[00:41] New bug: #259595 in samba (main) "synaptic unable to install - broken dependency samba-common" [Undecided,New] https://launchpad.net/bugs/259595 [00:49] What is the best way to get a package listing with versions? [01:28] kirkland: ya, I posted a couple of links concerning RAID, and since you're the raid master now, the props go to you :-) [01:36] pschulz01: dpkg -l > yay_packages.txt [01:50] sommer: Ta :-0 [01:51] sommer: Thanks.. together with a git controlled /etc this gives what I want. [01:55] np [01:56] sommer: What is the state of the current 'ubuntu-server' documentation? [01:57] pschulz01: here's a link to the most updated content on the web: http://doc.ubuntu.com/~mdke/test/serverguide/C/ [01:57] pschulz01: there's newer content in the bzr branch: bzr branch lp:ubuntu-doc [01:59] sommer: I am in the position where I can offer some paid work to some local 'interns'.. and updaing the serverguide would be good use of this resource. [01:59] pschulz01: sweet, the apache section needs some love... I was thinking of adding instructions for a2enmod, a2ensite, etc [02:00] pschulz01: also a big help is to just work through the current sections and make sure things like package names, path names, etc are still valid [02:02] sommer: That could e a good introductory excersise. [02:02] pschulz01: also a good learning experience :) [02:03] sommer: These guys a keen, but complete newbies. [02:04] heh, I remember those days... ah youth === anonimous is now known as newtonglez [02:24] could someone point me to a guide to set up raid1 post install without loosing data? ... i tried this one > http://gentoo-wiki.com/HOWTO_Migrate_To_RAID < but im having problems booting to the new raid partition [02:51] New bug: #259609 in samba (main) "package samba 2:3.2.1-1ubuntu1 failed to install/upgrade: subprocess post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/259609 [03:07] good morning guys :) [03:17] morning [03:17] what's the topic? [03:18] how to install mysql connector for java? [03:19] ugh, that's hard [03:19] i dont know [03:19] anyone? [03:36] does anyone know how to change the default runlevel in ubuntu? [03:40] michalski: What problem are you trying to solve? [03:41] ScottK: when my computer boots it goes into runlevel 3, I need it to go into 5 [03:41] *want it [03:42] Why? [03:42] I'm not sure what you think this will accomplish? [03:43] I jst recently customized the runlevels using sysv-rc-conf and now it send it off to 3 [03:43] (it will send me into a graphical session?) [03:43] Right because in Ubuntu 2 - 5 all do the same thing. [03:43] unless customized... [03:43] Yes. [03:44] I guess I still don't understand why. [03:44] I would want to do this? [03:45] Yes. [03:46] Note that unless you're on Dapper you're using Upstart with sysv emulation and not a try sysv init system. [03:46] different environmental conditions change around my server causing me to want to change runlevels at moments notice, eg: powergoes out; server going down for maintenence; unauthed user detected, ect... [03:47] pardon? [03:48] Ubuntu doesn't use the traditional sysv inits. It has Upstart instead. Upstart emulates sysv very nicely, but it is a different system. [03:48] (what is upstart.....) [03:49] * ScottK hands michalski http://en.wikipedia.org/wiki/Upstart [03:49] ah [03:50] so why would this be a problem...we can still change the booted runlevel right? [03:51] My recommendation is that althought the sysv init emulation in Upstart has been very well tested for normal use, you might be making different assumptions about normal use than the Upstart author did. [03:51] I would assume, but I've never needed to do it. [03:51] hmm [03:53] BTW, I didn't know until I read it there that Fedora had started using it. [03:55] hey I found it! /etc/inittab [03:55] theres a section in there for default runlevel [03:55] ...if the file existed :s [03:59] I think that only even exists if you upgraded the system from Dapper. Dunno if it still does anything or not. [04:01] i've only been using ubuntu since 7.04 [04:03] i was just thinking....if you can specify the runlevel at boot (via manual input to grub), can't I just modify /boot/grub/menu.lst? [04:11] ok back in 10 gonna try it === wantok is now known as `6og [08:31] rndc: error on bind restart?? [08:54] hey [08:54] some distro's force users to have strong passowords [08:54] is it possible to automatically enfore a few rules in all users passwords? [08:55] i mean that will change then from now on [08:56] <`6og> yes it is. [08:56] <`6og> pam-cracklib would be a place to start [09:27] Hi there! I have gone through the testing steps at https://wiki.ubuntu.com/EncryptedPrivateDirectory#Testing . I have got this: http://nopaste.biz/49942 "not such device" [09:28] sven-tek: okay, does /home/shaggy/Private and /home/shaggy/.Private exist? [09:29] yes, they are directories [09:32] sven-tek: ls -alF ~/.ecryptfs [09:33] http://nopaste.biz/49943 [09:36] sven-tek: which mount.ecryptfs_private [09:36] sven-tek: ls -alF `which mount.ecryptfs_private` [09:37] sven-tek: mount.ecryptfs_private [09:38] here it is: sven-tek@bluebox:~$ which mount.ecryptfs_private [09:38] /sbin/mount.ecryptfs_private [09:38] sven-tek@bluebox:~$ ls -alF `which mount.ecryptfs_private` [09:38] -rwsr-xr-x 1 root root 9632 2008-08-15 21:12 /sbin/mount.ecryptfs_private* [09:38] the .last filenameit is backgrounded red [09:40] But file seems to be okay: file /sbin/mount.ecryptfs_private [09:40] /sbin/mount.ecryptfs_private: setuid ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), for GNU/Linux 2.6.8, dynamically linked (uses shared libs), stripped [09:42] oh moment, iam running a custom bulid kernel. Are there any modules involved? [09:45] kirkland, sorry its my fault! === mgdm_ is now known as mgdm [10:16] `6og, thanks. have some example for ubuntu? I found just some redhat stuff that claims to have been patched in some specific way [10:40] hello, I have installed pptp vpn server, but my clients suffer from very low service, what u think the reason could be [11:17] This may be a dumb question, but I just have to ask: After installing drupal in Hardy, how can I mount a site with it? I have no idea on how to set up a site with it. Is there a sh script or something to run in a desired directory? [11:17] I'm clueless, really... [11:18] normally I just untar the file from drupal.org in a directory and there I install it, but since I'm using Ubuntu server I wanted to figure out how that works [11:18] thx in advance [11:21] hubuntu: Look at where it installs it: dpkg -L drupal [12:04] clusty, no i dont, sorry [12:04] Kamping_Kaiser: !!! [12:05] :O [12:12] found it. Thanks jpds! Now I just have to figure out how to make another drupal site on, say, a user home... anyone? [12:38] hubuntu, that sounds like the sort of stuff which would be covered in the docs ... (/usr/share/doc/$packagename/) [12:40] yes Kamping_Kaiser, I found out. Thx [12:41] hi everyone [12:41] hi didrocks [12:42] I am interested in working on ufw integration (now that removing multiuser tag is done :)) [12:42] I saw this page, can I directly edit it to assign, let's say apache2 and openssh for a begining ? https://wiki.ubuntu.com/ServerTeam/Roadmap#UFW%20Package%20Integration [12:44] didrocks, probably it's ok, but it would be better to coordinate your effort with Jamie: https://wiki.ubuntu.com/JamieStrandboge [12:45] in that way he will know, and if there's work done into that, he will surely know where to point you at [12:45] yes, he seems to not be there at this time. I will send him an email, thx :) [12:46] hi everyone, good afternoon! [12:46] hi byte_slave [12:46] hi byte_slave [12:47] how can i completly uninstall a package even the /etc config files [12:47] i used apt-get remove --purge packagename but /etc files remain [12:47] erase those [12:47] no prob to do that manually? [12:48] not really, but you mjst be sure nothing else is using those files [12:48] what program is it you want to uinstall? [12:48] i installed mailx and sendmail [12:48] but i want completly uninstall all this and start from zero [12:49] so i have yet files in /etc/mail and /etc/mailrc [12:50] you should be careful there... anyone has some advice to share? [12:50] byte_slave, you can either reinstall it, or `dpkg -l $package` and remove the files listed by hand [12:51] great, i ran that seems cleaned :) [12:58] dpkg -l is really useful, I can't believe I had forgotten such a feautre existed. Can it be used to find a certain library as well? [12:58] like libjpeg or something? [12:59] libjpeg.so or what the name was (needed for livecam.sourceforge.net) [13:37] morning [13:38] * delcoyote hi [13:52] question: i removed mailx and sendmail but sendmail command and mail command still in /usr/bin is that supposed? [13:56] New bug: #259740 in bind9 (main) "DNS: not able to use IPv6 Resolvers" [Undecided,New] https://launchpad.net/bugs/259740 [14:05] Did somebody manage to boot from an iso with last ubuntu-xen-server with an HVM kernel VM ? [14:06] Tryed all the tap:aio: / mount / losetup to loop methods [14:07] none made the trick, stuck at bios after : Booting from CD-ROM ... [14:11] i forgot the apt-get autoremove i guess :) now my question above is answered === dusty_ is now known as dusty-- === dusty-- is now known as dusty--- [14:49] How do you boot into single user mode? [14:49] a la init 1 ? [15:02] hellow [15:03] you guys know a solid SATA RAID card that works perfectly with hardy server? [15:03] i know about areca, but that's 400 EUR :) something a bit cheaper but still reliable out there? [15:07] a secondhand areca/3ware ? ;-) [15:09] <\sh> don't compare 3ware and areca...*grmpf* slow but stable vs. fast but unstable, at least our arecas here :( [15:16] init-top in the initrd, is that called before init is executed or just after? [15:16] We have a couple of new arecas here that seems to work fine. The previous test a couple of years ago made us not want them at all though. So things might have improved. [15:17] But out of choice I'd go for a 3ware card [15:17] after updating the update-initramfs do I need to reboot.. unfortunately on server udont get the reboot request like in client :-( [15:17] hm [15:17] and what about lsi megaraid? [15:17] that's a lot cheaper :) [15:18] areca = 490EUR, 3ware +- 270, lsi 170 [15:18] ive seen megaraid modules a lot, i think dell uses those, right? [15:18] well, if you want to go the software raid path, you can get by with a resonably cheap plain sas/sata card [15:18] nah no software raid [15:18] i have software raid now [15:18] megaraid in general is horrible, but there might be a decent one or two under that name too [15:18] maswan: define horrible? [15:19] verwilst: weird crappy management, strange performance issues, not very reliable [15:20] hm [15:20] maswan: dell uses those too eh? [15:21] verwilst: yes, but with a different firmware [15:21] oh [15:21] so a second-hand 3ware would be my best bet? [15:21] So it'd depend on how old a dell raid it is, etc. Some of the older ones were really bad. [15:23] I don't keep up much with recent dells, we don't buy them anymore [15:23] well my server is fully intel [15:23] but i want to get rid of my software raid [15:24] i have 1 raid 5 now with 3 disks [15:24] and /boot on software raid = troubles i heard [15:27] ah, well, software raid1 is fine, but other raid levels might be troubles. [15:27] but if you want reliable, I'd recommend 3ware [15:27] maswan: http://www.descom.be/shop.php?cPath=87 < but which one :P [15:27] so many to choose from [15:27] 4ports , that's for sure [15:27] <\sh> smartarray ;) but this means switching to hp [15:28] i guess all of them are supported? [15:28] <\sh> but imho they are using lsi nowadays too [15:28] verwilst: sure. the higher numbers are usually faster. [15:28] \sh: since quite some time, IIRC. but totally different firmware. [15:29] <\sh> maswan: yes...I'll wait for my new dl365 + p800i [15:29] <\sh> it will replace these areca storage machines [15:30] \sh: our latest storage adventure is a dl185 running opensolaris and zfs. there the p800i had crap performance, so we replaced it with a plain sas card and suddenly it was very speedy. [15:30] <\sh> attached to it a msa60 + 12x 750GB sata drives... [15:36] <\sh> maswan: I just need the p800i because of the external storage... [15:49] Is there a way of removing apparmor? [15:53] Anybody using Xen on ubuntu-server? [16:14] MUST a bind authoritative dns server be a dns caching server inorder for local network computers to be succesfully served a webpage hosted within local network ? [16:15] or can a dns server NOT B a caching server? [16:51] So i locked myself out of root form my server (located far away) by overwriting my group permision when I moved myself to the ftp group i created, and this was before i knew the install didn't set the root pw. anyways is there a way to run apache web server from within my home directory without installing it ? [16:51] I can't access the config files for it where the are from my account === p0w4h` is now known as p0w4h [18:02] g'morning [18:18] should I ideally set the delay_warning_time of Postfix to something other than 0h? [18:19] AlexC_: Why do you want to change it? [18:19] 4 hours is sort of normal, if you need it at all. [18:19] I'm just wondering if I should actually need it, ideally [18:20] nodoubt: all BIND instances have a cache.. :) If the machine is authoritative for everything it serves, there is no need for it to allow recursion for anyone (which would mean it never got non-authoritative data into the cache) [18:20] default is 0h, which is off basically [18:20] AlexC_: it's zero by default because many people don't want the spamage from the mail server telling them that 4 hours has gone by [18:20] User can't do anything except complain to an admin. Decide if that's a bug or a feature. [18:21] =) I'll keep it off then [18:31] what is the difference between using the basic server install cd and jeos? [18:33] jeos is optimized for vms [18:37] if i created a lvm from two drives and the second fails, what will happen? [18:42] frith: you mean al Volume Group made of 2 PV? [18:43] yes [18:43] i haven't really used lvm [18:43] frith: the use it and trye that [18:43] you can try it with filebased volumes [18:44] saying that i don't know what is the best configuration to go with for my server [18:44] lvm is only volume manager, if you want protection you need to setupo lvm over RAID [18:45] i have one raid5 volume ~400gb and one mirrored pair 300gb [18:45] frith: well, we dont know the purpose of your server [18:45] me either :) [18:45] i think i want to use kvm [18:45] never used kvm only xen [18:46] kvm seems to be the way to go [18:46] i want to test out different stuff [18:47] i guess i will forget about the 300gb for now [18:48] i am not sure about its reliability [18:50] frith: why? [18:50] it is raid1, isnt it? [18:51] we use sofware raid1 in most of my and customers servers [18:58] jmedina, i rescued both disks from a blown disk array [18:58] the 300gb drives that is [18:58] i figured it out [18:58] raid pair for home and the raid 4 volume for vm's [18:59] now i need to think of a good domain name [19:05] New bug: #259827 in apache2 (main) "Please sync apache2 2.2.9-7 (main) from Debian unstable (main)." [Wishlist,Confirmed] https://launchpad.net/bugs/259827 [19:13] hello, is the pptp vpn server should be the gateway for others servers ? [19:18] hmm i thought opterons supported hardware virtualisation [19:20] not all [19:21] sigh [19:21] i guess its not big deal [19:21] only if you want to install windows [19:22] with paravirt works well with opterons at least with xen [19:22] we have about 30 vms running with paravirtualization in one server [19:24] wow [19:24] how many cores? [19:25] 2 dobule xeon [19:25] its funny, i can buy a 32gb server with 8 cores with 4tb for 2.5k [19:27] are they all pretty much idle? [19:28] 2.5K? [19:28] £2500 [19:29] oh haha [19:29] pretty cheap, I remember 2 years I spent 5k for 8 cores, 4GB and 1TB [19:29] frith, I thought you meant $2500 [19:29] i don't know how much systems are in the states [19:30] SSH question, is it possible to define either the use of public keys or passwords on a per account/user basis? E.g. a user designated "has key" should not be able to login with a passwords, temporary low-level accounts are allowed to login with a password. Non-designated accounts should not be able to connect over SSH. [19:32] i should ebay my current servers and replace them with one machine [19:32] Brucevdk: I think in the newests releases of openssh you can manage groups [19:33] ssh has been able to do groups for ages [19:33] allowgroups iirc [19:34] or group allow [19:35] That still wouldn't allow the finely grained permissions I was talking about though, would it? [19:35] http://www.openssh.com/txt/release-5.1 I think it is match groups and match addresses [19:35] frith: yeap they only use global configs, not per user [19:42] Think I'll just drop this idea for now then, when I use `AllowGroups ssh` does Deny all other groups? [19:43] not nessecarily, you would just need to be part of the ssh group to be able to login [19:43] ,via ssh [19:45] hello can any1 help me if possible pls? [19:46] i have a server with the internet working on it... eth0 with a static ip adress, i have there 4 more networks cards each one with another ip adress all up working, eth0 has: 191.191.2.1,eth2:192.168.0.2,eth3:192.168.0.3,eth4:192.168.0.4. [19:46] all are static ip's [19:47] after those 4 cards will be clients, how can i make the routes so i can give to the clients another class of IP's, like: 192.193.2.* [19:56] New bug: #259844 in samba (main) "Samba share permissions not migrated to Intrepid, not working?" [Undecided,New] https://launchpad.net/bugs/259844 [19:58] how can i get the current speed of my cpu's ? [19:59] cat /proc/cpuinfo [20:00] but does that give the speed if they have throttled down? [20:00] oh yes it does [20:01] right time to get power management working [20:02] powersaved ? is that the best thing to use? [20:17] ah there we go :) [20:22] leonel: Are you around. [20:23] jmedina, and others: thanks for the help, I've decided to go with AllowGroups + PasswordAuthentication for now. If needed I'll try and see if more finely grained permissions are possible. Thanks again [20:28] Brucevdk: you are welcome [20:40] kirkland: just commented un update-motd [20:40] kirkland: it has a minor issue i posted there everything you need [20:44] brb [20:52] is it wrong of me to arrange the disks in my rack so i can write words? [21:07] New bug: #256563 in openvpn (universe) "Arbitrary code execution in 2.1 prior to 2.1-rc9 (dup-of: 256621)" [Undecided,New] https://launchpad.net/bugs/256563 [21:28] ooo compiling a kernel is like watching pain dry [22:04] hmmm, not very impressive, only slowed down my server from 2.2ghz to 1.8 [22:04] surely it can go down to 1ghz or something [22:40] any idea why the cups web interface would freeze when selecting make/model of the printer? [22:41] slavik: have you check the logs? [22:41] Aug 20 17:37:55 upserv kernel: [11226.466619] audit(1219268273.160:13): type=1503 operation="inode_permission" requested_mask="a::" denied_mask="a::" name="/dev/tty" pid=8392 profile="/usr/sbin/cupsd" namespace="default" [22:41] I mean cups logs [22:42] E [20/Aug/2008:16:31:54 -0400] CUPS-Add-Modify-Printer: Unauthorized [22:42] no idea why [22:43] when I select 'HP' in the make list, the web interfaces takes a while to respond (it just comes up with a blank page after that) [22:43] E [20/Aug/2008:16:31:16 -0400] PID 6902 (/usr/lib/cups/cgi-bin/admin.cgi) crashed on signal 9! [22:43] that also seems relavant [22:44] * jmedina loves kde printer interface [22:44] * slavik loves gnome printer interface [22:44] but this is a server install (no X) [22:46] slavik: have you tried lpadmin? [22:47] hmm [22:47] I used it to add printers, but what I tell it the make is so that it uses the proper driver? [22:48] then again, the first time I tried to add a printer, it worked ... [22:48] should I try purging cups and reinstalling it? === hessml|away is now known as hessml|away|away [22:55] it stops working when I click 'Continue' on the page to select the manufacturer (or upload the ppd file) === hessml|away|away is now known as hessml|away