[00:00] <verbal> ScottK: that worked. thanks a lot
[00:00] <ScottK> verbal: No problem.  Glad I could help.
[00:00] <mathiaz> ScottK: https://blueprints.launchpad.net/ubuntu/+spec/server-lucid-sogo-calendar
[00:00] <ScottK> Thanks.
[00:58] <jmdz> hello anybody here???
[00:59] <KurtKraut> jmdz, you should make your question right away
[01:01] <jmdz> :)
[01:01] <jmdz> first is how many months does the karmic have in server support is it the standard 1yr
[01:04] <ScottK> jmdz: It's 18 months (which is the standard(
[01:05] <nyk2005> I fail to install mysql-server. I already installed and removed it before. Then purged it. But now the config script fails to start the mysql daemon and exits. When I start mysqld_safe from the command line, I get this error: "Can't start server: Bind on TCP/IP port: Address already in use". But no other mysqld is running and I checked with netstat -ap, and the port is free! What can it be?
[01:09] <smackkd> how do i make my eth1 renew dhcp without bringing down eth0
[01:10] <jmdz> ScottK thanks
[01:10] <kane_> smackkd: ifup eth1 ?
[01:10] <smackkd> aaah
[01:10] <smackkd> thanks
[01:11] <nyk2005> it should default to skip-networking, then it works! strange to set it to localhost... ?
[01:51] <Joshhhhhh> I just installed ubuntu Server 9.10, and have created a new user after the installation was done. My user doesn't have any sort of bash history or console autocomplete. Is there something I missed?
[01:52] <kane_> Joshhhhhh: is your shell bash? and do you have a home dir for the user?
[01:53] <Joshhhhhh> kane_, I did specify the home directory for the user, but I didn't think about the shell not being bash..
[01:53] <Joshhhhhh> let me check.
[01:54] <Joshhhhhh> kane_, my new user's shell is /bin/sh, hmm, I never had to specify that as an option before
[01:54] <kane_> type 'which bash'
[01:54] <kane_> then 'chsh'
[01:55] <kane_> it'll ask you for your new shell, and just give it the full path to the output of 'which bash'
[01:56] <Joshhhhhh> kane_, should I wait for my compile job to finish?
[01:56] <kane_> Joshhhhhh: just open a new terminal and do it there :)
[01:57] <Joshhhhhh> kane_, I would be. I was just wondering if it would affect my running compile jo
[01:57] <kane_> unless it's spawning a new user session, no
[01:58] <Joshhhhhh> kane_, woot, it works. Thanks a bunch
[01:58] <kane_> no worries Joshhhhhh
[01:58] <Joshhhhhh> I've got a lot of configuring to do, I'm migrating to ubuntu-server after 10 years of Gentoo :)
[01:58] <kane_> Joshhhhhh: welcome to the fold :)
[01:59] <Joshhhhhh> I'm having a hard time getting familiar with all the repository complicated mess
[01:59] <Joshhhhhh> In Gentoo, if it was in portage, it was available, one way or another
[02:00] <kane_> Joshhhhhh: it can be a bit tricky in the beginning, but there's actually a very simple table that illustrates it: http://en.wikipedia.org/wiki/Ubuntu_(operating_system)#Package_classification_and_support
[02:00] <Joshhhhhh> kane_, oh, awesome, thanks
[02:01] <kane_> Joshhhhhh: you'll notice a lot of software vendors also set up their own apt-repo for their product, which you can add to your apt sources.list and it becomes magically available
[02:02] <kane_> Joshhhhhh: rabbitmq for example: http://www.rabbitmq.com/debian.html#apt
[02:02] <Joshhhhhh> So what happens in the situation of: E: Couldn't find package liblame-dev
[02:03] <kane_> E: can't install till it can be found
[02:03] <Joshhhhhh> That means it's probably available, but on a diferent package source?
[02:03] <kane_> Joshhhhhh: the supported repos from canonical tend to not suffer from this, but 3rd party/unsupported stuff may depend on libraries not releases
[02:03] <kane_> it's the apt equivalent of 'no such file or directory' :)
[02:03] <Joshhhhhh> hmm
[02:04] <kane_> Joshhhhhh: it's not happened to me in years, but it's usually a packaging mistake when it does
[02:04] <Joshhhhhh> oh, it seems the package was modified to libmp3lame-dev
[02:04] <Joshhhhhh> I'm following old instructions, apparently
[02:05] <kane_> that's certainly possible; you'll often find that if 'libfoo-bar' got renamed or split out, a place holder package is put in place that pulls in the new version
[02:05] <Joshhhhhh> oh, I think I saw that earlier. I can't quote the log entry, but....
[04:00] <w3rd_> hey guys, Im running ubuntu server and i have a LAMP server running
[04:00] <w3rd_> i have installed some third party application for monitoring my ups
[04:00] <w3rd_> after i upgraded i ran into issues
[04:01] <w3rd_> the webserver for the admin panel on my ups software is clobbering all httpd access
[04:01] <w3rd_> so i cant seem to access my files for my site
[04:02] <russlar> w3rd_: do you need the web panel on the UPS software?
[04:02] <w3rd_> no
[04:02] <russlar> ok
[04:02] <w3rd_> i would like to remove the software to access the files on my site
[04:02] <w3rd_> i have tried to change web root via httpd.conf
[04:03] <russlar> remove the software completely, or just turn off the web portion?
[04:03] <w3rd_> and it will default to the path for the software running the admin web services for the ups
[04:03] <russlar> editing httpd.conf probably won't help
[04:03] <russlar> the ups app is hanging on to port 80
[04:03] <w3rd_> well when i disable the service for the ups web admin it clobbers httpd completely and disables all functions
[04:03] <ScottK> w3rd_: You should probably talk to the people you got the 3rd party software from.
[04:03] <w3rd_> that is what i was thinking
[04:03] <w3rd_> that is where im at with this whole thing
[04:04] <bogeyd6> you need to change the port the ups app is listening on
[04:04] <w3rd_> we have tried changing the port on the software to allow for attempting access elsewhere
[04:05] <w3rd_> but i still am not able to access the files in web root
[04:05] <w3rd_> i had some issues starting httpd after the upgrade
[04:06] <w3rd_> i guess the upgrade from apache 2, to 2.2 has some issues with naming conventions
[04:06] <w3rd_> but i was able to get the darn thing to start finally once i modified the modules...
[04:07] <w3rd_> but does that make sense at all that i cant access web root, could it be an issue with the way that i modified the configuration file? that doesnt seem to hold if i am able to access port 80 traffic for the ups software, right??
[04:49] <some_name> hey guys, sorry about that.. had to bounce earlier
[04:49] <some_name> basically i was whining about the issue with the UPS software
[04:51] <some_name> my server was running a webserver and running a client application that utilizes webservices and after i upgraded i was no longer able to access the content in webroot
[04:52] <some_name> im only able to see the webserver content for the admin panel of my ups client app
[04:53] <some_name> i was hoping that if i removed the application this would allow the httpd to free up..
[04:54] <some_name> is there way to check for what services utilize port 80 and which are first to be accessed if there is some sort of priorirty established for web services?
[06:13] <android6011> I want to use this for a home server. but for energy reasons i want it to sleep after an hour of inactivity. if another computer tries to access it via network (http,ssh etc) will it auto wake up or does it depend on bios?
[06:15] <twb> android6011: that's dependent on hardware
[06:15] <android6011> ok
[08:18] <porkpie> i guy's I have an urgent problem....    I issue the command reboot or shutdown -r now but the server doesn't respond
[08:18] <porkpie> It's like something is stopping the process from running
[08:18] <porkpie> It's a production server as well .....
[08:19] <\sh> moins
[08:19] <porkpie> are there any other commands I can use to force a reboot
[08:21] <porkpie> \sh: Morin
[08:21] <porkpie> Morning even
[08:22] <porkpie> Can you help with this problem please ....  I don't have access to the physical server to reboot it
[08:22] <twb> What was that container (a.k.a. zone/jail) system that's in recent mainline kernels?
[08:23] <porkpie> what process would stop a reboot or shutdown command
[08:23] <\sh> porkpie, can you paste the output after you issued the "reboot" command?
[08:23] <twb> Ah, LXC
[08:24] <porkpie> \sh: that is just it .... no output at all
[08:24] <\sh> porkpie, not even anything in /var/log/messages or /var/log/syslog?
[08:26] <porkpie> \sh: http://pastebin.com/m11ebe484
[08:28] <qman__> porkpie, seems to me like it IS shutting down, but something is holding it up
[08:28] <qman__> possibly a program that won't let go of an IO operation, or maybe failing hardware?
[08:28] <porkpie> qman__: it a virtual system
[08:29] <porkpie> qman__: vmware
[08:29] <qman__> when are those messages in relation to when you ran reboot?
[08:30] <\sh> porkpie, access to the vmware console? do it the hard way...I can't see anything related to your reboot in our paste
[08:30] <porkpie> \sh: I am just waiting for the reboot now  ....
[08:30] <porkpie> \sh: sent a request
[08:31] <cemc> can I raid1-ize an existing partition on a disk without losing data?
[08:33] <qman__> cemc, not that I'm aware of, raid requires a different partition type, then puts your filesystem on top of the raid
[08:33] <qman__> it might be possible but would require some trickiness
[08:35] <qman__> cemc, you might be able to build half a raid1 on the other disk, copy the data, then repartition the original disk and join it to the array
[08:36] <qman__> of course I wouldn't try such a thing without backups
[08:39] <cemc> qman__: mhm
[08:53] <ESEDU> destination "printer1" is not accepting jobs
[08:53] <ESEDU> whats wrong
[08:53] <ESEDU> It is a HP Laser Jet 1320 printer
[08:54] <cemc> ESEDU: it's a network printer, right?
[08:54] <ESEDU> right
[08:55] <cemc> is it on? :) do you have ping to it?
[08:55] <ESEDU> yes its on
[08:55] <ESEDU> but its not on network, its connected usb
[08:57] <cemc> I see. we have one here and last time I had some similar error it was because of the network. on USB i don't know
[08:57] <ESEDU> how do you add a driver to it
[08:57] <ESEDU> u know
[08:57] <ESEDU> like lpadmin -p printer1 -v /dev/usb/lp0 -m "?????"
[08:58] <ESEDU> the -m is for  model, and it always says unable to copy ppd file, when i try to put something there
[08:59] <ESEDU> I know its someting so simple i just dont get it
[09:16] <cemc> hm, I have cups installed, I think it went easy with that, it has some web interface, or something
[09:17] <cemc> HP LaserJet 1320 hpijs, 3.9.8 - this is what I have at 'Make and Model' (in gnome System / Administration / Printing)
[09:18] <cemc> (it's a Karmic desktop...)
[09:27] <ESEDU> Im working a command line, no GUI :/
[09:35] <ESEDU> how do Install driver for my printer
[09:37] <ESEDU> over 8 hours, ive been trying to get this work
[09:37] <ESEDU> no progress
[09:37] <ESEDU> how hard can it be to install a printer
[09:37] <ESEDU> apparently very hard.........
[09:46]  * \sh doesn't use printers../me is using PDF and email ;)
[09:57] <ESEDU> -.-
[10:01] <acalvo> ESEDU: follow the how-to for CUPS
[10:01] <acalvo> does anyone here use Nagios?
[10:01] <ESEDU> done that
[10:01] <ESEDU> there is no lpadmin instruction
[10:02] <acalvo> ESEDU: so? use the web administration panel to add it to your CUPS
[10:02] <ESEDU> I dont have a GUI
[10:04] <acalvo> you have already have a WEB-GUI
[10:04] <acalvo> localhost:600
[10:04] <ESEDU> nothing comes up with that command
[10:05] <acalvo> sorry, localhost:631
[10:06] <ESEDU> so i just write localhost:631 on the command line
[10:07] <acalvo> it's a WEB-GUI
[10:07] <ESEDU> what is
[10:07] <acalvo> you have to use an internet browser, such as firefox
[10:07] <ESEDU> how do i open it
[10:07] <ESEDU> but i only have command line
[10:08] <acalvo> well, you can use lynx then
[10:08] <acalvo> or -- even better -- use a browser in a machine that is on the same lan as the server
[10:09] <ESEDU> can u be more specific?
[10:09] <ESEDU> how
[10:10] <ESEDU> I got it open on lynx
[10:14] <acalvo> ESEDU: CUPS provides a web interface to manage it. Search the main page of CUPS for more information (or the tutorial in help.ubuntu.com)
[10:35] <Omahn> Does anyone know of a guide to setting up Linux Virtual Server on Ubuntu?
[10:41] <ESEDU> what is the windows equivelent of cups
[10:43] <atomic_1> the Print Spooler service ? :)
[11:04] <johe|work> has anyone mentioned nagios problems with 9.10. I have an problem with duplicate entries of checkcommands.cfg. but grep command shows only one definition in one file
[11:06] <atomic_1> did you grep the entire config folder
[11:26] <johe|work> atomic_1, yes i did
[11:43] <incorrect> can anyone suggest a virtual host provider offering 9.10 in japan?
[11:44] <incorrect> hongkong maybe
[12:06] <ESEDU> how do I get files off my memory stick
[12:06] <ESEDU> form the command line
[12:07] <ESEDU> i put my memory stick in the usb and in /dev there is now a sdb
[12:10] <ESEDU> anyone?
[12:12] <twb> ESEDU: mount /dev/sdb /mnt && cp -a /mnt /root/
[12:12] <twb> Or some variation on that theme
[12:45] <yoshi765> DAMN FREENODE
[12:53] <ESEDU> after specifing the printers on the server, how can a client use it
[12:56] <len> hi guys, can any1 help me? I got 2 computers sharing files using NFS but something is wrong, I can't list or copy files between those computers. The ls /mnt/computer1 is enough to make my console freeze
[12:57] <len> is there any log I could check from NFS?
[12:57] <mok0> len, check /var/log/syslog for clues
[13:01] <acalvo> I need a regular user to be able to manage a service (slapd), however, setting it on the sudoers file cause to fail when trying to load the configuration for that service, because it's owned by another user
[13:01] <acalvo> how to solve that?
[13:01] <len> mok0: seems that my remote server isnt responding. But I can ping it
[13:03] <len> acalvo: have u tried to put him in a group taht can manage that service?
[13:03] <len> taht*
[13:03] <len> that**
[13:04] <len> mok0: how can I get more info about this issue?
[13:04] <mok0> len, have you installed the nfs-server?
[13:05] <len> yes, It were perfect before the shutdown I needed to do on friday
[13:05] <mok0> len, are you cross-mounting between macines A and B?
[13:06] <len> mok0: nope. I had no entry on /etc/exports the computer2 till today.
[13:07] <mok0> len, what is the content of your /etc/exports on the server?
[13:08] <len> mok0: /fabrica <my computer2 ip>(ro,sync)
[13:08] <acalvo> len: is there any other solution that does not compromise security?
[13:09] <mok0> len, as root, create a directory somewhere, and try to mount computer1:/fabrica on it
[13:09] <len> yes I have this dir
[13:10] <mok0> len, mount kshanmu1@utmem.edu this_dir
[13:10] <mok0> huh?
[13:10] <mok0> len mount computer1:/fabrica this_dir
[13:10] <mok0> paste-buffer pollution :-)
[13:11] <mok0> Karmic upgrade broke my copy-paste using mouse-buttons
[13:12] <len> mok0: I already have a dir mounted, but I cant list it right now... Do you want me to try to mount it again in another directory?
[13:12] <mok0> len, no the server is hung
[13:13] <mok0> len, umount -l /the_hung_dir
[13:13] <mok0> len, then shutdown client and reboot server
[13:14] <len> mok0: I did that before comming here
[13:14] <mok0> len, ok, on server, what does exportfs say?
[13:15] <len> 1 sec, I need to kill some ssh. I tryed to use tab complete on mount dir and console froze again
[13:15] <len> I think Im gonna restart it again
[13:16] <mok0> hehe yeah, NFS is one of the few things that makes rebooting necessary...
[13:17] <len> client is rebooting now
[13:17] <nyk2005> I installed postgresql, but now /etc/init.d/postgresql-8.4 start doesn't do a thing. The executable seems to be missing.
[13:18] <len> My NFS server (computer1) exports: /fabrica 192.168.13.253(ro,sync)
[13:18] <len> and I have /fabrica dir
[13:19] <mok0> len, ok
[13:19] <mok0> len 2 secs
[13:20] <mok0> len, is lockd running?
[13:20] <len> on client?
[13:20] <mok0> len, on server
[13:21] <len> seems so. I got kblockd/0 til /3 and lockd
[13:21] <mok0> len, does /fabrica function normally on server?
[13:23] <len> yes, people are accessing it now through samba
[13:23] <mok0> ok
[13:23] <mok0> len, try mouting machine1:/fabrica /mnt/xxxx (manually)
[13:24] <mok0> len, I mean on the client
[13:24] <len> mok0: yeah I figured it out :)
[13:24] <mok0> len, good :)
[13:25] <mok0> len, look for error messages in the log. It
[13:25] <mok0> It is often permission problems
[13:25] <len> its mounted
[13:26] <len> mok0: may my bkp script is breaking nfs ?
[13:26] <len> im going to try to move some files around just like my script
[13:26] <mok0> len, bkp?
[13:27] <len> back up
[13:27] <mok0> ah
[13:27] <mok0> len, do you use autofs?
[13:27] <len> mok0: no, I mount the dir on script, then I umount it on the end
[13:27] <mok0> len, ah.
[13:28] <mok0> len, for backup, I suggest you use rsync
[13:28] <mok0> len, look at "rsnapshot"
[13:28] <mok0> len, then you can avoid using NFS, and rsync is more efficient
[13:29] <len> I had plan of using rsync
[13:29] <mok0> :)
[13:29] <len> but I didnt had the time to read how to configure
[13:29] <mok0> len, rsnapshot is very quick to set up
[13:30] <len> Im not a linux expert, and I had to make some services to work the fastest way possible, so I did the way I know
[13:30] <mok0> len, yeah, that's the way most of us have learned everything we know :-)
[13:31] <len> mok0: BRB.
[13:32] <mok0> len check out http://rsnapshot.org/
[13:33] <mok0> len, you can then create a system that lets users see their own backup directories on the server, using autofs. It can work almost like "Time Machine" on the Mac.
[13:35] <len> mok0: back
[13:35] <len> mok0: what happens if a file get corruped on server?
[13:35] <len> will I be able to recover it?
[13:36] <mok0> len, uhm, that depends
[13:36] <mok0> len, you mean with rsync?
[13:36] <len> yes
[13:37] <len> I mean, the file may get corrupted or deleted and my fear is that he syncs that too and I lose some important file
[13:37] <mok0> len, if the file content has changed, then yes
[13:37] <mok0> len, rsnapshot keeps N days of backups
[13:37] <mok0> len, we use 40 here
[13:38] <len> mok0: cool I'll run for rsnapshot
[13:38] <mok0> len, it uses hard links to save space
[13:39] <len> sorry about the newbie quest: But what is hard link?
[13:39] <mok0> len, it's a directory entry that points to the same data area as the "mother" entry
[13:40] <mok0> len, however, if you remove the "mother" file, you do not delete the file
[13:40] <mok0> len, that's different with a symbolic link
[13:41] <len> mok0: I got it. :)
[13:42] <mok0> len, the "link number" is shown by ls -l, as the number after the -rw-rw-rw- thing
[13:43] <mok0> Unix is sooo cool :-)
[13:44] <len> I really like it, but I have a lot of trouble with some hardware, like my sound. And still have lot to learn yet
[13:44] <len> and games.... I miss playing so much :P
[13:44] <mok0> sound... O_o
[13:45] <mok0>  len, get an xbox :-)
[13:45] <mok0> I prefer not to have games on my computers :-)
[13:46] <len> I had a problem with sound too. I bougth that Sound System With USB, but seems that only work with musicmatch... kinda annoying!
[13:46] <len> mok0: here in Brazil we get xbox or ps3 3x the original price
[13:46] <mok0> yikes
[13:46] <speedo> hello, is someone using boot from iscsci i iBFT there?
[13:47] <mok0> len, the only problem I
[13:47] <mok0> 've had with sound is that the master volume tends to be muted when the machine has booted
[13:48] <len> for example if I try to import ps3. From taxes I'll have to pay ps3 price + 60% of its value because of importing + anothers taxes + the profit from who is vending
[13:48] <mok0> Haven't had time to figure out what to do about it, just  a minor nuisance since I rarely reboot
[13:49] <mok0> len, take it up with G-12 or whatever :-)
[13:49] <mok0> len, I've the impression that Brazil taxes foreign goods quite heavily. I guy wrote the same about a Garmin GPS
[13:50] <mok0> s/I/1/
[13:51] <len> mok0: it's heavy. Very often we see people crossing the board or entering here without saying that is bringing something inside the country so it wont have to pay
[13:51] <len> I got my note that wax =X
[13:51] <len> note = notebook
[13:51] <mok0> :-X
[13:53] <mok0> ... uhm, I can understand why you want to play games on your computer then :-)
[13:53] <len> hahaha, true
[13:54] <len> hey, Same problem without executing the script
[13:54] <nyk2005> I installed postgresql, but now /etc/init.d/postgresql-8.4 start doesn't do a thing. The executable seems to be missing.
[13:54] <mok0> nyk2005: check if the package is intact
[13:58] <len> mok0: no deal with nfs. what's dpkg cmd to check if it's ok?
[13:58] <mok0> debsums
[14:00] <mok0> or rather, debsums postgresql-8.4
[14:00] <len> mok0: no other command? I dont have it installed and no internet connection on it
[14:00] <mok0> len, I don't understand...
[14:01] <nyk2005> mok0: how?
[14:01] <mok0> nyk2005: ^
[14:01] <mok0> where "^" means "see scrollback"
[14:02] <nyk2005> ok!
[14:02] <len> mok0: to check if my nfs pkg is ok. I remember something with dpkg -<param to check if its ok> <pkg name>
[14:02] <mok0> len, don't know that one
[14:03] <mok0> len "man dpkg" ?? :)
[14:03] <len> yeah... lazy mode off
[14:03] <len> it was on XD
[14:04] <nyk2005> mok0: it's all there, but doesn't include any binaries.
[14:04] <mok0> len, if you don't have internet, how can you be on IRC ? :-P
[14:04] <len> Im not using the server on this computer
[14:05] <mok0> nyk2005: postgresql-8.4 contains the binary "postmaster" which is what you need
[14:05] <len> sound strange... I meant Im using 3 computers, and the only one with internet is this one with IRC
[14:05] <len> the server/client linux doessnt have conenction
[14:05] <nyk2005> mok0: not in that package here...
[14:06] <mok0> nyk2005: what distro are you using?
[14:07] <nyk2005> mok0: newest ubuntu
[14:07] <nyk2005> mok0: server
[14:07] <mok0> nyk2005: what does dpkg -l postgresql-8.4 say?
[14:08] <nyk2005> mok0: installed
[14:08] <mok0> nyk2005: what does debsums postgresql-8.4 say?
[14:09] <nyk2005> mok0: it lists files in /usr/lib and /usr/share
[14:09] <nyk2005> mok0: and they're OK
[14:09] <mok0> nyk2005: did you upgrade from jaunty or earlier?
[14:09] <nyk2005> mok0: no, just installed yesterday
[14:12] <mok0> nyk2005: what does sudo /etc/init.d/postgresql-8.4 start say?
[14:12] <nyk2005> mok0: nothing.
[14:12] <mok0> huh?
[14:12] <nyk2005> mok0: and I checked the initscript, the exec it tried to run doesnt exist
[14:13] <nyk2005> mok0: do_ctl_all doesn't exist
[14:13] <mok0> nyk2005: what about /usr/share/postgresql-common/init.d-functions
[14:15] <len> nyk2005: just a simple question: Did you installed using apt or did you donwload it from pg's site?
[14:16] <nyk2005> len: aptitude
[14:16] <nyk2005> mok0: this exists, but the do_ctl_all not
[14:16] <mok0> nyk2005: do_ctl_all is a function defined inside that file
[14:17] <mok0> nyk2005: that file should be source by /etc/init.d/post.... etc
[14:17] <mok0> sourced
[14:18] <acalvo> I need a regular user to be able to manage a service (slapd), however, setting it on the sudoers file cause to fail when trying to load the configuration for that service, because it's owned by another user
[14:19] <mok0> acalvo: perhaps you can make that regular user part of the same group
[14:19] <len> acalvo: why create a grp that can manage the servier and put him on that grp wont work?
[14:19] <mok0> len, beat you to it :-)
[14:19] <len> mok0: I gave this sugestion earlier, so I'm still wining :P
[14:20] <mok0> nyk2005: I also installed a fresh karmic server the other day, and postgresql works out of the box
[14:20] <acalvo> yes len, I know
[14:20] <acalvo> but the security hole it can open is too risky, I think
[14:20] <mok0> acalvo: what security hole is that?
[14:20] <len> alcavo, what's the diference of put him on sudoers?
[14:20] <acalvo> making an user part of a grup that has more privileges
[14:20] <len> same risky on my point of view
[14:21] <acalvo> in sudoers you can only access to the service
[14:21] <acalvo> not to the files/directories owned by the user that runs the command of the service
[14:21] <mok0> acalvo:  /etc/sudoers gives you quite granular control over priveledges
[14:22] <mok0> acalvo: you can also manually let the user own the files that he/she is supposed to edit
[14:22] <len> acalvo: and cant you create a new grp that have a restric permission to do what u need, and then u add the user to this grp?
[14:22] <acalvo> my goal is to let a user (which is another service) control another service
[14:23] <acalvo> anything else
[14:23] <acalvo> all files are owned by the user who runs the service
[14:23] <acalvo> not the user who would manage the service
[14:23] <mok0> acalvo: unless the process check ownership of it's controlfiles, anyone can own them
[14:23] <acalvo> that's the thing
[14:24] <len> now its too deep to me... Ill just read and learn :P
[14:24] <acalvo> in the config files of the service, it runs as another user, which controls the permissions of its configuration files
[14:24] <mok0> acalvo: what service are we talking about?
[14:24] <acalvo> sldapd
[14:24] <acalvo> slapd
[14:25] <nyk2005> mok0: strange...
[14:25] <mok0> acalvo: ah, I am not familiar with it
[14:25] <mok0> nyk2005: I'd try to dpkg --purge postgres and install it again
[14:26] <acalvo> mok0: most services run as another user (like mysql), so the configuration files are owned by that user
[14:26] <acalvo> it's a sanity check, I think
[14:26] <acalvo> so nobody else can run the service
[14:26] <acalvo> at least not with that configuration
[14:26] <mok0> acalvo: OK, I see
[14:27] <acalvo> if I put the user who has to have control to that service in the sudoers file
[14:27] <acalvo> it can lauch the binary
[14:27] <acalvo> but can't read the configuration files, because the binary is owned by him
[14:28] <acalvo> however, I don't see any other solution that what you both suggested
[14:29] <mok0> acalvo: It's worth trying
[14:29] <len> acalvo: another sugestion, maybe u can create some init's file that starts and stop the service, and give the control of this file to ur user. Inside this file you may try changing user to start the service....
[14:30] <len> mok0: seems I solved the problem with nfs
[14:30] <mok0> len, good!
[14:30] <len> mok0: the problem were the cable
[14:30] <nyk2005> mok0: didn't work...
[14:30] <acalvo> len: good point, but init files already do that
[14:31] <acalvo> I'll go straight and add the user to the group owner of the files
[14:32] <mok0> nyk2005: Hm, give us some more information to work on
[14:32] <len> acalvo: I read something about ACL, Access Control Lists you may put 2 or more grps to a file or group of files
[14:32] <len> with different permissions
[14:32] <acalvo> you're right, I've read something about it too
[14:33] <acalvo> but it would involve stopping a production server
[14:35] <nyk2005> mok0: config dirs are also empty..
[14:36] <mok0> nyk2005: what config dirs?
[14:36] <nyk2005> mok0: /etc/postgres*
[14:36] <len> acalvo: another solution, you may use RWXR-X--- and you can put the grp without modifying the configuration
[14:37] <mok0> nyk2005: uhm, you should have a directory "8.4" in there.
[14:37] <nyk2005> mok0: not there.
[14:38] <mok0> nyk2005: something is seriously wrong with your installation
[14:40] <nyk2005> mok0: yes, somehow!
[14:41] <nyk2005> mok0: it's also strange that the auto-partitioner only assigned 2gb to the root partition, but 12gb for each of the many tmpfs partitions.
[14:41] <nyk2005> mok0: already almost ran out of space when installing texlife
[14:41] <acalvo> len: it's worth a shot
[14:41] <mok0> nyk2005: yes, that's very odd, and wrong
[14:42] <mok0> nyk2005: how much space have you got on your disk?
[14:42] <nyk2005> mok0: maybe because the server has a very small primary hdd (only 40gb SD) and 24gb of ram, so it failed to partition correctly...
[14:43] <mok0> nyk2005: you want just 2 partitions, /boot (512Mb) and / (the rest)
[14:43] <nyk2005> mok0: argle.. :) why did it do that so wrongly?
[14:43] <mok0> nyk2005: I don't know... weird
[14:44] <nyk2005> mok0: where are the tmpfs defined? I didn't see them in fstab. So I can mount the other partions at /var or /tmp
[14:44] <mok0> nyk2005: never mind the tmpfs, they are in memory
[14:44] <nyk2005> mok0: yeah, I read so also, but why is then only 2gb available to root? hmm...
[14:45] <mok0> nyk2005: can you do a "df -h" ?
[14:45] <nyk2005> mok0: maybe a broken install try partitioned something away..
[14:45] <nyk2005> mok0: only shows 2gb for / and the tmpfs
[14:46] <nyk2005> and the raid
[14:46] <mok0> nyk2005: could be... I always make the partions manually, they are never right for me
[14:46] <_ruben> a large chunk is probably used for swap (based on memsize)
[14:46] <nyk2005> mok0: strange is: /dev/sda6 is / and there's no /boot
[14:46] <mok0> nyk2005: can you paste the line with / ?
[14:46] <nyk2005> /dev/sda6             2.3G  1.9G  370M  84% /
[14:47] <mok0> nyk2005: ah
[14:47] <nyk2005> it's very wrong, but still don't why 370megs shouldn't be enough for psql...
[14:47] <mok0> nyk2005: a tiny partion :-)
[14:47] <mok0> nyk2005: it might have gone over while installing
[14:48] <mok0> nyk2005: here's my df -h line:
[14:48] <mok0> /dev/sdd3              28G  1,3G   26G   5% /
[14:48] <nyk2005> mok0: true... because of the packages..
[14:48] <mok0> nyk2005: right
[14:48] <acalvo> will, it didn't work
[14:48] <nyk2005> so your's would also fit in 2gb.. :)
[14:48] <acalvo> I don't understand
[14:48] <acalvo> if I run the init script as root
[14:48] <acalvo> everything works
[14:49] <acalvo> however, the init script starts the process with another user/group
[14:49] <mok0> nyk2005: it would, but I am not sure I could install in 2 Gb... besides, you also need some space for postgresql's data
[14:49] <nyk2005> mok0: I put all data on a separate raid5
[14:49] <acalvo> so it should be able to read its configuration
[14:49] <mok0> nyk2005: ok
[14:50] <nyk2005> but you're right, I should probably reinstall :(
[14:50] <mok0> nyk2005: I think so... you'll be scrounging for disk space from now on
[14:50] <nyk2005> mok0: or at least mount that unused partitions as /var and then reinstall all packages. Can apt to that?
[14:50] <mok0> nyk2005: but what happened to the rest of that disk?
[14:52] <mok0> nyk2005: if you mount that partition and copy /var to it, then empty /var and mount the partition as /var...
[14:52] <nyk2005> mok0: must have been partitioned before and then only used free space. I installed over a remote control interface, which had bad connection and was slow and unstable..  so had to try more than once to get it installed. it was very annoying... tried 3 debian ISOs, none worked. only ubuntu for some strange reason, if I turned off acpi...
[14:52] <mok0> nyk2005: I think you need to reinstall :-)
[14:53] <mok0> nyk2005: it will be faster than doing all the other shit
[14:53] <mok0> acalvo: probably the binary runs suid to that user
[14:54] <nyk2005> mok0: or copy the whole root to the larger, unused space!
[14:54] <nyk2005> mok0: I try to fix it because I can reinstall from where I currently am, my upstream is to slow for virtual ISOs.
[14:54] <nyk2005> I mean can't.. :)
[14:54] <acalvo> maybe
[14:54] <mok0> nyk2005: yikes.
[14:54] <acalvo> quite confusing
[14:55] <mok0> indeed
[14:56] <mok0> nyk2005: but you will need to re-install every package I think. You can use dpkg --get-selections to save your current list of packages
[14:58] <nyk2005> mok0: cool, thanks! that will fix it!
[14:58] <mok0> nyk2005: good luck :_)
[14:59] <len> mok0: time to eat
[14:59] <len> thx for help
[14:59] <mok0> len, np
[14:59] <nyk2005> mok0: thanks! but why is /boot on root?
[15:00] <mok0> nyk2005: The installer normally puts it on its own little partition
[15:00] <mok0> nyk2005: In the olden days, the boot partition had to be the first one, I don't think it matters anymore
[15:01] <mok0> nyk2005: some sysadms like to unmount /boot when the system is up
[15:01] <nyk2005> mok0: ah yes, that what it's for...
[15:01] <nyk2005> mok0: so doesn't matter that much... except I have to be even more careful when moving the root partition...
[15:02] <mok0> nyk2005: it will be ... challenging :-)
[15:03] <mok0> nyk2005: if you are clever, you can join partitions without destroying what's on there
[15:04] <nyk2005> mok0: but at least R and mysql are running, which is what I'm currently using for work, so the system is not that broken.
[15:04] <mok0> nyk2005: heh, Linux is rather robust
[15:05] <mok0> nyk2005: things are hard when you don't have physical access to the hardware
[15:05] <nyk2005> mok0: yes, the tools for remote installing are very bad!
[15:06] <nyk2005> mok0: I mean except you have a very fast upstream..
[15:06] <mok0> I read something on the interwebs where a sysadm upgraded a running Debian system to Ubuntu :-)
[15:06] <nyk2005> mok0: What would be cool is an open netinstall server... doesn't this exists? The main problem are the virtual CDs (isos)...
[15:07] <nyk2005> mok0: well ok, that also seems tricky...
[15:07] <mok0> nyk2005: we have one that's used locally for upgrades etc.
[15:08] <nyk2005> mok0: can you also use it for new installs?
[15:08] <mok0> yes
[15:09] <nyk2005> cool
[15:09] <mok0> you set the new machine to boot from the net, and it fires up the installer
[15:10] <Zodling> greetings everyone
[15:10] <Zodling> anyone try useing iscsi install with a "Broadcom NetXtreme II 5709" ?
[15:21] <Fenix|work> Greetings and salutations
[15:24] <Fenix|work> I need some syslog-ng advice on setting up a central syslog server on an ubuntu-server box.  Can anyone assist?
[15:25] <Fenix|work> I'd like to have a web based log viewer, and an email notification system for serious errors that are logged.
[15:25] <Fenix|work> It also doesn't have to be syslog-ng, if someone has a better approach, but whatever the logger is needs to support syslog.
[15:31] <Boohbah> Fenix|work: awstats?
[15:32] <Boohbah> Fenix|work: web-based log viewer for all types of logs? how bout apache?/
[15:33] <Fenix|work> apache logs, or using apache server to browse the logs in directory list?
[15:35] <Fenix|work> I'm planning on creating a central logging server to log all of my ubuntu boxes, as well as using a syslog type service on all of my windows boxes to push their logs do the central server
[15:35] <Fenix|work> that's a helluva lot of logs so I'd like something that'll make my life a little easier and has a GUI interface to help parse through logs.
[15:39] <acalvo> Fenix|work: skunk
[15:39] <Fenix|work> ?
[15:40] <Fenix|work> splunk?
[15:40] <acalvo> Fenix|work: that
[15:40] <acalvo> nagios + splunk
[15:42] <_ruben> we use SEC to turn syslogs into scheduled mails
[15:42] <_ruben> filtered that is
[15:42] <Fenix|work> SEC?
[15:42] <_ruben> Simple Event Correlator
[15:42] <_ruben> !info sec
[15:44] <Fenix|work> hmm
[15:44] <Fenix|work> splunk I think will be costly for me...
[15:45] <Fenix|work> I think I'll be dropping more than the 'free' version will allow.
[15:48] <Fenix|work> Out of curiousity, what are the benefits of dumping logs into a database instead of a log file?
[15:49] <Fenix|work> and what sort of performance hit would be incurred?
[15:54] <_ruben> that'd depend on the db backend i guess .. and it'd allow for nice searching capabilities .. i wouldnt be a fan of it tho most likely
[15:58] <Fenix|work> _ruben, what do you think the performance hit would be... would by mysql on the logging server, not on a dedicated database server
[16:14] <_ruben> Fenix|work: i'd expect it to be dramatic compared to plain text files .. tho that's just a wild guess really
[16:17] <Bilge> The ubuntu documentation recommends running `passwd -l root` to lock the root account and force the use of sudo
[16:17] <Bilge> But this makes all root cron scripts fail
[16:17] <Bilge> CRON[2696]: User account has expired
[16:19] <jcastro> ttx, ping me when you have a few minutes, I'd like to show you how to schedule things in the summit system
[16:19] <Bilge> Isn't there a better way to do this
[16:34] <sivang> anybody knwo if KVM can be used on an Atom machine ?
[16:35] <sivang> anybody knows if KVM can be used on an atom machine ?
[16:44] <jpds> sivang: Do: egrep '(vmx|svm)' --color=always /proc/cpuinfo - and find out.
[16:44] <Fenix|work> acalvo, how do you use nagios and splunk?
[16:45] <jpds> sivang: https://help.ubuntu.com/community/KVM - has more information.
[16:45] <acalvo> Fenix|work: take a look at the nagios homepage
[16:47] <Fenix|work> ok, so nagios + splunk + rove and I may be a happy camper
[16:47] <acalvo> rove?
[16:47] <acalvo> what does rove?
[16:47] <Fenix|work> Rove is a company
[16:48] <Fenix|work> they have a mobile admin app
[16:48] <acalvo> oh
[16:48] <Fenix|work> and remote control of things from a mobile phone
[16:48] <Fenix|work> http://www.nagios.org/products/enterprisesolutions/rove/
[16:48] <Fenix|work> was looking at Rove for something else...
[16:49] <Fenix|work> they used to have another name... started with a K or something
[16:52] <sivang> right , so my system does not support virualization
[17:13] <roytech> hi
[17:15] <roytech> hi
[17:26] <Fenix|work> what's the default syslog dameon on ubuntu server?
[17:28] <roytech> hi guy's
[17:28] <roytech> im a newbe
[17:28] <roytech> and i have some basic q's
[17:28] <jpds> roytech: Ask away.
[17:29] <roytech> ill give you basic run down ok?
[17:29] <roytech> i wish to create a sever on a 100% linux
[17:29] <roytech> enviroment
[17:30] <roytech> to do the following
[17:30] <roytech> to be a fille server
[17:30] <roytech> and to be a user logon
[17:31] <Bilge> Cool story bro
[17:31] <roytech> and to enable a roaming profile for staff
[17:31] <roytech> what is the best way to do this
[17:31] <roytech> i have installed ldap
[17:31] <roytech> samba
[17:31] <roytech> and dns
[17:32] <Fenix|work> roytech, so the user accounts aren't hosted on the server
[17:33] <Bilge> Why is your grasp of English so shit when you're from England
[17:33] <roytech> i would like them to be not sure how to though
[17:33] <jussi01> !ohmy | Bilge
[17:33] <Fenix|work> you are using microsoft windows server for user accounts now?
[17:37] <Fenix|work> if you're planning on using Ubuntu server to host your user accounts and have samba as the PDC... then 'roaming profiles' is irrelevant.  You use either Samba or an NFS mount to mount their home directory on their local system and whenever they log in they have their home directory available to them.
[17:37] <Fenix|work> but a true roaming profile with their profile settings may be difficult.
[17:38] <Fenix|work> I'd check the samba and nfs guides and see what you find
[17:39] <Fenix|work> the Samba docs may have all the info you need to set up samba as a windows domain controller replacement.
[17:39] <Fenix|work> by the way, you'll also need winbind
[17:41]  * Fenix|work thinks that about covers it.
[17:52] <roytech> brb sorry have put kids to bed
[17:56] <bogeyd6> !activedirectory | roytech
[18:19] <roytech> i'm not using any windows machines
[18:19] <roytech> only ubuntu linux
[19:38] <roytech> hi
[19:39] <roytech> hi
[20:13] <filemro3b> hi every body
[20:29] <Mark21> Hello, I did setup Ubuntu 9.10 amd64 server on a system. After that I did connect multiple iscsi targets to it. Now I am looking for ways to use it (read: I need to know what device that I get with "multipath -ll" is what device in /dev/mapper/ (so I can put data on it)
[21:12] <SirStan> With Postfix; does virtual_map or aliases support a one to many relation on emails to recipients?
[21:15] <SirStan> nm clearly it does.
[21:41] <|rt|> I think I have a bug to report against samba in karmic but want to chase it down a bit more before writing it up....can someone tell me what file dpkg-reconfigure samba uses to make it's changes?
[21:41] <|rt|> right now if you install samba on a fresh karmic server and then run sudo dpkg-reconfigure samba it will error out with a sed error
[21:43] <|rt|> sed: -e expression #1, char 140: unknown option to `s'
[21:43] <|rt|> that's the error it gets
[22:10] <tonyyarusso> |rt|: My suspicion is debian/postinst
[22:10] <tonyyarusso> |rt|: (line 60/61)
[22:11] <|rt|> someone had already filed the bug
[22:18] <bogeyd6> http://www.theregister.co.uk/2009/11/12/microsoft_patents_sudo/
[22:52] <adurity> if I've got apache2 up and running, should I be able to `sudo netstat --ip -lp`and see the listening process
[22:53]  * Fenix|work grumbles
[22:55] <Fenix|work> I'm trying to install 9.1 and when trying to install the base system, I'm getting a debootstrap error of Failed to determine the codename for the release
[22:56] <Fenix|work> using a USB dvdrom
[22:57] <Fenix|work> anyone with any suggestions?
[23:03] <Fenix|work> hmmm... seems that the usb dvdrom became unmounted
[23:03] <Fenix|work> mounting /dev/sr1 /cdrom solved my problem
[23:52] <cj> mdz: looks like you were the last to touch https://wiki.ubuntu.com/KernelTeam/EC2; you have any idea how one might get involved in moving things forward?