[00:31] sick of wireless, I'm just going to buy a pair of these "Belkin 1GBPS Powerline 2 Networking Adapters", even with the Wifi connected using connection manager, it drops loads, okay, maybe if I get another dongle it'll be okay, but I can't be bothered with the messing around anymore :) and 1GBPS is a lot better anyway :P [00:34] RandomCake: network-manager is evil [00:36] yup twb, I'm sure it'll all be much better when I can get on the network without the thing :) [00:36] anyway, it's late, thanks for the help, night! [00:36] http://paste.debian.net/142520/ [00:36] RandomCake: ^^ [00:37] I'll have a go tomorrow twb, thanks :) hopefully I'll have some luck! [01:17] anybody here with experience with cobbler and images? [01:23] !anybody [01:23] A high percentage of the first questions asked in this channel start with "Does anyone/anybody..." Why not ask your next question (the real one) and find out? See also !details, !gq, and !poll. [01:32] twb: what's up? [01:32] ninjix: NFI [01:33] ;) [03:01] Does anyone know a good application to monitor response time for pages on web servers [03:01] ? [03:01] curl [03:01] twb: I was thinking something more along the lines to monitor them constantly and create graphs and all that [03:06] jetole: serverdensity.com pingdom.com etc. [03:07] PandorBox: thanks but looking for something I can run myself since I want to monitor individual web servers and not the load balancer [03:07] (also, serverdensity is giving out t-shirts & USB men for free at the moment, lol) [03:09] "USB men" ? [03:10] Reminds me of that guy in #emacs who wrote an elisp interface for his vibrator [03:10] yeah I was wondering about that too. Just figured it was like a girls blow up doll that plugs into the usb port on their computer [03:10] jetole: http://luneta.sourceforge.net/ [03:10] lol @ #emacs [03:10] PandorBox: neat [03:11] PandorBox: what language is that? [03:11] twb: http://i.imgur.com/BLiEj.jpg - pull his head off and he turns into a 1gb flash disk [03:11] jetole: PHP [03:11] (I only suggested it because it came up on freshmeat the other day) [03:11] PHP must die [03:11] no I mean what spoken language is that? Mexican? [03:11] jetole: hahaha [03:12] jetole: .mx speaks es, silly [03:12] :) [03:12] kidding aside though I can't read/write/speak/understand spanish [03:12] jetole: a friend is working on something related at the moment - http://python-log-probe.readthedocs.org/en/latest/screenshots.html# [03:12] with the exception of common swearing [03:12] jetole: it's either es or pt [03:12] wtf is pt? [03:13] Looks like pt to me [03:13] Portguese [03:13] oh [03:13] same thing to me [03:13] They're very similar -- probably more similar than English and German [03:14] I understand spanish and portguese as well as chinese, russian, aramaic, etc [03:14] 0.o you understand aramaic? [03:14] ITYM Mandarin [03:14] PandorBox: I understand aramaic as well as I understand chinese and russian, yes [03:14] so I'm trying to install off this ppa https://launchpad.net/%7Emarceloshima/+archive/ifolder and I get file not found w/ apt-get update off the mirror....I'm running lucid....any ideas? [03:15] Ik hou van polyglotten? [03:15] bokmal? [03:15] actually I think chinese is very similar to english. I get stuff made in china all the time and it all says "made in china" which seems very similar to how we would say manufactured in china in english [03:15] twb: dutch [03:15] jetole: I think that's enough whiskey for you [03:16] I suck at northern europe [03:16] twb: I'll take that under advisement but mr. Crown Royal is making a very compelling argument [03:16] twb: I'm doing Rosetta Stone Dutch course to help a friend out [03:17] so I'm thinking in regards to my original question about monitoring http latency, smokeping maybe? [03:17] PandorBox: is that some Ubuntu thing? [03:17] [03:17] PandorBox: I asked because lp has some translation thing called rosetta IIRC [03:17] anyone have any thoughts on smokeping to monitor http latency? [03:18] speaking of rosetta yeah my dad bought a learn mexican rosetta package [03:18] CD's and guides and all that bunched into a package [03:18] * PandorBox hasn't used smokeping [03:19] from what I am reading it can create graphs from curl output and I vaguely remember someone once telling me to use curl for this [03:19] IMO httping would be a better tool for the job [03:20] * jetole checks it out [03:20] btw, I got my nagios plugin published today :D http://exchange.nagios.org/directory/Plugins/Databases/SQLServer/MSSQL-Job-History/details [03:21] jetole: ... why not just use nagios then? [03:21] PandorBox: I want to measure response times so I can see when servers are slow to respond etc [03:24] if you're open to generating the data yourself and already have a nagios install - httping or siege [03:24] the movie with denzel washington and bruce willis? [03:24] oh wait that's The Siege [03:25] * jetole googles siege [03:25] though I probably dont want to generate it myself [03:25] if smokeping can do it then good enough I guess [03:25] :) === ivacau_ is now known as ivacau [06:42] I have a setup which involves 1 frontend apache proxy with 5 backend apache servers that actually hold the applications. Clients have FTP servers on the backend servers and have been connecting directly to those till now. I am looking for a ftp proxy solution so that i can route each client to proper backend while putting all backend servers in DMZ and clients only connecting to frontend apache box. [07:26] New bug: #885998 in facter (main) "facter upgrade crashes puppet" [Undecided,New] https://launchpad.net/bugs/885998 [07:54] I have a setup which involves 1 frontend apache proxy with 5 backend apache servers that actually hold the applications. Clients have FTP servers on the backend servers and have been connecting directly to those till now. I am looking for a ftp proxy solution so that i can route each client to proper backend while putting all backend servers in DMZ and clients only connecting to frontend [07:54] apache box. [07:58] I have a quick question about the Ubuntu EC2 images; when I right-click and choose connect, AWS suggests I connect with ubuntu as the username. However, when I try to spin up an Ubuntu image that I've edited, AWS suggests root as the username, even though you need to use ubuntu to connect. Is this an Ubuntu-problem or AWS-problem? === zz_jhelwig is now known as jhelwig [08:50] http://extras.ubuntu.com oneiric release signatures cannot be verified because NO public KEY how to fix this issue [09:00] runasand: it's a aws problem with assuming one size fits all. most ubuntu ec2 images use ubuntu as the default user some bitnami images use bitnami as the default user... etc etc [09:04] airtonix: so I guess I should talk to some aws people, thanks :) === jhelwig is now known as zz_jhelwig [09:36] New bug: #811422 in apache2 (main) "Exploitable integer overflow on x86 in mod SetEnvIf, leading to buffer overwrite" [Low,Incomplete] https://launchpad.net/bugs/811422 [10:16] Is there still an alternate download for AMD64+mac ? I want to install 64bit. [10:24] I've set up an Ubuntu box as a router between two networks. 192.168.199.0/24 is on eth0 and 192.168.111.0/24 is on eth1. I've enabled ip forwarding and I can ping between the networks. The 199 network is effectively the WAN and there's a route to 192.168.199.253 which is the gateway to the internet. My question - when I enable NAT with iptables -t nat --append POSTROUTING -o eth0 -j MASQUERADE everything dies. [10:24] Either iptables -t nat --append POSTROUTING -o eth0 -j MASQUERADE is wrong or I'm missing a step which Google isn't helping me find. [10:25] Can anyone point me to some instructions for doing this or suggest what I've done wrong? [10:29] stetho: which guide are you following? [10:31] Loads. There's 5 in the ubuntu forums alone, blog posts, server fault, howtoforge. They're all pretty much the same - the only thing about them is they deal with routing to an ISP (so eth0 is external in the proper sense of the word - in my set up it's still a private network). [10:51] New bug: #886048 in samba (main) "“check password script” option does not reliably pass parameters" [Undecided,New] https://launchpad.net/bugs/886048 [11:27] morning all [11:37] I got some zombies showing up when I boot server - using the ps command, how can I tell which particular program/service/whatever is causing those zombies ? [11:39] SilfenX: lets have a look at them [11:39] SilfenX: look at the parent pid too [11:41] PID TTY STAT TIME COMMAND [11:41] 1 ? Ss 0:00 /sbin/init [11:42] SilfenX: are you saying init is zombied ? [11:42] or the parent [11:46] https://bugs.launchpad.net/ubuntu/+source/libxml2/+bug/147144 quite weird that i can reproduce this on 10.04 lts with latest updates. [11:46] Launchpad bug 147144 in libxml2 "xslt:copy element is broken in 2.6.29" [Medium,Fix released] [12:04] given that i'm running 10.04 lts, what practical options do i have for up- or downgrading that package to try out another version to see if the bug doesn't happen with those? === heidar_ is now known as heidar === jason is now known as Guest12115 === Guest12115 is now known as jasef === jason is now known as Guest1785 === Guest1785 is now known as jasef [13:34] how i can redirect http://domain.com:8888 to httpS://domain.com:8888 automatically ? [13:35] greetings all. which release was the first one to support lvm snapshots? [13:36] sorry snapshot merging [13:38] no idea ? [14:26] hi [14:30] hi, i am running sybase ASE using ubuntu. i am trying to get pam authentication going ... anyone here, who did this before? [14:31] i have a prob when running the ase process as non-root user, then pam gives me problems, specifically the error pam_unix(ase:auth): unix_chkpwd waitpid returned -1: No child processes [14:32] Does Ubuntu Server run reasonably well on an HP Microserver? [14:42] ball: after searching for a few HP microserver modells.. I don't see why it wouldn't [15:06] hi all, how can I install a base gnome for my ubuntu 10.04 server? [15:07] ersi: Thanks [15:08] i upgraded from 8.04 to 10.04 now grub fails saying ALERT! /dev/md1 doesn't exist [15:08] ball: No guarantees though :) [15:08] trying to troubleshoot it now but not getting very far, any help would be appreciated [15:30] dork: I recall something like that happening to me [15:30] trying to remember how I got past it [15:31] you're trying to boot from a mdadm mirror, correct? [15:32] right [15:32] i did a dist upgrade from ubuntu 8.04 to 10.04 [15:32] and it fails booting saying alert /dev/md1 not found [15:34] did you try using a live Ubuntu disk, apt-get install mdadm then mount the array and chroot in to make sure that your ramdisk has the modules? [15:35] ninjix: that's what i' trying to do now but i'm having probles with the chroot environment [15:37] ok, what are you chroot env issues? [15:38] ninjix: right now i need to recreate and mount it [15:38] recreate? the chroot or the array? [15:39] the array, chcking fstab to see which is which [15:42] i've got md1 assebled and mounted to /tmp/root [15:42] but this looks like the boot part [15:42] has the kernel/grub dir etc inside of it [15:49] yes, that's the boot [15:50] what other md do you see in /dev/mapper [15:50] ninjix: ive got it all set up now [15:50] nice [15:50] it wsa using the mdadm.conf from the live cd i had to get the uuid's via blkid [15:50] ok [15:50] so i've got it mounted [15:50] gonna chroot up now [15:51] ok chrooted up [15:51] are you familiar with mounting the proc, sys and devpts? [15:53] i did it on gentoo once [15:53] lol [15:55] sudo mount -o bind /proc /var/chroot/proc [15:55] ? [15:56] Daviey: I heard you planned to sell your charms but I'm not 100% sure how to interpret that. [15:59] ninjix: ? [16:09] I have a zombie process - how can I find out which program/service is causing it? I have run 'ps -1' but it only lists a bunch of numbers, not the actaul name of the program/service ... [16:10] hi all, I have Mobility Radeon HD 5000 Series video card, since hd output doesn't work, is there a driver to install? (vga works) [16:11] dork: hi, I'm back [16:11] accidentally over-taxed my workstation's IO [16:12] dork: did you get the chroot mounts setup? [16:13] I was going to recommend you do: (chroot) sudo mount -t proc proc /proc; sudo mount -t sysfs sys /sys; sudo mount -t devpts devpts /dev/pts [16:16] please can anyone help me? [16:35] ninjix: i'm chrooted in [16:35] ninjix: though i tried to install grub2 this time and i think i made things wors [16:35] ninjix: but i'm chrooted into it now [16:35] with /prov /dev mounted [16:37] you're moving out of my scope of experience. I haven't every tried upgrading grub >> grub2 in chroot [16:38] well it's a chroot environment [16:38] should be the sae right [16:38] so what can i do here any suggestions? [16:39] what is in the /etc/initramfs-tools/modules file? [16:41] also, I recommend making a backup of your /root and /boot [16:41] all comments [16:41] i have a backup [16:41] there are only comments in the modules file [16:41] ok [16:41] I have a zombie process - how can I find out which program/service is causing it? I have run 'ps -1' but it only lists a bunch of numbers, not the actaul name of the program/service ... [16:41] any orchestra/juju gurus here today? [16:42] you want to add some mdadm modules :) [16:42] raid0 raid10 raid456 multipath raid1 md-mod linear [16:42] each get their own line [16:43] then you want to do: sudo update-initramfs -k all -u [16:45] ok [16:45] updating [16:45] why did it remove those modules [16:48] hi! is it possible to, using orchestra, manage older ubuntu versions? [16:48] how old? [16:49] mgw, lucid (latest lts) [16:49] I believe so, using backports [16:49] that didn't work [16:49] assuming orchestra is in the backports for LTS, mgw [16:50] I think it is... [16:50] ninjix: that didn't work, could grub2 be causing issues now? [16:50] i a have a couple of lucid servers, but i am not able to find the ubuntu-orchestra-client package for them [16:50] but maybe i'm wrong :-) [16:50] dork: did the initramfs complete successfully? [16:50] mgw, fwiw, not much is actually *in* the backports. [16:50] they dont backport everything, iirc [16:50] ninjix: yes [16:51] ninjix: though the error is different now [16:51] that should have worked if grub behaving [16:51] mgw, lucid-backports has no package with orchestra :| [16:51] ninjix: now it says it can't find a dev w/ a uuid address on it as opposed to /dev/md1 [16:51] ok, it's not backports mndo [16:51] it's ppa [16:52] https://launchpad.net/~orchestra/+archive/ppa [16:52] fysics: wpa.conf? I thought most of that stuff was done in /etc/network/interfaces ... [16:52] dork: that sounds like grub2 is almost configured correctly [16:52] but no idea how or if it works [16:52] I"m having plenty of fun on oneiric [16:54] dork: if you setup your chroot again, you can use blkid /dev/mapper/md0 to get the UUID [16:54] smoser, roaksoax — either of you here? [16:55] ninjix: setting it up again now will have the uuid in 1m [16:56] my grub2 skills are survival level at best [16:57] ninjix: i'm assembling the md0 and md1 based on the output of blkid [16:57] that's right right? [16:57] it assembles/mounts find [16:57] fine [16:57] that sounds promising [16:59] ok chrooted back in [16:59] ninjix: i don't miond going back to grub1 just to get this box back online [16:59] this is impacting mail and dns [17:00] that might be the quickest route for a return to service [17:00] blkid /dev/mapper/md0 [17:00] ppa:orchestra/ppa? has only to packages for lucid and it does not even build :( [17:01] show no results [17:01] *only two packages [17:01] then you can compare your backup of /boot and get your server right side up :) [17:01] ninjix: ok im in now, blkid /dev/mapper/md0 shows nothing [17:01] ninjix: just delete grub2 froi the chroot? [17:02] I'd try apt-get remove [17:09] ninjix: i removed it but grub fails and it shows a uuid/gpt based address still [17:09] and not /dev/md1 etc [17:09] like it's supposed to [17:09] even though i removed grub2 [17:10] did you run update-grub after uninstalling grub2? [17:10] no [17:13] dork: here's a link to a tool that saved my bacon once [17:13] http://www.howopensource.com/2011/08/reinstall-recover-grub-from-ubuntu-live-cd-usb/ [17:14] it's a grub2 tool, though [17:14] yah [17:14] i have no cd's and no usb drives on me [17:14] and im in the iddle of the largest data center in the world [17:14] heh [17:14] but i will have a look [17:15] is this a physical box or a virt? [17:15] did update-grub run successfully from chroot? [17:16] you should see the old grub1 /boot/menu.lst date modified change [17:18] yeah [17:18] it changed [17:19] nothing failed from upgrade-grub [17:19] peek inside it and make sure it looks sane [17:19] there is no upgrade-grub actually [17:21] open this link [17:21] http://rominux.blogspot.com/2009/11/tutorial-reinstall-grub-1-and-grub-2-on.html [17:21] try those GRUB1 commands [17:21] those grub commands would break my stuff if that addressing is wrong... [17:22] yup [17:23] ... [17:23] ok so how do i make grub look for md1 instead of those uuids [17:23] because obviously that is wrong [17:24] I don't recall grub1 using UUID, I think that is a grub2 feature. Can some other ops admin confirm? [17:25] that's why I suspect you still have grub2 in the MBR [17:27] ok so how do i fix that [17:27] reinstall/setup grub1 [17:28] im a little bit worried because of this [17:28] Checking if "/boot/grub/stage1" exists... no [17:28] it sees it in /boot/grub/stage1 [17:29] might want to try asking over int #debian [17:30] might be a sys ops with more experience with restoring grub1 [17:31] I know you're in an outage state right now so I don't want to give you poor advice on how to proceed [17:31] the people in #debian chastise people for asking for ubuntu help [17:31] and for asking about everything else [17:31] this is insanity. [17:32] i can't believe an upgrade fucked me this bad [17:32] should be more current experience with backing out of grub2 back to grub1 since grub2 only became the Debian default recently with Squeeze [17:32] i've got it back to doing exactly what it was doing before [17:33] which is? [17:33] can't find md1 insted of that uuid address [17:37] dork: previously you stated that you couldn't find update-grub in the chroot [17:37] right [17:37] maybe you need to chroot again and apt-get uninsta/reinstall grub [17:37] i did [17:37] that package doesn't come with it [17:37] Checking if "/boot/grub/stage1" exists... no [17:37] Checking if "/boot/grub/stage1" exists... no [17:37] oops [17:40] then perhaps grub2 is your path [17:41] scroll down about halfway on this page: http://www.howtoforge.com/how-to-set-up-software-raid1-on-a-running-system-incl-grub2-configuration-ubuntu-10.04-p2 [17:41] it has some instructions that might help you [17:45] you don't really know eh? [17:46] :( I'm at my limit of knowledge of this [18:04] Hello guys, I'm using a PPTP VPN as a client, I want my external ip address (public one, ppp endpoint) to accept connections on port 80 and redirect it to my machine (internal ip), I've done this in the past but somehow it doesn't work now, any of you have a clue on what iptables rules I should put in? Thanks [18:24] this sucks [18:32] dork: have you tried anything new? [18:40] ninjix: people in grub are saying grub 1.98 is broken [18:40] and to upgrade to oneiric's 1.99 [18:41] i tried grabbing the .debs from the ubuntu repositories [18:41] but there were a bunch of dependency issues [18:41] yeah I can imagine [18:41] kind of screwed and my boss is pissed [18:41] what about going gub2 [18:41] i did that before, when i first ran into it [18:42] that's why i had grub2 on there [18:42] and grub2 failed [18:42] so i don't know what to do heh [18:42] and thep eople in #grub are idle [18:44] hmm... well you could start your services in the chroot and then take some time to research how to fix the boot loader [18:44] let you limp along until later this weekend [18:56] ninjix: i didn't think about that [18:56] ninjix: i wish i considedred starting these services in the chroot a long time ago [19:08] did you get your mail and dns services running in the chroot? [19:14] dork: any luck? [19:30] ninjix: still doing upgrades [19:41] does anyone have a tutorial or page explaining how to install an 11.10 cloud client without PXE? I wanted to get a demo going but I don't have access to the DHCP server === zz_jhelwig is now known as jhelwig [20:46] ghexsel, if you're talking about orchestra, the ubuntu server CD has it as a boot option [20:54] At my job we have a webserver running ubuntu server. Now we want to create a back up web server to help us keep constant up time and to back up our information just in case something happens. If I were to use clonezilla for instance and clone the server to another drive and then change the network config to broadcast a new IP would that cause any problems, not sure if the actual host name is used anywhere since it's diffe [20:54] rent host names, or is there an easier option for that. Also would a simple rsync cron job be the best option to make sure the data on both stay the same, we use trac so only one folder ever gets changed really, or is there a better option for that [20:56] rsync is good for files [20:56] you'd have to change the hostname on the copy [20:57] I don't know anything about trac, but if it uses a database, you'd be better off linking the database servers in some way than trying to use rsync for it [21:00] can anyone help me get grub working again? doing an upgrade from 8.04 to 10.04 has ruined my grub setup and now i can't get it working. i can chroot into the environment fine just can't get grub working [21:04] it never connects... we got some hardware and creaated a subnet [21:08] qman__ : sorry I walked away. Trac does use a database, SQLite I believe but don't take my word on it. What do you mean by linking the database servers? Also where is the hostname located at? [21:08] yuck, sqlite [21:08] the hostname is in /etc/hostname and /etc/hosts [21:08] it must be updated accordingly with the new settings [21:09] ok...thanks [21:09] my recommendation is, if at all possible, you should upgrade to a real database server and not use sqlite [21:09] i was looking for that file but couldn't find it. Also trac can use other databases but it's built to work with one primarily and it's really easy to set up with it [21:09] but I don't know if you can/how you would, as I don't know anything about trac [21:10] anyway, the idea here is [21:10] understandable. I will look into more. I know it can use MySQL so I may try and switch it over or something [21:10] you upgrade to a real database server, on both servers, then link the two together [21:10] so that both servers always have the current database [21:10] so it's two seperate database servers "sharing" a database [21:10] yes, updating each other in almost real-time [21:11] interesting...i will look into that then:) [21:11] mysql can't go both directions [21:11] it can only have a master and slave [21:11] but other servers may be able to [21:11] well it would only need to be master slave [21:11] only reason we are doing this is primarily just in case of hardware failure or power outage [21:11] then a mysql master/slave should work fine [21:12] pretty much nothing will be done with the back up server b/c theoritically the master server should be up and functional again [21:12] k...thank you [21:12] you can take the slave server, change one setting in the configuration, reload mysql, and it's now the master [21:12] if you script it, it'd be about a three second failover process [21:13] the slave server has the caveat that it's read-only [21:13] so to be fully functional you have to 'flip the switch', as it were [21:13] but you'd be up to date, no lag [21:13] ok...I just looked it up and it is SQLite but trac also supports PostgreSQL and MySQL [21:14] I don't know if postgresql offers a better solution here or not [21:14] I only know mysql fairly well, but I also know that it's really lacking compared to others in most serious DBMS areas [21:14] Ok. I'll do some research and decided what to do later [21:15] i also have a friend that was a MySQL developer so I'll ask him for his opinion when he is available [21:19] Hey, stupid question: I want to be able to give somebody a ftp link so they can view it in their browsers, but with my shared hosting they have to log in as "anonymous@mydomain.com", not just "anonymous" like most browsers send. What link can I give them to send the username? [21:31] is vmbuilder still maintained? [21:32] rich_: I do hope so - is there reason for doubt? [21:32] rich_: soory, I misread - fwbuilder - no idea about your question === zz_jhelwig is now known as jhelwig [21:35] Hi, I'm running 10.04 and have what looks like hardware trouble from a new raid array. one drive throws errors every few days; then the array goes Read-only. warm boot won't fix it. Cold boot does. It's a remote box. ... [21:36] ... I have a firmware update for the drives, and have applied it to two of the three drives in the raid array. The third drive does not appear in /dev [21:36] Can you suggest a way to get that drive back, and apply the firmware update? Without suggesting reboot - warmboot has failed previously. [21:37] there really isn't any way [21:37] if it's not there, it's not there [21:37] check dmesg to get more detail [21:37] qman__ won't come back until BIOS tries to find it? [21:38] you can hot plug sata, and the kernel will rescan if it detects a hotplug event [21:38] but again, that's hardware level [21:38] you can probably initiate a rescan but it won't be there if the hardware doesn't see it [21:40] hardware issues really need to be sorted out before the box goes remote [21:52] guntbert most people don't. I think vmbuilder used to be the preferred way for a lot of people to build virts. using something called jeos, but it's hard to find any recent activity on the project or recent docs. [21:53] rich_: I see ... [21:54] qman__: +1 for intermittent hardware problems being sorted out before a box goes remote. :-) === tlocalhos is now known as localhost [22:37] alright [22:37] last stab here [22:37] i upgraded from 8.04 to 10.04 [22:38] after that, apparently the kernel couldn't map out my software raid /boot and / partitions [22:38] so it starts to boot the file system etc but it fails trying to mount [22:38] any ideas on how i can resolve this i've been working on this for 12 hours [22:42] grub1-2 migration is buggy as hell [22:44] dork: try to boot on an install cd with grub2, such as lucid, and reinstall grub on the disks in question