=== [1]Ivan2-Ian is now known as Ivan2-Ian [00:16] In rsyslogd, "$RepeatedMsgReduction off" prevents lines like "Oct 20 11:09:45 plum last message repeated 2 times" [00:16] What is the equivalent on hardy, in sysklogd? [00:40] hallyn: the gettext.sh crud shouldnt pop up now [00:49] is there a possibility of live migration of kvm vms running on lvm? [00:51] New bug: #878597 in rabbitmq-server (main) "rabbitmq-server fails to uninstall" [Undecided,New] https://launchpad.net/bugs/878597 === koolhead17|zzZZZ is now known as koolhead17 [00:57] not with kvm; libvirt might be able to do something clever, but I wouldn't bet on it [01:02] New bug: #878600 in rabbitmq-server (main) "'service start rabbitmq-server' does not fully detach from parent" [Undecided,New] https://launchpad.net/bugs/878600 [01:11] hello gays [01:11] m4v say me this is a gay channel [01:11] ubuntu sucks [01:11] ubuntu is the sith of debian [01:11] jajajaja [01:11] ubuntu is windows [01:12] m4v is #ubuntu-es [01:12] is the op ^ [01:22] hi [01:27] synflag: you are a m4v lover? [01:27] yes [01:27] ubuntu server is the sith of ubuntu, and ubuntu is the sith of siths [01:27] no me creo que todos estos users piensen que ubuntu server sirve de algo [01:28] hi to all [01:28] i have a trouble in ubuntu server, apache2 no work fine :S [01:29] maybe for the LOT of bug in ubuntu? [01:29] i don't know [01:29] best choice, RHEL clon, CentOS, scientific linux =) [01:29] Unity_wow: so use them [01:29] Ubuntu sucks in all banches [01:29] yeap twb [01:30] ubuntu == shit [01:30] What's an m4v when it's at home? [01:30] Wikipedia seems to think it's an iTunes wurble [01:31] twb: m4v say this channel is gay [01:31] m4v is a nickname [01:31] -NickServ- Information on m4v (account m4v): [01:31] -NickServ- Registered : Jul 16 14:04:39 2007 (4 years, 13 weeks, 5 days, 11:27:02 ago) [01:31] -NickServ- Last seen : now [01:32] m4v is op of #ubuntu-es [01:32] I see. [01:32] is the @operator of ubuntu-es [01:32] i can say ubuntu sucks like a windows [01:34] m4v tell me here is gay party [01:34] i have a big 8===================0D [01:34] who want? [01:34] open your ubuntuer´s asshole ( )*( ) === Unityx_ is now known as Fbdev === Fbdev is now known as Dr_Juan0 [01:52] for some reason, I can start libvirtd with the -d and the -l flag manually and it works fine, but I can't start it with libvirt-bin with the -l flag... I don't know why [01:52] anyone have any suggestions? === Dr_Juan0 is now known as hax0rbuntu [02:02] anyone actually here? [02:31] question about ubuntu server 11.04 [02:31] tried installing gui [02:32] but after that i rebooted and tried to ssh back in and was unable to reconnect [02:33] was hoping maybe you guys could help me figure out what i am doing wrong here exactly [02:33] installing a gui [02:33] i knew that was coming [02:33] hear, hear [02:33] aviandroid: do you have physical access? [02:33] but seriously, you probably installed network manager, which probably broke your networking configuration [02:33] qman__: ah, good catch [02:33] no just remote access twb [02:33] NM needs to be taken out back and shot [02:34] you are probably right qman__ [02:34] un canal en español para server please [02:34] is there something i should do beforehand [02:35] before installing the gui that is [02:35] alguien habla español [02:35] if you must, use --no-install-recommends and only grab the stuff you need [02:35] but in all seirousness [02:35] i sshed in and did and update then proceeded to install gui probably not the best idea i am sure [02:35] if you only have remote access, why do it? [02:35] alguien habla español [02:36] un canal de ubuntu-server en español [02:36] Solo #ubuntu-es [02:36] ? [02:36] oh okay i saw that command somewhere does it allow you to go down a list and select which packages you need [02:36] no [02:36] it simply does not install recommended packages, and only installs the dependencies to what you choose [02:37] you need to look at the list of what it's installing and skim for network manager [02:37] deter, busca en un canal de debian server puede que te ayuden [02:37] i guess qman__ because i am just not that savvy yet using the cli or terminal for everything [02:37] everything is done on the terminal anyway, there are no GUI tools to manage server functions [02:38] and you cannot get to the GUI without something like VNC, which is complicated to set up on a headless machine [02:38] X0rfbserver isn't that complicated, it's just difficult for the kind of people inexperienced enough to WANT it [02:39] GUIs belong on client machines, you can open multiple SSH sessions and use your local web browser for looking up/testing things [02:39] qman__, iwas using remmina from ubuntu desktop to connect [02:39] cant you just ssh -X and run x apps to your local display? [02:39] or ssh -XC ideally [02:40] unless you're running something like LTSP, there really is no point in doing so [02:41] you have to run the same commands either way [02:42] qman__, maybe i can try using the stripped down version you mentioned earlier no-install-recommends [02:43] i could try and use it without a gui but i foresee a lot of banging of the head on the desk [02:44] as I said, the GUI does not change anything that you actually have to do [02:44] the difference is whether you're using a direct SSH or a gnome-terminal over VNC [02:44] what about using i saw mentioned called webmin perhaps [02:44] !webmin [02:44] webmin is no longer supported in Debian and Ubuntu. It is not compatible with the way that Ubuntu packages handle configuration files, and is likely to cause unexpected issues with your system. [02:45] oic [02:47] it depends on your goals and needs, but if you really intend to administer a server, you need to get your head around it sooner or later [02:47] better now than when things actually depend on it [02:48] you are right i just was expecting to jump in head first [02:48] was just trying to put a foot in the water i guess you could say and then ease on in [02:48] and you can, there's tons of step by step documentation [02:49] the command line interface is the standard, it's what everything uses [02:49] qman__, i hate to say it but i might be asking a lot of questions [02:55] qman__, should i at least update to 11.10 [02:55] that depends [02:56] are you on a VPS or some other non-bare-metal system? [02:57] if so, avoid unless there's documentation from your provider on how [02:57] if you're on bare metal, you shouldn't have issues, but there's always the risk it won't come back online [02:58] it is a dedi [02:59] should i begin by installing fluxbox or maybe openbox to help assist [02:59] no [02:59] okay [02:59] that's getting into slightly more advanced configuration even for X servesr [03:00] servers* [03:00] if anything stick to mainstream ones, like xfce or lxde, but still, it's not needed at all [03:00] anything you'd do in a GUI should be done on the client end [03:00] web browsing and such [03:02] okay well what should i do starting out [03:02] the box is clean right now [03:02] what are you planning to run [03:04] well, regardless, you'll have to get familiar with some basic commands and an editor [03:04] web server vmware linux training transmission [03:04] nano does the job but I use vim, and lots of people use emacs [03:04] i have used vim and emacs before but very limited [03:04] both of the latter aren't exactly pick up and use, you have to do some reading [03:05] oh yes definitely [03:05] well, LAMP is pretty well covered in the server guide [03:05] yeah i noticed that as being pretty popular [03:05] vmware isn't supported here, KVM is the supported virtualization option [03:06] KVM that is right i knew that sorry [03:07] that's also covered in some detail, with the libvirt route, in the server guide [03:07] it doesn't cover everything, but it's a really good starting point [03:09] libvirt what is that exactly [03:10] an interface to virtualization [03:10] meant to be generic but it's mostly used with KVM [03:11] oh okay i see [03:20] i see if you do an upgrade and lose ssh connection there is another daemon started as backup [03:24] qman__, i very much appreciate your help [03:30] Okay, so I thought I knew, but clearly not: how do you set a user's primary group? [03:31] usermod -g [03:34] qman__, are there any good book publishers or videos you recommend reading watching [03:34] should i use the ubuntuforums.org [03:34] can't say I do, oreilly is generally good but most publications are redhat centric [03:35] ubuntu forums are a good place for support [03:35] thank you sure do appreciate it [03:35] aviandroid: the Ubuntu server guide; it's in /topic [03:42] twb, thank you === Amgine is now known as BadTypoGod [03:57] qman__: Thanks. aviandroid: read the ubuntu server guide and actually follow the steps (don't just read it and skim over). If you come to a part you don't completely understand, try using manpages, askubuntu.com, ubuntuforums.org and here to get answers. The best way to learn is to dive in, and make an effort to learn things completely as possible. Good luck! [04:00] Tohuw, I appreciate your advice going down that path as we speak [04:03] I have a new UPS. It has both serial and USB; which should I use? [04:03] aviandroid: Excellent. :) Enjoy! Much of it is actually quite rewarding and even intuitive as you start to learn it. Except for LDAP, because LDAP was invented by Satan to torture the living. [04:03] Tohuw: krb is worse [04:04] A homogeneous openssl-backed openldap environment isn't too bad [04:04] twb: The larger part of my job is managing active directory deployments of enterprise software, usually for law firms who hire beavis, butthead, ren or stimpy to set their infrastructure. I feel these pains. [04:04] If you bring AD or samba or FDS or domino into the mix, then [04:05] Tohuw, linux has restored my love for computing that i just about lost [04:05] aviandroid: you'll get over it [04:05] As an exercise, I created a secondary domain controller using samba and ldap. Then I decided: never again. [04:06] twb, you are probably right but don't be a killjoy i am on a high right now [04:06] aviandroid: Good. Linux restored my love for computers as well. You will have your frustrating moments that stretch for days, but there's a silver lining: you are using a capable kernel and a configurable system. [04:07] Tohuw, you are absolutely right [04:08] well guys it has been a pleasure talking to you all but i better get going [04:08] look forward to talk to you guys again soon [04:09] have a good one [04:15] Poor guy. He's all excited about his server, but he hasn't tried setting up virtual users in postfix yet. [04:15] There will be tears [04:16] In before correction: dovecot-postfix [04:16] dovecot-postfix tried to rape me when I installed it [04:16] So I just set up the equivalent five lines of config myself [04:18] Yes, Your Honor, Mr. dovecot-postfix did try to touch me... I'll show you on the doll where. [04:18] Incidentally, what are you using as your backend for virtual users in dovecot and postfix? LDAP? A database? Moon rocks? [04:19] I don't have any virtual users at this time [04:19] Some of my poor custoemrs are using ZCS because they have stupid calendar requirements [04:19] zcs? Zimbra? [04:19] yea [04:19] Poor souls. [04:19] Beats scalix or sogo [04:20] At least zcs monkey-patches *postfix* [04:21] It would be tremendous if someone could create a cohesive, easy to deploy open-source mail server that can seriously compete with Exchange. But I'd also like a million dollars and an unending supply of Scotch, so we have to pick our battles. [04:23] Actually I'd prefer to use just apple calendarserver (which is python foss) [04:23] AFAICT the rest of the job is already handled correctly by postfix and dovecot [04:26] Not even close. Exchange has no real open competitor in Enterprise due to its integration capabilities, unified communications and deployment. It's unfortunate but true. [04:26] For small-scale mail and calendar usage, yes, there are options. [04:28] hi, in /etc/hosts, 127.0.0.1 WEBSITENAME localhost WEBSITENAME localhost ? [04:33] Well, I have edited /etc/hosts and /etc/resolv.conf and I don't remember what I changed exactly but it resulted me unable to localhost... :S :( [04:34] if anyone decided to help me in this, I'll be very appreciated :-) [04:36] atdprhs: post your hosts and resolv.conf to a pastebin. [04:36] atdprhs: and what do you mean "unable to localhost"? Pinging localhost returns nothing? [04:40] okayz, give me 5 minute please [04:52] Tohuw, pastebin.com/YexQvKnB [04:57] Tohuw, are you here? [04:58] atdprhs: That pastebin is empty. [04:58] Also, you didn't answer my question: [04:58] pastebin.com/YexQvknB [04:59] atdprhs: and what do you mean "unable to localhost"? Pinging localhost returns nothing? [04:59] I get replies [04:59] but opening the webpage [04:59] is not working [04:59] That is not a DNS issue. [04:59] If you are pinging localhost and getting replies from localhost, the DNS is working, for all intents and purposes. [05:00] hmm [05:02] Tohuw: uh, localhost isn't (normally) resolved with DNS [05:02] Tohuw: it typically resolves via the files/compat nss method, because it's in /etc/hosts [05:03] Tohuw: re exchange integration -- do you mean integration with outlook o,r what? [05:07] I looked around Tohuw and still I don't know [05:08] it was working till I restarted the server few days ago [05:08] and it was the first restart to do after my last edit to the file [05:08] but then suddenly, i can't open the localhost after [05:08] so that why I suggested it could be one of them [05:10] here is an error when I restart apache2 pastebin.com/3WLgN3Dc [05:11] atdprhs: so fix apache config first [05:12] twb: I was being lazy and using "DNS" to refer to name resolution in general. I'll admit that's confusing. And integration, I'm referring to AD integration and unified messaging integration and third party integration and all the other mountains open competitors would have to climb. I need sleep now. Guten Nacht. atdprhs, as twb said, fix Apache, that is your current problem. [05:14] file /etc/apache2/conf.d/fqdn has nothing but egyptianprojects.com localhost [05:15] atdprhs: apache seems to think otherwise [05:15] atdprhs: try grep -r'ing over /etc/apache2 [05:15] r'ing? [05:16] grep -r [05:16] grep -r /etc/apache2 [05:17] right? [05:18] I'm sorry for my stupid question but everytime I come here, I learn a lot from all of you, to be honest, I'm very thankful for this channels to be existing :-) [05:19] I'm not going to teach you to use basic unix tools; there are other people who can do that. [05:21] I know, twb, I walked around and did my own on the server, I configured it to be web server almost 1 year ago, and I didn't touch it for long except for one thing that really bothered me, but then I came and tried it and walked around a lot of websites till I lost hope then I left it the way it was and then I didn't do anything but till I had to restart once and then no localhost... I am new to Ubuntu-Server and I love it, I was using in the past [05:21] windows server, but when I seen ubuntu, I liked the way it is [05:22] so that's why I transferred my work to Ubuntu and started PHP applications, and leaving microsoft's products [05:25] I tried both grep -r /etc/apache2, and it pauses, then I tried grep -r /etc/apache2/apache2.conf, and it pauses... [05:25] I have to press ctrl+c to get out [05:25] because it does nothing [05:28] well, thank you twb [05:45] twb, do you know what this could be resulted from? [05:45] so i can research for it [05:47] twb it listens to 80 and 443 and 443 [05:49] <_ruben> atdprhs: you're missing a rather important parameter to grep: the string you want to grep for... [05:50] ?? [05:51] grep -r listen /etc/apache2 [05:51] <_ruben> `grep foo bar` search for `foo` in a file called `bar` ... doing `grep foo` will search for `foo` on stdin [05:51] <_ruben> might wanna add a -i as well for case insensitive searching [05:53] grep -r a -i listen /etc/apache2 [05:53] _ruben, you know what problem I am having? [05:54] <_ruben> i only read a small part of the backscroll, so no ;) [05:55] well, i remember i edited /etc/resolv.conf and /etc/hosts long time ago, but didn't restart, back later when I restarted, and tried opening localhost, it doesn't work [05:55] when I restart my apache2, it gives me the following error [05:56] pastebin.com/tPj7F39M [05:57] I am sure I didn't edit the configuration of apache2 [05:57] only those two files I told you [05:58] <_ruben> pastebin the output of cat /etc/apache2/conf.d/fqdn and grep -r fqdn /etc/apache2/ [06:00] pastebin.com/hhv8NDsm [06:02] here is pastebin for the two files I edited: pastebin.com/bL3g2cN0 [06:03] <_ruben> i'd just move the fqdn file to some place "safe" and see it restarting apache works .. i got no clue as to what that file is supposed to accomplish, sure aint valid apache2 config lines [06:03] <_ruben> eww .. notworkmanager [06:04] "It looks like you're trying to set up a bridge! Would you like me to delete all interfaces?" [06:04] I configured at the very beginning the server to egyptianprojects.egyptianprojects.com [06:05] twb, are you talking to me? [06:05] _ruben, do I move or copy the file to somewhere safe? [06:05] atdprhs: no [06:06] <_ruben> atdprhs: move (or consider the pastebin your backup ;)) [06:06] alright, done [06:07] warned namevirtualhost *: 80 has no virtualhosts [06:07] <_ruben> warnings aren't as critical as errors :) [06:07] yup [06:07] :-) [06:07] <_ruben> tho it's a fairly common one tho [06:09] ok [06:09] what is it with fqn? [06:09] fqdn? [06:09] localhost doesn't run even without it [06:10] <_ruben> define "doesn't run" .. can't connect? or what? [06:10] not found [06:11] <_ruben> using a browser to test i guess? [06:11] yes [06:12] <_ruben> what does 'curl -v http://loclahost/' say ? [06:13] spelt correctly [06:13] <_ruben> that'd help, yes ;) [06:14] i'm restarting the server [06:16] i'm instaling curl [06:17] connection refused at 1.... [06:17] and refused again at 192.168.1.2 [06:17] and refused again at 127.0.0.1 [06:17] (That server's ip is 192.168.1.2) [06:18] hello there [06:18] anyone here ? [06:18] <_ruben> atdprhs: when restarting apache2, you only get the virtualhosts warning? no errors? [06:18] yes [06:18] only warnings [06:19] I tried again [warn] NameVirtualHost *:80 has no VirtualHosts [06:20] I'm sure I edited the two files wrong, my worse mistake, I lost their back-up... [06:22] <_ruben> try: sudo lsof -iTCP:80 -sTCP:LISTEN [06:23] no TCP state names in: TCP [06:25] hi atdprhs [06:25] anyone know about this command ? [06:25] lshw |grep disk [06:26] I read the resolv.conf, the website is hosted locally on the server, and I'm pionting out the domain to my server, and the server local ip is 192.168.1.2 [06:27] in resolv.conf search websitename nameserver localhost ? [06:27] what it does exactly ? [06:27] <_ruben> iuytfr: it lists most of your hardware and then shows only the lines that have 'disk' in it [06:28] I'm sorry iuytfr, what I know about greb is that it searches.. [06:28] Anybody used nut with USB before? The drivers are yelling at me: Can't claim USB device [ffff:0000]: could not detach kernel driver from interface 0: Operation not permitted [06:28] iuytfr: man lshw [06:28] How do I tell who has a lock on the USB device? [06:28] perfect [06:28] iuytfr: it lists hardware. [06:28] now my question is [06:28] <_ruben> atdprhs: do you run a nameserver (like dnsmasq/bind/etc) on that box? [06:28] by doing lshw |grep disk [06:28] Google suggested hal was to blame, but hal isn't on my box [06:29] i got this : *-display [06:29] *-disk:0 [06:29] *-disk:1 [06:29] *-disk:2 [06:29] *-disk:3 [06:29] *-disk:4 UNCLAIMED [06:29] *-disk:5 UNCLAIMED [06:29] and my question is why i got 2 disk unclaimed ? *-disk:4 UNCLAIMED [06:30] iuytfr: please don't paste into the channel. see the guidelines linked in the topic. there is a pastebin. [06:30] <_ruben> atdprhs: also, /etc/resolv.conf shouldn't containt any website specific stuff or anything, it's server global config file to control how dns lookups are handled [06:30] ok [06:31] do you understand my question ? what mean unclaimed for disk [06:31] do you have something with your hardware ? [06:31] samething [06:31] iuytfr: I'm not talking about your problem [06:31] so I should only leave nameserver 192.168.1.1 and let it take dns automatically? [06:32] or leave it blank? [06:33] iuytfr: i'm not sure grepping is really the best strategy. try reading the output of lshw indented under each of those -disk lines. [06:33] or leave it blank? [06:33] <_ruben> atdprhs: it should point to whatever the ip address of your nameserver is [06:34] alright, now curl replies with html and web [06:34] but not found page [06:34] i have the samething even w/o greping [06:34] unclaimed what exactly it means with disk ? [06:36] i don't know, and neither does the manpage. [06:36] okayz, hosts [06:36] LocalServerIP websitename localhost ? [06:36] iuytfr: http://ezix.org/project/wiki/HardwareLiSter has some info [06:37] like 192.168.1.2 is the ip for the server _ruben, that I already have the problem with and it is the same server that hosts the website [06:38] <_ruben> "192.168.1.2 websitename" "127.0.0.1 localhost" .. on seperate lines .. or alternatively "127.0.0.1 localhost websitename" [06:41] alright, "192.168.1.2 websitename" "127.0.0.1 localhost" [06:42] how can I echo my server's name? [06:42] o [06:42] h [06:42] nevermind [06:43] I did exactly as you said _ruben [06:44] why putting fqdn makes it gives error [06:45] when restarting [06:45] why invalid command? [06:45] <_ruben> because the contents of the fqdn file isn't valid apache config syntax [06:45] always has been that way since I started the server [06:45] Problem solved -- the driver setuid(nut)'s by default [06:48] in resolv.conf [06:48] "nameserver localhost" "nameserver 192.168.1.2"? [06:51] okayz [06:51] I have virtualhost already has the directory and everything set right [06:51] why not found? [06:52] Good Night All [06:56] how do you get ftpmail installed? [06:57] strickly: since AFAICT it's not part of Ubuntu: you don't. [06:58] ah ok because I wasn't finding it anywhere [06:58] and nothing came up thru google about it [06:58] only the man pages [07:03] _ruben [07:03] are you here? [07:04] well, # a node is marked as UNCLAIMED if no specific support for it has been loaded (or lshw has been unable to identify the driver) [07:04] rcsheets: [07:04] iuytfr: cool. that's what it means then. [07:04] what is the consequence on server ? [07:05] can i reinstall apache2 from scratch? [07:05] i suppose that you can't use those disks, unless/until you load a driver for them? [07:05] <_ruben> atdprhs: i am, sortof, trying to get some work done as well [07:06] <_ruben> atdprhs: sudo apt-get remove --purge apache2 && sudo apt-get install apache2 [07:06] but the problem is the server is in production w/o any problem rcsheets [07:06] i can't understand why i got this error [07:06] iuytfr: using all of the disks, including the ones that are "UNCLAIMED"? [07:06] yesx [07:06] what was the error? i think i missed that. [07:07] *-disk:4 UNCLAIMED [07:08] *-disk:5 UNCLAIMED [07:08] thanks _ruben, I'm unstalling and installing now [07:08] well, right, but... you just decided to run lshw one day and noticed that? [07:08] or was there something that led to this? [07:08] hopefully it works [07:08] yes correct rcsheets [07:08] iuytfr: perhaps there's actually no problem then. [07:09] _ruben, any suggested ideas? [07:10] well i thing i understand [07:11] 4 disk on server with raid 1 [07:11] the kernel load only in case if there is a problem on disk [07:11] am i right ? [07:17] _ruben, I enabled GUI, and I installed teamviewer, would you like to please take a look? === smb` is now known as smb [07:33] iuytfr: i don't know. i suppose it would depend on how you've got the raid set up. [07:34] yes i m still confused [07:34] why it gives same for memory [07:34] in the same way i try to grep memory [07:35] same error [07:35] unclaimed [07:48] I'm tired of this problem :( [08:17] morning all [08:35] I did it!!! [08:35] I DID IT!! [08:35] :D [08:36] I reconfigured apache2 and the server from the beginning :D [08:37] * josePhoenix gives atdprhs a high-five [08:40] thanks josePhoenix :D :D *High five back :D [08:53] How do I get Ubuntu (11.04) to start apache on boot? I have to run it manually every time. There are no errors in the log file. [08:54] $ update-rc.d apache2 enable [08:56] HarryPanda rocks [08:56] * HarryPanda confused [08:57] thanks HarryPanda, I'll do that now. [09:07] morning o/ [09:11] HarryPanda: that worked, thanks a mil! [09:36] hi all!! [10:21] hey koolhead17 o/ [10:22] hey lynxman howdy!! :) [10:22] * koolhead17 bows to Daviey [10:22] hah [10:23] Daviey: someone looked at that bug and related it to another one!!1 keystone folks :) [10:26] koolhead17: I'm not certain it's an upstream bug TBH [10:27] Daviey: yeah its not. cos from source keystone was running smoothly :P [10:37] heh [10:37] Daviey: thanks for closing this https://bugs.launchpad.net/ubuntu/+source/openstack-dashboard/+bug/854732 :D [10:37] Launchpad bug 854732 in openstack-dashboard "Broken openstack-dashboard package on Oneiric due to unmet dependencies (python-django-openstack)" [Undecided,Fix released] [10:38] np [10:38] * koolhead17 will get some more karma power now!! :D [10:41] \o/ [11:27] New bug: #878780 in mysql-5.1 (main) "package mysql-server-5.1 5.1.49-1ubuntu8 failed to install/upgrade: ErrorMessage: subprocess new pre-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/878780 [11:27] how can i be sure that my file isn't corrupted during transfer? take eg. md5sum before and after transfer? [11:27] trapmax: you could try checking the md5sum before and after transfer. [11:40] return -E_BRAIN_FULL [11:40] RoyK: lol :) [11:41] * RoyK is attending puppet master training [11:44] nothing to do with Jim Henson then? [12:43] Hi all. I just ran some update on my ubuntu server, and my MySQL will no long restart. I did some research around the net, and the advice seems to be to reinstall MySQL. Is that going to completely wipe out all my data, or can I safely reinstall with data in tact? [12:44] morning [12:44] why believe random data on the internet? [12:44] why not fix the problem? [12:45] answer from many ubuntu related forums [12:45] but if you would like to help me debug, I'd be more than happy [12:45] but this is a prod. server, so I'm in a bit of a rush... Just need the fastest fix [12:46] you haven't even bothered to describe the issue yet [12:46] and you want fast? [12:46] and you should never use a FAST fix on a production machine [12:46] cause the whole definition of fast, is it doesn't correctly solve the issue, and causes more issues later [12:47] OK.. All I know is when I try to start eh Mysql service I get "start: Job failed to start" [12:47] there are three ways to do things, fast, right and cheap, pick two :) [12:48] takamarou: what do the logs say? any errors in /var/log? [12:48] I've had issues with mysql not logging to the mysql log files [12:48] greppy, all empty [12:49] normally just starting mysqld manually shows the errors then [12:49] http://1.bp.blogspot.com/_SqhhJb_P3Kk/Sp_frVGGgXI/AAAAAAAAI9Q/BbBF9Rw5fac/s400/pick+two.jpg [12:49] suppose I should mention this, I just ran some system updates [12:49] try running, /usr/sbin/mysqld [12:50] http://pastebin.com/tDENT9Zs [12:51] sounds like /var/lib/mysql is owned by the wrong user [12:51] or you where not root when you ran that [12:52] sorry. Just ran it with sudo, no output. Logs are still empty [12:53] uhh, just tried to make a directory in my home directory.. got: [12:53] mkdir: cannot create directory `test': No space left on device [12:53] your disk is full? [12:54] guess reinstall mysql isn't enough then, reinstall the os :) [12:54] come on, clean up your drive [12:54] if you think you really do have disk space available, but it's not really free, you might have to reboot [12:54] patdk-wk, I have 200gb free [12:54] it wouldn't say out of space, if there was space [12:54] and, I just did a restart... then this problem began [12:54] are you out of inodes? [12:55] not sure.. what's that? [12:55] out of files [12:55] df -hi [12:55] http://pastebin.com/3s5bsMsP [12:56] heh? [12:56] where is the rest of the output? [12:57] cat /proc/mount [12:57] hold on.. tried another restart [12:59] hey zul [13:00] koolhead17: hi [13:00] patdk-wk, no such file or directory [13:00] cat /proc/mounts [13:00] and also, that was the entire output of df- hi [13:00] df -hi* [13:00] if that was the entire output, where is your root fs? [13:01] cause it only shows /boot [13:01] http://pastebin.com/vvaYi4kn [13:02] zul: https://bugs.launchpad.net/ubuntu/precise/+source/keystone/+bug/878282 [13:02] Launchpad bug 878282 in keystone "keystone database connection error" [Undecided,New] [13:03] koolhead17: cool ill fix it [13:03] takamarou, what is this rootfs stuff? [13:03] oh, I see it now, webnet-root [13:03] so where is the df -hi that includes /, webnet-root [13:04] patdk-wk, I'm really not sure. I don't actually know what df -hi does [13:04] same thing as df -h [13:04] except shows how many more files you can have [13:04] you have missing filesystems from it, so something is seriously wrong, you your copy/paste was bad [13:05] let me rephrase, I don't know what df does [13:06] patdk-wk, that's the entire output http://pastebin.com/PWaZTzKa [13:06] it is one of the most basic of commands ever, as much as ls, cp, mv, rm, ... [13:06] that is better [13:06] can you make directories now? [13:06] no\ [13:06] same output [13:07] what does /proc/mounts show now? [13:07] http://pastebin.com/1NExGx6B [13:08] not sure [13:08] that says you should be able to write [13:09] and df says your not out of space/inodes [13:09] :/ [13:09] but it's failing [13:09] ehh, weird.. [13:09] my only guess, at this point, cause I haven't seen this issue before [13:09] sudo mkdir works.. but I'm in my home directory, so I should have rights? [13:09] would be to reboot to a cd, and run a fsck -f on your root fs [13:10] oh? [13:10] yeah [13:10] so that's weird, but I wouldn't think that's what is breaking my mysql [13:10] it's not permissions, or it would of said that [13:10] so I still dunno [13:10] ok [13:10] so fsck you think? [13:10] maybe some apparmor thing, or something [13:10] but I won't be able to help you really [13:11] but it's affecting more than mysql, so I seriously doubt reinstall mysql would fix it [13:11] damn [13:11] what are the chances fsck breaks my system? [13:12] unlikely [13:12] great@! [13:12] but since sudo mkdir worked, I don't think fsck would fix it, it might [13:12] I would give fsck a higher change than mysql reinstall [13:12] but still a low chance [13:13] chance :) [13:13] if it was apparmor, there would be denials in the kern.log/audit.log [13:14] ya, I kind of doubt it's really apparmor [13:14] but only thing I can think of that could do something like that [13:15] hold on... gonna switch to my laptop so I can sit in front of the actual server... [13:17] back [13:19] so, if it's apparmor, what direction should I go with that? [13:21] I need my user "deploy" to be able to write to /var/www/apps/advicecapital and the folders in there, how do I do? [13:21] takamarou, easy, service apparmor stop [13:21] patdk-wk, That is easy :) [13:21] atleast to test it [13:29] not apparmor [13:44] znow: I am not a security expert, but I tend to think that it isn't a good practice for „normal users” to have write rights on web server folders. For development purposes: first I'd add a new group, assign the user to this group, change the group of the mentioned directory (and all of its subdirectories) and add write right for this group on these directories. And after the development finishes, remove the user from the group. [13:53] patdk-wk, any more advice? fsck came back clean, and turning off apparmor does no good....? [13:54] takamarou, nope, I'm just taking wild guesses, that might have an impact [13:54] but I really dunno, as I have never had that issue, ever, since I started using linux in '92 [13:55] that's comforting :) [13:56] Maybe the group membership was „damaged”? [14:07] zul, the gettext.sh problem being the gnome-keyring WARNING? or somethign else? [14:08] hallyn: its the libvirt-guests init script [14:08] ah [14:08] i thought you were just going to delete that before building the .deb? [14:09] hallyn: its commented out in the debian/rules now [14:09] otherwise it conflicts with the pre-stop in libvirt-bin.conf [14:09] https://bugzilla.redhat.com/show_bug.cgi?id=717957 [14:09] ok [14:09] bugzilla.redhat.com bug 717957 in libvirt "libvirt-guests script requires gettext but it wasn't installed" [Medium,New] [14:10] hallyn: ill finish it off today [14:12] zul, ok, no hurry from me, thx :) don't forget about the TtyAt path (in ubuntu16 in p) :) [14:12] yep yep [14:12] :) [14:26] New bug: #878877 in tomcat6 (main) "package tomcat6-docs 6.0.24-2ubuntu1.7 failed to install/upgrade: error writing to '': No such file or directory just after a fresh install" [Undecided,New] https://launchpad.net/bugs/878877 [14:31] sorry guys, i'm new to ubuntu... i'm using ubuntu server 10.04 and am trying to set up the LDAP server, but in trying to set it up (see https://help.ubuntu.com/10.04/serverguide/C/openldap-server.html, i'm at the backend.example.com.ldif step), i'm not sure how to create a file with that text through the commandline [14:33] Ian_: just open vi and cut/paste the text or create it locally and scp it to the server and copy into place [14:33] * koolhead17 is back [14:33] kim0: hey there [14:33] lynxman: i found glance doc awesome!! :) [14:33] i doubt much help is needed in that front :D [14:35] koolhead17: ooh very nice, I really like glance :) [14:36] lynxman: its like what i can do is add examples in that docs, they have nice detailed description already mentioned :D [14:40] learn something new every day [14:40] unfortunately, it's on a separate machine, so i guess i'm rewriting [14:41] thanks for the help [14:56] New bug: #878898 in backuppc (main) "package backuppc 3.1.0-9ubuntu1 failed to install/upgrade: Unterprozess installiertes post-installation-Skript gab den Fehlerwert 1 zurück" [Undecided,New] https://launchpad.net/bugs/878898 [14:57] when I try to do ssh git@github.com from my ubuntu server, it says permission denied publickey, but i have uploaded my public key? [14:58] hi, trying to set up a vm host machine, I'm done setting up KVM and libvirt (assuming successful since virsh -c qemu:///system list returns the required table headers), but when I try to build a virtual machine with "sudo vmbuilder kvm ubuntu --suite maverick --flavour virtual --arch amd64 --libvirt qemu:///system --ip IP --hostname internal --part vmbuilder.partition --user user --name "name" --pass pass" I'm stuck with Calling hook: boots [14:58] trap [14:59] my vmbuilder.partition has 3 lines in it "root 10000\n swap 2000\n ---" [15:00] jdstrand: im going to be uploading a new libvirt later today do you want to have a look at the debdiff after i finish writing these specs? [15:05] zul: if it passes qrt and hallyn is ok with it, go for it [15:06] jdstrand: k === semiosis_ is now known as semiosis [15:51] testing two one [16:00] thanks kirkland, nice lesson [16:01] nebajoth: thanks! [16:04] Curious, is there a definitive statement on the separation of juju and orchestrator? Juju is a project to automate orchestration of ubuntu clouds, but currently they seem separated? [16:05] Uhh, orchestra [16:08] Juju is awesome from my "tests" with it... I just don't really understand why orchestra is being discussed as a separate brand, and no one has really clarified the situation that I've seen [16:14] hey I know this question isn [16:14] 't ubuntu related, but what would be the easiest way of providing failover or a static point for users. [16:15] example.org should be always acessible, even if one of my IPs(Servers) fails... [16:15] using DNS for failover seems wrong. [16:15] this is why they invented bgp/ospf/... [16:15] but the IPs are server bound (MAC) so I could't change it [16:15] but do you really mean ip, or server? [16:15] those are two totally different things [16:16] you can easily move ip's between servers [16:16] hell, I do it hundreds of times per second [16:16] I suppose you have never looked at failover stuff [16:16] patdk-wk I can't, my hoster isn't enabling it [16:16] like, bgp, multipath, loadbalancers, ipvs, vrrp, ... [16:16] if you don't host your own crap, dns is the only option [16:17] depends what you mean with host your own crap [16:17] lunitik, juju doesnt' require orchestra. They serve distinct purposes. [16:17] I have a cluster of servers, but can't access the networking stuff [16:17] scalability-junk, control of the network, your servers run on [16:17] juju can work in a cloud environment. Orchestra is bare metal install/turn on. [16:18] Juju can work with Orchestra provisioned machines or EC2 provisioned etc. [16:18] the next best solution would be [16:18] two load balancers, in dns [16:18] and the load balancers go to the real webservers [16:18] but your still going be stuck to dns, at some level [16:18] patdk-wk and if one load balancer fails, the other one would be used ? [16:18] no [16:18] both would be used all the time [16:18] thought so :D [16:19] if one fails, HOPEFULLY the other would be used [16:19] cause browsers are suppost to be smart enough [16:19] cause that depends on dns [16:19] but if a webserver goes down, no one will notice, cause of the loadbalancers [16:19] supposed, so I could sort of work with dns, but not for sure :P [16:20] to have better protection on the loadbalacners, you need control of the network, for vrrp at a min [16:20] patdk-wk yeah I just have problems to accept that the lbs are unfailable ;) so I wanted to make sure. [16:20] who said they are unfailable? [16:20] but it's a lot less likely they would fail, compared to the webservers themselfs [16:21] and normally you should take webservers down for updates, much more often then loadbalancers need [16:21] patdk-wk: yeah, but the lot less likely part is most of the time unfailable for a lot of people... [16:21] heh? [16:21] why? [16:21] cause it wasn't setup correctly in the first place? [16:22] setup server, install loadbalancer, it should be fine from then on, except if a disk fails [16:22] Hi, I would like to set up a Ubuntu server for my private e-mail and maybe some webspace. I will be installing that on a VM. Which smtp/imap server would you guys recommend. I would like to have some kind of webmail interface like nutsmail or something similar. [16:22] I currently have a ISPConfig setup but I want to learn a bit more about all the different elements [16:22] so I figure setting it up myself would be a better way [16:22] ispconfig uses courier and postfix but they seem a bit meh [16:22] patdk-wk: yeah if a disk fails or any other hardware problem ;) anyway you helped me a lot. any suggestion for a loadbalancer? [16:22] medberry: ahh, I always forget about public/hybrid cuz they aren't as fun to play with [16:22] there are craploads of options :) [16:23] probably with ssl endpoint and only incoming proxying o0 [16:23] medberry: makes sense though, thanks for your feedback! [16:23] haproxy would be a big one, but that is web only [16:23] np, hth [16:23] ipvs does anything, but is much more complex, cause it's not just web [16:23] ok [16:23] I've thought about exim and cyrus as a combination, what do you think of that [16:24] hans what ever your comfortable using [16:24] I wouldn't touch those two myself [16:24] but I really hate cyrus, and don't like exim much, though it's much better than qmail [16:24] Well, I would like to hear some suggestions, I'm currently not comfortable with anything :) [16:25] dovecot is always a win :) [16:25] I just want somewhat flexible setup with a nice webmail interface [16:25] postfix is good [16:25] patdk-wk any solution for loadbalancing income but not outgoing data? so user->lb->server->user route instead of outgoing traffic going through the lb again. [16:25] so you would say postfix for smtp and dovecot for imap ? [16:25] there shold be ubuntu guides on setting that up [16:25] finally. the 11.10 server guide is now online [16:26] scalability-junk, how much traffic? [16:26] yeah, I can probably figure out setting it up if there some decent documentation. Just looking for some input for my intended use [16:26] patdk-wk not much I'm probably more curious about it, then it is needed right now. [16:27] a few TB per week [16:27] scalability-junk, setting it up for direct user return is possible, but also makes your setup very very complex, and extreemly hard to fix issues [16:27] and as a normal server these days can handle 10gbit of traffic easily [16:27] it's normally not needed [16:27] ah ok I thought it would be better/easier [16:28] I would look at probably haproxy first [16:28] unless you need more than web load balancing [16:28] so the easiest thing would be using 2 load balancers as proxy and ssl endpoint end redirect to the backend servers [16:28] yep [16:28] thought about nginx as an easy solution. [16:28] nginx would be a webserver [16:29] yeah but could be used as lb and ssl endpoint. [16:29] I wouldn't attempt to bastardize it into a load balancer [16:29] though I'm sure you could [16:29] why not use something made from the ground up for that purpose? [16:29] ah ok, so HAproxy is more the way to go [16:29] ipvs would give you the direct return path [16:30] but I doubt your network can handle that [16:30] ok, till I'm not google I don't need it :D [16:30] cause it does stuff kind of like vrrp does to move ip's [16:30] ok [16:31] just to be curious. if 2 lbs as a records in the dns could be sort of failover, if the browser is smart enough, it could work for 2 webservers too right? [16:31] yep [16:31] * scalability-junk I love that kind of stuff :P [16:31] the difference is only [16:32] if you need to take a webbrowser offline, everyone notices, with the lb's no one does [16:32] only if you have to take an lb offline will they, and that shouldn't happen often [16:32] not nearly as often as the webservers [16:32] ok great [16:33] still not sure how to handle ssl. I'm the always ssl kind of guy. [16:33] you can even run the lb's on the webservers themselfs [16:33] if you really want [16:33] just reboots then would get you [16:33] but using ssl endpoints and http after that isn't the best solution i reckon [16:33] depends on what your doing [16:33] if you control the network, it's fine [16:33] yeah that's the problem [16:33] if your hacked, well, it doesn't matter if it's encrypted or not, they have the keys [16:33] ^^ [16:33] scalability-junk: Error: "^" is not a valid command. [16:34] using ipsec instead of ssl lowers the overall overhead [16:34] cause instead of making a new ssl connection each time, it all goes over the same ipsec connection [16:35] sounds reasonable. [16:36] but if I would go for ssl I would need one lb for each domain to enable ssl loadbalancing for each domain directly to each webserver [16:36] damn that is more overhead than I would like... [16:40] scalability-junk, no you don't [16:40] read the damned manual :) [16:42] I don't need one loadbalancer for each domain? [16:43] if I have the ssl endpoint on the webserver. all loadbalanced traffic would be encrypted and so the loadbalanced wouldn't be able to decide which webserver to use. or am I totally wrong? [16:43] patd-wk thanks for your advice ;) [16:44] scalability-junk: with separate IPs for the web servers allocated to the load balancer, you'll only need one lb [16:44] you load the ssl certs on the loadbalancer [16:44] I trying to tolerate logcheck in lucid, but it's rules seem overly inadequate for use with postfix-dovecot+amavis. Is anyone aware of a third party update to logcheck rules tailored for lucid? [16:44] or the loadbalancers does it via ip [16:45] philsf: normally logcheck generates a rather large amount of output, something like fail2ban is bound to be better - block unfriendly hosts and read the logs if something goes bad [16:45] just my 2c [16:45] RoyK, that's the point: logcheck wasn't supposed to generate a rather ammount of output :/ [16:46] patdk-wk yeah I could load the ssl certs into the lb, but then I would have the http traffic to the webservers (don't like that on a network I can't controll) and via ip sounds great, haven't heard of that :( probably have to look at HAproxy closer [16:46] why would it be http traffic? [16:46] philsf: it generally does, and that's why I'd recommend using something a bit better, such as fail2ban. you'll see it quite clearly if something goes really bad and when it does, read the logs [16:46] things like "mail delivered to local user johndoe" is exactly what logcheck should ignore [16:47] RoyK, fail2ban deals with a completely different thing, maybe you mean logwatch or other log analyzer? [16:47] philsf: it should also ignore stuff like "machine at x.x.x.x was trying to use me as a relay" [16:48] patdk-wk user->https->lb (ssl endpoint) ->http-> webservers ->http -> lb (ssl startpoint) -> https --> user [16:48] user -> https -> lb -> http -> webserver [16:48] user -> https -> lb -> https -> webserver [16:49] philsf: my philosophy is, when it comes to logs, don't read them unless something goes bad. use some tool to scan them for BAD stuff, but don't waste time reading about some minor incident that won't hurt the system after all [16:49] this is why logs have error levels :) [16:49] so you decrypt the ssl and encrypt it (seems like a lot of resources for ssl handling, cause the de/encrypt part is done on the webserver again [16:49] pay attention to alert, crit, error [16:49] maybe some attention to warning [16:50] scalability-junk, I don't, but you would want to [16:50] or setup ipsec :) [16:50] encryption isn't that heavy really [16:50] not anymore [16:50] patdk-wk: an SSL load balancer is quite handy on heavy sites [16:50] RoyK, that's exactly what logcheck is all about [16:50] normally a good 1000mbit per cpu [16:51] I'll take a look at it, thanks for your help. [16:51] RoyK: how would you do it? [16:51] * RoyK goes back to reading up on puppet [16:51] RoyK: why not chef? [16:51] no idea what that is [16:55] Royk: similar to puppet http://www.opscode.com/chef/ [16:56] scalability-junk: I'm attending puppet master training, and it looks very good indeed, so I really don't need another tool... [16:57] hello all, anyone have to for a newb web hosting question? [16:57] RoyK: It was more a question of why puppet ;) [16:57] "anyone have" time, that is [16:58] scalability-junk, that chef doesn't look well documented [16:58] patdk-wk I haven't started to look into one of them closer, so I'm curious about which someone could prefer ;) [16:59] i need a recommended host for a site (wordpress) that will average 200 users, maybe a few thousand page clicks per day [16:59] wordpress.com with a custom domain o0 [16:59] what, something virtual? [16:59] well, outside of wordpress.com [16:59] a few thousand page clicks per day seems really really small [17:00] yeah I would go for shared hosting. [17:00] ya, it's not huge [17:00] should be fine with the right caching tool [17:01] any specific host you'd recommend over others? [17:01] trying to stay away from godaddy :) [17:01] depends on what you know [17:02] server-side? not much [17:02] BigBiff_: will you be administering the server yourself, or do you just need somewhere to place a bunch of wordpress sites? [17:02] hostgator is pretty easy (cpanel) [17:02] i wont be administering myself [17:03] i was looking at hostgator, bluehost [17:03] i just wanted to ask people who were more knowledgeable than me ;) [17:04] BigBiff_: I guess most people in here run their own servers :P [17:04] that's why I'm asking here, heh [17:05] also why we can't answer, cause we don't use shared hosting :) [17:05] * RoyK doesn't know much about such hosting services for that sole reason [17:05] ahh, hehe [17:05] I would go with hostgator I reckon, great support and easy to use [17:05] cool, i'll give them a shot [17:05] and remember caching ;) [17:06] on shared hosting, that speeds up everything a lot [17:06] google searches for "hosting reviews" are fairly biased, and seem to be setup by the hosts themselves [17:06] good recommendation, sir [17:06] thanks for your time, guys -happy hosting [17:06] see ya [17:16] scalability-junk: the real reason was just that I found a training session about it and wanted to learn something easily, getting paid on the way, and puppet really looks good [17:17] ah ok is juju something similar or totally different aproach? [17:21] juju seems very ubuntu centric [17:21] puppet can be used on most platforms [17:21] that's true, but ubuntu is the future... ;) [17:22] hallyn: libvirt being uploaded [17:22] not really, it's the current for some, but it can't do everyhting [17:22] RoyK: I know, ok so probably I go into puppet... [17:23] scalability-junk: we have all sorts at work, mostly ubuntu, some redhat, some fedora, some solaris/openindiana, an old HPUX box, some old Vax stations, you name it [17:24] I doubt I can (or want to) do much about those VAXes, nor the old HPux machine, but the rest would be nice to keep under central control [17:24] RoyK: why so many different osses? [17:25] mostly because of history [17:25] except the openindiana machines, which do ZFS storage [17:25] ok so mirgating all to one os wouldn't be an option? [17:25] no [17:25] well, possibly, but it'll take years [17:26] tons of custom software written for this or that [17:26] ah ok [17:26] some systems that were meant to last five years, including an old solaris 8 machine [17:27] still in service, no money for upgrades etc [17:27] wow solaris 8 not bad [17:27] even some win98 workstations [17:27] I think we got rid of the last winnt4 machine lately [17:29] ^^ [17:29] scalability-junk: Error: "^" is not a valid command. [17:30] zul, cool [17:31] scalability-junk: those VAX machines are from the early or mid 1990s :P [17:32] super-highspeed DEC Alpha CPUs [17:32] clocked at an astonishing 400MHz or so [17:32] wow sounds better than my quadcore i7 server ;) [17:33] * RoyK thinks i7 in a server, unless it's a home server, is a bad idea [17:34] i[357] doesn't support ECC memory, and in a server, you'd want that [17:36] depends what you do with the server I imagine. [17:37] for something in production, ECC is vital [17:37] for some home server, it doesn't matter much [17:37] ECC is great yeah, but for delivering content the price point counted, perhaps I will bite my self over it after some time, but right now it feels great [17:42] heh? ecc memory is cheap [17:43] wasn't cheaper for me [17:44] who said cheaper [17:44] it has like 3 extra chips [17:44] but when the difference is like 10% [17:45] for me as a student the 10% had a huge point :D [17:46] scalability-junk: I'm not using ECC for my home server either [17:46] I have piles of extra ecc memory [17:46] I should put it on ebay [17:47] patdk-wk: only problem with memory is that they change all the time, so whenever a new mobo arrives, you need new memory [17:47] yep [17:47] or perhaps you can use the old chips, but then degrade performance, add issues ....... [17:48] howdy. [17:48] repeating myself from last night, hoping someone is still awake this time. [17:48] your still using 90's stuff, so maybe you will buy it :) [17:48] patdk-wk: lol [17:49] patdk-wk: those machines won't get upgrades, nor replacements if they die.... [17:49] I just upgraded my ubuntu-server, everything works nicely, except some dbd issues when using netatalk (yes, sharing to mac) [17:49] Ive done that, hate it, but well :) [17:49] easier to get a stick of ram, than scrap a server before it is ready [17:49] air_: upgraded to onieric? [17:49] trying to scan the drive using dbd -s throws a memory error. [17:49] RoyK: yep. [17:49] air_: tried that myself, reinstalled lucid after 10 minutes [17:49] * patdk-wk knows nothing about netatalk/apple/i* [17:50] I was afraid that netatalk would be something people try to forget. :D [17:50] forget? never bothered to learn it :) [17:50] air_: there seems to be a bug in netatalk on onieric, and I really have far more use for a server that works than waiting for a bug to be fixed in some package not very many uses [17:52] RoyK: yeah, I noticed it's a beta tagged version. :/ [17:52] netatalk/oneiric uptodate 2.2~beta4-1 [17:53] patdk-wk: that's what non-mac users tend to do :) [17:53] air_: well, if you're looking for something stable, go with the LTS releases [17:53] they generally work far better [17:54] I do that for production, this is just a home server. [17:54] and for a server, do you really need the cutting (or bleeding) edge? [17:54] I go with lts, then if I need something newer, I move it into my ppa [17:54] this goes for my business and personal servers [17:54] air_: I did the same, and reinstalled once I found the bug :P [17:54] * scalability-junk has oneiric in production... on one server :P [17:54] (and yeah, I don't do AFP on anything else.) [17:54] RoyK: :) [17:54] you could try it backwards [17:55] air_: I didn't find time to report the bug - could you report it, please? [17:55] upport an old version to 10.10 [17:55] RoyK: how far did yuo trace it? memory error on trying to scan drive in dbd? [17:55] IIRC there was a segfault [17:55] air_: start afpd manually with --debug et al [17:56] IIRC it segfaults [17:56] Oct 20 20:56:33.056083 [4708] {dbif.c:648} (E:CNID): Failed to associate name index: Cannot allocate memory [17:56] then run 'ulimit -c unlimited' and try again and [17:56] huh? [17:57] that's another issue than what I saw [17:57] happens when running 'dbd -sv /mnt/' [17:57] air_: report a bug [17:57] My / drive works nicely in netatalk, the large raided drive fails. [17:57] hm.. interesting [17:58] I really didn't find time to look into it, as I said, I just reinstalled lucid to get something that worked [17:58] RoyK: http://pastebin.com/6it0sfyP [17:58] air_: can you please post a bug report on that one? [17:59] air_: how much memory do you have in the box? is it 32 or 64bit? [18:00] 64b, it's only 2gig in this one. [18:00] should suffice [18:00] It's not swapping at all, so I don't see the reason for the out of memory error? [18:00] neither do I [18:00] It's more likely something else failing, and it throws wrong error. [18:01] !bugs [18:01] If you find a bug in Ubuntu or any of its derivatives, please file a bug using the command « ubuntu-bug » - See https://help.ubuntu.com/community/ReportingBugs for other ways to report bugs. [18:01] (or, something else failing when trying to allocate memory) [18:01] air_: I don't know the internals about this (nor much else in the ubuntu development), so please file a bug [18:01] RoyK: sure, will do. [18:02] gotta put one kid to sleep first. :D [18:51] RoyK: filed it. https://bugs.launchpad.net/ubuntu/+source/netatalk/+bug/879020 [18:51] Launchpad bug 879020 in netatalk "afpd's CNID DB fails after upgrade" [Undecided,New] [18:58] are the UEC images missing support for ipvs on purpose? [18:58] is there any easy way for me to add support for this without remaking them from scratch? [19:00] Ryan_Lane: can you clarify what you mean by missing support for ipvs? [19:00] Ryan_Lane: install linux-virtual-extras I think [19:00] hmm. seems the kernel modules are there [19:00] but are broken [19:00] [ 714.343506] ip_vs: Unknown parameter `conn_tab_bits' [19:00] Ryan_Lane: maybe ipvsadm is too old? [19:01] this is all lucid. it's working on normal installs, not with UEC image [19:02] I guess I can try upgrading the kernel [19:04] Ryan_Lane: you sure the module was loaded correctly? [19:04] the module wouldn't load [19:04] because of the dmesg error I posted [19:05] * RoAkSoAx off to lunch [19:07] Ahh [19:08] heh. I can't just upgrade the damn kernel either [19:08] I have to upgrade the image, and rebuild the instance [19:08] awesome [19:14] Ryan_Lane: #ubuntu-cloud might have some tips for you on upgrading kernels btw [19:14] cool. thanks [19:18] anyone worked with capistrano on a ubuntu vm? ive upload my .ssh key from my home dir, etc... and when I try cap deploy:cold it says permission denied (public key) ? [19:25] Ryan_Lane: ip_vs module loaded just fine for me on lucid ami-6b965902 2.6.32-318-ec2 [19:26] 2.6.32-32-server [19:26] that's the kernel I'm using [19:26] I'm upgrading the kernel right now, hopefully [19:26] we'll see :) [19:27] modinfo ip_vs [19:27] srcversion: B215B2428B37B5E2A34F26B === mconigliaro_ is now known as mconigliaro [20:05] I have deployed my app to my server on linode, setted up virtual hosts (I think I have done it correct). but I get 403 forbidden when I enter the domain I set up as virtual host, but I can see my favicon in the bar,? [20:06] robbiew: ping for the monitoring spec for p i just re-used the oneiric one [20:06] zul: what aboyut thje cobbler one [20:06] RoAkSoAx: do you want to do it or want me to do it [20:08] * cwillu_at_work stabs apt-get with a tarball [20:08] for some reason, apt-get is pretending that it doesn't understand how to install dependencies [20:08] I have no idea why [20:08] http://paste.pocoo.org/raw/495747/ [20:09] (this is in an image build script similar to rootstock) [20:09] "cups : Depends: libavahi-client3 (>= 0.6.16) but it is not going to be installed"... so why don't you install it then? [20:10] stgraber, do you mind pushing http://people.canonical.com/~serge/lxc-fix-bindhome.debdiff as a fix for bug 879052 ? [20:10] Launchpad bug 879052 in lxc "user dir binding breaks on nonexistent user" [Low,Confirmed] https://launchpad.net/bugs/879052 [20:11] if someone has not noticed the openstack-dashboard is changed to horizon so new path is https://github.com/4P/horizon/ [20:11] :P [20:14] stgraber, uh oh, bad dep5 tag in there. lemme fix real quick [20:14] znow, you sure it's not cached? [20:14] Flatline_: dunno :) [20:15] znow, you can see if you have firebug or that chromium develoepr thingy under network tab [20:15] znow, or you might try clearing your cache [20:15] ctrl+shift+del is the shortcut for cleaning the cache [20:16] * cwillu_at_work jots that down [20:16] (fixed) [20:16] flatline: where is that? [20:16] guntbert, what? [20:16] flatline: "ctrl+shift+del is the shortcut for cleaning the cache" [20:17] * cwillu_at_work continues stabbing apt-get [20:17] guntbert, on chrome/chromium I suppose it would work on firefox too [20:17] lemme check [20:17] Perhaps this isn't a good day to start planning my migration from lucid [20:17] Flatline_: its first time I entered that domain [20:17] New bug: #879052 in lxc (main) "user dir binding breaks on nonexistent user" [Low,Confirmed] https://launchpad.net/bugs/879052 [20:18] Flatline_: it was my impression that this combination is captured by the OS to initiate a reboot [20:18] znow, umm can you paste your virtual host configuration? if you use an application server behind apache/nginx (apache as the proxy) you might need a tag I guess [20:18] hallyn: i just noticed that debian has a new lxc [20:18] Flatline_: yeah, where is it now I find that file? [20:18] :p [20:19] zul, yeah, we should sync at some point [20:19] guntbert, ah, its ctrl+alt+del on windows [20:19] its pretty complex with all those different 3 word directories in ubuntu :p [20:19] hallyn: we should [20:19] znow, are you using apache? [20:19] Flatline_: yup [20:19] zul, and then send remaining delta back to debian [20:19] znow, it should be something like /etc/httpd/conf/httpd.conf [20:19] i'm not gonna get to that today though [20:19] znow, just do a "sudo updatedb && locate httpd.conf" [20:19] Flatline_: my bad - sorry, I read ctrl+alt+del *blush* [20:19] updatedb might take a few seconds [20:20] zul, last i looked debian had changd its rules file (to something much nicer) too, but i couldn't get it to work easily [20:20] guntbert, :) [20:20] Flatline_: yup, sec :) [20:20] quick question, just to be on the safe side, does vmbuilder generate a random mac address? [20:20] Flatline_: sites_enabled yeah? [20:21] Flatline_: https://gist.github.com/1302239 [20:21] znow, checking sec [20:23] hallyn: agreed [20:25] znow, I'm not sure if this would work but can you try adding this: http://pastebin.com/R1wZYQbz between and [20:26] znow, requires a httpd restart [20:27] Flatline_: same 403 error :/ the file is named advicecapital.dk - but the domain is ac.danielg.dk - is there any probs with that? [20:28] znow, nope, I don't think so [20:28] Flatline_: okay, hmm this is odd [20:28] Flatline_: do I need to set anything up in hosts file or? [20:30] znow, umm, do you have a .htaccess file in your /var/www/apps/advicecapital/current/public folder? [20:30] znow, nah, that shouldn't be required [20:30] Flatline_: im gonna check, sec [20:31] znow, can you also paste apache error and access logs , they should be somewhere under like /var/log/httpd [20:31] Flatline_: no .htaccess - im gonna paste logs sec [20:33] znow, also can you check if you have a file like x.php seeing that file manually like http://domain.ext/x.php [20:33] Flatline_: https://gist.github.com/1302292 [20:33] Flatline_: its a rails app [20:33] no php stuff in there :) [20:34] aaah [20:34] apperantly your webserver doesn't serve the index page right up [20:34] I've never worked with rails before but I guess setting the right extensions in httpd.conf would solve your problem [20:35] try putting up an index.html in your html folder, that should be served correctly [20:35] if you confirm that, your virtual host configuration is right, but you have to configure apache to serve your rails pages too [20:37] Flatline_: sec [20:46] Flatline_: I just tried to add the site in sites-available aswell, tried do "a2ensite www.advicecapital.dk" and "a2ensite ac.danielg.dk" but it says the sites doesnt exists [20:49] I never used ruby or never needed to set it up on apache, but your problem is that your apache doesn't execute/serve your ruby files (it might just be the index extensions problem) maybe you should ask on ruby channel, I'm sorry I can't help you further [20:50] znow: the argument given to 'a2ensite' must be a file that exists in /etc/apache2/sites-available [20:50] rcsheets: ERROR: Site ac.danielg.dk not properly enabled: /etc/apache2/sites-enabled/ac.danielg.dk is a real file, not touching it [20:50] it says that [20:51] znow: /etc/apache2/sites-enabled/ should never contain any real files. it should only contain symlinks that are put there by a2ensite. [20:51] ah ok [20:51] znow: since a2ensite sees a file there, it is not touching it. [20:51] so I just delete the files in there? [20:51] okay [20:51] sec [20:51] znow: i can't tell you whether you should delete those files, because i don't know what's in them. [20:52] znow: you might want to back them up. but there should not be (regular) files in that directory. [20:52] rcsheets: virtual hosts :) [20:52] okay, a non regular appears now [20:52] rcsheets: atleast it has "blue" color, of the name [20:52] anyhow, I do need to add a line in the hosts file it appears? [20:52] znow: normally a change in /etc/hosts would not be needed just because you reconfigured apache. [20:53] rcsheets: hmm :/ [20:53] znow: why were you going to edit /etc/hosts? [20:53] rcsheets: read on the internet it might would be necessary, but I wont then :) but I do still get an 403 error when accessing ac.danielg.dk === shirgall is now known as Guest56801 [20:56] rcsheets: I have read* [21:00] rcsheets: you there? [21:00] <_jfb> RoyK: ping. [21:01] <_jfb> Hey folks, did a fresh install of 11.10, sshfs is hanging now after inactivity. Is there a KeepAlive setting I need somewhere on the client? [21:01] znow: on the phone. [21:01] on the server, I normally set TCPKeepAlive yes [21:02] <_jfb> can I set that in home directory config file if I don't have root on the server I'm connecting to? [21:02] in the server [21:02] so sshd_config [21:02] _jfb: pong [21:02] _jfb, you should be able to keep your connection alive by sending null packages [21:03] I'm not sure how it's done on sshfs, but possible [21:03] <_jfb> Flatline_: yes, I think that is what I want, where do I set that? [21:03] so, you don't need root perm on server [21:03] _jfb, dunno, have faith in google :) [21:03] * RoyK is off to bed [21:04] <_jfb> Flatline_: thanks... [21:04] <_jfb> RoyK: early night! [21:04] Need a little help with samba4 & vsftpd logins. Anyone have any experience? [21:04] _jfb, https://help.ubuntu.com/community/SSHFS, last part [21:05] _jfb, apperantly it works on sshfs too [21:05] <_jfb> Flatline_: perfect, thanks. That's the one I set before, I think I wiped it... this time I'll put in in ~/.ssh/config [21:06] _jfb: working on autoconfiguring ~/.ssh/config etc with puppet :) [21:06] <_jfb> RoyK: cool. I'm sure life is going to be (even more) smooth once you get back!! [21:07] anyone have experience working on amazon EC2 [21:07] _jfb: I guess starting with nordre etc will be good [21:07] all packages, sudoers and whatnot controlled centrally [21:08] ea k [21:08] l;s [21:08] ops [21:09] heh, I normally do that server side :) [21:09] Flatline - I've had a very short play with EC2 [21:09] if those scientistst could learn scipy, it'll make our life a bit easier :P [21:10] royk, my openindiana backed mythtv has been working great :) [21:10] _jfb: what is it IDL or matlab can do that scipy cannot? [21:10] <_jfb> RoyK: good question. [21:11] <_jfb> RoyK: For what we do, nothing really. [21:11] the licensing regime with matlab is a PITA [21:12] so is IDL [21:12] and AFAIK neither of those do parallel processing very well [21:13] Can anyone help with samba4 & vsftpd logins? Struggling with "user_sub_token=$USER" and "local_root=/home/$USER" [21:13] is samba4 in ubuntu? [21:14] rcsheets: there? [21:14] Built from git [21:14] then it really isn't an ubuntu thing :P [21:15] Yea, thought I'd ask though! [21:16] znow: what's going on now? [21:16] rcsheets: well just trying alot of things to fix this 403 error :p [21:17] znow: did you try reading the error log? [21:20] rcsheets: [Thu Oct 20 20:27:13 2011] [error] [client 87.52.2.167] Directory index forbidden by Options directive: /var/www/apps/advicecapital/current/public/ [21:20] znow: then you need either an index.html (or whatever is appropriate given your DirectoryIndex value) or Options Indexes [21:20] rcsheets: well, an rails application does not need an index.html file.. :/ hmm sec [21:24] znow: well, you're getting that error because there's no index.html and you have indexes disabled. [21:24] znow: if mod_passenger or something should be handling the request, then that's not happening either [21:24] rcsheets: hmm, well, still an index.html is not needed in a Ruby on Rails app, where do I have indexes disabled? [21:25] znow: i think you should probably get basic, static html hosting working before you try to tackle rails [21:25] rcsheets: erhh okay [21:25] znow: they are disabled by default. the relevant Options directive does not include "Indexes" [21:26] if you add "Indexes" to the Options for the relevant directory, then you should at least be able to get a file listing. [21:26] RoyK: FYI - I installed netatalk 2.2.1 from source, no more issues. [21:28] rcsheets: so far so good, just created the index.html file in public folder, wrote something, and that displays properly when I access the domain [21:29] o/ [21:29] hi all got little issue with my server ubuntu when installing for some reason i see my nic cards when installing but after system install i dont see them anymore i've got broadcom 5721 ethernet card :S can someone help me ? [21:30] rcsheets: yup... [21:30] rcsheets: now more debugging on why the bugger wont redirect to the actual hmm [21:31] znow: i'm afraid i have to do work now, so i can't help anymore, but at least it's better :) [21:31] rcsheets: it is, and thanks for your time === Richard is now known as Guest91159 [21:33] I am a newbie to linux [21:33] I'm trying to install a network printer [21:34] Guest91159: That is a desktop question, are you in #ubuntu? [21:34] I'm needing to know where the driver files are put to get them to installinstall [21:34] yes [21:35] I downloaded them to the desktop [21:35] Guest91159: search the dash for system-config-printers ... it has a wizard there, the drivers are already on your system probably [21:36] anyone can help me with my network problem with ubuntu server? [21:36] Guest91159: even before you downloaded them [21:37] Guest91159: hit "add" and choose from the Networked Printers list... find should locate it automatically if the printer is turned on and on the same LAN [21:37] ok, I have a printer folder [21:38] znow: no problem. good luck with the rails bit. [21:38] I've done all that and no printer is listed [21:38] Guest91159: Ubuntu comes with essentially every printer on earth supported out of the box, you don't have to find drivers yourself... you are thinking like a Windows user [21:39] my home network is on a windows 7 machine [21:39] Guest91159: Again irrelevant since the printer is network capable... what brand and model? [21:39] Brother 2040 [21:41] Guest91159: Do you know the IP of the printer? [21:41] mh strange I'm getting a 9999% cpu in top for fuser [21:41] and the cpu ist higly used... any idea? [21:42] don't use fuse :) [21:43] where can I find [21:43] I'm not using it on purpose [21:43] is it part of top? [21:43] seems unlikly [21:43] no [21:43] it's for using filesystems that are not supported directly via linux [21:44] Guest91159: Your printer should tell you, the router may also in its web interface [21:44] patdk-lap: how would I check all available file systems? [21:44] hallyn: is that something that should be SRUed or is it fine being in Precise only? [21:45] mount [21:45] stgraber, well, it's low priority; but if ppl hit that it'll cause hours of pain and years of hatred [21:46] stgraber, so i was hoping to sru it. your thoughts? [21:46] /dev/sda1 on / type ext4 (rw) [21:46] proc on /proc type proc (rw) [21:46] sysfs on /sys type sysfs (rw,noexec,nosuid,nodev) [21:46] fusectl on /sys/fs/fuse/connections type fusectl (rw) [21:46] none on /sys/kernel/debug type debugfs (rw) [21:46] none on /sys/kernel/security type securityfs (rw) [21:46] udev on /dev type devtmpfs (rw,mode=0755) [21:46] devpts on /dev/pts type devpts (rw,noexec,nosuid,gid=5,mode=0620) [21:46] tmpfs on /run type tmpfs (rw,noexec,nosuid,size=10%,mode=0755) [21:46] none on /run/lock type tmpfs (rw,noexec,nosuid,nodev,size=5242880) [21:46] none on /run/shm type tmpfs (rw,nosuid,nodev) [21:47] sorry pasted bin link failed... [21:47] *pastebin [21:47] that shuts me up :) [21:47] hallyn: I'm fine pushing it to -proposed too, fix is small enough that it should go through fine [21:47] is that template even in natty? [21:48] not sure for natty [21:48] I don't know where to look [21:49] patdk-lap: any idea what I did wrong? [21:51] posted crap to the channel [21:51] not used a pastebin [21:51] hallyn: uploaded to both precise and oneiric [21:52] stgraber, thanks [21:52] guess i better fill out an sru justification then [21:52] patdk-lap: this is the output of mount: http://pastebin.com/GfgxQjzL [21:52] that doesn't show anything using fuse [21:54] o0 [22:37] New bug: #879134 in samba (main) "package samba 2:3.5.8~dfsg-1ubuntu2.3 failed to install/upgrade: ErrorMessage: package samba is not ready for configuration cannot configure (current status `half-installed')" [Undecided,New] https://launchpad.net/bugs/879134