02:07] webfox: chances are, you've got a locale set on your local machine that doesn't exist on the remote machine. but that's just me guessing.. [02:07] sarnold: that error is happening in this case as well. [02:08] Perhaps I should do a downgrade on Python.. [02:08] Currently running version 3.3.2 [02:08] webfox: that is unlikely to fix it; probably you can install a langpack or something similar to build your missing locale information [02:09] sarnold: do you mean for Python or the OS? [02:09] webfox: the os [02:09] how do I do that? [02:10] webfox: check out apt-cache search language-pack [02:10] langpack is not a known package [02:11] apt-cache or apt-get? [02:11] apt-cache [02:11] apt-cache helps you search your computer's local versions of the lists [02:11] 'apt-cache search' is easily 50% of why I switched to debian back in 1998 or 99.. [02:12] Oh, ok, there are tons of packages, I should filter this result somehow. [02:13] webfox: probably you want language-pack-pt -- you might also want language-pack-gnome-pt or language-pack-kde-pt if you use one or the other of those environments [02:13] (just guessing based on your hostname :) [02:14] sarnold: I don't use any UI, is it necessary to install kde or gnome? [02:14] webfox: no. skip those. :) [02:15] sarnold: how about if I install language-pack-us? [02:16] webfox: what LANG= do you have on that machine and your local machine? [02:17] sarnold: probably us-base. Is there acommand to verify that? [02:17] webfox: echo $LANG [02:18] en_US.UTF-8 [02:18] yeah this is the latest us-base language package. [02:18] oh wow, I hadn't expected that. [02:19] sarnold: yeah, I prefer english. [02:19] webfox: it seems most better admins do :) I had assumed pt_BR had leaked through from a client somewhere. sigh. :) [02:20] :) [02:21] sarnold: so, am I dead then? [02:22] webfox: nah, just some error messages will be extremely annoying [02:22] webfox: i'm out of ideas on making them go away -- follow the intructions there to file a bug report, hopefully someone else will know what to do to make those annoyances go away :) [02:23] sarnold: all right. Thank you for your help! [02:23] webfox: have fun :) [02:30] any reason to use 64bit OS if on a 2GB ram VPS? [02:31] MavKen: more registers, longer registers, better nx support in the page tables, and you can standardize on 64bit isos and tools and so forth [02:32] ok [02:39] can anybody tell me how to find the original mac address of cloned mac??? === onesandzeroes__ is now known as blargh____ [03:09] what is the most lightweight thing to install so that php scripts can send out mail? I have 12 different domains on a vps and want the scripts to be sent from their domain. I tried exim4 but all domains were appearing as from "domain2.com via domain1.com" only the primary domain would appear correctly [05:17] hia ll [05:18] why when i run this command: sudo a2ensite itsol.mx i get an error: ERROR: Site itsol.mx does not exist! the file exists: in /etc/apache2/sites-available itsol.mx thanks in ubuntu 12.40 worked, but in ubuntu 13.10 not any advice? [05:19] jak2000: no idea; can you run strace a2ensite itsol.mx and look through the output to see what it's doing? [05:24] sarnold: http://pastebin.com/SidFav4p [05:25] jak2000: cool! check out lines 1141 through 1145 [05:26] jak2000: I think you just need to rename the file to itsol.mx.conf [05:27] ok done and work [05:27] great :) [05:27] but its correct: in sites-enabled i see now a file: itsol.mx.enabled (in color cyan, the letters) [05:39] sarnold? [05:41] jak2000: yes? [05:42] when restart apache i get this error: AH00526: Syntax error on line 30 of /etc/apache2/sites-enabled/000-default.conf: AllowOverride not allowed here [05:42] jak2000: hrm, in apache 2.4 the rules may have changes [05:42] s/changes/changed/ [05:43] yes [05:43] :( [05:43] jak2000: looks like it can only be used in a block http://httpd.apache.org/docs/current/mod/core.html#allowoverride [05:45] comment the line 30 reload the apache and work [05:45] continue working [05:45] thanks sarnold [05:46] jak2000: cool; it would be worth making sure things are still correct.. [05:54] sarnold mmm bether uninstall 13.10 and install 12.04 [05:55] jak2000: uhoh, more problems? [05:55] yes [05:55] and reading reading [05:55] the 13.10 is unestable right? [05:56] jak2000: it'd be worth doing the reading eventually, because 14.04 will also have apache 2.4 [05:57] jak2000: but 13.10 will only be supported until july -- 12.04 will be supported for a while, you can make the transition on your own schedule if you use 12.04 [05:57] 13.10 should be stable, but 12.04 has been deployed a lot longer and has received more testing work [05:58] yes [08:58] i am using MRTG in ubuntu server i have already installed mrtg but when i try cfgmaker --global 'WorkDir: /home/http/mrtg' --output /etc/mrtg/mrtg.cfg public@ localhost to test there is an error is there any help please? === Guest65903 is now known as root_ === root_ is now known as SoulOfTheInterne === SoulOfTheInterne is now known as ToBeFree === _elia is now known as elia === a1berto_ is now known as a1berto [11:51] hi what's the simplest way to do an upgrade? from Ubuntu server in April, do-release-upgrade? [12:06] hi ppl [12:07] I want to change a permissions for a folder am I right in thinkink that this is what I need to run from cmd line. sudo chmod 0755 myFolder [12:08] also is 0755 and 755 the same? I've only ever seen 3 digits === freeflying is now known as freeflying_away === Mogwai is now known as Guest82922 === freeflying_away is now known as freeflying === freeflying is now known as freeflying_away [16:11] I just installed ubuntu on a new VPS. I am moving over 12 domains that have been on shared hosting. What can I install so that each domain can send out mail via their php scripts and it appear as coming from that domain? I tried yesterday and each virtual site was sending from the primary domain on the vps. Starting over today. [16:22] Trying to fix an old sad 10.10 server. It keeps hanging after it claims to be done with init, after it's remounted it's drives. [16:22] Any ideas? Even single user mode doesn't seem to work [16:23] kyle__: could you reinstall it with 12.04 or something? [16:23] that is - do you have its data on a separate partition? [16:24] RoyK: A not my server, B There's tons of (what looks to be) very bad custom code on here, and the person who wrote and maintained it has been gone for two years. [16:25] Most of the data is in /var, which is on a separate drive now. [16:25] vm or physical? [16:26] perhaps booting on a memory stick and doing an fsck on the thing would be a start [16:26] RoyK: VM. And I thought the same thing, so I booted from an iso image and fscked both drives. No issues with either (but it claims to have gone over 2k days without a check). [16:27] fsck -f ? [16:28] without -f it won't do much if the fs is flagged as ok [16:28] Hum. You know, I didn't do the -f. Will try again [16:34] MavKen: you can override php's From: address in your virtualhost configs, see http://serverfault.com/questions/165499/ [16:37] thanks [16:38] I have 12 sites that I am moving from a shared host to my new VPS... should I just create a user for each and then add public_html dir in each /home/user/ ? [16:57] That didn't do it. Is it possible the remount of / is masking /var? [17:18] any idea how to install 1204 on non-pae hardware? [17:21] Hi Guys, i am editing the sshconfig, Changed the port number etc on Ubuntu 14.10, Tried to disable the Root Login via SSH and keep getting an error message. [17:22] OliPicard: can I borrow your time machine? ;) [17:22] XD Why Roy? Is it on Automatically? [17:23] was just thinking about 14.10 ); [17:23] (; [17:23] ah ops 13.10 my bad [17:23] was thinking about 14.10 as well! [17:23] I've disabled root login on most of my boxes [17:23] works well [17:24] PermitRootLogin no < thats what ive set it to on my box but when i SSHD into the box it provides an error about a bad config. [17:25] PermitRootLogin no <-- that's from my sshd_config [17:25] perhaps you have some special characters there? [17:26] http://paste.reynir.dk/8 < thats my config :) [17:27] try to type the line in by hand - might be special characters around [17:27] also - why do you disable key auth? [17:27] and perhaps it wants its ciphers [17:27] thats what it came with when i got the server :) [17:27] Using a Digital Ocean Image :) [17:27] ok [17:28] just a hunch, I think PermitRootLogin goes in sshd_config, not ssh_config? ( you have the client config, rootlogin is a server option) [17:28] (per the first two lines of the comments/file) [17:29] shuno: Ah Yes Doh on my side thanks :D [17:29] sorry shauno [17:33] You really want to listen to RoyK about keys though. they're 'best practice' for very good reasons [17:36] How do you set the SSHD keys up? [17:36] Yup i agree, i normally wont be using SSH anyway :) i prefer to use a VNC [17:44] vnc to a server?!? [17:44] OliPicard: ssh-keygen & ssh-copy-id user@host [17:44] OliPicard: VNX, unencrypted, totally insecure, over the internet? [17:44] *VNC even [17:46] bekks: exactly - insanity [17:50] You dont need any security precautions then at all - that server will be taken over withing minutes. === chaos_ is now known as soahccc [17:56] Im double checking with the host [18:11] You should use ssh, honestly. [18:11] I will once i can get a stable connection thats the problem right now [18:11] i used to use PuTTY but my ADSL connection has been so bad recently. [18:13] Ok doing some research on the system it seems the VNC is on the HyperVisor its self. [18:13] What? No :) [18:13] Which hypervisor is that? :) [18:14] KVM [18:15] You can configure KVM to run VNC on localhost only, and NOT expose it to the internet. [18:26] hmm not herd anything back from the host, [18:32] any idea how to install ubuntu on non-pae? [18:36] one more question :D im editing the etc/sysctl.conf i noticed that some of the redefined items such as net.ipv4.tcp_max_sync_backlog=2080 have a gap after the = and some don't just wondering should i include the gap so for example backlog = 2080 [18:38] OliPicard: The gap doesnt matter, actually. [18:38] Awesome thanks bekks [19:05] I got it fixed. It was ureadahead. It caused everythign to choke up with /var on a separate partition than /. Ugh. And no (obvious) clean way of disabling it either. [19:08] hi. I manage the workstations and NFS server of a small university lab. Having /home shared is great because people can move around, but the desktops freeze if the NFS server goes down. [19:08] ancaster: That's kindof expected if the NFS server goes down. With their files & drive disappearing and all. [19:09] ancaster: Generally, you try and keep servers that are in use from going down.... [19:10] kyle__: Yeah... :-) I get that bit. Is there any way of making a fallback NFS server? [19:10] or another solution that doesn't use NFS at all? [19:10] Humm. I've never seen a way of doing a failback NFS server, but that's not to say it's impossible.... [19:10] hmm, okay. [19:11] Are you thinking of something like how windows does portable homes, so it syncs the users files on login locally, then syncs them back to the server on logout? That I haven't seen, but I suppose it's quite possible. [19:11] It's just not fun for anyone when the entire lab has to stop working whilst I sort out some problem. [19:11] Yeah, something like that could work. I was thinking of doing some rsync on logging in and out... [19:12] Or perhaps using gluster... but my experience with it hasn't be positive. [19:13] gluster is great in a few use cases, but outside of those, even when it __should__ work, it's painful or broken. [19:13] Heh. Good to know. [19:14] alrighty, so I'm not missing some bit of technology that everyone else knows about. :-) [19:14] I know you can use libpam-mount to mount a user's directory at the time they login... I wonder if it's possible to use that, or some other part of pam, to sync their directory... [19:14] Nope :) [19:15] You can have a backup nfs server that just takes over the IP of the broken one. [19:15] There's nothing identifying the server other than its IP. [19:16] Is it common to have the entire user's home directory mounted? We find there is constant traffic from gnome (just moving a window around even). [19:16] It's usually the whole home dir, yes. [19:17] soren: interesting. is there software you recommend that does this (IP failover)? [19:18] ancaster: Well, the plan only really works if you have the data synchronized between the two servers. Do you? [19:18] Otherwise, I imagine things will get.. interesting. [19:19] you should be able to knock something together with drbd (keep two nfs servers in sync) and heartbeat (for failover)? [19:19] soren: :-) true. well, I don't have anything in place to do this at the moment. [19:19] ancaster: Do that first. :) [19:20] soren: shauno. :-) okay. [19:20] It looks like you can use the pam_script module to run login and logout hooks per user. May be able to play with that. [19:23] kyle__: oh cool. thanks. something about this idea seems less complicated maybe. === zotta is now known as raray === raray is now known as raaray === raaray is now known as rarray [19:28] thanks all. I have some ideas to play with now. === cppCzar_ is now known as cppCzar [20:39] HI Guys, does anyone here have experince compiling sqlite for ubnutu? [20:39] on node.js [21:26] try #Node.js [21:37] Hey folks, I made a git noob error and overwrote a working copy. I've tried rebooting into a livecd and running extundelete (ext4 on an lvm), but no dice with --restore-all --after [date]. I tried debugfs -w and lsdel says there are no deleted Inodes in that directory. Am I SOL? Is there something I haven't tried or am missing? [21:38] psyferre: you need old commit from git? [21:38] xnox: nope - wasn't that smart. I didn't have my working copy committed. [21:39] psyferre: did you atleast $ git add ? files you wanted? [21:39] I was trying to resolve a merge problem and reset back to HEAD, forgetting that I had uncommitted changes on one file that I really wanted to keep. [21:40] I'm rather new to git (always used subversion) and just forgot about the uncommitted file. [21:41] So it was overwritten by what was in the repo and I lost a large amount of work. Any other way I'm missing to get at the old contents? [21:42] psyferre: Git doesn't do magic. It downloads and writes files, and it uploads files. It doesn't hide your old files someplace else. There may be some sort of undelete utility that could work, but that's a stretch. [21:49] kyle__: right - the undelete utilty is what I'm looking for. After doing some research I used sysrq to do an emergency sync, then umount/remount read-only. After some more research, I booted to a live CD and used extundelete, which only managed to undelete .git in that directory. I also tried running debugfs, but lsdel says there are no deleted inodes in that directory. [21:49] I'm casting about for anything that I could try that I don't yet know about. [21:50] psyferre: This one may help, again it's a stretch. http://www.cgsecurity.org/wiki/PhotoRec [21:50] There's one more I'm wracking my brain for, I'll let you know when I remember it's name. [21:51] kyle__: Awesome, thanks! I'll give it a try. [21:52] Hu... There is also an extundelete, and it's even in the ubuntu repos... [21:52] Haven't tried that one. [21:54] Yup, tried extundelete first - no dice, unless I've used it incorrectly [21:55] You probably didn't :/ Most of these are designed for the situation where you accidently delete or format a file you need. Not the situation where you wrote over it. If it's made as a new file, there's a chance the file overwritten will still exist, but it's not a terribly great chance. [21:57] Yeah... I figure it's a longshot. If I got lucky it'd save me several days worth of work, though. =-/ [22:01] Ouch. Sorry man. If you have a separate machine with a buch of space on it, there are things you can do to prevent this sort of problem in the future. HOldon, finding article... [22:03] Not the exact one I was thinking of, but this may help for the future. http://blog.interlinked.org/tutorials/rsync_time_machine.html [22:04] This is the one I was thinking of. It does essentially the same thing, but explains every working part in detail, so it's worth a read even if you don't want to do it. http://www.mikerubel.org/computers/rsync_snapshots/ [22:05] Thanks, kyle__. Most of my other servers have amanda backing up everything important every night. This was a new machine for a project that hasn't gotten to staging yet, so I hadn't set that up yet. Live and learn, I suppose. [22:06] psyferre: That second link is what I've done when I had several, shall we say, bad users, who would accidently delete things when they were panicked. [22:07] I had it take hourly snapshots which it kept for 7 days, Saturday through Thursday. And on Friday, since it was the day they screwed up the most, I had it do them every 15 minutes. [22:07] lol [22:07] That's fantastic. [22:07] Big samba server, used by everyone in the company. Saved their hides more times than I can count, and saved me a ton of trouble. [22:24] night [22:58] Hi! Various utils on my Ubuntu 12.04 server cant fetch things via HTTPS. `npm` and `docker pull` are examples. curl works fine, and HTTP urls work fine. Any ideas? [23:03] matt2000: http://i.imgur.com/3ezM6.jpg [23:05] more like http://codeworkx.de/wordpress/wp-content/uploads/2011/12/jacoj.jpg [23:05] sry [23:05] google images fuzzed me