[00:00] *****named.conf.options***** [00:00] no [00:02] mm [00:02] now this: grep options /etc/bind/* [00:02] http://paste.ubuntu.com/612998/ [00:07] that is [00:07] you are including options again in .local [00:07] /etc/bind/named.conf.local:include "/etc/bind/named.conf.options"; [00:07] delete that line from named.conf.local [00:07] and run named-checkconf again [00:10] /etc/bind/named.conf.local:39: open: /etc/bind/named.conf.local: too many open files [00:12] please, post the output from named.conf.local [00:14] half-asleep / jmedina: can you give me the command for it again ;) [00:14] just show me the content [00:14] o [00:14] k [00:14] did you edited named.conf.local by hand? [00:14] nano [00:15] that is extrange [00:18] http://paste.ubuntu.com/613002/ [00:19] if you didn't someoneelse did [00:20] never touched it... [00:22] mmmm [00:22] well there are some direct and reverse zones loaded [00:23] and other thing [00:23] named.conf.local incluids again include "/etc/bind/named.conf.local"; at the end [00:24] probably it is a old installation [00:25] purge the package and delete all the /etc/bind :) [00:25] aptitude purge bind9 [00:25] and install again [00:29] hmmmmm [00:29] * Starting domain name service... bind9 [ OK ] [00:29] jmedina: Thank you :) [00:30] :) [00:31] it looks like someone used bind in that server before you [00:32] Webmin => Servers => BIND DNS Server doesnt load :( [00:32] everything else in webmin works fine accept bind [00:33] jajaja [00:33] the it was webmin [00:33] :) [00:33] sorry webmin is not supported :) === medberry is now known as med_out [01:07] would anyone know why after setting up postfix and dovecot I can no longer check mail from the command line using 'mail' from the heirloom-mailx package? I see mail in my Maildir/new directory, but mail says I have no new mail :( [01:43] apparently Drobo shows up as series of 2tb disks in Linux? should I use LVM or is there something more preferable to make them work as one drive? [01:51] Hello, does anyone know how to close a screen session within ubuntu server? i can't find what i should type and the server i'm running on a screen session froze. [01:51] ^a d [01:51] qman__: Error: "a" is not a valid command. [01:52] I don't understand qman [01:52] hold control and press a [01:52] then press d [01:52] that just returns you to the main console, that doesn't close out of a screen session [01:53] that detaches from a screen [01:53] thought that's what you were after [01:53] if you just want to kill it, ps aux or screen list, and kill the pid [01:54] thats what i'm after =D [01:55] if it's really stuck, you may need kill -9 [01:55] but only do that if it won't stop with a regular kill [01:56] if it'll reattach (just screen -r), I'd try to kill the window with ctrl+k from within screen. that way you don't need to clean up leftover pipes [01:56] so i type in ps aux and then type..? [01:56] ah, ctrl+a, k rather [01:57] try that first [01:57] if it doesn't work, ps aux | grep [hung process name] [01:57] take note of the PID, then kill [PID] [01:57] qman__: pgrep [01:58] thank you guys, ctrl + a + k worked [01:59] i apoligize i'm new to ubuntu server, i want to do some more cool stuff with ti [02:00] that one's specific to screen. learning how to seek & destroy is useful too :) but if I can get screen to clean up it's own children, means less leftover cruft [02:00] !RUTE [02:00] documentation is to be found at http://help.ubuntu.com and http://wiki.ubuntu.com - General linux documentation: http://www.tldp.org - http://rute.2038bug.com [02:01] I don't know about you, but on my system something in my config makes screen leak memory like a mofo -- RSZ of 200MB by the end of the day [02:02] I haven't run into that problem [02:02] and I run a screen'd irssi 24/7 [02:02] on a box with 256MB [02:03] Yeah, I have talked to upstream about it and they don't know either. [02:03] i'm connecting to a vpn server that seems to affect my routes. you can see here: https://gist.github.com/990377 ... my question is; what should i do after this happens so that *only* traffic from port 80 will be sent through the assigned routes? [02:04] That's a question for #netfilter [02:05] it doesn't actually answer your question, but where it adds a route for 0.0.0.0 .. there's a good chance removing that route will do what you're actually trying to achieve [02:06] yeah [02:06] routes aren't set based on ports [02:06] so removing that route would allow me to pass nothing through the vpn [02:06] shauno: AIUI he wants *only* port 80, i.e. he needs netfilter to add marks and then "ip rule" to set up mark-based routing [02:06] it's possible to reroute traffic with some iptables magic [02:06] so sounds like: step 1) remove the route of 0.0.0.0 === j1mc_away is now known as j1mc [02:06] and 2) use iptables to do the rest? [02:07] pretty much [02:07] twb: I understand that's what he's asking. I'm not sure if that's actually the route to his problem [02:07] OK [02:07] wouldn't be the first xy problem [02:07] (as I prefixed "this doesn't answer your question"). [02:07] if you're trying to access known resources on the VPN, you don't need to do any of that [02:08] you just need to set your normal default gateway and then ensure the appropriate VPN routes are set [02:08] but if you actually want all web traffic to route through the VPN, then yeah, you need to do some fancy stuff with iptables [02:09] I'm assuming he's running into an issue like I get, where if I vpn to work, I can't access irc because work becomes my default route, and work block irc. fixing his routes so that only traffic destined for the vpn'd network goes down the vpn, and traffic destined for the public internet goes over his local link, would be a better solution to such a problem [02:09] so this can be done purely with iptables [02:09] (if my assumption is right, of course) [02:09] and i leave the routes that are pushed down in place? [02:09] no [02:09] If that's the case he just needs to not set up his VPN to be retarded [02:09] you use iptables to mark the packets you want to use those routes [02:09] shauno, that is what i'd like to do [02:09] and you adjust the routing table to notice that mark and use a seperate routing table for those [02:10] Ideally one woudl first tell the VPN client not to break the 0/0 route in the first place [02:10] and chances are [02:10] if the work VPN is not under your control, and is still set up that way [02:10] it's for a reason [02:10] check your company policy [02:10] well its the vpn provider; i just done need all my traffic to go through them [02:10] its for personal use [02:11] if you have control over the VPN server, tell it not to funnel all traffic [02:11] i sadly have no control over the vpn server [02:11] i just have control over what i do once i connect to it and after it pushes those routes down [02:11] New bug: #788419 in openssh (main) "package openssh-server 1:5.8p1-1ubuntu3 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1 useradd: cannot lock /etc/gshadow; try again later" [Undecided,New] https://launchpad.net/bugs/788419 [02:12] patdk-lap, so how would i adjust that routing table [02:12] route del default gw 1.2.3.4 [02:13] where 1.2.3.4 is the VPN server [02:15] so that will allow any traffic to flow *not* through the vpn [02:16] correct? [02:26] yes [02:26] assuming you still have your normal, internet default gateway set [02:26] if not you'll have to set it again [02:33] then the only thing i need to do is figure out how to use iptables to send traffic ... through ... ? [02:43] i'm getting: Thu May 26 05:41:03 2011 Note: Cannot open TUN/TAP dev /dev/net/tun: No such file or directory (errno=2) when i try to connect to openvpn server. from 10.04 server. [02:44] hey guys. i have 3 servers. 2 of the three can access a website the other cannot. they all go through the same gateway and have the same dns server. from the one machine that cannot ping the website using its url i can ping the ip address and it will work.. any ideas? [02:46] ...maybe here I'll have better luck. I am trying to upload a package to launcpad ( via dput ) but, I am getting a rejection that I don't understand ( https://pastebin.canonical.com/47851/). Any thoughts anyone? [02:49] * negronjl is done for the day === smoser` is now known as smoser [03:18] apparently Drobo shows up as series of 2tb disks in Linux? should I use LVM or is there something more preferable to make them work as one drive? [03:39] Hi! [03:41] I have a problem after an upgrade from 10.10 to 11.04 using do-release-upgrade - I do not have a text login any more on my default VC [03:41] tty0 seems to work fine though (using cat to and from it) [04:11] if Aison shows up, someone might point him at isc-dhcp-server-ldap === Jasonn is now known as JasonnAWAY [05:24] will mod_evasive work with apache2-mpm-prefork? [05:59] hey guys [05:59] i'm downloading some libraries to compile from source and wondering where the best place to store the source code is. I'm assuming /usr/src but figured i'd to see what others are doing [06:45] depends on the libraries :) [06:46] hah! [06:47] in this ffmpeg and various codecs [06:47] ^case [06:47] slicslak: Error: "case" is not a valid command. [06:49] perhaps the better question is, is using /usr/src a bad idea? [07:10] Suppose the root filesystem has a directory, /srv, on which is mounted a separate filesystem. *Without* unmounting /srv, is it possible to chmod 0 the mountpoint beneath it? [07:11] Hm, looks like I can with mount --bind === smb` is now known as smb [07:42] hey guys [07:42] im trying to update the motd on my server by changing and saving /etc/motd [07:43] yet every time i relogin to my server the motd message doesnt show but the server specs [07:43] and stat usage [07:43] do i need to install update-motd? [08:34] how bad an idea is it to remove the need to type a password for sudo? i have only key based auth to the server; so i'm not really worried about users logging in [08:52] flowbee: wait you mean sudo login via ssh [08:52] eagles0513875: no he doesn't [08:53] ok [08:54] flowbee: if your system is very locked down as to who can interact with it, which is sounds like yours is, then I think it's not too insane to remove password auth. however having an extra auth layer for sudo action is always a good thing :) [08:55] 16:42 im trying to update the motd on my server by changing and saving /etc/motd [08:55] eagles0513875: wrong; edit motd.tail [08:56] twb: ok [08:56] flowbee: personally I enable key-based ssh as root [08:56] flowbee: apart from bypassing that issue, it makes it less of a hassle to do e.g. ssh x tar -c etc | ssh y tar -x [08:57] I also add NOPASSWD for me, but not my users :-) [08:57] tsimpson: the downside of too MUCH auth is that then you get security by post-it [08:57] I prefer to enforce sudo use, as it's logged. but that all depends on the specific scenario [08:58] tsimpson: unless you NOEXEC, it isn't logged [08:58] At least, no more logged than sshd [08:58] and NOEXEC will break things like /etc/init.d/* [08:59] but every times sudo is used, it logs the user and command [08:59] (Hint: "sudo -i", or "sudo python -c ...", or "sudo whatever", where "whatever" is some GUI diagramming tool that happens to contain a python debugging shell.) [09:00] the point being that sudo is mostly for running a command, where ssh is mostly for logging-in [09:00] obviously not always, but generally [09:00] Maybe you use it that way, but an attacker that doesn't want his privileged commands logged is not going to run each one separately [09:01] I guess it depends if you're trying to log malicious users or accidental rm -rf's from other trusted but not-too-bright sysadmins [09:01] but the attempt to gain sudo is logged, and you can tell what account was comprised [09:02] tsimpson: well, the same applies for ssh [09:02] ssh tells you who logged in, sudo tells you who attempted to gain privileges [09:02] restricting sudo is a pre-requisite for security [09:03] I give up [09:08] Argh [09:08] printer admin installed hplip, which pulled in fucking console kit [09:12] once you set the rc.locals or whatever to make the /etc/init.d/ script active on startup... if i modify the script do i need to redo that rc.locals command? .. or will it just work [09:15] question: how can someone easy explain howto make a private cloud using virtual desktops directly from the private Ubuntu Server and dont use Amazom? [09:41] After successfully installing Ubuntu server 10.4.2 and configuring software raid all i get is a blank screen on booting! and this were the instructions that i have used https://help.ubuntu.com/10.04/serverguide/C/advanced-installation.html [09:54] morning === Nafallo_ is now known as Nafallo [12:14] damn ubuntu natty, now my dhcpd suxx :(((( dhcpd: getentry.c:46: ldap_next_entry: Assertion `ld != ((void *)0)' failed. [12:23] Hello, I need some help figuring out a LAMP problem [12:23] anyone can help? [12:23] guys just put your wuestion please, instead asking permission for that :) [12:23] soory [12:23] *question [12:23] The LAMP is installed and it is working, what I need is to create a "shortcut" from a folder in a NTFS partition and open the content from the browser. It gives me error 403 Forbidden, and I can't change the folder permissions. Anyone can help me [12:25] Hello, what difference betveen desktop ubuntu and server in case use as virtualization host (virtualbox or KVM) ??? [12:29] corumx: are you having problems with html files or executing php files? [12:29] no [12:29] aurigus: no [12:30] what user is shown as owner on your ntfs partition [12:30] and what user is apache running on [12:30] make them match and your error will be gone [12:31] aurigus: please help me change the ntfs partition ownership [12:31] frenzz, kindly check ubuntu community documentation [12:32] corumx: you set ownership in /etc/fstab on the mount point [12:33] in the options field: uid=apacheuser,gid=apachegroup [12:33] its not really a great solution but it should work [12:34] aurigus: I don't have the ntfs partition on fstab file [12:40] aurigus: I solved the problem [12:41] Beat it with a hammer? [12:42] aurigus: I just install the NTFS Configuration Tool I it works now. Possible that this tool changes permissions on the NTFS partitions. Thank you for the help ;) === nijaba_afk is now known as nijaba [13:14] kim0, ping [13:27] when using ubuntu-server as a guest, is there to use LVM? === JasonnAWAY is now known as Jasonn [13:35] koolhead11: hey [13:35] koolhead11 is everywhere :) [13:36] bencc lvm would probably make more sense on the "host" .. but it really depends on what you're trying to do [13:36] kim0: in that case I don't need it. just want a simple server as a guest. thanks [13:36] cool [13:36] kim0, hehe!! what did i do now!! :D [13:37] haha [13:37] hey TeTeT :) [13:37] TeTeT: woohoo o/ [13:37] hi gents [13:38] haha [13:39] anyone using mhddfs and have any thoughts about stability? [13:57] Daviey: i added a comment to the seeds sepc [14:06] zul, and i responsed :) [14:09] Daviey: maybe we should have a "grab bag spec" so that work items that doesnt fit really anywhere else can be put in one place and tracked [14:11] zul, I think that idea might have legs. [14:13] * Daviey ponders [14:28] I'm looking for a workaround for bug 769927 (can someone mark it as critical?). A workaround would be to unmount nfs shares automatically at shutdown before the would actually be unmounted. What's the opposite of rc.local in that regard? [14:28] Launchpad bug 769927 in linux "Kernel Oops : Dentry still in use (1) [unmount of nfs4 0:1d]" [Undecided,New] https://launchpad.net/bugs/769927 [14:32] hey guys Im looking at using x forwading for some apps. My question is it possible to keep an app running when i lose my network connection and just am able to reattach? [14:34] Doonz: NX can do that. [14:44] oh yes i forgot about that program [14:44] thanx [14:44] kees: Hey. OpenStack currently uses sudo to elevate privileges when it needs to do "stuff". This sucks for a number of reasons and we'd like to change it. Do you have any recommendations? The euca_rootwrap thing? userv? Just run everything as root? :) [14:46] Hello, anyone available here? [14:47] klzk: no, nobody is here today [14:47] I've installed Ubuntu 10.0.4 LTS server to tunnel ipv6 over ipv4 [14:48] :x [14:48] klzk: there are 309 people in this channel [14:48] But they might be not available x) [14:49] Might you be able to help me pmatulis? [14:49] !ask | klzk [14:49] klzk: 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. :-) [14:51] soren: i have a work item to look at the eucawrap thing for openstack [14:51] Kk, I've set up an IPv6 tunnel over ipv4 and pinging from the server is going great. But now i want to ping from my machines to the internet over this tunnel it won't work. If i Ping from ipv4 to this server i get response. Same thing goes if i ping Fe80 (internal) but when i try Ipv6 2001 from DHCP i don't get any response. i edited the file in /etc/sysctl.conf to accept ipv6 routing [14:52] zul: Cool! [14:53] also i forced to accept routing on /proc/sys/net/ipv6/conf/all/forwaring [14:53] ah mind the typo(s) please [14:53] after editing /etc/network/interfaces, what service do I need to restart? [14:53] sudo /etc/init.d/networking restart [14:57] zul: Not sure if it was on your todo anywhere, but I just uploaded a working Xen package to Oneiric. [14:57] which version? [14:57] 4.1.0 [14:58] lol - this norwegian news site just reported that Fedora 15 was so cutting edge it supports encrypted homedirs [14:59] hmm...i dont see it on oneiric-changes yet [14:59] soren: what did you do to fix it? [15:00] zul: "just" being the operative word. [15:00] soren: heh [15:00] It'll probably hit oneiric-changes within the next 30 seconds. [15:00] (the clock just having passed the hour) [15:01] soren: it also doesnt help that i was looking at natty-changes [15:02] :join #vimperator [15:02] zul: Probably not, but really, I didn't get the accept e-mail until after I made my "within the next 30 seconds" comment. [15:03] soren: cool anyways thanks [15:04] zul: Sure. [15:04] /etc/init.d/dhcp3-server: command not found !!! [15:05] m3asmi: natty? [15:05] pmatulis: what [15:05] N? [15:06] pmatulis: yes natty :) [15:07] m3asmi, /etc/init.d/isc-dhcp-server :D [15:16] New bug: #788623 in nagios3 (main) "Please sync nagios3 3.2.3-1 from Debian Unstable" [Undecided,New] https://launchpad.net/bugs/788623 === med_out is now known as medberry [15:39] soren: sudo is too wide, which is why I liked my eucarootwrap idea... it validated arguments, etc. alternatively, you could write a dbus backend to do the work (it would do the arg validation) [15:41] kees: wouldn't I just be reinventing userv then? [15:41] kees: sudo is too wide for what? === D0minat0r_ is now known as D0minat0r [15:42] kees: Actually, sudo lets you specify arguments that you're allowed to pass and whatnot. [15:42] kees: My beef with it is that it creates pam sessions and it's a little brittle for my taste in terms of configuration. [15:43] koolhead11: thinks a lot [15:44] koolhead11:thanks a lot ;) [15:46] ping zul [15:46] lynxman: whats up? [15:46] New bug: #788647 in rabbitmq-server (main) "Soft link to version 2.3.1 instead of 2.4.1 break plugin builds" [Undecided,New] https://launchpad.net/bugs/788647 [15:46] soren: i don't know userv. my wrapper came with helper tools to deeply validate args [15:46] zul: got all the rabbitmq stomp goodness for you [15:46] zul: https://bugs.launchpad.net/ubuntu/+source/rabbitmq-server/+bug/788647 <-- attached debdiff [15:46] Launchpad bug 788647 in rabbitmq-server "Soft link to version 2.3.1 instead of 2.4.1 break plugin builds" [Undecided,New] [15:46] RoyK: for openstack privileged operations [15:46] lynxman: thanks ill look at it laster [15:47] zul: and all the other packages here https://launchpad.net/~lynxman/+archive/ppa [15:48] kees: userv has a daemon that acts on your behalf. I/O is passed over a UNIX socket (IIRC). It sanitises environment and lets you limit args passed. [15:48] kees: It's a bit long in the tooth, so it probably does't support capabilities, though. [15:49] kees: It's a tool Ian Jackson wrote and I believe he said it had been formally audited. [15:49] tomcat uses it i think [15:49] interesting [15:50] zul: What, really? [15:50] soren: i think so [15:51] or maybe not [15:53] kees: just create wrapper scripts [15:53] kees: anyways euca rootwrapper is on the wi tracker for openstack [15:57] New bug: #788651 in python-boto (main) "Please merge python-boto 1.9b-4 (main) from debian unstable (main)" [Undecided,In progress] https://launchpad.net/bugs/788651 === Guest40648 is now known as davidm === davidm is now known as Guest68947 [16:11] kirkland: jinkeys, byobu just crashed on me it seems [16:12] i lost my ssh connection, reconnected, and it was gone [16:13] ooh, juicy, i've got a segfault in syslog [16:13] May 26 10:10:36 sergelap kernel: [658538.322656] screen[21148]: segfault at 0 ip 000000000042bce0 sp 00007fffa3a78f10 error 4 in screen[400000+58000] [16:15] hallyn: natty? or oneiric? [16:15] natty [16:16] opened bug 788670 [16:16] Launchpad bug 788670 in byobu "segfault in byobu in natty" [Undecided,New] https://launchpad.net/bugs/788670 [16:16] hallyn: technically it's screen that segfaulted, not byobu [16:17] hallyn: how did you launch? [16:17] ssh-agent byobu [16:17] well, of course, byobu is just config files... [16:17] guess i shoudl retarget then [16:18] i don't know that i can reproduce, it's a weird one [16:18] hallyn: these two lines are causing segfaults for a number of people: [16:18] layout save byobu [16:18] layout autosave [16:18] hallyn: the layout feature in screen is buggy [16:18] hallyn: let me check upstream git [16:18] what permissions do i have to set in the ftp dir to be used by wordpress for automativ upgrade? [16:18] ups sorry wrong window [16:19] hallyn: i bet it's fixed by 8cf5efc07048abee125a24652768f4b24fc761bf [16:19] hallyn: the last commit in screen's git [16:20] hallyn: were you launching a new session, or reattaching to an existing one? [16:20] attaching to an existing one [16:20] hallyn: yup [16:20] which may have been still up as the ssh session may not have timed out yet [16:25] From http://www.conservapedia.com/Wikipedia - "Wikipedia is a online encyclopedia[1] written and edited by an ad hoc assemblage of anonymous persons who are mostly, according to the Register (UK)[2][3], teenagers and unemployed persons." [16:25] rotfl [16:25] everyone should read conservapedial from time to time :D [16:27] hallyn: can you reproduce it easily, by starting another session, detaching, and then attaching in the same way? [16:28] dhcpd: Not configured to listen on any interfaces! [16:36] New bug: #788680 in samba (main) "package smbclient 2:3.5.8~dfsg-1ubuntu2 failed to install/upgrade: corrupted filesystem tarfile - corrupted package archive" [Undecided,New] https://launchpad.net/bugs/788680 [16:49] dhcpd: Not configured to listen on any interfaces! [17:33] if a project is failing to build on launchpad will it not install? https://launchpad.net/~igraph/+archive/ppa I add the repo & apt-get update & I don't see any packages for igraph or python-igraph is this due to the build failing? [17:34] natty 64bit is my version. [17:37] thesheff17: right, if it's not there you can't use it === unreal_ is now known as unreal [17:50] negronjl: ping [17:50] RoAkSoAx: hi. give me a few minutes. I"ll ping you in a few. [17:51] negronjl: k thansk ;) [17:55] hi [17:56] anyone have experience in adding freetype support to php? [18:03] teenage mutant evilsushi? [18:08] RoAkSoAx: hi. [18:10] RoyK: =D [18:30] lynxman: uploaded thanks [18:30] zul: yay :) [18:31] New bug: #788742 in mysql-dfsg-5.1 (main) "package libmysqlclient16 (not installed) failed to install/upgrade: trying to overwrite '/usr/lib/libmysqlclient.so.16.0.0', which is also in package mysql-cluster-client-5.1 0:7.0.9-1ubuntu7" [Undecided,New] https://launchpad.net/bugs/788742 [18:33] RoAkSoAx, funny that, I am looking at writing custom facts as we speak. Is it the same debdiff as yesterday? [18:33] is there some separate channel for ubuntu orchestra [18:34] What is Ubuntu orchestra? [18:34] https://launchpad.net/orchestra [18:35] hey RoAkSoAx [18:35] Daviey: check facter-customfacts-plugin in ppa:orchestra/ppa [18:36] Daviey: I just dropped a patch [18:36] negronjl, How about give me a hint what it does? :) [18:36] koolhead17: doesn't look like it, but as you can see here, it looks like some orchestra stuff is happening here [18:37] Daviey: facter-customfacts-plugin allows you to create custom facts by using a script called fact-add. [18:37] Pici: yeah just realized!! :D [18:37] Daviey: ie: fact-add [18:37] hi koolhead17 [18:37] Daviey: that is how we are currently adding new facts in orchestra [18:38] Pici: orchestra is the provisioning/config management, etc etc from Ubuntu [18:38] negronjl, interesting... what about the .rb script? [18:38] RoAkSoAx: Ah. Neat. [18:38] Pici: so it is just the integration of puppet/cobbler/etc etc into one single solution [18:39] RoAkSoAx: I was hoping it meant something like that, no need to completely reinvent the wheel. [18:39] Daviey: the script is so we can read the facts ( which are stored in /etc/facts ) and incorporate them into facter === hallyn is now known as hallyn_afk [18:40] negronjl, Hmm.. why do you need an additional command to add them? Surely being in RUBYLIB is enough? [18:40] RoAkSoAx: is munin there as well ? [18:40] Daviey: fact-add and fact-del just add the custom fact in /etc/facts where the filename is the new fact and the contents of said file is the value of said fact [18:41] negronjl, interesting... [18:41] Daviey: this way I can programatically add and remove facts without having to write a .rb file for each [18:41] customfacts.rb reads /etc/facts and adds it all into facter [18:41] Daviey: ^^ [18:42] negronjl, i see! [18:42] negronjl, elegant.... Are you putting this in oneiric any time soon? [18:42] I was working on something similar... and duplicating of work-- [18:43] Daviey: I am pushing it to bzr and my PPA for all to see/per-use/test. I'll need a sponsor ( hint hint ) :) [18:43] negronjl, happy to.. [18:44] negronjl, However, might need resync'ing with what is about to be in Oneiric [18:44] Daviey: no worries. It's a simple enough package. [18:45] negronjl, As we are starting to crack on with feature work now, we really need to be clear what each of us are working on to make sure we are complementing and not duplicating. [18:46] Daviey: I agree. Any place where current work is being coordinated === SteveCUG is now known as Steve[cug] [18:46] ? [18:49] RoAkSoAx: is there any documentation explaining about the components part and other stuff [18:50] negronjl, the blueprints? :) [18:51] * koolhead17 looks at Daviey [18:55] negronjl is there any documentation yet? [18:56] lynxman: ^^ [18:56] RoAkSoAx: not yet. we'll have it soon [18:56] koolhead17: ^^ [18:56] negronjl: thanks ;) [18:56] RoAkSoAx: thanks. :D [18:58] am more interested to see munin in the list too 4 monitoring :) === NG_ is now known as ng_ [19:43] is there any way of testing/debugging cloud-init userdata after instance has booted? [19:43] ^ smoser [19:43] adam_g: Error: "smoser" is not a valid command. [19:44] adam_g, yes. [19:44] the real question, though, is why smoser is not a valid command [19:45] misallocation of resources. i blame management [19:45] https://groups.google.com/group/ec2ubuntu/browse_frm/thread/d4d51238a2afb55b/bca69f3e89c88d0b?lnk=gst&q=fast+way+to+test+user-data#bca69f3e89c88d0b [19:45] danke [19:47] fyi, seed directory changed in natty [20:10] anybody else having problems with archive.canonical.com? [20:17] would someone mind hitting http://archive.canonical.com and letting me know the results? [20:18] * soren recommends http://www.downforeveryoneorjustme.com/ [20:18] Daviey: facter-customfacts-plugin is ready in lp:~negronjl/+junk/facter-customfacts-plugin [20:18] thanks soren [20:18] Daviey: check it when you get a chance and let me know [20:18] that's handy [20:18] jimbobco: Quite. [20:19] jimbobco: It has no sense of humour though... http://www.downforeveryoneorjustme.com/www.downforeveryoneorjustme.com [20:20] made me laugh :) [20:20] wonder if there's a loop there to be exploited === skrewler_ is now known as skrewler [20:38] negronjl: ok, will look shortly [20:50] I posted this in #ubuntu but was told this would be a more appropriate place to post it. [20:50] I'm really confused. Apache should always load the first virtual host if it doesn't match any others but mine is ALWAYS loading the second virtual host. my vhost file: http://pastebin.com/zeJjpMJb. Locations in question: http://50.57.73.250/ http://stage.dev.mvretail.com/ [20:52] BPower: Because the second one has a more specific [20:53] BPower: Make them the same (i.e. ), and you're golden. [20:53] soren, thanks. I just fixed that and it's working now for those two. [20:53] now i've added one more vhost and it's always going to the default. i'll paste the new file [20:54] http://pastebin.com/wrT1fMkB [20:54] going to http://dev.mvretail.com/ should say "We're live!" [20:54] going to http://stage.dev.mvretail.com/ should say "We're stage!" [20:55] and going to http://50.57.73.250/ should say "We're www!" [20:55] as you can see, all work except live [20:56] permission error? [20:57] make sure DocumentRoot /var/www/live/MerchantView/ is readable my apache i think [20:57] do ls -alh on /var/www/live [20:57] littlebearz, are you talking to me? the entire /var/www has the same permissions (I just did it) [20:57] oh, wait, which one doesn't work? [20:58] live [20:58] hm.. I don't see any thing wrong with your config, anyone else found anything? [20:59] I'm going to test swapping the last two and see what happens [21:00] thanks littlebearz, soren [21:00] maybe it's your vhost taking all the *:80, but i don't think that's the problem [21:00] let me dig out my config [21:01] nvm LOL, i have a nginx config http://pbin.xxw.ca/84fb3b462.php [21:01] BPower: That's a bit confusing. However, do you have a NameVirtualHost *:80 line elsewhere in your config? [21:02] maxb, ya, ubuntu-server has it by default in /etc/apache2/ports.conf [21:05] btw littlebearz maxb & soren, swapping the order didn't fix it or change any behavior at all [21:06] BPower: hm.. i might check if the folder exist and then the typo [21:08] littlebearz, no typos and folder exists [21:08] (just double checked) [21:11] BPower: i'm baffled, um. try to remove the / after MerchantView ? [21:13] BPower: Don't put port number in the servername. [21:17] littlebearz, soren, trying now thanks :) === skrewler_ is now known as skrewler [21:21] littlebearz, soren, applied both -- don't know which did it but now live.* works but dev.* still doesn't [21:27] BPower: What does your conf look like now? [21:33] soren, littlebearz, maxb: Found the problem [21:34] Since /etc/apache2/conf.d/fqdn was not created, it would use my server's hostname as the default for anything it didn't recognize [21:34] My server's hostname happened to be dev.mvretail.com [21:34] so it was hijacking the vhost [21:35] i solved it by creating /etc/apache2/conf.d/fqdn with "ServerName localhost" === skrewler_ is now known as skrewler [21:47] BPower oh thanks, I didn't know that , i thought apache auto detect localhost and uses that as FQDN [21:48] littlebearz, ya apparently it autodetects the actual hostname ...which seems like more of a bug than a feature to me. [22:00] BPower: experience is what you get when you didn't get what you wanted :) [22:01] lol nice === ng_ is now known as NG_ === skrewler_ is now known as skrewler === robbiew1 is now known as robbiew === SpaceBass is now known as away === away is now known as SpaceBass