[00:02] <defn> is there a better way to get a newer version of apache in lucid than building from source?  im being exploited by a proxy problem that exists in the ubuntu package
[00:03] <Pici> defn: You may be able to find a ppa that has lucid builds for newer versions.
[00:04] <Pici> !ppa
[00:09] <wizardslovak> hello people
[00:10] <wizardslovak> i have phpbb instlled on server , it worked but now its not
[00:11] <defn> Pici:
[00:11] <defn> thank you
[00:11] <wizardslovak> anyone?
[00:13] <wotwot> wizardslovak, you might find an answer in your logs
[00:14] <wizardslovak> yea well dont know which ones
[00:14] <piercedwater> Can anyone here help me with large storage options? (8+ TB)
[00:56] <benedikt> If I have one machine that I want to "duplicate", what do i have to copy to the new machine - roughly speaking
[01:27] <SpamapS> benedikt: back in the day, before I used config management systems.. I'd just boot to single user mode and rsync from the old box to the new one
[01:29] <quentusrex> Anyone else aware of a new issue with kvm and guests hanging on 'Booting from Hard Disk' ?
[01:29] <quentusrex> the server is unchagned and not updated, the guest is the latest packages.
[01:30] <benedikt> SpamapS: thats what im going to do., what should I do about /dev and /proc?
[01:31] <SpamapS> benedikt: skip them
[01:31] <SpamapS> benedikt: skip /sys too
[01:31] <quentusrex> SpamapS, what config management system are you using?
[01:31] <SpamapS> quentusrex: right now, none, because I'm not caring for any clusters. But I used to have my own, and then I did some fiddling w/ puppet. :)
[01:32] <quentusrex> aah
[01:32] <quentusrex> I've used git for the configs, and hooks to deploy
[01:32] <benedikt> SpamapS: and when i boot the new computer?
[01:32] <benedikt> nevermind, ill google and fiond out
[01:39] <SpamapS> benedikt: well you'll need to rename the new one and change the IP probably
[02:06] <qman__> benedikt, dd if=/dev/sda of=/dev/sdb
[02:07] <qman__> literally, that's it
[02:07] <qman__> obviously if you're running on the same network you'll want to change the hostname and IP
[02:07] <qman__> but there are no special tricks needed
[02:08] <qman__> ubuntu will 'just work' no matter what hardware you pop your install into
[02:08] <twb> qman__: you'd want to blow away /etc/udev/rules.d/*persistent*net*
[02:08] <twb> qman__: otherwise the new host would try to raise eth0 and eth1, but only have interfaces eth2 and eth3
[02:09] <qman__> yeah
[02:09] <twb> If the new disk is smaller, you'd also have to wrestle that.  And of course the dd should be done while sda is offline (e.g. boot a live cd).
[02:09] <twb> Other than that, it should be fine
[02:54] <AdamDV> What openssl command can I use to generate a public and private key?
[03:05] <twb> SSL doesn't have public keys, it has certificates.
[03:06] <twb> (Technically the public key is included in the private key and cert files.)
[03:06] <jmarsden> ahem... openssh genrsa   may be what AdamDV was looking for?
[03:06] <jmarsden> Make that openssl genrsa
[03:06] <twb> I use gnuTLS's certtool(1) command, because it tends not to implement esoteric commands that confuse me.
[03:06] <AdamDV> thanks jmarsden
[03:07] <twb> e.g. it defaults to RSA, so you don't need to pick the algorithm
[03:08] <jmarsden> AdamDV: You can do man openssl to see all the commands, or  openssl list-standard-commands to just get a straight list of them without explanations.
[03:32] <zul> SpamapS: tomorrow
[03:48] <RudyValencia> How do I set up a PPTP server on Ubuntu 10.04 LTS server version?
[03:52] <highvoltage> RudyValencia: this might help: https://wiki.ubuntu.com/VPN
[03:52] <RudyValencia> ty
[03:54] <jmarsden> RudyValencia: Try http://www.ubuntugeek.com/howto-pptp-vpn-server-with-ubuntu-10-04-lucid-lynx.html for one idea :)
[03:54] <RudyValencia> I think that was what I did
[03:55] <RudyValencia> https://wiki.ubuntu.com/VPN is for VPN clients
[03:55] <RudyValencia> I'm trying to serve VPN
[03:55] <RudyValencia> and that's not working
[03:56] <twb> Don't use PPtP unless you absolutely have to; it uses insecure crypto algorithms (or is is handshaking?)
[03:57] <RudyValencia> L2TP would require generating a security key
[03:57] <twb> If you have a choice you should be looking at openvpn or ipsec+l2tp
[03:57] <RudyValencia> I don't have IPSec set up on this network
[03:59] <RudyValencia> I don't know if Windows supports OpenVPN connections
[03:59] <RudyValencia> (this is so I can access my Windows shares remotely, actually)
[04:00] <jmarsden> RudyValencia: See http://openvpn.se/
[04:02] <jmarsden> But if this is just for one remote PC, ssh to it and set up to tunnel SMB: http://pigtail.net/LRP/printsrv/tunnel-smb.html
[04:02] <RudyValencia> The remote PC won't have SSH tunneling as it runs Windows
[04:04] <greppy> RudyValencia: openvpn works with windows clients.
[04:05] <RudyValencia> I figured out the problem
[04:05] <RudyValencia> no ms-dns entry in pptpd-options
[04:07] <twb> ppp openvpn you can do with just a shared secret
[04:07] <RudyValencia> hey
[04:07] <RudyValencia> oops
[04:08] <RudyValencia> got it
[04:08] <RudyValencia> I tested my VPN using my android phone
[04:09] <RudyValencia> it connects fine
[04:12] <jmarsden> RudyValencia: I have some Windows machines that have SSH, Cygwin has openssh and it works fine.
[04:12] <RudyValencia> ah
[04:12] <jmarsden> But if you have a working VPN setup now, cool, use it :)
[04:12] <RudyValencia> I've already got the PPTP setup working. Just realized I didn't declare a DNS server.
[06:56] <piercedwater> What is the status on USB 3.0 support?
[06:57] <piercedwater> As far as pci-e cards, etc
[07:00] <piercedwater> Hey guys, did you know the forums are a great place to search before asking a question
[07:01] <lifeless> yes, yes we do.
[08:14] <j0nr> hello...I have ubuntu 8,04 LTS server installed. I installed Roundcube with an apt-get... I don't think it is the latest version, based on having seen it run on ubuntu 10,04. I guess this is due to being on 8,04. Is it possible to force the newer version to install?
[08:43] <\sh> nijaba: I replied on your comment :)
[08:45] <nijaba> \sh: looking :)
[08:47] <nijaba> \sh: it's a fairly simple resseler agreement.  Don't see how this could be called sponsoring...
[08:58] <\sh> nijaba: as said, I think the words in the serverwatch article is not clear enough....I read about the reselling agreement, but the first sentence gave me a wrong feeling...:)
[09:04] <\sh> patdk-lap: some news on the bonding problem from yesterday, I think it's not upstart which is bugging me, but something which looks like a regression regarding the bonding drivers and/or the way how the bonding driver is loaded in if-pre-up.d/ifenslave , investigating more now
[09:32] <chrismat> How do you grep for non ascii characters in a text file?
[09:33] <chrismat> Ie I'm looking for high ascii values like swedish character åäö
[09:55] <nijaba> \sh: agreed that the introductory statement from serverwatch can be misread "involves some money"...  bleh.
[09:57] <\sh> nijaba: you could ask them to remove this sentence ;)
[09:58] <nijaba> \sh: and make a journalist angry at us...  not sure I want to do that :P
[10:00] <\sh> nijaba: well, if this article gives a bad taste because of some wrongly chosen wording, I would ask for a change or clarification
[10:01] <nijaba> \sh: truth is, I trust our PR department to know better than I
[10:03] <\sh> nijaba: sometimes ;)
[10:14] <novitololo> I've a question: apache2 I suppose means apache 2.0 right?
[10:14] <iclebyte-work> novitololo, correct
[10:14] <novitololo> ok thanks ;)
[10:16] <novitololo> Is it from the apache2 that the httpd.conf is empty?
[10:46] <sergevn> novitololo: yes, it resides in apache2.conf or /etc/apache2/sites-enabled/ what you are looking for
[10:46] <novitololo> sergevn: nothing thanks, I was just surprised when I opened this computer
[10:46] <novitololo> :)
[10:47] <sergevn> novitololo: iirc httpd.conf isnt used anymore
[10:47] <novitololo> I see, thank you :)
[10:48] <sergevn> novitololo: np
[11:22] <DiagonalArg> Hey folks!  Installed Ubuntu over a previous install, which was on a RAID1. Only option on Alternate Installer was "Erase Drive".  Took 30 hours for 168G???  WTF??
[11:23] <DiagonalArg> If it's zeroing out the drive, that's 3.2MB/s (counting writing to each drive, separately).  Any thoughts?
[11:23] <DiagonalArg> I would have hoped there would be a way to do this by just deleting the files, but not zeroing the drive (assuming that's what it was doing).
[11:25] <patdk-lap> guess he wants an out of sync raid1
[11:27] <jpds> patdk-lap: Potentially.
[11:27] <DiagonalArg> Got lost there for a minute.  Anyway, I'm back if someone has any thoughts
[11:29] <patdk-lap> guess giagonalarg wants an out of sync raid1
[11:29] <patdk-lap> it has to copy everything on drive 1 to drive 2, for raid1
[11:29] <patdk-lap> why it took your system 30hours for 160gigs, dunno, mine is much quicker
[11:30] <DiagonalArg> I don't think it was a sync issue.  The machine ran fine, I shut down, and ran the installer again (this time with 64 bit, the previous had been 32)(
[11:31] <DiagonalArg> When I ran the installer, the only option it gave me, if I wanted to install to the same place as the previous system, was "erase drive"
[11:32] <DiagonalArg> I'll add that after that 30 hours, the install went fine and now I've got 64-bit Ubuntu running fine.
[11:33] <DiagonalArg> Also, I've benched both disks.  They're SATA-I
[12:08] <azizLIGHTS> hi, should i make pptp vpn server or l2tp vpn server?
[12:10] <azizLIGHTS> any good guide for this to use on amazon ec2
[12:11] <_ruben> depends on your needs/wishes/etc
[12:23] <azizLIGHTS> i wish to use vpn for browsing in a firewalled scenario
[12:23] <azizLIGHTS> client is windows computers
[12:24] <azizLIGHTS> (no ports open)
[12:26] <azizLIGHTS> what is best for me? l2tp or openvpn or pptp
[12:30] <azizLIGHTS> what are these certificates adn why u need them for l2tp vpn
[12:41] <azizLIGHTS> can someone explain me what is "certificate authority" and why u need for set up of l2tp vpn server. im tyrying to understand concepts inside a tutorial but it doesnt explain
[12:42] <pmatulis> never heard of 12tp, sorry
[12:42] <pmatulis> use openvpn maybe, it's well known
[12:44] <azizLIGHTS> L2tp over ipsec
[12:52] <nimrod10> azizLIGHTS, will you be able to use l2tp if there are no ports open ? best case that I can think of , if the people administering the firewall allow port 443 outbound you could use openvpn
[12:53] <azizLIGHTS> the client is firewalled, the server wont be (i dont think.... im going to put latest ubuntu server on amazon ec2)
[14:05] <RoAkSoAx> morning all
[14:11] <failover> Hi, it's a normal behavior, quotacheck run after every reboot ?
[14:12] <failover> the tool starts at the end of boot process and take about 5 minutes before gimme the prompt
[14:41] <raubvogel> Who should own the cert and key used by postfix and dovecot in ubuntu? Which group should I also use?
[14:42] <aliverius>  am running a vm qith qemu at startup. where should it reside? somehwere in /var?
[15:32] <noobster> good mourning all. I am having an error I have never seen before. I woke up today to a..  init: ureadahead-other main process (912) terminated with status 4
[15:32] <noobster> I get about 4 of those, and the srv will not boot
[15:32] <noobster> !!
[15:34] <jmarsden> You *woke up* to that?  That'll teach you to sleep on the server, instead of on a bed :)
[15:35] <noobster> lmao
[15:35] <noobster> yea
[15:35] <noobster> kinda freeked out, has this ever happened to anyone?
[15:36] <jmarsden> I think I've seen that kind of message during boot after a power loss on a workstation (no UPS), but it didn't prevent boot.  But that's just from (old and not always reliable) memory.
[15:36] <jmarsden> Was that the last thing on screen before the server hung?
[15:37] <noobster> ctrl + alt + delete reboot the machine as expected, so its not frozen
[15:37] <noobster> I was on it @ about 3AM last night moving smb.conf and bind9 files and it seed to just crash
[15:37] <jmarsden> It may be doing a filesystem check that takes time, are you just being impatient with it?
[15:37] <noobster> I tried to ssh in and got an error denied
[15:38] <noobster> well it was still offline when I got up @ 9
[15:38] <noobster> or online w/ connection refused
[15:40] <jmarsden> OK... so I'd guess there really was some sort of crash or issue last night, but... after you rebooted it at 9, how far did it get through the boot sequence and how long did you leave it along before hitting ctrl-alt-del ?
[15:40] <noobster> jmarsden, about an hour
[15:40] <jmarsden> Unless you have a multi-terabyte filesystem somewhere, that should be plenty.  Hmmm.
[15:41] <jmarsden> If you are local to the server, can you boot it into single user mode?
[15:41] <noobster> i do have a multi-terabyte /home
[15:41] <jmarsden> Ah... well then maybe you need to wait several hours after a crash.
[15:42] <jmarsden> See if you can boot single user and manually check the filesystems.
[15:42] <noobster> jmarsden, onsite now, tried to hit esc but did not bring up grub
[15:42] <RoAkSoAx> zul: have you tested sheepdog yet?
[15:42] <zul> RoAkSoAx: not yet
[15:43] <noobster> is grub trap another key other than esc in 10+
[15:43] <zul> RoAkSoAx: problems?
[15:43] <jmarsden> noobster: I think I usually hold down left-shift, rather than pressing esc, to get grub boot prompt?
[15:43] <RoAkSoAx> zul: nah!! just was going through the documentation and seems pretty interesting
[15:44] <zul> RoAkSoAx: i dont have the hardware for it but its needed for a feature in openstack
[15:44] <noobster> single user mode == "recover mode"
[15:44] <noobster> ?
[15:44] <jmarsden> noobster: Yes.
[15:44] <jmarsden> recovery mode, with a y, I think it is called?
[15:44] <RoAkSoAx> zul: yeah but I'm presumming that it could be tested in VM's
[15:44] <noobster> same error
[15:45] <noobster> mountall: Event failed      init: ureadahead-other terminated with status 4
[15:45] <jmarsden> noobster: Doesn't sound good.  Can you boot from CD and check the filesystems that way?
[15:46] <noobster> mounted-tmp terminated with status 127
[15:46] <noobster> yea I am downloading ubuntu 10.04 desktop 64 now
[15:46] <noobster> 40 mins left
[15:47] <jmarsden> noobster: Gut instinct is that you might have some sort of disk-related hardware issue.  40 mins?  You;re downloading it via your cellphone? :)
[15:47] <noobster> lol eveyone in pounding out T1 since the server is down
[15:48] <noobster> is there a trap to see if fsck is running?
[15:48] <jmarsden> Ugh.  I have 10Mbit/sec here at home and 35Mbit/sec at work, so ... ISOs download pretty fast :)
[15:48] <jmarsden> If you can get to a second console window (alt-F2 or whatever) you can see if you can log in there and poke around?
[15:48] <noobster> i have 20 @ home, 1.5 @ this office
[15:49] <compdoc> my comcast can d/l an OS DVD pretty fast, but only if I use bittorent
[15:49] <noobster> nope, just a flashing _
[15:50] <noobster> ok so I will stop chattering the board until I have the cd booted
[15:50] <jmarsden> noobster: So... why aren't you carrying your collection of .ISOs on a usb key or on a pile of CD-Rs?  OK... so it hasn't got to the point where it runs all the gettys...
[15:50] <jmarsden> noobster: OK.  I'll probably be AFK (breakfast and then on my way to work) by then, but others here can probably help you out if you need further assistance.
[15:51] <noobster> ty 4 the help !
[15:51] <jmarsden> You're welcome.
[15:52] <RoAkSoAx> zul: have you been through the documentation yet?
[15:53] <zul> RoAkSoAx: nope i just packaged it
[15:53] <RoAkSoAx> ok
[15:54] <RoAkSoAx> zul: btw... the archives and PPA's do not accept tar.xz tarballs right?
[15:55] <zul> RoAkSoAx: doubt it...but i have never tried it
[15:55] <RoAkSoAx> zul: yeah at least PPA's dont'. Who do you think knows?
[15:56] <zul> RoAkSoAx: umm...an archive admin maybe :)
[15:56] <RoAkSoAx> loldumb me :P
[15:56] <bigjools> they don't
[15:57] <jmarsden> They (archive admins) don't know?  Or they (tar.xz files) don't work? :)  I'm guessing the latter, but either interpretation is possible.
[15:57] <bigjools> heh, the latter
[15:57] <jmarsden> :)
[16:08] <RoAkSoAx> bigjools: do you know if there are any plans to support them?
[16:08] <bigjools> no plans
[16:08] <RoAkSoAx> bigjools: what if requested? Would that be considered?
[16:09] <bigjools> of course, it would need to go via the right channel to get any weight though
[16:09] <bigjools> unless someone sends me a patch :)
[16:10] <RoAkSoAx> bigjools: any branch in specific that I should be looking at?
[16:11] <jpds> RoAkSoAx: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=556407
[16:12] <bigjools> RoAkSoAx: https://code.launchpad.net/~launchpad-pqm/launchpad/devel
[16:13] <bigjools> lib/lp/archiveuploader/ contains all the upload processing
[16:13] <RoAkSoAx> jpds: cool ;)
[16:13] <RoAkSoAx> bigjools: awesome! thanks
[16:13] <bigjools> np
[16:48] <zul> SpamapS: ping when you are around?
[16:52] <SpamapS> zul: pong , now I'm found!
[16:53] <zul> SpamapS: so handreak, you are good with me reviewing it now?
[16:54] <zul> handbreak even
[16:56] <SpamapS> handlersocket
[16:56] <SpamapS> zul: yes
[16:57] <zul> SpamapS: k ill have a firstpass this afternoon
[16:57] <SpamapS> zul: sweeeeeet
[17:00] <Lars_G> Hi alll
[17:00] <Lars_G> This is more a general question about ubuntu but if someone must know it's the server guys
[17:00] <Lars_G> I'm trying to install ubuntu over the network into a xen virtual machine
[17:01] <Lars_G> I downloaded the xen netboot image and made it accessible to the xen server over the web, and it boots
[17:01] <Lars_G> but I have an 10.04 LTS DVD I would like to use as my install source to avoid the veeeery slow internet connection ussage
[17:01] <compdoc> reasonable
[17:01] <Lars_G> I've mounted the dvd in a machine and made it available thorough http (at my address: http://<ip>/ubu3
[17:02] <Lars_G> But the installer wont recognize it as a valid mirror
[17:02] <Lars_G> wether I enter the url as http://<ip> and the diretory as either /ubu3/ or /ubu3/ubuntu
[17:02] <compdoc> cant just place the dvd into a drive on the xen machine?
[17:02] <Lars_G> what can I do to have it recognized as a "valid" mirror?
[17:02] <Lars_G> compdoc: nope, no dvd in the machine, and the installer offers no dvd option
[17:02] <compdoc> I think its best to mount the iso file
[17:03] <Lars_G> So is there no way to have the dvd shared over http to be recognized as a repo?
[17:04] <compdoc> xen, xenserver, kvm, and the others have the ability to install from an iso thats stored on the lan
[17:04] <compdoc> it mounts it as if it was in the dvd drive
[17:04] <Lars_G> So I'd rather rip the iso and do so over nfs or cifs (xenserver btw)
[17:04] <Lars_G> sigh
[17:05] <Lars_G> only detail last time I tried a dvd install the template didn't locate the right vmlinuz file in the dvd
[17:05] <Lars_G> so it couldn't install
[17:05] <Lars_G> is there NO way to make it work with the dvd shared over http? the process is halfway there
[17:05] <compdoc> OS template for the VM?
[17:06] <compdoc> you mean selecting the OS type?
[17:06] <Lars_G> XenServer installs PV machines only from templates they provide
[17:06] <compdoc> or what do you mean by template?
[17:06] <compdoc> yes
[17:06] <Lars_G> I could use "other OS" but it'll install in HVM and I want this machine to be PV
[17:07] <compdoc> you want a tip? no matter what OS youre installing, select Windows 2003 as the OS template, and it'll run
[17:07] <Lars_G> Because it's HVM same as the Other OS option....
[17:07] <compdoc> I still use xenserver for one customer, but I'm all kvm otherwise
[17:07] <Lars_G> And installs from a DVD instead of booting a kernel directly
[17:08] <compdoc> I like xenserver, but I need more control over networking, etc
[17:08] <Lars_G> I agree
[17:08] <Lars_G> I chose it for a very specific reason
[17:18] <piercedwater> What is the name of the ubuntu server usb probe tools used to monitor usb speed?
[17:27] <piercedwater> ok so i found hwinfo
[17:27] <piercedwater> but it only gives me the port speed while not in use..im trying to find what the actual speed is?
[17:29] <compdoc> I think sometimes the speed is listed in dmesg
[17:30] <compdoc> oh, usb - not sure
[17:30] <noobster> ok, got ubuntu desktop cd booted up on server, and can see all the drives. Still getting the mountall: Event failed      init: ureadahead-other terminated with status 4 error when booting off drives. what is my next step? I hope I dont have to reinstall.
[17:31] <noobster> jmarsden,
[17:31] <noobster> mount -t ext3 /sda1 /mntsda1
[17:31] <noobster> & 2
[17:32] <noobster> no probles & can see the content of sda /etc
[17:35] <smoser> what package would be the proper place to put a ttyS0.conf ?
[17:35] <smoser> basically to run a getty on ttyS0
[17:36] <smoser> upstart owns the other ttyN.conf files.
[17:46] <soren> smoser: I think I have a bug open about that somewhere.
[17:47] <soren> smoser: Way, way old. Like two years or something.
[17:47] <soren> smoser: Well, it was for hvc0, but the idea is the same.
[17:48] <soren> smoser: https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/372864 but apparantly the talk about where to put the file was on IRC :(
[17:49] <smoser> soren, youdont have irc logs ?
[17:49] <soren> smoser: I'd say it belongs in whatever package the other ones are in (upstart currently).
[17:49] <soren> smoser: I do.
[17:49] <soren> Gigabytes worth.
[17:50] <soren> I don't feel like wading through them, though. The gist of it is that Scott didn't want it in the upstart package.
[17:50] <soren> I'm not sure I ever got a good reason out of him.
[17:50] <soren> Anyways, ttyS0 is harder than hvc0.
[17:51] <soren> Let me see if I can find the bug..
[17:51] <smoser> why would ttyS0 be  more difficult ?
[17:52] <soren> I'm looking for the bug reference.
[17:52] <soren> The short version is that hvc0 is specifically for a console.
[17:52] <soren> ttyS0 is a general purpose serial port.
[17:53] <soren> Some people got really upset that network-manager tried to probe for modems on all serial ports, because they had stuff connected to said serial ports that got into a rather unfortunate state if random stuff was sent to them.
[17:53] <smoser> i dont know... your suggestion that "hvc0 is specifically for a console" is not really true.
[17:54] <soren> Same will happen if we stick a getty on ttyS0.
[17:54] <soren> How do you figure that?
[17:54] <soren> You know what vc stands for, right?
[17:54] <smoser> maybe you use hvc0 for other things. and having the guest write random data to it will break those things.
[17:54] <smoser> its a general two way communication mechanism.
[17:55] <smoser> but, yes, i'm mostly just arguing. i see the point.
[17:55] <smoser> the bug says that both kvm and xen have /dev/hvc0
[17:55] <smoser> but i do not see such a think in kvm
[17:56] <soren> Maybe you need to specify it explicitly these days.
[17:57] <soren> $ kvm --help | grep vc
[17:57] <soren> -chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]
[17:57] <RoAkSoAx> hggdh: did you end up testing powernap again?
[17:59] <soren> smoser: I really do wish we could stick a getty on ttyS0 by default.
[18:02] <hggdh> RoAkSoAx: no, not yet -- got busy with other tests. Will check on it now
[18:03] <RoAkSoAx> hggdh: ok cool ;)
[18:04] <soren> smoser: My google-fu is failing me.
[18:04] <smoser> try bing
[18:04] <smoser> :)
[18:04] <soren> hah
[18:04] <soren> smoser: Someone complained. I'm not sure what the outcome was.
[18:06] <zul> lol
[18:16] <noobster> hi all
[18:17] <smoser> anyone have a suggestion... I'm trying to look for media in cdrom devices.
[18:17] <smoser> my initial try just tried 'mount /dev/cdrom' to test and see if there was a mountable cd there.
[18:17] <noobster> I am having trouble with a server. last night @ 3am, I was scp files from it and it crashed. I looked like it tried to reboot, but gave me a ssh error connection refused.
[18:18] <noobster> smoser, mount /dev/cdrom /mnt/cdrom
[18:18] <smoser> that takes like 18 seconds in this kvm vm.
[18:18] <smoser> if there is no cdrom there.
[18:18] <noobster> ?
[18:19] <noobster> got in today and the server has stopped @ init: ureadahead-other main process (912) terminated with status 4
[18:19] <smoser> ie, i have a /dev/sr0 and /dev/cdrom device in /dev. but there is no disk in it.  If i try "mount /dev/cdrom" it will take 18 seconds to return.
[18:19] <noobster> I did ctrl + alt + delete, and the system reboots
[18:20] <noobster> i am kinda freeked out
[18:20] <smoser> 'dd if=/dev/sr0 bs=1024 count=1024' (or hte like) will return immediately, saying "no medium found"
[18:20] <noobster> my boss is fixing to skin me lol
[18:20] <smoser> is there a better way that i can test if there is media ?
[18:20] <noobster> i downloaded, burned and booted from the cd rom
[18:20] <smoser> noobster, lucid ?
[18:21] <noobster> yes
[18:21] <noobster> 10.04.1lts
[18:21] <smoser> i suspect that a disk that you have in /etc/fstab is not there
[18:21] <smoser> and init is sitting around waiting for that disk
[18:21] <noobster> funny thing I was able to rebuild the entire md array no proble, except for /home thats an md/lvm
[18:22] <smoser> if you booted off live cd, try mounting the device, and removing /home form /etc/fstab
[18:22] <noobster> I got the md up, and it found vg VGHOME1
[18:23] <noobster> 2 things, how do i rescan/add LVHOME1 out of LGHOME1 w/out data lose, 2ed if there is nothing wrong w/ these drives, why will my server not start
[18:23] <noobster> I have tried recovery on the 2 kerns there nether boot
[18:24] <noobster> smoser, you want me to comment out /home lvm in /dev/md1/etc/fstab?
[18:24] <smoser> or you can add 'nobootwait'
[18:24] <smoser> to the options
[18:25] <smoser> but if that device isn't there, but is listed in /etc/fstab, init will wait for it
[18:25] <smoser> forever
[18:25] <noobster> /dev/mapper/VGHOME1-LVHOME1 /home           ext3    user_xattr        0       2
[18:25] <noobster> how should I adjust?
[18:26] <noobster> /dev/mapper/VGHOME1-LVHOME1 /home           ext3    user_xattr,nobootwait        0       2
[18:26] <noobster> like that smoser
[18:26] <smoser> yes.
[18:26]  * RoAkSoAx lunch
[18:26] <smoser> but you could just comment it out.
[18:26] <smoser> you can put it back later.
[18:27] <noobster> if /home is not there it gets stuck and I cant login
[18:27] <zul> smoser: you use can use to use the eject command to detect their was  cdrom or soemthing
[18:28] <noobster> can someone inform me of what 0 2 mean at the end, I forgot
[18:29] <smoser> i dont see how i would check to see if there was something there now.
[18:29] <smoser> noobster, man fstab
[18:31] <piercedwater> Does anyone know where I can find 64-bit Dell DSET or OMSA for Ubuntu?
[18:32] <piercedwater> Or something similiar...Im trying to get a diagnoses on my PE 1850
[18:33] <noobster> ok, the server came up, but with no /home directory
[18:34] <noobster> i can ssh in now though
[18:39]  * genii-around makes a note to remove "coffee" from his alert list
[18:43] <EvilPhoenix> heh
[18:55] <noobster> ok managed to get the server back up, thank you smoser for the nowaitboot option.
[18:55] <smoser> no problem.
[18:56] <noobster> one strange error I didn't get b4.. disconnected from plymouth
[18:57] <noobster> mountall: Disconnected from plymouth
[18:58] <noobster> has anyone seen this b4?
[19:07] <cohonen> okey guys,, i restart /etc/init.d/networking, and /etc/network/interfaces GETs ignored
[19:07] <cohonen> what the fu** BEEP is going on here
[19:08] <cohonen> how am i supposed to make semi static conf on my interfaces
[19:11] <cohonen> okey sorry,, i was in crack mode
[19:48]  * RoAkSoAx ... to hot outside today..
[19:56] <yoanis> hi there
[19:56] <yoanis> i just installed ubuntu-server 10.04 on a ppc machine
[19:56] <yoanis> the installer went fine
[19:57] <yoanis> but now i get this
[19:57] <yoanis> after reboot
[19:57] <guntbert> !enter | yoanis
[19:58] <yoanis> "/disk00:3,boot/vmlinux: unable to open file, Invalid device"
[19:58] <yoanis> guntbert: sorry about that
[20:00] <guntbert> yoanis: no problem, its easier to read that way - no experience with ppc here - sorry
[20:00] <yoanis> it's ok. thanks ;)
[20:22] <RoAkSoAx> zul: to build a package from ubuntu-cobbler bzr branch, should I just do bzr bd -- -sa and it will create the tarball?
[20:23] <zul> RoAkSoAx: it should
[20:24] <RoAkSoAx> zul: ok thanks ;)
[20:29] <RoyK> I see the binaries don't have symbols - is there a way to get a good coredump from a package without rebuilding it from source?
[20:38] <hggdh> install the needed dbgsym (or dbg)
[20:38] <jdstrand> hallyn: I hope you don't mind, but I recommended you for libvirt upload right in addition to what you asked for at the top of your application
[20:40] <jdstrand> hallyn: tip, you may want to subscribe to your applications page
[20:40] <jdstrand> s/applications/application/
[20:41] <RoAkSoAx> zul: ok so for example, when we have a .orig.tar.gz and in the source, we just debuild -S -sa, and it creates the .dsc and etc etc. So, if I the source is a bzr branch, what should I be doing to create the .orig.tar.gz and debuild -S -sa without actually having to have the dependencies installed?
[20:42] <zul> RoAkSoAx: what if you bzr bd -S
[20:42] <zul> and use something like pbuilder to build it
[20:43] <RoAkSoAx> let me try ;)
[20:47] <zul> right have a good weekend
[20:48] <RoAkSoAx> zul: you too!!
[20:54] <piercedw_> Can someone please suggest a diagnostics tool for a Dell PowerEdge 1850? I have tried Dell DSET and OMSA, neither of which are made for Ubuntu (64-bit).
[20:58] <guntbert> piercedw_: if I remember correctly they come as bootable iso files - just burn them...
[20:58] <piercedw_> DSET comes as a bin
[20:59] <piercedw_> Im just doing this for a home server...I dont have any experience with professional networks..just looking for something to tell my why my light ist orange
[21:03] <guntbert> piercedw_: yes, I see your problem now
[21:03] <piercedw_> yeah, its not exactly made for what im using it for, but i think i can get it to work
[21:03] <piercedw_> The amber light worrys me though
[21:04] <guntbert> piercedw_: did you see http://en.community.dell.com/dell-blogs/enterprise/b/tech-center/archive/2010/07/27/dell-openmanage-6-3-for-ubuntu.aspx  ?
[21:05] <piercedw_> awesome
[21:05] <piercedw_> ty so much
[21:05] <piercedw_> hope it works
[21:05] <piercedw_> :)
[21:06] <guntbert> piercedw_: Good luck :-)
[21:06] <piercedw_> thanks again
[21:08] <piercedw_> \o/
[21:15] <hallyn> jdstrand: thx, will do.  (in case someone comes by and badmouths me behind my back :)
[21:17] <jdstrand> hallyn: exactly ;)
[21:33] <piercedwater> I checked my system out via Dell OMSA, but I still have an amber light flashing. Any suggestions?
[21:35] <piercedwater> the only thing I can find is an alert from when I opened to chasis
[21:35] <piercedwater> chassis*
[21:46] <piercedwater> Its weird I have green checkmarks next to power supply, but I have a critical warning message under alerts
[21:47] <compdoc> critical doesnt sound good
[21:48] <compdoc> a chassis alert just means you opened the case. and the feds have been alerted
[21:48] <compdoc> jk :)
[22:02] <njbair> how long do files remain in /tmp?
[22:03] <guntbert> njbair: until reboot normally
[22:03] <bluefrog> njbair,  until reboot except if you change things in /etc/default/rcS
[22:04] <njbair> so files are only deleted on reboot?
[22:06] <njbair> I suppose I could write a cron script to delete periodically between reboots
[22:07] <erichammond> njbair: How will you know if a file in /tmp is no longer needed for some process?
[22:08] <RoyK> njbair: or just use tmpfs for /tmp
[22:08] <njbair> erichammond: lsof ?
[22:09] <RoyK> njbair: if some process stores files on /tmp or some tmpfs volume, you can't just remove the file, you need to restart the process as well
[22:09] <RoyK> removing a file is just an unlink
[22:09] <RoyK> removing its filename
[22:10] <RoyK> the data won't be reclaimed until the file is closed
[22:10] <erichammond> njbair: Just because it isn't open, doesn't mean some process isn't going to reopen it.  For example, some shell scripts use mktemp to create the file, then after a while open it for further processing.  By "process" I mean "software system".  Some applications store temporary files and use them later.
[22:10] <njbair> That's true
[22:11] <njbair> I'm mainly concerned with making sure that more recent files persist between reboots
[22:11] <njbair> so TMPTIME=10 should do the trick
[22:11] <erichammond> If you're having a problem with /tmp filling up, perhaps track down what processes are leaving them around and fix them.
[22:12] <njbair> It's not that...
[22:12] <RoyK> njbair: /tmp is temporary - store the files elsewhere if you want them to persist
[22:12] <erichammond> If it's humans, then have them use a different directory and automate cleaning that up based on age.
[22:12] <RoyK> erichammond++
[22:13] <njbair> I have a headless server, which has a weekly cron job that creates a .tgz backup of all the important files in /tmp. Then I run my backup script on my desktop which uses scp to grab the most recent backup from the server.
[22:13] <erichammond> argh.  Just glanced at my /tmp and it has a lot of files which look like they were put there by a browser download.
[22:14] <njbair> but those backups start to add up between reboots
[22:14] <erichammond> njbair: "important files in /tmp" ? :)
[22:14] <njbair> backups of important files...
[22:14] <njbair> I chose /tmp because I figured it would save me the trouble of writing a script to delete the old ones
[22:15] <RoyK> njbair: get real
[22:16] <RoyK> using /tmp isn't a good excuse for writing bad software
[22:17] <njbair> In Red Hat, /tmp files persist for 10 days. I assumed the same was true in Ubuntu.
[22:17] <erichammond> RoyK: get nice
[22:18] <njbair> really
[22:18] <RoyK> erichammond: sorry, mate, but not removing temporary files you've created isn't very good programming
[22:18] <njbair> there's nothing bad about leveraging OS facilities rather than duplicating them
[22:19] <njbair> RoyK: so you're telling me you employ the full gamut of best practices every time you write a shell script?
[22:19] <RoyK> njbair: If I use temporary files, I remove them, because when that software may be ported to another OS or distro, it might not work
[22:20] <RoyK> njbair: I'm just saying _remove the temp files_
[22:20] <RoyK> relying on one distro's functionality, like redhat's, isn't very wise
[22:21] <erichammond> anyway... we're now understanding how things work and helping each other figure out the best options
[22:21] <njbair> RoyK: point taken. but chill dude
[22:21] <RoyK> :)
[22:22] <erichammond> Sounds like the recommended approach is use a directory other than /tmp and implement your own backup and cleanup cron jobs.
[22:23] <njbair> what are your thoughts on using /var/backups
[22:23] <njbair> actually, that might be in the rsync tree... that won't work
[22:25] <patdk-wk> heh, simple to clean it, find /tmp/ -type f -mtime +10 -delete
[22:26] <RoyK> patdk-wk: that's the quick-and-dirty fix :)
[22:27] <patdk-wk> royk it works well :)
[22:27] <patdk-wk> I wouldn't use it on a *desktop* machine
[22:27] <RoyK> I know - I have a few of those cronjobs at work
[22:28] <patdk-wk> I'm just happy find supports -delete
[22:28] <erichammond> patdk-wk: That picks up orbit and pulse things on my system, which I'd be hesitant to delete outright.
[22:28] <patdk-wk> for years I was stuck on a system that it didn't, had to use xargs and rm :(
[22:29] <patdk-wk> erichammond, ya, like I said, don't use it on a desktop :)
[22:29] <patdk-wk> but pulse, x11, orbit, ... don't exist on a server, so normally ok
[22:29] <patdk-wk> but ya, he should use a backup dir, then run something like that on the backup dir
[22:29] <erichammond> until somebody releases a server package that makes assumptions about how Ubuntu treats /tmp
[22:29]  * patdk-wk puts /tmp on tmpfs
[22:29] <patdk-wk> even on my desktop machines
[22:30] <alex88> hi guys..has everyone tried installing greensql?
[22:30] <RoyK> erichammond: not really necessary - if people use /tmp for large storage without cleaning up, well, there is only one part to blame
[22:31] <erichammond> RoyK: It doesn't have to be large to be important.
[22:31] <erichammond> "important" while a process is running
[22:31] <erichammond> (not important between reboots on Ubuntu)
[22:32] <erichammond> I suppose any time you have shared space used by thousands of people, there's going to be some conflicts about what it can be used for and how to clean it up :)
[22:32] <RoyK> erichammond: how long would you think it'll take to fill up, say, 1GB of /tmp if an applications wasn't severely badly written?
[22:33] <erichammond> RoyK: I don't understand the question, but I'm pretty sure the answer is: 7
[22:33] <patdk-wk> royk, one second :)
[22:34] <patdk-wk> cat big file | grep something | somethingelse
[22:34] <patdk-wk> any piped command uses /tmp like nuts
[22:34] <patdk-wk> that is how I normally fillup my 2gb /tmp tmpfs
[22:34] <erichammond> been fun, gotta run
[22:34] <patdk-wk> not realizing that sql.gz is >2gigs when I pipe it to mysql
[22:35] <RoyK> patdk-wk: I'm quite sure that was grep's fault - piping doesn't use /tmp
[22:35] <patdk-wk> hmm
[22:35] <AlexMax> Help!  I accidentally 'purged' nginx instead of uninstalling it and my config files are gone.  I have backups, but I can't reinstall it because it can't find /etc/nginx/sites-enabled/default
[22:35] <patdk-wk> maybe, I know I hit that issue often with pipes
[22:35] <AlexMax> how do i get it to install it fresh
[22:35] <patdk-wk> haven't hit it in the last year though
[22:35] <AlexMax> I'm using the ppa version of nginx
[22:36] <RoyK> patdk-wk: no chance of undelete on linux
[22:36] <AlexMax> I already tried aptitude -o DPkg::Options::="--force-confmiss" reinstall nginx-full
[22:36] <AlexMax> that didn't do anything
[22:36] <patdk-wk> undelete?
[22:37] <RoyK> patdk-wk: that was for AlexMax, sorry
[22:37] <RoyK> AlexMax: are your config files still there?
[23:12] <AlexMax> RoyK: Nope, they'r egone
[23:12] <AlexMax> I don't care about getting my old configs back
[23:12] <AlexMax> But how the heck do I reinstall nginx
[23:21] <AlexMax> Setting up nginx-full (0.8.54-4ppa4~maverick) ...
[23:21] <AlexMax> ln: creating symbolic link `/etc/nginx/sites-enabled/default': No such file or directory
[23:28] <ahe_> is there a way to create a completely virtual lan that runs on a single linux host for testing purposes? what i want to do is to create interfaces like veth0 ... vethn all with different ip addresses that behave like they were connected to one physical switch so you can also send arp packages on those devices
[23:47] <NativeAngels> Hello
[23:54] <RoyK> AlexMax: then they're gone
[23:58] <szpuni> how can i chage startup of rsyslogd deamon? i have edited /etc/default/rsyslog file to add startup options and i have changed -c4 to -c0 -r but when i restarted deamon old configuration was keept any body have any ideas why this is happening?