[00:21] chmod 000 .known_hosts ; chmod 644 when your done [02:16] Can anyone link to a blog or how to which quickly goes over multipath fibre channel in Ubuntu, and the tools to use [02:17] I've just assigned two disks the a server, atm it's two seperate disks [03:47] what generates /dev/disk/by-uuid symlinks at boot time? [03:48] udev, I imagine. [03:48] I just created a multipath disk mapping but the UUID is pointing to one of the single disks in the DM, and not the /dev/mapper node [03:48] If you're using dmraid, be very, very afraid. [03:48] so I cant use the UUID in fstab to mount it, I'm manually specifying the /dev/mapper entry [03:48] not dmraid, multipath-tools (multipath, scsi_id) [03:48] OK, cool. [03:48] ubuntu's pretty pathetic when it comes to FC SAN disk and MPIO [03:48] but I managed to get it to work [03:49] oh_noes: did you write an article about your adventure? [03:49] fibre's a bit above my means, but I'm still curious [03:49] I should, there's not enough blog posts about [03:49] LAMP, 40000000000000, FC SAN MPIO DM, 1 thats got typos and isnt correct [03:50] LAMP blows anyway. [03:50] Why "the kids" flock to mysql and php, I do not know. [04:02] hi guys I need some help on getting horde/imp to show up on my server I used the command line to install it by apt-get install imp4 [04:12] hey [04:12] i just installed the server for the first time (i am a desktop user for some time now) [04:12] hi guys I need some help on getting horde/imp to show up on my server I used the command line to install it by apt-get install imp4 [04:14] unfortunately during I chose to encrypt the homes on it and now I cannot use ssh keys to access it (that is at least my theory as to why keys don't work). what's the fastest way to disable encryption? [04:14] TimReichhart: do a sudo dpkg -L imp4 [04:15] that is gonna tell you where files are located [04:16] if apache is not aware of the location /etc/apache/conf.d/... you need to either make a symlink to imp inside /var/www or make an alias [04:16] hope this helps... [04:23] well see Im seeing its listed under /etc/horde3 and /etc/horde/imp4 [04:24] but no web pages? [04:24] that is correct [04:24] so I have to create a sym link [04:25] ...weird. i never used it but had several web projects installed and all had some "meat" to them [04:26] i cannot help you more i guess. might want to try installing from the horde people. not sure how complicated that is [04:26] ya this is my first time even doing horde/imp [04:27] TimReichhart: try the official version. they prolly have a nice step by step howto. its also likely that its newer version [04:28] ...takes a while for stuff to reach ubuntu packages [04:28] i have the newest server 9.04 so [04:28] nvm i found it [06:00] Is there a channel for LTS *desktops*? [06:29] twb: #ubuntu or #kubunt should be able to respond to LTS questions [06:29] Har har. [06:32] then perhaps the forums or questions section on Launchpad would be the place to try. [08:57] anyone alive? [08:57] billybigrigger, yes? [08:58] my grub messed up on me, now my raid wont mount [08:58] fdisk shows sda1 and sdb1 [08:58] http://pastebin.com/m5ecd9eba [08:59] but i can't mount either sda1 or sdb1 [08:59] and md0 isn't showing [08:59] im in 9.04 live cd right now...i just a few files off it immediatley and i can deal with the grub issue in the morning [09:08] how do i re-create md0? [11:51] hi, how can i show ssh sessions? [11:53] i got an error so putty got killed, now i need to return to that ssh session [11:55] You can't [11:55] You might want to investigate the 'screen' program for use in the future [12:01] ? [12:01] what do you mean? [12:24] Hello! [12:24] anyone here using Intels RMM2? [12:24] my keyboard does not work with my custom kernel. It works with a ubuntu live cd [12:24] any idea what modules i am missing? [12:44] i've lost my /dev/md0 device for my raid 1 setup...fdisk shows sda and sdb as linux raid autodetect...now how do i go about re-creating my raid device md0? [12:45] you could use MAKEDEV if you know the numbers [12:45] but the driver should create that device [12:45] if I'm not mistaken [12:45] what kind of raid is it? [12:45] raid 1 [12:45] i set it up during 9.04 install [12:45] so software raid 1 [12:46] ah, software raid! [12:46] lvm? [12:46] no clue [12:46] i just picked the defaults all the way though :P [12:46] through [12:46] probably with mdadm [12:46] you have something important on it? [12:46] it was all setup in ubuntu install, so i don't know the process of setting it up, or configuring it when its broken [12:47] lol yes of course, there's always something important on the disks that go down! :P [12:47] ha! [12:47] wait a sec. how did you lose it in the first place? [12:48] did you just delete it manually? [12:48] i don't know [12:48] no [12:48] ahh now that i think about it, hard system lock up, then when i rebooted /dev/md0 isn't there [12:48] not shutdown properly [12:48] so you didn't erase it. good. [12:48] this means there is a problem with it, and this is probably why the device is missing. [12:48] so you do need to fiddle with mdadm probbly [12:48] probably [12:49] ok, thats new to me, i've never heard of mdadm [12:49] http://ubuntuforums.org/showthread.php?t=556108 [12:50] does this sound familiar? [12:50] did you read those 3 posts? [12:51] here's a nice article on how this is done: http://beginlinux.com/server_training/server-managment-topics/1000-raid-1-on-ubuntu-804 [12:51] yes I read them, there is no solution there [12:51] but he talks about a similar problem to yours [12:51] ya, so am i creating a new raid array? [12:51] to replace md0? [12:51] creating the array wont format the disks at all right? [12:51] not a new one [12:52] you want to preserve the old one right? [12:52] yes, only mke2fs will erase the device [12:52] so you should be safe [12:52] If you can, always make a copy of the files [12:52] this way you have something to go back to [12:52] well i thought that was the point of raid [12:53] i have 2 x 500gb disks in that array [12:53] now the array was borked so i loose all that data? :P [12:53] kind of sounds pointless to run those disks in raid, might as well have a TB of storage :P [12:53] like i can't even mount sda or sdb by themselves... [12:53] so they're useless right now... [12:55] http://pastebin.com/m35b78e72 [12:55] oops, take a look at that [12:56] am i doing something wrong? im trying to create md1 with sda and sdb [12:56] so its saying its already part of an array [13:06] ok [13:06] i got the array started with mdadm [13:07] Skaag::: now fdisk reports.... [13:07] Disk /dev/md0 doesn't contain a valid partition table [13:07] so is my data gone? [13:07] or whats my next step, just create an ext3 partition and mount /dev/md0? [13:12] never mind [13:51] New bug: #380197 in openvpn (universe) "pkcs12 doesn't work in jaunty" [Undecided,New] https://launchpad.net/bugs/380197 === asac_ is now known as asac [14:38] Hi [14:38] What's Mark Shuttlewords emailaddress? [14:41] Jeeves_, its stored on launchpad [14:42] Ah yes, i'll try to look it up there [14:42] launchpad is slow :) [15:08] do you guys think i need swap for dom0 that runs nfs server only (8gib ram)? [15:48] ivoks, heya master how's it going [15:48] good, you? [15:49] ivoks, good too :) [15:49] ivoks, do you have time to help me out in a couple questions? [15:49] i do, but not much [15:49] couple of minutes [15:50] ivoks, ok that would be enough.. How do I know if I should do updates? [15:50] updates? [15:50] what kind of updates? [15:51] ivoks, updating the package to a newer upstream version [15:51] there is debian import freeze [15:51] https://wiki.ubuntu.com/DebianImportFreeze [15:53] ivoks, so after Debian Import Freeze, I can start working on updates? [15:53] ? [15:53] are you talking about mergers from debian? [15:53] or something else? [15:54] oh, from upstream [15:54] ivoks, no. Something else. For example, filling a bug such as: Please update ipvsadm to new upstream version 1.25 [15:54] basicaly, debian is our upstream [15:54] but for packages which we care about more than others [15:54] those can be done directly from upstream [15:55] all updates should be done before feature freeze [15:55] after feature freeze, only bugfixes are going in [15:55] ivoks, right, so I should just pick those packages that I want them to be in the latest upstream version and update them? [15:56] yes, but they won't go in unless there is a strong reason [15:56] ivoks, for example this package: https://bugs.launchpad.net/ubuntu/+source/ipvsadm/+bug/375485 [15:56] Launchpad bug 375485 in ipvsadm "[needs-packaging] Please upgrade ipvsadm to 1.25" [Wishlist,New] [15:57] is it packaged for debian? [15:58] ivoks, in debian is: 1:1.24-2.1 [15:59] so, that's a no [15:59] you could work on that, if you want [16:00] ivoks, yep i'm on it already :) so what is the workflow in an update?? [16:00] but note that debian might package it before our debian import freeze, so it still can get in trough merge/sync process [16:00] version should be ..-0ubuntu1 [16:00] so that when debian creates ...-1, we could merge it [16:01] anyway, this is for #ubuntu-motu [16:01] not here [17:42] how do i delete raid devices that i created incorrectly so that i can re-configure them properly? [18:25] Does ubuntu-server have support for SAN over fiberoptic cards? [18:33] on 7.10 server i have a small problem... in /etc/sysctl.conf i have uncommented net.ipv4.conf.default.forwarding=1 to enable routing. However when I cat /proc/sys/net/ipv4/ip_forward 0 is returned. Is there something that I missed in 7.10 when it comes to doing routing? [18:37] how do i delete raid devices that i created incorrectly so that i can re-configure them properly? [18:50] Hi, I'm working on a transparent proxy server in bridge mode. It intercept all http request and forwards to squid. I'd like to do the same thing for DNS requests and then keep the result in cache for others clients. Is there a way to do that ? [18:50] W8TAH, i had to delete them with mdadm if i can remember correctly [18:51] W8TAH, after u have mucked it up it's a bit tough to get the tui (text user interface) to install it correctly again. [18:51] W8TAH, in the end I had to do it the command line way to fix it [18:51] or use other clean drives [18:52] ok -- [18:52] ugh [18:52] LOL [18:52] i'll see if i cna go via command line to run fdisk [18:52] worst comes to worst i'll grab a gentoo boot disk and use that to kill the partitions [18:52] and start over [18:53] Does ubuntu-server have support for SAN over fiberoptic cards (like Q-logic) ? [19:02] Do any of you know where to find a port of openssh-server 5.1 for 8.04? [19:39] I have a load balancer that does icmp, http and traffic checking. I am not responsible for the load balancer, but right after the LB comes my squid proxy and then the LAN. I am havint intermitent connection to the internet on the lan. When I remove the proxy internet seems normall. I dont know which logs to check since I cant see anything unusuall in messages and in syslog. Has anyone had simmilar problems? [19:39] one thing [19:39] it seems to be when the logs are rotated [19:42] does anyone have problems with logrotation and squid ? does anyone experience an internet connection hang? [19:42] rsr: do dyou use squid log rotate options? or logrotate? [19:43] jmedina when I installed squid through apt-get it seemed to have set up a system itself [19:43] jmedina I havent set any logrotate system so I assume the ubuntu packages do by default [19:47] does anyone know how I can disable squids autorotate funcion on ubuntu server? [19:48] 8.10 [19:54] rsr: just edit squid.conf and put rotate option to 0 [19:56] jmedina : strange... it is already set to 0 [19:56] jmedina : it defaults 0 [19:56] jmedina So some other tool must be doing it [19:56] ill be right back [20:05] Hi, I'm working on a transparent proxy server in bridge mode. It intercept all http request and forwards to squid. I'd like to do the same thing for DNS requests and then keep the result in cache for others clients. Is there a way to do that ? [20:11] Max007, it's called dns forwarding [20:12] i think bind does it [20:12] it's very simple to do [20:13] AnRkey: I need it to be fully transparent. I can't change any configuration on the lan computers or servers [20:13] AnRkey: The bridge must intercept all dns queries and forward them to dnsmasq [20:13] or bind [20:13] 1sec [20:16] Max007, i would set up a dns proxy/cache [20:16] a quick google finds many howtos for that [20:17] the problem.... [20:17] I tried dnsproxy with dnsmasq [20:17] but I'm not sure it works.. [20:17] getting iptables to do the rest for you transparently by redirecting all requests through the bridge for dns to the cache/proxy [20:18] iptables -t nat -A PREROUTING -i br0 -p udp --dport 53 -j REDIRECT --to-port 53000 [20:18] iptables -t nat -A PREROUTING -i br0 -p tcp --dport 53 -j REDIRECT --to-port 53000 [20:18] dnsproxy's using port 53000 [20:18] on a client pc use dig www.google.com for a test [20:18] hold on, i'll pastebin the results [20:18] ok that's fine, then iptables must nor forward all outbound port 53 traffic to that server's port 53000 [20:20] Max007, i know what u need to do, i just need an hour to get my son to sleep, he's bouncing up and down here [20:21] be back in a bit [20:21] will msg u [20:22] http://pastebin.com/m3e42e509 [20:22] ok [20:22] AnRkey: ok [20:52] Max007, sudo apt-get install dnsmasq [20:52] u'r done there [20:52] now u just need to mess around with your iptables [20:52] i'm looking around for u, brb [20:54] try something like this http://www.google.co.za/search?q=iptables+dns+redirect [20:55] you want iptables not to allow dns packets through the bridge, it should instead forward all traffic that comes in on the internal port to the dns server of your choice on whatever port [20:55] ok [20:55] let me try something [20:56] my techy gets it going with one line in his iptables rules file [20:56] i cant get into the box that it's set up on, soz it must be offline or something [20:59] ) [20:59] it almost work [21:05] Max007, http://advices.latkos.eu.org/advices-and-information/20/ [21:05] yeah that's what I tried [21:06] I think it work but not sure [21:07] http://pastebin.com/m66b9a2fa [21:07] look at that [21:07] hi all [21:07] I don't understand why ns1.reference.qc.ca is responding to 192.168.1.100 ... [21:07] what is wrong with this mdadm? http://dpaste.com/49158/ [21:09] and dnsmasq doesn't seem to cache the answer [21:10] i did the same nslookup again and 192.168.1.2 still ask 192.168.1.1 for the answer.. it should be in cache [21:10] yeah, that's not working at all [21:11] it's still going straight through to the dns server without being redirected [21:12] hmm [21:12] I can ask my techy 2mrw for u if you want, the kid's an iptables ninja [21:12] I think it's beeing redirected [21:13] default dns server on the windows pc is 66.46.75.162 but if you look at tcpdump output nothing is going to this server [21:13] no there are 3 hosts involved in your logs, there should only be two if it's working... am i nuts for saying this? [21:14] no [21:14] oohh, good point though on the ip not showing up in the dump [21:14] but if ns1.reference.qc.ca is responding to 192.168.1.100, there should be something going from 192.168.1.100 to ns1.reference.qc.ca [21:14] am i nuts ? :P [21:15] i'm reading through it again [21:15] :P [21:16] what are the two ip's of the bridge [21:16] only one ip [21:16] 192.168.1.2 [21:16] ok only on for the bridge interface then [21:16] yep [21:16] ok, 192.168.1.1 is the adsl router or something? [21:17] yes [21:17] linksys router [21:17] odd way to set it up but it should work fine [21:17] and default dns server for the bridge [21:18] ok then is the router set to cache dns requests? cause thats is what it looks like [21:18] no, i want the bridge to cache dns request [21:18] i know that, but is the router caching as well? [21:18] no [21:18] well... i dont think so [21:18] what is the bridge interface's gateway set to? [21:19] 192.168.1.1 [21:19] ok getting closer i think... [21:19] ok what iptables rules did u add, can u paste that to pastebin? [21:19] yep [21:20] brb, getting my lighter [21:20] http://pastebin.com/m22edc821 [21:20] ahhh, i finally get my sweet mary jane after a week of HELL [21:21] ok then that's working [21:21] it forwarded the request to your bridges gateway i think [21:22] and your adsl is caching the requests [21:22] the problem is that most adsl modems hang or give probs when u put large amounts of dns requests through to them from a large amount of workstations. [21:23] the adsl router is caching the requests, that is [21:23] AnRkey: there's no adsl [21:24] AnRkey: it's an E10 connection and there's only me on the network (it's a test environment) [21:24] ok then it's a cable router? [21:24] we love mary jane :) [21:25] AnRkey: linksys befsr41 router configured with static ip address [21:26] AnRkey: I'll check tcpdump on the dns server (ns1.reference.qc.ca) to see if the linksys router is sending a request to it [21:26] ok soz about that, we only have had adsl routers down here in south africa not used to other setups [21:27] i get the idea though [21:27] pc > bridge box > linksys router > cable modem? [21:28] you need to change your iptables rules slightly to include a destination address. make it the bridge's ip [21:28] make sure that dnsmasq is listening on all ip's and interfaces in it's conf file [21:28] pc > bridge box > linksys router > cisco router [21:29] yep, linksys router is sending request to ns1.reference.qc.ca [21:29] dnsmasq's listening on all ip [21:29] so it's just the iptables now [21:29] ok then u'r a breath away from getting it working [21:29] adjust that rule a bit and it should work [21:30] I think it's not possible to set an ip to REDIRECT target [21:31] no it most dfntly is [21:31] we use it on at least 4 sites [21:31] ok [21:32] lets google it [21:33] one box even inspects packets for all sorts of nasties. It's at a local law firm, their internet connection to their mail server and so on. [21:33] google rox [21:33] it's how i found my pipe :) [21:33] i love my pipe [21:33] hehehe [21:33] ok i dont know how much more use i'm going to be :D [21:34] lol [21:34] looking around for u anyway, but if your traffic all gets redirected to redtube.com then it's not my fault u didnt check my ip tables rule before using it :P [21:35] haha [21:35] ahhh, found the problem... that linksys only has a windows xp and vista logo in the compatibility section of the places site [21:36] i wonder how many people buy based on what logo is on the box [21:37] REDIRECT [21:37] This target is only valid in the nat table, in the PREROUTING and OUTPUT chains, and user-defined chains which are only called from those chains. [21:37] It redirects the packet to the machine itself by changing the destination IP to the primary address of the incoming interface (locally-generated [21:37] packets are mapped to the 127.0.0.1 address). [21:37] If I understand that right, it redirect to the local ip [21:38] ok but then what is your bridges dns server set to? [21:38] cat /etc/resolv.conf [21:39] 192.168.1.1 [21:39] i'm looking in my mikrotik router for clues about the iptables rule, i use a dns trap here too [21:39] Hi. I installed LAMP on Ubuntu 9.04 via tasksel, but it's not parsing my php pages - just giving them for download. How can I make it interpret them? [21:40] i dont think it's caching then, just forwarding [21:40] your box is doing the request for the client it seems [21:40] AnRkey: that's what I think too.. no caching [21:41] however a test request from a second pc should use a cached request if dnsmasq is working [21:41] 1.100 sending dns request to his dns server, bridge intercept it, send to 192.168.1.1 [21:41] so to set the ip of the parent dns server change the /etc/resolv.conf file's entry [21:41] but there's no caching [21:41] now u just need to switch to bind instead of dnsmasq [21:42] i cant think now but are you sure that dnsmasq caches? [21:42] but i dont understand why ns1.reference.qc.ca respond to 192.168.1.100.... [21:42] AnRkey: yes dnsmasq is a caching dns server [21:42] yeah that got me too [21:42] but it doesn't seem to cache right now [21:43] i know that using bind as a dns cache is super easy [21:43] there's nothing going from 192.168.1.100 to ns1.reference.qc.ca... It's weird that ns1 respond to 1.100 [21:43] try bind quick [21:43] AnRkey: really ? I didn't know bind can cache [21:43] lemme see quick [21:45] here from my google notebook [21:45] http://www.zytrax.com/books/dns/ch6/#caching [21:46] bind is bullet proof [21:46] if done some crazy shiz with it [21:46] i've* [21:47] Hi. I installed LAMP on Ubuntu 9.04 via tasksel, but it's not parsing my php pages - just giving them for download. How can I make it interpret them? [21:47] i think it,s normal that ns1.refe... respond to 192.168.1.100... 1.100 is asking ns1.ref so it waits a response from ns1, not the bridge box [21:48] vadi21: might have to a2enmod php (to enable the phpmodule) [21:48] or the module might be called php4 or php5 [21:48] vadi21, sudo apt-get remove php5-common --purge and then sudo apt-get install php5 again [21:48] tadah [21:48] a2en says those don't exist, going to try reinstall [21:49] vadi21, u do a net install? [21:49] tasksel [21:49] the bridge box must change the "from" ip address to fit what the windows box is waiting for (sorry for bad english... lol) [21:49] not a net install, no [21:49] tasksel? [21:49] what about it? [21:49] I used tasksel to install it [21:49] ok i need to read :P [21:50] but no, it's still giving them for download [21:50] a2enmod says php5 is enabled. [21:51] apt-get clean ; apt-get remove php5-common --purge -y ; apt-get install php5 ; a2enmod php5 [21:52] vadi21, of course... tasksel... just never used it directly, but the ubuntu installer does when i do my net installs, that's why i asked about it [21:52] afraid the same thing. [21:53] I was trying cherokee before, but removed it since. Wonder if that messed anything up [21:53] vadi21, u basicaly need to get php5 installed properly and then apache must have it loaded [21:53] what do the apache logs say? [21:53] pastebin them [21:54] which one? error.log? [21:54] the last 50 or so lines after a restart of apache [21:55] error: http://paste.pocoo.org/show/119817/ access: http://paste.pocoo.org/show/119818/ [21:55] ack, that's not access [21:55] http://paste.pocoo.org/show/119819/ is [21:55] AnRkey: got to go. Thanks a lot for the help ! Can I leave you my email address in case you ask your tech guy ? [23:32] hi [23:41] my roundcube database is messed up, and I don't know how to fix it [23:42] for webmail === stiv2k_ is now known as stiv2k