[00:00] marks: it's wherever you've configured the servername. Normally you'd have a file for it under /etc/apache2/sites-available/, sometimes in 'default' [00:00] except I got the parties in the conversation back to front. *cough* [00:00] BigRedS, and jkg, it is, lets see it, one moment === cmagina is now known as cmagina-afk [00:03] kirkland: grrr give me a sec [00:03] ivoks: was away.. in the bus.. checking the link now [00:04] kirkland: what's the error message though? [00:04] RoAkSoAx: http://pastebin.ubuntu.com/567165/ [00:07] BigRedS, and jkg, an alias without, WWW? [00:07] anyone ever tried using SLOG with zfs-fuse? [00:08] marks: yeah [00:08] exactly the same as the ServerName line, but ServerAlias in place of ServerName, and no www. [00:08] marks: do you have a line like "ServerName www.whatever"? if so add a line under it, "ServerAlias whatever" [00:08] the www is just another subdomain as far as apache is concerned - it has no special meaning [00:08] No i don't, and odly, my HTTPD is empty. [00:08] wtf am I in this channel [00:08] marks: apache's central config file is now /etc/apache2/apache2.conf [00:09] but sites are in /etc/apache2/sites-available [00:09] kirkland: yeah, i'll get that in the build, thanks. (I'm trying to debug test break libvirt 0.8.7 on armel first) [00:09] marks: httpd.conf is kept for compatibillity, but not used [00:09] RoyK: it's the glamour and the excitement, isn't it? [00:09] marks: you need to get the ServerAlias in the right VirtualHost, though, else it wont work [00:10] Its the global one i need to edit [00:10] jkg: not really [00:10] marks: the global what? [00:10] the default site is /etc/apache2/sites-available/default [00:10] The global domain, like not a hosted one. Would that be in the apache2.conf? [00:10] the core not-site-specific config is in apache2.conf [00:10] hallyn: with that, vmware and std work [00:10] marks: no. /etc/apache2/sites-available/default [00:10] hallyn: still no go with qxl [00:10] hallyn: did we need to add a configure option to qemu for that? [00:11] kirkland: not my fault this time :P :) what does 'file ./natty-desktop.img' show you? [00:12] RoAkSoAx: $ file natty-desktop.img [00:12] natty-desktop.img: QEMU QCOW Image (v2), 6442450944 bytes [00:13] kirkland: no, i didn't see a configure option for qxl [00:13] looked... [00:13] hallyn: k [00:13] kirkland: in maverick and before it used to show "QEMU Image" instead of "QEMU QCOW Image", and that's why it doesn't recognized as an image [00:13] BigRedS, can i show you a pastebin of what i have? I don't think its right. [00:13] kirkland: seriously? -vga std works? with a lucid image? [00:14] RoAkSoAx: okay, just look for "s/qemu/i" [00:14] hallyn: sorry, no, natty guest [00:14] marks: yeah, go for it [00:14] kirkland: just to sanity check, debian/install did not previously exist right? [00:15] BigRedS, http://pastebin.com/fW6WEVC1 [00:16] hallyn: uh, in what? vgabios? [00:16] marks: is that the virtualhost you want displayed at your domain? Just to check, do you have DNS records for both the www. and the normal domain in place pointing to that machine? [00:17] hallyn: right, "make install", I think, handles it today [00:17] hallyn: perhaps it's the Makefile that needs to be patched [00:17] kirkland: no, i just mean - you created a new debian/install file right? [00:17] BigRedS, how so what do you mean [00:17] marks: So long as you've not specified a ServerName or Alias with that domain, that's the vhost that should be displayed [00:17] hallyn: i did not [00:17] marks: what's the actual domain name you're using? [00:17] hallyn: i built locally, and sudo cp'd the files into the place qemu expects, as a temporary hack [00:17] I'm using http://tm0.org [00:17] Which doens't work without WWW [00:18] kirkland: will upload tomorrow. Will also include the Messaging Indicator :) [00:18] RoAkSoAx: rock ;-) [00:18] RoAkSoAx: oh, do one more thing ... [00:18] kirkland: sure [00:18] marks: works for me [00:18] RoAkSoAx: create a $HOME/.cache/testdrive/kvm.log [00:19] RoAkSoAx: that appends each kvm message run [00:19] RoAkSoAx: sometimes, I want to lookup the kvm message i used to launch a vm [00:19] RoAkSoAx: and use that again [00:19] marks: though now I can't get a dns lookup for it? [00:19] kirkland: ok cool [00:19] RoAkSoAx: like when -u file.img breaks ;-) [00:19] BigRedS, That might be my configuration [00:19] marks: both work for me, I suspect it's something local to you :) [00:20] kirkland: sure thing. Been wanting to add more logging support. I guess it is about time :) [00:20] BigRedS, is it possible that domain propingation was slower on one? I did recently add another domain in webmin without WWW [00:20] RoAkSoAx: cool [00:20] marks: it's possible [00:20] you can use the 'host' command to see the current records as far as your machine is concerned [00:20] RoAkSoAx: perfect, thanks! [00:20] hallyn: okay, no rush, tomorrow or this week is fine [00:21] marks: hosts tm0.org for example [00:21] hallyn: would be nice to do it before our meeting on Wed [00:21] kirkland: no probs ;) [00:21] hallyn: btw, will you be able to make it to that Happy Hour? [00:21] That might be it, i'll try through some proxies, thanks BigReds, and i will try the hosts [00:24] kirkland: I'm not sure. I"ll try [00:24] hallyn: RoAkSoAx: take care dudes, time to give the wife some attention for V-day ;-) [00:25] kirkland: enjoy!! [00:26] ttyl [00:27] i have a server which won't reboot [00:27] any ideas? [00:29] BigRedS, wonderful, it is just propinigation, thanks so much for your help? [00:29] ivoks: so you need another person at the site to build the network before you can continue ? [00:30] ibm watson is in jeopardy now! [00:33] marks: Ah, perfect! No problem! [00:34] i have an ubuntu system which will not listen to shutdown -r now or init 6 - anything else we can try? [00:34] iclebyte-work: is it doing anything at all? [00:35] it says it's shutting down but it wont go down [00:36] So, I'm new to postfix, I'd like to set up postfix through gmail. I saw it references /etc/postfix/main.cf, but I have /etc/postfix/master.cf, is this still relevant and/or is the wiki up-to-date? [00:36] define "through" [00:36] sorry, the wiki references /etc/postfix/main.cf [00:36] iclebyte-work: at a guess, it's waiting for a service to finish. what's at the end of syslog? [00:37] h00k: postfix uses both files [00:37] twb: as in...use google as an smtp relay, from what I understand. [00:37] and more [00:37] BigRedS: okay, so it is separate. [00:37] BigRedS, nothing relevant to a shutdown [00:37] BigRedS: Because the wiki was old, I wasn't sure if the file had been changed from main to master [00:37] Hm, that should be trivial except for having postfix know your gmail username and password. [00:38] Also gmail might still rewrite your messages so they all appear to come from you. [00:38] twb: that's okay, I don't mind [00:38] h00k: ah, main.cf configures most of postfix, master.cf how it interacts with other apps [00:38] Personally I use msmtp for this purpose. [00:38] twb: I'm setting up Nagios for the first time [00:38] BigRedS: ah, makes sense. [00:38] iclebyte-work: I'd start looking at each service you're running, see if any are crashing/crashed for some reason [00:39] iclebyte-work: have you a real-life monitor on it? [00:44] kirkland: heh, my first attempt at fix didn't work. No worries, I'll do it later. (For vgabios) [00:49] h00k: apt-get install msmtp-mta [00:51] Then put something like this in /etc/msmtprc http://paste.debian.net/107660/. [00:51] Then you can either put the password in cleartext in ~/.netrc or /etc/msmtprc, or type it every time. [00:55] all right, fixed [00:58] twb: would I be able to use that through Nagios? [00:58] AFAIK nagios uses mail(1) which uses sendmail(8) which is the interface msmtp-mta provides. [00:59] twb: okay, and what would the syntax be if I included the password in /etc/msmtprc? [00:59] 1 grunt nagios3: Warning: Attempting to execute the command "/usr/bin/printf "%b" "***** Nagios 2.6 *****\n\nNotification Type: PROBLEM\n\nService: HTTP\nHost: soy.cybersource.com.au\nAddress: soy.cybersource.com.au\nState: +WARNING\n\nDate/Time: Mon Feb 14 14:44:40 EST 2011\n\nAdditional Info:\n\nHTTP WARNING: HTTP/1.1 404 Not Found - 466 bytes in 0.004 second response time" | /bin/mail -s "** PROBLEM alert - [00:59] soy.cybersource.com.au/HTTP is WARNING **" +support@cybersource.com.au" resulted in a return code of 127. Make sure the script or binary you are trying to execute actually exists... [00:59] ...I get a lot of those because mail(1) isn't installed on grunt yet :-/ [00:59] h00k: I don't know, the manpage does. [01:00] h00k: if it's not obvious ask again [01:00] twb: roger. [01:09] twb: I tried testing without specifying a password, I'm getting a reply of: http://pastebin.com/60qxMh5Q [01:10] twb: (after I actually type in my password) [01:10] h00k: did you type in your password? [01:10] twb: yep. Also, /etc/msmtprc is http://pastebin.com/UUMwubR2 [01:10] That's saying you typed it wrong [01:10] "server message: 535-5.7.1 Username and Password not accepted." [01:11] It lies! [01:13] Shrug [01:14] twb: I'll futs around more, thanks for the alternative, though! [01:15] twb: it still looks like it's using PLAIN authentication and not TLS [01:15] * h00k googles [01:15] h00k: run it with -v [01:16] h00k: perhaps you don't have ca-certificates installed? [01:16] twb: I do, [01:16] twb: and -v gave me the same thing. hmm. [01:16] Then I dunno [01:17] "method PLAIN" probably means it's doing cleartext passwords *inside* the SSL tunnel [01:17] As opposed to e.g. doing digest without ssl, which sends an MD5 hash of your password that any MITM can see [01:18] http://pastebin.com/AcdkU2jv [01:21] twb: turns out I'm an idiot. my email was wrong. [01:22] * h00k facepalms [01:22] twb: thanks for the assistance. [01:30] how do i run a cgi file? [01:30] goddard: execute it. === Kiall|AFK is now known as Kiall [01:33] do i have to put it in a special place? [01:33] i have a cgi bin [01:35] No. [01:35] Executables are executables. [01:38] i keep getting a 404 error [01:38] goddard: you're using a browser? [01:38] i mean 403 [01:38] yeah its a emailer [01:39] Well, if you're doing it via HTTP then you need to set up your httpd correctly. [01:39] ive enabled cgi [01:40] ok almost got it i think [01:40] got a 500 error now [01:46] any ideas how to fix this? [01:49] I'm replacing a host that was serving some simple static HTML content, plus webalizer. Apparently the functional requirement for the latter is "when we make a press release, we want a rough idea of how many people looked at it". What's current best practice to meet such a requirement? My emphasis is on simplicity and security. [01:49] My current inclination is a cron job along the lines of grep -c 200.*/press /var/log/apache2/access.log >/var/www/hits.txt. [01:52] arg i got an inclination to break this computer rarrrr === Kiall is now known as Kiall|AFK [02:48] twb: I attach my companies web site to google analytics. It is just some small javascript. They have something called campaigns where you can just append some variables to the end of the url for tracking. It is really useful for what you are talking aobut. [02:50] Yeah, that's what $boss suggested [02:50] My main objection is that google already has too much power :-/ [02:51] haha true...I also use awstats if it is just a static page you are creating [02:52] it is basically just runs against your access.log file like you said [02:52] Okey dokey. [03:32] piwik is another option. [05:02] test, anyone please respond === Kiall|AFK is now known as Kiall [06:43] Hello, I suggest to have kerrighed (http://www.kerrighed.org/) in natty+1. That would most probably require adding another kernel flavour. [07:25] "Kerrighed is a Single System Image operating system for clusters. Kerrighed offers the view of a unique SMP machine on top of a cluster of standard PCs." [07:25] ...so plan9 cpu pooling is fashionable now? [07:35] twb: plan9 cpu pooling ? [07:35] what's that ? [07:37] twb: nevermind I found about it on wikipedia [07:37] http://en.wikipedia.org/wiki/Plan_9_from_Bell_Labs [08:13] axisys: no, i preseed network configuration === airtonix_ is now known as airtonix [08:38] Can someone help me with some ipconntrack / connections through firewall issues [08:38] http://monitor.wingify.com/munin/visualwebsiteoptimizer.com/lb1.visualwebsiteoptimizer.com.html#Network [08:46] Sparsh: are you using iptables-restore? [08:47] I am not sure, its default ubuntu 10.10 settings. How can I find out [08:47] twd: I dont know.. How can I find out [08:48] Sparsh: if you don't know, then you aren't [08:48] Sparsh: how ARE you loading firewall rules? [08:49] twd: I am not doing anything. all I did was ported a ubuntu 10.10 64 bit image, install nginx on it, and thats it [08:49] I need to make all other changes to it [08:49] twb: eveything is default as of now, but I want to change it to get more out of my servers [08:49] Sparsh: then you have no firewall [08:50] twb: suggest me a good firewall [08:50] There only is one firewall: netfilter. [08:50] twb: this server would be front end and will get around a billion requests per month [08:50] There are a bunch of wrappers around it to make it more "user friendly"; Ubuntu's recommended one is ufw. [08:51] Because I know what I'm doing, I prefer to write the ruleset by hand and call iptables-restore on it before the network comes up/ [08:51] okay [08:51] I dont want to ask you a lot of questions but would love to read about all these things in more detaisl... what all should i read about to get a complete knowledge of whats happening to my server and how can i optimize it, [08:52] BTW, I just did this [08:52] root@li235-57:~# sysctl net.ipv4.netfilter.ip_conntrack_max [08:52] net.ipv4.netfilter.ip_conntrack_max = 65536 [08:52] I guess 65536 is not enough for my server [08:52] If you don't have a firewall, then you aren't tracking connections in the first place [08:53] but my munin graphs show [08:53] nginx? [08:53] http://monitor.wingify.com/munin/visualwebsiteoptimizer.com/lb1.visualwebsiteoptimizer.com.html#Network [08:53] as a proxy [08:53] ? [08:53] yes [08:53] nginx as reverse proxy [08:54] http://paste.debian.net/107682/ <-- that routes a whole /24, and it doesn't get anywhere near the default conntrack limit. [08:54] and the problem is...? [08:55] i just see that you have a problem, but i don't see what the problem is [08:55] s/see/read/ [08:55] yes, even I dont know the exact problem but I want to be sure [08:55] that the server is fine [08:55] and configured properly [08:55] and can handle 700-800req/second with nginx [08:56] and wont have any ipconntrac related issues / or firewall issues [08:56] Sparsh: if you want to test that, then make 800 requests a second and see what happens [08:56] I tried that [08:56] it can [08:56] and ipconntrac stopped [08:56] its visible in munin [08:56] it can handle much more than 800req [08:56] if you look http://monitor.wingify.com/munin/visualwebsiteoptimizer.com/lb1.visualwebsiteoptimizer.com.html#Network [08:56] atm, your server is idle [08:57] at the moment, this server is behind another nginx [08:57] just because it gave some problems which munin recorded [08:57] it could be problem with a munin plugin [08:57] It is behind another nginx right now: http://monitor.wingify.com/munin/visualwebsiteoptimizer.com/dev1.visualwebsiteoptimizer.com.html [08:58] well.. the actual numbers of server are kinda same [08:58] of 2GB, you only use 0,4GB of RAM [08:58] traffic is at the peek 6mbit/s [08:58] cpu is idle [08:58] load is 0.1 [08:59] Sparsh: are you suggesting that the gaps in http://monitor.wingify.com/munin/visualwebsiteoptimizer.com/lb1.visualwebsiteoptimizer.com-fw_conntrack-day.png are points where your host stopped responding to HTTP GETs altogether? [08:59] No, that didnt happen [08:59] it was still responding but became slow [08:59] time_wait increased [08:59] I don't understand what the problem is [08:59] I dont know how to debug [08:59] I can increase traffic again but should know what a/ how to debug whats happening [08:59] http://monitor.wingify.com/munin/visualwebsiteoptimizer.com/lb1.visualwebsiteoptimizer.com-fw_conntrack-day.png is this graph normal? [09:00] so you had a problem? [09:00] I don't know. I'm a sysadmin, not a webdev. [09:00] response time was bigger? [09:01] yes [09:01] so the situation is [09:02] is the http load time plugin working? [09:02] it says that load time is 0s [09:02] no, I more keep and eye on response time using external services like pingdom [09:03] all I want to make sure is,- [09:03] IMO don't bother doing anything until users start actually complaining [09:03] I mean, obviously they're ALWAYS complaining, but wait until they start complaining about THIS [09:03] well.. if i put a lot of traffic to this server directly, and point DNS to it [09:03] I want to be confident [09:03] that this can face that much traffic [09:04] there is another server which is old now, which is doing all teh traffic [09:04] Sparsh: why? Are you building a hospital call center website or something? [09:04] i have a question about renice. if you have a process that consumes large cpu resource and you launch another cpu demanding process with lowest priority, would it still affect eg slow down the first process even in a bit? [09:04] Maleko: the OS will balance it, but yes, it will still take a slice of the cPU [09:04] Maleko: it won't queue it [09:05] twb: no, [09:05] but if you have modern cpu, with multiple cores [09:05] twb: it serves a javascript which is included in other websites and I dont want them to sloe down [09:05] AIUI each process gets a slice of the CPU time, and the scheduler uses nice to decide what proportion [09:05] Sparsh: next time use varnish :) [09:06] I am using Varnish [09:06] as well.. got 2 servers [09:06] nginx is reverse proxy as I dont want all the requests to hit varnish [09:06] ivoks: unfortunately that requires you to install cc on the host :-/ [09:07] twb: ? [09:07] ivoks: varnishd calls sh -c 'cc -o/var/tmp/vclXXXXXX.o /var/tmp/vclXXXXX.c' before dlopening it [09:08] Rather than parsing /etc/varnish.vcl, it turns it into a C translation unit and compiles it. [09:08] Sparsh: http://us.generation-nt.com/answer/bug-565565-munin-node-plugins-fw-conntrack-fw-forwarded-local-cause-severe-network-lags-big-firewalls-help-168991771.html [09:09] ah, cc as a compiler :) [09:09] Sparsh: http://munin-monitoring.org/ticket/843 [09:09] ivoks: try http://bugs.debian.org/565565 -- same content, canonical location [09:10] interesting bug [09:10] Sounds like a kernel misfeature [09:10] In any case, calling conntrack -S instead will DTRT, it says [09:11] ivoks: what did *you* think "cc" meant? [09:12] Command & Conquer :) [09:12] Sheesh [09:12] Creative Commons [09:12] :) [09:12] C.C. from Santa Barbara tv show [09:13] Admittedly I was talking about C&C networks (nowadays more like C3I) today, and I could see that going bad the same way... [10:22] morning [11:26] New bug: #719239 in mysql-5.1 (main) "Debconf should give a warning about .my.conf" [Undecided,New] https://launchpad.net/bugs/719239 [11:31] how does one install new kernels in ubuntu-server ? [11:35] apt-get update / upgrade? [11:38] if its available from the repositorys it will be apt-get upgrade'able, or you can compile your own kernel [11:41] kernels won't install from just an upgrade, you need dist-upgrade or full-upgrade [11:41] <_ruben> apt-get dist-upgrade in case there's been an abi bump in the kernel package [11:41] since it requires installing new packages [11:54] apparently linux-image-2.6.35-22-virtual is alerady installed, I am using 35-25, I need to switch back to 22...how can i do that ? [11:56] teddymills, the answer would depend how you are consuming it, is this xen, kvm, ec2? [11:57] Ec2..i see /boot/grub/grub.cfg with the grub entries...isnt there a grub command to say what entry to use? or does it take the first i nthe list ? [12:01] default=0, and it is loading 35-25 (not what I want) 35-22 kernel is number 4..counting from 0..so if i set default=4 that should work [12:02] or it could kill the ec2, very bad :( [12:04] i think it will play it safe on this one, and do it on a test instance first. the instance can wait a day. it will take many people a week to put the instance back together [12:26] New bug: #719260 in openssh (main) "X509 support for openssh" [Undecided,New] https://launchpad.net/bugs/719260 === oubiwann_ is now known as oubiwann [12:41] ivoks: do you have a blog or link with some more details? I wish that link has some examples for newbies [12:52] I'm installing the LAMP task in 10.04. Three times, it asks me if I want to set a password for the MySQL root user. Why? [12:57] jo-erlend: it checks 3 times during that particular install whether root has a p/w [12:57] yes, but why do I have to answer the same question over and over? [12:58] jo-erlend: you input a p/w and it asks again you mean? [12:59] I chose not to use a password. It continues its work, and then it asks me for that password again. I choose not to use one, and a little while later, it asks again. [12:59] jo-erlend: right, i saw the same just yesterday. probably if we put in a non-blank p/w the first time it would not prompt again [12:59] jo-erlend: but you'll need to verify that to be sure === hito_jp0 is now known as hito_jp [13:20] JamesPage: i noticed you added the amd64 lucid tests can you add the i386 ones as well [13:45] axisys: details for what? [13:49] JamesPage: we are running out of disc space on hudson... === FkCek is now known as FkCek|a [13:57] morning all [14:02] hggdh, call up dallas then [14:02] patdk-wk: will keep that in mind ;-) [14:03] hggdh: are you guys still suffering from winter armageddon [14:03] zul: not anymore, but it is still cold -- 70F/18C [14:03] :-) [14:04] yeah thats cold...:P [14:05] hggdh - I'll take a look at the space issue - I can prob make a minor amend to the package to make it a bit more economic [14:05] Hey there guys. Trying to set up a certificate with freeradius. I'm generating it from a server.cnf and client.cnf file. I'm curious if anybody knows which values I'm supposed to edit in these files prior to running sudo make (which makes the certificate) [14:06] JamesPage: meanwhile I am claning up a bit of space so that Hudson can run again (right now it seems it died a most horrible death) [14:06] gah - sounds nasty. [14:06] do we have anything in the artifacts created by each build that could be compressed as well? [14:10] hggdh: here it is 14C/47F [14:11] JamesPage: certainly the d-i syslog, each goes from 500K- 1M [14:11] RoAkSoAx: now *this* is something... Miami, is it not? [14:12] OK - I'll update the config to compress this file post test execution. [14:12] hggdh: yes Miami [14:12] JamesPage: also, how about just keeping the last 'n' runs? [14:12] ('n' to be discussed) [14:13] JamesPage: a good candidate, probably, is the console output [14:13] that would be good as well - that would be a Hudson thing rather than usit but it should be possible [14:14] hggdh: are you bouncing it at the moment? [14:14] JamesPage: I will in 3 min, just finishing applying updates (Hudson was dead on the water, anyway, 503 error) [14:15] prolly something about wanting some space to write something or other, rather picky programme [14:15] :-) [14:18] JamesPage: rebooting now [14:19] hggdh: let me know when its back up - I can then re-config the jobs to make then kinder on disk space. [14:19] I'll also push a new version of the package to the PPA later today to help as well (removing the build number from the workspace directory - no longer needed) [14:21] JamesPage: well, it is up, but still sick [14:22] lemme log in and take a look - alot of the jobs will be holding artifacts that they should not be. [14:24] ivoks: well.. do you setup a dhcp in same network? [14:25] kirkland: let me knwo if you see any more problems with my vgabios package; if not i'll go ahead and clean them up and put them on p.c.c for you (with fixed changelogs) to dput [14:27] axisys: of course [14:28] kirkland: drop-off duty, bbl [14:28] ivoks: ok [14:30] someone I can be in form of the asterisk server [14:33] Hey [14:33] I found a program that can play music on my server's audio speakers [14:33] and it allows me to stream my music to any device across the internet [14:33] :D [14:35] kirkland: ping [14:39] hggdh: man where those installer logs taking up alot of space! [14:43] JamesPage: yes... amazing what we do for data, is it not? [14:44] hggdh: its still compressing by there is more than 50% free space in the filesystem now. [14:45] JamesPage: super! [14:46] JamesPage: now, why does it return a blank page on the web? [14:46] hmm - I don't have that issue [14:47] perhaps cache on my side. I will clear it [14:48] no, it's just firefox misbehaving. All is kosher now [14:54] kirkland: never mind :) [14:55] RoAkSoAx: here now [14:56] hallyn: go ahead and push them all to p.c.c [14:56] hallyn: i'll pick them up, build locally, test, and upload [14:59] ivoks: is there a way to record my lvm create operation.. so I can put that in preseed ? I already built a box with lvm2 .. is there a way to generate a pressed for disk parition section based on the lvm2 I already have on a machine otherwise? [14:59] hggdh, zul: I've created i386 versions of the lucid regression ISO tests in Hudson; I've not run them yet but looking at the amd64 versions they should be OK. [15:00] axisys: you can make lvm map files [15:00] JamesPage: thank you [15:00] JamesPage: cool ill run them this afternoon thanks! [15:00] ikonia: how to do I do it [15:00] ikonia: ? [15:00] axisys: look at vgexport [15:00] be very careful [15:00] ikonia: these are not production systems yet [15:00] ok, don't be careful [15:01] ikonia: lol [15:01] kirkland: bug #717643 [15:01] Launchpad bug 717643 in qemu-kvm "package qemu-kvm 0.13.0 noroms-0ubuntu12 [modified: usr/share/man/man1/qemu-user.1.gz] failed to install/upgrade: trying to overwrite "/usr/bin/qemu-system-ppc64" wich is also in package qemu-system 0.13.50-2011.02-0-0ubuntu1" [Undecided,New] https://launchpad.net/bugs/717643 [15:04] ikonia: ok.. not following .. vgexport will just export my volume.. when i run vgexport -a .. it says rootvolg0 has active logical volumes [15:05] axisys: you can't export while the volumes are active, you want to look at the -m option [15:06] I think -p -s -m $mapfile rootvolg0 will allow you to do it while the volume is active [15:06] rusty with this on Linux [15:08] axisys: you can, but it's not automatic; you'll have to write the rules your self [15:11] ikonia: let me try that.. sorry was on the phone [15:12] ikonia: vgexport does have -p or -s or -m option [15:14] ivoks: there is an example lvm on line 66 here http://pastebin.com/xXxx0n3q [15:14] what are those numbers .. like 500 10000 1000000000 [15:15] may be I change them to something like mine.. [15:15] i got those example from http://www.pro.instalinux.com/cgi-bin/coe_profiles.cgi [15:19] hggdh: Now that you guys testing candidates for 10.04.2. shouldn't http://iso.qa.ubuntu.com/qatracker/dllist be updated? [15:20] axisys: i'm not sure... i always partition disk manually [15:21] RoAkSoAx: I do not have access to it; additionally I am not familiar with the procedures... sorry [15:21] but I think it will get updated very soon [15:24] axisys: http://debian.ues.edu.sv/compartidos/neozerosv/ciberues/partman-auto-recipe.txt [15:24]