[00:03] <SpamapS> mathiaz: ping
[00:14] <maek> SpamapS: thanks
[00:23] <Danawar2> #hunt
[00:23] <Danawar2> Sorry.
[02:28]  * smoser heads out for a long weekend
[02:29] <zul> mathiaz: dude happy canada day....stop working ;)
[03:28] <pictureaday> Anyone available to help me out (i'm having boot issues)
[03:28] <smoser> a happy canada day to all
[03:29] <pictureaday> :(
[05:26] <kman> Hello all.  Is this a good place to ask a question about Ubuntu login problems?
[05:29] <twb> That depends where the problem is
[05:35] <kman> twb so there is someone out there.  I am having a double login issue.  I forgot the password, went through the recovery console, reset the password and now it prompts for login id, then the password, password again, and then back to userid prompt.  I can't log in.
[05:36] <kman> I searched the web and found others complaining about this.  May be something about a bug in the gdm.  Not sure how to get around this.
[05:37] <twb> I can't help you with gdm
[05:37] <kman> was hoping some Ubuntu expert would be able to point me in the right direction.
[05:38] <twb> Can you reproduce the problem in the getty?
[05:38] <kman> is there a better IRC I should try for this question?
[05:39] <twb> #ubuntu, I suppose.
[05:42] <chrismsnz> kman: that doesn't sound right
[05:43] <chrismsnz> are you getting help in #ubuntu?
[05:45] <kman> HI there just switched to #ubuntu to see if anyone there can help.  Thanks for the suggestion.
[05:45] <chrismsnz> I can try give you a hand if you like
[05:45] <kman> sure.
[05:45] <chrismsnz> just let me know - I'm idling at the end of work on a friday _b
[05:46] <chrismsnz> ok
[05:46] <chrismsnz> so you forgot your password?
[05:46] <kman> What would you suggest?
[05:46] <chrismsnz> How did you reset your password? Did you go in via recovery/single user mode?
[05:48] <kman> I'll give a few details if you don't mind.  Am putting together a server for experimenting for a client.  Had a pile of parts - Mobo, CPU, RAm, hard drive etc.  So put it altogether booted the system and it comes up with the login prompt for Ubuntu.  No idea what the password is.
[05:49] <kman> Yes I went in through recovery, listed the users, set the password for the one I wanted.  then exit, resume normal startup and now it does the Id prompt followed by 2 password prompts. and then prompts for the id again.
[05:49] <chrismsnz> okay
[05:50] <chrismsnz> so is this in the graphical screen (GDM?) or the text screen (terminal/tty)?
[05:50] <kman> It is the text screen on this version.
[05:51] <chrismsnz> right - does it have the ubuntu version there?
[05:51] <kman> being new to Ubuntu and coming from the Windows world of GUI's I will want the graphical interface.
[05:52] <chrismsnz> sure
[05:52] <kman> I think its 9.04.  Restarting now to see for sure.
[05:52] <chrismsnz> ah ok, sure
[05:52] <kman> Yes 9.04 tty1 prompt
[05:53] <kman> There is no network connection if that makes any difference.
[05:53] <chrismsnz> shouldn't matter
[05:53] <chrismsnz> so you type in your username and it prompts you with "Password:"?
[05:54] <kman> Yes right now it says <login:
[05:54] <kman> I enter the user id and it prompts with Password:
[05:55] <chrismsnz> and when you enter the password?
[05:55] <kman> I enter the password I just set in the recovery mode and it prompts again for the Password:  then it goes back to the login: prompt.
[05:55] <chrismsnz> right
[05:56] <chrismsnz> well... that's not normal behaviour. If your password is incorrect it should ask you for your login again directly instead of your password again
[05:56] <chrismsnz> after a pause
[05:57] <kman> Almost no pause.   Nothing noticiable.
[05:57] <twb> There's little pause because he's authenticating against flat files on the local host
[05:57] <kman> If it is a pause it is at most 2 seconds.
[05:57] <twb> Most likely is that you've mistyped the new password
[05:58] <chrismsnz> kman: it might be worth going through recovery again - or a complete reinstall of a current version
[05:58] <twb> Get root again, reset the password with "passwd fred", then run "login" and try to log in.
[05:58] <kman> No I've retried a few time thinking that myself.
[05:58] <twb> kman: you might have mistyped it *when setting the password*
[05:58] <chrismsnz> twb: my systems always gave me a pause - i always assumed it was some sort of rate limiting feature
[05:59] <twb> chrismsnz: it might do that, too
[05:59] <kman> just going through recovery mod enow.
[06:00] <chrismsnz> also - which user did you reset the password for? it may be a system or disabled user and not allowed to log in at all
[06:01] <kman> I am at the root prompt.  ls /home to see the accounts.  I then select the kman one I expect to log in with.  passwd kman
[06:03] <kman> I get the Enter new UNIX password:  I enter kman to keep it simple.  then it prompts with Retype new UNIX password:  I retype kman.
[06:03] <chrismsnz> cool
[06:03] <chrismsnz> type "login" from the recovery prompt
[06:03] <kman> It then indicates passwd: password updated successfully.
[06:04] <kman> I just tried the login.  prompted twice and then says  "Segmentation fault"
[06:04] <chrismsnz> ooh
[06:05] <chrismsnz> might be looking at a hardware problem there chief, or a corrupt install
[06:06] <chrismsnz> like i said, I've never had a system prompt me twice for a password - and a segfault from the login/bash process is dodgy indeed
[06:06] <kman> I'm surfing on segmentation fault right now to see what is out there.
[06:06] <chrismsnz> haha - it just means the program crashed
[06:08] <chrismsnz> are you at a root prompt in recovery at the moment?
[06:08] <kman> If I installed the OS with a different mobo and moved the hard drive to a new box would this cause the problem?  I bet it's having problems talking to the hardware.
[06:09] <chrismsnz> possibly - if it was problems with the hardware you'd probably notice them earlier in the bootup sequence
[06:09] <chrismsnz> do: grep kman /etc/passwd
[06:10] <chrismsnz> and tell me what it says after the last ":" on the line
[06:11] <kman> kman:x:1000:1000:kman,,,:/home/kman:/bin/bash
[06:11] <chrismsnz> ok
[06:11] <chrismsnz> now type /bin/bash
[06:11] <chrismsnz> and see if it breaks for you
[06:12] <kman> no problem it just accepts it with same prompt after I type that.
[06:12] <chrismsnz> this might not work, but try: "su - kman"
[06:13] <kman> That seemed to work.  The prompt changed to the $.
[06:13] <chrismsnz> yep, it means you're a user now
[06:13] <chrismsnz> ok, type exit
[06:14] <chrismsnz> then reboot
[06:14] <chrismsnz> try logging in again normally
[06:16] <kman> I'm remembering now.  This was my experiment with a samba server to share files between windows and Ubuntu.  Mobo did not support Sata drives so I put it aside.  Didn't want to buy IDE drives.  Now I have a different mobo that supports the sata.
[06:17] <chrismsnz> it shouldn't really matter - the version of ubuntu is fairly recent and autodetects hardware on boot
[06:17] <chrismsnz> you might be better off with a clean install mate _b
[06:18] <kman> I'm beginning to agree.
[06:18] <chrismsnz> i take it you still can't log in?
[06:18] <maek_> sorry im confused. does preseed = kickstart?
[06:19] <kman> Just tried to restart normally and still getting the double prompt.
[06:19] <chrismsnz> maek_: pretty much
[06:19] <chrismsnz> maek_: preseeding can automate the install - partitions, packages etc...
[06:20] <maek_> chrismsnz: do you know of any good preseed resources that would show all the options and an example of two for said option. do you serve up a preseed file with the kernel option of ks=http://path/to/preseed ?
[06:20] <twb> preseed and kickstart are two different solutions to the same problem.
[06:20] <maek_> does kickstart work just as well or is it kind of sticked into ubuntu?
[06:20] <chrismsnz> kman: the only thing I can think of that may have happened is that the system was configured to use /dev/hd* as the filesystem, which may not exist using sata drives (they become /dev/sd*)
[06:21] <twb> Personally I'd be using preseed because it has been in d-i longer, and the file format isn't so fugly
[06:21] <chrismsnz> but that still doesn't explain everything.
[06:21] <kman> chris still can't log in.  I'll try a clean install.  That should eliminate some potential issues.
[06:21] <chrismsnz> if there's nothing of importance on the system, reinstall with current version and call me in the morning
[06:21] <maek_> twb: ok. I have a chance to start over. I just got the go ahead to move our dev and staging to ubuntu and id like to do everything the debian/ubuntu was as much as possible
[06:22] <chrismsnz> maek_: have you had a look at this? https://help.ubuntu.com/10.04/installation-guide/i386/appendix-preseed.html
[06:22] <twb> maek_: to invoke preseed uses different boot: parameters; see the appendix of the installation-guide
[06:22] <chrismsnz> it's a good starting point
[06:22] <maek_> chrismsnz: thanks
[06:22] <kman> Thanks Chris.  YOu down New Zealand way?
[06:22] <chrismsnz> yep
[06:22] <kman> I visited there for 1 week about 20 years ago.  Nice place.  Wish I'd stayed longer.
[06:23] <chrismsnz> it's awesome when the weather is good! :)
[06:23] <maek_> and flight of the concords
[06:23] <chrismsnz> hehe - gotta love those guys
[06:24] <chrismsnz> little known fact: they pitched their show to one of the networks here and were turned down before HBO
[06:24] <chrismsnz> bet they're kicking themselves now!
[06:24] <kman> Gotta go.  Thanks for the help.
[06:24] <maek_> yeah, I talked with this girl from nz said they were more famous here then in nz
[06:24] <chrismsnz> pleasure, kman
[06:25] <chrismsnz> maek_: yeah, i mean they're popular here, but they have a following in the US really
[06:25] <maek_> chrismsnz: yeah I saw them perform in vegas about 2 weeks ago
[06:25] <chrismsnz> ahh awesome!
[06:26] <chrismsnz> I wanna see some shows in vegas
[06:26] <twb> chrismsnz: you're not the same .nz chris that wrote an httpd in forth?
[06:26] <maek_> we have lots of shows
[06:26] <chrismsnz> haeelll no
[06:27] <chrismsnz> hey maek_ - have you heard of a show over there called "Scoundrels"?
[06:27] <maek_> if anyone knows what cobbler is for kickstart is there a similar program for preseed? it manage all the bits for you like setting up the pxe dirs and dhcpd stuff and generating the preseed from a template
[06:27] <twb> maek_: di-netboot-assistant or FAI are the things I've seen
[06:28] <twb> maek_: obviously for preseeding, you can simply examine the seeds of an installed system
[06:28] <maek_> chrismsnz: no just a chain of pubs called Scoundrels
[06:28] <chrismsnz> sweet
[06:28] <maek_> twb: thanks.
[06:29] <chrismsnz> When I used netboot/preseed I put it together myself... might checkout di-netboot-assistant
[06:30] <twb> Yeah, di-netboot-assistant is only SLIGHTLY more useful than annoying
[06:30] <twb> It's dead easy to use dnsmasq for PXE booting
[06:30] <chrismsnz> yeah, i had dnsmasq built into the router so that was pretty simple
[06:30] <twb> And now, you should be able to send the preseed file over TFTP, so you don't need HTTP
[06:31] <sponzor> what is wrong? http://pastebin.com/fY5AWXzS
[06:31] <chrismsnz> anyway i'm out of here
[06:31] <maek_> oh cool
[06:31] <twb> Meaning that you can just put the preseed files in /srv/tftp next to the netboot kernel/ramdisk
[06:31] <maek_> chrismsnz: thanks for the help
[06:31] <chrismsnz> bye
[06:31] <chrismsnz> np
[06:31] <twb> sponzor: Ubuntu now uses upstart as its init, so everything you know about SysV init goes out the window
[06:32] <twb> (Including policy-rc.d, the bastards.)
[06:32] <sponzor> damn i dont know notting about upstart :/
[06:32] <lifeless> twb: patches considered ?
[06:32] <twb> sponzor: go read the service(8) and restart(8) manpages, as that output instructed you to do.
[06:33] <twb> lifeless: the Debian upstart maintainer wasn't enthusiastic about it IIRC
[06:33] <lifeless> ah well
[06:33] <twb> lifeless: I don't interact with Ubuntu devs
[06:33] <sponzor> can i change it back to init.d?
[06:33] <twb> lifeless: what I'm doing at the moment is diverting /sbin/initctl and replacing it with a stub (basically /bin/false)
[06:34] <twb> I don't even know if that's working, really.  The postinst runs "start foo" and that tries to talk to some dbus-type daemon that isn't running inside the chroot
[06:35] <sponzor> becose service doesnt work like it should? http://pastebin.com/4ZLMXzmq
[06:35] <maek_> twb: where would the seeds from the install be?
[06:37] <sponzor> common this is confusing :/ http://pastebin.com/bxfVVbn9
[06:39] <twb> maek_: debconf --get-selections
[06:39] <twb> Er, debconf-get-selections, from debconf-utils
[06:40] <maek_> twb: thanks
[06:42] <scar> sponzor, did you try /etc/init.d/hostname restart?
[06:43] <sponzor> scar: that is the first thing i try :)
[06:44] <sponzor> is this a bug?
[06:46] <twb> Technically /etc/init.d/hostname restart SHOULD work
[06:46] <twb> It's a legacy compatibility layer
[06:46] <twb> sponzor: you cannot change back to init.d
[06:49] <sponzor> in my case it doesnt. i think i should downgrade to ubuntu 9...
[06:56] <twb> sponzor: what are you actually trying to do?
[06:56] <twb> Set the active hostname from /etc/hostname?
[07:00] <sponzor> yes
[07:00] <sponzor> the hostname is ok now i m yust confused with new feauteres :P
[07:09] <twb> Just cheat
[07:09] <twb> hostname $(cat /etc/hostname)
[07:24] <kai> hi folks
[07:29] <maek_> stupid question. coming from redhat I can do service inetd restart - how do I restart inetd
[07:36] <kai> maek_: /etc/init.d/inetd restart
[07:36] <kai> with sudo, of course
[07:36] <kai> if you use a recent ubuntu install it might complain about using a different set of commands, and tell you what to use now
[07:36] <kai> :)
[07:40] <maek_> kai: thanks, the reason Im confused is there is not init script for inetd in /etc/init.d
[07:40] <basso> Hello, can anyone help me getting an ftp server up and running?
[07:41] <maek_> I have /etc/inet.conf with my tftpd entry and /usr/sbin/inetd is running, just dont see how its getting started
[07:42] <kai> interesting, that's new :)
[07:42] <kai> you can tell I haven't used inetd in a while
[07:42] <kai> check out man update-inetd
[07:42] <maek_> :)
[07:42] <maek_> ok thanks
[07:42] <kai> that seems to be a tool to update the inetd.conf, and that also tells inetd to reload the config
[07:43] <maek_> yeah I was thinking just hup it but seemed like there would be a slicker way
[07:43] <maek_> thanks
[07:44] <kai> I'm sure this is somehow related to the new startup stuff
[07:44] <kai> but I haven't really tried to understand that
[08:09] <huats> morning
[08:22] <Mateo_> Hi everyone !
[08:23] <Mateo_> I'm trying to setup subdomain with virtualhosts but every subdomain redirects me to the folder of the domain...
[08:23] <Mateo_> this is my conf: http://pastebin.com/1u1wy21D
[08:26] <Jeeves_> Mateo_: Can you try this? http://pastebin.com/5GzHEANE
[08:30] <Mateo_> Jeeves_: now i have a 403 error even if my folder project is chmod 755...
[08:30] <Mateo_> Jeeves_: this is the stat of my index.html file: Access: (0755/-rwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root)
[08:32] <rahman> Hi, in our university we have some academic library services that only accessible with the ip addresses of the university. We want the academic stuff to use these services whereever they are, not only from the universit network. But as these services are private we need to do this with user authentication. So, can a proxy server (like squid)  used to achive this purpose? Or there is some other ways to do this?
[08:42] <Jeeves_> Mateo_: How's the permission of the upper directories?
[08:42] <Mateo_> Jeeves_: the same
[08:43] <Jeeves_>  Mateo_ What's in your errorlog?
[08:43] <Mateo_> i'm trying to chown the folder for www-data, but i don't remember the format... sudo chwon www-data:www-data my/folder ?
[08:46] <trapmax> y
[08:47] <Mateo_> Jeeves_: [Fri Jul 02 09:44:53 2010] [error] client denied by server configuration: /home/kunedev/projects/worldcup/index.html
[08:49] <Mateo_> Jeeves_: i dont' know how, but now it looks like it works ...
[08:51] <Mateo_> Jeeves_: thanks for your help !!!
[08:51] <cpf> mase_wk: Hi there :)
[08:52] <mase_wk> cpf: hey
[08:52] <cpf> So, what I need to know is how to "start" mysqld without the service/init.d commands...
[08:52] <cpf> So I might catch an error.
[08:52] <mase_wk> firstly can you dpkg -l | grep mysql
[08:52] <mase_wk> and pastebin the output
[08:54] <cpf> http://pastebin.com/MTcM1hTP
[08:55] <mase_wk> k looking
[08:56] <cpf> Trying to figure out some things in the logs. Nothing important yet...
[08:56] <mase_wk> hmm ok
[08:56] <mase_wk> so have a look in  the init script, there will be a mysqld binary somewhere, it may even be in your PATH
[08:56] <mase_wk> run that , i think there is a debugging more
[08:56] <mase_wk> mode*
[08:58] <mase_wk> i take it you just did an apt-get install mysql-server to install this ?
[08:58] <cpf> When I run it, it freezes... Haven't found any relevant debugging in man also.
[08:58] <cpf> Pretty much, it was installed by tasksel.
[08:59] <cpf> I could always try a reinstall...
[08:59] <mase_wk> just try running that
[08:59] <mase_wk> on the off chance something wasn't configured correctly
[09:00] <cpf> I'm trying to run the mysqld command, it's hanging...
[09:00] <mase_wk> not really sure why your having an issue to be honest
[09:00] <cpf> Oh, you mean reinstall? :D
[09:00] <cpf> Me neither.
[09:00] <cpf> It's weird in a whole, yesterday everything worked like a charm.
[09:00] <mase_wk> wait,  did you get prompted for a password during install ?
[09:00] <mase_wk> for the root user ?
[09:00] <cpf> Yes.
[09:01] <mase_wk> hmm ok
[09:01] <cpf> And I used it yesterday.
[09:01] <mase_wk> did the server crash at all ?
[09:01] <cpf> No.
[09:01] <cpf> But reinstall did fix it...
[09:01] <mase_wk> hmm
[09:01] <mase_wk> odd
[09:01] <cpf> Very.
[09:02] <cpf> I'm not even sure it overwrote my config files.
[09:02] <mase_wk> it shouldn't have
[09:02] <cpf> It didn't... Configs are as I set them.
[09:02] <cpf> Hmm, going to take a reboot, if it fails at that point, it's probably because of that.
[09:09] <cpf> Ok, so reboot didn't break the mySQL
[09:10] <cpf> Guess problem is solved. Although the problem wasn't found.
[09:10] <cpf> Thanks anyway :)
[10:40] <Roxyhart0> hi there, somebody has intalled gnarwl sucessfully?
[10:46] <oCean_> Roxyhart0: it's in the repos isn't it? (never used it myself)
[10:47] <Roxyhart0> it is for autoreply ...i am not sure if work fine in ubuntu
[10:47] <oCean_> Roxyhart0: don't know. It's in the repos, so installation should not be a problem
[10:48] <Roxyhart0> ok, thanks
[11:17] <KnightStalker> Hello
[11:17] <KnightStalker> I saw my postfix sending(receivng worked) not working so,I searched google and finded out to test host -t mx gmail.com for example,I saw it failing,any reasons?thats showed via postfix tail " Name service error for name=gmail.com type=MX: Host not found, try again"
[11:19] <soren> KnightStalker: Busted DNS setup, probably.
[11:19] <soren> KnightStalker: What's in
[11:19] <soren> /etc/resolv.conf?
[11:19] <KnightStalker> nameserver 192.168.1.1 ^_^
[11:20] <KnightStalker> is that incorrect?
[11:21] <soren> Depends.
[11:21] <kai> KnightStalker: depends, is 192.168.1.1 your name server's IP?
[11:21] <soren> Does 192.168.1.1 have a nameserver on it.
[11:22] <soren> and is ^_^ something you wrote here or is it in the file, too?
[11:22] <KnightStalker> lol thats something I wrote here oO
[11:23] <soren> KnightStalker: What does "host -t mx gmail.com 192.168.1.1
[11:23] <soren> " do?
[11:23] <KnightStalker> wait a sec
[11:23] <KnightStalker> same
[11:23] <KnightStalker> ;; connection timed out; no servers could be reached
[11:24] <soren> Either 192.168.1.1 isn't running a nameserver or you're firewall is too restrictive.
[11:24] <KnightStalker> actually how to find out what is running my nameserver?
[11:25] <soren> You ask your network administrator.
[11:25] <KnightStalker> well 192.168.1.1 points to my router
[11:26] <soren> They often run a caching dns server. Yours may or may not.
[11:28] <KnightStalker> does having this on hosts.deny have to do anything with this error?    ALL : 192.168.0.0/32 : deny
[11:40] <KnightStalker> Oh thanks for the tip,soren,I found it out via my router
[11:49] <Mateo_> Anyone used to work with ISPconfig please ?
[12:10] <KnightStalker> oh well,host -t mx now works but postfix still having issues
[12:10] <rahman> Hi, in our university we have some academic library services that only accessible with the ip addresses of the university. We want the academic stuff to use these services whereever they are, not only from the universit network. But as these services are private we need to do this with user authentication. So, can a proxy server (like squid)  used to achive this purpose? Or there is some other ways to do this?
[13:08] <zul> morning
[13:28] <zul> are you serious?
[13:29] <sommer> morning
[13:40] <_chris_> hej all
[13:41] <_chris_> when i use 'top' i can sort by %Mem by pressing shift+m , is there also an option to sort by %cpu ?
[13:43] <_chris_> ah ok just found out with shift+o i can select by what he should sort^
[14:13] <_ruben> _chris_: and shift-p for the (default) sorting by %cpu
[14:23] <mika__> hi, i'm setting up an ubuntu server machine which will be a web server to manage support tickets (using otrs)... what should i install on a server for it protection, maintain, tools, etc?
[14:24] <jpds> mika__: Apache with mod_ssl?
[14:25] <Guest92581> i did a "cp -v source dest" so I could see what it was doing... however it doesn't provide an average speed so that I can determine how long i'm goinig to have to wait... i didn't see an option in the cp man page... is their a means to determine that?
[14:25] <jpds> Guest92581: No.
[14:26] <jpds> Guest92581: But if you want that, why not use something like rsync?
[14:27] <Guest92581> i plan to once i have removed all the unessary files and learn to use rsync.. i didn't realize that rsync would provide me that for an initial copy.. is it as fast for the inital copy as cp?
[14:27] <mika__> jpds:  mmm...yes..what else? now i've just installed htop and uptimed.... and i'm installing  postfix (which config do you suggest to use considering that all the emails will be sent directly from the server without using external smtp: internet site or internet with smarthost?)
[14:28] <jpds> Guest92581: rsync -aPhv --delete-after sauce dest ?
[14:31] <Guest92581> jpds, guess i shoulda used rsync.. the cp has been running so I don't want to stop it now... or since rsync does incremental copies should i stop it and restart the copy with rsync so that i can determine the transfer speed?
[14:31] <jpds> Guest92581: rsync hashes everthing and continues where things left off.
[14:32] <lamont> mika__: the diff between those two is just relayhost being set, causing mail to  all be sent to $relayhost, instead of trying the ultimate destination directly
[14:32] <mika__> lamont: what do you suggest to use?
[14:32] <jpds> Guest92581: Or you could use something like io{stat,top} to see the disk activity (if the kernel supports it).
[14:33] <lamont> Guest92581: speedwise, about the same for the initial copy.  the other thing that rsync gives you is matching timestamps
[14:33] <lamont> mika__: depends on how the mail that host delivers to the net needs/wants to be sent
[14:33] <Guest92581> lamont, ah so the cp will touch every file?
[14:34] <lamont> if you want it all to go through some central mail hub, then with smarthost.  if there's no need or desire, internet site would tend to be better
[14:34] <lamont> Guest92581: I'm not aware of an option to fix the timestamps with cp.. then again, once one has rsync, one stops using cp
[14:35] <lamont> at least for large tree copies
[14:35] <mika__> lamont: the mail will be the ticket response generated from otrs, and they will be sent to all addresses (gmail, hotmail, etc)... having my host to be the ultimate destination would cause delivery problems?
[14:35] <thesheff17> does anyone use vmbuilder w/ a vmbuilder.parition file...I'm trying to just use the example from the documentation and it keeps bombing out.  http://paste.ubuntu.com/458315/
[14:35] <lamont> large tree syncs tend to be disk-bound, so any inefficiencies of the way rsync does the extra checking around the initial copy are generally moot
[14:37] <lamont> mika__: so you have an email that you want to deliver to gmail...  do you want this machine to directly connect to google's machines, or does it want to dump that task on some other machine in your infrastructure (the relayhost)?
[14:38] <mika__> lamont: this is the only machine i've in the infrastructure...so i think it would be deliver directly to gmail... right?
[14:38] <lamont> right.  therefore, you want 'internet site'
[14:39] <mika__> lamont: ok, thank you :)
[14:39] <mika__> lamont: do you think that would be delivery problems? i've already set the spf on the domain name adding the machine ip
[14:40] <lamont> depending on your isp, no
[14:42] <lamont> that is, if you have a commercial ISP account, they probably don't block outbound traffic to port 25.  if this is a home machine at the end of a consumer-ISP link, odds are good that you need 'internet with smarthost' and the ISP's mail server as the relayhost (because many ISPs block outbound port 25 from consumer links)
[14:42] <mika__> lamont: ok, thank you :) (it's a commercial isp)
[14:43] <mika__> lamont: do you have any other suggestion? like stuff to install, etc
[14:43] <lamont> I tend to avoid making such suggestions
[14:46] <mika__> lamont: ok..
[15:06] <Guest92581> i've got about 6 large directories that have within them many of the same files. Some of the files are only located in one subdirectory.  I need to consolidate all of the directories into one directory that contains only one copy of each of the files.  How may I accomplish this?
[15:13] <sgsax> there should be an option in rsync to accomplish this
[15:14] <steffan> Guest92581: remove the duplicates first using fdupes or similar
[15:14] <steffan> !info fdupes | Guest92581
[15:15] <sgsax> rsync -av --ignore-existing /source/dir/ /dest/dir
[15:16] <sgsax> I think something like that should work
[15:16] <sgsax> I forgot to add --progress in there
[15:17] <kirkland> hallyn: did you find https://wiki.ubuntu.com/SyncRequestProcess ?
[15:18] <kirkland> hallyn: get that filed, and i'll pull the trigger on the actual sync
[15:30] <kirkland> hallyn: seabios uploaded
[15:31] <kirkland> hallyn: what's the location of your qemu-kvm branch?
[15:48] <kirkland> hallyn: and qemu-kvm's multiboot fixes uploaded
[15:51] <failover> Hello, i setup 3 dns servers ( bind ), one as master and two as slave, the slave A update the zones when i change the serial on Master, the slave B just update zones if first i remove /var/cache/bind/* and them restart the bind daemon
[15:51] <failover> Any ideas ?
[15:53] <maek> whats the new startup stuff called? upstart? is there a place I can read how to use it, so I can figure out the slick way to restart inetd?
[15:57] <zul> upstart.ubuntu.com
[16:03] <hallyn> kirkland: thanks!
[16:03] <hallyn> kirkland: so since it turns out there was already a seabios bzr tree, will that automatically get updated?
[16:07] <kirkland> hallyn: yeah, i think so
[16:09] <hallyn> groofy
[16:09] <hallyn> groovy
[16:22] <hallyn> kirkland: sync request filed
[16:30] <kirkland> hallyn: bug number?
[16:30] <hallyn> kirkland: bug 601100
[16:47] <hallyn> jdstrand: are you considering merging 0.8.2 libvirt when it comes out (in 1-2 weeks)?
[16:51] <serverhorror> hi, just started with UEC. I downloaded images thru the web interface and then removed everything again with euca-deregister problem is now that i can't download the images from the store again (cause the webinterface still says the images are installed) and I can't find the button to push so that the webinterface knows that the images aren't actually available
[16:51] <hggdh> OK. This did not go as expected. Brazil played like <$@&%+>
[16:52] <serverhorror> oh and yup I know I could manually download and re-register the images. But I unfortunately specifically need the web interface part
[17:00] <mathiaz> ttx: yo!
[17:05] <ttx> mathiaz: yoyo!
[17:05] <mathiaz> ttx: catchup mumble?
[17:06] <ttx> sure, quick one
[17:07]  * ttx just realizes that mumble crashed a few hours ago
[17:24] <Delemas> I'm running slapd using a slapd.conf. It isn't logging where it should or anything at all. What else other than the init and config could be affecting slapd logging?
[17:25] <Delemas> I'm running 10.04 LTS and slapd 2.4.21-0ubuntu5.
[17:32] <failover> Hi, my bind setup is return a lot of errors like this:
[17:32] <failover> Jul  2 13:17:32 faraday named[2115]: error (network unreachable) resolving 'www.google.me/A/IN': 2001:500:4f::1#53
[17:33] <failover> Someone know why?
[17:35] <serverhorror> failover:  do you have ipv6 on that bind server?
[17:36] <failover> i don't think so
[17:38] <serverhorror> failover:  " resolving 'www.google.me/A/IN': 2001:500:4f::1#53" <- look up the IPv4 record of www.google.com on the server "2001:500:4f::1" (which is an ipv6 server). And if you don't have IPv6 that of course fails. Now the question is why is your bind thinking that this IPv6 server can answer your request...
[17:39] <failover> good question, just using basicly the default lucid conf
[17:40] <failover> oh
[17:41] <failover> this is a default behavior: listen-on-v6 { any; };
[17:44] <serverhorror> failover:  you can deactivate that _or_ get a tunnel from sixxs (if you can't get "native" IPv6 from your provider). And get yourself familiar with IPv6 (IMHO the better solution in the long run)
[17:50] <failover> i set listen-on-v6 from any to none, but still getting the errors...
[18:06] <pedrocr> I'm trying to install ubuntu server from a USB disk
[18:06] <pedrocr> it boots fine but then starts searching for a cdrom
[18:06] <pedrocr> instead of installing from the USB disk
[18:13] <Daviey> pedrocr, What version, and how did you put it on the usb stick?
[18:13] <pedrocr> Daviey, 10.04 and I used the GUI usb creator in my 10.04 desktop installation
[18:13] <pedrocr> gave it the server iso and it installed it fine
[18:14] <Daviey> pedrocr, Odd.. that is the way to do it.. There was a known bug in some earlier releases.. but that should work in lucid :/
[18:14] <pedrocr> it boots just fine but keeps trying to use the physical cdrom
[18:14] <Daviey> yeah.. that was a bug in previous versions
[18:14] <pedrocr> should I disconnect the physical cdrom device?
[18:14] <Daviey> no
[18:16] <pedrocr> any suggestions?
[18:16] <pedrocr> I'm using ubuntu-10.04-server-amd64.iso by the way
[18:16] <Daviey> pedrocr, When you get to the boot screen where it asks if you want "ubuntu server ; Ubuntu enterprise cloud" etc
[18:17] <pedrocr> yep
[18:17] <Daviey> try f6
[18:17] <Daviey> then press Esc
[18:17] <pedrocr> yes
[18:17] <Daviey> add "cdrom-detect/try-usb=true" to the end of the kernel line
[18:17] <pedrocr> it has that already
[18:17] <pedrocr> didn't work anyway
[18:18] <Daviey> hmm
[18:18] <Daviey> odd
[18:18] <pedrocr> I even tried to open a shell and mount the drive on /cdrom but it didn't work anyway
[18:18] <Daviey> pedrocr, Do you have another USB pendrive you could try?
[18:18] <pedrocr> sure
[18:19] <Daviey> pedrocr, Can you try that, and report back.. I might be not still be around; but try
[18:19] <pedrocr> I'm doing that right now
[18:19] <pedrocr> should I try the alternate image instead?
[18:23] <pedrocr> that worked
[18:23] <pedrocr> strange
[18:23] <pedrocr> it was able to mount the disk just fine
[18:25] <pedrocr> should I use swap in a server and how much?
[18:25] <pedrocr> I will have 4 disks in raid should I put a swap partition in each of them?
[18:48] <thesheff17> pedrocr...I only create 1 swamp file max 2GB now....you should control each application on how much RAM it uses....I have even heard of people disabling swap all together
[18:51] <maek> zul: thanks. re upstart
[19:05] <pedrocr> thesheff17, thanks
[19:05] <pedrocr> anyone know how to check how long a partition format will take in the installer?
[19:06] <pedrocr> I'm formatting a 4.5TB raid5+lvm partition and wondering how long it will take
[19:06] <pedrocr> 20 min have passed already
[19:06] <thesheff17> haha long
[19:06] <pedrocr> any idea?
[19:07] <thesheff17> I would give it at least an hour....not only does it have to format all the drives it is also starting to construct the raid5 in the background.....I haven't used it raid5 w/ lvm...will lvm wait till the raid5 build is complete....if that is the case it could be several hours.
[19:08] <pedrocr> I see
[19:08] <pedrocr> I'll probably leave it for tomorrow then
[19:09] <thesheff17> yea I would check back every hour or so
[19:10] <pedrocr> hah, done
[19:10] <pedrocr> 30 min apparently
[19:10] <thesheff17> nice
[19:10] <thesheff17> the raid is prob still constructing in the back ground. you can switch to another console and do cat /proc/mdadm
[19:10] <thesheff17> and that will tell you the state of the build....so if your machine feels sluggish this is prob why.
[19:11] <pedrocr> yep, md1 still going
[19:11] <pedrocr> the 5GB raid1 is done of course
[19:12] <pedrocr> finishes in 2241 minutes
[19:12] <pedrocr> :)
[19:12] <thesheff17> 37 hours :)
[19:12] <thesheff17> wow
[19:13] <pedrocr> it's probably overstated as the volume is being hit by the installer right now
[19:13] <thesheff17> true
[19:13] <thesheff17> yea once you have the os boot check it then
[19:14] <pedrocr> yep
[19:16] <thesheff17> hehe what are you doing with your new 4.5TB machine?
[19:16] <thesheff17> samba?
[19:21] <pedrocr> thesheff17, just backups mostly
[19:21] <pedrocr> got to run
[19:21] <thesheff17> nice...ttyl
[19:58] <peeps[work]> is there any easy way to share folders over samba?  I tried editing smb.conf, but this is a nightmare the man page is huyndreds of pages long
[19:59] <peeps[work]> when i try to view the workgroup from a windows client, i get an error that it cannot get access
[19:59] <qman__> peeps[work], sharing folders is easy, there is an example in the configuration file itself
[19:59] <thesheff17> yup
[19:59] <thesheff17> one sec when I get my smb.conf
[19:59] <qman__> the manual is huge because samba is a lot more than just a file sharing software
[20:00] <thesheff17> this is all I have in my smb.conf
[20:00] <thesheff17> [raid1]
[20:00] <thesheff17>        comments=shared
[20:00] <thesheff17>        path=/mnt/share1
[20:00] <thesheff17>        guest ok = yes
[20:00] <thesheff17>        browseable = yes
[20:00] <thesheff17>        writable = yes
[20:00] <thesheff17> sorry for pasting that
[20:00] <Pici> !paste
[20:01] <qman__> that will work, but it is not at all secure, just so you know
[20:01] <thesheff17> then use smbpasswd username
[20:01] <thesheff17> to set the password for the samba user you want to access the share
[20:01] <qman__> fine for your home network but if you're running in a production environment you're going to need to tighten it a bit
[20:01] <thesheff17> yes this is for my home network
[20:07] <thesheff17> erichammond: thanks for that response about vmbuilder and tmpfs..I got some help from the ubuntu-server mailing list maintainer of that project.
[20:07] <peeps[work]> ok i took out all the comments so that the file is legible. this is what i have right now.    http://paste.ubuntu.com/458438/
[20:07] <peeps[work]> i have a user named
[20:08] <peeps[work]> "share" and i want to share a folder under that home dir
[20:08] <peeps[work]> i just want to use the user's regular linux password for authentication, and be able to access this from windows client
[20:09] <thesheff17> I believe you just have to use smbpasswd to set the samba password for that user
[20:10] <thesheff17> I have always had to change the permission to that folder as well to get it work....chmod 777 /home/share/shared
[20:10] <thesheff17> restart samba and see if works
[20:10] <qman__> yeah
[20:11] <qman__> it synchronizes them by default, but you have to set it once
[20:11] <qman__> you don't need to make the folder 777, you just need to give the user access
[20:11] <qman__> either chown it or add the user to the group owner
[20:12] <peeps[work]> http://paste.ubuntu.com/458439/
[20:12] <peeps[work]> don't know how to use smbpasswd
[20:13] <thesheff17> I think it is in the samba tools package
[20:13] <thesheff17> apt-get install samba-tools
[20:13] <qman__> sudo smbpasswd -a share
[20:13] <qman__> and put in the password you want
[20:14] <ribo> is the ppa keyserver.ubuntu.com down? I can't seem to add any PPAs with add-apt-repository with: gpgkeys: HTTP fetch error 7: couldn't connect to host
[20:14] <peeps[work]> thesheff17, qman__ what i mean is, I tried to use it and got an error, see my paste
[20:14] <qman__> yes, you need the -a
[20:14] <thesheff17> oh yea
[20:14] <qman__> but only the first time
[20:14] <guntbert> ribo: you can use any other keyserver as well
[20:14] <peeps[work]> qman__, i did use -a.
[20:15] <qman__> well, that error is interesting
[20:15] <qman__> looks like a kernel panic
[20:15] <thesheff17> yea I have never seen something like that before
[20:15] <peeps[work]> do i need a different "passdb backend" or something, i just used the default
[20:15] <qman__> something is seriously wrong with your system, hardware or software
[20:15] <guntbert> ribo: for instance http://pgp.mit.edu/
[20:16] <qman__> that is not a normal error nor is it a configuration error
[20:16] <qman__> might be hard disk failure or corrupt binaries
[20:16] <peeps[work]> argh
[20:16] <peeps[work]> what is best way to check HDD integrity
[20:17] <ribo> guntbert: hrm add-apt-repository doesn't take any arguments but the ppa archive
[20:17] <qman__> fsck, maybe check the SMART information
[20:17] <peeps[work]> are you guys sure my settings  look sane?  http://paste.ubuntu.com/458438/
[20:17] <qman__> yes
[20:17] <qman__> that is not a problem with settings
[20:18] <peeps[work]> so what should I do now
[20:18] <qman__> the backend crashed in an unexpected way
[20:18] <guntbert> ribo: but you can download the key and add the repo to your sources in "the old style"
[20:18] <qman__> might try apt-get reinstall for samba
[20:18] <ribo> ah ok
[20:18] <qman__> err
[20:18] <qman__> sudo apt-get install samba --reinstall
[20:18] <qman__> I think
[20:18] <ribo> I did it the hard way, made a hosts entry for keyserver.ubuntu.com :P
[20:18] <peeps[work]> will a reinstall overwrite my settings
[20:19] <qman__> only if you purge the package first
[20:19] <qman__> a simple reinstall leaves the config files alone
[20:19] <peeps[work]> did a reinstall, same error
[20:21] <qman__> well, the error is most definitely caused by the binaries
[20:21] <qman__> so either you've got a corrupt file, a bad compile (bad package), or some other bug
[20:22] <qman__> could be caused by any number of things
[20:22] <qman__> a google search indicates that it's not likely a bug in the standard package
[20:22] <qman__> corrupt files are more likely, or if you compiled it yourself
[20:23] <peeps[work]> hmm, where do routine filesystem check logs go.  i have a headless server, so I never see when they run
[20:25] <peeps[work]> er,nm i think i see them in /var/log/fsck  not much to them
[20:26] <peeps[work]> ok, forcing another fsck...
[20:29] <peeps[work]> hrm, i did a "sudo shutdown -Fr now" but i don't think it checked anything
[20:30] <peeps[work]> er, i guess this version of shutdown doesn't have -F
[20:31] <guntbert> peeps[work]: Use: sudo touch /forcefsck
[20:31] <guntbert> then reboot
[20:31] <peeps[work]> ok
[20:36] <peeps[work]> if i had serial connection, could I connect while it was doing fsck?
[20:36] <peeps[work]> instead of ssh over ethernet?
[20:40] <qman__> yes
[20:41] <peeps[work]> ok i did a fsck.  i don't know how to view the results.  i see /var/log/checkfs and /var/log/checkroot files.  but neither of these were touched today
[20:41] <qman__> well, only if you configure it a certain way
[20:41] <peeps[work]> i'm sure it did some check because it took a long time before I could reconnect over ssh
[20:44] <peeps[work]> hmm, i found this launchpad entry.  https://bugs.launchpad.net/ubuntu/+source/samba/+bug/302605
[20:44] <peeps[work]> says removing secrets.tdb fixes it.  what is that file
[20:45] <qman__> that's the tdbsam password database
[20:45] <qman__> worth a shot
[20:45] <qman__> I dismissed it because that's quite old and an isolated incident, but it could be the same case for you
[20:46] <peeps[work]> this machine has gone through many distro upgrades, so maybe it got left with an old version of the file or somthing
[20:47] <peeps[work]> looks like it works now
[20:49] <peeps[work]> i will never understand how the windows share protocol can be so freaking slow on a LAN
[20:52] <peeps[work]> still, i would like to know how the fsck routine went, any comments on that issue?
[21:02] <peeps[work]> where'd everyone go :-(
[21:41] <goldins> hi I'm running apt-mirror so that I can run my own copies of the repositories and it downloads all the 32 bit and 64 bit packages but only makes the indexes for the 32 bit packages. how do I make it make the indexes for the 64 bit packages?
[21:45] <Italian_Plumber> is there any gee-whiz way to "make sure" that ubuntu server is using all four cores of a quad core processor?  Or does that even make sense?
[21:46] <goldins> Italian_Plumber: well how do you mean 'using' ?
[21:46] <thesheff17> goldins this is what I have in my mirror.list
[21:46] <thesheff17> http://paste.ubuntu.com/458482/
[21:46] <thesheff17> and that should work for 64 bit
[21:46] <goldins> Italian_Plumber: if you run some sort of hardcore process four times, check and see if top says that it's using 100% (or close to it) for each PID
[21:47] <goldins> that is four instances
[21:47] <thesheff17> it also has Hardy in there...edit accordantly
[21:47] <thesheff17> Italian_Plumber: you can do cat /proc/cpuinfo
[21:47] <thesheff17> and it will show you your processes...
[21:47] <thesheff17> most programs though are not multi threaded
[21:48] <Italian_Plumber> I want to know if the OS is ignoring some of the cores.  If they are idle.
[21:48] <thesheff17> as long as you see them in that output
[21:48] <thesheff17> it won't
[21:48] <Italian_Plumber> I moved from an Athlon XP 1800+ to a AMD Phenom 9460 and didn't see as much of a perfromance increase as I was expecting.
[21:48] <thesheff17> and top tells you allot
[21:48] <goldins> also numa-ctrl --hardware
[21:49] <thesheff17> most systems are not CPU bound
[21:49] <goldins> thesheff17: so for the 386 ones mine just says 'deb'...
[21:49] <Italian_Plumber> this is my top: http://pastebin.com/uDPBuweM
[21:50] <thesheff17> so this 96.0%id
[21:50] <thesheff17> is how idle your whole machine is
[21:50] <thesheff17> goldins I would change it to deb-i386 and deb-amd64
[21:50] <thesheff17> deb is prob just 32 bit
[21:51] <thesheff17> Italian_Plumber: so if you have 4 cores on that machine and you see that number at 75%...it means 1 core is 100% used and so on
[21:53] <qman__> Italian_Plumber, cat /proc/cpuinfo
[21:54] <qman__> if there are four processors, it's using them
[21:54] <qman__> I have never heard of ubuntu not using all cores of a multicore processor though
[21:55] <goldins> thesheff17: :-/ still doesn't work
[21:55] <qman__> I don't know what it's tuned to, I think maybe up to 64 by default
[21:55] <thesheff17> what is the error?
[21:55] <qman__> but the default kernel most definitely supports and uses quad cores
[21:56] <qman__> thesheff17, I gave up on apt-mirror and just used squid as an apt proxy
[21:56] <goldins> thesheff17: I mean, the index doesn't get generated so when I run apt-get update on my desktop I get a 404
[21:56] <qman__> it's easier to set up, only downloads what you need, and is totally automatic
[21:56] <goldins> this is my mirror.list
[21:57] <goldins> http://pastebin.com/n6p6QQxk
[21:57] <thesheff17> I would use your local mirror so us.archive.ubuntu.com
[21:57] <qman__> with some iptables trickery, you can set it up transparent
[21:58] <qman__> so there's no client configuration at all
[21:58] <goldins> thesheff17: I'm at NIH.gov, my connection to ANL.gov is much much faster
[21:58] <thesheff17> chances are they are only mirror a subset
[21:59] <goldins> I'm pretty sure they don't, I mean, why would they?
[21:59] <Italian_Plumber> here's the output to that command: http://pastebin.com/SVUHcV0h
[21:59] <thesheff17> bandwidth...not sure
[21:59] <goldins> they have 10Gs
[21:59] <Italian_Plumber> so yeah looks good
[21:59] <thesheff17> hehe
[22:00] <goldins> also they're on Internet2
[22:00] <thesheff17> well I can try to mirror off you mirror file and see what happens
[22:00] <goldins> we're also on Internet2
[22:00] <thesheff17> can you pastbin the exact error
[22:01] <goldins> I mean it's just this like 20 times (once for each repository):  W: Failed to fetch http://block/ubuntu/dists/lucid/main/binary-amd64/Packages.gz  404  Not Found
[22:01] <thesheff17> yea for some reason it isn't on the mirror :-/
[22:01] <goldins> that's obnoxious.
[22:01] <goldins> well I guess I can use us.
[22:01] <thesheff17> maybe they are only mirroring 32 bit?
[22:01] <guntbert> goldins: use another mirror?
[22:02] <goldins> I'm pretty sure I've used them on my desktop directly
[22:02] <goldins> but not certain
[22:03] <goldins> they're also one of the few mirrors that mirrors all of the ISOs as well
[22:03] <thesheff17> qman the thing a local mirror solves is that I update my mirror on the 1st of the month...roll all the patches to the dev environments...and then roll to prod when I know everything works....a proxy is nice if you can always have the latest and greatest but so many packages change so fast... I don't want my prod servers to have any problems.
[22:03] <thesheff17> if you take my mirror list and just do a find and replace and see if that helps at all....
[22:04] <goldins> well I'm trying US now
[22:04] <thesheff17> k
[22:07] <thesheff17> also maybe they removed karmic since 10.04 is out now
[22:09] <goldins> there is something wrong with nih's proxy that makes my connection to us.archive.ubuntu.com fluctuate wildly :-/
[22:10] <goldins> my inclination is to blame websense
[22:10] <thesheff17> hehe
[22:11] <thesheff17> once you get the whole archive though it will be quick to update everyday
[22:17] <thesheff17> goldins does it look like it is working?
[22:32] <goldins> thesheff17: it's certainly downloading a whole lot :-/
[22:32] <goldins> I'm not sure if it's redownloading everything as a matter of principle
[22:33] <thesheff17> it will eventually tell you how many GB it needs to download
[22:35] <ciao> Ciaooo
[22:35] <ciao> !list