[00:19] <fbc_> after installing php5-curl my server still won't execute curl statements in my PHP
[00:20] <fbc_> any ideas why?
[00:34] <Tellmarch> did you restart it? '^^
[00:37] <fbc_> Tellmarch, yeah
[00:37] <Tellmarch> maybe there is some module to enable with a2enmod? truth is i don't even know what is curl :p
[00:37] <fbc_> Tellmarch, hmm let me check
[00:38] <TheRealJeanLuc> Hey, I'm having a bit of an issue with Ubuntu 10.10 Server (32-bit). I'd like to install the ubuntu-desktop package, but when I run 'sudo apt-get install ubuntu-desktop' I'm told that the package could not be found. When I run 'sudo apt-get update' it takes  VERY long time, but ultimately only "Ign" and a couple of "Err" lines are listed
[00:38] <pmatulis> TheRealJeanLuc: pastebin your sources.list file
[00:38] <fbc_> TheRealJeanLuc,  does the server have an internet connection?
[00:39] <pmatulis> TheRealJeanLuc: and, yes, you need an internet connection
[00:39] <fbc_> TheRealJeanLuc,  if you ssh into the server and type "ping yahoo.com" do you get replies?
[00:40] <twb> Bah, you should test against example.net
[00:41] <fbc_> Tellmarch, doesn't look like there are any curl related modules to enable with a2enmod.
[00:41] <Tellmarch> sorry, don't know what can be missing then '^^
[00:42] <TheRealJeanLuc> here's the sources.list:
[00:42] <TheRealJeanLuc> http://www.texticle.us/9/
[00:43] <TheRealJeanLuc> I can't ping out due to campus network security restrictions (that's normal), but I can ping devices on-campus
[00:43] <TheRealJeanLuc> there's definitely an internet connection, I can wget files on the internet fine
[00:44] <pmatulis> testicle?
[00:44] <Tellmarch> can you paste the result of the sudo apt-get update?
[00:45] <Tellmarch> maybe not if it's VERY long time;.. :p
[00:45] <TheRealJeanLuc> I would paste the result, but I've let it run for about a half hour without it finishing (which I'm sure is due to the same problem)
[00:46] <twb> pmatulis: OBVIOUSLY you're meant to read it "Text Icleus"
[00:46] <Tellmarch> do you at least get a few lines fast enough?
[00:46] <twb> Boy, that pastebin is dumb.  It puts all the line numbers first, then has all the lines underneath
[00:48] <TheRealJeanLuc> Tellmarch, I do. a Get, two Ign, a Get, and then it goes in to the long series of ignores
[00:49] <Tellmarch> just ignores, no errors?
[00:49] <TheRealJeanLuc> just ignores.
[00:49] <Tellmarch> strange.
[00:49] <TheRealJeanLuc> when I let it run for a very long time, I saw a couple of Errs eventually, after probably 30-45 minutes
[00:49] <TheRealJeanLuc> but not continuous errors, just on occasion
[00:49] <Tellmarch> Connection failed errors?
[00:50] <TheRealJeanLuc> yeah.
[00:51] <Tellmarch> and wget fetch those files without trouble?
[00:52] <Tellmarch> wget http://archive.ubuntu.com/ubuntu/dists/maverick/Release.gpg
[00:52] <Tellmarch> for instance?
[00:53] <TheRealJeanLuc> yeah. it saved fine.
[00:53] <Tellmarch> straaaaaaaaaange
[00:53] <TheRealJeanLuc> my first thought was that the repo servers had gone down somehow, but they seem to be working fine
[00:54] <Tellmarch> must be something wrong in your configs.
[00:54] <Tellmarch> in /etc/apt/apt.conf or something
[00:58] <TheRealJeanLuc> I don't know apt well enough to know if this is normal, but in /etc/apt I have no apt.conf, just an apt.conf.d with a few files with some individual configuration blocks
[00:59] <Tellmarch> ye that's normal
[01:00] <Tellmarch> but "something" must be wrong in it, i don't know what though '^^
[01:04] <Tellmarch> maybe you can try temporirarly to just move the apt.conf.d folder to apt.conf.d.back, create an empty apt.conf.d or something and try the apt-get update...
[01:05] <Tellmarch> or i just don't know :p
[01:06] <TheRealJeanLuc> hahah. k. it's a really weird problem, particularly odd since this is a fresh install. I'm starting to wonder if the install messed up.
[01:07] <Tellmarch> if it's a fresh install you can try reinstalling all, it's fast :p
[01:09] <fbc_> Sounds like you need an open internet connection.. somethign is being filtered.
[01:09] <TheRealJeanLuc> that's what I was thinking, but when I wget files manually they come through just fine
[01:10] <TheRealJeanLuc> and the university uses Ubuntu desktop and server heavily, I can't think they wouldn't have noticed if they're network devices are screwing with updates
[01:10] <TheRealJeanLuc> *their
[01:10] <fbc_> Not sure if repo fetched are UDP or not..
[01:10] <fbc_> WGET I'm pretty sure is TCP.
[01:11] <fbc_> Kinda like a triviaftp which is why they are blocked.
[01:11] <fbc_> maybe...
[01:11] <fbc_> I know that you can try switching your repos from http to ftp and see it that helps.
[01:12] <TheRealJeanLuc> I'll give that a shot. thanks for the tip.
[01:13] <Tellmarch> i'm no expert, but there is a lot of tcp traffic when i do apt-get update. so looks like tcp to me...
[01:13] <Tellmarch> but yeah you can try ftp
[01:14] <Tellmarch> though usually http has more chance of going through than ftp...
[01:16] <TheRealJeanLuc> well, I at least know that I can FTP out (from this machine, even)
[01:16] <TheRealJeanLuc> huh. that had mixed results. it started out with a bunch of error lines, but it appears to actually be working despite the errors.
[01:22] <fbc_> +1 for me
[02:12] <bcessa> hi there, is there a channel where I can ask specific questions about running an ubuntu server on the cloud, particularlly using aws?
[02:15] <erichammond> bcessa: You might try #ubuntu-cloud or ##aws
[02:15] <bcessa> erichammond: thnx, I'll do that
[02:33] <fluvvell> What do the processes ksoftirqd/n and watchdog/n and migration/n  provide in the 10.04 server installations ?
[02:33] <fluvvell> I have 7 such groups run by root.
[02:38] <jmarsden> 7 is a little odd.  One per CPU core is usual.  Are you sure you don't have 8?
[02:41] <fluvvell> jmarsden, to be clear I should have noted 0-7 ie 8 groups ;-)
[02:41] <jmarsden> fluvvell: I don't know exactly what these processes/tasks do, but they are normal in Ubuntu, desktop or server.
[02:42] <jmarsden> On my quadcore desktop here I see 0-3...
[02:43] <fluvvell> jmarsden, i must have two quadcore cpus in this ibm System x3100 M3-4253D2X   - I hadn't really noted properly when I installed it.
[02:44] <jmarsden> fluvvell: grep ^processor /proc/cpuinfo   # will confirm that
[02:44] <twb> grep -c, even?
[02:45] <fluvvell> Count shows 8.  Pleasing, but a little disturbing, Its displaying some slowness for win clients accessing samba, thought I'd configured it the same as its predecessor,
[02:45] <fluvvell> Not much good when the new server appears to run slower than the old.
[02:46] <jmarsden> fluvvell: Time to tune it.  Meanwhile, http://www.tin.org/bin/man.cgi?section=9&topic=ksoftirqd has info on what ksoftirqd is.
[02:47] <jmarsden> http://linux.die.net/man/8/watchdog  # has info on watchdog
[02:50] <jmarsden> The samba docs include a chapter (45) on performance tuning, might be worth exploring for your slowness issue: http://www.samba.org/samba/docs/man/Samba-HOWTO-Collection/speed.html
[02:54] <twb> So guess what I just discovered
[02:54] <twb> Purging ufw from lucid causes it to disable the firewall -- EVEN IF ufw wasn't involved in firewalling
[02:55] <twb> This is especially great when I happen to be removing ufw from a chroot, on a production server running AT A PRISON
[02:56] <jmarsden> Grin!  This is why we should do our learning and trying out new things on a test machine in a safe and controlled environment... I don't think a live production server in a prison counts as a "safe and controlled environment" :)
[02:57] <fluvvell> jmarsden, thanks - reading it now - urr once samba.org wakes up!
[02:57] <jmarsden> fluvvell: You're welcome
[02:57] <twb> jmarsden: well, yes, we did test things
[02:57] <twb> jmarsden: but you tend not to notice when the firewall falls into "allow all" mode on the test box, or you just assume it was one of the other idiot devs that did it
[03:06] <fluvvell> urk, ufw is installed by default! I never noticed that when I installed shorewall
[03:12] <twb> fluvvell: it's installed but not enabled
[03:12] <twb> fluvvell: BUT, purging it doesn't check whether it's enabled first
[03:13] <twb> IMO its postrm should say "am I enabled?  If not, leave the firewall the hell alone"
[06:50] <jetole> Hey guys. I don't know if this is off topic or not but I am not getting much help in #ubuntu. I have many servers. I also just installed ubuntu 10.10 to my workstation at the office. I did ssh-keygen and created a passphrase-less key. I copied this key to the server (either through ssh-copy-id or manually editing the authorized_keys file). I am being prompted for a passphrase for my key. Whatever this pass phrase is supposed to be, I don't know it. ...
[06:50] <jetole> ... doing ssh -v shows that this happens after we send the key to the server once already. Does anyone know what the hell is going on?
[07:01] <gobbe> if it asks passphrase it seems that you gave passphrase during key-generation?
[07:01] <gobbe> or did you just press enter when it asked passphrase?
[07:04] <_ruben> !jaunty
[07:04] <jetole> gobbe: I just pressed enter and I have re-created the key probably a dozen times tonight
[07:04] <_ruben> figured it'd be eol'ed, time to slap some fellow sysadmins
[07:05] <jmarsden> jetole: Also check which keypair is being used here, do you have another keypair (since you have many servers) that it might be asking for a passphrase for?
[07:05] <jetole> _ruben: if it's on a server then slap them for installing it in the first place. A server should only have a LTS edition i.e. 6.06, 8.04, 10.04
[07:06] <jetole> jmarsden: no. This is a new install after a reformat on a work station. I am doing the initial setup for everything. Using ssh -v I can see that it is sending the key to the server, then it asks for a passphrase afterwards like perhaps the server didn't like the key so it's assuming the key is encrypted and then asking
[07:07] <twb> GRAAH
[07:07] <jetole> I have completely deleted and re created the ~/.ssh directory tonight and the workstation might as well be new so I don't think it's getting keys from anywhere else
[07:07] <twb> I just noticed that upgrading ifupdown on my lucid router MADE IT WRONG
[07:07] <jetole> lemme pastebin something and see if maybe that helps
[07:07] <twb> To wit: it added an "auto lo" line, when there was already a PERFECTLY GOOD line "allow-auto      lo unmanaged managed sandbox dmz isp0 internode"
[07:08] <twb> Stupid ubuntu! >>>:-/
[07:08] <jetole> twb: it happens
[07:08] <twb> Good thing this headless router didn't *reboot* before I caught the error, otherwise it'd be bricked.
[07:08] <jetole> and Stupid Ubuntu seems to be a common phrase these days but still better then RHEL/Fedora/CentOS/SuSE
[07:09] <twb> jetole: yes, but not as good as Debian
[07:09] <twb> At least they *test* their releases first :P
[07:09] <jetole> twb: I thought so but debian can't seem to install on lvm on crypt on raid 0 as I just found out since I was going to install debian on this machine
[07:10] <jetole> I tried 5.07 and 6rc1 and both failed what ubuntu did years ago
[07:10] <jetole> I was pretty let down
[07:10] <jmarsden> jetole: pastebin output from an ssh -vv you@server.example.com attempt might reveal something...
[07:11] <jetole> jmarsden: oh... ok, I was just reviewing an ssh -v in vim but I can redo it with -vv
[07:11] <twb> jetole: well, Debian 6 is the same vintage as 8.04
[07:12] <jetole> twb: ok. I don't know if 8.04 handled this but I used this same setup on two other machines except raid1 instead of raid0 since ubuntu 9.10
[07:13] <jetole> jmarsden: http://pastebin.com/vB9uRnVs
[07:13] <twb> jetole: There's no Debian equivalent of 9.10
[07:13] <twb> jetole: the next release (squeeze) will equate to 10.04
[07:14] <jetole> twb: I didn't know but the rc1 that failed was the next release
[07:14] <jetole> it was dated 01/12/11
[07:14] <twb> Oh right
[07:14] <twb> Did you report the bug?
[07:15] <jetole> twb: no but I had a thurough conversation with it about both issues with a few people in #debian who can report it
[07:15] <jetole> I know. poor answer
[07:16] <twb> jetole: should've used #debian-boot; they're the installer guys
[07:17] <jetole> jmarsden: see lines 73 and 74 @ http://pastebin.com/vB9uRnVs
[07:17] <jetole> twb: wish I knew. I really wanted to move to debian because I wanted the rolling releases and I was very let down
[07:18] <jetole> I hate upgrading ubuntu (which always breaks something) or being stuck with a system that doesn't have the latest packages
[07:18] <twb> Sounds like you want arch
[07:18] <jetole> I used debian about a 6 - 8 years ago and was thinking about migrating back but I will have to wait to the next time I reinstall an OS to try again
[07:18] <jetole> no I don't
[07:19] <jetole> debian does rolling releases in the testing branch
[07:19] <jmarsden> jetole: I think lines 77 and 78 are "interesting", and that is where this diverges from my ssh -vv trace...
[07:19] <jetole> jmarsden: they are very interesting indeed. Don't suppose you know what it means?
[07:20] <jmarsden> not yet, googling :)
[07:20] <jetole> jmarsden: I am too. also, does debug1 refer to the first -v and debug2 to the second / -vv ?
[07:20] <jmarsden> Yes
[07:22] <jetole> jmarsden: "This is not an error. OpenSSH first tries to read the private key with a no encryption. That fails, so it prompts for a passphrase." http://fixunix.com/ssh/73558-pem_read_privatekey-failed.html
[07:24] <jetole> jmarsden: I don't have ssh-agent running. Do you think this is the issue?
[07:24] <jetole> ah nevermind. I am going to test it
[07:24] <jmarsden> That is a difference between your setup and mine, so sure, test it :)
[07:26] <jetole> Well that didn't work
[07:27] <jetole> be back in 10. I need to step out for a smoke quickly
[07:29] <jmarsden> OK.  Just tested here setting my passphrase to empty (using sshkeygen -p ) and it worked fine.
[07:37] <jetole> jmarsden: yeah it has always worked fine for me too. I have done this for a long time with empty passphrase ssh key
[07:37] <jmarsden> If you do it with an actual passphrase for the same workstation and server, does it work fine?
[07:38] <jmarsden> (in other words, is the zero-length passphrase really the issue here, or is the problem elsewhere?)
[07:40] <jetole> Got it
[07:41] <jetole> I don't know what it had to do with what but here are the startup services I initially disabled and just re-enabled now
[07:41] <jetole> Certificate and Key Storage, Secret Storage Service, SSH Key Agent
[07:42] <twb> jetole: try with -oBatchMode=yes
[07:42] <jetole> twb: I got it working by re-enabling those startup applications in gnome
[07:42] <twb> Sigh
[07:42] <twb> Flipping gnome crunk
[07:43] <jetole> Sigh is right. Why should my openssh client be compiled against gnome
[07:43] <jetole> jmarsden: thanks a ton though for the help
[07:43] <jmarsden> jetole: No problem, glad you found it.
[07:43] <jetole> still a little upset because of where I found it
[07:44] <jetole> but hey, at least it's working
[07:44] <twb> jetole: it's not, but gnome will be hijacking shit
[07:45] <twb> jetole: if you didn't run gnome, you wouldn't have any problems at all
[07:46] <jetole> twb: yeah well I tried kde again for the first time in a while tonight and... meh. Don't feel like going back to xfce, haven't used fluxbox in a decade though I did like it then but it seems pretty minimal still... I don't know
[07:46] <twb> jetole: no, I mean don't run a GUI
[07:46] <jetole> and the new ubuntu one.. epiphany, entropy, whats it called
[07:46] <jetole> oh well yeah
[07:46] <jetole> then there's that option
[07:46] <jetole> lol
[07:46] <twb> What is the point of going through preschool and learning to read and write if you're going to click on things
[07:47] <jetole> twb: I don't know. The voice recognition on my cell phone is probably more literate then I am
[07:47] <jetole> heh
[07:47] <twb> Kids these days
[07:47] <jetole> twb: what ripe old age are you?
[07:47] <twb> Yes.
[07:48]  * jetole rolls eyes
[07:48] <jetole> I've been using ubuntu for 13 years since I was 13 so I think I have surpassed kid status
[07:48] <jetole> also took my first programming class, C++ (not C but thank god it wasn't basic) when I was 13
[07:49]  * jetole wonders what else I want to customize on this workstation
[07:52] <jetole> hmmmm... considering going back to fluxbox
[07:52] <jetole> off topic though so never mind
[08:06] <erwin_wahl> hello
[08:07] <erwin_wahl> is here anyone active
[08:08] <erwin_wahl> quit
[08:34] <shey> Can anyone assist with apache2 setup? php pages will not load. Mods are all enabled... restarted, still no go..
[08:35] <gobbe> have you followed docs from ubuntu.com?
[08:35] <shey> all of them.
[08:35] <gobbe> !lamp
[08:35] <shey> I've been reading for hours now..
[08:35] <gobbe> that one
[08:36] <shey> giving it a shot now. thx.
[08:43] <shey> I got all that done already, even userdir works.
[08:43] <shey> but php pages refuse to load.
[09:28] <binBASH> moin \sh
[09:36] <\sh> moins
[09:45] <sidd_mak> how is ubuntu server different the desktop ??
[09:51] <Tellmarch> sidd_mak, no graphical interface (in the default installation)
[09:51] <Tellmarch> and few different options in the kernel
[09:53] <sidd_mak> Tellmarch : but we can install X on server right
[09:53] <Tellmarch> yes, it's possible
[09:53] <Tellmarch> it's the same packages, so...
[09:53] <Tellmarch> you can have the same as on your desktop
[09:53] <sidd_mak> Tellmarch : is the kernel enhanced for server
[09:53] <Tellmarch> yes
[09:54] <sidd_mak> so desktop n server uses different kernels right??
[09:54] <Tellmarch> • The Server Edition uses the Deadline I/O scheduler instead of the CFQ scheduler used by the
[09:54] <Tellmarch> Desktop Edition.
[09:54] <Tellmarch> • Preemption is turned off in the Server Edition.
[09:54] <Tellmarch> • The timer interrupt is 100 Hz in the Server Edition and 250 Hz in the Desktop Edition.
[09:54] <Tellmarch> same kernel, but different options
[09:56] <sidd_mak> so if preemption is not there so no jobs are preempted n they a processed at once
[09:57] <Tellmarch> honestly, i don't even know what it means, i'm no specialist of this :p
[09:57] <Tellmarch> but i trust that it's better for servers :)
[09:57] <sidd_mak> ok thanx man
[10:00] <lambda_x> anybody here is using libvirt/kvm? Mine virtual machines hang during virsh save, I need to restart libvirt to get access to them again, virsh resume brings it back - i can access them through vnc but I cannot type, network interface does not answer too
[10:02] <\sh> zul: your feed aggregation is somehwat broken on p.u.c
[10:10] <HelpMeSamba> hello everyone, i just updated my karmic koala to lucid and I can't connect to my samba shares anymore can anyone help me please? I've tried unstalling the 3 packages to enable personal file sharing and ticked the share box and i'm still unable to access my samba server shares
[10:25] <Tellmarch> i'm trying to use rails on ubuntu-server 10.04... should i install the packages of ubuntu for rails etc., or use gem install instead?
[10:26] <gobbe> i would go with ubuntu-ones if there's no special need to use gem ones
[10:27] <Tellmarch> well, apparently they aren't exactly up to date... and i can't follow my tutorial because of this...
[10:28] <Tellmarch> for instance 'rails server' doesn't work.
[10:32] <Tellmarch> bon ok j'ai trouvé le guide pour la vielle version :p
[10:32] <Tellmarch> je vais voir si ça suffit à faire tourner l'application que j'ai...
[10:39] <Tellmarch> raté.
[10:48] <ara> hello guys!
[10:48] <ara> I am trying to install ubuntu server (natty daily) in some servers with preseed, but in most of the systems it is stuck because it asks for a partition table to use
[10:48] <ara>  could it be that the prior installation (also natty) corrupted the partition table somehow?
[10:48] <ara>  this is the preseed file that I am using: http://paste.ubuntu.com/556098/
[11:40] <pmatulis> ara: try to install non-natty
[11:48] <JamesPage> Hi ara: hggdh spotted this yesterday in the automated server ISO testing.
[11:48] <binBASH> Hi all, I setup openvpn to accept duplicate cn in certificates. I need some clients to have static ips so I've setup own certifates for these and put them in the client config dir. They are getting static ips now. For  the other clients I use a cert with a generic common name. But for those clients the ips of the static clients get assigned as well. What is wrong?
[11:48] <JamesPage> it looks like d-i has changed in someway - I'm not sure how far he got investigating it.
[11:49] <ara> JamesPage, thanks, I am also with Colin having a look to this. I will talk with hggdh when he gets in
[11:50] <JamesPage> ara: np - shout if I can help in any way
[11:51] <JamesPage> does anyone have any experience in setting up ipv6 in virtual machines running under qemu-kvm/libvirt?
[12:05] <TREllis> horrah for cobbler in the archive \o/
[12:06] <lynxman> yay for cobbler
[12:18] <Name141> What is it you put after your crontab for it not to send you mail? > /dev/null 0 or something ?
[12:18] <greppy> Name141: > /dev/null
[12:19] <Name141> greppy: that's all I need? like /home/name141/bnc/bncchk > /dev/null
[12:19] <Name141> and it'll stop sending me mail after it restarts it?
[12:20] <greppy> yes.
[12:20] <Name141> greppy: I'm seeing '>/dev/null 2>&1' on google searches , what's the difference ?
[12:21] <greppy> that lets stderror through to be emailed but not stdout.
[12:21] <Name141> I'm not understanding ?
[12:22] <greppy> Some programs will output errors on a different file handle, known as STDERROR instead of STDOUT.
[12:23] <pmatulis> Name141: http://tinyurl.com/6ar7yrz
[12:23] <Name141> pmatulis: bbq
[12:25] <Name141> so anyway, to stop it fully > /dev/null ?
[12:26] <greppy> yes
[12:30] <_ruben> not really, just >/dev/null would only "ignore" stdout, not stderr
[12:38] <ara> JamesPage, hggdh: bug 705377
[12:38] <ara> colin is having a look to it
[12:56] <hggdh> ara: yes, just bypassed it on my local version of Hudson
[12:56] <hggdh> ara: and... good morning :-)
[13:15] <iclebyte-work> has anyone used openqrm before?
[13:15] <pmatulis> Name141: bbq?
[13:19] <Name141> [06:18:49:AM] <Name141> greppy: I'm seeing '>/dev/null 2>&1' on GOOGLE searches , what's the difference ? [06:21:28:AM] <pmatulis> Name141: http://tinyurl.com/6ar7yrz
[13:19] <Name141> yeah , made the same sense
[13:19] <Name141> (as telling me to use google when I was using google)
[13:21] <jdstrand> twb: re ufw purge-- this is bug #581744, fixed in maverick. would you mind adding a comment to that bug, saying it affects you on lucid, and I can do an SRU
[13:21] <twb> jdstrand: thanks
[13:22]  * twb tries to log into lp
[13:22] <pmatulis> Name141: the first hit from the search i gave you contains the answer
[13:27] <twb> jdstrand: done via signed email
[13:41] <lambda_x> is there anyone who have libvirt/kvm with *working* save and restore vm?
[13:42] <jdstrand> twb: thanks
[13:44] <twb> lambda_x: I think I did it once with plain qemu
[13:44] <twb> lambda_x: a long time ago
[13:45] <lambda_x> twb: thats my point. nobody uses ubuntu-server default libvirt / kvm virtualisation technology because it is unstable and does not fit any production environment
[13:46] <lambda_x> 1.5 a year and none of important issues have been solved wtf...
[13:46] <twb> I don't use it because 1) I don't have hardware VT; and 2) I hate libvirt getting in my way.
[13:46] <twb> AFAICT libvirt's mainly for people who like XML and don't like knowing what they're doing
[13:46] <lambda_x> nah
[13:47] <lambda_x> its unix way
[13:47] <lambda_x> you have a tool, then another tool to manage first one, then graphical tool
[13:47] <lambda_x> its all ok
[13:47] <twb> No, only losers have a graphical tool
[13:47] <lambda_x> but libvirt sucks
[13:47] <twb> libvirt is the webmin of virtualization.
[13:47] <lambda_x> twb: some losers have bosses who need graphical tool to see it is working
[13:48] <twb> You can't solve a social problem (the boss' stupidity) with technology.
[13:48] <twb> It'd be far better to swap in a new boss
[13:48] <twb> jdstrand: that's odd, the ticket says *I* created it.
[13:49] <greppy> twb: not always a possibility, which is too bad.
[13:53] <jdstrand> twb: I think it is cause you didn't creat a separate task, but a bug against a different 'distribution'. that's fine, I cleaned it up
[13:58] <twb> OK.
[14:10] <zul> jdstrand: umm...cobbler is still in binary new?
[14:12] <jdstrand> zul: yes, it wasn't ready when I last checked before eod yesterday. I can look at it now
[14:12] <zul> jdstrand: thanks
[14:19] <RoAkSoAx> Morning
[14:20] <iclebyte-work> morning
[14:20] <iclebyte-work> or afternoon as it is in my case
[14:23] <jdstrand> zul: fyi, this looks weird to me:
[14:23] <jdstrand>   -rwxr-xr-x root/root      3411 2011-01-19 15:40 ./etc/cobbler/cobblerd
[14:24] <jdstrand> zul: looking at it, it seems to be a sysv init script. I'm thinking it is misplaced?
[14:25] <zul> jdstrand: interesting
[14:25] <jdstrand> zul: I won't nak it, but will file a bug
[14:25] <zul> jdstrand: thanks ill get it fixed
[14:28] <jdstrand> zul: fyi bug #705441
[14:28] <zul> jdstrand: thanks
[14:31] <jdstrand> sure
[14:31] <jdstrand> zul: accepted
[14:32] <zul> jdstrand, \o/
[14:38] <raubvogel> Easy question of the day: how do I define a /28 network in bind? Specifically, how do I do the zone definition in named.conf.local?
[14:40] <RoyK> bind/dns doesn't care about netmasks
[14:41] <gobbe> just configure ips
[14:44] <m_tadeu> hi...my server stops after a day or so...it loses the ethernet connection, so no ssh, webserver, no ping...I try to connect a screen with no success. I try to reboot with the 'reboot' command, nothing happens. The logs don't say a thing
[14:46] <m_tadeu> what can  this be, or where can I check for trouble?
[14:46] <gobbe> eh
[14:46] <gobbe> so it does nothing?
[14:46] <gobbe> i would go and chek memory
[14:46] <m_tadeu> nothing at all...at least at the interfaces
[14:48] <gobbe> well...reboot does nothing?
[14:48] <m_tadeu> nothing....I mean the reboot command...
[14:49] <m_tadeu> pressing the shutdown button does nothing, which means the APM is not responding
[14:50] <m_tadeu> hot shutdown works hehe
[14:51] <raubvogel> So, gobbe and Royk, if I have, say, 128.122.42.80/24, my reverse zone will be simply called 42.122.128.in-addr.arpa
[14:51] <patdk-wk> sure
[14:52] <patdk-wk> anything smaller than a /24 you have to define it per ip
[14:52] <raubvogel> The ip for the network itself (.80)?
[14:53] <RoyK> 128.122.42.80/24 isn't a valid network
[14:54] <RoyK> 128.122.42.0/24 is, though, and 128.122.42.80/28 is (iirc)
[14:54] <m_tadeu> what's the command for mem check?
[14:54] <RoyK> but bind doesn't care - just create the zone file for 128.122.42.0/24 and just register the hosts
[14:54] <RoyK> m_tadeu: reboot and choose memtest86 :P
[14:55] <m_tadeu> RoyK: that I know....but isn't there a command for that?
[14:57] <RoyK> m_tadeu: no - because kernel has already control over the memory after bootup, and you don't want that
[15:01] <blackxored> Hello guys, any of you have experience associating an iphone client to a freeradius-enabled wireless setup?
[15:03] <m_tadeu> RoyK: I see...well the grub menu is not showing up...the timeout is set for 10secs
[15:04] <raubvogel> Royk, I meant 28 :)
[15:06] <RoyK> m_tadeu: hold down shift
[15:08] <m_tadeu> RoyK: thanx...now as I remember this is a non-stop memcheck, right?
[15:18] <compdoc> you running memtest86?
[15:19] <m_tadeu> I am
[15:19] <compdoc> yeah, its non-stop, but if you suspect ram, you should let it run a few passes
[15:20] <m_tadeu> ok...finished the first pass now
[15:20] <compdoc> Ive seen ram pass one test, but fail after a few
[15:20] <compdoc> but its also possible your ram is ok
[15:21] <m_tadeu> I'll leave it for the afternoon, then
[15:21] <compdoc> its nice to run over-night to test a system. they also have programs that stress-test a system
[15:22] <m_tadeu> can you point me to those?
[15:22] <m_tadeu> guess I'll write a doc for testing procedures
[15:22] <compdoc> I dont use any - I look for bad caps on older systems, or burned components. If they look ok, I just run memtest86+
[15:23] <compdoc> having a problem?
[15:23] <m_tadeu> in deed...the server completely stops responding after a day or so...not a fixed time
[15:23] <m_tadeu> but I must not use it for a day or so
[15:24] <compdoc> how old is it?
[15:24] <m_tadeu> 3months old
[15:24] <compdoc> so its all new parts?
[15:24] <m_tadeu> yep
[15:25] <jeremytoo> I have recently built a 10.04.01 LTS server with / on a RAID1 setup.  When I boot it, I never get a grub prompt -- it simply goes straight to booting.  Is this by design, or have I misconfigured something?
[15:25] <compdoc> some ram needs the voltage set higher than normal - its usually printed on the module. make sure you set it righ
[15:25] <compdoc> right
[15:26] <compdoc> theres also a bios setting that causes raid controllers to not work right, at least for me
[15:26] <m_tadeu> I'll check after memtest
[15:26] <compdoc> its called PCI Latency Timer
[15:27] <sjm> jeremytoo, now that you mention it, I think I had the same thing happen to me recently (RAID off a 3ware card).
[15:28] <jeremytoo> sjm:  did you ever get it to show the grub menu?
[15:28] <sjm> jeremytoo, let me boot it up, it's here beside me.
[15:28] <m_tadeu> I'm having the same thing happening here...no grub menu
[15:30] <compdoc> what partition type you guys using? msdos? gpt?
[15:30] <sjm> nope, no grub
[15:30] <jeremytoo> compdoc:  I set everything up through the GUI -- so whatever the default installer uses
[15:31] <sjm> I installed off the mini.iso:  bare minimum and then added things through aptitude.
[15:31] <compdoc> that should work
[15:31] <m_tadeu> ext4
[15:31] <compdoc> Im running memtest on a system Im building right now
[15:31] <jeremytoo> I booted off the server install CD and set it up, two 1 TB SATA drives in a raid1 array.
[15:32] <jeremytoo> yeah, I'd like to be able to access memtest86 in a pinch...
[15:32] <kirkland> smoser: ping
[15:32] <compdoc> is it a software raid?
[15:32] <jeremytoo> compdoc:  yes
[15:32] <sjm> Yes, I'd like the grub menu too.  Mine's hardware raid
[15:33] <compdoc> I use 3ware - great controller
[15:33] <sjm> I have a 3ware too
[15:34] <compdoc> but I dont boot off my arrays. I like a small drive for the OS
[15:35] <sjm> I have a two drive raid 1 for the OS and a 4 drive raid 10 for data.
[15:35] <sjm> but it's all off the 3ware controller.
[15:35] <sjm> but, back to the question... what happened to the grub menu?
[15:36] <compdoc> yours was working?
[15:36] <sjm> no, It was a new install (off the mini.iso)
[15:36] <compdoc> oh
[15:37] <sjm> (I had a debian install before that did show the menu, IIRC)
[15:38] <compdoc> dont remember - is there a setting in the 3ware bios to set a drive to bootable?
[15:38] <sjm> But it's not just 3ware, jeremytoo has software RAID
[15:39] <compdoc> yeah, and you could both have completely different problems for that reason
[15:39] <jeremytoo> it would appear that on bootraid setups, the grub-installer is doing SOMETHING to grub.cfg which precludes a grub menu being displayed
[15:39] <jeremytoo> I'm trying to find out if that's a happenstance thing, or by design?
[15:40] <jeremytoo> my previous 10.04.01 box booted off RAID1, but was an upgrade, so it had a grub boot menu
[15:42] <compdoc> in the mobo bios, it sees the raid and you have the system set to boot it?
[15:42] <jeremytoo> it boots 100% fine
[15:43] <JamesPage> zul: when do you reckon Samba 4 might make it to the main archive?
[15:43] <jeremytoo> the bios do not see the raid, compdoc, because I'm using softraid
[15:43] <jeremytoo> (boot partition is on /dev/md0)
[15:43] <zul> JamesPage; hehehee
[15:43] <zul> JamesPage; its still alpha
[15:43] <JamesPage> zul: yep; thought that might be your response :-)
[15:44] <sjm> compdoc, yes, the mobo bios sees the 3ware raid and it is set in the boot order.  my boot partition is "/dev/sda1"
[15:45] <compdoc> well, if no one in here can help, you might try the mailing list - lots of ppl read it
[15:47] <gyppo> hello?
[15:47]  * genii-around hands gyppo a coffee
[15:47] <gyppo> thanks...
[15:47] <gyppo> I was wondering if someone here could point me straighrt
[15:47] <gyppo> I'm trying to set up postfix...
[15:47] <compdoc> point to what?
[15:48]  * jeremytoo points due west for gyppo
[15:48] <gyppo> and, while I can telnet 192.168...... 25
[15:48] <gyppo> I can't telnet from outside the router
[15:48] <gyppo> I'm pretty sure I'm port-forwarding 25
[15:48] <compdoc> you isp could be blocking 25 - mine does
[15:48] <jeremytoo> gyppo:  as does mine
[15:49] <gyppo> ah. any way to check?
[15:49] <compdoc> try forwarding port 2525 to 25 on the server, then see iff telnet 2525 works
[15:49] <gyppo> of course, excellent idea :)
[15:49] <gyppo> what if it is my isp?
[15:50] <gyppo> I assume there's a common solution
[15:50] <gyppo> ?
[15:50] <gyppo> Will I still be able to smtp pop etc, or will it need tweaking?
[15:51] <compdoc> you can buy a business plan, that allows all ports, or use a company like dyndns.org to send mail to your port 2525, or whatever port you like
[15:51] <compdoc> well, not just any port, but they allow several to choose from
[15:52] <jeremytoo> gyppo:  how many users do you intend to serve?  As lame an answer as it is, gmail w/ custom domain name is free, and easy to administer
[15:52] <sjm> many also have started using a default port 587 rather than 25.
[15:52] <compdoc> pop and the other ports usually work. just not 25
[15:52] <gyppo> default port 587? Thanks, that's something. But won't that just move the problem
[15:52] <compdoc> of course your mileage may vary
[15:53] <gyppo> and thanks, jeremytoo, but I'm trying to get away from google :)
[15:53] <jeremytoo> gyppo:  I'm on comcast -- I had to migrate my SMTP server off to geekindustries.com when comcast sealed down port 25.
[15:53] <compdoc> yeah, comcast used to allow 25
[15:53] <sjm> jeremytoo, try editing the timeouts your /etc/default/grub, run "sudo update-grub" and reboot.
[15:54] <jeremytoo> sjm:  that's what I'm thinking.
[15:54] <sjm> I just got a boot menu
[15:54] <gyppo> yeah, I just tried 2525->25, works fine
[15:54] <gyppo> pesky isp
[15:55] <gyppo> thanks guys
[15:55] <compdoc> btw, its dyndns.com, not .org
[15:55] <jeremytoo> gyppo:  check your Terms of Service -- you may be violating the ToS by running an SMTP server, exposing yourself to the risk of immediate termination of service and financial penalties
[15:58] <sjm> jeremytoo, I think I found what is happening...
[15:59] <sjm> jeremytoo, check this page, section 4, bullet "hidden": http://ubuntuforums.org/showthread.php?t=1195275
[16:00] <sjm> jeremytoo, I think you (and I) are really getting the menu, but it's hidden.  hold down SHIFT while booting to see the menu.
[16:01] <jeremytoo> sjm:  I commented out two lines in the grub menu to slow boot slightly and display the menu:
[16:01] <jeremytoo> #GRUB_HIDDEN_TIMEOUT=0
[16:01] <jeremytoo> #GRUB_HIDDEN_TIMEOUT_QUIET=true
[16:02] <jeremytoo> sjm:  actually, those changes went in /etc/default/grub
[16:02] <jeremytoo> there, now the loggers and google can find that answer the next time I lose it.
[16:03] <sjm> jeremytoo, I don't think it will slow the boot at all (that's probably GRUB_TIMEOUT), but it will show it.
[16:04] <jeremytoo> I read through the grub config file -- it was actually bypassing that timeout entirely before -- setting GRUB_HIDDEN_TIMEOUT=0
[16:04] <jeremytoo> so I'd have had to be REALLY fast on the shift key
[16:04] <sjm> I'm being dragged into knowing grub2 rather than just grub
[16:05] <jeremytoo> I still miss lilo.
[16:05] <jeremytoo> as crappy as it was
[16:06]  * geekbri hisses at jeremytoo
[16:06]  * jeremytoo smirks and promises to watch his language.
[16:06] <sjm> jeremytoo, I assumed that the hidden timeout was the time that it would stay hidden rather than showing and when set at zero was "infinity...and beyond" while the GRUB_TIMEOUT sets the actual wait.
[16:07] <jeremytoo> could be.
[16:07] <jeremytoo> I just built this core i7-2600k sandy bridge machine and it boots so dang fast you don't have ANY time to think
[16:07] <geekbri> jeremytoo: the only language i was offended by was "lilo"
[16:07]  * jeremytoo chuckles.
[16:07]  * jeremytoo dusts off his turbolinux 3.01 CD
[16:07]  * sjm wonders how jeremytoo is going to watch the sound waves of "language"
[16:07] <jeremytoo> through an oscilloscope, how else?
[16:08] <sjm> jeremytoo, the page says you can just hold down the shift key during the whole boot process.
[16:09]  * sjm is getting jealous while booting his "fastest" Pentium D server.
[16:14] <sjm> jeremytoo, initial testing says you might be right.  the GRUB_TIMEOUT might only take effect once the menu is shown.
[16:14] <jeremytoo> this is my main KVM server, so now that it's working, I'm done testing ;)
[16:15] <sjm> jeremytoo, mine is still in the development area, so I have been rebooting it after a few changes to see the effects.
[16:15] <jeremytoo> sjm: nice.
[16:38] <Naia> v10.04 server, 64bit. Failed on installation of kxsldbg-0.4. Was installing it as part of a web development application Quanta Plus. Error at: http://paste.ubuntu.com/556198/
[16:44] <sjm> Naia, you might be missing a package or two.  check here: http://astbook.asteriskdocs.org/en/2nd_Edition/asterisk-book-html-chunk/asterisk-CHP-3-SECT-8.html
[16:47] <Naia> Error: sudo yum install gcc-c++. Setting up Install Process. No package gcc-c++ available. Nothing to do.
[16:48] <Naia> Same error with recommended libstdc++-devel install.
[16:48] <sjm> naia, yum?  are you on ubuntu?
[16:49] <Naia> Aye. But the page you recommended said to type yum. I'm only a week old on Ubuntu.
[16:50] <takamarou> Hi all.. I'm trying to set up ubuntu 10.10 on a 3(ish) year old server I've got.  The server has two dual-core Xeon CPUs in it.  After a fresh install, I'm seeing that the server is only utilizing one of the two CPUs.  Could anyone walk me through enabling that second CPU?  Thanks!
[16:51] <gobbe> did you install from server-media?
[16:51] <takamarou> me?  I installed from a 10.10 disk
[16:52] <gobbe> yep, but is it server-media?
[16:52] <gobbe> or alternative or what?
[16:52] <takamarou> Server, I believe.  I downloaded the ISO directly from the ubuntu site, server section
[16:53] <gobbe> ok, and how do you believe that only one cpu is used?
[16:53] <takamarou> I'm using that 30 day landscape trial, and only lists one of my Xeon Processors under hardware
[16:55] <gobbe> hmmh
[16:55] <sjm> Naia, try:  sudo aptitude install libstdc++6-4.4-dev
[16:55] <gobbe> takamarou:  sudo lshw -c cpu
[16:55] <takamarou> but, sudo lshw | grep Xeon shows up with 2 processors.. it appears so anyways..  Perhaps landscape is just wrong?
[16:55] <Naia> May I paste 3 lines here?
[16:55] <gobbe> takamarou: upload output to pastebin
[16:56] <gobbe> it might be that landscape is wrong
[16:56] <takamarou> yeah, that definitely outputs two cpus
[16:56] <sjm> Naia, I saw your paste from the ubuntu list.  Is it something different?
[16:56] <gobbe> landscape is still quite poor, far away from satellite/spacewalk
[16:56] <Naia> Yes. New errors.
[16:56] <sjm> s/list/channel/
[16:57] <sjm> Naia, keep it in pastebin for now, i would say.
[16:57] <takamarou> gobbe, are satellite/spacewalk good alternatives to landscape?  I'm actually a bit put off by the high cost on landscape
[16:57] <gobbe> takamarou: they don't work with ubuntu :/
[16:57] <takamarou> ouch.  Know any other good ones?
[16:57] <Naia> sjm: new error pasted. http://paste.ubuntu.com/556212/
[16:58] <gobbe> takamarou: sorry, no :-/ I remember that some guy was trying to make spacewalk-fork to work with ubuntu
[16:58] <takamarou> cool.  I'll look into it.  Thanks for the help.  I suppose I'll trust my own servers output before I trust landscape from now on :)
[16:59] <sjm> Naia, you didn't type it in correctly.  you typed: "libstdc__6-4.4-dev" and not "libstdc++6-4.4-dev"  (underscores rather than plus signs)
[16:59]  * Naia blushes.
[17:00] <gobbe> takamarou: http://administratosphere.wordpress.com/2010/01/28/system-management-software-spacewalk-and-landscape/
[17:00] <Naia> sjm: done.
[17:01] <sjm> Naia, now can you try the:  ./configure  and  see if it runs without errors?
[17:01] <Naia> sjm: I should try reinstall of original software, now?
[17:02] <Naia> sjm: checking for X... configure: error: Can't find X includes. Please check your installation and add the correct paths!
[17:05] <sjm> Naia, sudo aptitude install xlibs-dev   ( You need to install the development packages to be able to compile)
[17:06] <Naia> Not sure what those are. I'm sorry.
[17:06] <sjm> Naia, install the xlibs-dev package and then try ./configure again.
[17:07] <Naia> sjm: Package xlibs-dev is not available, but is referred to by another package.
[17:09] <raubvogel> anyone using opendkim? If so, has anyone experienced the not being able to verify signatures using opendkim-testkey mentioned in http://stackoverflow.com/questions/4102383/debugging-opendkim-postfix-bad-signatures-and-verification-failures?
[17:10] <sjm> Naia, what package are you trying to install?
[17:11] <Naia> sjm: http://paste.ubuntu.com/556217/
[17:12] <sjm> Naia, if it is kxsldbg, can you just install the one already packaged:  sudo aptitude install kxsldbg-kde3  ?
[17:12] <Naia> sjm: Couldn't find any package whose name or description matched "kxsldbg-kd3"
[17:16] <sjm> Naia, would installing the package  "kdewebdev"  meet your needs?
[17:17] <Naia> Let me paste what started this fiasco.
[17:20] <Naia> sjm: http://paste.ubuntu.com/556223/
[17:23] <sjm> try this:  aptitude install kdewebdev cervisia
[17:24] <sjm> Naia, ^^ that should get you those two packages without having to compile them.
[17:24] <Naia> http://paste.ubuntu.com/556225/
[17:25] <sjm> Naia, do you want it for kde3 or kde4?
[17:26] <Naia> Unknown answer. I'm only a week old at this ... and all I wanted was to do web editing. ;)
[17:26] <ideaman> Can anyone tell me why sudo might take 6 seconds to respond every time?
[17:28] <sjm> try this instead:  sudo aptitude install cervisia kxsldbg-kde3
[17:28] <RoyK> ideaman: strace it
[17:28] <sjm> Naia, try this instead:  sudo aptitude install cervisia kxsldbg-kde3
[17:29] <ideaman> RoyK: wow thanks, I'm still a noob, didnt even know about that. Give me about 10 minutes. thats about how long after a fresh reboot it takes to act up.
[17:30] <RoyK> wtf?
[17:31] <RoyK> 10 minutes to boot ubuntu server?
[17:31] <Naia> sjm: Couldn't find any package whose name or description matched "kxsldbg-kd3"
[17:31] <ideaman> no no
[17:31] <ideaman> after it boots, i dont have this issue until it's been up for 10 minutes
[17:32] <RoyK> ideaman: what are you running on this thing?
[17:32] <RoyK> any heavy services?
[17:32] <ideaman> nada in top
[17:32] <RoyK> how much memory?
[17:33] <ideaman> 2GB, 250HD
[17:33] <ideaman> only it's just an asterisk box
[17:33] <RoyK> asterisk .. ugh
[17:33] <RoyK> just basic chan_sip etc?
[17:34] <ideaman> yea very basic not even connected to the outside world yet though
[17:34] <sjm> Naia, go into synaptic and add the universe repository.
[17:37] <RoyK> ideaman: well, let's wait some more minutes to see if the error occurs again
[17:37] <Naia> sjm: I type universe reposit and it comes up with Kubuntu-restricted-addons, Lubuntu-restricted-addons, xubuntu-restricted-addons and ubuntu-restricted-addons. Only one installed is ubuntu-restricted-addons.
[17:37] <RoyK> ideaman: next time, don't just reboot to fix an unknown problem - better spend some time digging into it :)
[17:38] <Naia> sjm: please don't shoot me.
[17:39] <sjm> Naia, are you using synaptic or aptitude or apt-get?
[17:39] <Naia> Synaptic Package Manager.
[17:39] <m_tadeu> which monitoring tools do you recomend? I want to monitor cpu/mem/disk, http, voip, ssh, mysql
[17:40] <sjm> Naia, in the menu, go to Settings ->  Repositories.  Under the Ubuntu Software tab, check the second line that has "universe" at the end of the line.
[17:42] <sjm> m_tadeu, depends on what you want and your environment.  You *could* do it all with tcpdump.
[17:42] <Naia> sjm: I'm sorry. I see no settings tab and I looked under System>Preferences and System>Administration.
[17:42] <sjm> no, inside Synaptic.
[17:42] <Naia> Duh
[17:43] <Naia> Is already checkmarked ....
[17:43] <m_tadeu> sjm: I'd like to have averages, peaks, notifications for critical status etc...a web client would be nice too :)
[17:43] <sjm> Naia, then you need to "reload" and search for kxsldbg-kde3
[17:44] <Naia> sjm: Please define reload.
[17:44] <Naia> Duh
[17:44] <sjm> m_tadeu, I think Nagios can get you most of that.
[17:44] <sjm> m_tadeu, maybe cacti
[17:45] <Naia> sjm: Found it. Installed. That error gone from Quanta. Only one left is the cervisia.
[17:46] <Naia> sjm: Synaptic says cervisia (cvs client for KDE4) is already installed.
[17:47] <sjm> Naia, you might be seeing conflicts between kde3 and kde4 stuff.  Unless you are going to use CVS with Quanta, I wouldn't worry about it.
[17:47] <Naia> Is there a reason I should?
[17:47] <m_tadeu> sjm: cool...thanx...gonna read docs about those
[17:48] <sjm> Naia, if you don't know what CVS is, then you probably won't be using it.
[17:48] <Naia> True enough. Thank you for all your help. :) I appreciate the patience.
[17:51] <Naia> sjm: I looked it up. Nah. I don't worry about tracking changes with the little work that I do. Thank you again for all your help!
[17:51] <sjm> Naia, you're welcome.  Have fun.
[18:08] <ideaman> RoyK: still there?
[18:12] <RoyK> ideaman: back
[18:12] <fullstop> Hi all.  I'm running ubuntu-server 10.04 LTS, and I've run into this issue: https://bugzilla.redhat.com/show_bug.cgi?id=605757
[18:13] <fullstop> I'm getting "OOPS-" messages when searching ubuntu's bug database, so I don't know if a bug report exists already.
[18:13] <ideaman> so the error is back
[18:14] <fullstop> anyway, it is supposedly fixed in qemu 0.12.5... so I rebuilt the appropriate packages from 10.10 and put them on my KVM host.
[18:15] <fullstop> I'll know by tomorrow if the problem is gone or not.
[18:15] <RoyK> ideaman: ok
[18:15] <RoyK> ideaman: check memory available for a start
[18:16] <ideaman> RoyK: memfree in top?
[18:16] <RoyK> or just 'free'
[18:16] <RoyK> pastebin free
[18:16] <RoyK> !pastebin
[18:17] <ideaman> RoyK http://pastebin.com/8XH2UsBu
[18:18] <RoyK> ideaman: pastebin the command resulting in this error, please
[18:18] <RoyK> it certainly isn't lack of memory
[18:18] <shey> Can anyone supply help with LAMP? I cannot seem to get .php files to load, always asked to download. Yes, I have followed all instructions as per the How-To... added the modules, restarted, rebooted, kicked it a few times.... nothing...
[18:19] <ideaman> done
[18:19] <gobbe> shey: upload your configs to pastebin
[18:19] <RoyK> shey: a2enmod php5?
[18:19] <shey> RoyK: yeah.
[18:19] <RoyK> ideaman: we need an url :P
[18:20] <shey> gobbe: apache2.conf & httpd.conf?
[18:20] <ideaman> sorry
[18:20] <ideaman> http://pastebin.com/SrNzKEUW
[18:20] <gobbe> shey: yea
[18:21] <RoyK> ideaman: does `sudo true` do the same thing as in terms of the delay you're getting?
[18:21] <shey> ok, brb, also.. I dont know if this means anything, but I enabled userdir.
[18:21] <RoyK> shouldn't matter
[18:21] <shey> ok, upping conf's now.
[18:21] <RoyK> shey: but php is disabled by default for userdir
[18:21] <RoyK> shey: check the userdir config file
[18:22] <ideaman> yes
[18:22] <RoyK> ideaman: can you pastebin `strace sudo true`, please?
[18:23] <ideaman> RoyK, so sorry, I have to jump on a webinar with a customer. Will you still be around in 30?
[18:23] <RoyK> probably
[18:24] <ideaman> ok, I'll message here as soon as its over. thx
[18:34] <takamarou> Hi all.  I've got a new ubuntu server 10.10 install going.  I'm adding users using 'groupadd -d <dir> -m <user>'.  For some reason, now when I log in as those users my bash .profile is not being loaded?  Anyone ever seen this before?
[18:35] <Wolfsherz> shouldn't you use adduser before groupadd?
[18:35] <qman__> adduser is the preferred method because it does all the little extra bits like that for you
[18:35] <takamarou> uhh, thats what i meant
[18:35] <takamarou> im using adduser
[18:36] <Wolfsherz> try just using "adduser <username>"...
[18:36] <qman__> adduser does not have those arguments
[18:36] <qman__> see man adduser
[18:36] <qman__> you're probably thinking of useradd
[18:37] <shey> gobbe: http://pastebin.com/nwE98EAL
[18:37]  * RoyK uses useradd :P
[18:38] <fullstop> I wish there were only one...
[18:38] <RoyK> shey: erm - did you check the userdir config?
[18:38] <takamarou> that worked
[18:38] <takamarou> thanks
[18:38] <fullstop> I am always confused as to which one I should use.
[18:38] <RoyK> fullstop: use the one you know
[18:39] <RoyK> useradd -m someuser ; passwd someuser
[18:39] <fullstop> coming from RHEL, it's not always easy.
[18:39] <RoyK> or adduser someuser and go through the interactive Q/A
[18:39] <RoyK> fullstop: well, welcome :)
[18:40] <fullstop> One of them, by default, will create an account which can sudo.
[18:42] <qman__> uh, no
[18:42] <takamarou> really? I don't want that..
[18:42] <takamarou> and it's not useradd
[18:42] <qman__> only the first account created on the system, automatically, will be added to the admin group by default
[18:43] <qman__> the admin group is given sudo privilege
[18:43] <fullstop> qman__: the first user after an install or during the install?
[18:43] <qman__> during
[18:43] <takamarou> so, if it's better to use adduser instead of useradd for creating users... is there something better than usermod for adding users to groups?
[18:43] <qman__> takamarou, adduser
[18:44] <qman__> adduser <username> <groupname>
[18:44] <takamarou> and deluser is good for deleting.. right?  What about adding groups?
[18:44] <shey> RoyK: ammended. http://pastebin.com/3zNXcyTg
[18:44] <qman__> addgroup
[18:45] <qman__> this is all in the manual for adduser, btw
[18:45] <takamarou> boy.. this book I bought sucks!
[18:45] <fullstop> qman__: I just tried w/10.04, and neither creates with admin..
[18:45] <fullstop> takamarou: you should have bought a vacuum cleaner.  Then you'd be happy.
[18:46] <RoyK> shey: ops - check php5.conf
[18:46] <RoyK> there's where it's disabled for userdir
[18:46] <qman__> think of it like this
[18:47] <RoyK> http://pastebin.com/HEdaFAWU
[18:47] <qman__> adduser/addgroup is the automated, easy way
[18:47] <qman__> useradd/groupadd/usermod/etc are the manual way
[18:47] <qman__> both are correct but the latter requires you to make sure all your options are right
[18:48] <takamarou> sounds like I will stick with the automated way :) thanks guys
[18:48] <RoyK> qman__: useradd -m works well on servers - makes a homedir and copies the skeleton there, then it's up to you as the admin to add group membership and other stuff
[18:49] <takamarou> RoyK, but, as I've just learned, that skeleton is buggy.  The bash profile doesn't load.. which pretty much sucks if you're on a server
[18:49] <RoyK> takamarou: works for me (tm)
[18:49] <takamarou> odd..  This is a completely fresh install, too.
[18:49] <RoyK> takamarou: or did you run useradd without the -m flag?
[18:49] <takamarou> no.  useradd with the -m flag
[18:50] <shey> RoyK: I cannot seem to thank you enough!
[18:50] <RoyK> :)
[18:50] <shey> ffs, HOURS!!!!!!! I spent scratching my head.
[18:50] <RoyK> shey: I had a little issue with that userdir+php myself the other day
[18:50] <shey> that should really be in LAMP server how-to
[18:51] <RoyK> imho it's a bit paranoid to disable php for userdir
[18:51] <RoyK> but then, it's way more secure
[18:51] <shey> well, not really if you look at the amateur shell user.
[18:51] <shey> indeed.
[18:52] <shey> again, thanks for your help man... =]
[18:52] <RoyK> with php in userdir, a user can easily create a nasty php script allowing anyone to run anything, so if a local root exploit is found, no, not if, when, well, there goes your server
[18:53] <shey> true
[18:53] <RoyK> shey: np
[18:53] <shey> But I am sole user of this server, so I have no need to disable it.
[18:54] <shey> ok now to slim down Ubuntu Server a bit.. I'm sure I'll be back when it comes to ROR time.
[18:56]  * RoyK wonders wtf some people tend to want to 'slim down' their installs
[18:58] <m_tadeu> still no errors in memtest....most probaly the mem is just fine
[18:59] <m_tadeu> what else can be stoping my server? any tips?
[19:00] <RoyK> m_tadeu: what's happening?
[19:01] <m_tadeu> after a day of not using it, I can't access to it at all...no ssh, no http, no ping :)...no keyboard, no screen, nothing...but the server is on
[19:02] <RoyK> m_tadeu: after rebooting, do you find anything in the logs?
[19:02] <RoyK> m_tadeu: also, if you have a syslog server somewhere, try to configure the box to log there
[19:02] <m_tadeu> nothing relevant on logs...no errors no warning
[19:03] <RoyK> and - if a panic or oops happen, the server won't restart automatically unless you tell it to
[19:03] <RoyK> add vm.panic = 60 and vm.panic_on_oops = 1 to /etc/sysctl.conf
[19:04] <m_tadeu> I never saw a panic yet...but I guess it would be logged, right?
[19:04] <RoyK> run sysctl -p to activate that
[19:04] <RoyK> a panic can't be logged if there is a problem with the disk
[19:04] <RoyK> some panics are logged, but most won't
[19:04] <m_tadeu> I see....that would be solved by remote logging
[19:04] <RoyK> or a network console
[19:04] <RoyK> or serial console
[19:05]  * RoyK sticks to serial consoles for that
[19:05] <RoyK> get a nullmodem cable, plug it into your server and some other box running some serial software, anything, minicom or even screen will do
[19:06] <RoyK> configure linux to use that, and the chance of catching a panic output will be increased by a wee bit
[19:07] <m_tadeu> hehe...a wee bit :)
[19:07] <m_tadeu> I'm gonna let memtest run for a little while...guess 12 passes is quite enough...then I'll start on that
[19:08] <RoyK> in the meantime, get a serial cable :P
[19:08] <RoyK> if you haven't found any errors after 12 passes, I guess it might be something else
[19:09] <m_tadeu> if it's a problem in the disk, I would get it if I stress it
[19:10] <RoyK> btw, is autosleep or similar cpu powersaving stuff enabled in the bios?
[19:11] <RoyK> that may lead to headaches
[19:11] <m_tadeu> good question...let me check that
[19:16] <fluvvell> anyone know how to assign a blkid to a partition that seems to not have one?
[19:18] <m_tadeu> I tought every partition should have one
[19:18] <m_tadeu> are you checking as sudo?
[19:18] <RoAkSoAx> kirkland: were you able to test PowerNap?
[19:19] <fluvvell> my raid1 array has dropped one of its members, and I build by blkid.  Yes I thought they all should have them too. Yes to sudo
[19:19] <patdk-wk> only partitions that have crap have id's :)
[19:19] <patdk-wk> a partition that isn't formatted with something (ext3/4, lvm, xfs, ...) won't have an id
[19:20] <fluvvell> patdk-wk, its not formatted ist a linux_raid_member  - or should be
[19:20] <fluvvell> *its
[19:20] <patdk-wk> well, raid counts too
[19:20] <RoyK> patdk-wk: man blkid
[19:21] <fluvvell> Royk, blkid reads and displays the id, are you aware of anything that can set it ?
[19:21] <patdk-wk> royk, I dunno why you said that to me
[19:22] <patdk-wk> if blkid doesn't give into for it, it means blkid doesn't know what it is
[19:22] <RoyK> patdk-wk: sorry - nick fsckup
[19:22] <fluvvell> from the manual:  blkid  has two main forms of operation: either searching for a device with a specific NAME=value pair, or displaying NAME=value       pairs for one or more devices.
[19:22] <patdk-wk> if blkid doesn't know what it is, it's probably corrupt
[19:22] <patdk-wk> or not formatted :)
[19:23] <RoyK> hm.. shouldn't fdisk create a blkid?
[19:23] <patdk-wk> no
[19:23] <RoyK> if it's just filesystems, using blkid for md replacements won't work
[19:24] <patdk-wk> it's not just filesystems, basically it's any structure
[19:24] <patdk-wk> filesystems, swap, lvm, raid, ...... all create a structure on the disk, therefor have blkid support
[19:25] <patdk-wk> fdisk doesn't make structures (unless you count those 40 or so bytes in the first sectore)
[19:25] <RoyK> sure, but if two drives are mirrored and one dies, how can one make a blkid on the new replacement drive?
[19:25] <fluvvell> output of fdisk -lu   http://pastebin.com/ia9TB5XH
[19:26] <patdk-wk> royk, each md partition would have an blkid
[19:26] <fluvvell> RoyK, sort of my question really.
[19:26] <patdk-wk> in my case:
[19:26] <RoyK> fluvvell: can't you just replace the drive with the device name?
[19:26] <patdk-wk> /dev/sda: TYPE="isw_raid_member"
[19:26] <patdk-wk> /dev/sdb: TYPE="isw_raid_member"
[19:26] <patdk-wk> /dev/mapper/isw_bgebcdgcaf_Volume01: UUID="2ef294e7-5015-4b60-86d4-683eb22ddc61" TYPE="ext4"
[19:26] <patdk-wk> /dev/mapper/isw_bgebcdgcaf_Volume05: UUID="53681bc0-51fc-487d-b7cd-8639a9e6e5c3" TYPE="ext4"
[19:26] <patdk-wk> but that isn't a md raid, it's a dm raid
[19:27] <fluvvell> RoyK, I guess.  My /etc/mdadm/mdadm.conf  has  ARRAY /dev/md0 level=raid1 num-devices=2 UUID=6d4c2a80:85d76dbd:e368bf24:bd0fce41
[19:27] <RoyK> fluvvell: md should be detected by the kernel
[19:27] <kirkland> RoAkSoAx: on my todo list for today;  i'm going to test it with a few different machines
[19:27] <RoyK> without need for a config file
[19:28] <fluvvell> RoyK, Ive read that. Was just making sure I guess.
[19:29] <RoyK> fluvvell: how does your /proc/mdstat look?
[19:30] <fluvvell> http://pastebin.com/DRnhhLyr
[19:30] <RoyK> fluvvell: doesn't look like you've replaced sdb - have you?
[19:31] <fluvvell> No, its a brand new drive - brand new server I installed 10 December
[19:31] <fluvvell> It just seemed to not start the array when I rebooted it last night
[19:31] <fluvvell> sorry: started in degraded
[19:32] <RoyK> fluvvell: well, seems sdb has issues anyway, since md has offlined it
[19:32] <fluvvell> RoyK, Yeah. <sigh>
[19:32] <RoyK> fluvvell: check the logs to see if there have been i/o errors on that drive
[19:34]  * RoyK points fluvvell to http://www.howtoforge.com/replacing_hard_disks_in_a_raid1_array
[19:36] <fluvvell> RoyK, I did, no I/o errors.  There is  "Jan 20 20:02:39 wilma kernel: [   14.173301] raid1: raid set md0 active with 1 out of 2 mirrors"
[19:36] <fluvvell> when I rebooted
[19:37] <RoyK> fluvvell: mdadm --manage /dev/md0 --add /dev/sdb1
[19:38] <RoyK> if the partitions are setup like I would do it
[19:38] <RoyK> that is
[19:39] <fluvvell> RoyK, thanks,       [>....................]  recovery =  0.7% (762944/102398208) finish=15.5min speed=108992K/sec
[19:41] <m_tadeu> RoyK: I have power management set as "active power management"(whatever this means I'm trying to check). should I allow the OS deal with it? or disable it at all?
[19:43] <RoyK> m_tadeu: just disable it
[19:44] <m_tadeu> crap...can't disable it
[19:46] <m_tadeu> I it has a "max performance" option...guess this should be the closest to disabling
[19:47] <RoyK> I've seen this happen with win2k8 boxes - they died while idle
[19:47] <RoyK> haven't seen it on linux machines, though
[19:47] <RoyK> but then, the similar 24-core machines we have on linux rarely sleep :P
[19:48] <m_tadeu> damn thing is that I have to wait 2 days for testing
[19:48] <RoyK> why?
[19:49] <m_tadeu> because it takes more then a day idle to freeze
[19:50] <m_tadeu> well I don't know exactly how much time it takes
[19:50] <RoyK> well, just wait it out :)
[19:50] <RoyK> debugging such errors is a PITA
[19:50] <RoyK> beleive me, I know :)
[19:50] <m_tadeu> :)
[19:51] <m_tadeu> well...gonna try this out and hope for the best
[19:51] <m_tadeu> thanx for the info
[19:58] <elb0w> If I need to execute commands on startup should I make chkconfig startup script? or the remote?
[19:59] <RoyK> an init script should do, or even rc.local
[20:05] <lieuwe> i just pluged in an usb-stick that i want to automount on boot, how do i do this?
[20:10] <lieuwe> halp?
[20:11] <RoyK> lieuwe: http://tinyurl.com/4g9nkj5
[20:11] <lieuwe> RoyK: i'm doing that, but i kinda need to hurry, so i'd like answers as quickly as possible ;P
[20:13] <qman__> add it to fstab with 'auto' option
[20:13] <mrothhh> hi I want to offer ubuntu desktop on  20 dumb terminals
[20:13] <mrothhh> is their any thing I nee dto do or a product
[20:14] <qman__> mrothhh, you want LTSP, it should be on the ubuntu alternate disc
[20:14] <RoyK> lieuwe: a quick google is often faster than waiting for someone on irc.....
[20:14] <lieuwe> RoyK: i'm googling too,
[20:14] <mrothhh> I have 20 dumb termnals that have rdp client
[20:14] <mrothhh> HP dumb terminals
[20:14] <lieuwe> RoyK: parralel processing :P
[20:14] <mrothhh> oh ltsp
[20:15] <RoyK> mrothhh: install ltsp
[20:16] <qman__> I only mentioned alternate disc because it has (had?) an easy one-step installer
[20:16] <qman__> I haven't used it since 8.04 though
[20:18] <lieuwe> is it possible to get cp to show a progress bar?
[20:18] <RoyK> afaik the alternate disc exists on 10.04 as well
[20:18] <pmatulis> lieuwe: yes, rsync
[20:18] <RoyK> lieuwe: use rsync -P
[20:19] <pmatulis> RoyK: slow on the draw there
[20:19] <Tellmarch> rsync.... it shows a progress bar per file right? not a global one?
[20:19]  * RoyK kicks pmatulis in the leg
[20:20] <RoyK> Tellmarch: per file, yes, it's recursing, no chance to show a global one with that logic
[20:20] <Tellmarch> well, windows does it (lol)
[20:21] <RoyK> windows does it, but it uses a lot of time to count files before it starts
[20:21] <RoyK> and windows can't continue from an interruption
[20:21] <RoyK> whereas rsync -P can
[20:22] <Tellmarch> i actually never thought about using rsync instead of cp...
[20:22] <RoyK> even without --partial (part of -P), rsync will continue, but not for single files
[20:22] <Tellmarch> i've been using it for years instead of scp, but for cp, too strange an idea
[20:22] <Tellmarch> lol
[20:32] <jcastro> zul: your link to your cobbler blog post is broken on planet ubuntu
[20:33] <elb0w> What does ubuntu use instead of chkconfig
[20:33] <elb0w> ?
[20:33] <zul> jcastro: yes i know...me and wordpress dont get along
[20:36] <RoyK> I and wordpress get along well :)
[20:37] <RoyK> lenios: ipv6 didn't work?
[20:44] <lenios> RoyK, it's my autoidentify that is taking too much time
[20:48] <pmatulis> elb0w: chkconfig, update-rc.d, editing upstart job filenames
[20:49] <pmatulis> elb0w: there is no central good-working tool right now (mostly due to upstart)
[20:49] <RoyK> lenios: what is autoidentify?
[20:50] <lenios> xchat has my identify password, and is taking care of identify for me on connect
[20:50] <lenios> a little late, that is
[20:53] <RoyK> lenios: didn't you say it was sudo that spent time?
[20:53] <lenios> ?
[20:54] <RoyK> lenios: sorry, I was thinking of something ideaman was saying
[20:56] <RoyK> ideaman: ping
[21:28] <cr3> hi folks, is there a recommended way to automatically provision a configuration for ec2 instances? for example, I'd like to have my own packages and some tweaks to the configuration in ec2 and enable other folks to reproduce the same environment easily
[21:30] <cemc> I have a LTSP+dbus problem un 10.04. there's an update for dbus and dbus-x11 package. I'm running chroot /opt/ltsp/i386 apt-get upgrade, and it fails with: http://pastebin.ubuntu.com/556323/ . any ideas?
[22:38] <mrothhh> I can't seem to build a 200 dollar 4 bay ubuntu nas
[22:38] <mrothhh> from parts at newegg or microcenter
[23:01] <gyppo> hello?
[23:02] <gyppo> does anyone know of a way to send mail from a server, without using the isp's relay, when the isp blocks port 25~?
[23:02] <estacion03> can anyone help me change the uri of my printer?
[23:04] <gyppo> quiet in here...
[23:04] <gyppo> I wouldn't know where to start, estaction03, sorry
[23:05] <estacion03> can anyone help me change the uri of my printer?
[23:08] <estacion03> I have one ubuntu comp hooked up to my pc with a printer and it works fine my other unbuntu (a carbon copy) will not print the uri are different any thoughts
[23:11] <RoAkSoAx> kirk/win 9
[23:11] <RoAkSoAx> klik
[23:11] <RoAkSoAx> lol
[23:13] <estacion03> I have one ubuntu comp hooked up to my pc with a printer and it works fine my other unbuntu (a carbon copy) will not print the uri are different any thoughts
[23:18] <progre55> hi guys. I have some servers behind a corporate firewall in the office, but now it's 12am and I'm at home.. the servers are connected to another database server in the cloud, and I need to disconnect them from there, but dont have access to the office servers. how do I block those IPs? any suggestions, please?
[23:18] <progre55> they are connected to port 2
[23:19] <progre55> 22*
[23:22] <thesheff17> progre55 are they just connected through ssh?
[23:23] <estacion03> can anyone help with printing networking
[23:23] <progre55> thesheff17: yep.. tunneled through ssh to port 3306
[23:24] <thesheff17> progre55: if you ps aux | grep ssh you will see the process running the tunnel and you should just be able to do kill -9 pid.
[23:24] <thesheff17> and it will disconnect them.
[23:24] <progre55> thesheff17: but they are using "autossh" which would automatically reconnect..
[23:25] <thesheff17> I would look into ufw which is a simple firewall program then and you can just deny the IP's
[23:25] <progre55> I know the IP address.. could I just block the IP?
[23:25] <thesheff17> http://ubuntuforums.org/showthread.php?t=823741
[23:26] <progre55> oki, thanks a lot
[23:26] <progre55> let me have a look
[23:26] <thesheff17> sudo ufw deny from <ip address>
[23:29] <thesheff17> estacion03: I don't know much about printing....you may want to check out the #ubuntu room.  That is where many of the desktop users hang out that may know more about printing.
[23:30] <estacion03> thesheff17, so this wouldnt be a server issue since this problem is over a lan
[23:30] <estacion03> ?
[23:32] <thesheff17> estacion03: well I'm not sure about the printing aspect....I guess if you are using ubuntu-server to print you can ask here but I'm not sure how many people here print with ubuntu-server.  I would check /var/logs and the services used to print and start from there...Also search google.
[23:33] <estacion03> thesheff17, thats why im here my searches have been unfruitful
[23:33] <thesheff17> have you looked here: https://help.ubuntu.com/community/Printers
[23:33] <thesheff17> there is a Sharing printers section.