[00:41] I just installed 12.04.02 on brand new hardware that has UEFI. I used custom partitioning (always do), and it doesn't boot. Now I see that I need a UEFI partition? [00:42] shodan45: iirc, you need something like 100 megabyte FAT partition right up front [00:42] right, that's what I'm reading now.... but is that absolutely needed? [00:42] it's the only OS on the box [00:43] i think it is, unless you can convince your bios to boot in a compatibility mode [00:44] hmm, I think I could do that, but I'd rather figure out UEFI [00:44] if you've got the time, that sounds like a good idea [00:44] it's not going away, anyway :) [00:48] ScottK: re "abandoning servers" .. Thats not just Canonical. The entire industry is changing. [00:49] ScottK: and its not just giant IT shops. Small shops, < 20 server shops, are finding it useful to have a self service "cloud" rather than angry gray beards watching over all the resources. [00:50] ScottK: But really, Canonical is following the money. If there was money in SMB servers, they'd be there. [00:50] * sarnold . . o o O O ("It took a lot of anger to get my beard this grey") [01:05] the whole industry is facing a big change right now, not just Canonical. Chasing the future is way better than attempting to catch up to the past. [02:21] I have a printer problem, details are here http://www.linuxquestions.org/questions/linux-software-2/fa-printing-shared-printer-win7-output-like-php-or-html-or-the-paper-4175465040/ [03:06] SpamapS: sure, but the stuff on the metal has to work for the cloud to be viable, so it's in many cases a false, short sighted choice to assume that basic infrastructure will continue to support the cloud stuff even though it's ignored. That'll be true for awhile, but not forever. [04:02] I have a printer problem, details are here http://www.linuxquestions.org/questions/linux-software-2/fa-printing-shared-printer-win7-output-like-php-or-html-or-the-paper-4175465040/ [04:58] I have a printer problem, details are here http://www.linuxquestions.org/questions/linux-software-2/fa-printing-shared-printer-win7-output-like-php-or-html-or-the-paper-4175465040/ === x3cion2 is now known as x3cion [09:04] what is correct, file or archive? [09:54] what is the relation or difference between cluster and cloud ? [10:05] I'm creating some users on a new server with default shell /bin/bash, but there's no directory/file colours. How can I enable this? [10:11] you may want to paste this in .bashrc http://pastebin.com/3iUWgC0D [10:12] hxm: great, thanks [10:12] np [10:26] is it possible to create an initial directory structure with /etc/skel ? [10:27] i want to put the .ssh/authorised_keys in by default [10:29] for every user? [10:29] yep .. every user added with useradd [10:29] sounds kinky [10:35] ha ha [11:06] authorised_keys → authorized_keys [11:06] en_GB → en_US === schmidtm_ is now known as schmidtm [11:28] zul: https://code.launchpad.net/~yolanda.robla/python-novaclient/havana/+merge/168021 [11:28] https://code.launchpad.net/~yolanda.robla/cinder/havana/+merge/167726 [11:31] I have a printer problem, details are here http://www.linuxquestions.org/questions/linux-software-2/fa-printing-shared-printer-win7-output-like-php-or-html-or-the-paper-4175465040/ [12:00] I have a printer problem, details are here http://www.linuxquestions.org/questions/linux-software-2/fa-printing-shared-printer-win7-output-like-php-or-html-or-the-paper-4175465040/ === baffle_ is now known as baffle [13:00] jamespage: ping [13:00] zul: morning [13:01] jamespage: good morning! so ceilometer needs a newer kombu, which is find and dandy [13:01] but the new version of kombu depends on a fork of amqplib [13:02] so im not sure how to go about this === hggdh_ is now known as hggdh [13:41] hey, I'm trying to upgrade a ubuntu server but the kernel is too old [13:41] if i do [13:41] linux-image-3.2.0-45-generic-pae [13:42] apt-get install !! [13:42] then I get a message telling me to apt-get -f install [13:42] but that needs to install libc6 [13:42] whicH I can't do because kernel is too old [13:43] help! [13:43] I need to basically "undo" the upgrade [13:45] anyone? === wedgwood_away is now known as wedgwood [13:55] jamespage: http://people.canonical.com/~chucks/ca/ [14:13] zul: https://code.launchpad.net/~yolanda.robla/python-keystoneclient/havana/+merge/168077 [14:30] hi [14:30] in case someone here has an answer to my question I post this link to my policyd question http://askubuntu.com/q/305282/136822 [15:00] something bad going on with the ubuntu-server image today, looking at it now [15:07] zul: https://code.launchpad.net/~yolanda.robla/quantum/havana/+merge/167230 [15:07] https://code.launchpad.net/~yolanda.robla/heat/havana/+merge/168098 [15:52] does anyone know why qemu-nbd -c successfully mounts a qcow image, but after qemu-nbd -d (which responds with "disconnected"), the qemu-nbd -c proc is still in the process list? [16:12] ShapeShifter499: While unfortunately I don't have much helpful input on your question, please do post full questions in the channel, as it improves the quality of the channel logs and lessens reliance on outside resources that may not necessarily be immutable links or readily viewable by all users. (An obvious exception, of course, is when it is appropriate to use a pastebin for configurations files, logs, etc.) [16:12] (Do feel free to include a link to the question as well, of course) === plars is now known as plars-afk [16:43] adam_g: ping [16:44] zul, hey [16:44] adam_g: can you take a look http://people.canonical.com/~chucks/ca/ [16:46] zul, ya [16:46] adam_g: thanks [16:49] zul, +1 [16:49] adam_g: thanks [16:51] Is it possible to bind the ports needed for NFS server to fixed number? I cannot firewall my setup. There is nothing about this in the community wiki article [16:54] Womkes: hrm, I thought in practice 2049 was used nearly exclusively? [16:55] Well there are two fixed ports [16:55] 111 and 2049 [16:55] then there are a few ports that are assigned randomly [16:55] for nlockmgr and mountd [16:55] these I need to fix so I can firewall them [16:56] I could also whitelist the source IP [16:56] but I don't prefer to do that [16:56] I like to be very specific with the firewall rules === plars-afk is now known as plars [17:39] why is apt telling me that it "kept back" quantal kernel packages on a fresh install of 12.04.2? [17:40] hi, trying to sort out something in the precise server image today [17:40] http://cdimage.ubuntu.com/ubuntu-server/precise/daily/pending/report.html shows that there were some uninstallable binaries for maas [17:41] but shouldn't those show up in http://people.canonical.com/~ubuntu-archive/testing/precise_probs.html? [17:47] maybe a timing glitch between when those were generated, seems to work ok now [17:47] but maas package is updated [17:48] In case anyone is interested, the company I work for is looking for a full-time devops/sysadmin in Seattle. PM me if interested and I'll send you the job description. [18:27] adam_g: http://people.canonical.com/~chucks/ca/ [18:27] zul, +1 [19:14] greeting all. Is there any equivalent to /proc/mdstat in the sysfs world? [19:14] I know that /proc is being deprecated so I wanted to use the appropriate sysfs resource instead. [20:06] a file was deleted from one of our servers via a 3rd party ftp program. the file appears to be gone, but the disk usage doesn't reflect this. is there anything i can do to trigger a refresh without rebooting the server, or some other thing to consider [20:09] baniir: check ls -l /proc/*/fd/* -- perhaps a process still has the file open [20:09] baniir: if the file had multiple hard links, perhaps only one link has been removed [20:10] sarnold: yep, thanks; just ran lsof | grep deleted and found my file. it appears several times. do you know how to close these properly [20:11] baniir: well... your programs may be leaking file descriptors, in which case this problem should be fixed. but if they legitimately have it open and you just want the disk space back, you could kill the process. when the last process holding it open dies, the disk space will be reaped [20:16] sarnold: you mean the program that was used to open the file? a colleague used mac os transmit for that… looks like apache is keeping the file open (it's a log file) [20:20] baniir: here's how you can get apache to rotate log files; the important part is 'apachectl graceful': http://httpd.apache.org/docs/2.2/logs.html#rotation [20:30] sarnold: thank you. attempting that command, i get "Internal error: Error retrieving pid file run/httpd.pid" [20:31] baniir: hrm. I wonder if you can do "service apache graceful", or if 'reload' or 'restart' is tied to a graceful restart... [20:31] sarnold: i have /var/run/httpd/httpd.pid [20:48] any suggestions on how to deal with a mount gone bad? kill −9 won't kill it, ctr-alt-del won't reboot === mrmist is now known as mist === wedgwood is now known as wedgwood_away === broder_ is now known as broder [23:50] I was trying to find a help channel for smartmontools, but was unable to find one so I suppose the next best place is here. Anyways, I am trying to setup smartd to run, and scan my devices, and then e-mail me when or if there is an error or some other issue. However, I have a MegaRAID controller, and I am not sure what to do with the /etc/smartd.conf file when those come into play. Does [23:50] anyone here use smartd in their setup?