[00:23] Does anyone here work with Java apps? seems like there are packages for things like jboss and activemq, but they are all hooked up to (have dependencies on) the open source JDK rather than Oracles [00:23] do people in the real world actually use openjdk for these apps? [00:24] or can i ignore the fact that things like gcj are dependencies of them? === n0ts_off is now known as n0ts === matsubara is now known as matsubara-afk [00:37] kirkland, ping [00:38] bigjools, was having issues copy and pasting within byobu. [00:38] it was about to get ugly, so i figured i'd ask you [00:38] :) [00:38] o/ kirkland [00:38] heh [00:43] ianp_: thats because oracle's JDK is not re-distributable [00:49] Hello everyone, I have an idea about a webpage where people would publish things they want to sell or rent but the problem is in the insert into of the database [00:50] how is this type of thing normally solved? do I have to approve every post? or I record the IPs of the last 5 mins and only let users post from 5 to 5 mins? [00:55] ffunenga: there are a TON of systems and that is often how those sites differentiate themselves [01:07] SpamapS: can you point me to one please? [01:08] craigslist and ebay spring to mind quickly ;) [01:14] sarnold: no... I'm talking about the system to prevent database flooding [01:14] or equivalent types of attack [01:16] is the last 5 minutes the best idea? [01:17] ffunenga: trivially vulnerable to spammers who write once every six minutes [01:17] or have access to botnet of thousands.. [01:41] ffunenga: the point is, this is how websites make their money [01:41] ffunenga: so its not likely that you'll find many ready-made systems for this [01:42] ffunenga: check out fail2ban and mod_security [01:47] those also combined with a few tricks and a catchapa [01:48] I really like reverse catchpa's (should figure out how to spell that again, it always confuses me, just like they do to me) [01:48] .. where they force _you_ to scribble a word illegibly? :) [01:49] no [01:49] where you add extra form boxs that should NOT be filled in [01:49] ah! hehe [01:55] SpamapS: Thanks, mod_security seems to be promising [01:56] patdk-lap: didn't think about captchas, nice one. What "few tricks" are you talking about? === n0ts is now known as n0ts_off [02:49] hello all, question. I want to make a script watch for a file to appear and then execute some commends when the file appears. What is the best route for this? A crontab every minute was my first idea but wondering if there is something better. [02:49] LeChacal2: apt-cache show incron [02:50] it'll use the inotify mechanism to only poke your program when a watched file / directory is actually modified -- saving gobs of drive, CPU, heat and power [02:50] that doesn't sound very safe [02:51] can't use sudo? [02:51] patdk-lap? [02:51] running whatever happens to be in the file [02:52] "execute some commands when the file appears" ... not "execute the commands in the file" :) [02:52] ah, ok [02:52] good paranoia though. I like that. [02:52] looking at incron seems like I still have to have a script running all the time waiting for incron feedback [02:53] yes the file would be empty and the script would read the title of the file to know what to do [02:55] LeChacal2: it'd be something like /path/to/directory/ IN_CLOSE_WRITE,IN_MOVED_TO,IN_MODIFIED /path/to/script $# [02:57] sarnold, hmm I didn't look close enough at incron I guess, thank you, I will read more on it [02:57] LeChacal2: the incron program would be running all the time, yes, but it would be _asleep_ until the kernel alerts it to a new or modified file. Then it would execute your script. If you get 10 new files per minute, this might take more disk and cpu, but significantly faster response time. if you go ten minutes between files, this will take one tenth the disk and cpu, and be _significantly_ better response time. [02:59] sarnold: this would just be a "backdoor" remote access system for excuting commends on a server so it would be like once in a bluemoon so not worried about the cpu/disk usage [03:00] LeChacal2: then absolutely incron :) [03:00] .. and be careful of patdk-lap's warning! executing things found in random files can be bad for your health [03:01] yes i got that, the script would something like, look for file named shutdown in folder than shutdown the server not look in this folder and run the commends in the file [03:29] Anyone use mutt? [03:32] pentagon: yes [03:33] sarnold: I have postfix going now how do I configure mutt to grab i's mail. [03:33] pentagon: is postfix doing local delivery? does it go to /var/spool/mail/pentagon? [03:35] for your .muttrc, you'll set spoolfile=/path/to/spool/file and then set folder=/path/to/other/folders -- if you're using procmail to split mail into folders before delivery === groupcat_ is now known as groupcat [03:52] sarnold: I am just trying to keep it simple. Is it ok to change sudo postconf -e 'home_mailbox = Maildir/' to sudo postconf -e 'home_mailbox = Mail/' [03:52] pentagon: probably; I don't know postfix as well as I should, though. [03:53] and you'll want to ignore my advice about spoolfile if you're using maildir :) I forget how that works too, but I seem to remember it was easier than I xpected. [03:56] sarnold: How do I initialise spoofiles or maildirs? Doesn't seem to say anything on the help page. [03:57] pentagon: spool file, just touch the file and check the permissions [03:57] Maildir, eh.... get procmail or postfix or mutt to do it? :) [03:58] I am looking for a .muttrc in the home but do not find it. sarnold [03:59] Where would the docs be stored? [03:59] pentagon: it won't be there by default; you'll need to create one to suit your needs [03:59] pentagon: man muttrc [04:27] Im interested in setting up a headless dedicated ubuntu server. As far as specs go, Im looking at 3.0 GHZ or higher, at least 4 GB of RAM and ideally 4TB of storage. I've been looking at my options with different computers. does anyone have a recommendation as to what kind of system will work well for my needs [04:31] UH [04:32] What do you mean what kind of system? Ubuntu Server is probably one of the more documented OS's [04:32] and probably the most supported. [04:33] Oh, you mean what hardware? [04:33] yes [04:33] Depends what you plan on doing with the server. You said 4tb so I'm going to assume a media server? [04:33] Or rather, file server? [04:33] yea [04:33] Ah. OK :) Well... [04:33] 3.0Ghz could be overkill. [04:34] ok thats not set in stone [04:34] If you aren't transcoding, then most of the IO would be Hard Drive related, and not CPU. [04:34] yea [04:35] My media server (Ubuntu Server, 2.4ghz, 4gb ram, 1 raid5, and 2 raid0's) runs top notch for a media server connected to a few XBMC clients. runs very little CPU. [04:35] I run transmission on it, but mostly used as a NAS. Very little CPU usage is going on. [04:35] It's all mostly hard drive usage that's happening, so as long as you get some decent disks, you should be fine on a barebones system. [04:36] 1gz or more, 2gb of ram or more and you should be perfectly fine. [04:36] anything i need to keep in mind as far as hardware goes [04:36] hey guys hows it going tonight? any apache guru's have some mod rewrite questions with magenta === n0ts_off is now known as n0ts [04:37] devslash: not really. make sure to get a gigabit ethernet card, and a gigabit ethernet router, and if you are using a switch, make sure its gigabit. home media centers will stall out and buffer a bit on 10/100 megabit connections. [04:38] how do i know if my router is gigabit [04:38] are you connected directly via a wire on your home computer? [04:38] as in, the computer your on right now. is it wired or wireless to the router? [04:38] i currently have an older laptop as my server [04:38] which is connected via an ethernet cable [04:38] I'm not on that computer now [04:38] the computer you are on RIGHT now. is it wired or wireless? [04:39] wireless [04:39] windows or linux? [04:39] linux (Ubuntu) [04:39] connect wired [04:39] if you can, [04:39] yea [04:40] like i said my existing server is connected via ethernet [04:40] ok [04:40] log into it via SSH [04:40] and run this command: ethtool eth0 | grep -i speed [04:40] devslash: if you install 'ethtool' you'll be able to use it to determine if your NIC has negotiated gigabit or not with your switch [04:40] if its 10 or 100, no good. if its 1000 your good. [04:41] shouldn't ubuntu have ethtool installed default? i don't ever remember running it on my server but i do have it. [04:41] also, i am going to be switching ISPs soon and will be using my apple router (yea Iknow they aren't the best) as my router with my cable modem [04:41] 100MB/s [04:41] but [04:42] thats with my current DSL modem. my router might be gigabit. [04:42] it wouldn't matter. it would only show the link speed between server->router [04:42] not server->modem. its the link speed. [04:42] yes [04:42] i'd suggest buying a gigabit switch or upgrading to gigabit router. [04:43] currently I'm using my dsl modems' router capability for wifi [04:43] ah ok. [04:43] but I'm switching to cable [04:43] anything on your network will only run at 100mb, even if its capable of gigabit. you'll get a max speed of about 12 megabytes per second. as with gigabit you'll hit 125 megabytes per second. [04:43] the iso I'm signing up with is giving me a modem that doesn't have wifi capabilities so i am going to switch to my airport extreme router [04:44] err isp [04:45] You could, if your feeling dangerous, is get a switch with jumbo frames support (or if your router supports jumbo frames), and turn them on with every device within your network. your speeds could theoretically increase quite a bit with large file transfers (aka movies). [04:45] i think thats beyond what i know how to do [04:46] devslash: k. well, make sure to get gigabit speeds for a media server. [04:46] otherwise you'll run into a lot of bottlenecking. [04:47] so if i connect to a gigabit router thats connected via ethernet to a plain old cable modem, would there be any kind of bottleneck between the modem and router ? [04:48] you wouldn't care; the modem <-> ISP link is going to be way slower than even 100 [04:52] maybe this is a dumb question but does gigabit require different ethernet cable [04:52] a bit [04:53] cat 5e is minimum required for gigE, but most people deployed that for 100 anyhow [04:54] devslash: it is because of the twists in the wire similar to pots same reason DSL gets a lot of line noise [04:54] oh i see [04:54] cat5e just has more twists so more copper more expensive [04:55] brb [04:56] sarnold: im getting RCPT rejected when trying to receive mail from the outside I take it this is because my hostname is set different locally than my isp assigns me, any suggestions? [04:56] pentagon: eh? I thought RCPT was used when you were _sending_ mail ... [04:57] it's been years since I've done SMTP by hand, but I thuoght it was MAIL TO: followed by RCPT FROM.... [04:57] i'm too tired to be giving smtp advice. :) sorry pentagon. [04:57] I assume it's short for recipient, when it is showing in the mail.log as rejected. [04:59] can you copy-paste the whole line? that'll mean more to someone... [04:59] (like I said, tired, maybe not me.) [05:03] AaronMickDee: hey [05:04] i plugged in my airport extreme and went into a utility app that tells you about the hardware. i noticed that for ethernet (I connected my laptop to my AE via ethernet cable) it shows Media Subtype:1000baseT. that means its gigabit right ? [05:06] is anyone that i was talking to a few minutes ago still here ? [05:06] devslash_: I am. [05:07] that is gigabit right ? [05:07] devslash_: yay :) [05:08] good [05:08] i wonder if its causing any bottlenecks with my existing server at 100MB/s [05:09] devslash_: so 100MB/s is 800Mbps, or flatlined gigabit ethernet. [05:10] ok... [05:10] vs gigabit that is 8000Mbps ? [05:11] 1000Mbps --> ~120 MBps [05:11] huh [05:11] it's a bit confusing :) [05:11] 100 is megabit per second; 1000 is megabyte per second [05:11] err [05:11] stupid [05:11] when i do ethtool eth0 | grep -i speed it return "100Mb/s" [05:11] like I said, sleepy :) [05:12] The average user has no use for gigabit. [05:12] that's 100 megabit [05:12] right [05:12] but for my purposes I'm running a streaming file server essentially [05:12] that'll get you roughly 12 megabytes per second traffic [05:12] Ever seen the movie matrix ? [05:12] yea [05:13] Well the more bandwidth the more resolution the matrix gets, so how will you know reality ? [05:13] I'm streaming music files so nothing especially big. i know that my slow internet speeds are 1 factor which I'm fixing by upgrading yo a faster internet plan [05:14] devslash_: and you'll probably love cable over your dsl. all dsl users I know aren't real happy. my cable _flies_. (Not like South Korea or Sweden fast, but good for the US fast.) [05:14] you know how fast my fucking dshell is ? 3MB/s [05:14] the new one is a bit faster at 10 [05:15] sarnold: how fast is your plan ? [05:16] See the bottleneck? [05:16] Gigabit router to run a 3MB/s connection. [05:16] so do you think my slow isp speeds are the biggest bottleneck ? [05:16] or the 100Mbps or both [05:17] devslash_: I routinely get 2MB/s downloads from ubuntu, torrents sometimes slightly more. [05:17] ok [05:17] but [05:17] sarnold: 1000Mbps -> 125MBps :P - caps matters > but you won't /get/ 125MBps most of the time, you generally max out 80% or so of the links nominal capacity [05:18] lifeless: in fact, I've never gotten more than 80MBps on this, I assume it's the cheap-ass switches we bought. :) [05:18] though it might also be the cheap-ass 5e we used. [05:18] since I'm running a file server serving music files (mp3s so most files are nominally small) do you think my isp or 100Mbps is causing a bottleneck [05:19] devslash_: depends what you're doing :) downloaded torrents off the internet, definitely the ISP. [05:19] no [05:19] most of what i use it for is streaming media files , music [05:19] devslash_: streaming four movies from your NAS to your xboxes, playstations, iphones, and laptops, that's all local. [05:19] i stream over the internet [05:19] ISP [05:20] thats what i thought. that should be fixed soon [05:20] once i get 10Mbps that not be an issue [05:20] that won't be a issue [05:24] Welcome to the matrix, enjoy your STAY. [05:24] whoa...... [05:25] Tomorrow verizon wireless cmos eyes will be swarming you like sentinels. [05:28] I had to beg my isp to give me a lower bandwidth. [05:42] Eh, its working but not as well as I would liek. [05:42] Seems I have to put in my hostname manually for destinations since I have a generic locqal hostname. [06:04] CrypticSquared: nice nick [06:06] um... thanks? [07:03] wtf does `grep -q` do? the man page says "do not write to STDOUt, exit immediately with zero status if the match is found" that would mean to me that if I don't find the string, return true. I'm trying to decode this bash code in my rc.local of a new VM I've purchased. http://paste.scsys.co.uk/208541 [07:04] it looks like if "some conditional" it dumps the SMBIOS that contains the VNC password and resets it for toor: [07:06] this conditional runs every time I reboot my VM since it's in /etc/rc.local... [07:07] /etc/issues says I can delete the toor user, and I don't think chpasswd works on non existent users, so I -think- it sets the toor password back to whatever matches the specific handle in the SMBIOS [07:22] We can remember it for you wholesale. [07:22] I figured out who the we is. [07:23] What is the difference just close you eyes and do whatever you want. [07:23] Can't tell the difference. [07:38] three18ti: "grep -q" returns true if it finds what it's looking for and false if not. [07:38] three18ti: Just like "grep" (without -q). It just doesn't output the matching lines. === smb` is now known as smb === mcclurmc_away is now known as mcclurmc === tuv0k is now known as darthanubis === rvba is now known as Guest78352 [08:11] New bug: #1057289 in nova (main) "Ubuntu cloud archive missing nova-novncproxy" [Undecided,New] https://launchpad.net/bugs/1057289 [08:12] New bug: #1057290 in nova (main) "Ubuntu cloud archive missing swift-plugin-s3" [Undecided,New] https://launchpad.net/bugs/1057290 [08:12] New bug: #1057303 in samba (main) "package samba-common 2:3.4.7~dfsg-1ubuntu3.10 failed to install/upgrade: le sous-processus script post-installation installé a retourné une erreur de sortie d'état 10" [Undecided,New] https://launchpad.net/bugs/1057303 [08:12] New bug: #1057347 in apache2 (main) "a2ensite suggest files which do not exist" [Undecided,New] https://launchpad.net/bugs/1057347 === mcclurmc_away is now known as mcclurmc [08:44] soren, cool thanks. === _rvba is now known as rvba === cpg is now known as cpg|away [10:38] <_ruben> what would be a nice place for a script to set irq affinities for nics? init script hooked to the net-device-up signal or so? === tuv0k is now known as darthanubis [10:49] _ruben: in /etc/networks/interfaces, 'up /run/this/script', perhaps? [10:50] <_ruben> guess that would do as well yeah [10:51] <_ruben> the pinning of nic queues to specific cores did fix the uneven distribution btw :) [10:52] nice [10:52] how did you do that? [10:53] I thought that is why /usr/sbin/irqbalance was made [10:53] <_ruben> patdk-lap: irqbalance didn't manage to create an equal balance, cpu0 was getting the majority of the workload [10:54] <_ruben> or perhaps it'd even out better under higher loads, who knows [10:54] <_ruben> with irqbalance cpu0 would be 6+% si, and the others around 1% si [10:54] that is a good thing :) [10:55] unless the cpu is too loaded, it's better to load up one core [10:55] and keep the other cores lower [10:55] then you can use turbo boost or whatever they call it [10:55] <_ruben> RoyK: i used the printf snippets from http://www.metaflows.com/technology/10-gbps-pf_ring-2/ as an example basically === n0ts is now known as n0ts_off [10:56] <_ruben> patdk-lap: well, the total (si) load dropped as well, even tho i was afraid the opposite would happen [10:57] _ruben: but patdk-lap's got a point. cpu cache is used far more efficiently if only one core is active [10:58] you may even save power by clocking down the others [10:58] <_ruben> i dont pay the power bill ;) [11:00] it's even good for the environment ;) [11:00] you will sleep better at night :) [11:00] <_ruben> once the load increases significantly on these boxes, i might revisit this stuff === n0ts_off is now known as n0ts [11:10] any nody here? :D [11:10] any body here? :D [11:10] <_ruben> no [11:10] !ask [11:10] Please don't ask to ask a question, simply ask the question (all on ONE line and in the channel, so that others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) See also !patience [11:10] hi guys, last night my server went down for an unknown reason, if it was a kernel panic, does it leave any trace? just to know if it was a kp or a power outage or similar [11:12] hi, my application with google maps API work fine on private IP, but the problem when accesed form public IP. must have a configuration on ubuntu server? [11:13] *from [11:16] tcp_cungkr1nk: when I was messing with gmaps apis a few years ago, your api key was tied to the hostname and url, I don't know if they still do that. [11:17] greppy i'm not set the hostname and url on my gmap api key [11:18] <_ruben> hmm .. seems irqbalance.org is down .. wondered if there was any details on if/how its behaviour depends on overal load [11:36] <_ruben> ah, ipv6+ssl appears to be broken for irqbalance.org [11:50] anyone can tell me how I can set mysql to restart automatically on server restart? :) [11:52] I have to restart mysqld manually on my vps everytime my provider updates their root server [11:56] 'service mysql start' to /etc/rc.local ? === gary_poster|away is now known as gary_poster === zz_Kiall is now known as Kiall [12:13] daniel_-: afaik mysql starts automaticaly [12:14] if not: sudo update-rc.d mysql defaults === sjerk22 is now known as ScriptJerk === matsubara-afk is now known as matsubara === Kiall is now known as zz_Kiall [12:39] jdrab: thx [12:41] my ubuntu server hangs on apt-get update -> "Unpacking replacement dpkg" is taking forever. any suggestions? [13:02] jibel, qatracker fix: [13:02] http://paste.ubuntu.com/1230366/ [13:02] just a logic error if --append is not given. [13:34] theres a config file for port allowance right? [13:42] what would everyone recommend for DNS Server? [13:47] * andygraybeal awaits a response too :) [13:48] what was your question andygraybeal? [13:48] briggz, i'm awaiting a r3esponse to your question also ;) [13:48] about the dns :) [13:49] lol [13:49] have you looked at bind? [13:50] I'm setting up a server for a intro to linux class [13:50] not really, ive heard of it, but also heard it's complicated to setup [13:50] yea, i need a lifevest. [13:51] I was going to go with that originally but I thought other people might have other ideas of fairly decent secure alturnatives. [13:51] is that what your currently playing around with? === zz_Kiall is now known as Kiall [14:01] well.... i cheat and i use dns interface from openhosting and rackspace ... plus i've been messing with zentyal [14:01] hi all [14:01] briggz, --* [14:02] errr... --^ [14:02] is it possible to recompile the 12.04 kernel in ec2? [14:02] there is no ec2 branch on the precise kernel as there is on lucid [14:23] smoser: hey, can you make sure https://wiki.ubuntu.com/QuantalQuetzal/TechnicalOverview/Beta2 is good please? [14:24] k. reading. [14:24] add anything awesome. :) [14:24] ta [14:32] smoser: Do you have the script for updating the iso tracker from jenkins testing? [14:33] i use this http://bazaar.launchpad.net/~jibel/+junk/qatracker/ [14:33] from lp:~smoser/+junk/jenkins2isotracker/ [14:33] thanks [14:33] (ec2 is alrady done now. and utlemming has set it up to be more automatic on that too) [14:34] note, there is a bug in qatracker [14:34] basically you have to pass --append [14:34] jibel, ^ [14:34] fix bug with [14:34] http://paste.ubuntu.com/1230366/ [14:35] hi how can I recompile precise kernel on ec2? [14:35] smoser: if you have it already setup, can you run it for the iso tests? [14:35] smoser: i'll discuss with jamespage about trying to automate this [14:36] where are results, daviey? [14:37] smoser: https://jenkins.qa.ubuntu.com/view/Quantal/view/ISO%20Testing%20Dashboard/ ? [14:38] hm.. ok [14:48] Daviey, i can't automate this at all, but i'll see if can't one time populate [14:49] smoser: Why can't it be automated ? [14:50] i'm not doing it now [14:50] oh sure, i get that! [14:50] :) [14:50] I mean, going forwards.. i wondered if jenkins could JFDI? === dax_roc_ is now known as dax_roc [15:01] New bug: #1057573 in socat (universe) "Socat depends on libreadline5 and fails to link properly." [Undecided,New] https://launchpad.net/bugs/1057573 === n0ts is now known as n0ts_off [15:11] smoser, I pushed the script to lp:~ubuntu-qa-website-devel/ubuntu-qa-website/qatracker-tools [15:11] why the hell are we still subscribed to socat? [15:11] smoser, feel free to submit LPs [15:11] MPs [15:12] adam_g: ping awake yet? [15:13] jibel, you can't submit MP on junk branch [15:17] Daviey: maybe you'd like to talk with website people about server/cloud pages for 12.10 and whether you want to push lts/cloud-archive/quantal and in which combinations. [15:18] Daviey: also I gave them heads up that there probably will be just "ubuntu server" (64 bit) and not "32 bit / 64 bit" variants. [15:21] Hi everyone, I just set up Ubuntu server and installed vsftpd on the system. I am trying to create user accounts on the system with no access to the shell but it seems when I attempt this they lose the ability to connect to FTP as well. Here is the command I researched/built: useradd -d /home/user1 -s /sbin/nologin -G ftp user1 [15:21] xnox, unless you want to point to arm server too from there :) [15:21] Any ideas? [15:22] ogra_: hmm... http://www.ubuntu.com/download/desktop doesn't have any links to ARM, but the server one does... and it goes to "arm for desktop & server" [15:22] =/ [15:23] yeah, desktop is only a reference image while server is an actual product for arm [15:25] smoser, you can propose a merge against ~ubuntu-qa-website-devel/ubuntu-qa-website/qatracker-tools isn't it ? it is not a junk branch [15:25] http://bazaar.launchpad.net/~jibel/+junk/qatracker/ [15:26] i was just using that [15:26] i didn't know there was another. [15:26] smoser, ah,ok, it's obsolete now, I moved everything to a more 'official' branch [15:26] [17:11] smoser, I pushed the script to lp:~ubuntu-qa-website-devel/ubuntu-qa-website/qatracker-tools [15:27] modulo utc+2 [15:27] xnox: urgently? [15:28] Daviey: between now and release day. [15:28] Daviey: I think beta2 is more urgent right now =) [15:28] Daviey: could you look at the upload of maas-enlist i just made pleasE? [15:28] roaksoax: for b2? [15:28] Daviey: skaet is waiting on you for b2. [15:29] xnox: yeah [15:29] Daviey, yeah, we're missing manditory server testing results and its critical path for releaseing the images [15:30] ugh [15:30] smoser: Did you post everything which jenkins covered? [15:31] Daviey: no not necessary for b2 [15:31] :) [15:31] thank goodness :) [15:34] Daviey, no. i'm working on it. [15:34] its aPITA === n0ts_off is now known as n0ts [15:39] smoser: ugh [15:39] sorry, i thought it would be easy [15:39] i was trying to at least model [15:39] what data we'd need and such [15:39] skaet: we do have 100% mandatory [15:44] Daviey, http://pb.daviey.com/jGFS/ [15:44] thats the maaping i have so far [15:44] any idea where I can get the ec2 12.04 kernel source? [15:45] kontinuity, its just the -virtual kernel (which is just the -server kernel) [15:45] so your question is generalized to [15:45] "where can i get ubuntu kernel source" [15:45] Daviey, amd64+mac is missing 3 manditory results. [15:46] smoser: thanks! I am trying to compile the ec2 kernel, any pointers? I am following this guide for now http://wiki.freeswitch.org/wiki/Amazon_EC2#Updating_Kernel_Timer_to_1000HZ [15:46] http://iso.qa.ubuntu.com/qatracker/milestones/238/builds/24189/testcases [15:47] well, 'git checkout --traac -b ec2' is no longer valid [15:47] https://help.ubuntu.com/community/Kernel/Compile [15:48] but as you mentioned the ec2 branch is no longer valid for Precise [15:48] so can I build from the main branch [15:48] ? [15:49] Hey! I just remove my swap partition to resize another, then I recreated another patition for the swap but what do i have to do to tell my server where is my new swap partition? [15:50] AlexO: mkswap on the partition and add it to /etc/fstab [15:50] and remove the old one [15:50] smoser: any clues? [15:51] kontinuity, yes. its just the '-server' kernel. so follow any instructions you see for that. [15:52] smoser: awesome thanks [15:53] RoyK: Thanks ! Do I need to reboot? [15:53] AlexO: no, just swapon -a [15:54] RoyK: Thanks again working just fine :) [15:54] how can i force apt-get to resolve the dependencies and install the deps for a given package? [15:54] its been being stupid for a while now, so... [15:54] AlexO: I'd recommend using the swap partition's UUID in /etc/fstab - use blkid to find it [15:55] RoyK: When I did mkswao /dev/sda5 It return me the UUID [15:55] AlexO: also, run swapon -s to make sure the old partition isn't hanging around, if it is, you may have to rebot [15:55] AlexO: ah, ok [15:56] AlexO: or at least, swapoff that one [15:56] RoyK: Ok Thanks :) [15:57] AlexO: did you swapoff the old one before you repartitioned? [15:57] RoyK: Nope I used gparted(iso) to remove it and when I do a swapon -s I only have one swap partition listed [15:58] ok [15:58] if you removed the partition while running swap on it, it might have been a slight problem... [15:58] Yeah I guess ^ [16:06] http://dalkescientific.com/writings/diary/archive/2007/06/01/lolpython.html [16:09] zul: ping on sponsoring bug 1054306 ? (debdiff for augeas bugfix) [16:09] Launchpad bug 1054306 in netcf "ncftool list fails" [High,Invalid] https://launchpad.net/bugs/1054306 [16:12] Daviey, is someone working on those last manditory test cases for amd64+mac or do you consider it good enough as is to ship? [16:18] skaet: amd64+mac not a released item [16:18] :) [16:19] Daviey, can you either add it to the community set, or remove it from the daily builds then. Prefer to save testing focus for those that are then. [16:19] ok, marking it as ready. [16:24] Daviey, https://wiki.ubuntu.com/QuantalQuetzal/TechnicalOverview/Beta2 <-- have you finished making your edits? [16:30] hallyn: sure i can get it in a little bit [16:30] zul: thanks [16:30] stgraber: would there be any chance of getting augeas added to the server package list? [16:31] hallyn: need to upload openstack final first though [16:31] zul: np, i'ts not that urgent, long as it makes release :) === matsubara is now known as matsubara-lunch [16:34] hallyn: I think the problem is that for some reason libvirt isn't in the ubuntu-server packageset, if it was, augeas would be pulled in automatically [16:34] stgraber: d'oh === n0ts is now known as n0ts_off [16:49] I have "d-i tasksel/first multiselect edubuntu-desktop" in my preseed file, can someone give me some pointers as to why it isn't installing edubuntu-desktop? [16:55] rjune: hmm, that's how I have it in my preseed notes, perhaps something changed in d-i recently [16:55] rjune: which ubuntu version is that? [16:55] I'm trying to install precise [16:56] hmm, that's how it's done in https://help.ubuntu.com/12.04/installation-guide/example-preseed.txt too [16:56] *sigh* [16:57] stgraber: happen to have any insights on why that could happen? ^^^ [16:57] highvoltage, are you doing network installs? [16:58] highvoltage: do you have universe and multiverse enabled? [16:59] can you guys see what I type? [16:59] highvoltage: oh, and the task is edubuntu-desktop-gnome btw :) [16:59] briggz: pong [16:59] http://www.stgraber.org/download/preseeds/weblive-edubuntu.seed is a working seed to install edubuntu (you probably want to change the password and drop the late_command though) [17:00] I joined another channel, it tells me it can't send to channel when I try to type a question [17:00]