=== _Techie_ is now known as _TechAway_ [00:14] Hey guys quick question. I'm making a ubuntu network based on an older machine. I was planning to virtualize the network, but, for the base machine should i install the server software directly or could i run virtual box off of puppy linux what seems like the best way to go? [00:15] Would it surprise you to find that people here might recommend running Ubuntu as the base distro? [00:16] not really [00:16] but just curious as to if i was virtualizing [00:16] but that works thanks [00:22] jeeves_: Hi! [00:22] Would you mind leaving my nick alone? :) [00:22] jeeves_, mine should be "Jeeves_Moss" === jeeves_ is now known as jeeves_moss [00:22] Thanks :) === Jeeves__ is now known as Jeeves_ [00:23] Jeeves_, better? [00:23] Much :) [00:23] ty [00:23] Jeeves_, cool [00:23] Jeeves_, got a sec to see if you can help me take a stab @ this script? [00:25] a sec, maybe [00:25] But it's currently 1:25 AM where I live, so I'd like to go to bed soon :) [00:29] Jeeves_, ok, one sec [00:30] Jeeves_, it's only d-loading the first index.html page (not finishing the list of URLs in the URL input text file), and it's only doing the first page of the index.html it's pulling [00:32] jeeves_moss: I've just joined again [00:32] so i've got no clue what you're talking about :) [00:33] Jeeves_, it's only d-loading the first index.html page (not finishing the list of URLs in the URL input text file), and it's only doing the first page of the index.html it's pulling [00:35] That's what you said before [00:35] I've got no context, no script [00:36] oh, sorry. one sec [00:36] So i've really got no clue what you're talking aout :) [00:36] Jeeves_, http://pastebin.ca/1960490 [00:42] Hmm, that strange [00:42] +'s [00:42] Jeeves_, ??? [00:43] That should work, if you ask me [00:43] # [00:43] $WGET wget -r -p -nd -nc -l"$DEPTH" -t5 -H --domains=images.4chan.org -A.jpg,.jpeg,.gif,.png,.pix -erobots=off "$URL" -e robots=off [00:43] That does /usr/bin/wget wget [00:43] which isn't good [00:45] Jeeves_, ok, crap. you know it's been a LONG day! [00:46] Jeeves_, any ideas on why it's not following the page links? I want it to scrape the images out of images.4chan.org ONLY. [00:46] :) [00:46] -erobots=off [00:46] You say that twice [00:47] Jeeves_, and the recursive scrape of the linked page under the start? [00:47] No clue [00:48] Never been looking for hot girls like this :) [00:48] s/hot/stupid/ [00:48] ok, thanks. I'm trying to site scrape as a "proof of concept". So far, I can't get it to scrape anything further than the seed page [00:48] jeeves_moss: Sure :) [00:49] Jeeves_, lol. I'd like to find a way of writing all of the file names to a flat file so I can skip the ones already d-loaded. [00:50] Jeeves_, thanks again for your help. I think I can find a "round about" way to make it do what I want. [00:53] Can I configure Server with a root account and get rid of sudo, or will that break stuff? [00:59] njbair: That will not break stuff [00:59] So, yes you can! [00:59] I imagine just `sudo passwd root` and then `apt-get remove sudo` should do it, eh? [01:05] can anyone here help me restore a borked system upgrade? [01:05] the process is still running though, i can see it via ps aux [01:05] root 26055 0.0 4.9 42856 25088 hvc0 S Oct11 0:00 /usr/bin/python /tmp/tmpzQNQJU/mave [01:05] rick --mode=server --frontend=DistUpgradeViewText [01:06] how do i resume that process? [01:12] anyone? [01:25] I'd suggest let it run if it shows signs of doing anything? [01:28] no [01:29] it was in an ajax console for a vps (linode.com) and i thought originally that i killed the process...with a ctrl-z, but i paused it [01:29] so then i accidentally [01:29] err i ran the command fg which showed the process and ctrl-c out of that [01:29] so i have no idea how to resume this process... === Pilif12p|afk is now known as Pilif12p [01:46] New bug: #659602 in mysql-dfsg-5.1 (main) "package mysql-server-5.1 5.1.41-3ubuntu12.3 failed to install/upgrade: Package is in a very bad inconsistent state - you should reinstall it before attempting a removal." [Undecided,New] https://launchpad.net/bugs/659602 === jjohansen is now known as jj-afk [02:20] ok i'm trying to resume a borked system upgrade...attempting $ sudo dpkg --configure -a results in this error [02:20] dpkg: status database area is locked by another process [02:21] how can i resume the system upgrade? [02:22] is it safe to remove /var/lib/dpkg/lock? [03:11] hello i just upgraded to from 10.04 32bit and now when i run vnc-client from windows to ubuntu-server, im not able to connect to the desktop, I went to Adminstrations or Preferences > Remote Desktop and it appears to freeze once i set allow access to this desktop and set password for it. I've even installed x11vnc as well but that was not working as well, what to do ? [03:11] edit: upgraded from 10.4 to 10.10 32bit [03:11] RCola, you available to assist me again [03:12] is php5 =5.3 ? [03:12] what ver of php is in the pkgs now [03:13] I need php5 [03:13] really 5.3 and 5.3 fpm [03:17] yes it is 5.3.3 [03:18] I dont fins fpm [03:18] why does php5 install apache [03:19] I only want phpe and some modules [03:19] using nginx and not apache [03:19] its wrong to force apache on a install [03:19] is there a way to only install php5 without apache [03:20] i agree, i dont like that either [03:21] is there no way to reject parts and just get what you need [03:21] i dont think mine installed apache but i did see it installed some apache files [03:21] The following extra packages will be installed: [03:21] apache2-mpm-prefork apache2-utils apache2.2-bin apache2.2-common libapache2-mod-php5 libaprutil1-dbd-sqlite3 libaprutil1-ldap libcap2 php5-common [03:21] ssl-cert [03:21] thats just wrong [03:22] yea that sucks.id like to remove all that extra crap too [03:22] yeah [03:22] who are the pkg devs [03:23] it needs to be delt with [03:23] try in #ubuntu-dev [03:34] is there a plan to upgrade php to 5.3.3 on ubuntu-server 10.04 [03:35] rneese: a full version upgrade generally doesn't happen after release [03:35] rneese: If you don't want apache, don't complain, just don't install it [03:37] wemaxb the issue is php should be its own pkg and not force apache with it [03:37] It is. [03:37] then the meta php5 pkg [03:38] And php5-fpm is available in Ubuntu 10.10 [03:39] rneese: the meta package is for people that just want the defaults, if you want something else, you need to install the appropriate package from the dependencies first (php5-cgi | php5-fpm) and then install php5 or just install php5-common afterwards [03:39] i dont find php5-fpm for 10.04 [03:39] there will always be defaults [03:39] whats the difference between php5-cgi and php5-fpm? [03:39] rneese: we already said it's not in 10.04 [03:40] i read one uses fast-cgi one uses regular cgi, which is most desirable? [03:42] is there a nginx-devel [03:42] this is a old ver of nginx [03:42] add the ppa if they have one [03:42] rneese: I think MTecknology is familiar with nginx [03:42] howdy [03:42] the ver in the pkgs is old [03:43] from what I have seen [03:43] http://wiki.nginx.org/Install [03:43] there's a PPA for the latest stable and dev [03:43] dev recently moved to stable so development is just a copy [03:44] looking for the ppa [03:44] not finding the ppa [03:44] i just gave you a link that gives you the ppa [03:46] ok [03:46] rneese: what you said about php.. php-apc depends on some apache pieces and I think a couple other parts of that meta package do too - but php5-fpm and php5-cgi and php5-cli don't [03:46] I haven't figured out why the apache deps yet [03:49] ok [03:50] well I just installed 8.52 nginx [03:50] thats the lates [03:50] cool === cs278| is now known as cs278 [03:51] it has non of the php stuff in it [03:51] man this layout compaired to bsd is ugly [03:51] nginx shouldn't come with php.. [03:52] they're two separate things [03:52] but the config file should have the php settings in it [03:52] and it does not [03:53] yes it does [03:53] there's fastcgi_params and an example of how to use it [03:54] http://wiki.fusionpbx.com/index.php?title=Ubuntu_Install#Nginx_Setup [03:54] none of the lines on that page are in the files [03:54] well.. for starters... [03:55] in the first code block - I already see examples of things you should never ever do [03:55] http://wiki.nginx.org/Pitfalls [03:55] well this is right out of a working nginx conf file [03:56] that it was pulled from [03:56] ooh! it suggests some REALLY ugly stuff [03:57] that stuff is horrible [03:57] rneese: read that wiki page I just sent you [03:58] I am and its nothing like we do on bsd . I am doing this on ubuntu virtual box for a friend [03:59] and this is just not what I would expect [03:59] bsd guys roll their own package same as anything else [03:59] most distros have these pitfalls in them [04:00] we tend to use ports not pkgs [04:01] and the config file layout is diff [04:02] here's the thing..... [04:02] if you don't like it [04:02] file a bug [04:02] bugs.launchpad.net/nginx [04:02] <|rt|> rneese: I think that bsd's have a bit more consistency compared to Linux distro's but different isn't always wrong [04:03] Right! The maintainers can't ignore your bug, or claim that it's a feature, if you don't report it! [04:04] :P [04:04] <|rt|> rneese: the way that Debian and it's derivatives setup their apache config is way different than redhat and it's derivatives....although in that case I think that the redhat way is wrong :) [04:05] <|rt|> j/k it's just different but I do prefer the way that debian and it's derivatives approach it [04:05] |rt|: speaking of which... I really need to chat with the debian nginx maintainer sometime about marging the changes in the ppa.. [04:05] merging* [04:08] ok [04:09] i have spawn-fcgi installed but not finding the startup script [04:11] Just use php-cgi and write your own inti script... or copy/paste from the nginx wiki [04:11] the cool new thing is php-fpm [04:11] thats in the nginx repos [04:11] !info php5-fpm [04:11] php5-fpm (source: php5): server-side, HTML-embedded scripting language (FPM-CGI binary). In component universe, is optional. Version 5.3.3-1ubuntu9 (maverick), package size 2875 kB, installed size 7624 kB [04:11] it's also in 10.10 [04:12] but this is only 10.04 [04:12] so we opted to use spawn-fcgi [04:12] 22:11 < MTecknology> thats in the nginx repos [04:12] spawn-fcgi only starts the processes, there's a lot of extra leg work [04:13] http://wiki.nginx.org/Configuration [04:16] ok [04:19] ok thanks [04:24] rneese: btw - there is #nginx if you need more nginx specific help [04:24] that works [04:24] got it up and the website is working [04:25] did you use a socket instead of a port for the php connection? [04:26] that tends to make things much cleaner and easier to work with - especially when you get into really funky and weird setups like i have [04:26] each website is its own user, each user is inside of a jail, each website uses its own php process, blah, blah [04:27] yes === Pilif12p is now known as Pilif12p|afk [05:21] MTecknology: its all working thanks [05:22] rneese: :) [05:23] its working flawlessly [05:23] and the full install on a vps is 1 gig [05:29] anyone have any idea why ssh wont come up on boot? [05:29] just upgraded from lucid to maverick [05:29] it's a headless box so have to blindly login w/ an attached keyboard and do a /etc/init.d/ssh start to login [05:29] how can i select which packages are installed using a preseed file [05:29] whats stranger is that no services are visible (nmap) before I start ssh [05:30] but after I start ssh they are visible as expected [05:32] New bug: #659655 in openssh (main) "openssh-server fails to come up on maverick (upgrade from lucid)" [Undecided,New] https://launchpad.net/bugs/659655 [05:32] that be me uvirtbot [05:35] is it possible to slect which packages are installed using a preseed file for ubuntu server? [05:39] anybody have pam-mysql working with vsftpd? I could use some pointers. [06:10] lets say i add a ppa repository and run my webserver from that. do i need to manually run apt-get update every time i want to check to security updates and such? [07:34] My server offers Ubuntu installation via PXE booting and approx apt package caching :D [07:34] how cool is that? [07:39] does anyone know if xen supports Ubuntu 10.04 (Lucid)? [07:48] RudyValencia: approx isn't cool at all [07:48] Oh? [07:48] It works for me. [07:49] One of them even allows clients to inject arbitrary packages into the cache [07:49] I was considering apt-mirror for my packages but balked at the expensive amount of downloading I'd need to have a mirror of just main [07:49] It's only me using it on my LAN [07:49] RudyValencia: main of one arch, one release, no sources is only 8GB or so, IIRC [07:49] Yeah but I'm on a 1.5Mbit/sec connection here [07:50] You only download it once [07:50] Right [07:50] But I just don't have a time to let it download. [07:50] It'll eat my connection up. [07:50] Er, throttle it. [07:51] So, approx also downloads the packages only once. [07:51] Good morning all. I am trying to setup NIC Bonding in Ubuntu 10.04. In every previous release of Ubuntu i have sucesfully managed to do this by following the instruction in the following guide http://www.howtoforge.com/network_bonding_ubuntu_6.10. Things seem to be slightly different in 10.04 and i am yet to sucesfully manage to bond eth0 and eth1. Well i kind of got it working, but it fails on each subsequent restart. Anyway long story short, i [07:52] I'm not distributing Ubuntu to a computer lab or to a whole organization with this server, just using it for PXE install caching [07:53] (When I finish using my local apt configuration, I just copy over the original sources.list with Ubuntu servers. [07:57] New bug: #659700 in clamav (main) "package clamav-base 0.96.3 dfsg-2ubuntu1.1 failed to install/upgrade: il sottoprocesso vecchio script di post-installation ha restituito lo stato di errore 1" [Undecided,New] https://launchpad.net/bugs/659700 === _TechAway_ is now known as _Techie_ [08:22] <_ruben> pfft, mirroring only main is for wussies, i just mirror main/universe/multiverse incl -updated/-backports/-security for gutsy - maverick ;) [08:35] how do i upgrade to 10.10? sudo do-release-upgrade says no new version [08:36] <_ruben> andriijas: check its config, it probably is looking for the next lts, which doesnt exist yet [08:36] _ruben: which file is this? [08:40] <_ruben> andriijas: the prompt setting in /etc/update-manager/release-upgrades [08:40] <_ruben> change lts to normal [08:41] thx === mgolisch_ is now known as mgolisch === doko_ is now known as doko [09:56] <[Ben]> Hi everybody. I am looking for some help on how to bypass the problem with the P410 raid controller (HP DL370 G6) with ubuntu 64bits, showing no logical disks (and even if I "disable" the raid, seeing no disk at all). Thanks. [09:57] DL370? [09:58] <[Ben]> Yep [09:58] <[Ben]> data:image/jpg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD/2wCEAAkGBhISEBQUEhAUFBUVFhwZFBcUGRkXGBgZHxUZFBQXGBcZHCYgGBkjJRgVHy8gJCcuLCwsFSAxNTAqNSYrLCkBCQoKDQwOGg8PFS4kHx81KSk0LC8tNjUuLDUpLDAsLikpKTMsKTUtLCw1NSwpLSs1KSktKSwsLCk1LCkpLiksLP/AABEIAE8AlgMBIgACEQEDEQH/xAAcAAAABgMAAAAAAAAAAAAAAAAAAwQFBwgBAgb/xABIEAABAwEDBgcLCQgDAQAAAAABAgMRAAQSIQUGEzFBVBQiI1GRk9EXMkJhcYGSobKz0gckM3N0grHBwhY1UmJyouHiQ0RTNP/EABcBAQEBAQAAAAAAAAAAAAAAAAABAgP/xAAdEQEAAgIDA [09:58] <[Ben]> AAAAAAAAAARExQQIh8FES/9oADAMBAAIRAxEAPwCaLflFthF91d1MxJBOOJ2DxHopqZz7yesSm2NETBg7do9YrOeLt2zzAPGOChI+jXsqBHrfpJXdSj5ziECBAaAOA54mqJ7OfNgH/ca6a1Of2Tt9Z9KoFU2paokoETiAZxjnpA/Yi3PKKXP8QGHk6atIsP3Qcm78z6VYPyh5N35j0qrVaJI74jxik9sd4seb+041FWd7omTd+Z9Ksd0bJm/sel/iqsWJErAJIGMxr706qeDk8pQhyOKTgZJ5pB59YqCx5+UXJm/selWvdIyZvzPSeyoPzRtCE2K1h1aQFNJCAtQSVK0qr4bva1xcmMYidlIX3WMbl6ZxMwmZ2SrybNdWkvtP3dJyZvzXSeyh3Scmb810nsquDqjsV/cO2jMouN8HaDV7TydKSrilN0 [09:58] <[Ben]> lRqItYzuj5M35npPZWvdKyXv7PSeyq1WtRvi7JMKu3ee6YiPNXW53P2cqahSFcg0DolJOMKkcTwwYmceerTKaO6Vkvf2ek9lDul5L39npPZVcXljZPT/AJrFiKNINLeKMZg46sNRqNRFysd3TMl7+z0nsrB+U3Je/s9J7Kra4RpHY+jJOjBOMSYwkmhky1NpQrSJKpOERzGdfmok9La2W0pcQlxCgpC0hSSNRSRKSPKDQpuzSM5Pskbu17pNCgQ5+qiyj+v9C6r619Gr7Qfd1Pnyjq+aD6wewuoDJhtX1593VgObiuP9z86bcqvKBTdSFEmADJ5ySIP8tLXjx/u/nSVwS8yP58fRXWpwRV9mxbxmS2krgpuwdZW2J14q4xx8flovKJJRxkBKgTIAIOISRM+U0ptwi0qA1aXDyX2aJyyOOvzewisdtz+ [09:58] <[Ben]> yifP+Bp3fTDSTeGMECTPfao1CKaLL34x5/wNdPl61DRxoEoLi0uAjYm4kBOAHNP3vPVYK80SgWO16QpALQuXykEq0qr4aveHFybuMROykL2gxu3pkzBVdmdmP4baXZpOIFhtWkUkBSEBAWpIKlB5RWGpxvQUTdxiJpA4WsboXrxIJAJnYL35VdJsgdmcJ6VdtGZSDfB2tEF6eeVkm7dunVJjXFByScAvpV8VZykBoWw0HA7/wApKjEXcYlUa6zLpxxOPfCC1glYuyTCojnumI8equwz0VZi6LtxXJtjkSmJunDk/CnXj5a5G3SVgCSYN2Nc3IEePVXYZ4WmzqWghSXORaHJKBxAPF4hELGE4zWoc505V0J2JPr7azYEN6QaZKijbBM+1Wzym9iVjzn4q1smjDgLiFlG0SZ9oVmW+OYJ1J47sCEEnRyc [09:58] <[Ben]> J5qRDvfvGl5HHdIBDZJ0YJxiTAiSeam9Pe+eiTlbLM393WP7M17pNCsZl/u2x/ZmvdJoUQ2fKan5jenvXEHyzKP1eqoFyk3cTAVN5y9qjwSI9VTp8qg+ZJ+uTPj4q9fqqBbfrV9YPYrUBytJ4yf6PzFInl8o2f5v0qpblBkpUidqJHTQyTkwPu4uNICE3iXVXAeMEwk/xYnXsBqoZ7SrlST/H+pmi8pgqUuNZj2Uc9dLac1fnSEpfsxSptayrSHRghaAEqVEBwxIG3Cict5D0DjKg4ysOLjkl6SIU2BfwwmcOeDzVC3JWezKCgY59qeY+OnfKS0FKLipJCSubuBwkDbs1+MV1VqbIBnRCfF+EiKZ84moZSCUk304p16/8ANRR2aDyU2O16RQCVtJCApQBWoOqK0tz4UFExjEUgdW1jdSvWZIJAJnYLw [09:58] Yep; just looked it up. [09:58] <[Ben]> wJsdsvqCUraCUAqAvqDqitLYPhQUTHipI+7Z8YbcnaQSAT4hfHk1bKuk2bXDOoK9I/FWcolJZbDSVh0fSEqwIumYJWRrit3FtnUhzpPx0XlK4plAaQsOzxyVYFMHUSsjmrMunHE498JLXJdTEkwQI1zdIEequvzztdnW/KVJc4jYlpQIkNnAFJAChtxrkLYklwBMklJCY1zdUAAeeYrr88rbZ1upKVId5JsS2oESEniyCAFA68a1DnOnMOrGwL6f9q2ycpAcBdQso2gKx9sVq84nYFel/tWcnuNhwF1ClI2gKx8Xhisy3xzAkA3ndYQSdGCoE3ZJAiTjqpr0uzz+uKc547sTcUToxIJAkkCAT4q1ycniKw8L8qJOVpMz0Rk+yDmszXuk0KMzX/wDhsv2dr3aaxRDH8qY+YTzOoP4g+qagMrvlW3low5 [09:58] <[Ben]> siXGrqkpUJGCgCOg1zgzXY3dnX/5o7KtiDH3yVCcQBE4YUnFovAiFJE+EICtcEDbGOMbans5rWfdmerR8Na/spZ92Z6tHZSxAD1oKUEBSlJJ7xMkE7Dd1Hy1q07Cm+NdAWlShjhgdYGEifxqfzmlZ91Z6tHw0BmjZt1Z6tHw0ERtZZa8N9KvEoE484hNN2W7c0tACFIJvpPFBBicdYGFTiMzrLuzPVt/DWrmZ9m3Znq2/hoIazPt7bdltgcdSkOMpShKlRpCHVlSUjngp7aQv5QssmESdqgoAHzX/Nq2VOAzRs26sdWj4az+xtm3ZjqkdlLSu7V/dtbc4A+kPjoZRtDamUJaCkug8dRWIKYOAN8jm2VPxzOs26sdWjsrKszrNE8FY6tHZUbjlSvFrVecEGSUkCCNd1QGPlIrtc8LdZ1PA3kujRNiWlA [09:58] <[Ben]> ACgdeNSj+yFm3Vjq0fDWwzQs26sdUjsq2xSBH3mNiF9KvjrWyOM3wVoWUbeMqfbqfDmfZt1Y6pHZQTmfZt2Y6pHZUaiam1emjyjpmEFR0YKpN2TAiSdUVrY3gEHEYmrD/slZ91Y6pHw0Y3m20BAZbEaoQkflRJ7PmaK5yfZDz2dr3SaFOFjbutoAEQkD1UKAxaZovg4o6hQElgUODijqFATwehwcUdQoCiyKwbOKOoUBHBxQ4MKPoUBHBhQ4MKPoUBHBxQ4OKPoUBHBhQ4MKPoUBHBhQ4MKPoUGEjChWaFB/9k= [09:59] <[Ben]> -_- [09:59] <[Ben]> sorry [09:59] <[Ben]> http://www.quadsel.in/productimages/1260509460.jpg [09:59] <[Ben]> I have two bays with 8 2,5" drives [09:59] :( [09:59] <[Ben]> I've looking for a solution, the only valuable thing learned is that it needs cciss to work [09:59] <[Ben]> it works on 32bits, but failed on 64bits ... [10:00] <_ruben> that'd be rather odd [10:00] <[Ben]> I was looking everywhere for 64 bits drivers, but it seems only Red Hat and Suse are able to handle this in 64bits (are they ?) [10:01] <[Ben]> I have two version of Ubuntu 10.04, 32 and 64bits, the 32 bits shows me the disks, and the 64bits just failed ... [10:01] <[Ben]> I tried a debian 5.0.6, too (in 64bits) with the same results [10:03] [Ben]: You can't "modprobe cciss" ? [10:05] <[Ben]> It's when the install is asking me to set the partitions, I can't do a modprobe at this time, can I ? [10:06] Perhaps try moving to another tty? [10:06] <[Ben]> I'll try this right now. I saw this thing, too : http://cciss.sourceforge.net/ [10:07] Last sentence of the first paragraph. [10:09] <[Ben]> Yes, I can read :p. But still, it doesn't work as advertised in 64bits mode. [10:09] <[Ben]> I'll try a modprobe (no ssh yet, so I have to go downstairs in the server room, brb) [10:15] New bug: #659784 in squid (main) "package squid 2.7.STABLE9-2ubuntu5 failed to install/upgrade: ErrorMessage: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/659784 [10:20] <[Ben]> dmesg : cciss - compat_alloc_user_space [10:20] <[Ben]> module unknown (he said) [10:30] [Ben]: Weird. [10:33] <[Ben]> as you said ... [10:34] <[Ben]> What surprise me the most is that NOBODY is talking about this problem. Like I am the first to install ubuntu on a Proliant with this fairly known controller [10:34] So, the cciss module isn't loaded? Or not available at all? [10:41] <[Ben]> how can I know ? It just said what I showed you [10:41] <[Ben]> shown (?) [10:41] dl370 is certified to work on ubuntu === ivoks-afk is now known as ivoks [10:41] <[Ben]> Yes I know ... [10:41] ScottK: no :/ [10:41] <[Ben]> but maybe Ubuntu 32bits (it works on Ubuntu 32bits) [10:41] nope, 64bit [10:42] http://webapps.ubuntu.com/certification/hardware/200804-267/ [10:42] which version are you installing? [10:42] <[Ben]> It's the tower version, the "ML" [10:42] <[Ben]> Ubuntu 10.04 LTS server [10:42] <[Ben]> 64bits ("amd") [10:43] cciss module is available in ubuntu 10.04 64bit [10:44] [Ben]: do you have access to this server? like, keyboard and VGA? [10:44] [Ben]: no private messages [10:45] [Ben]: Is this a DL370 or ML370 ? [10:48] [Ben]: you'll gave to site in front of that server or setup an installer with network access [10:48] s/gave/have/ [10:49] <[Ben]> It's a DL 370 G6 [10:51] can you start the installer and sit in front of the server? [10:52] <[Ben]> Yep [10:52] do it [10:52] <[Ben]> I'll move with a laptop downstairs [10:52] <[Ben]> brb [11:02] <[Ben]_> back [11:02] <[Ben]_> @ivoks : I am in front of the DL 370 G6 [11:02] [Ben]_: start the installer [11:03] <[Ben]_> done [11:03] did you get to the partitioning? [11:03] <[Ben]_> Yes [11:03] alt+f2 [11:04] <[Ben]_> ok [11:04] <[Ben]_> shell active [11:04] is cciss module loaded? [11:04] lsmod | grep cciss [11:04] <[Ben]_> nothing appears [11:05] modinfo cciss [11:05] is cciss there? [11:05] <[Ben]_> (test : /bin/sh: modinfo : not found) [11:05] you should get a lot of output [11:06] hm [11:06] modprobe cciss [11:06] <[Ben]_> FATAL : error inserting cciss (/lib/modules/2.6.32-24-generic/kernel/drivers/block/cciss.ko): [11:06] <[Ben]_> Unknown symbol in module, or unknown parameter (see dmesg) [11:07] heh? [11:07] <[Ben]_> dmesg : cciss : unknown symbol compat_alloc_user_space [11:08] what image did you download? [11:08] 10.04 or 10.04.1? [11:08] <[Ben]_> I think 10.04 lts, I'll check [11:08] <[Ben]_> how to tell ? (I know for debian) [11:08] by the image name? :) [11:09] <[Ben]_> lol [11:09] you have the image on your laptop? [11:09] <[Ben]_> nope [11:09] <[Ben]_> but can scan the USB drive on the server [11:09] does anyone know which kernel is in 10.04 and which in 10.04.1? [11:10] <[Ben]_> 10.04.1 [11:10] hm... ok [11:10] would it be a pain to try 10.04? [11:10] <[Ben]_> I can try the 10.10 out today (?) :D [11:11] <[Ben]_> No, I can just download it and make different USB drives, I am stuck with a non-working server, so I'll give a try [11:11] do it [11:12] <[Ben]_> I tried Debian 5.0.6 64bits and it didn't work (for your information) [11:12] <[Ben]_> SO you want me to try the 10.04 and/or the 10.10 ? [11:12] 10.04 [11:12] cciss loads on my 64bit lucid [11:12] how can I get my ldap back :( [11:13] bdb(dc=alvhaus,dc=ch): Program version 4.8 doesn't match environment version 4.7 [11:13] Aison: ? [11:13] I no longer can start ldap because of this error [11:13] and so I get segfaults in nss [11:13] etc... [11:13] Aison: what ubuntu version is that? [11:13] the new one [11:13] 10.10 [11:13] environment version? [11:13] what's that... [11:14] no idea [11:14] my slapd starts just fine [11:14] is this a new install or upgrade? [11:14] upgrade [11:15] slapd is compiled with libdb-4.8 [11:15] <[Ben]_> Why upgrade when it works :p ? [11:16] now, where would it pick up 4.7 as an environment version [11:16] I guess the database files are from 4.7 [11:17] probably [11:17] so the environment version of the database files have to be upgraded also [11:17] there should be a tool like db_recover or so [11:17] db4.8-util [11:17] but [11:17] don't do that yet [11:18] do you have /etc/ldap/slapd.conf maybe? [11:18] Good morning all. I am unsucesfully tyring to setup NIC Bonding in Ubuntu 10.04. I managed this with ease in prior releases. In 10.04 the bond appears to work, but then fails to come up on subsequent reboots. Is anyone able to advise on a possible "how to" for Ubuntu 10.04? Many Thanks [11:19] ivoks, no, just /etc/ldap/slapd.d/ [11:19] there's my config [11:19] tgwoollard: that's probably cause of upstart [11:20] Aison: do you have some custom files there? [11:21] ivoks thanks. I'll have a dig and see if i can get it working. [11:21] ivoks, well, I added new schemas, added stuff for ldap replication, etc... [11:22] Aison: is 4.7 mentioned anywhere? :) [11:22] grepping... [11:23] no [11:24] hm [11:24] db4.8_upgrade: Program version 4.8 doesn't match environment version 4.7 [11:26] ivoks, Aison: bug 658227 details a workaround for this issue; the newer version of slapd in Maverick uses 4.8 [11:26] Launchpad bug 658227 in openldap "upgrade process does not upgrade underlying BDB format from 4.7 to 4.8 (so slapd aborts with "Program version 4.8 doesn't match environment version 4.7" error message)" [High,Triaged] https://launchpad.net/bugs/658227 [11:26] however the version in Lucid uses 4.7 (and the upgrade does not accomodate this) [11:26] oh, thx [11:27] JamesPage, the funny thing is, I get segfaults on other places, because libnssldap is not working ;) that's not nice also [11:27] Aison: there's a workaround how to fix it [11:28] yes, upgrading to 4.8 ;) [11:28] oh, i see... JamesPage pasted the link :) [11:28] this is how you should do it [11:29] stop slapd [11:29] install db4.7-util and db4.8-util [11:29] go to /var/lib [11:29] copy ldap to ldap.backup [11:29] run db4.7_checkpoint -1 [11:29] then db4.7_recover [11:30] then db4.8_upgrade *.bdb log.* [11:30] er.... no log.* :) [11:30] just db4.8_upgrade *.bdb [11:31] we should export database on update [11:32] clean it, and import it on install [11:36] ivoks, my ldap is working again [11:36] :) [11:36] had to upgrade var/lib/ldap/accesslog also [11:37] that's replication log, i guess [11:37] yes [11:38] problem solved [11:38] \o/ [11:38] we should fix this in natty and maybe even backport to lucid and maverick [11:39] Its been raised as a SRU for Maverick [11:41] fix is available? [11:41] Not yet [11:41] then how can we have sru? :) [11:42] /SRU/potential SRU/ [11:42] ok [11:48] hm... everything needed is already in the package [12:19] <[Ben]_> @ivoks : running 10.10 Ubuntu with HP DL370 G6 seems to be ok ! [12:19] <[Ben]_> I can see the drives [12:20] <[Ben]_> Two possibilities : [12:20] <[Ben]_> 1) the usb key installation is not suitable for this kind of product (hdd layers incompatibility ?) [12:20] <[Ben]_> 2) The Usb Key installation was corrupted (I MD5( [12:21] <[Ben]_> I MD5ed the iso of the 10.04 LTS before using the tool to put it on a usb Key [12:21] <[Ben]_> Anyway, God bless Ubuntu 10.10 :p [12:21] <[Ben]_> Thank you for your help and see you soon [12:32] for several weeks, I changed the motherboard / cpu in one of my servers [12:34] its a 64bit cpu now, but the system of course is still running on 32bit [12:35] is there a way to change to 64bit? eg. install a new harddrive, install new ubuntu and the copy somehow the whole configuration and install the same packages? [12:41] brb === _Techie_ is now known as _TechAway_ [12:44] <[Ben]_> it's really complicated due to the facts that all the drivers are written in 64bits instead of 32bits [12:44] <[Ben]_> so your software and /home, etc can be "imported", sure [12:44] <[Ben]_> but for packages and drivers, it's a different thing, in my opinion (depending on the software) [12:50] Hi guys, I got a question. I installed ubuntu server edition some days ago and configured a softwareraid 1 on 2x 1TB hdd drives - I had to set the partitions on both HDDs when installing software raid 1 - now one of the HDD crashes and I want to install a new HDD (also a hdd from another manufacturer) as second drive again, but how do have to install it? Will the sync-prozess start itself or... [12:50] ...will i have to start it manually? do I have to set the partions before the installation of the hdd by hand? [12:51] Or will raid set the new partitions itsself? [13:00] xxfog, you need to create a partition on the drive the same size as your other one, of type linux raid autodetect, then use mdadm to add it to your array [13:01] it will then rebuild itself [13:03] xxfog, tutorial here, start at number 4: http://www.howtoforge.com/replacing_hard_disks_in_a_raid1_array [13:03] skipping the bits about md1, if you only have one raid [13:05] [Ben]_: it should work with lucid [13:12] Hi all, can anyone tell me how I'd locate my php install location on a server? I've tried locate php, and trawled through all that, and also whereis php, but nothing appears [13:12] daffy_dowden: you are new to linux? [13:12] ish [13:13] I thought it'd be in /etc [13:13] only configuration files go in /etc [13:13] you do know how programs are installed on linux? there's no single directory [13:13] I'm aware of that [13:13] So how can I find the executable? [13:13] so, what exactly are you looking for? executable? [13:13] which php [13:13] which php [13:14] which doesn't return anything, as mentioned [13:14] then you didn't install php CLI executable [13:14] install php-cli or php5-cli package [13:14] ok, so how are the php sites on the server being run? [13:14] that's what I don't get [13:15] through mod_php [13:15] apache's mod_php [13:15] They run though an apache module [13:15] right, ok [13:15] maybe he should tell us why he's "looking for php" [13:15] and that can't be run directly? [13:15] Are you trying to execute a PHP file via the terminal? [13:15] yes [13:15] Then you need php-cli [13:15] then you need php5-cli [13:15] Or that [13:15] cheers [13:15] then you need php-cli :) [13:15] then you need php-cli [13:16] you might want to consider php5-clie [13:16] cli [13:16] :) [13:16] next! [13:16] lol [13:16] I don't even see php-cli available in my aptitude, I think the only package left is php5-cli [13:17] If i install php5-cli will apache still use mod_php then? [13:17] Yes [13:17] good [13:17] It's just another method of using php [13:18] out of curiosity, could you force it to use another method? [13:18] There's php-cgi I believe [13:19] Would that run as a daemon? [13:19] Like if I were hosting say a python site with fastcgi and nginx? [13:19] I don't know, you'd have to ask somebody that has experience with php-cgi [13:20] fair enough. Thanks for the help [13:21] there's fastcgi [13:22] and fcgid [13:22] and you can use those with php5 like you would with say python or ruby [13:23] ? [13:24] yes [13:25] it's better for performance and server [13:25] but it lacks some features [13:25] ahh cool [13:25] you can't use .htaccess for example [13:25] right [13:25] (with php values) [13:25] that makes sense [13:26] Do you know why it's better for performance vs mod_php? [13:26] cause apache doesn't start with php module every time someone vistis your website [13:27] I see, thanks [13:28] Is there a way to disable SSH password authentication for a specific user? I want to switch my user account over to public key authentication, but I have 2 other users that I don't want to affect by disabling password auth [13:28] and then you don't have to use apache prefork, which is slower than apache worker [13:28] jumbers: yes [13:29] pmatulis: How would I go about doing so? [13:29] jumbers: man sshd_config and investigate the Match keyword [13:29] what's the difference between prefork and worker? [13:29] daffy_dowden: prefork is used only cause of php [13:30] daffy_dowden: how many cpus do you have? [13:30] on this server - 1cpu 4 cores [13:30] that's 4 cpus [13:30] pmatulis: Thanks [13:30] worker scales better than prefork [13:31] prefork is not threaded (cause php doesn't support it) [13:31] qman__: there are 3 partitions but i don`t know there size for sure. so i should better control the working harddrive and add same partitions on the new drive right? [13:31] while worker is threaded and therefor scales much better [13:32] so say a site I had on this machine were to become incredibly popular, it'd be better to use fastcgi to get the benefits of the multiple cores? [13:32] xxfog, the command given on that tutorial will copy the partition scheme from the working drive to the new one [13:32] daffy_dowden: yes [13:32] daffy_dowden: and varnish [13:32] is that similar to memcached? [13:33] no, varnish is reverse proxy [13:33] ah ok [13:33] with varnish and static content, your server will be able to provide, almost, unlimited connections [13:33] right [13:34] for more info, my bank account is... :) [13:34] soren: hi! would you mind looking at and commenting on my patch for bug #659532 ? [13:34] Launchpad bug 659532 in vm-builder "vmbuilder XML is incomplete with libvirt 0.8.3 (No bootable devices)" [High,In progress] https://launchpad.net/bugs/659532 [13:34] so, if I understand this right, a request would come through varnish first, then if it's static get served via apache worker, if dynamic go via a fastcgi thread? [13:35] ivoks: lol [13:35] * soren buries his head in shame [13:35] soren: what did you do now? [13:36] zul: 12:34 < jdstrand> soren: hi! would you mind looking at and commenting on my patch for bug #659532 ? [13:36] Launchpad bug 659532 in vm-builder "vmbuilder XML is incomplete with libvirt 0.8.3 (No bootable devices)" [High,In progress] https://launchpad.net/bugs/659532 [13:36] soren: ah yeah :) [13:36] :] [13:37] daffy_dowden: if it's static, it will be served by varnish, which cached it [13:38] ivoks: cool, otherwise it goes to the app server, in this case fastcgi, right? [13:39] right [13:40] thanks man, much appreciated! [13:40] jiboumans: who from your team can join us for this UDS session? https://blueprints.edge.launchpad.net/ubuntu/+spec/appselection-server-n-kolab [13:40] qman__: so i can insert a blank new hdd without setting up the partitions of type raid? [13:41] jdstrand: Looks lovely. [13:42] soren: cool, thanks. so, one thing I wanted to be sure about was --raw. I had a time figuring out how to properly use it. I'd like to make sure I use libvirtxml_fsimage.tmpl. what is an invocation that would? [13:44] jdstrand: Create a 5G disk image. Pass it to vmbuilder with the --raw arg. [13:44] jdstrand: It basically just tells vmbuilder to use that file (or device) instead of creating a new one. [13:45] soren: right. I thought I did that in some iteration of the patch before I modified libvirtxml_fsimage.tmpl and saw qcow2 pop in there (ie, it used libvirtxml.tmpl [13:46] soren: but I won't be able to reproduce that at this point, so I could have jacked something up [13:47] soren: alright, cool. thanks for your help :) I'll take care of maverick and lucid (when it is required). can you handle natty if you are doing a separate upload rather than a pocket copy (I can do a pocket copy)? [13:48] <[Ben]_> @all : ubuntu 10.10 server installing process on DL 370 G6 (hp), seems to be blocked - 83% on "configuration of Linux-image-2.6.35-22-server" [13:48] jdstrand: Better do a pocket copy for now (althought I doubt anyone is running natty anyway). [13:48] <[Ben]_> other tty reachable but no pertinent info [13:48] soren: sure, np [13:48] [Ben]_: just curious, but why would you be running a non-LTS release on a server? [13:49] <[Ben]_> @RoyK because LTS didn't recognize my P410i raid controller ... [13:51] New bug: #659880 in php5 (main) "libapache2-mod-php5 will consume all available memory when installed and activated" [Undecided,New] https://launchpad.net/bugs/659880 [13:51] RoyK: Because you sometimes need newer software than available ? [13:51] [Ben]_: k [13:52] Jeeves_: well, obviously, but mostly the 'old' software works just as well, and I wanted to ask in case he was making a mistake [13:52] <[Ben]_> Anyway I am stucked with a non-working raid controler or a non-installing system :D [13:54] [Ben]_: did you already say what make and model of server this is? [13:54] <[Ben]_> HP DL 370 G6 (P410i and P410 controler) [13:54] [Ben]_: k [13:55] <[Ben]_> cciss driver not loading with 10.04.1, will try to download the ISO again (but was "sure" it was ok because of the MD5 hash) [14:02] I did not find an apparmor profile for RoundCube Webmail for my Ubuntu 10.04 server, therefore I plan to write my own profile. Where can I list the profile once it is completed so other people don't have to redo the same work? [14:03] Hi I'm trying to make Pacemaker start my KVM/QEMU machine but it gives me an error. I suppose the reason is that the init script is not OCF or LSB compatible, but up-start. Is there any VirtualDomain resource for up-start jobs? Or am I completely wrong? [14:05] I'm using Ubuntu 10.04 by the way. [14:05] david506: submit a bug with feature request [14:05] toshko: i didn't understand your question [14:05] toshko: do you have problems with starting upstart or virtualdomain resource? [14:05] ivoks : thanks [14:06] starting pacemaker :/ [14:08] as I understood the "virtualdomain" resource of pacemaker tries to start/monitor/stop the libvirt-bin "init" script, which is in fact converted to upstart job in ubuntu 10.04 [14:08] virtualdomain? hm... i don't think so [14:08] let me check [14:09] as i see there is no support for upstart jobs (only for LSB/OCF) scripts in the pacemaker [14:10] VirtualDomain RA doesn't check if libvirt is running [14:10] you should make sure it's running [14:11] you can start it manually, with anything RA or upstart plugin for cluster-agents [14:11] cluster-glue [14:11] fsck... [14:11] i belive i've created packages for lucid, that support upstart [14:11] let me check [14:11] I notice that Ubuntu 10.04 does not come with IPTables enabled with rules by default, I was surprised by this decision as IPTables adds an important layer of security. Is there a Wiki article discussing not activating IPTables by default in detail ? THanks [14:12] david506: Erm, some people would rather choose which firewall software they'd like to use instead? [14:12] david506: ie, Shorewall, ufw, ... [14:12] The same could be said about apparmor vs selinux, no ? [14:12] (Not saying that they shouldn't use iptables). [14:12] it's not the same [14:13] iptables is enabled [14:13] ok [14:13] you can use [14:13] it [14:13] but no one can give you sane default that will protect you [14:13] [Ben]_: the DL380 G7 is fully certified by Canonical. this system uses a P410i [14:13] Ok [14:13] it's up to sysadmin to create its own rules [14:13] and by default, all ports are closed [14:14] so, what would you firewall by default? :) [14:14] [Ben]_: please open a bug if you can't get it to work [14:14] david506: ivoks is right. however, ufw is installed by default and if you want to add firewall protection, you need only do 'sudo ufw enable'. see man ufw for details [14:14] [Ben]_: Bug would be best; a DL380 G6 I admin has a P410i controller and it Just Works. [14:15] toshko: is libvirt running? [14:15] david506: it will make some decisions for you (that you can customize later if desired) [14:15] toshko: by default, it should [14:15] New question, I have 6 servers. 1 public net Squid reverse proxy, 2 private net web servers ( feed squid ), 2 private net database servers ( feed web server ) and 1 webmail/SMTP/POP3/IMAP server. Should the mail server be public or private? If it's private, should I use the squid machine with iptables to forward appropriate ports using NAT MASQUERADING ? [14:15] ufw might be nice on desktops.. but running it on a server causes (at least for me ) a real mess rule-wise. [14:15] well it's running when i start it [14:15] david506: https://wiki.ubuntu.com/SecurityTeam/FAQ#UFW [14:16] ewook: how so? [14:16] thanks jdstrand [14:16] but isn't the idea the resource to start it on the second machine when the first falls down [14:16] jdstrand: the rules thats generated via ufw isn't how I would do them myself :) [14:16] toshko: it is, but VirtualDomain RA starts and stops VMs, not the libvirt [14:17] the text under UFW answers my original question about not enable iptables by default. It makes sense. [14:17] ewook: can you elaborate? [14:17] jdstrand: also, the order and the really... almost stupid chain-layout isn't something I like either. [14:17] jdstrand: :) [14:17] ouch [14:17] jdstrand: relax... [14:17] jdstrand: don't swear [14:17] ewook: what order? you can change the order with 'insert' [14:17] ivoks: thanks, then i missunderstood [14:18] I was told that I needed a firewall anyway as there are many IP based protocols that don't rely on open ports [14:18] toshko: upstart plugin was developed for pacemaker [14:18] ivoks: so what is the script VirtualDomain monitors? [14:18] jdstrand: sure I can. but it's still a mess how ever you choose to see it. [14:18] toshko: i just can't find it :) [14:18] toshko: it monitors VM [14:19] ewook: you are of course free to not use ufw. if you aren't going to provide useful comments, it cannot be improved [14:19] I'll be back in twenty minutes [14:20] jdstrand: I will provide useful comments when I am able and have time for it. Just a penny regarding using ufw and other "smart" wrappers for iptables. [14:20] ivoks: well if it does not monitor the "libvirt-bin" script, then i shouldn't need upstart job resource [14:21] ivoks: how does it monitor the VMs? [14:21] toshko: that's right [14:21] toshko: look at the RA, it's a shell script [14:22] ivoks: ok thanks very much :-). Will do! [14:29] <[Ben]_> @all : unable to boot the 10.04.1-server-amd64.iso writen on CD [14:30] <[Ben]_> has anyone the same problem ? I burned two different CDs with two different burners. 10.10 is fine and 10.04.1 on USB Key is ok (but faulty for other things) [14:33] unable to boot? [14:36] <_ruben> pxe boot/install ftw ;) [14:53] ivoks: I've looked at the RA and it seems that it uses virsh to monitor and stop/start the VM. Thanks :-) [14:53] New bug: #656738 in landscape-client (main) "apache2 restart fails, but exit code is zero" [Undecided,Invalid] https://launchpad.net/bugs/656738 === xfaf is now known as zul [15:00] hi [15:00] I installed ubuntu-server in my laptop but I dont know to work on it.. how to setup a home server for six systems? [15:01] is there any way to work GUI in ubuntu-server? [15:03] techbreak: Well, you could install one, but it's not really recommended with it being a server and all. [15:04] Nyxation, where can I get good how to's or tutorial for ubuntu-server? for setting up home server? [15:04] * patdk-wk wonders exactly what is the definition of a homeserver [15:06] techbreak: That really depends on what you're going to be using it for. [15:06] <_ruben> patdk-wk: a server that resides in a home? :) [15:06] heh :) [15:06] Nyxation, I wanna setup a server among six systems at my home.. any suggestions? [15:07] my *home server* does dns, dhcp, tftp, ftp, smb, nfs, mythtv, bgp, ospf, http, https, ssh, ipsec, iscsi, auth, ntp, (hmm, sure I'm missing a few) [15:08] techbreak, to do what? act as a firewall? fileserver? webserver? .... [15:08] normally your have an howto for each one [15:08] patdk-wk, webserver [15:09] patdk-wk, and if its possible together with firewall too [15:09] well, webserver almost doesn't even warrent a howto these days [15:09] firewall though, can be more complex, depending on how you want to do it, with dhcp/dns and whatnot [15:17] Daviey: I am proposing we close bug 610987 INVALID [15:17] Launchpad bug 610987 in euca2ools "euca-describe-availability-zones output is a hint, not totally accurate - and should be documented" [Low,Confirmed] https://launchpad.net/bugs/610987 === todd_ is now known as Todd [15:26] hggdh: o/ === rgreening_ is now known as rgreening [15:28] Was about to file a other-server-n-sq spec catchall for UDS, to cover "QA workflow redux", "Foundations bug triage overlap", "SRU tracking", potentially more apport hooks etc. [15:28] hggdh: want to file it or should I ? [15:28] hggdh: looks like we'll have a separate one for testing autmation (the hudson stuff) and another for UEC testing [15:29] I meant other-server-n-qa [15:30] ttx: give me a minute, have to digest what you said [15:31] ttx: what about renaming the UEC testing to Cloud testing? [15:32] hggdh: cloud is a bit vague. UEC is more precise [15:32] hggdh: but whoever files it gets to decide how he names it [15:33] hggdh: want me to file the Server QA catchall session ? [15:34] (you'll probably lead it) :) [15:34] ttx: please do, and we can then discuss all involved [15:34] sure [15:34] Hey folks, I have a quick question. Running a file server and I'm trying to change permissions. So far I did sudo chgrp %groupname% %foldername%, sudo chown %username% %foldername%, and finally sudo chmod 770 %foldername% the problem is that the permissions don't carry over to the subfolders. how can I change the subfolder access? couldn't find much on it but I may not be searching for the right thing online. [15:34] thanks! [15:34] * hggdh is unsure on what has been going on [15:35] jauntypaperclip: sudo chown -r and chmod -r [15:35] jauntypaperclip: The -R option will do that (see man chgrp) [15:35] OH! recursive duh thank you! [15:35] oops -R, not -r... thanks alvin for the correction [15:35] duh on my end. thanks folks [15:36] I love case sensitivity :-) [15:36] wtf is this console-kit-daemon? [15:36] heh [15:41] did something change in nfs between lucid and maverick? [15:41] my exports had to change [15:41] have to have an export for each mount point on server instead of just using the crossmnt option [15:48] hi...i've installed dovecot+postfix on my ubuntu 10.04 server...i need to create a mailbox for webmaster@mydomain.com, but the documents are baffling, probably because i'm missing one simple bit of understanding...do i need to create a ubuntu user called webmaster then do something to it? === sailerboy_ is now known as sailerboy [15:51] spotter: can you pastebin old/new fstab? [15:51] !pastebin [15:51] For posting multi-line texts into the channel, please use http://paste.ubuntu.com | To post !screenshots use http://tinyurl.com/imagebin | !pastebinit to paste directly from command line | Make sure you give us the URL for your paste - see also the channel topic. === mgolisch is now known as mgolisch_ === mgolisch_ is now known as mgolisch [15:52] New bug: #659958 in squid (main) "Squid requires restart after Network Manager made connection to network" [Undecided,New] https://launchpad.net/bugs/659958 [15:52] export [15:53] not fstab [15:53] at least all I did was change something on server [15:53] can anyone give me a hint so i can narrow down the docs i'm reading in trying to understand this...would webmaster be a virtual user for dovecot? [15:55] wtf. I can't find _any_ references to console-kit-daemon under /etc, but still the bastard starts on every bloody bootup [15:58] RoyK: maybe /etc/ConsoleKit [16:03] hey, what can I use to connect to a vpn on an ubuntu server? [16:05] DormantOden: openvpn is pretty good [16:05] (should of said pptp vpn :P) [16:05] found pptp-linux, that the only one? [16:06] New bug: #659975 in mailman (main) "mailman breaks CC field incorrectly" [Undecided,New] https://launchpad.net/bugs/659975 [16:06] ignore my question...i created a ubuntu user...seemed the mailbox was already configured...the simplicity is totally awesome :) just hope the security is ok [16:14] DormantOden: vpnc if it's cisco [16:15] hello all - could someone help me with an apache/php problem? [16:20] does anyone actually chat on this server? lol [16:20] no [16:21] /list will tell you how few of us are here [16:22] thanks for the tip ivoks [16:22] be careful with that command [16:23] dryliketoast: There are a bunch of Apache and PHP specific channels. If you ask your actual question here you're more liklely to get a response. [16:23] (Also, you should be using alis , not /list. /msg alis help list) [16:25] #ubuntu is the only channel with 1000+ users :) [16:25] No, there are others. [16:25] But Ubuntu is the largest. [16:25] /m alis LIST * -min 1000 said so [16:27] ivoks: oh, nevermind then [16:27] i managed to get to the bottom of my problem on another channel - thanks all for helping an IRC n00b :) [16:28] well, you didn't ask the question here at all [16:30] SpamapS: around ? [16:30] ivoks: yes #ubuntu is really useful as well [16:37] ttx: o/ [16:39] SpamapS: mumble ? [16:45] ttx: on now, sorry, forgot i had tea brewing ;) [16:49] anyone good with iptables? [16:50] I want to pass traffic from a certain port range to a vpn =) [16:52] hi...apologies for the n00b question...my server has a silly hostname...what are the consequences of me changing this? i believe i need to change something in /etc/hosts too although that doesn't contain any reference to the name in /etc/hostname [16:53] is server 10.10 an lts release? [16:54] no === jj-afk is now known as jjohansen [16:55] Hey awesome peeps, I have one more question. I'm trying to set up a group for my file server. I currently have access however another windows machine that I'm using does not. I tried doing sudo chown ubuntuserveruser /srv/samba/share/group folder, sudo chgrp groupname /srv/samba/share/group folder, sudo chmod 770 /srv/samba/share/group folder I ensured that the other user is added to the group but no dice. Please and th [16:55] ok...i'll try to be less vague...does anyone know if i change the server hostname, will i need to re-install or re-configure apache2, postgresql, postfix, dovecot? [16:56] ^ to add where ever you see a space I backslashed to accept the space [16:56] jauntypaperclip: Error: "to" is not a valid command. [16:56] just to add, to add where ever you see a space I backslashed to accept the space [16:59] Hello [17:02] hello, i am trying to install ubuntu 10.04.1 x64 server, downloaded server image several times and burn on few cd/dvd and it always halts on "starting up partitioner 47%". tried on 2 diferent PC same result. is there a option to try install from USB key or from net ? any help is appriciated [17:03] m1r1: have you compared md5sums for the downloaded file? [17:04] and the burned disc? [17:04] that's the best way to verify it [17:05] i have not verify file or disc, how could i do that ? [17:06] what kind of operating system is the iso located on? linux/windows? [17:06] 1st was on linux, now on windows [17:06] both same error on partitioner 47% [17:07] window's doesn't have md5sum installed, if you can run it on a linux box, the command is md5sum then it'll generate a number for you [17:07] erimar77: sadly , no linux box available here :/ [17:08] m1r1: sure, something like this can work http://www.pc-tools.net/win32/md5sums/ [17:08] m1r1: and the list of hashes is here https://help.ubuntu.com/community/UbuntuHashes [17:09] basically, it verifies that nothing was corrupted in transit [17:09] erimar77: if somthing was corupted in transit, how much is posibility that it breaks on same location in install ? [17:10] couldn't tell you [17:10] this is step one... the easiest to figure out [17:10] erimar77: i will redownload again and then verify before burning [17:10] many thanks m8 === bigjools is now known as bigjools-afk [17:13] m1r1: another thing to try is a generic desktop version and see if you can get to the "try it out" desktop [17:13] if the numbers don't match, your download is corrupt or incomplete [17:13] which can cause any number of problems [17:16] is there a reason why apache2-ssl-certificate missing from ubuntu server 10.04? have i not installed something, or is this still an open issue? [17:18] nevermind...i've found what looks like a useful doc on help.ubuntu.com [17:18] Hey awesome peeps, I have one more question. I'm trying to set up a group for my file server. I currently have access however another windows machine that I'm using does not. I tried doing sudo chown ubuntuserveruser /srv/samba/share/group folder, sudo chgrp groupname /srv/samba/share/group folder, sudo chmod 770 /srv/samba/share/group folder I ensured that the other user is added to the group but no dice. Please and th [17:18] Hmm. I've just migrated a filesystem, but need to 'fix grub' - i.e. install it. [17:18] i.e. I added a new drive to a system, copied the root FS, fixed the fstab, etc etc - but I don't know how to straighten out the bootloader. [17:20] I'm currently in the installer 'rescue' mode, no root FS loaded. [17:22] ..oh, there's a helper in the installer. [17:26] how bad is it to use the maverick repository from lucid to maverick to upgrade the cups package? [17:26] what type of problems can I expect? [17:28] very, very bad problems [17:28] your package versions will get horribly messed up [17:28] if you're going to go that far, manually download the cups packages you want and install them with dpkg [17:28] don't add a maverick repository to lucid [17:31] ubuntu lucid server appears to be blocking udp ports by default for some reason [17:31] I'm starting freeradius and if I use any port but the official freeradius port, its fine [17:32] okay my bad [17:32] it's something stupid freeretard is doing [17:43] folks, isn't there a way to do a echo 1 > /proc/sys/net/ipv4/ip_forward using sudo and not sudo su - ? [17:43] alcy: there is [17:43] alcy: tee [17:44] ivoks: cool, thanks [17:44] echo "1" | sudo tee /proc/sys/net/ipv4/ip_forward [17:45] qman: thank you ... just to clarify your saying download the cups1.4.4.deb and just use dpkg to install [17:45] is there a recomended place to download newer packages from for lucid? [17:46] ivoks: cool. [17:48] ivoks: but is that a work around, or a std. way of manipulating such system variables ? [17:48] digitolx, the backports repository is best, a PPA second best, and manually installing the way I explained would be third best [17:50] is there a public PPA or other repo that contains a more up to date version of the ec2 api tools? [17:50] can someone please tell me wtf console-kit-daemon is and how I can disable that? [17:50] alcy: standard way is /etc/syslog.conf [17:50] bah [17:50] /etc/sysctl.conf [17:50] alcy: ^^ [17:51] alcy: set some.var = something in /etc/sysctl.conf and run sysctl -p [17:51] ivoks: or rather, why doesn't a direct sudo work, where as piping the value to tee works ? ...yeah I could've changed sysctl but still, it would be nice to know this [17:51] RoyK, IIRC, that's what was causing a memory leak for me in jaunty, I removed the package it was a part of and it solved the problem [17:52] with no other noticeable effects [17:52] alcy: cause you are piping as user [17:52] qman__: it still leaks in maverick [17:52] alcy: and tee is a program which you run as root [17:52] qman, understood.. if the first 2 are not available for cups 1.44 is there a recommended place to get a deb.. or can you use dpkg to install src ? sorry for the qustions I'm just trying to get this right in my head [17:52] digitolx, you should get the deb from the ubuntu packages site, under maverick [17:52] ivoks: ah cool ! thanks :) [17:52] it may depend on other maverick packages [17:53] qman: thank you very much [17:53] RoyK: thanks for the tip. [17:53] but at least that way you don't have to worry about packages outside of that dependency chain [17:56] right on,, I was debating on enabling maverick then disabling once I updated the cups packages .. being that there are more than one package depending .. ie libcups2-dev ,, cups-bsd, cups-client.. ect... [17:56] that's a very bad idea, because once apt-get is aware of all the new versions, you will have dependency hell [17:57] just do it manually [17:57] You can also rebuild it.. which would be awesome if you did so, and then filed a backport request if it worked without changing the source. ;) [17:58] will do,,, since that is the case should I get all the packages that are installed now affiliated with cups or just the main cups1.4.4 [17:58] well, you want to replace as few packages as is necessary [17:59] so I would download the main cups, and attempt to install it [17:59] then find what it complains about, and attempt to install those [17:59] rinse, repeat until happy [17:59] cool,, think my questions are over for now lol.. thank you very much === bigjools-afk is now known as bigjools [18:01] re: consolekit, I'm still not exactly sure what it's good for [18:01] supposedly it keeps track of user sessions, but my system works just fine without it [18:02] and the internet is not clear on what uses it or why [18:02] qman__: I just removed the whole thing on this Maverick test box, and it runs well without it [18:03] seems if running a box with rapid cron execution, we have a few of those, consolekit eats up a large chunk of memory over time [18:07] hello, i am trying to install ubuntu 10.04.1 x64 server, downloaded server image several times and burn on few cd/dvd and it always halts on "starting up partitioner 47%". tried on 2 diferent PC same result. MD5 hash was checked with latest download and it was correct and also fails on partitioner 47%. is there a option to try install server from USB key or from internet ? any help is appriciated [18:08] m1r1: perhaps hardware problems? the cd drive? [18:08] m1r1: anyway - try memtest86 first [18:08] qman: one last question.. would this be the same process going back to karmic ? [18:09] digitolx, yes [18:09] roger [18:09] RoyK: tested on two diferent PC's, same error occurs on both "47%" halt [18:09] same cd drive? [18:09] or cd [18:09] this last machine is brand new from store with new cd rom [18:09] wasted about 6 cd's already :) [18:10] _burnt_ with the same cd writer? [18:10] m1r1, did the md5sums match up? [18:10] also, sometimes a full speed burn won't work, try setting it lower [18:10] qman__: yes md5 its corect [18:10] cdrecord -speed 10 [18:10] or something [18:10] might help [18:10] RoyK: downloaded from two diferent machines , burned on two diferent writers (dvd and cd) [18:11] same sort of hardware? [18:11] RoyK: i give a shoot on 8x [18:11] as in same type of server machines? [18:11] ok [18:11] RoyK: two completley diferent machines [18:11] wierd [18:11] yea [18:11] * RoyK has installed quite a few lucid servers [18:12] i didnt manage to pass 47% for few days now :) [18:12] shit [18:12] well, try with 8x [18:12] if that fails, try with a usb pen [18:13] if that fails, pray to the nearest god and hope [18:13] yes, that what i was planing to if this fail, but dont know procedure for server install from USB [18:13] lots of factors can cause a bad burn, slowing it down improves the chances of success [18:13] m1r1: if you have an ubuntu desktop machine, it's easy [18:14] low quality media, worn out drive, dust, voltage fluctuations, cosmic rays, planets aligning, etc [18:14] m1r1: otherwise, the ubuntu 10.10 download pages has some links [18:14] indeed, especially planet alignment [18:14] don't forget the mass of the person next to the machine [18:14] :D [18:15] that for sure aint issue here :) [18:15] burning on 8x , disc-at-once [18:16] <_ruben> removable media, bah .. pxe boot/install for the win ;) [18:16] _ruben: takes a little longer to setup, though [18:17] well i need server for pxe :D and it wont install :D [18:17] unlock the drawer with the key inside [18:18] RoyK, http://fedoraproject.org/wiki/Desktop/FastUserSwitching#ConsoleKit -- it provides some magic for gvfs and similar, but there is still no explanation why it exists on systems without X [18:18] Hey sorry for reposting, wondering if someone is able to help. I'm trying to set up a group for my file server. I currently have access however another windows machine that I'm using does not. I tried doing sudo chown ubuntuserveruser /srv/samba/share/group folder, sudo chgrp groupname /srv/samba/share/group folder, sudo chmod 770 /srv/samba/share/group folder I ensured that the other user is added to the group but no d [18:18] qman__: I just installed the damn thing [18:18] s/installled/uninstalled/ [18:19] yeah [18:20] problem solved - NEXT! [18:21] <_ruben> RoyK: a little, yes, which is earnt back within a few server installs, especially when messing with lots of test servers/vms :) [18:21] * jauntypaperclip bounces up and down on her weird fitness ball chair [18:22] jauntypaperclip, your message is getting cut off, as your IRC client is not smart enough to line break at the limit [18:22] regarding your problem, have you tried rebooting the windows client? [18:24] windows is very stupid when it comes to credential management, and once it decides to authenticate as guest or as the logged-on user, it will refuse to do otherwise until you log off/log back on or reboot [18:25] 8x burn done, going to test 47% again :) [18:26] m1r1: check the cd for defects in the installer [18:26] m1r1: maybe you have a bad batch of CDs [18:27] I've actually found that feature to not be reliable [18:27] it has passed discs that don't match the md5sum [18:28] qman__: it doesn't check md5sum i thought. just some kind of basic defect finder :). otherwise it would take a very long time [18:29] pmatulis: how you mean bad batch od cd's ? [18:29] failed sucsesfully on 47% again [18:29] bad media [18:30] failed successfully, i like that [18:30] good outlook [18:30] cant be all cd's i tryed are bad (they not same brand, or even bought at same store) [18:31] m1r1: did you check for defects as i suggested? [18:31] pmatulis: how to check for defects ? [18:31] m1r1: the installer should have an option [18:31] sorry qman__ I'm using the webchat in case you need to know here's the second half of the message. I tried doing sudo chown ubuntuserveruser /srv/samba/share/group folder, sudo chgrp groupname /srv/samba/share/group folder, sudo chmod 770 /srv/samba/share/group folder I ensured that the other [18:32] pmatulis: got it [18:32] will check now [18:32] I will try what you suggested right now though! Thanks! :) [18:32] the other user is added to the group but no dice. Please and thank you! *argh!!! darn webchat [18:34] jauntypaperclip, have you created a password for that user with smbpasswd? [18:34] qman__: yup sure have. That's why I thought it was weird that I couldn't get in on the other machine. I can do it on mine. [18:35] like my own windows box aside from the ubuntu server [18:35] hello, https://bugs.launchpad.net/ubuntu/+source/vim/+bug/572627 [18:35] Launchpad bug 572627 in vim "ftdetect scripts not loaded from directories added to runtimepath" [Undecided,New] [18:35] is there a way of writing the file names already d-loaded to a file so the next run of wget won't re-download the same files? [18:36] I tried to comment syntax on in /etc/vim/vimrc to fix the issue but pathogen still not detects new files [18:36] any idea ? [18:37] qman__: is there something I should be doing to fstab? in 9.04 I had to add acl to fstab... but I don't think that's still needed. Just for my own curiosity. :) [18:37] jauntypaperclip, not for basic samba file sharing [18:38] does your share in smb.conf have any allowusers statements? [18:38] qman__: hmmmm but what about managing users and groups? [18:40] qman__: ; add group script = /usr/sbin/addgroup --force-badname %g ?? it's still commented... [18:41] jauntypaperclip, to add a user to a group, use `sudo adduser username groupname` [18:41] ok found the fix ! [18:41] qman__: I've added all the users to the groups... [18:42] jauntypaperclip, you can verify by running `grep ^groupname /etc/group` [18:42] i have checked install CD burned on 8x for defects, it replyed that cd is ok [18:43] qman__: yup already parsed it out of the group file using grep to validate and all my peeps are there :( [18:43] jauntypaperclip, pastebin the share from /etc/samba/smb.conf [18:45] qman__: http://pastebin.ubuntu.com/512535/ [18:46] sorry :( I kinda left the comments in from the sample config === _TechAway_ is now known as _Techie_ [18:46] it helps my backup in case of emergency [18:46] change guest ok to no [18:46] again, windows is fairly stupid and will attempt to authenticate as guest [18:46] and some windows installs will while others do not [18:47] also, you're probably going to want create mask to be 775 [18:47] or 770 [18:47] but that's up to you [18:47] ahhh uncomment it huh? *facepalms* [18:47] yeah I gotta get around to changing it to 770 [18:48] no, not under the printers share [18:48] under your share [18:48] you have "guest ok = yes" [18:48] change it to no [18:48] ohhhhh!! oops thanks. Thought you meant global [18:49] many windows clients will attempt to authenticate as guest first, and with that option, they will succeed, then not ask for any credentials [18:49] so, i : 1. downloaded server 10.04.1 x64, 2. verified md5 corect, 3. burned with 8x speed disc-at-once , 4. tested disk for errors is ok... install halts on 47%... any tips how to proced with install ? [18:49] and subsequently fail to access any of the files [18:50] m1r1, do the two computers you tried to install on have the same type of disk controller? [18:50] qman__: gotcha. though I entered the credentials. I was able to validate that the user can see their folder with permissions set so that only they may see it and not see my folder for instance. I confirmed that it worked so I don't think it's credentials [18:50] qman, you mean ide/sata ? [18:51] m1r1, no, I mean the same brand and model [18:51] or, if they're using the same exact hard drive [18:51] qman__: i dont know :( [18:51] hardware is diferent [18:52] what I'm getting at is, a failing disk or an issue with a particular type of disk controller might be the problem [18:52] qman__ is the man! [18:52] but if they're not the same disk, and they're not the same type of controller, that's not it [18:53] qman__: the first one was older PC , core2duo , this one is quad 8200 (brand new from store) [18:54] m1r1, what chipset does each use? [18:54] for older one i dont know, sent it back, for new one i would need to look [18:57] also, how much time did you give it at 47%? [18:57] and how big is/are your hard drive(s)? [18:59] qman__: the guest thing worked. you are in fact... the man! [18:59] glad to hear it [19:01] qman__: hdd is 1tb , when it comes to 47% there is no activity on PC at all, no matter 1h or 15min, just halts [19:02] i went to bios and set port 1-4 AHCI (was on IDE), but still same result [19:03] will load now live cd to see what chipset (if it manage to load) [19:03] http://29.media.tumblr.com/tumblr_la67s29j0Y1qdr7qxo1_500.jpg [19:08] qman__: Sorry for ever doubting you even for a half minute. lol. The windows thing authenticating as guest after you've defined the user id and pass is so weird. I may set up a batch file if utterly required. But yeah thanks again!! [19:26] thanks for help people, i return this box and try tomorow with other one hopefully [19:31] New bug: #660105 in apache2 (main) "when deflate is enabled, please also compress CSS and JS by default?" [Undecided,New] https://launchpad.net/bugs/660105 [19:33] m1r1: no luck with the new image? [19:35] RoyK: no luck at all :( === _Techie_ is now known as _TechAway_ [19:35] m1r1: can you try to install 8.04? [19:36] RoyK: will return this box to shop and get new one during this week i hope, its acting real weird :/ [19:37] I'd try 8.04 first if I were you [19:37] perhaps there's some local issue [19:37] since two different boxes fail.... [19:37] yes, both on 47% which is most strange :/ [19:37] indeed [19:39] what i found out, this box is AMD pehonm and old was core2duo so, as qman__ sugested maybe could be bad hdd controler, but strange that same block on 47% for both [19:40] AMD is new from shop , recived yesterday === jsalisbury is now known as jsalisbury_brb [19:44] m1r1: do you have any chance to test 8.04? [19:47] RoyK: no more empty cd's :( [19:48] RoyK: only USB stick, but dont know how to preprare for server install from usb key === Pilif12p|afk is now known as Pilif12p [19:50] m1r1: http://www.ubuntu.com/desktop/get-ubuntu/download <-- instructions there - it applies to building an USB boot device from an iso image, not necessarily u10.10 [19:51] RoyK: thanks, will cehck it out now [19:51] * RoyK is quite sure he said that some hour or two ago [19:53] RoyK: there is no 8.04 in that list [19:53] in usb installer list [19:54] what OS is your PC running? [19:54] win7 [19:55] http://www.pendrivelinux.com/universal-usb-installer-easy-as-1-2-3/ [19:55] it's all on that page [19:55] so i can also format usb and put 8.04 server install with universal installer ? [19:56] i read this is only for live cd roms, or am i wrong ? [19:58] it shouldn't be a problem [19:58] it's just the bootstrap that needs a little fiddling [19:59] I want to know how to set up sendmail on linux server so I can make php send emails, can anybody help? [19:59] is it possible to turn on tab completion after the sudo command? if so how? [20:00] martin___: no need for sendmail - that's ancient history - apt-get install postfix [20:00] that'll fix most of it [20:01] no configuration then? :RoyK [20:01] martin___: should work well out of the box [20:01] clayd: google for bash custom tab completion [20:01] RoyK: i ordered intel quad procesor and got amd quad, will return it tomorow, and get intel during this week, then i retest all [20:01] Royk: thanks :) [20:01] m1r1: imho the AMDs are just as good, though [20:02] well, depends on model, obviously [20:02] yes, but firm shipped wrong pc [20:02] RoyK: I have been looking around and cant seem to make it work. I uncommented the lines in bash.bashrc then rebooted and still no go? [20:03] I think you need to configure it, clayd [20:03] ahhh. [20:03] how can I display the stats like I see when I ssh into ubuntu server? [20:05] cdubya: landscape-sysinfo [20:05] nice, thanks === ian_brasil_ is now known as ian_brasil === jsalisbury_brb is now known as jsalisbury [20:17] cdubya: if you install webmin you can see this all graphically, best of all, its free! [20:19] !webmin [20:19] webmin is no longer supported in Debian and Ubuntu. It is not compatible with the way that Ubuntu packages handle configuration files, and is likely to cause unexpected issues with your system. See !ebox instead. [20:19] martin___: we generally recommend ebox, not webmin [20:20] SpamapS: I assume its personal preferance? [20:21] martin___: read that line from ubottu again === _TechAway_ is now known as _Techie_ [20:28] Does anyone know if its possible to use extended attributes over NFS? [20:28] it is [20:42] martin___, yeah, I've installed webmin but it's been a long time. :-) [20:43] cdubya: well apperently it isnt supported, however i havnt had any problems [20:43] martin___, no worries. Thanks! [20:48] dear god, highlight! [20:50] can anybody tell me how to install ebox with all modules? thought i woulde give it a try? [20:55] martin-: as I see it they regard their product more as a standalone tool based on ubuntu than as something to add to ubuntu [20:57] I'm not a fan of web interfaces that I didn't write myself [21:00] martin-: an on top of this: ebox uses only its own config files, not the standard ones from the system [21:00] *and ... [21:06] guntbert: what? [21:07] martin-: I don't think it is wise to put ebox on top of a running system and expect the system to still work [21:11] guntbert: no idea what you're talking about [21:13] martin-: sorry, my client lead me to a faulty nick completion - there are more martin... around [21:13] martin___: as I see it they regard their product more as a standalone tool based on ubuntu than as something to add to ubuntu [21:14] is it an issue if i put a 64-bit server as a Distcc helper to a 32-bit client ? [21:14] martin___: and on top of this: ebox uses only its own config files, not the standard ones from the system [21:15] guntbert: so what are you saying, i shouldnt install it? as i have been told i should, webmin is no longer supported apparently [21:17] martin___: yes, webmin is no longer supported (for a long time now) - but I was very dissatisfied with ebox, I removed it as soon as I saw how it handled the system configuration [21:18] martin___: I cannot give real advice here - just my own experience shows that ebox seem in no way to be a tool like webmin [21:18] tbh if it works and doesn't create security issues then I'm happy :guntbert [21:20] martin___: the problem for me was that you seem to be completely sold to that system once you start using it - I didn't see a way back - with webmin it is enough to not use it or unistall it and the system will behave exactly the same [21:22] good, because i have just uninstalled webmin on my production server with a view to maybe install zentya (formally ebox) that i am currently installing on a test server [21:32] SpamapS, ping [21:32] other-server-n-upstart-server-enhancement [21:40] is anyone aware of a public PPA or other repo that contains a more up to date version of the ec2 api tools? [21:48] hi, seriously is using ubuntu + tomcat going to be an hassle or its quite simple [21:48] i don't like not having gui [21:48] maybe i'm not clear [21:49] i want to have a tomcat server, i was wondering if i use normal ubuntu and add tomcat am i going to suffer a lot from it === ivoks is now known as ivoks-afk [21:55] ScottK, i wonder if you could bump https://bugs.launchpad.net/lucid-backports/+bug/649742 [21:55] Launchpad bug 649742 in lucid-backports "backport maverick ec2-api-tools to lucid" [Undecided,Confirmed] [21:55] or anyone who is backporter [21:55] * ScottK looks [21:56] Madwill: The Ubuntu tomcat deb should work well for you, actually. [21:56] smoser: Done. [21:57] thanks [21:57] ScottK, gracias. so is there another way to get a bump on such things othe rthan pinging you ? [21:57] smoser: Currently probably not (and I don't mind) as we are very short of people looking over backports. [21:57] in fact i'm going for openfire rpm which is sort of build on tomcat [22:01] Can anyone explain how I can use `route` to route all traffic through an interface? [22:02] ScottK: what's actually involved in that? === _Techie_ is now known as _TechAway_ [22:11] ajmitch: Not a lot. It's just reviewing bugs to make sure they've been properly tested. [22:11] ajmitch: The biggest thing is checking to make sure rdepends are tested (if any). [22:11] DormantOden: You need to set the default route to use that interface [22:11] The test standard is builds, installs, runs. [22:12] But you can't break rdepends. [22:12] That and we want an upgrade path, so if you want to backport Natty to Lucid, you probably also need to backport to Maverick (for example) [22:12] makes sense [22:18] Is webmin ok to use with ubuntu server? [22:19] Sure [22:19] Ive heard it isn't supported anymore? [22:19] Ive heard it isn't supported anymore? [22:20] does anybody use it? [22:20] It's still supported and I use it [22:20] great thanks :) [22:21] They've made it a lot nicer to look at too [22:21] Since the ancient days [22:21] haha, I wouldnt know, im not that ancient myself :) [22:21] It used to look like this http://www.obsceneart.com/images/icons-icons-icons.png [22:22] wow :D [22:22] old [22:26] New bug: #660199 in samba (main) "package samba-common-bin 2:3.4.7~dfsg-1ubuntu3.2 failed to install/upgrade: subprocess installed post-installation script returned error exit status 2" [Undecided,New] https://launchpad.net/bugs/660199 === _TechAway_ is now known as _Techie_ [22:29] murphy_: To be clear: It's not supported here. We'll send you to ask the webmin people if you have any problems. [22:35] ScottK: Thanks, i just wanted to make sure it would cause any problems to my production server [22:35] murphy_: I wouldn't make that assumption. [22:35] ScottK: sorry? [22:35] In the past, I've helped people fix postfix setups where webmin ate the config file. [22:36] I've no idea why, but I wouldn't assume it's safe. [22:38] murphy_: it's not in the archive for a reason [22:41] My ubuntu server installation does not give a login prompt on boot, I have to press CTRL-ALT-F1 before logging in. Is there a fix for this? I'm using Lucid. [22:44] eroick is it a fresh install or have you used the server for a while? [22:45] chrismat: fresh === _Techie_ is now known as _TechAway_ [22:46] Will everybody stop talking in riddles and just TELL me whats wrong with it :ScottK :jcastro [22:47] murphy_: I don't use it, so I don't know exactly. I can tall you that I've helped people fix postfix configs that got eaten by webmin. [22:47] !webmin [22:47] webmin is no longer supported in Debian and Ubuntu. It is not compatible with the way that Ubuntu packages handle configuration files, and is likely to cause unexpected issues with your system. See !ebox instead. [22:47] it's not a riddle, it's just not recommended on debian/ubuntu because it doesn't work [22:47] That's the official position. [22:47] ... reliably. [22:47] It does work fsvo work, but then stuff happens. [22:47] well, he did say _production_ system so I assumed reliably. :) [22:49] so can i configure a tty to appear right after boot? I'm using serial to get into this machine, and I cant send the CTRL-ALT-F1 over GNU Screen (or can I?_ [22:55] eroick: if serial console is configured correctly, you'll get a login on ttyS0 or whatever you're using [22:56] RoyK: I get the boot messages, and it sits at Starting NTP server ntpd [ OK ] [22:56] but I know that CTRL-ALT-F1 will get it to a login prompt, i tried it in qemu [22:57] configure a tty in /etc/init/ttyS0.conf or something [22:58] the serial console in kernel won't setup a getty for it [22:58] example http://pastebin.com/09k2y9uZ [22:59] and then just do screen /dev/ttyUSB0 baudrate? [22:59] from the computer i'm connecting from [23:00] or minicom [23:00] do I need to do an rc-update? [23:00] I'm not sure if screen supports serial stuff [23:00] never used that, though [23:01] minicom does [23:01] i've been using screen, it works to display all the boot messages, i just don't get a TTY [23:01] but I don't get a tty with the same image in qemu... [23:02] dunno then [23:02] I haven't been using qemu a lot [23:02] * RoyK just uses serial consoles for physical boxes including good old RS/232 === erichammond1 is now known as erichammond [23:06] What am I doing wrong... http://pastebin.com/eE0jZ20P Error=salesperms.sh: 4: SalesDocuments[0]=/media/data/Sales: not found [23:06] salesperms.sh: 28: Bad substitution [23:09] RoyK: thanks.. i'll try this when i have a chance. one more question: do you know where the startup of ttys is controlled? [23:09] /etc/init [23:09] thre are scripts there starting them [23:09] or config files, even [23:09] ok, i'll check those too [23:10] the one I pastebinned was my /etc/init/ttyS0.conf [23:11] can you send CTRL-ALT-* from minicom? [23:12] why would you? [23:12] if a serial console (aka getty) is set on a serial port, it'll work [23:13] how would you send it? [23:13] sorry - why would you want to? [23:13] to change tty [23:13] on a serial console, you just have one tty [23:13] hmm, ok [23:14] virtual ttys is a "new" thing that came up only 20 years ago or something [23:15] alright [23:15] im not sure what my serial connection is doing... it's monitoring the entire boot process [23:15] ie, before a tty even exists [23:15] that's kernel output [23:16] you need to launch a tty on the serial link to be able to log in [23:16] alright, i'll make the ttyS0.conf [23:16] and will I need to add that to start up somehow? [23:17] iirc init will read through that dir and start whatever is set there [23:17] ok [23:17] thanks [23:23] I am getting errors when starting apparmor about missing PROC and HOMEDIR variables. It's as if the files in tunables/* were not being parsed. [23:23] Found reference to variable HOMEDIRS, but is never declared [23:24] dlublink: does the profile in question have "#include " near the beginning of it? [23:25] nope [23:25] dlublink: then you need to add it, in order to get the definitions included. [23:25] Sorry, my bad, I copy/pasted a script and didn't notice that because it starts with a # [23:26] Wonderful, thanks [23:26] Fixed [23:59] my movies on samba shares sometimes stutter, how do I know if its my network bandwidth or resource strain on server thats causing it