[00:25] when installing freenx to ubuntu-server 11.04, should i install the desktop first? i want the minimal xfce desktop available to freenx [00:34] the PPA repo seems to not be available? trying to follow https://help.ubuntu.com/community/FreeNX on ubuntu server 11.04 [00:39] Failed to fetch http://ppa.launchpad.net/freenx-team/ppa/ubuntu/dists/natty/main/source/Sources 404 Not Found [00:39] they dont have one for 11.04 yet [00:40] can i change natty to one of the other dists, and have itwork without any problems? [02:10] Hello all. Would appreciate some upstart wisdom: What's the recipe for telling a job to stop when another, *non-upstart* daemon is stopped (More broadly, instruct the job: "You stop running if this job stops running")? [02:13] amit: it needs to recieve a signal to stop on [02:14] so you'll have to hack the non upstart job to emit a signal on shutdown, then catch that signal with your upstart [02:15] afaik there's no clean way to have it poll the other job to see if it's still running, though certianly you could put pgrep in a while loop [02:17] http://upstart.ubuntu.com/getting-started.html === aguaithefreak is now known as aguair [03:25] 1 plum.cyber.com.au mdadm: RebuildFinished event detected on md device /dev/md1, component device mismatches found: 12800 [03:25] ...when my monthly raid rebuild happens, should I worry that it finds some mismatches? [03:47] hey [03:47] Can someone give me tips for the best IRC-Chat-Bot and IRC-Server ? [03:51] DerEnsBoss: for what use case? An internal office IRC server? [03:51] For my Privat Network [03:51] ! [03:52] So you will have what, maybe a couple dozen users at most? [03:52] And they'll be more-or-less trusted friends/family/coworkers/whatever? [03:52] Its for Test [03:52] How it Works [03:52] to set it up [03:52] So it's for pedagogy. [03:53] for myself to learn [03:53] In that case you are best choosing a simple implementation [03:53] ok [03:53] Thanks [03:54] FWIW I run ircd-irc2 because it's the oldest (and therefore, presumably, easiest) variant in Ubuntu [03:54] Or you know, you could just bugger off in the middle of a conversation [05:33] Hi guys. Have a question about ntp. I've installed the ntp client, and our servers generally line up. However after about 3 to 4 days I start getting 1+ seconds of drift between them. Is there a way to force them to update every 10 minutes? We use down to millisecond timing in our cluster, so it causes a lot of problems with our data storage when the times are out of sync [05:34] here's my conf file. https://gist.github.com/28938913f9a0947a2406 [05:36] toddnine: ntp or ntpdate? [05:37] Per ntp.conf's manpage, it appears to poll every 10s by default (maxpoll) [05:40] toddnine: oh, you're using some remote NTP server [05:41] toddnine: deploy a local one, then have *it* look at your country's RR NTP servers, and have the rest of your servers look at it. [05:41] twb: I'm on aws, does it provide an ntp service for it's Dcs? [05:41] Possibly you can set them up as peer-peer, but I haven't tried that [05:41] toddnine: I do not provide support for AWS, sorry [05:42] (Above was assuming your cluster was on a fast LAN, not distributed across "th cloud") [05:42] Yeah, they're all in the same zone, so the same dc [05:42] Shrug [05:51] toddnine: https://forums.aws.amazon.com/thread.jspa?messageID=53658 may be relevant?? [05:55] I wonder if the problem is the VM technology is not giving the VM every tick [05:56] twb, I think it's more that installing VMWare server2.0 on Ubuntu 10.04 is a huge pain in the ass! [05:56] jeeves_moss: huh? [05:56] jeeves_moss: he's using amazon stuff [05:56] But yeah, steer well clear of vmware, especially vmware server [06:04] twb, awww [06:17] jmarsden: Yeah, unfortunately that's outdated [06:18] those files don't exist within the distribution any longer [06:18] toddnine: OK... but it sounds like you have an EC2/cloud-specific issue, so asking the cloud/EC2 community is probably a good way to go. I [06:18] I [06:19] m not sure milisecond sync across the could is a "normal" expectation, to be honest... [06:19] s/could/cloud/ [06:19] jmarsden: Agreed. Didn't get any response on the aws channel either [06:20] I'm not 100% convinced that's the issue, I just noticed we get it a lot less when our clocks our synced. We're getting jobs that are executing multiple times in our queueing system, and it seems that it's due to the cluster not correctly syncing delete operations [06:22] toddnine: presumably you are paying Amazon money [06:22] toddnine: so call their tech support people [06:22] Also maybe get a DBA to fix the code so it doesn't rely on timing belts [06:22] twb: Doesn't work that way mate :) [06:23] We're not using a db, we're using distributed storage. The quorum doesn't return properly if the timestamps don't line up [06:23] Whatever === koolhead11|afk is now known as koolhead11 === smb` is now known as smb [09:37] morning all [09:38] morning jamespage [09:47] I'm a new Ubuntu Server Administrator and i am needing some guidance. [09:48] kbrown90: be more specific [09:48] i've recently installed Ubuntu Server 10.10 and got Apache 2.2 webserver up and working... now i'm stuck on trying to get a mailserver up and running. [09:49] i've got MySQL install as well. [09:49] !serverguide [09:49] The Ubuntu server guide may be found at http://help.ubuntu.com/10.04/serverguide/C/ [09:50] but i'm having problems with getting the server to give me the settings and configuration files for postfix. [09:51] Run postconf [10:51] everytime i use the command "telnet localhost 25" my server kicks back as it's trying to connect to "::1..." which i don't think is right and i'm not sure why? could someone help? [10:52] i'm trying to set up Postfix on my Ubuntu Server 11.04 [11:06] kbrown90, Try using a hostname that is only defined in IPv4, or "127.0.0.1" if you don't want to hunt one down. Then again, depending on your usage scenario, you might consider not responding to IPv6 a bug. [11:20] kbrown90, did you checked this https://help.ubuntu.com/community/Postfix before asking here ? [11:20] can someone help me in modifying the topic here [11:20] yes. I followed all the directions from there. [11:24] When you install LAMP That should also include PHP right? [11:26] yes [11:32] my computer i use to test my webserver out on is trying to download my php files instead of displaying them... why? [11:32] kbrown90, restart apache once [11:32] http://howtoforge.com/installing-apache2-with-php5-and-mysql-support-on-ubuntu-9.10-lamp [11:32] i just restarted the whole system. and it still doing it. [11:33] see the link [11:44] i well, i just got a lot of info from using the "info.php" thing, but my php forum is still not comming up. Could there be a problem somewhere else? [11:45] kbrown90, is php5 running on php.info [11:46] and also check the apache log [11:47] It says "PHP Version : 5.3.5-1ubunntu7.2" [11:48] kbrown90, well check the php code then. :) [11:49] would earlier versions of PHP work if i run PHP5? [12:03] can i install and run MySQL4 while running the newer versions as well? [12:44] kbrown90: it's possible with some manual hacks, but you really shouldn't do this [12:44] kbrown90: mysql 4 is veeery old and EOL [14:17] kbrown90, make sure your forum software is up to date and PHP 5.3 compatible [14:18] there were some major changes from PHP 4 and less major ones from 5.2 [14:18] which can break code [14:28] Hi, I am having trouble to set master user authentication with dovecot. Here is 'dovecot -n' output http://pastebin.com/Mjufm8eS the problem is when I try to login with master user via imapsync authentication failed: Here is dovecot log with failed authentication: http://pastebin.com/p2wWGkTD [14:29] master user "zimbra" is in "/etc/dovecot-passwd.masterusers". The problem is dovecot try to authenticate ldap user "example" with the master users password and fails. [14:31] .. [14:43] New bug: #799775 in python-gflags (universe) "[MIR] python-gflags" [Undecided,Fix committed] https://launchpad.net/bugs/799775 === guampa_ is now known as guampa [15:19] Hi everyone, I am using the HTML2PDF library to create PDF's with PHP. I have noticed that on my local machine these PDF's render in 2-3 seconds but over Ubuntu Server they take over a minute. Does anyone have any idea what could be causing this or how to improve performance? [15:19] I use the Google Charts API which is what takes so long .. but I am new to this and not sure what I should change [15:28] My server keeps using the loopback as interface, how could I change that to eth0? [15:30] can you see what packages u installed with apt-get now days [15:30] in desketop the software center tells you an install history [15:30] aptituted does in /var/log/aptitude [15:30] but I can't see anything in apt-get [15:38] hola buenas alguien me podria ayudar porfavor [15:38] alguien que hable spañol [15:38] ? [15:38] !es | Pablito [15:38] Pablito: En la mayoría de canales de Ubuntu se habla sólo en inglés. Si busca ayuda en español o charlar entra en el canal #ubuntu-es. Escribe "/join #ubuntu-es" (sin comillas) y dale a enter. [15:40] ok [15:41] amm [16:38] has anyone tried phpmyadmin with debconf-set-selections [16:38] SpamapS, ping [17:31] Say, anyone know of a setup that'll give me what FreeNAS does -- that is, a web interface, AND zfs? [17:31] Note that I don't absolutely need the web interface itself to expose ZFS. [17:31] Or rather, I'd want it to expose swapping drives, but not necessarily expose snapshots. [17:34] does anyone know a good channel to ask domain-related questions (A/MX records, subdomains, etc.) ? [17:36] JoeCoder: searching teh channel list shows me a channel named #dns [17:36] thanks. [17:36] New bug: #805572 in bind9 (main) "package bind9 1:9.7.3.dfsg-1ubuntu2.1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/805572 [17:36] also, how do I search the channel list? [17:37] depends what IRC client you use, in XChat it's in the Server menu [17:38] trillian has "browse channels" but there are thousands and not even alphabetized. useless. [17:38] (and no search) [17:38] no option to search? [17:38] bah [17:39] use xchat instead ;) [17:40] I would if I spent more time on irc. I'm here maybe 15 minutes a week. [17:53] i'm trying to follow https://help.ubuntu.com/community/FreeNX on ubuntu server 11.04, but there is no PPA repo for 11.04... can i get around this somehow without any problems? === koolhead11 is now known as koolhead11|afk === med_out is now known as medberry === Bilge_ is now known as Bilge [18:55] split splat splatter? === RoyK^ is now known as RoyK === tsimpson_ is now known as ts2 [19:05] my server ran out of space. How can I use the command line to see where all of the space went [19:05] something like df -R, if it existed [19:06] JoeCoder: you probably want "du" [19:06] why -R? [19:06] RoyK: recursive [19:06] yeah [19:06] -R is used for recursion in other commands [19:06] df looks at filesystems, not directories [19:06] JoeCoder: cd / ; du -h --max-depth=1 [19:06] probably a start [19:06] man du for more info, it's quite powerful [19:07] aha, var/log/apache2 is 8GB [19:07] oops [19:07] lol [19:07] this is a test server; I've got only my own stuff running on it. [19:07] might want to look at log rotation and compression [19:07] still learning [19:07] JoeCoder: probably forgot to rotate the logs, I guess [19:07] as StevenR said [19:07] or I don't know how :) [19:07] i've heard about log-rotate? [19:07] man logrotate [19:08] hm... but iirc log rotation for /var/log/apache2 should be in there by default [19:08] of the file is named something.log [19:08] Say, what's the difference between ubuntu desktop kernel and server kernel? [19:09] DanaG1: very little - server version uses lower HZ (ticks per second) and perhaps some other tweaks - that's about it [19:09] I guess NUMA may not be in the desktop version either [19:09] And is there some ubuntu server web-interface app? [19:09] !webmin [19:09] 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. [19:10] What's recommended now? [19:10] erm ... what's the other? the bot used to name it [19:10] !ebox [19:10] zentyal is a web-based GUI interface for administering a server. It is designed to work with Ubuntu/Debian style configuration management. See https://help.ubuntu.com/community/Zentyal (Project formally known as eBox - including in Lucid/10.04). [19:11] Packages are still named ebox. [19:11] ubottu: no, 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. [19:11] ubottu: no, 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, see !ebox instaed. [19:11] ubottu: no, 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, see !ebox instead. [19:11] !webmin [19:11] 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. [19:11] RoyK: it already exists [19:12] var/cache/apt is 114MB. Do I need to setup somethign to occasionally clear it? [19:12] ikonia: saying no, blah to ubottu gives it a correction [19:12] it doesn't need a correction [19:12] ikonia: a correction that needs to be approved [19:12] it doesn't need a correction [19:13] ikonia: it should be corrected to inform users about ebox, which it didn't [19:13] it doesn't need it [19:13] obviously not, but sometimes it's better to inform users directly about alternatives than telling them to FGFI [19:14] not when ebox has caused almost as many problems in the past - it's missed off for a reason [19:14] hmm, then what else is recommended? [19:15] options are limited [19:15] I'm pondering between Ubuntu Server and FreeNAS. [19:15] The latter has ZFS and a nice web interface, but the former has support for things like amd64_edac. [19:15] And boatloads more packages. [19:15] 8GB is pretty big for a server that only I use. apache's error.log was 8GB. Every entry seems to be: "mysql: Error reading file 'UNOPENED' (Errcode: 22)" I think it got stuck in some kind of loop, but I don't know what caused it. [19:16] anyone here using ZFS fuse on ubuntu? [19:17] I hear the write performance is rather horible... [19:17] unless it's native....it sucks [19:17] apache's error.log is growing at a rate of 3MB per second [19:17] tail error.log [19:17] I once used ntfs-3g to copy to a nearly-full, highly-fragmented drive. [19:17] JoeCoder: what's hitting that ? [19:18] JoeCoder: turn the debug level down ? [19:18] It seriously went down to BYTES per second. [19:18] s/seriously/actually/ [19:18] all of the entries are, "mysql: Error reading file 'UNOPENED' (Errcode: 22)" [19:18] JoeCoder: tail -f [19:18] JoeCoder: whats referencing that file ? [19:18] should be the apache processes [19:18] I don't have a file named "UNOPENED" [19:19] lsof error.log [19:19] lsof ? [19:19] list open files [19:19] lsof gives you then PIDs keeping a file open [19:19] had to install it. [19:20] https://gist.github.com/febf747ffe1c35e82883 [19:20] apache is using mod-suexec and php with fastcgi [19:21] error.log.1? [19:21] that's the rotated one [19:21] that's the one that is being written at 3MB/sec. I mistyped before. [19:21] JoeCoder: any idea why sh/mysql is keeping that file open? [19:22] do you have a cron job using it or something? [19:22] nope [19:22] which version of ubuntu is this? [19:22] php calls the mysql and sh commands. [19:22] 10.04 [19:22] I'll take a look at my php code that invokes those. [19:22] JoeCoder: then it seems one of them is doing something like a tail -f and isn't killed on logrotate [19:23] does logrotate rotate based on time or size? [19:23] JoeCoder: you might want to change the logrotate post script [19:23] hmm, are there documented cases of ebox breaking things? [19:23] JoeCoder: kinda both. [19:23] JoeCoder: usually time, once a day or once a week or so [19:23] depending on the settings [19:24] but if some processes keeps the file open, and the postrotate script doesn't do its job, the process (apache) will keep logging to the same file even though the filename's new [19:25] this log continues to be written at 3MB/sec even after mysql and apache have been stopped [19:25] in unix, the filename is merely a pointer after all [19:25] maybe I have a runaway shell script. [19:25] why would a shell script _write_ to that file? [19:25] I have no idea [19:26] same for mysql [19:26] well, track down whatever's filling up your drive and kill it [19:26] the php code calls a shell script and mysql; maybe apache redirects their output before they're invoked. [19:26] and perhaps use nagios/icinga to monitor the drive fill so that you get an alarm before the disk is full [19:27] Really, does ebox break stuff? [19:28] it was a call to the mysql command line utility I had made to import a database dump, called from php [19:28] DanaG1: search the net, the bug database etc - personally, I only use the commandline for managing unix machines [19:28] oh, and that ebox looks like it's intended to serve as a router. [19:28] so it must have had its output redirected to that file by apache [19:28] mystery solved, except for why mysql got stuck in that loop. [19:29] DanaG1: ebox isn't a router, but I guess there might be stuff in there to configure linux as a router :P [19:29] I already have Tomato as my router. [19:29] s/as/on/ [19:30] DanaG1: then don't use your server as a router :P [19:30] hmm, so anyway, here are the use cases I need most: ssh server, samba, and backups -- using ZFS, I want snapshots. [19:31] And a Deluge server. [19:31] Maybe I don't all that much need a web interface, after all. [19:31] s/ all that// [19:32] DanaG1: if you plan to use zfs, you might want to test openindiana - things get a little faster when zfs is native... === mquin_ is now known as mquin [19:34] * RoyK has some 350TB on ZFS on openindiana - works rather well... [19:37] What's the userspace like? [19:39] Oh yeah, random amusing thing found in a user manual for the SuperMicro atom board: [19:39] System Management Architecture for Server Hardware. [19:39] SMASH. [19:39] I don't want a SMASH anywhere near my server! [19:39] supermicro has ATOM boards? [19:39] * RoyK didn't know [19:40] X7SPA / X7SPE. [19:40] No ECC, though. [19:40] I decided to go with an HP MicroServer instead. [19:41] the cost for Atom + case + drive tray + psu is greater than the cost for the Microserver. [19:41] And no ECC. [19:42] DanaG1: you won't get ECC without a CPU that can handle it [19:42] Right. The Athlon II Neo N36L does support ECC. [19:42] most CPUs have the memory controller built-in these days, and Intel only supports ECC on Xeons [19:43] for AMD it's a different and more happy story :) [19:44] Now if only AMD would make an IPMI chip that had a basic Radeon in it... [19:44] So you could get KMS on KVM, instead of garbage ASPEED or Matrox. [19:44] At that point, I'd rather use ssh or serial console. [19:45] DanaG1: using a remote graphical console? [19:46] am I? Right now I don't have the IPMI card for the thing. [19:46] but are you using X? [19:47] Right now, I have FreeNAS on a USB drive, and an old Ubuntu desktop install on the data drive. (I have another drive coming tomorrow). [19:47] So, I do have X, but I don't anticipate much needing it for admin stuff. [19:48] just try openindiana text install instead [19:48] you won't need the fancy X stuff anyway === juliux_ is now known as juliux [20:15] hi #ubuntu-server === Xalem_ is now known as Xalem [20:17] I install lighttpd on natty/11.04, I can start it without problems via lighttpd -f /etc/lighttpd/lighttpd.conf but /etc/init.d/lighttpd start returns "Warning: Fake start-stop-daemon called, doing nothing." which I suppose happen because it's using an empty initctl rather than initctl.REAL so I thought I should use upstart thus service but that's the same, it uses start-stop-daemon and not start-stop-daemon.REAL so now Im lost [20:17] (the goal is to have lighttpd start at boot, using /etc/rc.local failed too) [20:24] Utopiah: i believe there is a switch you need to toggle in the lighttpd conf file [20:24] Utopiah: actually, disregard [20:25] the lighttpd conf file given by the Ubuntu default package if for init.d rather than upstart as far as I can tell [20:27] (but then I just discovered upstart today so I might be completly wrong) [20:29] exact output is http://pastebin.com/raw.php?i=TJ6Y0Xw4 btw [20:33] Utopiah: that link is bad [20:35] I make it die too soon, sorry http://pastebin.com/raw.php?i=4cqHQKLi [20:59] Utopiah: maybe pastebin the upstart job [21:02] New bug: #805661 in ntp (main) "NO reference clock support. I need Oncore ref clock support in ntp." [Undecided,New] https://launchpad.net/bugs/805661 [21:02] Lot of people here suggest to use ssh for everything. I just setup a Ubuntu Server. I want see it as network drive on my local network on both Ubuntu and windows installation. Is it possible to mount ssh as a network drive on windows? Or i should just use samba. === smw_ is now known as smw [21:05] http://pastebin.com/WRFVaesh [21:26] donnie: use samba for local shares. [21:41] @greppy thanks. i completely messed up my smb.conf reinstalling samba wouldn't rewrite it so i deleted the samba folder. Now it seems it is gone permanently. Any way to get it back [21:41] Is there a way to pull a list of all installed packages from an Ubuntu installation that is accessible but unbootable (turned into a USB drive since it won't boot). All files look good to go. [21:41] donnie: uninstall and purge and then reinstall [21:42] donnie: this would also be why backups are so prized, even for "it's just a little edit, no big deal..." [21:52] ascheel, chroot, dpkg -l === andregondim is now known as Andre_Gondim [21:53] qman__: gotcha === viezerd- is now known as viezerd [22:23] qman__: seems to have worked just dandy. Thanks!