[00:01] it's 1am here, you know :) [00:03] ah sorry, you'll be sleeping soon. [00:03] well, don't wait on me. i'll figure it one way or another. i'll be here tomorow as well.. i need to get this done this week. [00:04] run pvcreate with -vvv [00:04] ok [00:04] it might give you a clue what's going on [00:05] thank you for the insight. [00:05] you've already created md0? [00:05] yes, it should be there.. i just deleted and recreated it in the installer [00:06] when i used the installer to set the LVM on it - it said the same thing in the command line (couldn't find /dev/md0 or it was filterd) [00:07] run pvcreate with -vvv [00:07] i'm not there yet, but when i get there, i will. [00:07] hey guys i am having issues with cron.d not working…. i have */5 * * * * www-data php -q /var/www/mydomain.com/tickets/public/api/cron.php [00:07] rebooting now [00:08] it doesn't seem to run [00:08] webPragmatist: use full path to php [00:08] and check /var/log/syslog [00:09] ivoks: okay thanks [00:10] oh i get Error: bad username; [00:12] mdadm -D /dev/md0 .. currently says it's rebuilding (38% complete) [00:13] AndyGraybeal: that could be a reason why it isn't working [00:13] okay, i'll wait for it to build. [00:13] try pvcreate anyway [00:14] oh idid..it says the same thing; and -vvv is lots of data, how do i redirect it into a file? [00:14] oh on boot up i noticed an error cannot find /dev/null ... i was like.. that sounds bad! [00:14] but it's there.. so i odn't know [00:16] 10 years and you still don't know how to redirect stdout? [00:16] bad windows sysadmin :) [00:16] well i tohught it was like > filename [00:16] 1> name_of_the_file 2>&1 [00:18] k i've never seen that before, thank you. [00:22] here's the pastie: http://pastie.org/1061328 .. there's a lot more devices it lists.. but i figure you'd get the drift from here - i think this is the meat of the matter. [00:23] it says partition table signature found, skipping. is that the problem? [00:23] yes, it is [00:23] is that resolved with parted? [00:24] mdadm --stop /dev/md0 [00:24] what disks/partitions are in /dev/md0? [00:24] which [00:24] i'm not sure, how do i find out? [00:24] i thought i removed all the partitions. [00:25] i thought i actually re-created the whole array when i re-installed. [00:25] i mean, i remember going through the motions atleast. [00:26] fdisk -l doesn't show anything [00:27] dmesg | grep '320 GB' [00:27] but maybe tha'ts cause i --stoped it [00:28] did you run what i said? [00:28] http://pastie.org/1061336 <--- yes [00:28] yes, sorry it took me a second [00:29] so, it'd sdb->sde [00:30] fdisk -l /dev/sdb [00:30] fdisk -l /dev/sdc [00:30] fdisk -l /dev/sdd [00:30] fdisk -l /dev/sde [00:30] check if there are any partitions on those disks [00:30] if there are, delete them [00:30] okay [00:30] it looks like the block size is different in both those pasties, the last one is 512bytes, and the first one is 4096bytes (i don't know if this has anything to do with it) i'm going to do the fdisk -l now. [00:31] they all have: Linux raid autodetect as the "System" [00:32] paste output of one of those disks [00:32] k [00:32] http://pastie.org/1061344 [00:32] delete those partitions [00:33] or [00:33] wait, don't [00:33] k, yea, i htink i need them for the RAID [00:33] not really, but anyway [00:34] okay, yea, i have no idea. [00:34] mdadm --zero-superblock /dev/sdb [00:34] mdadm --zero-superblock /dev/sdc [00:34] mdadm --zero-superblock /dev/sdd [00:34] mdadm --zero-superblock /dev/sde [00:34] ok [00:34] oh maye i need to turn md back on? [00:34] mdadm: Unrecognised md component device - /dev/sdb [00:35] did you stop the array? [00:35] you had asked me to, i haven't startd it back up yet [00:35] ok [00:35] try with sdb1 [00:35] should i start the array? [00:35] oh sdb1 nm [00:35] worked [00:35] i'll keep going [00:36] right [00:36] k, all done [00:37] should i try pvcreate now? [00:37] er i guess i should start the array, but i'll wait for further notice. [00:37] mdadm -A /dev/md0 [00:38] then pvcreate [00:38] k [00:39] after: mdadm -A /dev/md0 .. it says: mdadm: no devices found for /dev/md0 [00:39] gah! [00:39] mdadm -A --scan [00:39] lol, i'm stupid :) [00:40] same answer [00:40] you've cleand the superblock [00:40] of course it won't work [00:40] okay. [00:40] so recreate the array? [00:40] ivoks, sorry had an appointment, donno if it boots from the disk in fact because the grub screen never shows [00:40] mdadm --build /dev/md0 --level=5 --raid-devices=4 /dev/sdb1 /dev/sdc1 /dev/sdd1 /dev/sde1 [00:40] ah --level=10 for me [00:40] ivoks: heya...dovecot 2.0 is suppose to be out in a couple of days...whadayathink? [00:41] unless you think level=5 is better [00:41] ivoks, I even went in the bios to tell it to boot from second hard drive [00:41] lox_: did you replace the disk with boot partition? [00:41] zul: i was a bit off for some time; i have to catch up with what's new [00:41] ivoks, yes it has a boot partition with bot flag (the boot is in the raid1 too) [00:42] And I already copied the parytition table of sdb to the new sda [00:42] AndyGraybeal: i was under impression that you are building raid5 [00:42] raid10, and it says to use 'create' instead of 'build' [00:42] right, create [00:42] go ahead with that, otherwise, i'll take your recommendations. [00:42] ivoks: okies...i was tempted to [00:43] zul: we might :) [00:43] zul: give my just some time to examine everything [00:44] ivoks, everything is checking out; should i wait for rebuild or should i just go ahead and try pvcreate [00:44] ? [00:44] ivoks: no problem [00:44] AndyGraybeal: do pvcreate [00:46] ivoks: same response: /dev/md0 not found or ignored by filtering; .. the -vvv returns the same; partition table signature found, and fdisk -l reports the same: Linux raid autodetect. [00:46] time to hit the bead... [00:46] bed [00:47] alright, thank you for your help ivoks. [00:47] AndyGraybeal: then just wait for it to build [00:47] okay thank you [00:47] never did md level 10 [00:47] okay; i don't know what to think about it in comparison with 5 [00:48] paste output of cat /proc/mdstat [00:48] i used to do 5 with evering [00:48] http://pastie.org/1061358 [00:49] What would be a good way to do a VoD/File Server/Voip System [00:49] I was originally going to use xbmc for VoD [00:49] AndyGraybeal: wait for it :) [00:49] good night [00:49] but then I just went with simple file sharing [00:49] ivoks: thank you [00:49] for VoD and file server [00:50] but I feel like that is a very poor solution [00:50] Jinxed-: what is VoD ? [00:50] AndyGraybeal, Video On Demand [00:50] basically to allow for streaming video [00:50] aah ok, thank you === mdeslaur is now known as mdeslaur-afk [00:51] aka I don't want to broadcast, I want to allow remote users to request a stream [00:51] Any suggestions? [00:51] well, your aware of asterisk for voip, i'm sure. i've set that up before and got it to work.. so if i can do it, it must be fairly straight forward. [00:52] haha [00:52] yeah I was planning on using asterisk [00:52] i shouldn't be allowed to answer questions though, they should keep me in a cage and ask the on-lookers not to feed me. [00:52] i was even tempted to try * with vmware so I could use AsteriskNow [00:52] fileserver, i guess would be samba or nfs if your all *nix [00:53] i would use asterisk straight up, it's fairly easy. [00:53] the guys in the channel are pretty nice too [00:53] #asterisk that is [00:53] they told me to use AsteriskNow [00:53] but yeah I would be willing to try Asterisk with ubuntu [00:54] asterisknow is linux, vmware is not needed correct? i would us KVM and not vmware, but i'm sure that could be a holy war in a bottle. [00:54] someone mentioned trying to use zfs [00:54] zfs is more for bsd, but i'm sure someone here can answer better than myself. [00:54] haha, i actually was going to use virtualbox [00:54] for virtualization [00:54] so I could just basically have the two OS [00:54] i wouldn't run any server program in virtualbox; i would stick with kvm, but like i said, i shouldn't be answering questions. [00:55] haha [00:55] alright [00:55] you should wait for osmeone else to respond before you listen to me, seriously [00:55] It seems like simple file sharing (right click share folder) is a poor system [00:55] ivoks, so, no idea why it won't boot from sdb ? [00:55] lox_: ivoks went ot sleepytime [00:56] AndyGraybeal, oh, ok ... Can you help me with my raid problem ? [00:56] lox, no no no bro.. i'm trying to strangle myself with mdadm as we speak. [00:57] AndyGraybeal, so what server features should I install on top of the regular desktop installation [00:58] Jinxed-: i would do the opposite, install server, then install ubuntu-desktop ontop of that. but you'd probably want atleast openssh-server, you'd need the KVM stuff for virtualization if that's where you wanted to go. [00:59] Jinxed-: i run samba for fileserver but only because i got to run some windows boxes.. i'd probably do NFS otherwise... i've never touched ZFS, so i can't tell you what about that. [00:59] my finance ladies won't learn openoffice and i swear my hair falls out when i run MS Office in wine... so windows boxes it is (even if they are KVM'd) [01:01] ha, well currently i just have ubuntu desktop [01:01] my whole group is fine wiht openoffice accept the finance ladies.. makes me angry, they gang together.. saying that the calc, now called spreadsheet or something is too different and they don't want to learn something different. [01:01] i don't have any server features (except for vlan and samba installed) [01:03] sudo apt-get install openssh-server ... atlest. [01:03] you should read about KVM and libvirt; ubuntu has nice docs! [01:03] you happen do be able to donate a link? [01:03] i use kvm with libvirt, i like 'virsh' the command line that controls the virtual machines. [01:03] :p [01:03] yea hold [01:04] https://help.ubuntu.com/community/KVM [01:04] there is also #ubuntu-virt and they are huge help [01:05] and #kvm and #libvirt if i recall correctly [01:05] lox_: i'm building my array as we speak, 38% done. [01:07] Heya. :) [01:09] hi beefstake :) [01:13] AndyGraybeal, can i rebuild my array from livecd ? [01:14] lox_: if it's a mdadm raid array then you should able able to yep. [01:14] beefstake, hi, yes it is [01:14] Mhmm yep you should be fine. [01:15] beefstake, I needed to replace sda and do removed it from the array, shutdown, replaced disk and it won't boot from sdb [01:15] RAID 1 yep? [01:15] yes [01:15] Did you have the md device set as the boot device or one of the raid members? [01:16] beefstake, yes [01:16] a boot partition in raid1 and a / partition on raid1 too [01:16] Ahh kk yep. [01:17] beefstake, but won't boot from sdb and I am stuck there [01:20] lox_: What you need to to then is boot up the live cd [01:20] Then force assemble the array. [01:20] beefstake, done [01:20] how ? [01:20] http://gist.github.com/491501 [01:21] Ahh like sdX5 are the / partitions? [01:21] yes, what is wrong with it ? [01:21] Well you need to do both anways [01:21] So you do it like this: [01:22] Hmm wait, have you recreated the partitions on sda yet? [01:22] yes [01:23] using sfdisk [01:23] Kk cool. [01:23] mdadm --assemble --force /dev/md0 /dev/sda5 /dev/sb5 [01:23] http://gist.github.com/491506 [01:23] Should complain about not having a superblock. [01:24] true [01:24] mdadm: no recogniseable superblock on /dev/sda5 [01:24] mdadm: /dev/sda5 has no superblock - assembly aborted [01:24] Yep k. [01:24] mdadm --assemble --force /dev/md0 /dev/sb5 [01:24] We will add sdb in just a sec. [01:25] mdadm --re-add /dev/md0 /dev/sda5 [01:25] Try those 2 [01:26] $ sudo mdadm --assemble --force /dev/md0 /dev/sdb5 [01:26] mdadm: cannot open device /dev/sdb5: Device or resource busy [01:26] mdadm: /dev/sdb5 has no superblock - assembly aborted [01:26] sudo mdadm --re-add /dev/md0 /dev/sda5 [01:26] mdadm: cannot get array info for /dev/md0 [01:26] Wait, what. [01:26] Which is the drive that didn't fail? [01:26] sdb [01:26] That can't be right.. [01:27] I hotremoved sda, shutdown and replaced it, then rebooted [01:27] Hmm. [01:27] Has it won't boot, I loaded live cd and copied partitions to sda [01:27] yeah. [01:27] cta /proc/mdsat gives nothing from livecd [01:27] The issue here is that sdb is reporting that it doesn't have a raid super block. [01:28] what did you use to copy partitions? dd? [01:28] Yeah because no mds are running. [01:28] no sfdisk [01:28] lox_: aah.. *googles [01:28] Nah you shouldn't dd them, you should only copy acroos the partition layout with sfdisk. [01:28] This is what I have done [01:28] mdadm should resync the disks once you readd them to the array. [01:28] I used sfdisk and copied only the layout [01:29] The issue here is that sdb is saying it doesn't have a superblock... [01:29] it worked yesterday, I used exactly same process to replace sda (and of cource I waitied resync to finish) [01:29] (needed to replace both because of SMART errors)7 [01:29] Mhmm. [01:29] But it booted on sda when removed sdb [01:30] now it won't booton sdb, why ? [01:31] Is that normal (from livecd): [01:31] $ cat /dev/md0 [01:31] cat: /dev/md0: No such file or directory [01:34] $ mdadm --detail /dev/md0 [01:34] mdadm: cannot open /dev/md0: No such file or directory [01:34] is that normal ? [01:36] beefstake, can I try mdadm --create /dev/md1 --level=1 --raid-devices=2 /dev/sdb5 /dev/sdb5 ? [01:36] Or it may erase my data ? [01:39] Yeah don't use create. [01:39] Create is bad. [01:39] beefstake, why don't I have /dev/md0 when I boot with the livecd ? [01:42] sounds scary lox_ do you have your data backed up? [01:42] arf ... part of it [01:42] hopefully beefstake can help :) [01:42] i'm sure he can. [01:42] i'm on the edge of my seat hoping there is a good outcome. [01:43] The array was working so nicely ... [01:43] how long? [01:43] many month [01:43] nice [01:44] sudo mdadm --assemble --scan [01:44] mdadm: /dev/md0 has been started with 1 drive (out of 2). [01:44] mdadm: no devices found for /dev/md1 [01:44] cat /proc/mdstat [01:44] Personalities : [raid1] [01:44] md0 : active raid1 sdb1[1] [01:44] 476519872 blocks [2/1] [_U] [01:44] [01:44] unused devices: [01:44] beefstake, ? [01:45] that means all your data is there, correct? [01:45] I miss sdb5 [01:45] where all data is [01:46] oh tru my data is here [01:47] and boot is on it too [01:47] maybe back up now before things get worse [01:47] so sdb5 could be swap [01:47] AndyGraybeal, sure ;) [01:47] i hate living by a thread :) [01:48] AndyGraybeal, should I use dd ? [01:48] the fastest backup solution will be the best ;) [01:48] oh i have no idea, maybe that find and cpio or rsync to another drive just to be safe. [01:48] you should lisgtne to someone here, not me. i'm a newb [01:48] dd would be slow i assume. [01:49] you saw how ivok was scolding my ignorance :P [01:50] ;) [01:54] Backing up [01:55] What is the best way to handle swap partitions in a raid1 array ? I have added the swap partition to the raid1 aray is it theway to do ? [01:56] what are you using to backup (curious)? [01:56] rsync [01:56] I could have used cp -a [01:56] i've been reading that cp -a isn't a good solution, that find and cpio is better, or rsync. [01:57] but i'm not sure honestly [01:57] Basically rsync does what cp -a does with some feature like incremential backup [01:58] http://www.mikerubel.org/computers/rsync_snapshots/#Rsync [01:58] okay [01:58] AndyGraybeal, but for everyday use, I use rdiff backup [02:00] ah nice [02:00] yea, i just read now that rsync is a bad idea because of the overhead it might take while trying to 'sync' to a empty destination [02:00] so many variables :) [02:01] rdiff backup only backs up diff changes between files, so I can backup every two hours with it [02:03] I stil don't see why it is not booting from sdb alone, everything is there: grub, boot flag [02:12] okay, my problem is that i can't seem to build a LVM on top of a newly made array. when i try 'pvcreate /dev/md0' it says: "Device /dev/md0 not found (or ignored by filtering)" but that error is a little bogus because when i do pvcreate -vvv /dev/md0, it says: " /dev/md0: Skipping: Partition table signature found" .. so this is where i am stuck. i don't know what to do. fdisk says this: http://pastie.org/1061462 ... [02:13] if anyone can help me get rid of this GPT on my raid array, that might move me forward in creating an LVM on it. [02:15] AndyGraybeal, sorry I dono LVM, never foudn a use to it ... [02:16] lox_: thank you for the response. yea, i was going to give it a go, see how things worked out. [02:16] What do u need lvmfor ? [02:17] an added layer of failure (in most cases) [02:18] lowridah, that is whta I miss it seems, for me LVM is just usefull for resiing patitions [02:18] lox_: yea, i hear i can resize partitions on the fly too, that sounds nifty [02:18] lox_: i don't know, maybe it's another layer of confusion i don't need, but i was gonna give it a shot this time around. [02:18] AndyGraybeal, a long time I have not used many parition for one system.... [02:19] i want to have my /home and /srv as seperate partitions [02:19] i got 70 users and i don't know how things will go in the future, so we'll see. [02:21] AndyGraybeal, what is /srv used for ? I only saw ebox using it until now ... === jjohansen is now known as jj-afk [02:21] Guys.. I need some help with a EUC setup. I've been pouring over the docs and found that I shouldn't be makeing edits to the network stuff in eucalyptus.local.conf but should be using eucalypts.conf (acconding to open.eucalyptus.com).. hovever, the changes I make there never seem to get into the local file and the local file takes precidence it seems. [02:21] Now... the local files has a big warning about not changing it directly.. so WHERE THE HELL DO YOU CHANGE THINGS??? [02:21] I'm trying to setup some IP's that are public so I can see the machines from the outside [02:23] lox_: i'm gonna use it for all my users data.. aside from their /home (personal data) of course. my computer acts as a file server.. so that's where their shared files will go. [02:23] lox_: read .. file system hierarchy and it gives and explaination [02:23] lemme bring up the url [02:24] http://www.pathname.com/fhs/pub/fhs-2.3.html#SRVDATAFORSERVICESPROVIDEDBYSYSTEM [02:26] AndyGraybeal, thks for the link, bookmaking it [02:27] yea, i think i'm gonna head out -- i will get back on this in the AM here on EST. [02:28] lox, what is ebox? [03:40] hello people [03:45] jeez those f.... centos people are idiots [03:45] Lol [03:45] What did they do now? [03:47] wizardslovak: I found a lot of them nice to talk to now and then [03:48] huh [03:48] ask question ? [03:48] bs [03:48] i did and none to help , just started bitching about how untechnically i am and how english is barrier to learn , wtf? [03:51] Lol. [03:51] Fail. [03:51] What did you need to know? [03:51] I use CentOS a fair bit. [03:51] damn sudoers file [03:51] but forget it , ill use ubuntu from now on [03:52] beefstake, u have a few minutes? [03:52] also that ubuntu is as easy as windows , so no technicall people stop on debian , huh as i said idiots!!! [03:53] lox_: Yeah, you still having trouble with your raid? [03:53] beefstake, mdadm --assemble --scan found my md0 aray but not the md1 (swap) [03:53] I added sda to md0 [03:53] it syncs as I speak [03:54] lox_: Nice, you are home free then. :) [03:54] But what about md1 (swap) can I start without it (still in livecd session) [03:54] That's fine, your system will boot without swap. [03:54] ok [03:55] now what do u think of my weird problem of not being able to boot on sdb ? [03:55] md0 is in grub [03:55] boot is in md0 [03:55] always liked ubuntu people , so kind ;) thx whoever invented ubuntu [03:55] it has grub in it [03:55] Anyone have IS C config and Ubutun 9.x and upgrade to 10.x and feel good? or were there issue or major problem [03:55] and the boot flag is set on the partition [03:55] Mhmm.. [03:55] beefstake, I ask it because of what could happen if sda fails again ... [03:55] lox_: That shouldn't have been an issue though. [03:56] lox_: What should normally happen is you boot with the array in a "degraded state" [03:56] Meaning that a disk has failed. [03:56] Not big deal, it should just boot. [03:56] I think something went wrong when you were removing disks or something. [03:57] beefstake, you have to grub-install to each disk you want to be bootable [03:57] to the MBR [03:57] qman__: Ahh completely forgot to remind him of that. [03:57] otherwise the BIOS loads the MBR of your other disk and doesn't find anything to load [03:57] beefstake, thks bro, I hope it will boot after sync, I think the issue may be the bios ... [03:58] lox_: As qman__ said you need to do grub-install on each disk in the raid. [03:58] You will need to reinstall grub to the MBR of sda now you have replaced it. [03:59] do I just do grub-install /dev/sdb1 ? [03:59] from the livecd ? [04:00] You will probably have to mount it first but yep. [04:00] pretty much [04:00] I can't remember if there's more to it, but that's the idea [04:00] did anyone used ubuntu jetos? [04:00] and no, it doesn't need to be mounted [04:00] i mean jeos [04:01] qman__: Yeah? How does grub-install determine the bios disk of /boot then? O.o [04:01] sudo grub-install /dev/sdb1 [04:01] /usr/sbin/grub-probe: error: cannot find a device for /boot/grub (is /dev mounted?). [04:01] No path or device is specified. [04:01] I cannot mount sdb1 it is a raid partition [04:01] oh yeah [04:02] that won't work unless you're chrooted into your install [04:02] Hmm might have to do it by hand. [04:02] what will work though [04:02] is if you run grub, then set it up [04:02] Yep. [04:02] and manually install to each disk [04:02] but grub is only going to see md0. isn't it ? [04:03] no [04:03] quite the opposite, grub doesn't know about md0 [04:03] at least not at that stage [04:03] grub sees your physical disks [04:04] oh ok [04:04] what about grub-install --root-directory=/mnt /dev/sdb ? [04:04] with md0 mounted in /mnt ? [04:04] no, that's wrong [04:04] --root-directory is the root= parameter [04:05] tur [04:05] true [04:05] Nope. [04:05] lox_, http://ubuntuforums.org/showthread.php?t=224351 [04:05] follow that first procedure [04:05] --root-directory is for installation of grub image files to /boot/grub [04:05] so n oone used it [04:05] Jeos? Nah I haven't. [04:06] changing `setup (hd0)` for each disk you want [04:06] what software you need to setup ubuntu as media server? [04:06] qman__, thks [04:07] wizardslovak, that depends on what you mean by media server [04:07] NAS and able to play media in tv [04:07] qman__, will it apply to grub2 too ? [04:08] lox_, I'm not sure, grub2 is a whole different animal [04:08] I run lucid so it it grub2 I think [04:08] wizardslovak, do you just want to share the files and access them with an HTPC? or do you want to stream them like VLC? [04:08] lox_, did you upgrade from hardy, or did you fresh install? [04:09] like vlc , i dont want to do it yet, just want to know huh [04:09] this one was fresh as the upgrade killed my raid5 [04:09] wizardslovak, then the software you'd need is vlc-cli [04:10] thx gman ;) [04:11] lox_, see method 3 here: https://help.ubuntu.com/community/Grub2#Reinstalling GRUB 2 [04:11] that's how you chroot into your install, and then grub-install will work [04:12] qman__, but /dev/sdb1 won't mount, it is a raid partition [04:12] lox_, you wouldn't mount sdb [04:12] ok people i am out good night ;) [04:12] md0 already mounted [04:12] under 4. mount your system partition, you mount where your / is [04:13] so /dev/md0, /dev/md1, whatever it is [04:13] same goes for boot partition, where /boot is located if it's separate [04:14] qman__, so at point 9 I do grub-install /dev/md0 ? [04:14] lox_, no [04:14] you do grub-install /dev/sda [04:15] then sdb, then sdc, and so on [04:15] for each disk you want to be bootable [04:15] done ! [04:15] You saved my day, I wish I had u on talk earlier ;) [04:16] now it should be able to boot your system from any of the disks [04:16] qman__, I wait the sync to finish and I will reboot [04:17] qman__, the the step I miss when chnaging a HDD in an array is to reinstall grub once booted back with working raid [04:18] qman__, I shoudl have issue sudo grub-install /dev/sdb after changing sdb and I would have not had that mess .... [04:19] yeah, hopefully grub2 is a bit smarter than grub1 and can deal with a change in boot order [04:19] qman__: IME, no [04:19] extlinux can [04:19] with grub1, you had to not only install to each disk, but also change the config based on which one you were installing [04:20] qman__, I followed that guide: http://www.howtoforge.com/replacing_hard_disks_in_a_raid1_array [04:20] nothing about grub in that damn guide [04:21] lox_, if you didn't replace the disk grub is installed to, it doesn't need any changing [04:21] but sometimes that happens [04:21] Grub *SHOULD* be installed to *ALL* disks in the array [04:21] However, in the example you pasted, he fails sdb so doesn't notice because sda is (presumably) the default boot disk [04:22] qman__, sdb is a new disk I replaced yesterday, and today I needed to replace sda, as I didn't reinstalled grub on sdb =>fail [04:22] lox_: so the lesson you have learned is: don't trust some random guide you found to be perfect [04:23] both were having a lot of SMART errors (intensive use during two years) [04:23] twb, and ... reinstall grub after I chnage any disk in a n array ;) [04:23] thks guys [04:23] yeah, I lost a RAID 5 recently because two of the disks catastrophically failed within an hour of each other [04:24] lox_: just so [04:24] I run a raid 6 with a spare now [04:25] raid is not a backup, never forget that [04:25] I am afraid to run raid 5 again, upgrade to lucid broke it last time [04:25] qman__, I learned that when my raid5 failed .... [04:25] upgrading to lucid broke my framebuffer, of all things [04:26] qman__: that's a "feature" [04:26] now it won't go above 640x480 [04:26] even after installing grub2 and correctly setting the parameters [04:26] qman__: in fbcon, or X, or both? [04:26] no X, just (I assume) fbcon [04:26] grub2 framebuffers are irrelevant AFAIK [04:27] What GPU? [04:27] ATi Rage XL [04:27] ran vbeinfo, shows support up to 1280x1024 [04:27] So I guess it'd be radeonfb.ko [04:28] before I installed grub2, it was just giving me out of range on the monitor [04:28] You put something like "radeonfb" in /etc/initramfs-tools/modules, run "update-initramfs", and reboot -- unless it's compiled into the kernel, in which case you edit the boot: line [04:28] At least, that's how it works for fbcon with i915/KMS on an Intel GPU [04:28] ok [04:29] Obviously if you only need 4:3 resolutions, you can just pass vga=771 or vga=779 or so [04:29] widescreen would be nice, but I was just using 1024x768 [04:29] but before I put grub2 in there, the vga line was causing my monitor to go out of range for some reason [04:29] gentoo's wiki's "framebuffer" article has a table of vga=N [04:30] when with hardy it worked fine [04:30] tried several different ones, too [04:30] Shrug [04:32] at first I thought maybe plymouth was attempting to go into 1440x900, the monitor's native resolution, and the card didn't support it [04:33] but that doesn't explain why putting in grub2 changed the result, or why removing the vga line makes it work at default 80x25 [04:33] qman__: are you using digital (DVI, displayport, etc) or analogue (d-sub)? [04:33] analog [04:33] It's entirely possible the default timings for high resolutions are beyond the capabilities of your current monitor === bastidrazor is now known as bastid_raZor === deegee is now known as 92AAAT9PA === mdz is now known as 92AAAT9W3 === foxbuntu is now known as 14WAAZ29V === jussi is now known as Guest88035 [05:32] guys, witch tool would u recommend me to clean up a HDD with some SMART errors ? HDD regenerator? [05:33] If a SMART self test fails, your drive is dying and should be replaced. [05:33] It cannot be "fixed" with software. [05:34] finally got it working, the issue is that it was attempting to use 32-bit color, while the card only supports 24-bit [05:34] I don't know WHY it was trying to use 32-bit color, since I told it otherwise, but eh [05:35] lox_: as twb mentioned your drive is dying. [05:35] qman__: there's no such thing as 32-bit [05:36] twb, ok thks [05:36] it is replaced [05:37] It's really three eight-bit channels (rgb), and an eight-bit alpha channel in some transforms [05:39] twb, yeah but they're two separate modes [05:39] and the card doesn't support 32-bit [05:43] I don't know why it didn't work in the first place though, for that it would have to be the radeonfb thing you mentioned [05:43] because everything else I changed was grub2 settings [05:44] vesafb can only use standard VESA 2.0 modes, which are all 4:3 [05:45] it is a standard VESA mode, 1024x768 [05:45] okey dokey [05:46] New bug: #610313 in bind9 (main) "package bind9 1:9.7.0.dfsg.P1-1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/610313 === EvilTrek is now known as EvilPhoenix === Norkakn_ is now known as Norkakn [07:00] Hi. I have a 8.04LTS server with 2.6.24-19-server. When any process is outputting, it will freeze after a bit, and then usually resume. So, if I tail -f a file, it'll be fine for a bit, and then freeze for ~5 minutes, then be fine again [07:00] writes to disk were really high earlier today, but now they are pretty sane [07:00] Sounds like you have a high I/O load [07:01] 1.3%wa in top, is that high enough to cause this? [07:01] You want to measure it with iostat (iowait?) [07:02] Blk_read/s averages around 250, Blk_wrtn/s 300 [07:02] The first stanza reports the I/O at boot time. [07:02] Run it with an interval and then look at the second stanza [07:03] http://paste.scsys.co.uk/47160 [07:04] that work? [07:07] it seems to be related to putting things out on the screen. If I don't redirect the output of a command, it takes 90 seconds. If I > /dev/null, it takes 1.2 seconds [07:14] I'm in over ssh though, and time is still reporting that it only takes 1.2s, even though it takes over a minute to print it all [07:14] web traffic is light though === rmk_ is now known as rmk [07:32] i use kvm on a ubuntu hardy server but i'm unable to give my guest system (ubuntu lucid) a fixed ip address. how to do this? [07:33] i use libvirt to manage the guest system [07:35] toabctl: you could virsh console into the box and alter /etc/network/interfaces [07:36] mase_wk, i'm already logged in in the guest system. [07:36] mase_wk, but what is the network-interface name? eth0? [07:36] ifconfig -a will give you a list === taneli_ is now known as xampart [07:36] physically there's only one network interface connected to the wan [07:37] yep you can set up a bridge in the host and then tell virsh to get your guest to use the bridge [07:37] mase_wk, is there a howto for that? [07:38] well i think the libvirt site has a pretty good howto [07:38] i would suggest having a google for it [07:38] thats how i found it [07:39] mase_wk, and the bridge should have the public ip, right? [07:39] yes [07:39] if thats how you want to set it up [07:40] well actually it really depends what your trying to do. If you have a network and each guest should have a public i[p [07:40] or some portion of guests need a public IP then you set up a bridge [07:40] if your hardware supports it you can do PCI passthrough for NICs directly to guests === jj-afk is now known as jjohansen [07:52] Mogge [07:56] New bug: #610329 in bind9 (main) "name server reported by dig are wrong" [Undecided,New] https://launchpad.net/bugs/610329 [08:02] ^^^ "I don't understand how DNS works, let's blame Bind" [08:02] Jeeves_: Error: "^^" is not a valid command. === Q_Continuum_ is now known as Q_Continuum [08:07] twb: do you have any other ideas? [08:08] Norkakn: nope [08:27] morning [08:28] 'evening === reisi_ is now known as reisi [09:25] * SpamapS should not be as wide awake as he is right now [09:49] ttx: does bug triaging always get this far behind when we sit in a room for a week instead of working? ;-) [10:12] how accurate is the dovecot documentation for 10.04 as there is no mention of it [10:12] in the wiki at all [10:17] hi all [10:36] !hi TuxSax [10:37] !hi | TuxSax [10:37] TuxSax: Hi! Welcome to #ubuntu-server! Feel free to ask questions and help people out. The channel guidelines are at https://wiki.ubuntu.com/IRC/Guidelines . Enjoy your stay! [11:48] I set up an Apache server on Ubuntu 10.04, and I set up two Things. In one I have the Servername as mydomain.com and the other as asdf.mydomain.com. When I type in the ipaddress of the server in the browser it brings me to asdf.mydomain.com. So for some reason it is using that as a default. Is there a way to disable this behavior? [11:48] I want it to not show a webpage at all if I have not set up a vhost for the name entered. [11:51] zrbecker: The first fuond vhost is the default [11:51] So create a vhost with what you want it to show, and name the file 0000_default [11:52] And it will be the one which shows up [11:52] when I enable default it doesn't display my domain virtual host, it always just uses default. [11:52] Is there a way I can fix that? [11:53] SpamapS: yep -- that's how it goes when almost everyone forgets their usual chores :) [12:13] HI there, somebody have experience with Dante? i would like to know if it record the sites visited per users? [12:18] Roxyhart0: never used it but google analytics is really good at keeping track of unique visitors to your site [12:18] thanks [12:20] welcome [12:25] I am looking for a proxy p2p that also can record the sites visited by my users in case i got problems with copyright stuff. Somebody knows any tool? [12:25] On ubuntu i have three virtual hosts set up with these configurations http://pastebin.com/biiJkHHt can anyone tell me why it will only serv the default one? All three of them are in the sites-enable. And if I turn off default the two other sites display fine, but I want default to be served when a random host is requested. [12:28] Roxyhart0: Kerio Controkl [12:28] Control [12:28] thanks a lot! [12:29] zrbecker: http://httpd.apache.org/docs/2.1/vhosts/examples.html#default [12:29] hello, i have had to compile php from source, for gd to work correctly, i did this, but then this other coder did some stuff and now he gets:Use of undefined constant MCRYPT_BLOWFISH - assumed 'MCRYPT_BLOWFISH' but mcrypt is installed and all seems fine.. [12:29] First hit on Google 'apache2 default vhost' [12:30] Good morning. I am running the lucid desktop 64 live CD. Is there a way to update all the packages except the ones related to the kernel? [12:34] Kio control apparently works just in windows server [12:34] Roxyhart0: No. It's a vm-appliance [12:34] So you can run a KVM-vm with Kerio Control in it. [12:35] Or, you can run it on a normal server, obviously. [12:35] But you can't do anything else anymore than [12:36] sorry you mean run in a virtual machine? [12:36] Yes [12:37] vm with windows? [12:39] Roxyhart0: Kerio Control is Linux [12:39] ok, thanks i will read more [12:43] Jeeves_: bleh, I changed *:80 to _default_:80 like that website seems to be suggesting, but same results... [12:44] zrbecker: What's the hostname of the machine? [12:44] asdf.mydomain.com by any chance? [12:44] zrbecker.com [12:45] hostname -f [12:46] hmm the hostname is set as MainServer. I thought this was changed. [12:47] I remember some weirdness if one of my vhost's ServerName was set to the host's $FQDN. [12:47] Thought that might be what you were seeing, but perhaps not. [12:47] I forget the details, just wanted to offer the hint. [12:48] heh thanks. [12:49] hmm weird, not sure since when but now it is serving the default for only one of the virtual hosts, but servering the correct files for the other one. Oh computers how I love thee. lol [12:52] zrbecker: are you talking about apache? [12:52] yes [12:55] soren: In my apache2.conf file my ServerName was set to zrbecker.com I changed this and it fixed it. So I guess you were right. How weird. [12:56] Jeeves_: Thanks for the help too. [12:56] np === oubiwann` is now known as oubiwann [13:09] Jeeves_, question for you [13:09] i am sharing a folder over my network [13:09] and it was working [13:09] then i went to get breakfast [13:10] now i come back, and when i try to access my folder from my windows laptops [13:10] I get [13:10] is not accessible [13:10] you might not have permission to use this network resource [13:10] the network path was not found [13:10] any ideas what may have caused that/how to fix it? [13:11] Jinxed-: what exactly did you have for breakfast? [13:11] pancakes [13:11] wheat [13:11] 2x [13:11] eggs [13:11] scrambled [13:11] 2x [13:11] rockstar energy cola [13:11] 2x [13:12] well not quite true... my pancakes were like an inch thick each and the size of the plate so i only made i through 1.5 of them, and i am just starting the energy cola [13:13] Jinxed-: That involces Windows, so I've got no clue [13:13] involves [13:14] hmm [13:14] you don't think it is anything on the ubuntu side? [13:14] I've got no clue. Really. [13:14] The folder Im sharing is on my ubuntu machine (I got the vlan trunking working btw from yesterday :) ) [13:15] I've got no clue. Really. [13:15] And also, sharing folders via de Gui isn't really #ubuntu-server'ish [13:16] Jeeves_, Do you know of any better way to share video or to do VoD through a more server oriented approach? [13:17] I feel like sharing a folder will only work for a small number of computers [13:17] Jinxed-: streaming would be a better choice [13:17] you can even do it using VLC [13:17] I would like to be able to do streaming [13:18] but I guess there are many other options to stream video on a network [13:18] but I believe VLC only lets you broadcast a stream like a shoutcast that people could view [13:18] I want people to be able to request a video that they want to see and stream it [13:18] Jinxed-: How many videos will you be sharing, and to how many people? [13:19] umm the numbers could very between 5-1000 [13:19] people [13:19] and 600 gigs-5TB of video [13:19] well actually starting off less video [13:19] maybe like 50 gigs [13:19] I want to build a system that is scalable [13:20] I know its easy to build something for a personal network, but im interested in learning how to set up something larger [13:21] Jinxed-: flumotion, or something like that [13:21] Is it flash? [13:21] no, it will be most likely some compressed version of video [13:21] h.264 [13:21] xvid [13:21] i want to be able to capture raw footage from a camera [13:21] compress it [13:22] and share it over the network [13:22] flumotion can do that, afaik [13:22] i have the stuff to capture/compress it is the video on demand [13:23] Hi everyone! [13:23] Can anyone tell me if the developers of "software-sources" have an channel-sources? [13:23] Cause I want to run the process for selecting the best Mirror server, every time before making an update to some program, for obtain more speed downloading [13:23] stupid windows... i reset them and it can work [13:23] Jeeves_, I will check out flumotion [13:25] Jeeves_, that looks SWEET [13:48] Hello, I have deployed two lucid servers (just OpenSSH server) as VirtualBox VMs. One is i386, one amd64 [13:48] networking is OK in the amd64, but not in i386: I have no eth0 ther [13:48] lshw -class network reports the network interface as DISABLED [13:48] ifconfig only returns lo [13:49] etc/network/interfaces file has eth0 configured for dhcp [13:50] Can someone tell me how I can get my eth0 enabled again? [13:50] ringods: can you see what mii-tool shows? [13:52] ringods: try ifconfig eth0 up [13:53] Both the working and the non-working system report: eth0: no autonegotiation, 1000baseT-FD flow-control, link ok [13:56] ringods: try renewing your dhcp settings using dhclient [13:59] sudo ifconfig eth0 up; sudo dhclient [13:59] output tells me an ip address is assigned [13:59] running ifconfig tells me eth0 is up, but only having an ipv6 address [14:01] ringods: if you set a static ip, can you ping out? [14:01] ringods: might be the logical link in vmware is down [14:06] ringods: can it be a problem in the vmware host settings? [14:08] RoyK: with static IP, it works flawlessly: /etc/init.d/networking restart brings eth0 up with the static ip assigned [14:09] wierd [14:09] changing it to dhcp again results in eth0 down and (off-course) no succesful dhcp request. [14:09] morning all [14:09] ringods: can you try to dump network traffic to see to where the dhcp request broadcast go? [14:10] In the meantime, I also double checked the vbox guest settings: both are configured exactly the same networking wise [14:10] and vbox works? [14:10] yes, I have 2 karmics and the lucid-amd64 running succesfully [14:10] * RoyK blames vmware [14:10] only the lucid-i386 fails to initialize the network [14:12] does lucid-i386 work on vbox? [14:16] RoyK: except for the eth0, it boots correctly [14:17] initial lucid support came in vbox 3.1.6. I'm running 3.2.6 [14:18] I'd guess there might be something related to vmware, then [14:18] vmware? [14:18] I'm talking VirtualBox here. [14:26] oh, so it works [14:26] oh - I misunderstood :) [14:26] in bridge or nat mode? [14:26] nat, with custom vbox dhcp server [14:27] the 3 other VMs also use that custom vbox dhcp server, an they initialize correctly [14:28] AFAIK very little has changed in the Ubuntu DHCP client - just to be sure, have you tried to sniff the network on the server while doing a DHCP request to see where it gets sent? [14:28] it should use a normal broadcast [14:28] but then, using vbox with a custom dhcp server complicates things a bit [14:28] problem is that /etc/init.d/networking restart seems to bring eth0 down resulting in dhcp failing. [14:29] with the static IP, the same command works perfectly [14:29] does it work if you set dhcp and then reboot? [14:29] no [14:29] if I login, and check ifconfig: only lo, no eth0 [14:30] lemme try to install lucid32 on vbox [14:31] Here is how I added an additional network on my virtualbox host: [14:32] VBoxManage dhcpserver add --netname "10.100.164.0/24" --ip 10.100.164.1 --netmask 255.255.0.0 --lowerip 10.100.164.5 --upperip 10.100.164.254 [14:32] then reconfigure nic1 of guests as follows: [14:33] VBoxManage modifyvm --natnet1 "10.100.164.0/24" [14:33] why don't you just bridge? [14:34] I am simulating production setup... [14:34] will you be using vbox in production? [14:35] no [14:35] I use vbox to try out server configurations [14:35] then use bridging with vbox on a separate network [14:35] NAT won't help you simulating [14:35] well, the separate network is the problem here at the office... [14:35] I am limited in what I can do. [14:36] well, it's probably vbox that's making a problem here [14:36] I could get an address during install of lucid without problems with nat [14:36] but why does it work for the 3 other VMs then with the same network setup? [14:36] no idea [14:36] ask on #vbox [14:40] ok, will try there. [14:43] RoyK: then again, even on real systems, my lucid system frequently come up with only ipv6 address assigned to eth0. networking restart solves the problem, but it is a pain to have to intervene every time [14:43] is this a known issue? [15:17] ok... in effort to get kernel upgrades working for uec, I'm going to have to depend on grub. [15:18] i can do that either run-time or build-time. i would prefer run-time as it seems like it would be easier to fix. with a package upgrade. [15:18] Daviey, kirkland ^^ question directed at you two. [15:19] its not a big deal to have node controllers depend on grub-pc in my opinion, as they already will have grub-pc (so they can boot). [15:19] smoser, Hmm.. runtime? [15:19] "Depends" versus "Build-Depends" [15:20] smoser, euca-nc can add a depends on grub2-pc if you need it.. I don't see a concern there [15:20] It's surely already available to you tho :/ [15:20] right. [15:20] the issue would be when / if someone needed a loader other than grub2 [15:20] but that is probably small case. [15:21] Looking to the future, it should probably be grub-pc|grub-efi [15:21] the second part of the question is to use grub-mkrescue, or role my own "make a bootable floppy" utility. [15:21] Hmm.. good point about lilo tho. [15:21] using grub-mkrescue would pull in a dependency on xorriso [15:21] smoser, How many lines would you anticpate your homebrew would be? [15:21] < 200 [15:22] That sounds reasonable, would it be generic or tied to euca? [15:22] somewhhat generic, but i'm not terribly concerned about that. [15:23] smoser: would that still make sense when grub isn't the user's bootloader? [15:23] i'd just make it euca-specific as the shortest path, and have it as part of eucalyptus-nc [15:23] yeah.. i was wondering if it was generic - if there was somewhere better to put it. [15:23] twb, the goal of this is to create a boot floppy that will eventually load the kernel/ramdisk from the guest (in a vm) [15:24] so, yes, even when the host OS was not using grub2, it would be needed to have this function. [15:24] Daviey, well, grub used to have 'grub-mkfloppy' which would have sufficed. [15:24] but it has been replaced by grub-mkrescue , which depends on xorriso [15:25] the ideal situation, was for me to use grub-rescue-pc, which is a canned ISO ready to be used. [15:25] however, its an ISO [15:26] smoser, Can't you make it boot from CD? ie ISO? [15:26] which means i can't modify it, and i'd need to modify it a bit. I was hoping to cp grub-rescue-floppy.img my.img && mount -o loop my.img && modify it && umount . [15:26] ahh... CD = SATA [15:26] i can maek a boot cd, yes. [15:26] and grub-mkrescue will easily do that for me [15:26] but it depends on xorriso [15:26] my only hesitation in that route is runtime node controller dependency on xorriso [15:27] smoser: OK, does it make sense when hardware less than ten years old tends to lack floppy support? ;-) [15:27] xorriso = universe... [15:27] can't see a MIR being successfuly for that. [15:27] xorriso is build depend of grub2, so it would seem that it has to be main [15:28] twb, until kvm -floppy goes away it makes sense. and even then, at this poitn the output of grub-mkrescue is an iso that boots as a floppy or a one of those funny shiney discs that are also obsolete [15:28] Hum. [15:28] I always just use -kernel -initrd -append [15:28] twb, right. that was what i was hoping originally. [15:29] unfortunately, if you use -kernel and feed it a grub multiboot image , the grub multiboot image will boot, but will only see virtio and ide disks. [15:29] scsi discs are not seen. [15:29] and we need to support scsi disks. [15:29] If you have -kernel, why use a bootloader at all? [15:29] -kernel would have been so much easier. [15:29] smoser, Ah, scrub that - it's main! [15:29] Lucid = universe, Maverick = Main [15:29] cool. Now Eucalyptus does not fail to set the public IP sometimes [15:29] twb, thats the essential reason for this: [15:30] (keep in mind that it would be nice if stuff from Lucid could be easily backported to Lucid) [15:30] it fails to bloody get *ANY* IP address [15:30] Daviey: rmadison -uubuntu ftw? [15:30] kvm -kernel ... guest update & upgrade ... reboot ... old kenrel. [15:30] twb, you don't need -uubuntu [15:30] Well, *I* do, because I'm on Debian :-) [15:31] smoser: yeah, in analogous infrastructure I have a silly dance to avoid that [15:31] yes, the goal is to avoid all non-necessary silly dances === 92AAAT9W3 is now known as mdz [15:31] ttx still has a nice silly dance that solves problems though. [15:32] I need it [15:32] I don't suppose syslinux's chainloading support is any easier than grub's? [15:32] cjwatson, ping. [15:35] twb, i might look. [15:39] twb, i hadn't really considered that. i'll take a quick look. [15:44] i guess the benefit of going with syslinux would be no dependency on a boot loader that might affect the host [16:22] do most of you guys kill apparmor [16:22] I don't touch it [16:23] * EvilPhoenix lets it sit there minding its own business [16:23] well it just screwed me for a few minutes [16:23] I have no compelling reason to extend it, and I have not run into any issues with it [16:23] you were messing with it werent you [16:23] because i symlinked /var/lib/mysql to somewhere else [16:24] ehehehe [16:24] probly not a good idea [16:24] just stopped apparmor for the time being [16:24] i may just uninstall it === dendrobates is now known as dendro-afk [16:24] i don't need a heartattack [16:24] it's mostly a damage control measure anyway [16:25] how do i remove a GPT partition table from /dev/md0 ? [16:26] * EvilPhoenix never messes with things he doesnt have a reason to mess with [16:26] in gdisk /dev/md0 i tried, 'd' but it doesn't have a partition there, just the table. [16:27] or in other words, how do i clear /dev/md0 of any partition information? can i delete the raid array and recreate it and that will remove any partition information correct? [16:27] none of my mds have partition tables [16:27] you could zero it [16:27] or just format with whatever new one you want [16:28] why exactly do you want to? [16:28] well, i'm trying to do pvcreate /dev/md0 and it returns that there is: /dev/md0: Skipping: Partition table signature found [16:28] ah, just zero the first...I forget how much [16:28] where a partition table would be [16:29] dd if=/dev/zero of=/dev/md0 bs=1M count=1 [16:29] qman__: okay, i odn't know how to do that. [16:29] should do it [16:29] that will destroy any data in the first 1MB of the disk [16:30] ther'e snothing on the disk [16:31] qman__: okay thank you, i've never done this before [16:32] no problem, just making sure I'm not instructing you to nuke your important files ;) [16:35] I just started using flumotion and when I do ps -ef | grep flu I see that there are alot of flumotion processes running after I close it down... how do I kill all of them at once? [16:36] Jinxed-, killall flu [16:36] where flu is the process name [16:37] qman__, what if I don't know it exactly [16:37] I think regex works [16:37] user 4901 4480 5 11:27 ? 00:00:15 /usr/bin/python /usr/bin/flumoti [16:37] example line [16:38] just run `ps a` [16:39] hmm [16:39] behavior is a little different than I remember [16:39] ah, ps -e [16:39] qman__: that sort of worked here's what i got, in gdisk: Caution: invalid main GPT header, but valid backup; regenerating main header [16:39] no, that one's cut off too [16:39] from backup! [16:39] AndyGraybeal, hah [16:40] qman__: but pvcreate /dev/md0 worked! [16:40] qman__: do you think i should proceed? [16:40] AndyGraybeal, yeah, should be fine [16:40] *with LVM i mean [16:40] if it lets you create the volume and format it you shouldn't have an issue [16:41] gdisk is reporting: arning! One or more CRCs don't match. You should repair the disk! [16:41] qman__: i want to get rid ofthis error before i proceed. [16:41] can i recreate my raid array and start all over again? [16:42] some how get rid of any of this stuf [16:42] AndyGraybeal, you could, but zeroing the whole thing should be faster [16:42] lets zero the whole thing then. [16:42] dd if=/dev/zero of=/dev/md0 bs=2M [16:42] it'll still take a while, but not as long as recreating and resyncing the raid [16:50] qman__: thank you [16:51] i'm gonna try it now, i've been runing downstairs and upstairs, i got a crew of people running wires. [16:53] In OpenSSH is there a way to disable password authentication for a single user? [16:56] Jinxed-, ps ax --format args | cut -d" " -f 1 | grep '/' | awk -F/ '{print $(NF-0)}' | grep flu [16:57] i am trying to remove the unused package linux-image-2.6.32-23-server using apt but it keeps hanging on "Running postrm hook script /usr/sbin/update-grub" [16:59] I am not running server (should clarify from the start) but I was told that there would be more expertise on this in here [16:59] i have tried running /usr/sbin/update-grub by its self and that never finishes either, any ideas? [16:59] qman__: out of curiousity, am i writing 2M of 0's at a time to the disk? [16:59] AndyGraybeal, yes [16:59] it's a bit faster than not specifying a block size in my experience [16:59] I have a 802.1q trunk between my laptop and a switch [16:59] qman__: awesome, i'm not questions, i'm just curious. [16:59] *questioning [16:59] hi [16:59] and I have it configured in ubuntu in the /etc/network/interfaces [17:00] im looking for something that's GUI to do backups using RAR .... [17:00] however whenever my computer restarts [17:00] are there any tools for linux that does this? [17:00] or it's most command line tools [17:00] none of the configuration is there, and I end up throwing random network commands at it until it comes back [17:00] how do I get it to start when the comp starts [17:00] ltyr, since ubuntu server does not have a GUI, this is the wrong place to ask [17:00] try #ubuntu [17:01] Jinxed-, probably belongs in /etc/network/interfaces, though I've never done that kind of setup [17:01] Here is my current /etc/network/interfaces setup http://paste.ubuntu.com/469864/ [17:03] Jinxed-, I'm not positive, but I think you still should have an `auto eth0` there, even though it's unconfigured [17:04] qman will give it a shot [17:04] reboot time === jjohansen1 is now known as jjohansen [17:08] qman__, no luck :( [17:09] Jinxed-, what commands do you use to get it working? [17:10] qman__, I haven't quite figured it out what it is [17:11] phyfus, I found a few people with the same problem, but no solutions, sorry [17:11] qman__, ok I did sudo ifup vlan20 [17:11] sudo ifup vlan30 [17:12] qman__, ok well i'm trying a reboot to see if that helps [17:14] Jinxed-, that's it? if so, try changing auto vlan10 vlan20 vlan30 into three separate auto lines [17:15] you mean like [17:15] auto vlan10 [17:15] auto vlan20 [17:15] etc [17:15] yes [17:16] I can't find concrete information on it but 20 and 30 starting as down seems to point there === dendro-afk is now known as dendrobates [17:18] interesting === Jinxed- is now known as b0ot [17:19] qman__, after reboot I was able to finish removing the package using apt-get autoremove, but I got an error The link /vmlinuz.old is a damaged link Removing symbolic link vmlinuz.old you may need to re-run your boot loader[grub], how do i re-run my boot loader? [17:20] phyfus, that's safe to ignore, and only shows up when you remove all but one kernel [17:20] phyfus, but to do it manually, `sudo update-grub` [17:20] ok thanks [17:20] ubuntu support is ace! === Guest88035 is now known as jussi === jussi is now known as Guest14592 === Guest14592 is now known as jussi01 === jussi01 is now known as jussi [17:24] hello [17:24] I would like to learn html 5 and I am asking if anyone here knows about some learning materials around the web. [17:27] harrison: i would ask in #html or #web, i think w3c has their own irc server also with resources. [17:27] harrison, i feel like google might actually be a good place (I know that is usually not advicated to google it) but html 5 is very popular right now and I imagine html5 introduction or html5 tutorial would bring up some relevant starting poitns [17:28] ok qman__ no luck on the vlans starting on restart [17:37] thanks [17:38] I didn't know about the #web and #html chat rooms, whats the url of the W3C server? [17:38] *W3C IRC server [17:40] I thought the mpt2sas module made it into 10.04... [17:40] Hmmm. [17:41] Never mind, I have to rebuild my PXE initrd. [17:55] kirkland: hallyn: hi - is there a specific DebuggingKvm wiki page that outlines how importance should be set on qemu-kvm bugs? [17:55] mathiaz: not sure, but the guide is easy .... [17:55] mathiaz: critical == eats your data [17:55] kirkland: which guide are you referring to? [17:55] mathiaz: high == crashes the host [17:56] mathiaz: medium == crashes the guest (and most other "normal" bugs) [17:56] mathiaz: low == doesn't crash host or guest, but has undesirable, clearly buggy behavior, or if it does cause crashes, can be worked around [17:56] mathiaz: wishlist == feature requests, everything else lower priority [17:57] kirkland: people.canonical.com/~sergeh/binary.iso.lzma has the uec livecd fwiw [17:57] hallyn: wgetting [17:57] hallyn: HTTP request sent, awaiting response... 404 Not Found [17:58] gah [17:58] serge, not sergeh [17:58] sergeh was at ibm :) [17:59] people.canonical.com/~serge/binary.iso.lzma [18:00] mathiaz: kirkland: i tend to say if it can be worked around i call it medium, if it cannot for some ppl then i call it high... [18:00] lose data (in either guest or host) i would call critical. looks like i'm not in tune with kirkland [18:00] hallyn: pulling [18:01] oh, i guess we are - i was misreading [18:01] hallyn: i agree with ^ [18:01] of course, 'lose data' can be easy if you use a stupid fs so then it's not kvm's fault :) [18:03] hallyn: kirkland: I've updated https://wiki.ubuntu.com/DebuggingKVM [18:03] mathiaz: cool, thanks [18:05] SpamapS: i should have something up for you today [18:06] hello! i need help with UUIDs. partition sda5 suddenly has no UUID, how can I create and attach new one? tune2fs -U random /dev/sda5 didn't help [18:14] mathiaz: thx [18:19] qman__: that fixed the issue! thank you brother. [18:27] Can I use SCP to move a file on a remove server from one directory to another? [18:30] FunnyLookinHat: why not just us cp? [18:31] FunnyLookinHat: SCP is more from remote machine to local machine transfers [18:31] AndyGraybeal: I'm writing a cron'd PHP script to transfer files to another server - once the file is there I have to move it into the correct directory (to prevent another script from prematurely reading an incomplete file) [18:31] AndyGraybeal: I'm open to any suggestions... :) [18:32] yea, i'm not gonna say anything becaues i don't know any better, i'm fairly new to all of this. [18:32] even though it's been 10 years i've been meddling with it [18:32] heh- fair enough :) [18:33] someoen else will probably help, i'm no good at this stuff [18:34] FunnyLookinHat, you can use 'scp' to *copy* a file from one directory to another [18:34] smoser: but not to move - right? [18:34] scp: secure CP [18:34] cp: copy [18:34] lol - figures. [18:34] so, no, not really. [18:34] hi! I have a php that writes user input to a text file, problem is strange chars like euro sign are not written correctly, should I just change the locale? [18:34] Can you think of a good way for me to accomplish this, thought? [18:35] eduardo_f: I would bet so - but try ##php [18:35] thanks FunnyLookinHat [18:36] np [18:37] well, if you're needing to transfer a file across a network, you can use rsync. [18:37] rsync will transfer file and then do a rename when it is done. [18:38] for that explicit reason, FunnyLookinHat [18:40] smoser: whoahhhh - wait. It will do the rename? [18:41] yes. [18:41] smoser: i.e. if I do rsync -t /path/to/localfile /remote/server/file - the file won't show up in the filesystem until it's completely uploaded? [18:42] well, it will show up in /remote/server/.file.zcBoJz [18:42] or some name like that [18:43] and then when complete it shows up as /remote/server/file [18:43] No kidding... [18:43] wow. [18:44] Do we have an EC2 image that's closer to 10.04.1 than to 10.04 ? [18:45] Thanks smoser - I'll have to check that out [18:45] kim0, you can use a daily [18:46] they're there for testing purposes. [18:46] ah I'll probably do that [18:46] kim0, http://paste.ubuntu.com/469905/ [18:47] smoser: any preference for 64bit or 32bit over ec2 for small instance [18:48] well, small instance only works for 32 bit [18:48] :) [18:48] so , i suggest 32 bit [18:48] hehe [18:48] anybody here use flumotion? [18:48] be aware, that those AMIs "fall off" [18:48] (https://wiki.ubuntu.com/UEC/Images/NamingConvention) [18:49] we keep the latest 5 daily builds. if you rebundle, you're OK, but if not, you may have to change a reference to an AMI in your launching code. [18:50] got you [18:51] smoser: any reason why there's no images for 24th and 25th [18:51] lucid build twice per week. [18:52] development is "daily" [18:52] (development == maverick) [18:52] ok got you [18:57] Hi, need help with some noob questions. am i on the right place? (ubuntu 10.04 server amd64) [19:00] viusert: its best to just ask the question, sometimes there are people who have an answer, sometimes there are not. ;) [19:00] b0ot: whats fluomotion ? [19:03] i get a message in console "init: ureadahead-other main process (###) terminated with status 4" and blinking cursor under it. i think it is not ureadahead problem, how can i figure whats wrong? in /var/log many different logs, should i look for something specific? [19:03] SpamapS, flumotion is a video streaming server application [19:05] viusert: bug #522197 seems related to that [19:05] Launchpad bug 522197 in upstart "init: shouldn't log a warning/error when a job fails with a status in "normal exit"" [Low,Triaged] https://launchpad.net/bugs/522197 [19:06] b0ot: cool. [19:10] sommer, so, https://help.ubuntu.com/community/CloudInit has more info than the cloud-config syntax post. [19:10] cloudinit takes multiple types of input, one of which is cloud-config. [19:10] tell me how to review section, where do i see it ? [19:11] SpamapS: ping [19:11] New bug: #610544 in openldap (main) "slapd apport hook" [Wishlist,New] https://launchpad.net/bugs/610544 [19:11] SpamapS: #ubuntu-meeting, if you're around ... [19:11] smoser: it'll be here sometime tomorrow: http://doc.ubuntu.com/ubuntu/serverguide/C/uec.html [19:11] SpamapS: you should join #ubuntu-meeting [19:11] crap forgot [19:12] k [19:16] Anyone know how to find a stream if you are doing it over localhost? [19:17] FOUND IT === bastidrazor is now known as bastid_raZor [20:00] ivoks: you still ahve a papercut assigned to you, bug 321091 [20:00] Launchpad bug 321091 in bacula "Bacula fails to install correctly if mysql wasn't installed before" [Medium,Triaged] https://launchpad.net/bugs/321091 [20:00] ttx: yes, i know [20:01] ttx: i got my laptop back yesterday so i'm back in business since today [20:01] ivoks: do you think you can work on it by alpha3 ? That's end of this week [20:01] * ttx slaps the laptop thief [20:01] they had to replace my MB [20:01] never buy lenovo [20:01] that's just junk [20:01] 2 times in one year [20:02] that's the croatian lenovo. [20:02] it's chinese :) [20:02] ivoks: kirkland/smoser/mathiaz are pretty happy with theirs [20:03] ivoks: hey! [20:03] do we install recommends by default in maverick? [20:03] mathiaz: hey [20:03] ivoks: I was considering buying one, if only they would ship an x201 in france [20:03] ivoks: since Jaunty, yes [20:03] ttx: well, this is x200s and it's excelent, on paper [20:04] ivoks: maybe stop scuba diving with it ? [20:04] and it is awesome, when it works [20:04] :) [20:04] mathiaz: wondering if you have time to review the latest openldap-dit branch? [20:04] sommer: on my todo list [20:04] sommer: I've noticed your update [20:04] mathiaz: coolio, if there are more changes needed I should have time to get them in quickly [20:04] sommer: hi there [20:04] sommer: ok [20:04] ivoks: yo :-) [20:05] thanks mathiaz [20:06] nxvl: you have papercut bug 16953 on your plate -- if you think that's more than two hours work, then we should unnominate that one. If you can't make it happen by alpha3, maybe unassign yourself from it [20:06] Launchpad bug 16953 in aptitude "Aptitude: should accept both "Si" and "Sí" (when asking for confirmation)" [Low,Triaged] https://launchpad.net/bugs/16953 [20:07] interesting bug :) [20:07] ivoks: that's an itch I don't need to scratch, I don't have that letter on my keyboard. [20:08] how about these: lj nj dž [20:09] kirkland: could you do me a favor? Could you please take care of bug #608338 ? [20:09] Launchpad bug 608338 in libesmtp "Sync libesmtp 1.0.4-5 (universe) from Debian unstable (main)" [Wishlist,Confirmed] https://launchpad.net/bugs/608338 [20:11] ttx: it is [20:11] ttx: it needs a function that changes i for í and that means using ascii and bla bla bla [20:11] nxvl: ok, I'll wontfix it as a papercut then [20:11] ttx: the problem is that there is no easy way to do that because it uses the translated string from the pot [20:12] RoAkSoAx: sweet, I was going to file a sync bug for the same reason! :) [20:12] nxvl: yep [20:12] so thre is no way to say "this or that" [20:12] the only way is to strip the `'s out of it [20:14] New bug: #610557 in openvpn (main) "NFS Client Not Properly Unmounted on Halt/Reboot" [Undecided,New] https://launchpad.net/bugs/610557 [20:14] RoAkSoAx: done [20:21] kirkland: thanks :) [20:44] SpamapS: hehe I'm gonna update a MIR for libesmtp now that the CVE has been fixed [20:44] RoAkSoAx: its needed for collectd which I'm working on a MIR for as well [20:48] SpamapS: awesome, I need it for pacemaker :) [20:52] Is there a guide to setting up a mail server with the ability to use LDAP instead of unix accounts? === dendrobates is now known as dendro-afk [21:07] SpamapS: the only way i can boot system for troubleshooting is install CD with rescue mode. so uoy say the upstart is my problem? where it save logs? is there a way to look where it stops? [21:08] SpamapS: (sorry for bad english) [21:11] i have /dev/md0: LABEL="data" UUID="bef0f8b9-fc8a-4e81-96fa-8d868ef58e35" TYPE="ext4" -- how do i get this into fstab? === dendro-afk is now known as dendrobates [21:17] <_Techie_> i installed ubuntu server 10.04 last night and it cause my boot to hand at 'verifying DMI pool data' is tehre a workaround for this? [21:17] _Techie_: 1st boot after install ? [21:17] <_Techie_> viusert: yes [21:17] <_Techie_> i have checked and re checked all the hardware [21:17] <_Techie_> and it boots fine with that drive if i install win7 on it [21:17] _Techie_: maybe you should try no acpi (i'm a noob too) [21:17] <_Techie_> will have to try that tonight [21:18] <_Techie_> with luck i will have my parts arrive today, so i will have a compter assembled by tonight that i can use to create live USB's [21:32] I have one problem: I particioned the disks, when I will install the base system, i had this message: "The installer cannot figure how to install the base system No installable CD-ROm was found." I have one server HP Proliant ML110 G6. === Ng_ is now known as Ng [22:14] mathiaz: I took the 5 steps you put on the board and have been using them. Quite effective so far. [22:14] mathiaz: err, 5 GTD steps I mean [22:18] zul: SpamapS: i think you're probably the ones to ask: if i want an iscsi target (in a vm, for testing), what's the ubuntu-ish thing to install? [22:18] tgtd isnt it? [22:19] zul: freakin' cool, thanks! === unreal_ is now known as unreal [22:50] Is there a guide to setting up a mail server with the ability to use LDAP instead of unix accounts? [22:54] smoser: ping [22:54] smoser: re: https://bugs.edge.launchpad.net/ubuntu/+source/byobu/+bug/610134 [22:54] Launchpad bug 610134 in byobu "'byobu -S _name_' is broken" [Low,Confirmed] [23:07] SpamapS: glad to hear that! [23:08] SpamapS: if you wanna take it to the next level, I'd suggest putting David Allen 'Making it all work' book on your reading list [23:08] mathiaz: hi, Have you read 'switch' ? [23:08] mathiaz: if not, you should :) [23:09] lifeless: switch? hm - I haven't heard of this book [23:09] another heath book [23:09] lifeless: oh [23:09] I linked it in my blog a few days back [23:09] lifeless: the follow-up/next book from Made to Stickl [23:09] *stick* [23:09] yes [23:10] its very good [23:31] New bug: #610630 in php5 (main) "Please add Kolab php patches to Ubuntu php5" [Undecided,New] https://launchpad.net/bugs/610630 [23:41] SpamapS:/win 2 [23:45] SpamapS: egads, the title of the kslm (klsm?) talk has changed, doesn't mention that acronym at all any more... [23:47] * hallyn wonders whether that means he's given up on whatever patchet he was brewing === dendrobates is now known as dendro-afk [23:49] ups [23:56] hallyn: "Service Measurement in the Kernel" .. I think he's just not ready to present working KSLM [23:57] SpamapS: hm, ok, i thought last week he had KSLM in the title or abstract