=== medberry is now known as med_away [00:13] quentusrex, well the answer to your original questions seems to be here. /usr/share/pyshared/VMBuilder/plugins/libvirt/__init__.py [00:14] Hi, everybody. I was wondering which mailbox format works best for mass mailing (newsletter app) [00:17] ignarps: yea so its [1406457.588623] scsi 3:0:0:0: Direct-Access SanDisk Cruzer 7.01 PQ: 0 ANSI: 0 CCS [00:17] how does that help me out? [00:21] i've been using kvm/libvirt for since 8.10.. this is the first time i've read about (or noticed maybe) libcap2-bin ... what is the deal with this? what does it do? [00:22] i've never installed it, and i've used bridged networking [00:22] is it for bridging? i [00:22] i'm reading this: https://help.ubuntu.com/community/KVM/Networking#bridgednetworking [00:22] what's listening on port 139 on a fresh ubuntu server install if I didn't select samba server to be installed? [00:25] grout, try: dmesg | grep sd It should show up as a drive [00:26] hypetech, use lsof -i to find out [00:28] ignarps: looks like it's avahi [00:28] ah avahi nice [00:29] is avahi necessary if I'm not wanting network discovery? [00:31] hypetech, it is not needed if you don't care about auto discovery of network resources [00:31] ignarps: ty [00:31] you can still manually mount things [00:53] Cheers all, I've just completed a bare metal install of Lucid Server 64, what are some of the first things you do after a base install that make life when working with a newly installed Lucid box? [00:53] *easier [00:55] onecrazycat: heh.. I can't name them anymore because I rely on command-not-found to tell me what I need. ;) [00:56] So you don't do anything really, you just get to work and as needed you install packages? [01:00] onecrazycat, install etckeeper :) [01:01] then figure out how you are going to backup the data on your new server [01:01] Actually true I do always throw etckeeper in. [01:02] onecrazycat: unless you're only going to ever have one server.. its a good idea to put those "essentials" into config management.. then you don't have to remember them. [01:02] Like Chef or Puppet? [01:03] or cfengine. or throw your own :) [01:03] Would a gut repo supersede the use of etckeeper? [01:03] *git [01:03] Yeah for a long time I had a dir full of bash scripts named ##_something that I'd run after install [01:03] etckeeper can use git if you like. the nice part about etckeeper is it integrates into apt [01:04] Hrm, cool. [01:04] Reading about it now @ http://fnords.wordpress.com/2009/02/23/etckeeper-chronicles-1/ [01:04] onecrazycat: some kind of VCS with config management is considered the best practice these days.. "Infrastructure as Code" === eagles0513875 is now known as zz_eagles0513875 [01:07] SpamapS: Agreed wholeheartedly. A noble challenge. [01:07] onecrazycat: puppetmaster + etckeeper would be all you need [01:08] SpamapS: I'd maintain if you're messing with server configurations often enough you think you need etckeeper, you aren't doing it right. [01:09] ScottK: right, etckeeper is just for the puppet master box. [01:10] ScottK: though having it on the individual nodes can be eye opening as to how often people are having to work around puppet. :) [01:20] hosting your puppet config out of a git repo is also a smart move :) [01:20] I don't really "get" git [01:20] but I'm a bzr fanboi :-P [01:21] you don't git it? ;) === med_away is now known as medberry === asac_ is now known as asac === alan23424_ is now known as alan23424 [03:09] does anyone know anything about a patch for ndiswrapper? I'm getting errors installing it so from what I've understand from googling I need a patch. I can't figure out how to get/install it though. [03:12] econ: I recommend you just go buy a better wifi card for your server [03:12] If you're using ndiswrapper you're already running with scissors, trying to patch and reroll your own ndiswrapper is only going to make things worse === erichammond1 is now known as erichammond [03:16] yeah, I've been trying to get it working for so long now that I'm about ready to just give up and get a new one [03:17] That's certainly my recommendation [03:17] Unless you're a soup kitchen or something and can't afford the $50 for a new unit. [03:34] I am getting lot of these device-mapper errors http://pastebin.com/HUS2Kqiz .. [03:34] any idea what is going on? [03:35] I dont have any storage .. and I am using hardware raid [03:35] Well 251:* is LVM [03:35] Not sure what 252 is [03:36] mpt-status -i 2 [03:36] ioc0 vol_id 2 type IM, 2 phy, 67 GB, state OPTIMAL, flags ENABLED [03:36] ioc0 phy 1 scsi_id 4 FUJITSU MAY2073RCSUN72G 0401, 68 GB, state ONLINE, flags NONE [03:36] ioc0 phy 0 scsi_id 3 FUJITSU MAY2073RCSUN72G 0401, 68 GB, state ONLINE, flags NONE [03:36] Maybe CCISS or some shit? [03:36] so as you can see my disks are fine [03:36] axisys: WTF is mpt? [03:37] twb: that is for mptsas [03:37] 02:03.0 SCSI storage controller: LSI Logic / Symbios Logic SAS1064 PCI-X Fusion-MPT SAS (rev 02) [03:37] Blergh [03:37] Is that a fakeraid card that you've set to passthrough by creating 1*RAID0 arrays for each physical disk? [03:38] twb: i think so .. it has been a while since I installed the OS on it.. [03:38] i think i built a raid1 .. [03:38] because df -h / shows [03:39] /dev/sdb1 65G 19G 43G 30% / [03:39] http://pastebin.com/r9cFFYNj <-- dmesg [03:41] I am guessing I dont need the device-mapper [03:42] does not look like I am using it.. not sure why I installed it in the first place.. probably was drunk.. heh [03:43] lsmod | grep dm [03:43] dm_raid45 75532 0 [03:43] xor 4685 1 dm_raid45 [03:44] since I am not using any raid5 .. that is probably need to be removed or blacklisted.. [03:44] i guess i could blacklist it and reboot.. see if that removes the noise [03:46] everytime I start mutt -F custom.muttrc I get these panic messages .. http://pastebin.com/HUS2Kqiz [03:46] running mutt w/o that muttrc does not trigger it.. [03:47] could it be my fs is corrupted? [03:47] i dont know what it mean by run recovery .. [03:48] there is no recovery tool.. is it mean backup your data ? i dont follow.. i did not think it is mutt related [03:48] http://picpaste.com/mutt-eY5cmneY.png <-- this is the png I am referring to [03:56] boy was I wrong? those panic messages were related to mutt header cache.. removing the case and restart mutt got rid of thos panic errors [03:56] s/case/cache/ === squishy is now known as SquishyNotHere === zz_eagles0513875 is now known as eagles0513875 [07:27] hey all, i installed x and fluxbox (for kicks) and i cant seem to run it, [07:27] when i do startx [07:27] i get a black screen :( === evilmist is now known as mrmist [08:13] OMG.. someone PLEASE help me get postfix working correctly on my server!! [08:13] I've spent ALL freakin' day on this thing! [08:15] Nothing? No one? [08:15] I've gone through all the postfix guides and I still can't send an email through any user addresses on my server.. [08:21] New bug: #771662 in php5 (main) "php5 segfault when zend guard loader used " [Undecided,New] https://launchpad.net/bugs/771662 === kim0 is now known as kim0|vacation === airtonix_ is now known as airtonix [10:22] New bug: #761971 in mdadm (main) "no notification that one of the disks has been disconnected in a raid 1 configurtion" [High,Confirmed] https://launchpad.net/bugs/761971 === jjohansen is now known as jj-afk [14:11] morning all [14:20] TREllis: o/ [14:21] RoAkSoAx: o/ pong [14:21] TREllis: so I looked deeper into the mini.iso import [14:22] RoAkSoAx: recently? saw your bug update a few days ago [14:23] TREllis: basically, if the name contains the arch and we specify the --os-version, then it will be imported correctly and created distro/profile [14:23] TREllis: if the name does not contain a valid arch, then we need to specify both --os-version and --arch [14:23] TREllis: off course, after that we need to edit the profile and add a preseed [14:24] TREllis: so something like this: [14:24] sudo cobbler import --name=natty-i386 --path=/mnt/mini --breed=ubuntu --os-version=natty [14:24] or [14:24] sudo cobbler import --name=natty-mini --path=/mnt/mini --breed=ubuntu --os-version=natty --arch=i386 [14:24] and the way how I change the preseed for NQA: [14:24] sudo cobbler profile edit --name=natty-mini-i386 --kickstart=/var/lib/cobbler/kickstarts/nqa.seed --kopts="priority=cr sudo cobbler profile edit --name=natty-mini-i386 --kickstart=/var/lib/cobbler/kickstarts/nqa.seed --kopts="priority=cr [14:24] any ideas why my server is taking a while to respond when it has to verify my password with sudo ? [14:24] RoAkSoAx: hmmm that's exactly what I've been doing, but no distro or profile is created... let me double check [14:24] itical locale=en_US" [14:24] sudo cobbler profile edit --name=natty-mini-i386 --kickstart=/var/lib/cobbler/kickstarts/nqa.seed --kopts="priority=critical locale=en_US" [14:25] TREllis: note that --name=natty-i386 (and not specifying --arch) will create profile/distro [14:25] RoAkSoAx: that's exactly what I've been doing to work around the current traceback [14:25] TREllis: the traceback can be ignore [14:25] in manager_import_debian_ubuntu there's two functions process_repo [14:25] RoAkSoAx: oh, I could never get the profile/distro to be created, had to do those manually after an import [14:26] TREllis: one empty, one debian related... so I'm guessing that it is either an old function that never got completed, or a function that we will need to re-write in the future [14:26] RoAkSoAx: I think I posted this on the bug... but I think it's related to the mirroring code [14:27] RoAkSoAx: cobbler will use debmirror or yumdownloader to grab a whole archive mirror for a distro [14:27] TREllis: in my case I the profile for the mini ISO is created if A) --name contains arch --name=natty-mini-i386 *and* --os-version=natty is specified *or* B) --name= does not have valid arch, and --os-version *and* --arch are specified [14:28] RoAkSoAx: ah right [14:29] RoAkSoAx: funny thing is, with the alternate cd because it detects it correctly, don't need to specify any of those :) [14:29] TREllis: right, that's because the alternate has the .disk directory and contains the info [14:29] of the release/version/arch of the image [14:29] yeah [14:30] TREllis: in the case of the mini ISO, that information is not available so there's nothing much we can do but manually tell the os-version|arch [14:31] RoAkSoAx: yeah I saw the bug got marked invalid, tbh having to specify the options is fine, we just need to avoid the traceback (even if it can be ignore, it's still a bug!) [14:31] d* [14:32] TREllis: yeah I already have the patch for it, it is just commenting 1 line and adding 1 pass [14:32] TREllis: till the repo thingy is figured out [14:32] RoAkSoAx: actually, rechecking our script, it seems we already noticed it creates it if you specify arch in the name [14:34] TREllis: ;) now, the repo thing might be a left over from old code that never got finished/working or something went wrong when changes were started to be made for Ubuntu [14:34] RoAkSoAx: http://bazaar.launchpad.net/~orchestra/orchestra/trunk/view/head:/provisioning-server/usr/sbin/orchestra-import-isos :) [14:36] TREllis: cool, that's for the automatic import that kirkland was talking about? [14:36] RoAkSoAx: yep [14:39] TREllis: ok so browsing through the code I can see that the process_repos function was intended to be empty for now, and apparently the one with the logic to import debian repos is a leftover code from a merge [14:40] RoAkSoAx: ah-ha :) [14:42] TREllis: yeah that process_repos is old code when the imports only handled debian and the new code from chuck didn't really implemented, and apparently someone merged both codes together [14:43] RoAkSoAx: bingo cool :) [14:47] TREllis: and in the old code, that function doesn't seem to be used from within manage_import_debian_ubuntu.py [14:47] TREllis: so it is safe to ignore it [14:54] RoAkSoAx++ thanks for checking it out [14:55] TREllis: no worries! Will put them in a PPA later today [14:56] New bug: #771818 in squid (main) "SIGABRT on unexpected data" [Undecided,New] https://launchpad.net/bugs/771818 [15:06] New bug: #771698 in libnss-ldap (main) "/usr/bin/id does not show ldap groups" [Undecided,Incomplete] https://launchpad.net/bugs/771698 [15:11] Daviey, hey man, release day tomorrow, what's the word on the street? [15:13] rickspencer3, A disaster... we need to cancel the release [15:13] nah, just kidding. [15:13] We are generally looking pretty good.. [15:13] Daviey, excellent! I've been looking for a crisis to give my life meaning this week! [15:13] heh [15:13] brb [15:13] too late to back pedal, respin!! [15:23] i heard that openstack will be part of ubuntu 11.04, but packages.u.c does not have any result for packages === robbiew1 is now known as robbiew [16:22] RichiH: its in universe [16:23] RichiH: but the reason packages.u.c doesn't have it is the package names don't have 'openstack' in them [16:24] RichiH: http://packages.ubuntu.com/search?keywords=openstack&searchon=all&suite=natty§ion=all [16:26] SpamapS: aye, i found it but didn't think of stating it in here [16:26] thanks & sorry [16:27] hallyn: hi! I haven't kept up on the lxc stuff. would you mind commenting on bug #645625? [16:27] Launchpad bug 645625 in lxc "lxc container can power-off host machine" [Medium,Confirmed] https://launchpad.net/bugs/645625 [16:28] New bug: #645625 in lxc (main) "lxc container can power-off host machine" [Medium,Confirmed] https://launchpad.net/bugs/645625 [16:29] jdstrand: k [16:30] thanks [16:32] New bug: #611194 in euca2ools (main) "No secure way to protect against MiM attacks" [Medium,Confirmed] https://launchpad.net/bugs/611194 [16:37] hey guys anyone know or have a good link on how i can setup a network install for ubuntu server [16:37] eagles0513875: google for it [16:38] ok RoyK [16:38] there are multiple options [16:38] eagles0513875: is this a one time thing or are you going to do it on an ongoing basis? [16:39] eagles0513875: the latest version of Cobbler can provision and install Ubuntu [16:39] ahh ok [16:39] a lot of tools are also based around cobbler [16:39] And is available in universe for 11.04. :-D [16:40] humm ok so i install it on my desktop then just point my server to the ip of the server hosting the image [16:40] no [16:40] you don't install from an "image [16:40] nice job on researching there [16:41] would something like this also do the trick https://wiki.koeln.ccc.de/index.php/Ubuntu_PXE_Install [16:41] you need to look at options such as profile built installs, interactive etc [16:41] depends as SpamapS is this a one of thing, or a regular thing [16:41] "off" [16:43] TREllis: ok so further investigation, after importing an ISO/dist it should create a disable repo for such distro. It is not doing so because "process_repos" is not being called [16:44] TREllis: according to https://fedorahosted.org/cobbler/wiki/DebianDeployment this should be working, however, for what I can see this might have been disable when merging ubuntu changes [16:44] RoAkSoAx: it is, I use it with sat [16:44] I have used that on 10.04 builds [16:45] ikonia: right, are we talking about cobbler 2.1.0? [16:45] hang on - I'll check the exact version [16:45] 2.0.9 [16:45] SpamapS, hallyn... my friends with the dell san have a new question... :) They noticed about a week ago that they are seeing two copies of portmap running. [16:46] no symptoms and i'm still gathering info. just wondering what you make of that? if anything. [16:46] marrusl: two portmaps? thats weird. [16:47] ikonia: ahh, then that's the issue :) [16:47] SpamapS, he's going to see if he can reproduce it in test. right now they just see it in production. [16:47] I wanted him to turn up upstart logging to see what starts them, but he can't reboot reight away. [16:47] RoAkSoAx: ah I see, so it is the repo mirror code, just should be disabled by default for netinst and isn't [16:47] so we'll see. [16:48] RoAkSoAx: there is a 2.1.0 package pending, so I guess we'll find out [16:49] TREllis: yeah apparently in old code, when importing debian, it created a disabled repo_mirror, which now is not happening [16:49] RoAkSoAx: there is probably quite a bit of cobbler ubuntu/debian code clean up to do anyway as I knew a long time ago a friend in a previous job implemented the initial support, which got disabled due to lack of maintainers then our patches and before that someone else was merging some at the same time for Ubuntu support [16:50] TREllis: it looks like it was stagnent for a while then took a big jump recently [16:51] marrusl: since portmap binds to a port, it shouldn't be able to run twice, unless its running inside a container of some sort [16:51] TREllis: yeah that's what happened. Chuck send patches for different .py files for ubuntu/debian, someone right after that merged changes to a debian_ubuntu.py file, and apparently he noticed that the ubuntu/debian support was added, and then he merged chuck's patches into manage_import_debian_ubuntu.py [16:52] TREllis: I'm guessing that something went wrong there that now, disabled repos are not being created [16:53] SpamapS, good point. but yeah, little chance they are doing that. they said it was unkillable (or respawning, not sure) and comes back after reboot. [16:53] SpamapS, and nothing useful in the logs I have except them trying to kill it once. [16:54] RoAkSoAx: ah-ha :) [16:54] SpamapS, since there's no functionality loss, we can take it easy. hopefully he can reproduce in test. [16:54] we=me :) [16:54] TREllis: Oh well, I guess we'd need to re-write process_repos as I was thinking to support both debian/ubuntu [16:55] hm, shouldn't openstack be a meta package? [16:55] TREllis: bug again, this probably wouldn't have been noticed without your testing :) [16:58] RoAkSoAx: sooner rather than later :-) [17:01] pmatulis: its such a complex system.. not sure if a meta package is really the right way to go.. [17:01] $ apt-cache show eucalyptus [17:01] N: Unable to locate package eucalyptus [17:01] TREllis: indeed! you are adding items to my list of todo's though :P [17:02] RoAkSoAx: I'll buy you a beer at UDS ;-) [17:03] TREllis: hehe sounds like a plan ;) === NG_ is now known as ng_ [17:07] Good morning everyone. [17:07] Has anyone used the UEC Beginners Guide by CSS Corp Open Source Services? === ahs3 is now known as ahs3-afk [17:18] TREllis: alright, so it seems that cobbler no longer supporst adding apt mirrors [17:21] RoAkSoAx: ah, perhaps that was removed back when the first support of ubuntu&debian was stripped [17:22] TREllis: there's some logic in the code to handle apt repos but doesn't seem to be able to create one as it doesn't recognize apt as a breed for the repo [17:29] TREllis: fixed, lol! [17:36] so I have only about 3 users on my ubuntu servers, all of them have sudo access and we don't really need home directories. Is their anyway I could make the UID 0 for these users, remove the home directories so these users always use /root/ for their home dirs? [17:36] also I wanted to make the bash_history have the user name and time stamp [17:37] zul: quick question about cobbler. Should the creation of a repo mirror be done when importing an ISO cause right now is disabled even though reposync action is enabled?? [17:37] evilsushi: if you're going to do that, why don't you just have everybody log in as the same user? [17:37] RoAkSoAx, it should be done with debmirror iirc [17:38] zul: yeah the thing is that when importing the ISO the part that creates a disabled repo is ignored [17:38] RoAkSoAx: ok i might not have written that part yet :) [17:39] evilsushi: if you give multiple users the same uid, name-uid mapping is incredibly unreliable (it'll give the same answer for everyone. because they're all the same user). so the bash_history bit wouldn't work [17:39] zul: hehe, there's a written part for debian that creates the mirror for debian and debian-security updates, so I guess it is just matter of adding support for ubuntu mirrors [17:39] zul: which I'm doing now :) [17:39] will show you the patch once I have it done [17:39] RoAkSoAx, ok [17:40] nah it would work, if all users are UID 0 you could add something like this to the bashrc [17:40] PROMPT_COMMAND='echo $$ $STAFFID \ "$(history 1)" >> ~/.permanent_history' [17:40] this would give you a duplicate bash history type [17:42] I guess that'd depend where $STAFFID comes from. but I think sticking everyone in one account, and then reinventing per-user accounting, is a sign you're solving the wrong problem [17:43] yeah everyone on the same account is a nightmare for keys [17:44] why do you need uid = 0 anyway if they can all sudo? [17:44] just make them use the same home dir and keep using sudo [17:45] STAFFID=`logname 2> /dev/null` [17:46] if you have more than one user with the same uid, I wouldn't trust logname to tell the truth [17:46] I'd try it on a spare machine / vm. it can be interesting. you can edit passwd and shadow to give more than one user uid 0 [17:47] but it does completely destroy the system's ability to put a name on a face [17:48] name on a face? [17:48] mapping names to uid and back. most things will return the name as being the first match in passwd [17:48] name to a uid then === jj-afk is now known as jjohansen === ahs3-afk is now known as ahs3 [18:21] hello, I am getting an rpc error when I try to mount a volume from one of our nfs servers [18:21] http://pastie.org/1840045 [18:22] I was wondering if I could get an opinion on this tx [18:25] whoops! forgot to include the machine info.. here it is.. http://pastie.org/1840060 === hallyn is now known as hallyn_afk [18:28] bluethundr: já tentou colocar o ip da maquina no lugar do hostname? [18:29] loopback_br, I only speak english. sorry [18:29] eu falar um poquenho portuges [18:31] New bug: #771924 in postfix (main) "package postfix 2.8.2-1ubuntu1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 75" [Undecided,New] https://launchpad.net/bugs/771924 [18:32] bluethundr: is portmap running on the local machine? [18:33] bluethundr: and for that matter, statd ? [18:34] hi SpamapS, yes both portmap and statd are running [18:34] however I just tried to mount the same mount/nfs server from two other hosts (one ubuntu one centos) and got the same result [18:35] so I think this points to a problem with the remote machine [18:42] bluethundr: sounds like it. No errors on the other side? [18:42] SpamapS, just trying to ssh in.. having key issues..sigh :) [18:43] what would you recommend for aws ec2, maverick or upcoming natty? any noticeable differences? [18:45] bluethundr: sorry man, i forgot... i've already tryed to change the hodtname for the ip of the machine in fstab? [18:45] bluethundr: *you've [18:45] loopback_br, yep as a matter of fact I have [18:45] it times out when I do that [18:46] bluethundr: try 'showmount -E HOSTNAME' from the client side to see if its even succsfully exporting anything [18:46] er, showmount -e [18:47] I get this back [18:47] portmap getport: RPC: Success [18:51] bluethundr: what you put on your exports file? [18:52] loopback_br, I didn't setup the nfs server [18:53] * SpamapS is always amazed that 15 seconds after he posts to his blog, googlebot is already aware of it [18:54] bluethundr: your sdf doesn't have a partition? [19:00] are there well maintained postgres 9.x packages available for ubuntu or do people build from source? [19:00] loopback_br, yes it does but it just looks like an EBS volume if I'm not mistaken [19:05] bluethundr: i've never worked with EBS =/ [19:06] loopback_br, meh.. you get used to AWS after a while :) [19:08] TREllis: almost fixed... now the only issue is debmirror [19:08] kirkland: ping [19:08] RoAkSoAx: pong [19:08] kirkland: when you were looking for mirroring software [19:08] kirkland: did you run into HOME not defined in environment! [19:08] with debmirror? [19:08] RoAkSoAx: hmm, no, i don't think so [19:09] kirkland: I'm patching up cobbler to mirror archives [19:09] RoAkSoAx: cool [19:09] kirkland: but that error appears. It is not cobbler the one that throws the error but it seems debmirror is when running in a subprocess [19:10] RoAkSoAx: hrm, i don't think i've seen that [19:10] kirkland: when I run the same command in my terminal, it works because it finds $HOME defined [19:11] kirkland: but running via cobbler it fails, so I guess we'll have to use another mirroring sotware if I don't find a solution [19:11] RoAkSoAx: can't you just define $HOME? [19:12] RoAkSoAx: is this running as the cobbler user? [19:12] RoAkSoAx: cobbler's home is /var/lib/cobbler [19:13] kirkland: uhmmm that would make sense, though cobbler uses subprocess to launch debmirror, which I presume doesn't have a $HOME [19:23] bluethundr: rpcinfo -p [19:23] RoAkSoAx: then just hard code a HOME value :-) [19:28] kirkland: that's what I'm trying but doesn't seem possible [19:28] but will keep trying [19:28] aljosa: the Debian packages should build/install on Ubuntu, since Martin Pitt (an ubuntu tech board member) is the maintainer. :) [19:29] SpamapS: thanks for info [19:29] aljosa: I'd guess that 9.0 will land in 11.10 and then it can be submitted to backports for lucid/maverick/natty === ivoks is now known as banana === banana is now known as ivoks [19:31] ivoks: orange you glad I didn't /nick banana ? [19:33] eh? [19:33] ivoks: old dumb kids joke [19:33] ah nick = steal [19:33] or not? :D [19:33] say [19:34] those american jokes... [19:34] Does anyone know why Ubuntu 10.10 Server's minimal virtual install (kernel -virtual) does not contain the Hyper-V modules, if it is possible to restore those to that kernel, and if this is rectified in the next release? [19:34] its a knock knock joke. ;) [19:34] as i said... :D [19:34] Anpheus__: because -virtual is for the guests, not the hosts? [19:35] Yep, and these are the guest paravirt modules [19:35] Oh well then wtf ;) [19:35] start bashing at #ubuntu-kernel [19:35] loopback_br, thanks I appreciate the help but I think the EBS volume bay actually be what I need [19:35] I just talked to the senior and he let me know [19:35] ty again [19:35] Yeah, I don't know why the guest hv modules are missing from the -virtual build, but I ran into it when configuring 10.10 yesterday. [19:36] Does anyone know *why* 10.10's -virtual kernel doesn't come with those modules, or more importantly, does 11.04? [19:36] bluethundr: rpcinfo -p is to see if youe nfs server is working [19:36] Because if it's rectified in Ubuntu vnext I don't really care. [19:38] loopback_br, ok that's cool [19:40] bluethundr: see if it shows portmapper and nfs in the list [19:40] Anpheus__: #ubuntu-kernel is better place for that question [19:42] Anpheus__: vnext meaning, vtomorrow ? ;) [19:42] SpamapS: Yes, that one. :D [19:43] kirkland: done! :) Now cobbler also mirror's archive's for an imported ISO [19:43] TREllis: ^^ _> this also makes that no traceback is shown when importing a mini iso [19:44] Anpheus__: I don't see any thing with find /lib/modules/2.6.38-8-virtual -iname '*hyper*' on my ec2 instance [19:44] Try "hv" [19:45] And that's part of the problem. [19:45] $ find /lib/modules/ -iname '*hv*' [19:45] /lib/modules/2.6.38-8-virtual/kernel/drivers/staging/hv [19:45] Ok, is that 11.04? [19:45] nothing in that dir ;) [19:45] Dammit. [19:45] Anyone know where I can get source for crontab? [19:45] yes its 11.04 [19:45] Vixie cron [19:46] apt-get source cron ? [19:46] elb0w: dpkg -S `which crontab` [19:46] So someone went through the list of drivers to build the -virtual kernel, and basically culled things that they didn't think would be needed for a VM. [19:46] And in that process they excluded drivers that are *only* useful inside a VM. [19:46] Where can I file a ticket on this? [19:46] Anpheus__: doh! definitely should open a bug [19:46] Anpheus__: ask in #ubuntu-kernel .. they may already be aware [19:46] Channel is pretty dead [19:47] I asked though [19:47] Anpheus__: and otherwise, they'll make it clear where to put the bug / what to put in it. [19:47] but probably https://bugs.launchpad.net/ubuntu/+source/linux/+filebug [19:48] Anpheus__: note bug 560821 [19:48] Launchpad bug 560821 in linux "No hyper-v modules on linux-generic-pae/virtual" [Undecided,New] https://launchpad.net/bugs/560821 [19:49] Anpheus__: I'll mark that confirmed since you and 3 other people have asked for it. [19:50] What would cause the default virtualhost file to serve a particular document root, if the site is specified in another virtualhost? [19:51] Thanks SpamapS. [19:57] I'm looking into upgrading some incredibly old systems to new hardware and likely replacing the entire OS for a whole new steup. Something like 200 systems. One change is a move from using samba for authentication and profile storage to active directory. It sounds like we might end up needing to use Windows for the AD servers (we'd like to have local profile storage). The issue is.... I don't want to use Windows. [19:58] Any of you guys happen to know if it's possible to use ubuntu for the local profile/auth storage? [20:14] MTecknology: samba joins AD networks just fine and can be the file storage for profiles. [20:20] Hi everyone, Error: Unable to initialize gtk, is DISPLAY set properly? any ideas? (running 10.10 and executing pgadmin3) [20:29] TANATHOS: sounds like you're trying to run a graphical app in a non-graphical environment [20:34] pgadmin3 is a graph app? [20:34] TANATHOS: no idea [20:34] TANATHOS: but it sounds like it [20:34] :) no problem I will work around it [20:35] thank anyawy [20:44] hey all === marjo_ is now known as marjo [21:01] HI all [21:01] People, how I set ulimit in limits.conf? [21:02] dunno - tried it once - didn't work [21:06] i'm trying to add a network printer, i get it all set up, others have reported it as working, yet when i go to print something it prints multiple blank pages and i see the following in my messages log: http://paste2.org/p/1387772 [21:06] any ideas? [21:07] bad driver on the client [21:08] by others i dont mean other users of my printer, i mean others have set it up with success in their own environment [21:18] SpamapS: awesome :) [21:19] wow, I added an ip address to /etc/postfix/access as "ip_address deny" and also "ufw deny from " and it is still brute forcing my pop3 interface [21:19] urgent help please ? [21:20] you can always add the ip to /etc/hosts.deny [21:20] no? [21:20] or drop using ip tables [21:20] i think like iptables -j drop [21:20] or something similar [21:20] m4xx: ufw uses iptables [21:21] orudie: ufw status shows the blocked ip? [21:21] sry =[ [21:21] MTecknology, thats right [21:22] orudie: could you pastebin iptables -L ? [21:26] MTecknology, http://paste.ubuntu.com/599968/ [21:27] thanks for the info :) [21:27] good day all [21:27] Hello. I'm trying to install Perl's Digest::SHA1 module but it fails. After it's going to build and reads metadata I see "/bin/sh: cc: not found". Which packages provides cc? [21:28] Lenhix, gcc.. sounds like you need build-essential though [21:28] thx hazmat. Will try. [21:28] orudie: trying to block 109.169.55.114? [21:29] MTecknology, yeah [21:29] pastebin ufw status verbose [21:29] just for comparison [21:32] hazmat: gcc did it. Thx. [21:38] MTecknology, Anywhere DENY 109.169.55.114 [21:39] orudie: pastebin that output [21:39] MTecknology, thats the paste from ufw status [21:51] New bug: #772012 in cobbler (universe) "import sets incorrect tree path if arch is specified" [Undecided,In progress] https://launchpad.net/bugs/772012 [22:00] how many days to wait is it suggested until we update our servers to 11.04? [22:02] Daviey: some issues [22:07] vraa: wait until you tested 11.04, I would say? ;) [22:07] testing is for the weak! [22:07] then you should have upgraded 2 months ago! [22:08] ;-) [22:09] hmm [22:09] isn't it suggested you never update servers to 11.04 :) [22:09] but only to 12.04 [22:10] LTS right? [22:10] the problem is - i *must* have the latest version of mtr, gimp, and tuxracer on my vps! [22:11] * Patrickdk wonders if mtr even changed versions [22:11] I'm just looking forward to 2.6.42 [22:13] is there a changelog for the kernel versions anywhere? [22:13] Patrickdk: you can run servers on the interim stable releases [22:13] Patrickdk: its just that you will have to upgrade it, guaranteed, within 18 months of installing [22:16] it's useful if you need newer libraries or software versions... === ng_ is now known as NG_ [22:19] Indeed, though I'd like to see us make use of backports more now that it can be enabled and selectively utilized. === hallyn_afk is now known as hallyn [22:59] interesting.. mdadm does not exit non-zero if there are degraded arrays.. but I think it may have done so in the past [23:03] aha.. needs --test === gtaylor2 is now known as gtaylor [23:51] whats a better idea for vps os? 10.04 32 bit || 64 bit [23:58] how much RAM does your VPS have ? [23:59] I guess for most VPS 32-bit is probably better