[00:26] New bug: #759284 in vsftpd (main) "package vsftpd 2.2.2-3ubuntu6.1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/759284 === dendro-afk is now known as dendrobates [00:56] question: when manually run update and upgrade with apt-get, there are 3 packages kept back. If I run with aptitude, they are installed. Anyone know which is correct or does it matter? [00:56] the 3 packages are: linux-headers-server, linux-image-server, and linux - server [00:57] ahhh.. nvm. found the answer. [00:57] how odd, when i perform the initial three default schema installation for ldap as per ( https://help.ubuntu.com/10.04/serverguide/C/openldap-server.html ) via Fabric, it errors. But performing the exact same commands mannually succeds [01:12] dravekx: apt-get update && apt dist-upgrade [01:23] RoyK, thx :) [01:30] hi guys [01:52] i'm having some problems with sudo, why there is a significant pause between user input of sudo password and the expected terminal response when using sudo after the cache timeout ? [01:57] hey guys- i just fired up a new maverick server, and it doesn't appear to have add-apt-repository... is this normal? [01:59] it is on all my lucid servers [02:00] right, me too :) [02:08] i'm setting up two dhcp servers and i want one to be a backup… any good docs on how to do this? [02:09] just copy the config file out [02:09] or setup a proper backup system [02:09] up to you [02:09] yeah but i only want one to be a primary at any given time. [02:09] in other words, the backup should be dormant until the primary fails [02:09] unless this is the wrong approach. i'm by no means an expert with this sort of thing. [02:10] just google dhcp clustering [02:10] it's trivial [02:11] ubuntu clustering is trivial and works very well [02:11] that is, at least the dhcp part [02:12] ya, clustering dhcp is alittle bit to wrap your head around at first [02:12] but is pretty simple [02:13] actually i think i found what i needed here: http://www.madboa.com/geek/dhcp-failover/ [02:13] In case anyone else is interested. [02:13] can only do two servers though [02:13] It's pretty straightforward. [02:13] Two is all I need. [02:13] …for now [02:13] we're running that at the office [02:13] my only issue is, I dunno why, but when I set it up, all dhcp requests get delayed a good 10sec or so [02:13] works like a dream [02:13] one fails, the other takes over [02:14] and I do lots of netboots, so it annoys me [02:14] RoyK: what happens once your primary comes back up? [02:14] it resyncs [02:14] patdk-lap: then you must have done something funny [02:14] onecrazycat: they sync up [02:14] royk, ya, i figured, but can't figure out what :) [02:14] RoyK: Nice. So no manual intervention req'd then? [02:14] no [02:14] RoyK: You're a gentlemen and a scholar ;) [02:15] you can also setup dhcp ranges that are private per each instance [02:15] thank you, boy :) [02:15] and stuff served there don't failover [02:15] required for bootp [02:15] I wonder if that is what is screwing me, still have bootp stuff [02:17] it's not everyday I get called a gentleman, I must confess I feel honored [02:18] RoyK: Sap it up. :) [02:19] * RoyK hands onecrazycat a dram [02:19] RoyK: It's not every day I get a solid dram from someone in ubuntu-server IRC. [02:19] :) [02:20] that'll be an islay [02:21] does maverick server normally run apache2 + sendmail out of the box by default., or is that just my vpses config? [02:21] I'm pretty sure your VPS needs to enable that during install. [02:22] Ubuntu server (any version) doesn't run anything by default, that's the beauty of it. [02:22] jblz: iirc that's apache + exim or postfix [02:23] yeah, it's totally weird, it's got apache2 and "sendmail-mta", with one of the apache2 processes running as root [02:23] It could be for their administrative purposes… but this would be a question you could ask them to confirm. [02:23] well, i figured someone on here would know whether that was standard maverick behavior or not [02:23] not standard [02:24] good [02:24] thank you onecrazycat. [02:25] jblz: np [02:26] nite [02:30] they probably just set the lamp install by default [03:39] New bug: #759367 in php5 (main) "package php5-fpm 5.3.5-1ubuntu6 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/759367 === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [04:51] I have a server with dhcpd installed, but it isn't running… I inherited this box, and I'm curious, how can I tell if it's set to start at boot? I looked at the rc.* files in /etc/, and it's there as though it is set to run at boot… is there anywhere else this sort of thing can be set? [04:55] onecrazycat: what release? [04:55] onecrazycat: lucid? [04:56] As at lucid, dhcp3-server is still a sysvinit package, so it will be in /etc/init.d/dhcp3-server with SNN symlinks in /etc/rcS.d or rc[2-5].d [04:58] twb: hardy [04:58] Same applies then [04:59] twb: okay, this is where i'm a little confused. [04:59] twb: i ran a quick ls /etc/rc* | grep dhcp [05:00] and i get a lot of files in each of the rc[2-5].d directories === dendrobates is now known as dendro-afk [05:02] onecrazycat: /etc/network/interface has to say inet dhcp [05:02] extraclassic: let me check that === dendro-afk is now known as dendrobates [05:04] extraclassic: are you sure? My current dhcp server doesn't have this [05:04] extraclassic: that is, the server I know is running dhcp for sure doesn't have this. [05:05] i'm not sure what the file is called then...in debian it's /etc/network/interface so I guessed [05:06] extraclassic: ah okay, let me grep through that directory (/etc/network/) to see if i can find any references to dhcp on a known working server [05:08] try /etc/network/interfaces [05:08] yea nothing there... [05:10] onecrazycat: try this then - http://www.ubuntugeek.com/how-to-install-and-configure-dhcp-server-in-ubuntu-server.html [05:12] nevermind...don't think that's what you were asking [05:39] I have a Soekris net5501 board, which I believe has a Geode CS5536 southbridge. [05:39] I am trying to work out which/how to load the watchdog driver. [05:39] AFAICT it's the "geodewdt" driver, which hit mainline in 2008. But this module isn't present in my lucid install (2010, .32) -- so WTF? [06:02] hi guys... where can I find the list of recently installed updates on my ubuntu 10.04 Lucid?... I updated laptop and the touchpad stopped working === dendrobates is now known as dendro-afk === twister004_ is now known as twister004 [08:50] My logs show that ntpd regularly unbinds from 0.0.0.0:123 and binds to :123. [08:51] This is stupid. How do I tell ntpd to shut up, bind to 0.0.0.0:123 when it starts, and just stick to that? [08:51] twb: unfortunately you can't. [08:52] It's *especially* stupid because I have e.g. four IP addresses on 1.2.3/26, and it claims to bind onto 1.2.3 thrice [08:52] Bloody ISC needs to get a clue [08:53] I just finished replacing bind9 with ntpd and unbound, maybe I should throw away their ntpd too [08:54] twb: I hope you meant nsd instead of ntpd ;) [08:55] twb: openntpd is not bad, if you want to try out an alternative to ntpd [08:59] Er, yeah [09:00] Thanks for the +1 for openntpd [09:57] hi, i use a ubuntu server10.04 64bits as a router... So i activate net.ipv4.ip_forward=1 in /etc/sysctl.conf but it's really slow (10KB) and i have 50MB in the lan... Where should i look to solve this issue? [10:05] xro2: are both NICs gigE? [10:14] twb, i have 1Go and 10Go on the other side [10:14] NFI then [10:14] twb, NFI? [10:15] No Fucking Idea [10:15] twb, ok... [10:31] nobody else have an idea why my network is slow? [11:02] <\sh> Ng: ping terminator and grouping all terms via super+g (it doesn't work when you under unity ;)) any solution for this? [11:03] \sh: I've only been back from paternity leave for 2 hours, so I haven't had a chance to do any testing with natty yet. Perhaps mapping it to another shortcut would be a good short term solution until I figure out what's changed about Super in Unity :) [11:05] <\sh> Ng: good idea...and about the super key, it's being catched by the launcher first imho...eventually didrocks has a solution to this problem :) [11:14] hi, i use a ubuntu server10.04 64bits as a router... So i activate net.ipv4.ip_forward=1 in /etc/sysctl.conf but it's really slow (10KB) and i have 50MB in the lan... Where should i look to solve this issue? I forgot to tell that is a virtual machine on a ESX server [11:23] Hi guy. Does anyone know if there is a "pastebin" server available in the packages [11:35] when I have a "register=>xxxx:yyy@zzz" in sip.conf, should this peer be displayed in "sip show peers"? [11:45] Hi, I want to start jscoverage when my server starts (I have a script inside /usr/local/bin which is executable and just calls jscoverage-server with the appropiate arguments). So I have in my /etc/rc.local the line "/usr/local/bin/startJsCoverage &" but it doesn't seem to work, how am I supposed to be doing this? [12:51] Hi, i try to set ip_forwarding on a ubuntu server 10.04 64bits (on vmware)... I setted cat /etc/sysctl.conf net.ipv4.ip_forward=1... But the routing process is very slow... I get 50MBytes on each lan and only 20KBytes when i route the traffic from one lan to the other one... Have you an idea? [12:53] ls [13:03] someone could confirm me that there is nothing more than set ipv4.ip_forward=1 to activate routing... [13:06] xxron: You can try it at #ubuntu (but this channel would probably benefit you the more). [13:06] xxron: Sorry -- Slow traffich here. :( [13:06] traffic* [13:09] xxron: yes that's what you need to do and either restart or reload sysctl with the change === dendro-afk is now known as dendrobates [13:10] xxron, check your duplex settings [13:11] zoopster, patdk-wk, it's full duplex, mtu are the same, no iptables... [13:11] are you sure it's fullduplex on your nic AND the switch? [13:11] xxron: throw a packet sniffer on it and you'll quickly find the problem [13:12] Duplex: Full on eth1 and eth3 [13:13] patdk-wk, the problem is that it works on LAN... i get problem only when i try to route the traffic...... [13:13] 50MBytes doesn't sound like working [13:13] you should be able to get atleast 90MByte [13:14] what kind of system is it? [13:14] Mware VMXNET3 Ethernet Controller [13:15] that isn't a system [13:15] that is a nic driver [13:16] so, what do you mean by system? [13:16] motherboard, cpu, nic's, ... [13:17] patdk-wk, so it's a virtual machine on a ESX4.1 server, with 2 intel CPU, and VMXNET3 [13:18] that still doesn't tell me anything at all [13:19] patdk-wk, a virtual machine don't really have a motherboard and a nic... [13:19] virtual machines don't run on thin air [13:19] and yes, virtual machines do have motherboards and nics, in your case vmxnet3 nic's :) [13:20] and probably a vm version7 motherboard [13:20] but I'm talking about the real hardware it's on [13:21] patdk-dk, so you want the ESX hardware... === chuck_ is now known as zul [13:54] Daviey: we should be iso testing today shouldnt we? [13:55] zul, I'm not convinced that this is our real candidate... but cautionary testing is still worthwhile. [13:56] is ubuntu going to support PERC S300 anytime soon? [13:56] zul, At least getting most of the test cases covered would be a good sniff... but i don't think we need to break our backs on it, if there are other urgent things on. [14:16] Anyone got an idea why /sbin/blkid doesn't give any result when executed as non-root on 8.04? It works on 10.04 as non-root.. [14:16] And no, i'm not planning to upgrade the entire environment to 10.04. [14:19] laen, off the top of my head, no, but strace will show what's going on [14:21] laen, are you sure you're not root? because it doesn't work for me here on 10.04 as a non-root user [14:22] ppetraki: that's because your /etc/blkid.tab has no read permission for other. [14:23] ppetraki: Many distro's just give that permission. [14:24] * ppetraki needs a second cup of coffee [14:25] ppetraki: open("/dev/sda1", O_RDONLY) = -1 EACCES (Permission denied) ..auch :P [14:26] laen, yeah, I get that too, even when it works [14:27] laen, http://pastebin.com/b5sWS8Pm [14:30] ppetraki: your blkid does seem to work as normal user though [14:30] laen, sure it does [14:30] <09:30:42>ubuntu-lucid$ blkid /dev/sda1 [14:30] /dev/sda1: UUID="ca077257-f9ec-4b63-8b0c-b8376e6f068a" TYPE="ext4" [14:30] yeah [14:31] the stat calls go through, and there's a read on the blk device [14:31] I've never looked at the blkid source code but I suspect it's examining the partition table as part of the uuid generation [14:32] on you 8.04 system, is that block device delegated to something like mdadm? [14:32] Nope, direct disk, no software raid or lvm or something. [14:34] Wondering how or why the user doesn't have O_RDONLY access on the disk, in 8.04, and does on 10.04. [14:34] I'm not having a very bright moment here. [14:35] don't worry about it ;) [14:35] permission check on 8.04? [14:35] * ppetraki has blkid source code now [14:36] <09:35:50>src$ ls -l /dev/sda1 [14:36] brw-rw---- 1 root disk 8, 1 2011-03-28 17:54 /dev/sda1 [14:36] It's the old blkid 1.0.0 [14:36] yeah, same [14:36] that's funny, even my user isn't in the disk group [14:38] ppetraki: yep [14:40] can any body help me in doing mac based auth using freeradius? [14:40] i have done settings as per http://wiki.freeradius.org/Mac-Auth [14:40] but while connecting the server is ignoring my request [14:40] plz ? [14:46] is any body out there? [14:48] plz anybody anser [14:48] ppetraki: "Before kernel 2.6.24, and since kernel 2.6.24 if file capabilities are not enabled" ..guess i'm out of luck, it's using CAPABILITIES in 10.04, which isn't used in 8.04 as that is the .24 kernel. [14:48] laen, where'd you find that factoid? [14:50] ppetraki: I went from the manpages of OPEN(2), FCNTL(2), PATH_RESOLUTION(7) to CAPABILITIES(7), where that sentence was written. [14:50] ppetraki: Not that i found those manpages on Ubuntu though ;) [14:51] laen, well, they're part of the programmers manual. I have them, but I'm a kernel dev [14:51] So, does that sound logical to you then? [14:52] laen, still reading, but yeah [14:53] So, I'm wanting to look at a bit of system monitoring and diagnoses. The packages 'atsar' and 'systat' are both available to me. Does anyone have any opinion/ experience regarding either of them? [14:56] ppetraki: thinking about a workaround, how unsafe would it be in this case, to give blkid suid ..the customer application prefers not to use sudo [14:57] laen, well, adding suid to anything puts the system at more risk, instead, you may be able to install the lucid kernel on 8.04 (hardy?) [14:58] laen, actually populating the disk group might be enough [14:58] ppetraki: oh duh, indeed. [14:59] laen, then again, what does a regular user need blkid to begin with? [14:59] ppetraki: none, customer application [14:59] ppetraki: well, open() call still fails .. weird. [15:00] Oh wait [15:00] wrong machine ;). Done, works! [15:01] \o/ [15:01] Interesting problem :) [15:01] yeah, and I [15:01] I mean, the solution was a bit too easy, but.. cool :) [15:01] 'm still half asleep :) [15:02] Ow :) [15:18] ppetraki: got any good documentation to read into when i wanna try to understand the way CAP_* are used on distributions? [15:20] laen, besides the manpages? not really. Then again, I mostly tolerate userspace :-p, so I'm not the best person to ask === dendrobates is now known as dendro-afk [15:21] laen, that would be a great idea though [15:22] anyone know of a tool like grubby for grub2? afaik grubby is from grub1 days [15:22] or am I left with ... sed -i /etc/default/grub ;-) [15:23] ppetraki: :) [15:33] hallyn: SpamapS has been looking into some courier bugs. He might be the best one to 'care'. === dendro-afk is now known as dendrobates === zz_carcinogen75 is now known as carcinogen75 [16:04] Daviey, kirkland , zul [16:04] could someone review this, just to make certain its sane: [16:04] http://paste.ubuntu.com/593600/ [16:04] looking [16:04] i'm fairly sure it is, i tested by: [16:04] * run new instance [16:04] * kirkland looks [16:05] * sudo apt-get --purge remove grub-legacy-ec2 grub-pc grub-common [16:05] * sudo rm -Rf /boot/grub [16:05] looks ok [16:05] smoser: yeah, looks fine [16:05] * sudo dpkg -i grub-legacy-ec2_0.6.1-0ubuntu6_all.deb [16:05] smoser: i just noticed that: [16:05] this fails (thats the old version) [16:05] $ dpkg -S /boot/grub [16:05] dpkg-query: no path found matching pattern /boot/grub. [16:05] * rm -Rf /boot/grub [16:05] * sudo dpkg -i grub-legacy-ec2_0.6.1-0ubuntu7_all.deb [16:05] smoser: so that dir must be manually created by the installer, or something [16:06] kirkland, did you run it past foundations? [16:06] Daviey: nope, but i suggest to smoser that he might ask cjwatson about it, just for a history lesson, if nothing else [16:06] New bug: #759885 in cloud-init (main) "grub-legacy-ec2 does not create /boot/grub on install" [Critical,In progress] https://launchpad.net/bugs/759885 [16:07] cjwatson, ^^ would appreciate feedback [16:07] kirkland, good thinking. [16:08] cjwatson, the pastebin at http://paste.ubuntu.com/593600/ [16:11] so, should i wait on cjwatson ? [16:12] we're kind of gated on uec images build to get ec2 images out there. and that is gated on this bug. [16:12] bug 759885 [16:12] Launchpad bug 759885 in cloud-init "grub-legacy-ec2 does not create /boot/grub on install" [Critical,In progress] https://launchpad.net/bugs/759885 [16:15] Daviey, kirkland thoughts ^ [16:16] smoser: i don't think you need to wait, no [16:16] smoser: it's blocking our work, and me/zul/daviey all gave you a thumbs-up [16:16] smoser: so it's on us [16:16] smoser: it would be good to get cjwatson's opinion, if and when he can give it, but let's rock and roll with what you have [16:17] New bug: #756489 in samba (main) "smbd crashed with SIGABRT in __kernel_vsyscall()" [Low,Incomplete] https://launchpad.net/bugs/756489 [16:17] New bug: #759099 in samba (main) "smbd crashed with SIGABRT in close_file()" [Low,Incomplete] https://launchpad.net/bugs/759099 [16:17] meeting [16:17] I'll get back to you in a bit [16:17] ok. i just uploaded that. [16:18] so i assume i need a release admin "OK" anyway [16:30] i've installed festival on my 10.04 server machine, ran the update.rc and made/edited /etc/defaults/festival, yet when i try to run the festival init script it seems to do nothing, i see nothing in logs. [16:31] any ideas? [16:36] anyone know how i woudl see if a package is in a seed ? [16:36] uec seed was modified yesterday (2011-04-12 09:12:31 -0400) [16:36] to add lxcguest [16:37] but 'apt-get remove lxcguest && apt-get install ^uec --dry-run' does not indicate that lxcguest would be installed [16:38] i'm wanting to drop line 13 from http://bazaar.launchpad.net/~ubuntu-on-ec2/vmbuilder/automated-ec2-builds/view/head:/conf/natty-server.conf [16:38] but dont know that i can yet. [16:39] apparently festival.smc needs to be in /etc/ not /usr/share/festival/ [16:39] thanks =] [16:40] Question: My samba domain controller will not serve up scripts upon login, and acts very slow when not connected to the Internet. Once the net connection comes back up, everything is fine. Anyone know why this may be happening? [16:41] how do i change th kernel config parameter in ubuntu. I would like to change the value of CONFIG_VHOST_NET=y from m. [16:42] The DNC is ubuntu 10.04.2 for the above question [16:43] ppetraki: so, udev.log is basically filled with junk like http://paste.ubuntu.com/593611/ [16:43] ppetraki: (this is for bug 644489) [16:43] Launchpad bug 644489 in multipath-tools "constantly changes /dev/disk/by-id/{scsi,wwn}-* LUN symlinks with multipathing" [High,Confirmed] https://launchpad.net/bugs/644489 [16:43] ppetraki: why would /dev/sdc keep being marked as 'changed'? [16:44] could kpartx.vs.multipath be the problem here? [16:44] so the udev rules can run, and find that there's no real work to do [16:44] it's not exactly fine grained :/ [16:45] so what's the bug here, log spamming or some real hung processes? [16:46] well, lots of system time taken up, and symlinks from wwid by-id to /dev/sdX keeps changing [16:47] something is very wrong there [16:52] /dev/disk/by-id/wwid* is supposed to point into /dev/mapper... [16:52] not to /dev/sdX [16:52] the symlinks actually change, as in the signature, or they're just being recreated needlessly? [16:53] oh I see [16:53] so do we know what kernel event prompted all this traffic? [16:57] New bug: #759940 in mod-wsgi (main) "webdav modifies uri and path_info attributes" [Undecided,New] https://launchpad.net/bugs/759940 [16:57] ppetraki: no, [16:57] ppetraki: but hey, it looks like he's still using '-Q' flag to modprobe [16:57] hallyn, yeah, that's busted, and a quick fix [16:57] ppetraki: but soI wonder if it's evidence of bug 737027 [16:57] Launchpad bug 737027 in multipath-tools "kpartx udev rule is broken" [Medium,New] https://launchpad.net/bugs/737027 [16:58] well let's ask the guy to try '-q' i guess :) [16:58] that'll work :) [17:00] ppetraki: feh. he's the one who originally submitted that bug about -Q! [17:01] i don't know how, but I need tocome up with a better way to represent bugs for myself. Faster fetches, but also showing me what other open bugs people and packages have [17:01] all right, so on that note, i need to go check for burning bridges in kvm/libvirt for natty. [17:02] New bug: #759943 in mod-wsgi (main) "mod_wsgi.so-3.2 gives errors" [Undecided,New] https://launchpad.net/bugs/759943 [17:16] smoser: /boot/grub is created by the action of installing GRUB. We've tried to avoid putting it directly in packages because some people like to unmount /boot except when upgrading, to minimise risk of filesystem wear. Your fix is OK for beta-2, but I would prefer it if you created it in postinst or in update-grub-legacy-ec2 or something like that [17:17] cjwatson, :-( [17:18] sorry. i would have done that for sure. that was my original thought. [17:18] it is *really* not a big deal, though, as if you're installing grub-legacy-ec2, you're really on a virtual machine, and likely in EC2 where /boot is probably on / [17:22] darn, unity-2d doesn't work at all on kvm [17:23] smoser: right, hence why it's OK for beta-2. It would be good to clean it up for final, that's all. [17:23] cjwatson, ok. i'll do that. you'd just do 'mkdir /boot/grub' in postinst > [17:23] ? [17:24] mkdir -p, yes [17:39] i'm running ubuntu with /dev/sda5 linux swap . I just resized disk in vmware ESXi, when I load gparted it doesn't let me resize the partition, since right at the end of dev/sda1 is dev/sda5 then unallocated space [17:42] ppetraki: drat, -q didn't fix it for him [17:49] so right now the installer prevents you from having an upper case letter as the first character of a username [17:50] this is a bit silly. There's no technical reason this should be the case [17:50] I'd like to change it in a later version of Ubuntu (see: not natty), but I'm looking for examples of things this might break [17:51] cjwatson mentioned you folk may have encountered software that's more likely to have such interoperability issues [17:58] hello all [17:58] need help with something if someone would be so kind [17:58] Daviey: it has just being pointed out to me that the cobbler web setup for apache takes over the whole server basically [17:58] so, i've set up a Time machine with a samba share and netatalkd [17:58] (on my linux server) [17:58] but for some reason, the time machine backups never finish [18:01] zul, oh goody. [18:01] zul, is there a bug? [18:02] Daviey: its coming [18:16] New bug: #760012 in cobbler (universe) "cobbler-web unnecessarily overrides default DocumentRoot" [Undecided,New] https://launchpad.net/bugs/760012 [18:17] smoser, It would probably be a good idea to raise two blueprints when you have a chance... cloud-server-p-image-build-process and cloud-server-p-desktop-image (i think pitti will want to take that one over fwiw). [18:17] what would be in 'image-build-process' ? [18:18] smoser, I think two things... one you were talking about using new build platform earlier, and secondly handling of migration to centralised builds. [18:18] what do you think? [18:19] sure [18:21] New bug: #760019 in cobbler (universe) "A profile's --kopts-post option does not work" [Undecided,New] https://launchpad.net/bugs/760019 [18:26] anybody good with mailers? [18:26] need help with sendmail [18:29] how do I rollback a fs ? I want to put some data on /path/foo .. so I create a tar of /path/foo first and then make tons of changes.. what is the best way to roll back to orig /path/foo ? do I untar the old /path/foo to a new location and point /path/foo to it ? [18:29] in solaris zfs rollback does all the magic [18:29] rm -rf /path/foo [18:29] tar -xzvf pathfoo.tar.gz [18:29] tar -czvf preserves file structure [18:30] so if you run tar -xzvf at /path/.. [18:30] it will be as it was [18:33] NightDragon: ok .. thanks a lot [18:34] np [18:34] NightDragon: do I need a -p to preserve file mode and ownership ? [18:34] NightDragon: or that is default ? [18:35] if you didnt create the tar with -p i dont think it will matter === hallyn is now known as hallyn_afk [19:02] how do i change th kernel config parameter in ubuntu. I would like to change the value of CONFIG_VHOST_NET=y from m. [19:36] rnigam: sounds like you want to run a custom kernel [19:49] anyone help with a 10.04 bios RAID issue ? [19:49] i created a RAID 5 logical disk in my bios, but ubuntu is still reporting it as individual drives [19:49] is there something i need to do ? [19:50] yeah, not use fakeraid [19:50] either use linux SW raid, or true HW raid. [19:50] use use mdadm in ubuntu and don't use the BIOS raid [19:50] ok i figured that's what i have to do [19:52] ap0c, for any linux to detect your fakeraid, you need dmraid loaded [19:53] i did that, but it still didn't recognize my fakeraid [19:53] i just turned off the bios RAID and am going to do sf raid now [19:53] it's not my boot device so not really important [19:53] just 4 1TB disks going to use RAID 5 sf mdam now [19:53] these ok directions? https://help.ubuntu.com/community/Installation/SoftwareRAID === carcinogen75 is now known as zz_carcinogen75 [19:57] pmatulis_: Well I would like to add a vhostnet module to the current kernel. Is that possible? [19:58] * RoAkSoAx off to lunch === marass is now known as mcas [20:11] smoser: uec-publish-image is failing on maverick and natty images on euca i386 === hallyn_afk is now known as hallyn [20:18] hggdh, can you give more information please ? [20:18] jbernard: do you have libcgroup update for natty queued up for natty? [20:18] (with the security fix)? [20:19] smoser: opening a bug now. But I was preparing to beta2-test uec on i386, and -- in the process -- I load lucid, maverick, and natty kernels. [20:19] hallyn: it's in unstable now [20:19] smoser: lucid went OK, maverick and natty failed. Hum. I can try *not* using --use-loader [20:20] hallyn: so it should be ready for merging [20:20] jbernard: ok - have you talked to skaet about it? [20:20] hallyn: your patches should apply cleanly on top [20:20] (since we're late in natty cycle) [20:20] hallyn: i haven't [20:20] hallyn: is that the next step? [20:21] jbernard: sigh - way too late for my patches (the upstart ones, right?) to go in for natty, unfortunately [20:21] jbernard: i think so [20:21] jbernard: that is, if the change is going to be rejected for natty anyway you may as well save yourself the effort [20:21] hggdh, can i get at this uec somewhere ? [20:21] though with security fix I find that hart to believe [20:21] hard [20:21] smoser: if you have access to the uec test rig, certainly [20:22] what host? === dendrobates is now known as dendro-afk [20:22] smoser, santol [20:23] smoser: zeroed in --user-loader on uec-publish-tarball [20:23] what is santol's ip ? [20:23] just a sec [20:23] smoser: 10.55.55.7 [20:25] skaet: is it possible to sync libcgroup from unstable to fix a security issue? [20:25] how do I kill this process? [20:25] root 7647 27139 4 19:20 pts/0 00:00:05 [tar] [20:25] kill -9 7647 did not work [20:25] do I need to reboot ? [20:26] hallyn: if we can get the merge accepted, then we can upload a version with your patches to natty-updates potentially? [20:26] jbernard: yeah, we push it to o-series and then consider SRU to natty [20:26] hallyn: ok, I think that will work well [20:26] jbernard: and since there are real bugs fixed by it, SRU has a chance [20:26] cool [20:26] pls do talk to skaet though [20:27] hggdh, so where can i get credentials? [20:27] libcgroup is in universe. there shouldn't be a huge problem with that [20:27] oh. i keep forgetting about that [20:27] jdstrand: so is it subject to freeze now? [20:28] smoser . ~/uec-testing/scripts/users/admin/eucarc [20:29] i think you're on soncoya, not santol [20:29] hallyn: if its bugfix only, no. if there is more, you still need approval from a member of the release team [20:29] oh wait. no [20:29]