[00:00] as much as i read 1 will be the master and the other 2 are storage [00:28] is there an official way to make a file like index.html point to another file say zenlive.html ? [00:29] madwill: several ways [00:29] madwill: the most flexible way is to use mod_rewrite [00:29] madwill: or, another simple one, just use a symlink (and turn on FollowSymlinks === medberry is now known as med_out [00:37] How do I add kernel boot parameters? [01:17] tsrk: best is to add them in /etc/default/grub, then run update-grub [01:19] (add them to GRUB_CMDLINE_LINUX) [01:35] hallyn: Ok, thank you. What's the difference between ..._LINUX and ..._LINUX_DEFAULT? [01:42] g [01:42] er [01:42] oops [01:42] wrong win [01:43] (Ubuntu 11.04) I don't know if this is exactly a "server" issue, but I started with the server install. I then installed lxde on top, but every time I try to change the background on the lxsession, I get a segfault from X: http://pastebin.com/Gc9Jct0m [01:52] oh no, tsrk has skeedaddled [01:53] sjm: yeah best to go to #ubuntu-desktop with that [01:54] hallyn, ok, thanks. [02:04] hello i need help in this cfgmaker --output=/etc/mrtg/mrtg.cfg public@192.168.10.1 [02:04] have set my ip from router [02:04] i am setting mrtg in ubuntu server [02:06] wow. eyeos2 seems to have come a long way [02:07] think i am going to have to it up soon :) [02:07] i just hate that it requires openoffice in order to export to odt [02:07] well.. od* [02:13] Macer: eyeos? [02:13] habriel: you may prefer something newer, like collectd or munin or cacti [02:13] habriel: mrtg is pretty old-school [02:14] twb: web based "OS" [02:14] Oh, one of those DEs in js [02:14] well.. more of a UI than OS.. supposed to be open source cloud computing [02:14] yeah [02:14] Bugger that shite [02:14] they did quite a bit of work on it since i last tried it [02:14] twb: i don't know it seems to work fairly well :) [02:15] it is good for idiot proofing the way employees use their workstations heh [02:15] zimbra was actually supposed to be good for that but wow has it just gotten worse and worse [02:16] i think of all the things i have tried to figure out a web based solution for employees.. zimbra had to be about the biggest let down i have ever seen [02:16] especially nowadays.. where they can't seem to get things to work right.. like the Instant Messaging.. so their solution was to simply eliminate it [02:16] i hear they are removing their docs section too [02:16] twb so you recommended cacti [02:17] so i suppose it is just a glorified imap server with a web ui? :) [02:22] habriel: no because it's PHP [02:22] habriel: I definitely recommend collectd [02:23] Macer: the main things that non-free exchange clones provide you is calendaring and an outlook connector (outlook is crap at IMAP). [02:23] i am looking a tool monitoring easy [02:24] Macer: unfortunately they also like to reinvent the distro in /opt [02:24] twb: well zimbra concentrated on fixing their connectors than actually turning into something that was actually useful [02:24] Macer: I would recommend first looking at apple calendarserver (which is FOSS and in Ubuntu already), and replacing outlook with tbird. [02:24] which was a shame because it had so so so much potential.. i was a huge fan up until right before 7.0 [02:25] twb: i would never use zimbra again [02:25] I've rolled out zimbra and scalix a few times, and I would never ever do so voluntarily [02:25] not in the direction it is going now [02:25] Right! [02:25] heh [02:25] it seriuosly could have given exchange a run for its money if it didn't cnocentrate so much on trying to get it to work with exchange and only worked on its own thing as an alternative to exchange [02:25] but they wanted to bite into the ms customers [02:26] twb: do you have any guide for set collectd in ubuntu sever [02:26] which i think was a poor business decision. they should have pushed their web interface and its ability to IM, file share, etc all in one spot.. along with fixing their docs [02:27] instead they are eliminating im and docs... and i mean.. i don't see what good it is anymore as a "collaboration suite" without its ability to actually communicate and share documents. :) [02:28] it was supposed to be a one stop interface requiring nothing but a web browser and even a slower small powered client .. it could have been amazing.. i would have used it for everything and pushed it onto everyone i came across .. but now they totally suck [02:28] habriel: I don't, but basically you install collectd-core (without rrdtool) on your spokes, and collectd-core (with rrdtool) on your hub. [02:28] habriel: then you instruct the spokes to send to the hub, and the hub to accept info from spokes [02:28] so i am looking for alternatives and eyeos seems the closest to what zimbra should have been... :) need to set it up tomorrow and give it a real test run on my ubuntu server [02:29] habriel: then you instruct them as to what information to collect [02:29] if i can figure out how to get it to properly export od* then i am happy [02:29] habriel: finally, you set up collection3 or some other visualization, so you can browse to a web page and see graphs of the data [02:30] thanks twb i will try [03:01] Is there a default dhcp server included in ubuntuserver, if so where can I find it? [03:02] !info isc-dhcp-server [03:02] isc-dhcp-server (source: isc-dhcp): ISC DHCP server for automatic IP address assignment. In component main, is optional. Version 4.1.1-P1-15ubuntu9 (natty), package size 392 kB, installed size 980 kB [03:03] Calif: Is not installed by default. [03:03] I installed natty, does that mean I have it? [03:03] ok [03:03] crap [03:04] I gotta do something real quick ill be right back to ask some more questions about tihs... namely how to do I get it? And further is there a simple way to derive the download link from something without gui, in cli? [03:04] Calif: (most) services are not installed by default. [03:04] Calif: try running "tasksel" [03:04] Calif: you should also read the Ubuntu Server Guide [03:05] Brb sorry.. gotta step away for one min [03:05] brb,.., [04:30] hi guys i have a domain name pointed to my server public ip adn i wanted it somehow to be mydomain name for myweb server using publci ip any idea how do i do it with Vhost [04:33] More of an Apache question (see #httpd), but... [04:34] Are you trying to run multiple sites with different domain names on the same running instance of httpd? [07:00] how can i tell multipathd to recheck volume sizes? [07:41] hi RoAkSoAx === smb` is now known as smb [08:30] hi guys... im using mysqldump to backup some databases on my ubuntu server.... can I view some logs to ensure if the backup went through without any issues??.. please advise! [08:40] twister004, you mean automated via cronjob? [08:42] dc5ala...yes.. can I do a --log-error=filename? [08:43] twister004, in your crontab you can put a line like MAILTO=twister004... [08:44] dc5ala... what can I mail?.... which log file? [08:44] twister004, you will get the output of the command via mail [08:44] twister004, or you just redirect output of mysqldump into a file via ">" yourself when you run it [08:45] or better, append it with ">>" [08:47] dc5ala... can I just do a "sqldump -u -p --opt --skip-lock-tables --databases info_schema > /mnt/.../test.sql >> test.txt"??? [08:48] ops, sorry, think i did mistake, let me see :) [08:48] hi [08:49] twister004, i just see there is a --log-error=name parameter for mysqldump, may give that a try [08:50] dc5ala.. i tried that.. i get the following error "mysqldump: unknown variable 'log-error=test.txt'" [08:50] looks like mysql maybe an old version [08:50] can i know the history of iotop application ? [08:51] output of 'mysqldump -V' is "mysqldump Ver 10.10 Distrib 5.0.24a, for pc-linux-gnu (i486)" [08:51] you did prepend "--" ? [08:51] yes [08:52] i only checked the parameter list of mysqldump on a server of mine, says mysqldump Ver 10.11 Distrib 5.0.51a [08:52] dc5ala... will >> work? [08:53] yup.. so it's a version problem.. mine's old... I dont wanna update mysql coz this is production environment [08:53] have a look in the help if there is an log-error parameter [08:53] nop [08:53] ot's not there [08:53] it's* [08:53] that one is old server too :) [08:54] maybe there is a generic logging parameter? [08:54] nothing... only flush-log [08:54] nothing to do with logging error [08:55] willl the mysql.log or mysql.err file under /var/log/ help? [08:55] i have a look on another machine, what version is running there [08:55] how can I make changes in /etc/security/limits.conf take effect without restarting the server? [08:55] I need to increase the max allowed file-descriptors [08:57] dc5ala.. there's another machine with sql version "mysqldump Ver 10.11 Distrib 5.0.51a, for debian-linux-gnu (i486)"... hee I see the '--log-error' parameter [08:57] it's a version problem [08:58] here** [08:58] hehe, guess so [08:59] you may try redirect "stderr" maybe, e.g. mysqldump ... > out.sql 2>>error.log [09:01] twister004, the logs under /var/log/mysql are probably only used by the server and not by msqldump, not sure [09:02] dc5ala... 2>>error.log doesn't seem to be working [09:03] how comes? :) [09:04] if mysqldump writes anything to stderr it should end in that file [09:05] thanks [09:05] dc5ala.... 2>>error.log is working [09:05] !!! [09:05] thank you!!! [09:05] atleast it's getting redirected to a log file. [09:05] could you force and error to test it? [09:09] twister, oh, just see another thing i used in a script, mysqldump ... > db-$(date +%Y%m%d).sql [09:09] twister004, puts a nice timestamp in your filename [10:35] does "sudo more /proc/sys/fs/file-nr" give me the total used file descriptors on my server? [10:58] hi all [10:58] jdstrand, i have reported the bug as you suggested. [11:39] hi, does somebody knows how i can use clusterssh with ssh agent forward? [12:11] New bug: #777022 in mysql-5.1 (main) "package mysql-server-5.1 5.1.54-1ubuntu4 failed to install/upgrade: sub-processo script post-installation instalado retornou estado de saída de erro 1" [Undecided,New] https://launchpad.net/bugs/777022 === cloakable_ is now known as cloakable [12:36] New bug: #777033 in postfix (main) "installation lacks RELEASE_NOTES-2.6 documentation file" [Undecided,New] https://launchpad.net/bugs/777033 [12:50] how can i tell multipathd to recheck volume sizes? [13:10] koolhead11: thanks! [14:11] ttx: around? [14:11] ttx: just wondered how etherpad had held up at the OpenStack summit? [14:13] jamespage: it didnt blow up :) [14:14] zul: thats good :-) [14:14] thinks are looking OK for using it at UDS as well [14:14] jamespage: \o/ [14:26] New bug: #777096 in bind9 (main) "package bind9 1:9.7.0.dfsg.P1-1ubuntu0.1 failed to install/upgrade: el subproceso instalado el script post-installation devolvió el código de salida de error 1" [Undecided,New] https://launchpad.net/bugs/777096 [14:36] New bug: #777103 in groovy (main) "Merge groovy (1.7.10-1) from debian unstable" [Medium,In progress] https://launchpad.net/bugs/777103 [14:41] on 10.04 when do the rc2.d/S* scripts start, before, after or durint the stuff in upstart? === paul__ is now known as orudie === med_out is now known as medberry [15:02] New bug: #776299 in apache2 (main) "package apache2-mpm-prefork (not installed) failed to install/upgrade: subprocess dpkg-deb --control returned error exit status 2" [Undecided,New] https://launchpad.net/bugs/776299 === jjohansen is now known as jj-afk [16:30] Hello, I've got Gigabyte GA-H67MA-UD2H-B3 motherboard and have 2 disks configured as RAID 1 from BIOS. When I boot Ubuntu Server 11.04 64 bit on my computer it cannot find the disks. I manually go to Detect Disks step and system prompts me to select Driver. Can you tell me what driver should I select? [16:43] kirkland: ping [16:43] kirkland: is the dell laptop we tested powernap back in January a 4 core? [16:44] RoAkSoAx: nope, dual core [16:44] Vanuatoo, I believe that's "fakeraid" [16:45] Vanuatoo, dmraid package is what you need [16:46] New bug: #777193 in ethtool (main) "ethtool settings may not apply on boot with separate /usr" [Undecided,New] https://launchpad.net/bugs/777193 [16:46] kirkland: k thanks [16:46] Vanuatoo, caution, I'm not aware of many folks who use fake raid in production environments. Mostly it's used as a compatibility bridge for users who are dual booting from an existing Windows installation. [16:47] There is no fake raid in the list [16:48] Vanuatoo, that's the slang name, see https://help.ubuntu.com/community/FakeRaidHowto [16:48] SpamapS, the ethtool bug above is mine. thought you might be interested from a boot perspective. [16:49] Hello, I am looking for any documention or information regarding Ubuntu server MPIO. Can anyone here point me in the right direction or give me any useful advice? [16:49] Vanuatoo, I don't know if dmraid support ever made it into the installer. [16:51] Vanuatoo, wish I could give you more direct assistance though I don't have any first hand experience settiing up these fakeraid deployments. I've been avoiding them like the plauge, MD is a more robust and better supported option [16:51] What is MD? [16:52] For some reason I get this error in my apache log: "[Wed May 04 17:51:36 2011] [error] [client 192.168.10.105] File does not exist: /var/www/~rt", any way to narrow it down to the file I've stated that in [16:52] Vanuatoo, Linux's defacto software raid stack, managed by mdadm [16:58] Defghanistan, MPIO deployments are SAN specific, what are you trying to deploy? No, we don't have Ubuntu centric MPIO docs but the general available ones for other distros are applicable, for the most part. [17:00] We're working with a netapp fibre channel SAN, deploying ubuntu server to fulfill web and database roles === jj-afk is now known as jjohansen [17:08] hi all [17:50] quick question, guys.. if a server has just apache2-mpm-prefork installed is it an apache server? or would that be only machines with apache2 installed? [17:50] dumb question I know [17:52] Hey guys... any opinions on replacing dash with bash? I don't mean in general, I have a customer who wants to do that on their systems. [17:52] actually they asked specifically about what upstart uses for script stanza's. but i assume the more relevant question is what you link /bin/sh to. [18:13] hi all, I'm setting up a bridge I've configured br0 I'll also still have to configure one of my real interfaces also for example eth0 right? [18:21] New bug: #777261 in librdmacm (main) "Sync librdmacm 1.0.14.1-1 (main) from Debian unstable (main)" [Undecided,New] https://launchpad.net/bugs/777261 [18:22] to change an IP do I just edit /etc/network/interfaces and restart networking? [18:23] That's what the doc's say but this is a remote machine and I'd hate for it to not come back up [18:23] smoser: I *know* you already told me, but I am afraid I forgot: how does one boot a karmic EC2 kernel? With --kernel? And where are the kernel lists? [19:08] Hello, I installed "linux-scst" which installed linux-image-2.6.32-29-scst on my ubuntu 10.04 server, however when the system boots its selecting 2.6.32-31-server. I do not see the scst kernel in my grub configuration either. Any ideas on how I can boot the system to the scst kernel? Thanks [19:14] Hello, can someone help me with bridging connections..when i type ifconfig i get listed only primary device [19:16] Nikolasr, use brctl [19:17] yes..but when i type eth0 its says invalid device ignoring.. [19:17] :) [19:18] man brctl :) [19:21] i mean when i type brctl addif "primary" "eth0" (without quotes) [19:22] I got notting. [19:22] Please help me. [19:36] RoAkSoAx: ping [19:37] kirkland: pong [19:38] RoAkSoAx: i have a small but significant fix to powernap's load monitor [19:38] RoAkSoAx: i'm wondering if you mind doing the sru? [19:38] kirkland: sure [19:39] kirkland: just commit it to lp:powernap and I'll grab it from there [19:40] kirkland: or better yet, commit your fix, release a new version in oneiric, and I'll SRU your fix [19:40] RoAkSoAx: yeah, that's what i was thinking [19:40] RoAkSoAx: here's the fix i'm testing: http://paste.ubuntu.com/603409/ [19:41] RoAkSoAx: basically, if the load threshold is "n", we need to count the number of processors every time we run [19:41] RoAkSoAx: because if we're powernapping, and we're down to 1 processor, then we want to wake if load gets above 1 [19:42] RoAkSoAx: rather than wait until load gets to 16 (if we have 16 processors) [19:42] RoAkSoAx: i don't like the commands.getoutput("getconf _NPROCESSORS_ONLN") every time this runs though [19:42] RoAkSoAx: i want to do it more cheaply than that [19:43] RoAkSoAx: http://stackoverflow.com/questions/1006289/how-to-find-out-the-number-of-cpus-in-python [19:43] kirkland: yeah makes sense to me all of it [19:43] kirkland: though will importing the multiprocessing module be really cheaper? [19:44] RoAkSoAx: hmm [19:45] kirkland: and may be subject of errors [19:45] RoAkSoAx: hmm [19:45] RoAkSoAx: let me test [19:48] kirkland: let's just try it out for now, I don't think it will really make any difference [19:49] RoAkSoAx: my testing says you're wrong [19:49] :-) [19:49] kirkland@x201:/tmp$ time python 1.py [19:49] real 0m0.834s [19:49] kirkland@x201:/tmp$ time python 2.py [19:49] real 0m16.143s [19:49] 1.py: http://paste.ubuntu.com/603416/ [19:49] 2.py: http://paste.ubuntu.com/603417/ [19:49] :-) [19:50] kirkland: cool then! Let's just use multiprocessing then :) [19:55] hi all, I'm trying to point virt-install to my cdrom drive but -c /cdrom isn't working how should this be done? [19:56] ghostlines: this works in Natty [19:56] not before [19:56] at least it shouyld [19:58] hmmm, otherwise I must make an image of it? [19:58] ghostlines: and you might have to make the cdrom available through HTTP [20:00] damn, seems like I have to jump through hoops to use a local device [20:00] ghostlines: the thing is that virt-install does not correctly determine that a cdrom is an ubuntu source/distro [20:00] ghostlines: so this was fixed patched in Natty [20:01] ahhh [20:01] ghostlines: however, I've only tested this when making the loop mounted cdrom available through HTTP, not through /cdrom === ahs3 is now known as ahs3-lunch [20:02] I could swear that virt-manager allowed me to select devices after a vm was installed, real strange that I can't do it with virt-install [20:03] thanks for your help though, think I'm just gunna try and make an image of the dvd [20:03] ghostlines: virt-manager uses libvirt while virt-install uses python-virtinst AFAIK [20:04] ubuntu's docs needs to get updates as well the python-virtinst pkg doesn't exist in 10.10 it's just virinst [20:05] ghostlines: upstream is python-virtinst in ubuntu it is just virtinst [20:06] RoAkSoAx: we should also use os.getloadavg()[0] [20:06] RoAkSoAx: it's over twice as fast as reading /proc/loadavg [20:07] RoAkSoAx, and since when did /dev/cdrom move to /cdrom ? I gotta really keep up [20:07] kirkland: cool then. I guess we'll have to look throughout the code for tricks like this [20:07] ghostlines: I don't know actually [20:07] RoAkSoAx: yeah [20:08] RoAkSoAx: well, it might be my fault after all :-) [20:09] RoAkSoAx: i'm listed as the author of that file :-P [20:09] kirkland: hehe nah well neither you nor are/were python experts to know all the tricks and functions python already provides [20:09] hehe [20:13] kirkland: btw.. take a look at it, and tell me what you think: https://help.ubuntu.com/community/Cobbler [20:21] New bug: #777356 in postfix (main) "[fixed upstream, regression] Postfix smtp does not skip unknown XFORWARD attributes, causing SMTP syntax errors" [Undecided,New] https://launchpad.net/bugs/777356 [20:31] jamespage: re:etherpad -- no problem at all [20:32] jamespage: though I'd admit that not everyone was using it, and that we had only 2 sessions in parallel [20:43] kirkland: just thought of something. the idea of scale the threshold should only work when using the script that turns off the cores right? [20:43] kirkland: never mind it doens't really matter [20:43] RoAkSoAx: nah [20:45] RoAkSoAx: okay, my load monitor fixes work great [20:45] RoAkSoAx: i'm about to release powernap 2.7 [20:45] RoAkSoAx: are you happy with the other features you've merged for oneiric? [20:45] kirkland: yes, please release it [20:45] RoAkSoAx: done [20:45] kirkland: i was planning to switch to source format 3.0 but that can be left for 2.8 [20:46] as it is not high priority right now [20:46] RoAkSoAx: hmm, when you do, you'll need to fix a bug somewhere, either in bikeshed:release-build or somewhere else [20:46] kirkland: I've attached a debdiff for natty, could you please sponsor it? [20:46] kirkland: how so? [20:46] RoAkSoAx: as bzr bd fails for me when i do source format 3.0 and try to release-build [20:46] RoAkSoAx: you'll see :-) [20:47] kirkland: hehe ok [20:48] RoAkSoAx: are you doing the SRU justification or am i? [20:48] kirkland: just did it :) [20:48] sweet [20:56] hggdh, karmic images were never moved to pv-grub. [20:57] so you can't really boot a karmic ec2 kernel other than the one associated with the image. [21:03] eyeos 2.x was bust.. 1.9 is awesome [21:03] hi here [21:03] anybody willing to help me with phpmyadmin problem [21:04] i'm running fcgi as parser [21:11] kees: mk-sbuild oneiric is failing for me, bc awk isn't being installed so dpkg -i basefiles fails. Is this anything you know about offhand? [21:16] kirkland: so release-build should not do the first bzr bd? [21:16] RoAkSoAx: hmm [21:17] RoAkSoAx: well, that's always worked before [21:17] but, yeah [21:17] okay [21:17] RoAkSoAx: it needs to move that bzr bd to the end [21:17] RoAkSoAx: after everything else [21:17] RoAkSoAx: and it'll build the -0ubuntu1.deb [21:19] kirkland: ok cool! [21:19] RoAkSoAx: uploaded powernap_2.6-0ubuntu2.1_source.changes [21:19] kirkland: thanks [21:22] smoser: any chance of generating a karmic ec2 kernel for the -29? === smb` is now known as smb [22:00] Daviey: what do you think about dropping https://blueprints.launchpad.net/ubuntu/+spec/server-o-kvm-document-suggested-changes ? [22:04] zul: this is funny - the scheduler is working *so* hard to keep me out of convenient afternoon sessions (i have none in the last two hours) that i'm double-booked for the lxc session [22:05] hallyn: doh...talk to robbie [22:06] zul: yeah the one that's 3am central time may have quite a few ppl attending remotely from US so that one should be moved if possible [22:06] what you dont wake up that early? [22:15] zul: i may still be awake at that point, but unsure [22:18] hallyn: Not coming to UDS? [22:18] soren: alas, not this time [22:19] That was getting really annoying. [22:20] hallyn: But you'll miss out on all the Hungarian... err... stuff! [22:20] * soren can't think of anything uniquely Hungarian [22:22] soren: not to mention all the uds camaraderie. i'd love to be going. c'est la vie. [22:27] anyone know if libvirt supports CoW cloning of qcow2 based VMs via virt-manager? [22:42] hi people [22:43] anyone up here to stress their brain ? [23:23] soren: umm...hungarian gulash? === ahs3-lunch is now known as ahs3 [23:56] zul: Hungarian gulash is yummy