[00:01] I know I haven't changed any core LVM configs; just added/removed volumes. [00:30] ugh the hospital blocks social networking sites even on their guest network [00:31] tunnel your way to freedom [00:32] a3ist: whats the easiest way to do that, i assume something involving ssh forwarding? [00:33] set up one of your own computers elsewhere as a proxy [00:34] then tunnel to it and forward the traffic over ssh for security [00:34] ccheney: sounds good to me [00:35] or do you mean as a visitor/outpatient ? [00:35] any suggestions for a good simple proxy to run? [00:35] linxeh: visitor, my wife had our son this morning [00:36] and they have internet access ? [00:36] linxeh: yea :) [00:36] its not as uncommon as you'd think [00:36] here we have to turn off all wifi and mobile devices [00:36] but they run websense on the http [00:36] I hate websense [00:36] webnonsense [00:36] it miscategorises sites bigtime [00:37] oh cheney as a side note when you run the ssh tunnel you can add compression as well to help with latency [00:37] * ajmitch has used corkscrew for setting that up before [00:37] a3ist: oh ok, thanks, the last time i had to run a tunneled proxy was nearly 10 years ago and i have forgotten most of how to do it :) [00:37] ccheney: and congratulations :) [00:37] ajmitch: thanks [00:38] ssh -C -L port:localhost:port hostname [00:39] there are a ton of guides and stuff out there too for reference [00:39] i think i don't need corkscrew as they don't block other ports outbound at least not 22 anyway [00:39] a3ist: ok will take a look, thanks [00:39] then forwarding to a squid proxy on a remote host might work well [00:39] yup [00:40] it works for me for getting to the US music store for U1 :) [00:42] heh [00:51] there [00:51] there is a way to tunnel a full ipv4 network over ssh iirc, right? [00:52] if i did that i wouldn't need to change any configuration on any of my apps [00:52] ccheney: openvpn is the easiest way [00:52] smoser: As long as you can say it builds, installs, and run in the bug and it's not avoiding doing a proper SRU, I can approve it. [00:52] Give me a ping when it's ready. [00:53] ccheney: though you can do pppd + ssh too [00:54] SpamapS: ok [01:01] hi guys i have 60 client pc any apps that can help me deploy the installation of operating system- like ubuntu desktop..diskless and server base..? [01:01] SpamapS: does openvpn do the server side as well as client? [01:02] SpamapS: i installed it and it seems like a client, but maybe i misunderstand how it works [01:02] ccheney: yeah its a peer to peer connection [01:02] ccheney you can do vpn with NetworkManager but you need the extra plugin for the right type [01:02] like networkmanager-pptp [01:02] ccheney: its just a cmdline option whether to listen or conncet [01:02] from the client side, that is [01:02] not the host side [01:02] ok [01:03] ruben23: so these are NFS root? [01:08] SpamapS: the client pc are all empty newly purchase..i want to deploy ubuntu desktop or Ubuntu netbook editon using the server side.. [01:09] ruben23: but you said they are diskless, so how will they load their applications? [01:09] ruben23: you may want LTSP [01:10] SpamapS: sorry i mena disc less no DVD-Rom or cdrom.. [01:10] there are storage in every PC [01:10] ruben23: ah, are they all the same model PC? [01:11] ruben23: if so, SystemImager might work for you [01:11] SpamapS:there 30 units the same and the other batch same model also, but the 2 catergory are diffrent. [01:11] ruben23: anyway, good luck, I must be going [01:12] ruben23: right so you can make two "golden images" and then use systemimager to install them all [01:14] SpamapS:any how to do it..? [01:14] with sysmanager. [01:15] i mena sysimager [01:16] ScottK, i commented at https://bugs.launchpad.net/lucid-backports/+bug/589214 . [01:16] Launchpad bug 589214 in lucid-backports "backport maverick ec2-ami-tools to lucid" [Undecided,New] [01:16] Looking [01:17] will try to get the other tested soon, but really, the changes are very minimal over the previous versions (the essentially add a string 'ap-southeast-1'). [01:18] smoser: Approved. [01:18] Now it just needs an archive admin to execute the backport. [01:18] ScottK, so what is the appropriate version number ? [01:18] 1.3.49953-0ubuntu1~lucid1 [01:19] ok. what i got from https://wiki.ubuntu.com/SecurityTeam/UpdatePreparation [01:19] thanks. [01:21] smoser: Yes. That's for security/updates. Backports uses a different numbering scheme. [01:22] right. and the ~ is reasonable. [01:46] New bug: #591517 in mysql-dfsg-5.1 (main) "package libmysqlclient16 5.1.41-3ubuntu12.1 failed to install/upgrade: ErrorMessage: el subproceso script post-installation instalado devolvió el código de salida de error 2" [Undecided,New] https://launchpad.net/bugs/591517 [04:00] i can't connect my guest os after install [04:00] do know how [04:06] unewbie: what virtualization technology are you using? [04:12] kvm [04:12] twb: kvm [04:13] unewbie: are you using libvirt? [04:13] yes [04:13] vmbuilder when create guest os [04:13] OK, then you would access it via libvirt somehow. [04:13] how is that? [04:13] I don't know. [04:14] :D [04:14] any other way? [04:15] That would depend how kvm is invoked by libvirt [04:15] what's your suggestion [04:15] Normally when I use kvm directly, I just connect its text console to stdio. [04:15] it's my first virtualisation [04:16] what's text console? [04:16] Er, the bit without the pictures. [04:16] do i have to manage directly on the host server? [04:17] may be that's what i want [04:17] the text console [04:17] i don't want graphic [04:17] i only want to use ssh from my xp desktop [04:19] Yeah, so I would ssh in, and run kvm -curses -hda foo.img [04:20] -hda? the harddisk? [04:20] I think you should just read the libvirt documentation. [04:21] where? [04:22] I don't know. [04:22] Did you try reading the ubuntu server guide? [04:23] yes [04:23] but don't understand [04:24] i'm seeing the serialconsolehowto page is this what you mean? [04:24] i thought it was the serial port behind the server [04:24] not the text console [04:27] The serial console is a text console. [04:27] Probably libvirt is just shit and can't directly export a VGA text console via curses. [04:28] Most virtualization systems can't, so the libvirt people probably haven't added that into libvirt, even though your kvm backend supports it [04:29] what are using using for virtualisation [04:29] may be i cant switch to your way [04:29] can [04:32] so [04:32] if i have the 64 bit mini.iso [04:32] and idk if i have 64 bit and it loads [04:32] do i have 64bit? [04:32] cause all it says is semperon, but online it says 64bit... [04:38] what socket [04:38] twb: from that page, i have edit ttyS0.conf but i can't locate the file [04:39] What release is the VM running? [04:39] Unless you *install* using the serial console, you won't have a ttyS0.conf [04:39] if it's socket A, it is most definitely not [04:39] if it's socket 754, it's about a 70/30 chance [04:39] how to find the vm release? [04:40] kvm --version [04:40] QEMU PC emulator version 0.12.3 (qemu-kvm-0.12.3), Copyright (c) 2003-2008 Fabrice Bellard [04:40] is that? [04:47] err [04:47] back [04:47] I have [04:47] eh [04:47] lemme see [04:48] err [04:49] cnet says sixty four bit computing yes [04:49] it does not say the socket... [04:49] and since its in use im stuck on this slow tablet [04:53] Mgamerz: if the kernel boots, you have a 64bit capable chip [04:54] this pc was made in 05 [04:54] perhaps they did not label it as 64 as 64 was still semi new [04:54] First opterons were around '04 IIRC [04:54] 03 [04:54] wikipedia'd [04:54] :) [04:55] Opteron 1401.4 GHz1 MB800 MHz7x1.5584.7WSocket 940June 30, 2003[1]OSA140CCO5AG (B3) [04:55] :) [04:55] same link I had [04:55] i thought the pc server i had sucked [04:55] now really looking at the specs [04:55] no wonder its fuzzy in my head.. my first son was 3 months old.. probably hadn't slept for a few days when they announced. ;) [04:55] my pc sucks :( [04:55] Mgamerz: its ok, everybody's PC sucks now.. thats why we have the cloud. :) [04:56] but do you have the weed to support the cloud? [04:56] :) [04:57] * SpamapS coughs [04:57] aghhh [04:58] red screen of doom [04:58] step failed!! [04:58] oh the humanity [04:58] OMG [04:58] NO [04:58] UBUNTU NO [04:58] YOU DID NOT JUST DIE WHILE INSTALLING [04:59] :D [05:00] Mgamerz: red screen sounds odd [05:00] it said install step failed [05:00] :/ [05:01] i re-ssh'ed in [05:01] i have an installation of ubuntu server behind a private network. i have a staging server on the public network; what's the easiest way to push package updates to the private server? [05:01] webmin :P [05:01] for a start, i have copied all the packages from /var/cache/apt/archives to the private server [05:02] so i have a copy of all the packages i would need to update. i guess i also have to sync whatever info is downloaded when i do apt-get update on the staging server [05:02] eh [05:02] im not a guru... [05:03] so... i can't help you :s [05:03] panfist: "easiest" is probably to mirror the updates apt source to a USB drive and then sneakernet it over. ;) [05:03] lol sneakernet [05:04] yeah lol at least my ubuntu didn't just die :P [05:04] actually [05:04] i recoovered it [05:04] by resshing [05:05] manually installing stuff in miniiso killed the interface [05:05] it's either a.) sneakernet or b.) become an expert in linux firewalls overnight + lock down windows workstations or c.) say fuck it and just leave proprietary information [05:05] on the internet [05:05] C [05:05] i think the "lock down windows workstations" part is the scariest and most impossible [05:05] support open source [05:05] panfist: you don't need to be an expert to poke a hole in a firewall. [05:06] and piss off the rest :P [05:06] spamaps no i need to be an expert to make sure no one can poke a hole in mine [05:06] disable the network adapters... nobody will get in then [05:06] i can guarantee you [05:06] no wait [05:06] bet money on it [05:06] and that if anyone compromises a windows box operated by grandmothers they can't use that to poke a hole in my firewall [05:07] panfist: uh wha? [05:07] O_o [05:07] mgamerz so in other words, sneakernet [05:07] eh [05:07] sure [05:07] 85% installed [05:07] ok so [05:07] spamaps what? [05:08] panfist: the offline updating thing though, its a common issue for high security requirements. I'd say your best bet is to use something like 'apt-mirror' onto a disk and then just expose it like a cdrom in sources.list [05:08] i am installing 64 bit and i want to install a nvidia propriety driver (i need for video recording to server) [05:08] so i need the kernel source [05:09] i looked it up earlier and it was hella scary [05:09] on desktop version it was all preinstalled... [05:09] Mgamerz: you shouldn't need to do anything crazy for that. [05:09] git [05:09] the web page was scary long [05:11] Mgamerz: well that pre-installed version on the desktop is just a package ( I believe from the partners repository ) .. so should be easy enough to just apt-get install [05:11] the server doesn't have the hardware drivers... thing [05:11] jockey? [05:11] i couldn't find out the name of it [05:11] karmic was jockey [05:12] hmmm [05:12] grub 2 b or grub not2b [05:12] grub 2 is a nightmare [05:12] but it has ext4 support... [05:12] such hard decisions [05:14] finishing installation [05:14] here we go [05:15] linux takes security miles ahead of windows [05:15] wiping swap space for security [05:18] .... [05:18] this is taking a long time [05:20] here we go!! [05:20] .... [05:20] you cannot be serious. [05:21] damnit grub!!!! [05:23] it randomly does not work [05:24] and i cannot get it to load a burned cd unless i use the miniiso on my sd card grrrr [05:27] oh ffs [05:27] it wants me to reinstall it all [05:27] screw that [05:27] its floppy disk time [05:28] ERRR [05:28] ... it figures this is the one computer without a floppy drive [06:00] grub2 is so lame === lifeless_ is now known as lifeless [06:20] ... it fails to lload [06:20] completely [06:20] i hate this :l [06:38] and it wont mount the cd [06:38] this is just lovely [06:43] are TLER disks (like western digital raid edition, for instance) better suited for use in a softraid mirror than standard desktop-oriented sata disks? [06:48] rcsheets [06:48] Mgamerz [06:48] what do you mean by softward raid vs sata [06:48] like sata in a raid? [06:48] or just standalone [06:49] !enter [06:49] Please try to keep your questions/responses on one line - don't use the "Enter" key as punctuation! [06:49] oh, sorry, i didn't realize my question was ambiguously phrased. [06:49] twb [06:49] i have installed ubuntu like 4 times today [06:49] grub is a pos [06:49] let me try again. for use in a softraid raid1 mirror setup, are RAID Edition drives (like WD RE3, etc.) better suited than standard desktop sata disks are? [06:50] rcsheets: softraid doesn't care [06:50] you might, if you need specific MTBF or other criteria to be met [06:50] lifeless: i'm asking only with regard to the disk's firmware giving up faster when it can't quickly perform a requested operation [06:50] i'd just use normal sata [06:51] it'd be cheaper and probably not cost you much time... but thats my opinion [06:51] i'm not sure how it would cost me any time [06:51] so........... [06:51] there you go? [06:51] rcsheets: "raid edition" might simply mean the firmware does different block alignment or such. Write the vendor and ask. [06:52] this is taking ridiculously forever [06:52] twb: i'm sure the vendor will tell me to buy the more expensive disks. the feature i'm referring to is meant to avoid a disk dropping completely out of an array just because it's taking a long time to repeatedly retry an operation. [06:52] grub2 better work or i am going to throw something [06:53] oh... [06:53] hmm O-o [06:53] im not a linux expert [06:53] so... you shouldn't trust me linux softraid [06:53] rcsheets: I wouldn't know about that. [06:54] the theory, as i understand it, being that if the disk is in an array, a failure to (say) read a single sector isn't such a big deal, but if the disk tries for a long time to do that read and the (hardware) raid controller gives up waiting on it, then it will fail the whole disk [06:54] unnecessarily degrading the array [06:54] rcsheets: that might appliy to hardware raid arrays [06:54] i wasn't sure if the same logic applied to softraid as well [06:54] but wouldn't a mirror array just read the other disk? [06:55] frankly, if I have a disk that can't read data it wrote, anywhere on it, I want it to be removed immediately [06:55] +1 [06:55] then you must replace a lot of disks lifeless [06:55] rcsheets: its going to depend on what you're doing with it, there isn't a 'best' answer we can give you. [06:55] Mgamerz: no, MTBF is pretty high [06:55] Mgamerz: usually a batch goes at once though [06:55] ok, fair enough. [06:55] i have a karmic hard drive with 277 bad sectors [06:55] its... disconnected because of that :) [06:56] karmic was a nightmare for me :/ [06:56] lifeless: i'd rather have the disk continue operating in the array, rather than the array being degraded, personally. if i choose to replace the disk, i'll do that at a convenient time. meanwhile, i'd rather it just remap that sector. [06:56] rcsheets: well, as I say, it depends on your needs/config etc [06:57] sure. alright, thanks. [06:57] Sigh. [06:57] so... slow... [06:57] So I'm upgrading my workstation SOE from a 8.04 to 10.04 base. [06:57] i have spent 3 hours getting this pos to work [06:57] Now I get to deal with a whole NEW bag of places where Ubuntu-specifc postinst code assumes it's not a chroot. [06:57] should not have F'ed with it in the first place... [06:58] ! [06:58] is there a way to run a command if im like in rescue mode [06:58] as a user from a machine i have mounted a hard disk on? [06:58] like if i needed to update grub via rescue [07:00] Mgamerz: go talk to #grub and get them to explain how to fix your bootloader. [07:00] Mgamerz: reinstalling over and over won't help, nor will complaining [07:00] it worked last time <_< [07:01] if anyone happens to be interested, there is some discussion of a similar nature to my question here. http://arstechnica.com/civis/viewtopic.php?p=1144188 [07:03] i hate this installer [07:04] i cannog go to the partitioner to see the disk mounts and then go back to grub without it complaining to do it again. [07:05] eh [07:05] so [07:05] this is one of those rescue mode moments... [07:26] I was gifted an older dual xenon server, and I have been trying to install 10.04 server on it without any success. It won't boot from a known good CD. [07:26] anyone feel like helping me troubleshoot this? [07:27] Won't be able to do a ton, but I presume you've checked BIOS for boot order? [07:28] yeah. I actually just disabled everything but the CD ROM [07:28] and it comes up 'no boot media' or something? [07:28] boot failure, system halted [07:29] hrm, 'system halted' sounds like Linux, not BIOS... [07:29] me either [07:29] mine does that as well. [07:30] if you boot it without the CD in it, does it give the 'boot failure, system halted' message? [07:30] It just spins up and down [07:30] or something different? [07:30] lemme try [07:30] That will narrow it down to BIOS or CD. Might be the CD doesn't like some piece of hardware its seeing :-D [07:31] I have tried like 4 cd's [07:31] i think my drive is just dying. [07:31] the cd status light blinks on and off when the server first boots, but nothing after the intel screen [07:32] same deal [07:32] yep thats post blinking it [07:32] holmser_, ok so then we know it is booting, but at least its reading the CD [07:32] Q: isn't that just post? [07:32] Mgamerz, have a spare drive you can rob from another box, even temporarily? [07:32] eh [07:32] yeah, its just post. [07:32] well im using mini.iso [07:33] and loading the iso from flash from there. [07:33] Mgamerz, 'system halted' is something I know Linux uses...usually its 'BOOT MEDIA FAILURE insert disk' or something [07:33] except grub neverworks it seems... [07:33] I know its looking for a network boot, but I don't know how to do one [07:33] just hangs the system [07:33] Hrm. [07:33] can your box support usb boot? [07:34] i booted from an SD card [07:34] 128MB :P [07:34] Any other OS CDs either of you have laying around to test, make sure its making it past BIOS? (Don't need one you'll install, just to get up to a loading screen to confirm its disk and not hardware) [07:34] well mine will spin up and down and up and down for like 4 mins [07:34] then it will give up [07:34] I am positive the CD is good. I just installed on a different server not 2 hours ago [07:35] it reads it when mini iso mounts is... well used to, now it doesnt. i think its just dying [07:35] Mgamerz, since you're booting funny I don't know on yours, never tried that method. (I like the idea, but the config always looks like a PITA) [07:36] holmser_, ok so is the CD aborting/kernel panic'ing, or is the hardware having issues? That's what the second CD will help test. [07:36] it never checks the [07:36] CD [07:36] it checks it on post, then nothing [07:37] it doesn't even try to boot from the CD [07:38] So it IS hardware then. Any idea if CD drive is IDE or SATA, and if you have a similar drive to try s wapping in? [07:38] not to jump in on a thread [07:38] but [07:38] which kernel is best to pick? [07:38] generic server virtual image-gen... [07:38] I just run with defaults, I don't do too much messing around. *shrugs* [07:38] then the linux-image-#######-generic [07:38] Mgamerz: probably generic for your purposes [07:38] :S [07:39] K [07:39] include all drivers [07:39] yes... alright thanks [07:39] but yeah unless you're doing something weird or have limited hardware I'd just do generic [07:40] is a pos considered limited [07:40] well for server you don't need much. [07:40] as a base system, anyway [07:40] how much RAM? (Usually the limiting factor) [07:40] 768 [07:40] -32 shared [07:41] that should be plenty, unless you're using it as a huge VM host. [07:41] lol vm [07:41] yeah right [07:41] for a dumb console server you should be able to drop that shared to 8MB [07:41] if no GUI [07:41] if possible, probably via BIOS [07:41] i want to record tv with this... [07:41] at least try [07:42] well that doesn't require a GUI [07:42] it doesn't? [07:42] not generally [07:42] with a tv tuner ? [07:42] can be remotely controlled, IIRC [07:42] i had tightvncserver going for a while [07:42] Unless it'll use your (likely onboard) GPU for processing, no. [07:43] yep geforce 6100 [07:43] tightvnc to the host console, or a virtual one? [07:43] virtual X display [07:43] so i could config a few things from gui [07:43] other than that X never ran [07:43] mediatomb pisses me off tho... [07:44] it's database dies randomly [07:44] What was the name of the package for UEC? [07:44] yeah virtual you're not using the "GPU" at all, IIRC [07:44] OK, so I have a 10.04 desktop chroot on an 8.04 server. wget works, apt-get hangs and permanently fucks up all processes that attempt to do I/O on the underlying filesystem until I reboot. [07:44] * Ng looks for a PPA with qemu-kvm 0.12.4 [07:44] What gives? [07:44] had a windows box anyway set up that way...console ran like crud, remote ran smooth as silk. [07:45] There's nothing in dmesg, either. [07:45] what is... chroot? [07:45] im still linux nubcake :/ [07:45] Mgamerz: don't worry about it; you don't know enough to help. [07:46] twb: strace the apt-get, see what it's hanging on? [07:46] ... [07:46] Ng: yup, that's the plan (waiting for it to reboot again :-/) [07:46] Ng: also, to try without first mounting /proc and /sys within the chroot (which you need to do if you want apt-get to work.) [07:47] Mgamerz, its a very funny way of doing basically a VM. Really weird, roundabout way of doing it...but it "works" (for a better description, I'd hit up wikipedia) [07:48] Q_Continuum: that's something of a modernist interpretation :-) [07:49] Ng: sounds fun - should i put that on my production system? [07:49] rescue shell? [07:49] MTecknology: hmm? [07:50] Ng: 01:44 * Ng looks for a PPA with qemu-kvm 0.12.4 [07:50] !info qemu-kvm [07:50] qemu-kvm (source: qemu-kvm): Full virtualization on i386 and amd64 hardware. In component main, is optional. Version 0.12.3+noroms-0ubuntu9 (lucid), package size 2496 kB, installed size 7056 kB (Only available for i386 amd64 powerpc s390 lpia all armel sparc) [07:50] twb, not having looked at it in a long time myself it was the best description I could come up with for someone who doesn't know it :-D [07:51] MTecknology: ah right. I'm doing it specifically because I hit a memory leak bug in testing 0.12.3 and I want to see if it's fixed (a least one leak bug was fixed in .4) [07:51] MTecknology: I have no idea how people are using kvm in production, I've been running it seriously for less than 48 hours and I'm having nothing but problems ;) [07:51] Ng: I've had an enjoyable time actually :P [07:52] Ng: what host, and what guest? [07:52] twb: amd64 lucid [07:52] Ng: I notice high mem usage on server - but it's nearly all cached [07:52] Ng: both headless? [07:53] MTecknology: at least historically, qemu would allocate the guest's memory up-front [07:53] twb: both are ubuntu-minimal installs with a few things added, but libvirt/kvm are starting a vnc server for each guest's console, but I'm not really using that [07:53] Ng: surprising; I run a lot of qemu instances like that, and I generally don't have problems. [07:53] MTecknology: I was trying to figure out why the first guest I made was showing ext4 filesystem errors on every boot so I fired up 4 VMs overnight running spew to see what would happen and 3 died this morning from OOM ;) [07:53] Ng: http://pastebin.ubuntu.com/447043/ [07:54] MTecknology: my new guest os installed and i can ssh [07:54] unewbie1: great [07:54] thanks for the advices [07:55] but i have to stop the apparmor of the host [07:55] twb: it certainly is surprising. I think tonight I'll leave the machine running memtest for 8 hours or so to rule that out [07:56] unewbie1: glad you have it working [07:56] hi all [07:56] newbie question.. [07:56] Question: Can I have Ubuntu Server 10.04, at 4am each moning (i'm guessing using cron), reboot the server, enter Clonzilla (available in my GRUB menu), bootup the Clonezilla, execute a backup image of my main RAID to a large drive (I know I can do this part by command line and it work) then when complete, reboot the system again and run Ubuntu Server (from GRUB)? Can all these reboots etc and selection opti [07:58] Yosi: better question - what are you trying to achieve? [07:58] i want to automate the entire process [07:58] right now it all works but manually [07:59] have to be at the cpu at 4am [07:59] rebbot, backup, then reboot again into the system [07:59] ewe .. [07:59] offline backups? [08:00] By chance - you ever look into rsnapshot? [08:00] i want a cron job to execute a reboot, then select the grub item, lauch a command line and when done, reboot and lauch the system again [08:00] tried them all [08:00] only thing that works well is clonezilla [08:00] nothin in linux unfortunately works well with live databases and 100 % [08:01] mysql databases? [08:01] my EMC SAN does great LUN snapshots, but cost me a TON to license that feature for every host [08:01] my sql, php, apache [08:02] there are no apache databases.. [08:02] or php [08:02] the data gets corrupt with most options i tried, too many transactions per second [08:02] true [08:02] sorry [08:02] thats what i meant [08:02] just mysql [08:02] only luck with clonzilla when everything is offline or online with Snapshots on the SAN [08:03] thats what mediatomb does [08:03] it dies with transaction within transaction [08:03] I tried zmanda [08:03] doing an offline backup just for mysql makes no sense... with mysql you can copy the entire db files and only lose current transactions [08:03] all sorts of things.. nothing works well when the database is heavily accessed [08:04] Yosi: http://pastebin.ubuntu.com/447046/ [08:04] onjly other thing that worked without corruption is a mysql replication server [08:04] i will take a look at that thanks.. any way of automating wwhat i got now? [08:05] can CRON launch all these sequence of events [08:05] not without risk of breaking things - you'd have to script editing the files [08:05] at 4am, we offline so i don't mind being down [08:05] we are not a 24.7 shop [08:05] You'd have to alter the grub menu entires via a script - which is bad [08:06] what do u mean by breaking things? [08:06] ahhh [08:07] Just learn how to efficiently backup your database - I'd restored many times using that script I gave you - you only ever lose current transactions which are not yet in the files [08:07] Then no breaking :) [08:08] I'm gonna head to sleep though - Just trust me... changing a config, rebooting, changing a config, rebooting - you will break things [08:08] thanks [08:08] i will test it out [08:08] likely result in not rebooting [08:08] Ng: busted! /proc/6542/status: No such file or directory [08:08] fair enough [08:08] To test it out... take the outut it produces - and restore to a blank server [08:08] thanks [08:08] i will [08:08] g'night [08:08] night [08:12] Daviey: hey? [08:14] OK, now I think it's aufs' fault [08:23] chroot: cannot execute /bin/network-console: No such file or directory [08:31] hey guys [08:31] question what port do accessing the repositories use? [08:32] Mgamerz: that's caused by your parent shell being some kind of retarded thing [08:32] eagles0513875: apt-get typically uses HTTP [08:32] twb: talk about strange ok [08:32] thanks twb [08:50] Ng: \o [08:51] Daviey: you're king of virtualisation these days, right? ;) [08:52] Ng: I'm no longer physical in this world, i'm a virtual Daviey.. yes :) [08:53] awesome [08:53] sweet [08:53] how did you ascend? [08:53] Daviey: I was just curious how much we fancy SRUing kvm - I've found a memory leak that (subject to some testing proving the case) I think is fixed upstream [08:54] Ng: funny you say that, Serge is wanting to get an upload of kvm today [08:54] * Daviey checks if it was maverick or an sru [08:56] Ng: i can't find the email.. he must have said verbally.. Do you have a .patch or branch? [08:56] Daviey: I do not. I filed bug #591610 and linked it to the upstream bug, but they don't indicate which commit fixed it, just that it's fixed in 0.12.4 [08:56] Launchpad bug 591610 in kvm "memory leak" [Undecided,New] https://launchpad.net/bugs/591610 [08:57] so I've built that and am re-running my test (4 1GB VMs, some virtio some not, running spew(1) --read-after-write for about 8 hours) [08:58] hmm [08:58] Daviey: while I'm pestering you - I'm guessing you use ubuntu-vm-builder, so you're building VMs with ext4 by default? Have you seen any odd fs errors? the first VM I made was spitting out a few ext4 errors each boot [08:58] Ng: TBH.. mainly using ext3 atm [08:59] ubuntu 10.10 maverick mccain [08:59] Daviey: interesting. I had to hack vmbuilder to make an ext3 one, at soren's suggestion ;) [08:59] Ng: the fsync changes in dpkg make it so slow.. most of the testing i'm doing is on ext3 because of that [09:05] Ng: whilst that certainly sounds like a potential candidate for SRU.. The upstream bug doesn't give any indication where it was fixed.. Seems like someone might need to check the vcs logs between 0.12.3 and 0.12.4 :( [09:06] I only quickly skimmed the changelog for the 0.12.4 release and it seemed like it was all bug fixes [09:13] hrm, can I poke libvirt into running kvm with "-cpu host" or am I forced to specify an exact model? [09:14] alright [09:14] so i got the server working [09:14] but i dont have any repos. [09:14] Ng: What's the use case (for the "-cpu host" question)? [09:15] soren: well by default I'm just getting the lowest-common-denominator CPU for easier migration. I have no need of migration so I'd rather that all the interesting CPU flags were exposed to the guests for their optimal enjoyment [09:16] Ng: Makes sense. [09:18] Ng: This is using the notation? [09:19] Ng: Or are you referring to the machine attribute of the os->type tag? [09:21] soren: I guess the stuff. I'll set it to whichever pre-defined model is closest to an i7, I was just curious if there was a way to make it just set whatever flags/models are present in the host [09:22] Good day all , would someone explain this [09:22] Amazon EC2 fits perfectly with Ubuntu Enterprise Cloud's modularity, virtualisation capabilities, range of applications and optimised performance. [09:22] omg [09:22] no [09:22] UEC is package of softwares ? [09:22] i dont have nano installed [09:22] NOa [09:22] alabd: UEC allows you to build a private cloud on your own hardware [09:23] SpamapS: thanks is it pack of softwares ? how to install it ? [09:23] alabd: https://help.ubuntu.com/community/UEC [09:25] NG thanks what does this mean ? Amazon EC2 fits perfectly with Ubuntu Enterprise Cloud's modularity, virtualisation capabilities, range of applications and optimised performance. [09:26] alabd: Where do you see this? [09:27] my password is wrong [09:27] soren: http://www.ubuntu.com/cloud/public [09:27] an not get if EC2 is an image or is a software or is a concept [09:27] If you're not familiar with EC2, you probably don't need UEC. [09:28] wtf... [09:28] i can log in but my sudo password is different? [09:28] alabd: EC2 is Amazon's public cloud service. UEC has some of the same APIs, which is why they can be a useful match to people who need both types of cloud [09:30] Ng: Whilst i'm not "suggesting" this.. i have previously made a kvm wrapper script to work with libvirt, to inject more switches :) [09:30] hah [09:30] I found the problem with my chroot [09:30] aufs is happy if my script is fed an .iso, but not if it's fed a directory. [09:31] I haven't done that for twelve months, so I'd forgotten that aufs had this bug. [09:31] Solution: don't feed it a directory. [09:31] errr [09:32] how the heck do you use vi [09:32] you take time to learn it and then use it and cry when anything isn't vi :D [09:32] i need it since my sources list has only cdrom and my cd drive died [09:33] i need to get nano!! [09:33] Mgamerz: use cat > file (and make no typos!) [09:34] ED, MAN! MAN ED! [09:34] IT IS THE STANDARD UNIX EDITOR [09:34] nano>vi [09:34] false [09:34] we can all agree on that [09:34] truths [09:34] !iptables [09:34] Ubuntu, like any other Linux distribution, has firewall capabilities built-in. The firewall is managed using the 'ufw' command (see https://help.ubuntu.com/community/UFW), or 'iptables' (https://help.ubuntu.com/community/IptablesHowTo). GUI frontends such as Firestarter/Gufw (Gnome) or Guarddog (KDE) also exist. [09:35] Mgamerz: that won't work; nano needs stdout to be a controlling tty [09:35] Mgamerz: holywar in the making! [09:35] Ng: /usr/share/libvirt/cpu_map.xml has the list of cpu models libvirt accepts. [09:36] i should not have to use google to figure out how to type in a terminal [09:36] this is outrageous [09:36] Mgamerz: vimtutor [09:36] http://www.gnu.org/fun/jokes/ed.msg.html [09:36] soren: hmm, I guess I'll just pick the greatest/closest to the host CPU then :) [09:37] Ng: Or add one that matches your host CPU. [09:37] soren: editing files in /usr makes me sad, but sure [09:37] i have no repos [09:38] Ng: It should :) [09:38] so iam serverly limited in abilities [09:39] oh [09:39] lovely [09:39] i ruined my sources.list [09:40] Mgamerz: http://repogen.simplylinux.ch/ [09:46] how do I disable the post-login message that tells the system load, documentation links, etc.? [09:48] i mounted the iso and insgalled nano [09:48] thank goodness!! [09:50] Hi. I would like to redirect a port internal from port 4285 to port 443 with only iptables. Is it possible to do this with IPv6? [09:57] hadoop: It's generated by pam_motd (https://help.ubuntu.com/10.04/serverguide/C/pam_motd.html) [09:58] hadoop: you can disable it completely by reconfiguring the landscape client, by running sudo dpkg-reconfigure landscape-common then selecting 'Do not display sysinfo on login' [10:00] TREllis: thank you, that worked :) [10:00] hadoop: or you can customise it yourself by modifying the files in /etc/update-motd.d/ or silence it for one use by creating a ~/.hushlogin [10:00] user* [10:00] oh, I think I'll do it per user, no idea when I may require such info in future [10:21] New bug: #591642 in javassist (main) "Sync javassist 1:3.12-2 (main) from Debian unstable (main)" [Wishlist,Confirmed] https://launchpad.net/bugs/591642 === register is now known as Guest86842 [11:36] Good day all, is this true > UEC provide us easy cloud computing deployment and management on ubuntu [11:37] and does "Ubuntu is hugely popular on Amazon's EC2" mean ubuntu is installed on amazon servers ??? [11:38] NG^ [11:38] alabd^ [11:38] ng ^ [11:38] alabd: I don't think amazon have publically stated what their servers run, but what that quote means is that people are running Ubuntu inside EC2 - you can run whatever you choose and people are choosing Ubuntu :) [11:39] NG thanks does "Ubuntu is hugely popular on Amazon's EC2" mean ubuntu is installed on amazon servers ??? [11:40] sorry [11:40] ng and when they choose ubuntu they should use install UEC on their ubuntu yes ng ? [11:40] Hallo, [11:40] setting up a cyber in Togo. Set up a nfs-server and i want do control the access of some folder. How can i realize this? [11:40] and is this sentence true ? is this true > UEC provide us easy cloud computing deployment and management on ubuntu [11:42] alabd: choosing ubuntu on EC2 is perfectly valid instead of installing UEC on their own hardware - not many people have the same amount of hardware that Amazon has ;) [11:42] but being able to do either is a great choice [11:42] I suggest you read some more background articles about what EC2 and UEC are :) [11:44] see am humble true ?we can have cloud computing on ubuntu with UEC ok ? so when they choose their host on EC2 to be ubuntu , will amazon install UEC for them ? ng [11:44] alabd: people can install the images provided by amazon themselves [11:45] yes but before installing UEC do they can use amazon images on ubuntu ? binBASH [11:46] I dunno what virtualization amazon is using. [11:46] I don't have an EC2 account either [11:47] binBASH: UEC is virtualization ? [11:47] it make ubuntu a cloud server yes ? [11:47] alabd: With UEC you can build your own private cloud which is compatible to the Amazon EC2 api [11:47] so when we choose Ubuntu on amazon [11:48] so you have several servers, you can set them up as cloud. And let your users run virtual machines on them. [11:48] it means then we can use ec2 images on ubuntu [11:49] on those virtual machines ubuntu will be host or guest or both ? [11:51] a virtual machine appears just like a real server [11:51] so you can install what you want on your virtual machine [11:51] even windows for example [11:52] it's just like a real server. [11:53] see vm itself is installed on a server [11:53] yup [11:54] is that server a cloud server ? [11:54] yup [11:54] and UEC is installed on this server yes ? [11:54] right [11:55] so this server is ubuntu [11:57] thanks a lot binBASH now want to know amazon EC2 is instead of which part ? [11:59] instead of ubuntu server ? [11:59] instead of vm guest ? [12:12] alabd: Amazon EC2 is the same like Ubuntu UEC [12:12] a plattform to run virtual machines [12:12] ;) [12:14] ... on many many real servers === IPNixon_ is now known as IPNixon [12:21] but UEC is not a platform to run vm , it is a platform to make ubuntu a cloud server , isn't ? binBASH [12:23] With UEC you can put servers with hypervisors (xen/kvm) together into a cloud. [12:23] so a user requests to run a virtual machine in the cloud. [12:24] UEC will set it up and run it on a random server in your cloud [12:26] is it possible to use virtblk with 8.04? I guess this is faster than hardware virtualising to /dev/sda? [12:33] RoyK: For me virtio has the same speed [12:35] binBASH: humble thought UEC make a ubuntu server to be cloud server then we can connect to our app from client , we need more than one server to have cloud computing ? and why we need vm on server ? [12:41] binBASH: ok [12:43] alabd: cloud computing is really just virtual computing gone large - same thing [12:44] humble thought UEC make a ubuntu server to be cloud server then we can connect to our app from client , we need more than one server to have cloud computing ? and why we need vm on server ? RoyK [12:45] read the docs - you can do UEC on one server, but it's not recommended [12:46] thanks a RoyK binBASH God give you peace bye 4 now [12:47] Maybe he wanted something like google appengine [12:47] :p [12:50] Daviey: around ? [12:50] ttx: \o [12:50] Got a few minutes for a mumble talk ? [12:51] ttx: sure [13:18] zul: around ? [13:18] zul: yep [13:18] You're next on Mumble. [13:18] okies gimme a sec [13:20] ttx: just staring up my laptop [13:26] hggdh, hallyn: ping me when you are around [13:34] morning [13:35] binBASH: seems slow here [13:36] are you using kvm or xen? [13:36] or, that is, it's rather fast, but the system isn't very responsive [13:36] kvm [13:36] qcow2 file I guess? [13:36] yes [13:36] did you allocate the whole space? [13:37] or is it growing file? [13:37] hm.. seems like it's a holey file [13:37] that's good already [13:37] no, a file with holes [13:37] not allocated [13:38] so called spare file [13:38] but another VM seems way faster, and that's running 10.04 [13:38] ahh, I'm running 10.04 as well [13:38] hm. I weren't using the -virtual kernel [13:40] seems virtblk is in 8.04-virtual :) [13:41] hm. not that it helps much [13:43] binBASH: No, qcow2 images are not sparse. [13:44] soren: yeah, because sparse would be slower [13:45] A sparse file is a file that is sparsely allocated, ie not their entire size has been allocated on the filesystem. qcow2 images' size increases as needed. [13:45] binBASH: Uh.. No. [13:45] Not? [13:45] binBASH: Au contraire. [13:46] inside my vms I have same speed like on the real server for hdd [13:46] also on qcow2 [13:47] hi, wanted to know if ubuntu will not be supporting Xen as like in Red Hat? tnx in advance [13:47] binBASH: How is that an argument in this discussion? [13:48] a_m_y: Depends on what you mean. [13:48] soren: Well I read somewhere once that qcow2 should be slowest. [13:48] But I dunno if it's true [13:48] And that's mostly correct. [13:49] saw about the what RH that they will not be supporting or including Xen on their latest distro, [13:49] wanted to know if it is the same in ubuntu [13:49] tnx [13:49] "slowest" != "too slow for what someone is trying to do" [13:50] a_m_y: The question remains: What do you mean? [13:50] a_m_y: Ubuntu does not run as a Xen dom0. It does, however, run as a Xen domU. [13:50] soren: Well I am no virtualization specialist yet. Just started to use KVM two months ago :) [13:50] planning to use Xen and will be using ubuntu lts 10.04. worried that in future, ubuntu will be doing the same in Xen [13:51] as like RH that will not be including or support Xen [13:51] tnx in advance [13:51] I give up. [13:52] sorry soren, dont know if am throwing the right question or am asking a stupid question. sorry [13:57] a_m_y: Well, you just keep repeating the same thing. [13:58] a_m_y: that's not going to clarify anything. [14:03] how weird. I pushed a .changes for qemu-kvm to my private ppa, got a response saying the right one was accepted - qemu-kvm for maverick. But then I got notification of failure to build: of atlas (blas/lapack HPC libs stuff) on lucid! [14:06] hallyn: You don't usually get those? [14:06] I've install iSCSI target and exported an lv, is it safe to mount the lv and us it as a standard partition also? [14:08] soren: no [14:08] soren: even if it were a valid dependency, a build for lucid ought not be related to a maverick build of qemu :) [14:08] it sounds like a complete snafu on build system... [14:08] hallyn: No no, I mean.. [14:09] hallyn: Every night, we build a bunch of packages. [14:09] hallyn: for regression testing. [14:09] hallyn: atlas is one of them. [14:09] hallyn: A bunch of us get build failure e-mails. [14:09] Oh, I see [14:09] hallyn: the PPA's are way behind today, so these uploads from last night are just finishing now. [14:09] oops. then i shouldn't have re-dput [14:10] hallyn: Won't matter. [14:10] ok, then, got it :) off to look at altas build [14:13] You got the atlas build e-mail becuase you're an indirect member of ubuntu server qa. === register is now known as Guest16349 === txwikinger2 is now known as txwikinger [14:33] *sigh* http://paste.ubuntu.com/447201/ [14:36] What's vscan? [14:39] ttx: ping as requested [14:40] hggdh: on a call -- will pong you back in a few [14:42] soren: vscan is a virus checker for samba [14:44] Neat. [14:54] hallyn, ttx, mdeslaur: re bug #591423> I don't see anything that indicates the original problem is an apparmor issue. sure, when he used a wrapper to work around the issue the wrapper was (appropriately) blocked by apparmor, but that is a different issue. I commented in the bug [14:54] Launchpad bug 591423 in libvirt "qemu -drive boot=on flag causes boot to hang." [Low,Incomplete] https://launchpad.net/bugs/591423 [14:55] jdstrand: I agree with that [14:56] oh yes, I see now that you said as much in the bug [14:56] * jdstrand wipes the sleepies from his eyes [14:57] jdstrand: eh [14:57] hggdh: got 5 min for a quick Mumble check ? [14:58] ttx: yes, but one caveat: smoser and kirkland told me I sound like a chipmunk (sounds issues on my laptop) [14:58] ttx: we can try, though [14:58] let's try chipmunk [14:59] I hear a distant voice [14:59] hggdh: let's try IRC pm instead [15:00] Good day all , is this true > UEC provide us easy cloud computing deployment and management on ubuntu [15:03] jdstrand: duh, i see. i hadn't noticed he ran that by hand. [15:04] binBASH: hi ^ [15:05] alabd: Hi [15:05] is this true > UEC provide us easy cloud computing deployment and management on ubuntu [15:05] you asked that already some mins ago? [15:05] yeah [15:08] "Ubuntu Enterprise Cloud setting the standard for easy deployment and management" is written http://www.ubuntu.com/project/about-ubuntu , so can we say regard to that , UEC provide us easy cloud computing deployment and management on ubuntu [15:10] http://www.ubuntu.com/cloud [15:11] I am having a hard time configuring a simple kvm guest for a bridge network. I think everything looks good (both on the vm creation and the bridge config), but once the vm is started the new IP (the one of the guest) never get attached to the bridged interface. Any idea ? [15:11] alabd: If you want to setup your own private cloud environment it's probably easiest way [15:11] the vm was created using vmbuildr [15:12] +e [15:13] binBASH: no no humble wrote that sentence want to see if it is true if it is then can be written in a book [15:14] a book? [15:14] yeah am writting a ubuntu book in persian [15:14] ahh ;) [15:14] that's why ... [15:14] hallyn: got 5 minutes to Mumble ? [15:15] alabd: better ask kirkland about that [15:16] don't require to setup cloud but rather require to know something about it [15:17] binBASH: you mean you don't know if that sentence is true or not [15:17] alabd: I'm not using UEC [15:17] also humble don;t want use it [15:17] I don't need it ;) [15:18] also humble don't need it [15:18] alabd: I find it very important to understand the technology your writing about [15:18] alabd: it's clear you don't - so I would not advise asking people's advice then basing a book on this as I've told you in the 10 other channels you constantly ask about this stuff in [15:18] alabd: That seems like a paraphrase of the previous sentence, you shouldnt need a support channel to tell you that. [15:19] alabd: For me simple virtualization not in a cloud is sufficient === register is now known as Guest67701 [15:20] ttx: oh, sorry, yeah - need 1 min to set up headset [15:22] hallyn: Did you see Ng's bug earlier.. Are you doing KVM sru's or just Maverick atm? [15:28] Daviey: is that the boot=on one? [15:28] ttx: metacity must have messed with my settings. it works when i'm in dwm [15:29] ikonia: you see that am not writing about UEC , but rather am writing similar sentence that is written here about uec in 1 sentence http://www.ubuntu.com/project/about-ubuntu this sentence "Ubuntu Enterprise Cloud setting the standard for easy deployment and management" but humble don't wan copy it in book but rather want to rephrase it , so humble rephrased it to UEC provide us easy cloud computing deployment and management on ubuntu , now , wan [15:29] hallyn: hmm, no... a mystery memory leak that is fixed in later release [15:30] alabd: have you spoke to the translations team ? [15:31] hallyn: it was bug #591610 [15:31] Launchpad bug 591610 in kvm "memory leak" [Undecided,Triaged] https://launchpad.net/bugs/591610 [15:31] Hi! vmbuilder script is broken in lucid. [15:31] DirtyD: How so? [15:32] --firstboot and --firstlogin options are not recognized. "no such option" [15:38] did the launch of a kvm guest should fire /etc/qemu-ifup ? I think it should right ? [15:39] but --firstboot is still broken in vmbuilder(lucid) I can launch nadda. [15:40] binBASH: thanks a lot God peace you bye [15:48] would you guys implement active/active just "because" … it seems that there's little to lose and in the future you wouldn't have to switch file systems [15:56] can anyone point me to a webpage containing the kickstart package groups ? There's not a lot of info on these :( [15:58] basically ubuntuserver and ubuntu desktop just the same right? just desktop has gui and server has server apps? [16:00] if i put server apps like apache mysql in ubuntu desktop and remove gnome its like the server release and vice versa to server if i put gnome and open office [16:02] jmazaredo: They are similar, but not completely the same. I don't know all differences, but there are other kernel settings (like the I/O scheduler) in the versions. [16:03] but in laymans term they are the same right? [16:03] just some tweaks to the kernel and some little stuff [16:05] Hi, I am having trouble with likewise-open5 and ubuntu server 10.04, I have installed the latest version from the likewise-open ppa and the default repo version with the same issues.. I have problems with dcerpcd, lsassd, lwiod, and lwregd not starting on boot. Also, I have set the default_realm=mydomain.local and it is still requiring that I type domain\user to login via ssh. Any help with this would be appreciated. Thank you! [16:21] have you guys ever encountered this situation? After an install of ubuntu server 9.10 I don't have the admin group and the sudoers file setup for the admin group. Am I missing a package when I install ? [16:22] I'm doing the install using kickstart by the way , and I'm not removing any package, just adding openssh-server and vim [16:24] wierd [16:24] never seen that [16:25] that is, I've never installed using kickstart [16:25] in the normal install, you get prompted for a username+password [16:25] that user is added to the admin group [16:25] can you have skipped that somehow? [16:28] aaaa, maybe this could be because I enabled only the root account [16:28] I think that is why [16:28] >< [16:29] just boot on an install cd or a usb stick [16:29] mount the installed root, change the root password [16:29] or even, boot into single [16:29] I'll rework that kickstart file to add a regular user [16:29] it should give you root access without problems [16:29] ok [16:29] i can nuke the server no problem [16:29] :) [16:29] thank you very much for pointing this silly mistake [16:30] np :) [16:35] Hi, I am having trouble with likewise-open5 and ubuntu server 10.04, I have installed the latest version from the likewise-open ppa and the default repo version with the same issues. I have problems with dcerpcd, lsassd, lwiod, and lwregd not starting on boot. Also, I have set the default_realm=mydomain.local and it is still requiring that I type domain\user to login via ssh. Any help with this would be appreciated. Thank you! [16:53] ttx: ping...so maverick already has the virus checker built in we just have to add man pages basically [16:53] zul: slacker [16:53] ttx: yeah so sue me :) [16:55] vmbuilder is broke? === deslector is now known as deslector|afk [17:14] if I uploaded a pkg to ppa:serge-hallyn/virt, is a bzr-accessible source automatically created? (I'm looking at https://wiki.ubuntu.com/DistributedDevelopment/Documentation/GettingTheSource as it seems i have to go through bzr to get sponsorship?) [17:15] i.e. so after I dput ppa:serge-hallyn/virt package.changes, can i do bzr branch lp:serge-hallyn/virt/package or something as an equiv to apt-get source ? [17:20] really I don't understand [17:21] eerything seems to be well configured [17:21] New bug: #591802 in tomcat6 (main) "tomcat fails to start using a security manager" [Undecided,New] https://launchpad.net/bugs/591802 [17:21] but I don't have a bridged connection [17:24] mathiaz: hey [17:24] hey mathiaz [17:24] question on sponsorship if you dont' mind [17:24] hallyn: o/ [17:24] hallyn: go ahead [17:24] hello mathiaz [17:24] if I uploaded a pkg to ppa:serge-hallyn/virt, is a bzr-accessible source automatically created? (I'm looking at https://wiki.ubuntu.com/DistributedDevelopment/Documentation/GettingTheSource as it seems i have to go through bzr to get sponsorship?) [17:24] good morning folks! [17:24] i.e. so after I dput ppa:serge-hallyn/virt package.changes, can i do bzr branch lp:serge-hallyn/virt/package or something as an equiv to apt-get source ? [17:25] hallyn: nope - I don't think so [17:25] perhaps my first q shoudl be: [17:25] hallyn: however you can push your own branch to LP before doing an upload [17:25] hallyn: so instead of starting by apt-get source package-name [17:25] hallyn: start with bzr branch lp:ubuntu/maverick/kvm [17:26] hallyn: hack on the kvm src branch [17:26] hallyn: bzr push lp:~hallyn/ubuntu/maverick/kvm/fix-for-bug-123 [17:26] and requesting sponsorship for my package is done with a keyword in the branch (automatically) right? so i must do that? [17:26] hallyn: and then dput ppa:serge-hallyn/virt, [17:26] hallyn: you must create a merge request [17:27] hallyn: by going to the LP page of the branch you've just created [17:27] hallyn: there you can create a merge request [17:27] ok, thanks [17:35] anyone use SSH, and would like to direct me to a great tutorial?? [17:35] mathiaz: hm, so... I had created my package by doing uscan. that doesn't seem to play nicely with using bzr [17:35] = [17:35] mathiaz: im' tempted to rm -rf *; rsync -va ~/original_source . [17:35] ssh is easy [17:35] and then bzr push that [17:35] ok [17:35] hallyn: hm - is it a brand new package? [17:35] just install openssh-server [17:36] then point me in a great direction [17:36] if you are connecting to it [17:36] hallyn: which package are you working on? [17:36] already done that [17:36] alright so you have an ssh client like putty or openssh-client on another [17:36] it goes [17:36] have SSH running between two machines richt now [17:36] ssh username@ipaddress [17:36] mathiaz: it's qemu-kvm [17:37] and thats all there is to it [17:37] Hi, I am having trouble with likewise-open5 and ubuntu server 10.04, I have installed the latest version from the likewise-open ppa and the default repo version with the same issues. I have problems with dcerpcd, eventlogd, lsassd, lwiod, and lwregd not starting on boot. Also, I have set the default_realm=mydomain.local and it is still requiring that I type domain\user to login via ssh. Any help with this would be appreciated. Thank you! [17:37] well... assuming its a lan [17:37] I'm updating from 0.12.3 to 0.12.4 [17:37] hallyn: ok [17:37] hallyn: IIUC 0.12.4 is a new upstream release [17:37] im trying to figure out how to transfer files between the two machine [17:37] hallyn: with a new tarball? [17:37] ah. [17:37] yes [17:37] is one of the computers windows? [17:37] hallyn: ok - thanks for outlining the background [17:38] oh so i need to just start a new bzr alrogether? [17:38] yep [17:38] use winscp [17:38] it uses ssh [17:38] winscp [17:38] hallyn: so this is what I would do to get started: [17:38] mathiaz: sorry i assumed that was always the case... [17:38] and you can read both local and remote director [17:38] y [17:38] hallyn: bzr init-repo qemu-kvm/ [17:38] i use it all the time [17:38] hallyn: cd qemu-kvm/ [17:38] hallyn: bzr co lp:ubuntu/qemu-kvm maverick/ [17:39] hallyn: bzr branch maverick update-to-0.12.4 [17:39] hallo i want to use Openlanhouse (http://trac.openlanhouse.org/wiki) for an internet access room in Togo. The problem is that i can not register any machine on the server, but when i start the program there is always are windows with the clients to register, but it does not work! Does anyone have an idea??? [17:39] hallyn: so you end up with the current package in maverick in the maverick branch and you're working branch in update-to-0.12.4 [17:39] hallyn: and all the work will be done in update-to-0.12.4/ [17:40] i cant find a page on wireless options for /etc/netowrk/interfaces [17:40] like wepkey no power management :/ [17:40] hallyn: now how to update to 0.12.4 you need to import the new upstream release tarball [17:40] hallyn: let me look up some documentation [17:41] im using Cygwin right now as my SSH server on my windows machine. [17:41] mathiaz: well the process of updating, ignoring bzr, I've got down I think [17:42] so if I can just rsync the source into my new branch that would rock [17:42] I see that winscp is a Sftp/Ftp client [17:42] Im looking to have this windows machine as my ssh server [17:42] hallyn: well - importing an new upstream release tarball should actually be done with bzr itself [17:42] hallyn: there is one command that does that [17:42] hallyn: I'm searching for the command [17:42] oh. [17:43] it says scp stfp [17:43] but it still works over ssh [17:43] hmmm [17:43] i beleive [17:43] if you have an ssh daemon running [17:43] yep but is it not just the client and not the server [17:43] hallyn: ok - it's merge-upstream [17:44] I liked the uscan+debuild approach :) [17:44] thanks, i'll look up merge-upstream and give it a shot [17:44] i have done it from the desktop not the server but under places/connect to a server/ [17:44] hallyn: well in that case s/uscan/merge-upstream/ [17:44] i could connect to an ssh server... bjt it was linux tablet [17:44] never tried to a linux box [17:45] coolio [17:45] hallyn: you need to upstream tarball (which is automatically covered by uscan) [17:45] although, if you aren't using X i don't know how to. [17:45] im still newish to linux :/ [17:45] hallyn: and you run something like: bzr merge-upstream --version 0.12.4 ../qemu-kvm-0.12.4.tar.gz [17:46] ok, thanks mathiaz [17:46] google winscp ubuntu alternative or the like [17:46] hallyn: oh - you need to have the bzr-builddeb package installed [17:46] hallyn: as this package provides all the commands used to build packages with bzr branches [17:48] brb [17:57] eh [17:57] my wireless keeps saying link is not ready [17:57] any ideas why? [18:02] I found out how to do the file transfers with in a terminal enviroment.... thanks Mgamerz [18:09] hey guys, im in a bit of a pinch, i was wondering if i could get some help with an accidental install of server [18:10] an accidental install? [18:11] ya, ill explain [18:11] wanted to run apache as a back end on my desktop. [18:11] sude tasksel added LAMP [18:11] rebooted... desktop gone... only ubuntu server cli [18:11] oh [18:11] wierd [18:11] everything is still there.. and if i try going to tasksel again and installing "ubuntu desktop" says apptidude failed 100 [18:12] apt-get install ubuntu-desktop [18:12] mushroomblue ive tried that a few dosen times.. it failes to get files [18:12] i dont know where its trying to get them from.. but its not getting them. [18:13] T3CHKOMMIE: is this a desktop or server install? [18:13] it's either trying to look in /var/cache/apt for existing packages, or trying to connect to the network for more packages. [18:13] its originally a desktop install. [18:13] but now its server boot screen and everything. [18:13] my home folder is still the same [18:14] nothing has been formatted. but i cant xstart or anything. [18:14] strange - the LAMP install shouldn't do that - next time, just install apache2, modphp etc - tasksel might have done something nasty [18:14] apt-get update failes to get stuff as well. [18:14] but for now, try to start gdm at hand [18:14] gdm? [18:14] gdm is the one starting X in the first place [18:14] just type gdm as root [18:15] ok so, sudo /etc/init.d/gdm start? [18:15] yeah, that should work too [18:15] ok, what should this gdm do>? [18:15] it should start X [18:15] login window and so on [18:16] ok so sudo xstart isnt the same then? [18:16] that won't start the login window [18:16] use gdm [18:16] ok [18:16] thanks a ton RoyK [18:17] im giong to give that a go and hopfully recover... i know my way around server cli... but i dont have the time to dink around with it now.. im at work.. cant do work :S blah. [18:17] gotta reboot thanks for the help. see yall in a bit. [18:17] T3CHKOMMIE: did it work? [18:17] it won't help to reboot - you have to re-enable gdm first [18:17] machines dualbooted im in windows right now ;) [18:17] ah [18:17] :) [18:17] brb [18:18] how do i re-anable gdm for giggles? [18:18] don't remember - they changed that on lucid [18:18] crap [18:18] or which version is this? [18:18] 10.04 [18:18] k [18:18] http://ubuntuforums.org/showthread.php?t=1477439 [18:20] thanks! [18:20] brb [18:31] im back [18:31] i changed that gdm.conf file... [18:31] rebooted. [18:31] as root did gdm [18:32] said it didnt exists... [18:32] did a ifconfig... etho doesnt exist... so thats my program. [18:32] problem* [18:34] pastebin ifconfig -a [18:35] whats that for RoyK [18:35] ? [18:35] not familiar with pastebin [18:35] you just paste something there [18:35] and give us the link [18:35] pastebin.com [18:36] its a code site of sorts [18:36] oh [18:36] !pastebin [18:36] For posting multi-line texts into the channel, please use http://paste.ubuntu.com | To post !screenshots use http://tinyurl.com/imagebin | !pastebinit to paste directly from command line | Make sure you give us the URL for your paste - see also the channel topic. [18:36] i dont think thats going to work :/ my ubuntu partition cant get on the net work period. [18:36] does ifconfig -a show any other interfaces? [18:38] ifconfig shoes my Lo [18:38] thats it. [18:38] ifconfig -a ? [18:38] didnt do that -a [18:38] :S [18:38] what about your /etc/network/interfaces [18:39] AH thats how i get to it. i couldnt remember how to do that one. [18:39] ok, im giong to give that one a shot. ill be back. [18:39] also, ubuntu links eth0 to a mac address [18:39] see /etc/udev/rules.d/70-persistent-net.rules [18:40] is there a page on wireless options in that file? [18:40] in what file? [18:40] the udev one? [18:40] no [18:41] network interfaces [18:41] im in server and gotta set up wireless [18:41] dunno - never used wireless on servers [18:41] also need to turn wireless-power off [18:41] that is, I have, but that was sime time ago [18:53] Daviey: fyi, I've had the 4-way kvm spew test running for 9 hours and 48 minutes, which is at least an hour longer than the leaking run last night and there is no leaking going on :) [18:54] now back to investigating why I'm getting ext4 corruption in a couple of the guests [19:09] Hey guys, I guess I'll go ahead and ask. I just got fowarded to this channel from #ubuntu-devel, so maybe somebody here can point me in the right direction. The thread I posted yesterday should sum up what I'm wanting to know. http://ubuntuforums.org/showpost.php?p=9432432&postcount=1 [19:10] !here | saycheeze [19:10] saycheeze: Please give at least an overview of your problem *here* (all in one line) - you will get a much greater audience. If you have to use more than 3 lines, please use http://paste.ubuntu.com [19:12] Basically, I want to get involved in the development of Ubuntu. The thread explains my skills and weaknesses. [19:16] saycheeze: and they sent you *here*? this is a support channel -- did you see https://wiki.ubuntu.com/UbuntuDevelopment ? [19:16] Yep, that's exactly what I was thinking. [19:17] And yeah, I have. I'm just trying to figure out where to go from here (having all the tools and such) [19:17] I might have missed something in documentation that would explain things properly, there's no telling. I'll check it out again, anyways. [19:19] saycheeze: try #ubuntu-devel, its a channel with 324 users [19:21] 24 [19:21] Yeah, I got sent here by those guys [19:21] er, oops. [19:21] Go figure [19:21] saycheeze: I think they may not have read your actual post. -devel is really the best place to start for this sort of thing. [19:22] haha Yeah, that's exactly what I was thinking [19:22] It makes sense to talk about development in the development channel haha [19:22] Alright, guys. Thanks for the help, I'll see ya around! [19:25] how do you build a software raid 10 manually and mount it to /var. [19:25] <_ruben> wasnt this the original -server development channel? i heard rumours it being more and more about -devel than -support :) [19:26] <_ruben> gswain: man mdadm .. there's quite some options to choose from [19:26] i only found out about this in the server guide... so not according to the server guide [19:26] <_ruben> *confused* [19:28] i tried it with the installer several times and though it looks good when it boots up it complains about a degraded array and says its inactive. which makes sense because there is no way that i 20 minute install builds, syncs a 4x500gb raid ten array and formats it [19:29] <_ruben> syncing doesnt have to be completed prior to formating/installing/rebooting/etc [19:29] huh weird [19:30] even still it seems formatting should take longer than the 2 minutes the installer gives it [19:30] *shrug* [19:31] alright [19:32] so i am trying to install the proprietary nVidia driver for my chipset [19:32] and it keeps saying that it failed [19:32] I downloaded the 64bit linux GeForce 6 series (it says this is a GeForce 6100) [19:32] and it keeps failing [19:33] mgamerz: As Ubuntu Server doesn't use a gui, you'd probably be better off asking in #ubuntu [19:34] i'm going to install a gui [19:35] well [19:35] i need the driver because i'm going to use this to record TV [19:35] mgamerz: then install ubuntu desktop and add the needed services [19:35] GUI stuff is off topic here. [19:35] oh [19:35] ubuntu desktop and server are about the same thing [19:35] but people in here, don't use GUIs for their servers [19:36] well i won't mainly [19:36] i just want to have the driver [19:36] beause i need to have hardware acceleration for some features im using [19:36] it's easily installable from the ubuntu desktop gui [19:36] but that's a desktop thing, not a server thing === mgamerz is now known as Mgamerz [19:38] it should alreeady have the propriatory nividia driver from the install of the desktop install [19:39] if not, apt-get install ubuntu-restricted-extras [19:39] going through the install again and it lets me create mirrors just fine and i have 4 partitions left all the same size i can see them all and they are unassigned to any raid devices [19:39] yes [19:39] when i go into create raid 10 [19:39] it only shows me 3 of them [19:40] are they visible in /proc/partitions? [19:40] any chance the raid 10 in the server install cd is a little buggy [19:40] lemme check [19:41] no sdc doesnt appear to be partitioned [19:41] however [19:41] in the ui installer it shows it [19:41] as being partitioned in 2, a 10 gig and a 490.1 gig [19:41] gswain: Is that software RAID? [19:42] RyanP: yea [19:42] i know these drives are good because the raid 10 setup works fine for centos10 [19:42] same schema [19:49] also i dont get why sdc isnt partitioned as it shows in the ui as part of md1 which is a mirror for swap [19:50] is there something like mdadm --detail --scan that actually shows which partitions each array is using [20:01] New bug: #591875 in mysql-dfsg-5.1 (main) "mysql_install_db run without --user in postinst" [Undecided,New] https://launchpad.net/bugs/591875 [20:04] gswain: cat /proc/mdstat ? [20:35] mathiaz: http://www.debian.org/volatile/ [20:36] zul: oh cool [20:36] zul: thanks for the update [20:36] zul: feel free to jump into the conversation next time [20:36] mathiaz: sorry for eavesdropping [20:36] zul: well - you're not eavesdropping [20:36] I am [20:36] oops [20:37] zul: if we have the conversation in the channel it's meant to be public :) [20:37] zul: like IRC anyone can jump into [20:37] heh [20:37] im still getting use to it [20:37] zul: you basically maintain php for debian/ubuntu - right? [20:38] mathiaz: i have basically two earphones on right now [20:38] MTecknology: basically yes [20:39] zul: That's amazing. I tried tweaking it some and about shot myself. That is really a beast of a package. [20:40] MTecknology: yes it is...its caused me to miss many AA meetings [20:40] lol [20:40] zul: at least you're not a quitter [20:44] zul: Managing that best is amazing. Good work on it. If I had a clue I would have of3fered to try to help. [20:45] MTecknology: thanks [20:51] is there anyway to create partitions using MB/GB in fdisk.. ive no idea how many sectors or block i want a given partition to be [20:52] gswain: +3G [20:53] gswain: start cylindad = default; end cylindar = +20G - gives you a 20GB partition [20:56] oh sweet thanks [21:14] hello channel, I have a noobish question.. .when I run aptitude update, sometimes I get a positive or negative value in brackets at the end like this "Current status: 8 updates [+6]" ... what exactly does that value in the [ ] mean? [21:16] bventura: the change in the count during the last action [21:16] aha [21:16] thanks, i see [21:17] bventura: it took me some time to see the connection [21:18] yeah I've been wondering this.. it's a hard thing to articulate to google hehee [21:19] bventura: :) === dendro-afk is now known as dendrobates [21:30] RoyK, you still there? [21:30] yep [21:31] ok i check out /etc/network/interfaces eth0 didnt exhist... so i created it and set it to dhcp. [21:31] i got an ip and was able to tasksel and install ubuntu desktop. [21:31] unfortunatly, i cant get the desktop to load, even if i reboot [21:31] any ideas? [21:32] gdm doesnt get anything going either [21:32] :( [21:33] type gdm and press enter [21:34] yup i did that too [21:34] gave me some sort of display warning... flickered then back to cli [21:34] i dont think there is much hope for me on this one [21:36] T3CHKOMMIE: if you can reinstall and then add apache with 'apt-get install apache2' and so on instead of using tasksel, it'll probably clean up things [21:37] I really don't have too much experience with using desktops as servers or the other way around [21:37] reinstall like, boot from a cd install partition.. format? [21:37] if you're not really into linux, debugging this will take longer than a reinstall [21:38] ive been doing this for a while... i just didnt want to have to reconfigure everythign... [21:38] check the X logs [21:38] but thats ok. ill just burn myself another cd and get on with it. thanks for the help tho! i appriciate it. [21:38] it's probably failing on something [21:38] ok ill give that a shot. [21:39] thanks again guys! i really appricate the help! [21:47] im seeing reports of xfs running a little buggy on lucid? is anyone running it in production successfully here? [21:47] or have anything to add [21:49] hi guys, someone know how to solve the problem of connection refused for ssh in instances of UEC? [21:50] openssh installed? [21:50] port 22 open? [21:52] tyska: might be a good idea to euca-get-console-output for the instance [21:52] is ext4 generally considered safe now since its the default fs in ubuntu server? [21:52] gswain: yes [21:53] but, of course, it is your decision on which FSs to run [21:53] i was wondering if xfs might be a better option for storing vms on since it seems to excel at large files but im not sure how stable it is [21:55] hggdh: i have changed the public IP of my CC, im thinking could be this the cause of my error, before the change of the IP address everything was fine [21:55] tyska: it might be related, yes (routing?) [21:56] hggdh: yeah, but i have a private interface on my CC that i use to communicate with nodes, and the public interface (CLC) to have Internet Access [21:57] all CLC and CC in the same server [21:57] tyska: can you access the instance from the CLC? [21:57] using either of the IPs? [21:58] no, im trying to access from the CLC/CC server [21:59] i can ping but i cant connect with ssh [22:03] tyska: (1) euca-get-console-output; (2) ssh -vv to it, and check the output; sudo traceroute -T -p 22 [22:15] hggdh: i deregistered the node, reboot the servers, and know the error changed [22:15] hggdh: now its Permission denied (publickey). [22:32] I'd like to use kickstart to pre-seed installations, however I need to be able to have an encrypted / partition. I've tried the options listed on the fedora page for kickstart, but they are not working [22:32] anyone know how to encrypt / via the kickstart file? [22:38] Brian_H have you checked the kickstart known limitations in the server guide? [22:38] i think i recall that feature not being implemented as of yet [22:39] gswain, looking now [22:39] id love to use kickstart to preseed but it has some pretty bunk limitations thus far === dendrobates is now known as dendro-afk [22:40] its much easier and cleaner, at least from what I can see, than using preseed :/ [22:57] <84XAAGJVC> has anybody install Livezilla as server in ubuntu? === 84XAAGJVC is now known as rgotten [23:04] has anybody install Livezilla as server in ubuntu? === AdamDV is now known as Guest34939 [23:38] is it reasonable to expect a P4 with 512MB of RAM to be able to run Ubuntu Server 10.04 smoothly? [23:39] rottenrecords, depends what you do with the server [23:40] incorrect: well, I just want to use it as a development web server (so MySQL + Apache + SVN) [23:40] rottenrecords, it depends entirely upon what services you're trying to run on it [23:40] we have it up and running, but it's been really, _really_ slow [23:41] doing what? [23:41] i am sure it will be just fine [23:41] being idle [23:41] some tasks are cpu heavy (database queries), some are ram heavy, hd heavy, etc [23:41] i'm just talking about logging into it over ssh [23:42] from elsewhere on the lan or remotely [23:42] after i type in my username, it takes about 5 seconds for it to prompt for password [23:42] maybe you have balls DNS [23:42] that is nothing to do with cpu or memory [23:42] on the lan, from a computer 2 feet away [23:42] yeah [23:42] also, blacklist IPv6 if you don't use it [23:42] that is because ssh is looking up your host entry i bet [23:42] there is something you can do to turn that off [23:42] but you will find it only irritating once [23:43] a lot of those kind of ssh login authentication delays have to do with timeouts related to IPv6 lookups [23:43] i see [23:43] i will look into that [23:43] the other thing is, the temp installed Gnome of the server [23:43] however its not anything to do with how well a p4 can run you php code [23:43] also, depending on what you're trying to do remotely, you can run ssh with compression using the -C parameter [23:44] makes it easier when trying to do a graphical session remotely with something like vnc or x11 forwarding [23:44] now that i think about it, maybe dns/host lookup is the issue [23:45] because the php code is only taking 24-120ms to run [23:45] and most of that is due to the mysql queries [23:45] if you can't configure a working dns server just hack your hosts file [23:45] that's what i'll probably do [23:46] but should Gnome run very slowly on a server with these specs? [23:46] because it takes about 2 seconds just to open up a terminal window [23:46] depends if its out of ram [23:46] i have all visual effects disabled btw [23:46] * incorrect shrugs [23:47] i've not run a P4 for years [23:47] but i don't suppose its unusable [23:47] rottenrecords, doing gnome remotely isn't the best choice, but it can be done with certain servers if you do things like setting up server-side caching and limit colors from the client side [23:47] a3ist: i'm not doing it remotely [23:47] rottenrecords, in that context I meant "remotely" as any connection that's not a monitor/keyboard physically plugged into it [23:47] i only use Gnome when I'm on the workstation hooked up to the kvm the server is connected to [23:48] oh [23:48] well, try a lightweight DE like xfce (low reqs) or lxde (even lower) [23:48] both are full desktop environments [23:48] thanks, i'll check those out [23:49] going to even lower requirements, check out tiling window managers like awesome (I use it, and it is indeed awesome), dwii, xmonad, icewm, etc [23:50] this link isn't *technically* for ubuntu but a lot of the items are in the repos: http://wiki.archlinux.org/index.php/Comparison_of_Tiling_Window_Managers [23:50] these don't usually come with things like file managers or panels and stuff (some do) but can still run them [23:51] so you can pick and choose which components you want, to keep things lightweight [23:51] and since they're still running on x, you can still use graphical apps like gtk apps [23:51] hrmm [23:52] for context, I think the "awesome" window manager is about 4 megabytes altogether with a memory footprint of like 300kb [23:53] they're that lightweight [23:53] yea, i think that'd be a huge improvement over gnome [23:54] which doesn't really work right on this server anyway [23:54] like the user/groups admin GUI wont even open [23:55] and before it was locking us out whenever the system resumed from hibernate or whatever [23:56] well, you can turn that off in the screensaver panel in gnome [23:56] but I think ditching gnome on a computer with those specs would be your biggest priority :) [23:56] heh [23:57] well, thanks for your guys' help. i'm gonna get started on removing gnome off of here. [23:57] if you want to stick with a "normal" desktop environment instead of a tiling WM, like I use, lxde has the lowest overhead of any of the "full" DE's [23:57] up to you of course [23:57] good luck [23:58] thanks [23:58] peace