[00:00] Hi. I'm a support channel virus. please copy me to each support channel to help me spread over the whole freenode network. [04:23] how to do some thing that automatically checks wheather a proccess has not been runing for a specified time. if it isnt, runs it and logs in a file that it has made it run.? [04:35] working on a simple dhcp server my server ip is 213.137.73.4 and when I try to stop my dhcp server I get "address range 213.137.73.2 to 213.137.73.253 not on net 213.137.73.0/255.255.255.0!" [05:56] Is there something like the ubuntu desktop version of update-manager that I can run on my server? (of course apt-get update && apt-get upgrade --download-only) in a cronjob is available, but that's not what i mean.) [05:59] for one thing, update-manager knows how to only install updates from gutsy-security for example [05:59] (automatically without confirmation from security while other updates are left for admin action) [06:06] spowers: create a seperate /etc/apt/sources.list file such as /etc/apt/sources.list.security and use "apt-get -c /etc/apt/sources.list.security upgrade" [06:06] ooh, good idea [06:07] where sources.list.security contains only the repo's you want, such as security upgrades [06:07] i must admit i'm a little skeptical on the idea of automatically installing any updates, but i understand that the security team is very, very, very, very picky [06:07] very. [06:07] so it may well be safe to stick in cron [06:08] yes, a significant amount of attention is paid regarding security updates [06:08] that doesn't mean things don't go wrong on a minor level [06:08] I've seen updates, which get rolled back the next day, then rolled forward again the following day [06:08] but that is unusual and definitely not the norm [06:09] ok, since it seems like you're on a roll here, i have another question [06:10] i used to use just apt-get all the time, and sometimes aptitude when i wanted to get a more interactive view of packages on my system [06:10] yes, and? [06:11] but i'm getting ready to start deploying maybe half a dozen ubuntu server systems at work, and i'm thinking it would be a great idea if we could somehow update them all at once, or manage packages on them all at once [06:11] is there anything that's like, "synaptic for multiple servers" or something like that? [06:11] or is there another way of looking at the problem [06:12] i could easily um, clusterssh -X and run synaptic on them, as i'm not particularly averse to installing xlibs and even all the way up to gnome libs to get that functionality [06:13] https://help.ubuntu.com/community/Apt-Cacher-Server [06:13] that would be one option [06:13] hmm [06:14] while that is generally more designed for having the repositories local [06:14] you could probably configure it specifically for your needs for specific packages [06:15] again, by perhaps another cron script to point to that apt repository [06:15] oh, so it'd be a model where i'd have a local apt repo that my servers trusted [06:15] and i'd just drop the packages i wanted to force down the line into that repo [06:16] that's another good idea, but what if the packages have debconf questions [06:16] it would require a bit of tweaking but yes, I believe that would be quite possible [06:16] spowers: as I havent done it myself, I can only suggest it, the devil is always in the details ;) [06:16] what i'm really after is a kind of central admin console for a large group of servers.. and 6 is not so large, but i could see it being an itch i'd want to scratch down the line [06:17] especially if 6 turns into 60, as we're doing all virtual servers rather than physical boxes now [06:17] i'm being asked to deploy apps every week now, where it used to be a monthly thing :) [06:17] again, thats where using something like clusterssh would be of benefit [06:17] hmm [06:18] yeah, i've only used it in passing, but it seems like maybe investing some time in cssh would be all i needed [06:18] * faulkes- nods [06:18] heck, i'd probably be happy with that [06:18] i'd be a bit happier if they'd update it so it didn't look like 1988 [06:18] but otherwise, you know, it's great [06:19] well, it's unix (perse), all things are possible with a litle work [06:21] it's weird. i used to be all debian, and spent a lot of time doing debian admin stuff, and all was well.. then i took a job at a windows 2003 shop, and at work i quit doing so much debian, but things have changed at the shop and i've convinced the powers that be to put some ubuntu/debian stuff in production, but all the windows admin work has made me look at things from a different perspective [06:22] which i suppose is the windows weenie perspective [06:22] i get the impression that the ubuntu folks have done a lot to improve on debian as a server while i wasn't looking for the past few years [06:22] well, you work with what you are given [06:23] as for windows, I generally install cygwin and do what needs doing that way [06:23] or use that stupid shell thingy they have [06:24] one thing that i've found invaluable to me at work.. well, two things, the other one being rdesktop, and the first one being a project called smbnetfs [06:24] we have win2k3 servers and winxp desktops [06:24] smbnetfs places the "network neighborhood" on a fuse mountpoint [06:24] * faulkes- nods [06:24] * faulkes- nods [06:24] the configuration is quite odd, but what an effect [06:25] cat /dev/null > /smb/box/c$/autoexec.bat [06:32] apt-get -o Dir::Etc::SourceList=/etc/apt/security-sources.list upgrade [06:32] -c complained about garbage in the file (huh...) [06:35] either way ;) [06:37] it's a good idea, and i'm going to try to use it to destroy my server at home [06:37] (in an unattended fashion) [06:39] thanks for the advice [06:43] always good to try it out in a test environment first ;) [06:58] Bambi_BOFH eh? sounds to me like somebody is the PFY [06:59] faulkes-: that makes two comments in that vein in about 3 minutes :) [07:03] heh [07:03] yes well, when you spent enough time in a.s.r ;) [07:03] hehe [07:06] apt is all screwed up, how do I fix it [07:06] dpkg: error processing /var/cache/apt/archives/debianutils_2.22.1_i386.deb (--unpack): [07:06] trying to overwrite `/usr/sbin/add-shell', which is also in package passwd [07:43] I'm told that using tasksel's lamp-server option does setup & configuration beyond what installing the packages separately does. I've looked at tasksel's lamp-server task and it only seems to install apache2 and mysql-server. not even php5. for starters, does it install php? [07:45] iirc there was an issue about that but I can't remember what it was offhand [07:45] I would suggest searching launchpad [07:46] iirc installing php5 should add itself appropriately to your apache installation (although requires a restart of apache) [07:47] ahh, I was misunderstanding tasksel's 'task-fields' option, which explains the lack of php5. So my question is more, does tasksel do anything beyond if I were to install the packages separately (I understand how they'd be integrated if I installed them separately) [07:47] <_ruben> the lamp task installs apache+mysql+php just fine .. there were some issues with tasksel itself tho (hanging at end of install) .. using apt-get install lamp-server^ 'fixes' that [07:47] my friend says the Ubuntu team worked for 6 months on the lamp-server and believes it does a load of specific extra configuration [07:48] I'm not a tasksel expert, I couldn't say [07:48] thanks [07:48] iirc I think zul does work on php stuff [07:48] can anyone here provide me with a list of the dependencies provided by the virtual package 'lamp-server' please? (I'm not running Ubuntu) [07:49] virtual packages aren't listed at packages.ubuntu.com [07:51] Morning :) [07:53] thanks anyway [08:02] <_ruben> here's a... ow wait .. he left [08:14] good morning [08:16] morning [08:20] hmm, I should probably crawl into bed [08:21] nijaba: anything poll related we need to discuss before I head off into lala land? [08:22] faulkes-: just that I talked with limesurvey people [08:22] nijaba: oh, heh, I know, on the export, where we list the available distro's, ubuntu is listed last, I think we should list it first [08:22] they commited on a bugfix for monday :) [08:22] faulkes-: I'll fix that... (alphabetical order) [08:23] awesome on the response from limesurvey [08:23] faulkes-: in exchange, I am working on packaging limesurvey to ubuntu [08:23] nijaba: yeah, I thought it would be something like that [08:23] re: alphabetical [08:24] nijaba: hmmm, interesting, maybe I could work with you on that, or at least follow along (packaging it) [08:24] faulkes-: it will be my first as well... [08:24] I've been reading through the docs and such but I'm a newbie when it comes to it [08:24] faulkes-: just followed packaging 101 this week! [08:25] * faulkes- nods [08:26] ok, I'm gonna jump into bed here, I'll catch up with you later today at some point I imagine [08:26] hello..I am using ltsp across two servers with dhcp failover. My ldap auth server that uses libnss locks up on boot at klog. Any fixes out there for this? [08:27] good night faulkes- [08:27] achandrashekar: what version of the server? 7.10? [08:28] yes [08:28] 7.10 [08:28] i know there is bug report out there.. [08:28] achandrashekar: you should ask mathiaz or dendrobates when they are around, but right now I am afraid they are asleep [08:29] they are around in the morning.ermmm PST? [08:29] achandrashekar: they are both EST based [08:30] okay..will do. [08:30] its plague since fiesty..and i havent been able to search or walk my way past it....yet [08:31] I bet [08:32] tried ALL the recommended - switch S10 to S18 rc files, soft_boot yes, check /etc files for presence of klog and nvram...you name it...but no luck at all. [08:32] the only thing that works is to remove ldap from nsswitch and then im good [08:32] but then...of course no ldap [08:32] auth === cjwatson_ is now known as cjwatson [09:03] nijaba, faulkes-: w.r.t. ordering of choices in polls.. Being at the bottom is not necessarily bad. [09:03] nijaba, faulkes-: However, I find randomisation to be the best ordering mechanism for poll options. [09:03] soren: I was thinking about it as it is one of the options in LimeSurvey [09:04] Lots of folks click randomly, and randomising the options evens that out. [09:04] http://slashdot.org/pollBooth.pl?qid=740&aid=-1 [09:04] That illustrates it pretty well. [09:04] People *think* they click randomly. [09:04] They don't. [09:05] Randomising the options will thwart those people's attempts at messing with the poll. [09:06] The slashdot poll is slightly special, though. [09:06] The "real" cowboyneal options is always at the bottom, so I'm quite sure that's why most people favoured that one. [09:08] It's tricky if begin asked "how many foo have you got" and the options are "0-10", "10-20", ....., ">100"... If you randomise those choices, you screw with people's heads :) [09:08] soren: should only be done on text question, IMHO [09:09] Sounds like a reasonable, general rule :) [09:09] I read a study at one point about this. [09:11] I believe it recommended adding a bogus option to the list, and deduct the number of votes that option got from all the other vote counts as it was a reasonable assumption that that was the amount of bogus votes each of the other ones got, too. [09:12] soren: that might be a bit too much... [09:12] soren: would you be able to mentor faulkes- & I on packaging LimeSurvey? [09:13] nijaba: It's a php web app, isn't it? [09:13] soren: yep [09:14] soren: so it should be pretty simple [09:14] I've never done one of those, but sure. [09:14] The tricky part is database configuratino. [09:14] configuration, even. [09:14] soren: ah... right [09:48] moin [10:24] Hiya .. I'm trying to get a pretty old box fired up with Ubuntu-Server and it is going off about: No installable kernel was found in the defined APT sources. It is a Cyrix chipped box .. help? [11:00] hi all [11:00] I need to know how to setup CGI on Ubuntu 7.10 [13:00] Holla! [13:00] when i try to run this simple install script: http://pastebin.ca/912520 [13:00] i run into an apt error saying that apache is not configured yet. [13:01] due to my exit codes the installation aborts. [13:01] Any idea why or what i am doing wrong? [13:02] morning [13:03] <_ruben> spiekey: apache isnt valid afaik .. try apache2 [13:04] <_ruben> (and you have ssh specified on 2 different lines) [13:11] _ruben: at first i just used phpldapadmin and let the deps do the work, but then it came up with "apache is not configured yet"...i will reproduce it in a minute. [13:28] jjesse: I'm sorry to here you can't make it to UDS. [13:32] here...how can i avoid this error? http://pastebin.ca/912552 [13:33] dendrobates: too close to baby coming [13:33] my wife is due w/ first child to close to uds this time ;( [13:33] not :( that we are having a baby, but that i'm missing uds [13:34] jjesse: understood. next time perhaps. Although a six-month old is a bit of work, so it may be some time before we see you again. :) [13:35] <_ruben> spiekey: ah .. its on dapper .. no experience with that .. only gutsy here (and a test install of hardy) which does not have apache, only apache2 [13:35] jjesse: heh kids can be fun.. [13:36] _ruben: thanks anyway [13:46] dendrobates: i understand... six months hopefully :) [14:14] dendrobates: nice forum response [14:21] hi [14:23] I am running gutsy and I want to play a bit with linux-vserver . Do I have to patch and compile my kernel, or does any of the distro builds contain the vserver patches? [14:24] yes you have to compile your own patch [14:25] darn [14:27] i have got samba + ldap running, and if i reset my dapper box my MachineName$ get removed from my ldap database [14:27] any idea why? [14:28] or who? [14:31] <_ruben> linux-vserver as in LVS or some virtualization thingie? [14:47] faulkes-: thanks. [14:48] redguy_work: I believe that a pre-patched vServer kernel has been made available by RevolutionLinux in Universe for Hardy. [14:48] _ruben: virtualization thingy [14:48] nijaba: RevolutionLinux? [14:48] zul: a service provider in Canada [14:49] nijaba: ahhhh....those damn canadians [14:49] zul: damn you, then ;) [14:49] hey now [14:49] nijaba: heh [14:49] but they are much better than the Maudits Français ;) [14:49] don't me start up with the timbits and saying "eh" [14:50] s/don't me/don't make me [14:50] I'll send the mounties after ya [14:51] faulkes-: you're from Canada as well? [14:53] yes [15:02] elmo: I just saw the addition of a bunch of Dell HW by AndrewGlen-Young \o/. Should I had them the other wiki page as well? [15:03] mathiaz, soren: this includes the iSCSI NAS MD3000i ! [15:03] I suspect you're on the wrong channel, dude. [15:03] :) [15:03] who does bind9 for ubuntu? [15:04] lamont [15:04] woops... [15:04] hah hah! [15:04] zul: why? [15:10] how can i search for backport packages again? [15:11] lamont: https://bugs.edge.launchpad.net/ubuntu/+source/bind9/+bug/191685 [15:11] Launchpad bug 191685 in bind9 "[hardy]: bind9 does not remove /var/lib/bind on purge" [Undecided,New] [15:11] zul: meh [15:12] and yeah, iz bug [15:12] lamont: if you want I can do a debdiff and put it somewhere [15:12] it's more that the mkdir in postinst doesn't say -p [15:13] lamont: that was my thinking when I saw the bug [15:14] which is fixed in the next upload. [15:14] okies [15:15] the not-purging-configish-stuff-on-purge is a hold over from bind vs bind9 [15:15] it pisses off admins when switching from bind 8 to bind 9 nukes all their zone data [15:16] I'll upload that tonight === cjwatson_ is now known as cjwatson [17:39] zul: I think I've managed to setup ubuntu-server-bugs correctly [17:39] mathiaz: yay! [17:40] zul: let's see if it works as expected. [17:40] hopefully.. [19:14] man I'm really on the fence about the server guide being installed by default bug [19:14] do you not want it installed? [19:15] on the one hand I don't think it's a big deal to have it installed because personally I look to the web first [19:15] but from some of the comments in the bug others may look to local copies first [19:15] i thought there was great discussion at uds that it was a problem it wasn't included? [19:16] sommer: think of the case where you are building a server and you don't yet have network connectivity [19:16] having a local copy of the guide may be useful [19:16] faulkes-: right, the package is on the cd though, just not installed by default [19:16] or you are not allowed to have internet access [19:16] the other part of the bug was about not having a non-gui doc package, which has been solved [19:17] sommer: I think of it this way, the server edition is in fact, one very large package which is installed [19:17] and if you look at it like that, almost all other packages install documentation (/usr/share/doc type area) by default [19:18] granted, thats just my opinion [19:18] faulkes-: heh, I'm having a hard time visualizing that one [19:19] I'm good either way, and can see both sides of the argument [19:19] sommer: ok, if you "aptitude search " the listing on the left sometimes has a v, indicating it is a virtual package, a meta [19:19] think of the entire server edition as one big meta ;) [19:20] but decision is yours, I've never found it to be a problem [19:20] I guess for me the deciding factor is it worth the amount of work needed to install it by default, and since I'm not the one doing the work... [19:21] I'll be happy if the new docs are on the web site at launch day ;-) [19:21] aye [19:23] mathiaz: are you the one who'd update the seed, or whatever is needed? [19:29] also if the guide isn't installed by default do you solve the motd bug by giving instructions on how to install and view the guide? [19:29] seems like that could make for a long motd... heh [19:32] heh [19:32] cat server-guide.txt >/etc/motd [19:32] that'll teach'em [19:32] heh, longest motd ever [19:33] and just in case they decide to do a network install and ssh in [19:33] cat server-guide.txt >/etc/issue.net [19:40] faulkes-, jjesse: thanks I think I'm off the fence [19:40] glad i could help [19:40] or at least have a solid opinion :) [19:49] mathiaz: ping [19:49] mathiaz: CC Meeting is starting in 10 minutres [20:03] arg [20:03] not installing manpages is a cruel joke =( [20:06] pwnguin: If we aren't installing man pages it's a bug. [20:06] "man: command not found" [20:07] i have to assume it's intentional [20:11] Hi! I've just installed ubuntu-server and I'd like to test it a bit. First problem! I can't edit files in /etc/. I'd like to set up network, IP and such. It appears I'm not in sudoers group. [20:11] How do I get standard linux root user. I miss it. :) [20:11] !info root [20:11] Package root does not exist in gutsy [20:11] !root [20:11] Do not try to guess the root password, that is impossible. Instead, realise the truth... there is no root password. Then you will see that it is 'sudo' that grants you access and not the root password. Look at https://help.ubuntu.com/community/RootSudo [20:12] you can boot into recovery mode and add yourself to sudo [20:12] why wasn't this done at setup? [20:12] it is [20:12] at least I havent encountered a case where it wasn't [20:13] this box is just installed. And I've added user at setup. And it's not in sudoers. [20:13] 7.10 [20:13] I believe during setup, the there is a specific option to add yourself to sudo [20:13] as I said, boot into recovery mode and add yourself, that would be the easiest fix [20:14] Is there a way to do this without reboot? [20:19] how lame ... system is useless [20:19] reinstall is needed [20:19] I fail to see how doing a simple reboot is such a problem [20:20] faulkes-: and what do I do after reboot? [20:20] you select recovery mode, which will bring you to a root prompt [20:21] faulkes-: default grub menu doesn't have selection or does it? [20:21] tiny: if you hit esc [20:21] ah k [20:21] a grub menu should appear [20:21] thanks will try [20:21] then add yourself to the admin group [20:21] strange though [20:21] i thought it added the new user to sudoers by default [20:26] It did at one box at work ... the only difference in install method was that I installed that box over PXE [20:26] and now I need to check the syntax of sudoers [20:26] you should only need to add your account to the admin group [20:27] cat /etc/group | grep admin [20:27] install process should have done that :) [20:27] admin:x:: [20:28] there are possible reasons why it didn't [20:28] man sudoers [20:28] however, getting upset about it won't solve the problem [20:29] I know ... I miss the good'ol root account [20:29] simply edit /etc/group, add yourself to the admin group, you should be set [20:29] you can still have it [20:29] sudo su [20:29] faulkes-: k I will, ty [20:29] problem solved. [20:29] man sudo_root [20:31] sudo is lame and reasons for it are not justified [20:31] that is your opinion [20:31] yes [20:32] I don't have an admin group [20:32] weird eh [20:32] im starting to doubt you installed ubuntu :P [20:32] uname -a [20:32] one sec [20:32] please ... I'm not such a noob [20:33] cuz this seems like a hard one to screw up, and ive been trying lately [20:33] the last line of /etc/sudoers should read %admin ALL=(ALL) ALL [20:33] i managed to get the installer to lock up on the partition manager [20:33] if you don't have an admin group, simply edit /etc/group andcreate one [20:33] or use the tools written for the express purpose of doing this for you [20:33] faulkes-: you we're talking about /etc/group earlier [20:34] yes, I was [20:34] both need to exist [20:34] /etc/sudoers tells sudo which groups can sudo [20:34] the line %admin ALL=(ALL) ALL in /etc/sudoers says anyone in admin group can sudo [20:35] therefore, if you do not have an admin group in /etc/group you must create one and add yourself to it [20:36] This distro is far from polished .. I was under wrong impression. [20:36] still waiting on that uname [20:36] stuff like that should be auto set at install [20:36] it is [20:36] i just installed ubuntu server amd64 twice over ssh [20:36] both have an admin group and sudoers [20:37] pwnguin: I need to walk from one pc to another ... I can't ssh into box because it's in rescue mode [20:37] tiny: no? invoke-rc.d ssh start? [20:38] /sbin/ifconfig eth0 1.2.3.4 ; /etc/init.d/ssh start [20:39] obviously replacing 1.2.3.4 with the ip address of the server [20:41] ok ... I've "fixed" stuff. Added a line: username ALL=(ALL) AL [20:41] to /etc/sudoers [20:42] pwnguin: Linux orion 2.6.22-14-server #1 SMP Sun Oct 14 23:34:23 GMT 2007 i686 GNU/Linux [21:33] * faulkes- joys [21:34] fusesmb fun [21:34] and from what I've been able to attend of UDW so far, I've really enjoyed myself [21:35] good week so far, now if only the purple elephants I keep seeing stop trying to feed me twinkies === `6og is now known as Kamping_Kaiser [21:44] is there a sources.list somewhere that holds the usual suspects? somehow all i got was a cd src line [21:46] you mean fro /etc/apt/sources.list? [21:46] yes [21:46] hmmm [21:46] iirc wasn't there a sources generator dealie listed somewhere on the wiki? [21:52] Anyone know how I get g++ with the "-m32" option on Ubuntu Server 64-bit to use the /usr/lib32/stdc++ library instead of /usr/lib/gcc/x86_64-linux-gnu/4.1.3/libstdc++ during compilation? GCC uses the proper 32 bit libraries and everything runs fine, but g++ is dieing on this one library for some reason... [21:54] is this the right channel for this type of question? Or should I ask in the developer channel maybe? [22:38] thebusby: Another channel is more likely to give you quick answers. [22:38] thebusby: It's not server specific at all. [22:42] soren: I still have no luck with ubuntu-vm-builder [22:43] soren: It creates a root file that is 124k [22:43] mathiaz: I'll upload a new version before I go to bed. [22:43] mathiaz: And a newer version on Monday, probably. I hope to spend some time cleaning it up over the weekend. [22:44] soren: thank you! [22:44] soren: great. Thanks ! [22:49] soren: what do you think about bug 193531 ? [22:49] Launchpad bug 193531 in libvirt "pxe booting not supported" [Undecided,New] https://launchpad.net/bugs/193531 [22:51] faulkes-: following up on a few discussions, I've made quite a few edits to the survey. [22:51] mathiaz: Hm... 8139 ought to have the PXE boot rom. [22:56] soren: so how do I set the domain to use the pxe boot rom ? [22:57] libvirt should be doing that for you. Hang on. === jmtt_ is now known as jmtt [23:07] mathiaz: Er.. I actually explicitly remove them. [23:07] We don't have the source, so we can't ship it. [23:07] mathiaz: I'll look into it. [23:07] soren: ok. [23:08] soren: do you think this will be fixed for hardy ? [23:09] mathiaz: I'll know in a few minutes :) [23:13] mathiaz: Can you easily test it? [23:14] soren: how come ? [23:15] soren: what would I need to do ? [23:17] mathiaz: Install the etherboot package, unpack /usr/share/etherboot/rtl8139.pxe.gz as /usr/share/kvm/pxe-rtl8139.bin [23:17] mathiaz: And let me know if it works. [23:17] soren: I can do that. === jmtt_ is now known as jmtt [23:21] mathiaz: hi [23:21] mathiaz: sqlite is not a build-dep, it's just dep for bacula-director-sqlite, which we could drop, and leave only sqlite3, if it gets into main [23:22] ivoks: right. [23:22] ivoks: we could actually leave it in universe [23:22] soren: I'm not sure if it works. [23:22] sqlite? [23:23] soren: I'm having a hard time figuring out the boot sequence [23:23] sqlite stayes in universe, and sqlite3 moves to main; if i understand it correctly [23:23] soren: I don't think it works. [23:23] soren: at least creating/defining a domain works [23:23] ivoks: correct. [23:23] yay! [23:23] ivoks: so we can leave bacula-*-sqlite in universe [23:24] right [23:24] mathiaz: Hm... It might actually be rtl8139.zrom.gz you want to unpack and drop in there. [23:24] ivoks: and move bacula-*-sqlite3 in main [23:24] i'll check if sqlite3 could even conflict/replace sqlite [23:25] so, if someone had sqlite, he could now upgrade to sqlite3 [23:25] and drop sqlite version; if that's not possible, i'll just leave sqlite in universe [23:26] soren: ok - it boots from the network [23:26] soren: but can get an ip adress [23:26] soren: s/can/can not/ [23:27] mathiaz: Do you see anything on your dhcp server? [23:28] mathiaz: Are you bridging? [23:28] soren: hum - it works actually [23:28] \o/ [23:28] soren: if I cancel the pxe boot and ask a second time to boot it works [23:28] soren: it gets an ip adress. [23:28] soren: but not the first time. [23:28] *shrug* [23:29] Do you think it's an important use case? We don't support it right now (as etherboot is in universe) [23:29] soren: well - I'll use it to do automated test install [23:30] soren: it's usefull when you have lot's of vm and want to control where they boot from. [23:30] soren: but I'm not sure it's such a common use case. [23:31] We can add a README.PXEboot and tell people how to make it work. [23:31] soren: yes - I think that's enough. [23:32] Cool. Will do. [23:33] mathiaz: iirc there is a reason for the behaviour you're seeing w/ pxeboot dhcp / reboot [23:33] I don't remember it offhand but I do recall something about it [23:33] i'll let you know once my brain defuddles [23:34] faulkes-: well - I see a question if I wanna boot from the network or not [23:34] faulkes-: but it's strange that the first run doesn't work and subsequent one work [23:34] faulkes-: anyway - it does what I want :) [23:34] * faulkes- nods [23:35] I just remember encountering that kind of behaviour before when I was working in a very pxeboot dependent org [23:35] * faulkes- will have to troll through the 'ol memory melon [23:35] but I have course to attend, that should jar it loose, I'll let you know === Seeker`_ is now known as Seeker`