[00:06] <lownoize> hi
[00:06] <lownoize> is anybody here running vsftpd under ubuntu 10.04 server x64?
[00:08] <lownoize> and has an idea what could be the problem for the bug 590537
[00:14] <Moltar> Jordan_U: I finally got the system re-installed from the DVD, here's the boot info script output if you have a second to look at it:  http://pastebin.com/z4ZZhyW2
[00:14] <zoooom> I setup ubuntu to run an ET server. I can ping with it's IP but not it's name.  Do I need to run DNS on it?
[00:17] <NightDragon> okay, so since when is wpkg 'easy software deployment'?
[00:17] <NightDragon> is WPKG really as easy at it gets?... its about as easy as defusing a bomb
[00:35] <Jordan_U> Moltar: Interesting, and when you try to boot what happens?
[00:42] <Moltar> Jordan_U: It just says 'boot failed: Hard disk 0'.  Not terribly helpful.
[00:43] <Moltar> Jordan_U: That message is coming from the BIOS.  On these IBM boxes, it's uEFI, not really BIOS.
[00:46] <Jordan_U> Moltar: If you hold shift during boot is "GRUB" ever printed to the screen?
[00:47] <randy_> ?
[00:47] <Moltar> Jordan_U: Let me go try that.
[00:51] <Moltar> Jordan_U: No, it still just says Boot Failed.  It's like it doesn't recognize the MBR or something.
[00:58] <Jordan_U> Moltar: Are you sure that you're booting in BIOS mode rather than UEFI mode?
[01:00] <Moltar> Jordan_U: Pretty sure.  If I reduce the size of the RAID array so it's less than 2TB, I can install and I'll end up with a fdisk formatted drive that works just fine.  I don't have to change anything in the BIOS to do that other than delete a disk from the RAID array.
[01:01] <Jordan_U> Moltar: Maybe the BIOS code is trying to be too smart for its own good. Try asking in #grub.
[01:02] <Moltar> Jordan_U: I've tried using the grub-efi-amd64 package and re-creating the /boot partition as msdos so that the EFI BIOS can find it.  That allows me to set up an EFI boot option with the grub.efi image file.  That gets rid of the 'boot failed' message, but leaves me stuck after the initrd has been loaded.
[01:03] <Moltar> Jordan_U: Thanks for looking, I'll go try the #grub group.
[01:05] <Jordan_U> Moltar: There's been a lot of work kernel side recently to get video with EFI going well. This probably isn't very usefull to you on a production server, but maverick would likely work better when booting via UEFI.
[01:08] <Moltar> Jordan_U: Thanks, I'll at least try downloading and installing it.  It'll be worth a shot to see if it will boot at least.
[01:10] <Jordan_U> Moltar: You're welcome.
[02:11] <ader10> I can't use pureadmin to modify virtual users. pureadmin doesn't recognize that I have a user ftpuser and a group ftpgroup already made and it keeps trying to make them
[02:13] <ader10> And becauseit can't create the user and group, it won't even open the virtual user window
[02:13] <ader10> Any help on fixing pureadmin would be appreciated
[02:50] <zoooom> I setup ubuntu to run an ET server. I can ping with it's IP but not it's name.  Do I need to run DNS on it?
[02:53] <qman__> zoooom, you don't need to run DNS on that server necessarily, but you do need to create a DNS entry for it, or whatever other name service you're using
[02:53] <ader10> zoooom: If it's behind a dynamic IP it would be easiest to use a dynamic dns service
[02:55] <zoooom> maybe I'm getting it wrong but If I'm using DHCP from my router shouldnt I be able to ping it by name?
[02:55] <qman__> not likely
[02:55] <qman__> your router would have to have a dynamic DNS on it
[02:56] <qman__> most stock firmwares don't
[02:56] <qman__> dd-wrt can
[02:56] <zoooom> no I mean on my local network
[02:56] <qman__> yes
[02:56] <ader10> zoooom: likely
[02:56] <qman__> your router would have to run a DNS server that can update itself based on the DHCP leases it's handing out
[02:56] <zoooom> yea  it doesnt respond to it's name  but I can ping 10.0.0.101
[02:57] <qman__> most routers' stock firmwares can't do that
[02:57] <qman__> dd-wrt has the software to do it
[02:58] <zoooom> OK I'll just use the ip then    thanks
[03:22] <ader10> Can anybody help me figure out how to get symlinks working with pure-ftpd
[03:22] <ader10> My user can't even see them
[03:22] <ader10> (virtual user)
[03:23] <ejat> bug 523148 , is it will be fix in LTS ? or maverick ?
[03:29] <spartan07> crontab not working. running LTS 8.04. tried editing 2 ways 1) crontab -e and 2) nano /etc/crontab
[03:30] <spartan07> when I run the command on the terminal it runs perfectly
[03:33] <bogeyd6> spartan07, paste command
[03:39] <spartan07> for testing:  echo "Nightly Backup Successful: $(date)" >> /home/seg02/vTigercrons/test.log
[03:40] <spartan07> need for work:  /var/www/vtigercrm/cron/modules/com_vtiger_workflow/com_vtiger_workflow.sh
[03:41] <spartan07> bogeyd6, the way it is in crontab -e : 05 * * * * /var/www/vtigercrm/cron/modules/com_vtiger_workflow/com_vtiger_workflow.sh >>/home/seg02/vTigercrons/workflow.log 2>&1
[03:42] <bogeyd6> oh
[03:42] <bogeyd6> there is your probb
[03:42] <spartan07> im running tail -f for the echo command running every min
[03:42] <spartan07> ??
[03:43] <bogeyd6> 05 * * * * /var/www/vtigercrm/cron/modules/com_vtiger_workflow/com_vtiger_workflow.sh > /home/seg02/vTigercrons/workflow.log 2>&1
[03:44] <spartan07> bogeyd6, not 2 >?
[03:44] <spartan07> isnt >> append?
[03:45] <bogeyd6> im telling ya
[03:46] <spartan07> ok 1 sec running
[03:46] <bogeyd6> 05 * * * * /var/www/vtigercrm/cron/modules/com_vtiger_workflow/com_vtiger_workflow.sh >>/home/seg02/vTigercrons/workflow.log 2>&1
[03:46] <bogeyd6> wtf
[03:46] <bogeyd6> 01 00 * * 1 /root/archive-mail > /var/log/archive-mail.log 2>&1
[03:46] <bogeyd6> 01 01 * * * /root/zimbra-backup > /var/log/zimbra-backup.log 2>&1
[03:46] <bogeyd6> that is my crontab on the zimbra server, works like a champ and the >> was causing the commands not to work
[03:47] <bogeyd6> Linux zcs 2.6.24-28-server #1 SMP Fri Jun 18 14:47:02 UTC 2010 x86_64 GNU/Linux
[03:47] <spartan07> running 8.04?
[03:49] <spartan07> no go
[03:50] <bogeyd6> kk
[03:50] <spartan07> bogeyd6, echo running every min. not running :(
[03:51] <bogeyd6> you are trying to echo from a crontab command?
[03:52] <spartan07> bogeyd6,                   01 * * * * echo "Nightly Backup Successful: $(date)" > /home/seg02/vTigercrons/test.log
[03:52] <spartan07> just for testing
[03:53] <bogeyd6> uhm
[03:53] <bogeyd6> why not just put it into a .sh file and lets it run it that way
[03:54] <spartan07> just doing it to test the crontab. the other command is running .sh file but its not running either
[04:12] <spartan07> bogeyd6, wierd it ran but it truncated and time the command ran was 20:01:01 PDT 2010.
[04:13] <spartan07> its 21 20:11:02 PDT 2010, so its running but not running every hour
[04:15] <spartan07> sorry every min
[04:15] <twb> You haven't asked it to run every minute
[04:15] <twb> You've asked it to run on the first minute of every hour
[04:15] <spartan07> LOL
[04:16] <spartan07> how would every min be?
[04:16] <twb> * 1  * * *  echo foo
[04:16] <twb> There should be a header present when you run "crontab -e"
[04:17] <spartan07> no header
[04:17] <spartan07> https://help.ubuntu.com/community/CronHowto
[04:17] <spartan07> it says minute(s) hour(s) day(s)_of_month month(s) day(s)_of_week user command
[04:18] <twb> Oh, sorry, every minute of every hour would be "* *  * * *  echo foo"
[04:19] <spartan07> 01 *    * * *   is right for every min right?
[04:19] <fallous> no
[04:19] <twb> No.
[04:19] <fallous> all asterisks for the time
[04:19] <spartan07> * *   * * *  is every min?
[04:19] <fallous> yes
[04:20] <fallous> 1 * * * * would be the first minute of every hour, * 1 * * * would be every minute during the 1am hour
[04:23] <spartan07> what about every 5 min? sorry for the basic quesitons I thought I had this figured out
[04:24] <spartan07> echo command is running every min <--awesome
[04:24] <fallous> comma-separated
[04:24] <failover> My bind is eating 160mb of ram, this is right?
[04:25] <failover> About 200 machines use this server as nameserver
[04:25] <fallous> so for every 15 minutes, say, you'd do 0,15,30,45 * * * * echo foo
[04:26] <spartan07> ok so 5 min would be 0/05 *  * * *
[04:26] <fallous> */5
[04:27] <spartan07> fallous, thank you
[04:27] <fallous> np
[04:30] <twb> fallous: dnsmasq certainly doesn't need that much ram
[04:37] <fallous> erm, that was failover not me ;)
[04:38] <failover> :)
[04:39] <twb> Sorry
[04:44] <spartan07> fallous, how would I tag a date to a log  for example the cron */5 * * * * /var/www/vtigercrm/cron/modules/com_vtiger_workflow/com_vtiger_workflow.sh > /home/seg02/vTigercrons/workflow.log 2>&1
[04:49] <fallous> I script pretty much exclusively with perl, not shell
[04:52] <spartan07> ok
[04:53] <spartan07> thank you guys for the help. crons working now. I guess they were always workign correctly
[04:53] <spartan07> LOL
[04:53] <ajmitch> */5 * * * * /var/www/vtigercrm/cron/modules/com_vtiger_workflow/com_vtiger_workflow.sh >  /home/seg02/vTigercrons/workflow.log
[04:53] <ajmitch> sorry, was trying to append date
[04:54] <ajmitch> $(date +\%Y.\%m\%d.\%H\%M) in the filename
[05:08] <FOCer> can't start mysql on 10.04 after making some configuration changes, even after reverted the changes and reboot the server it won't start either.
[05:12] <FOCer> any help would appreciated.
[05:42] <silentwhisper> is there an easy way to setup a mail server
[05:42] <silentwhisper> most of the tutorial are way to long to follow
[06:05] <KurtKraut> silentwhisper, if this mail server need to exchange e-mails with other servers around the internet (like @gmail.com), yes, it is a pain in the ass and very long process you need to follow.
[06:05] <KurtKraut> silentwhisper, everything that differ from the steps the tutorials provide will lead to all your e-mails being labelled as spam
[06:39] <ader10> Can anybody help me figure out how to get symlinks working with pure-ftpd? My user can't even see them (virtual user)
[06:39] <qman__> ader10, symlink paths are relative to the environment you're in, so if they point to something that exists outside of your chroot, or is in a different location because of the chroot, they won't work. Use mount -o bind if you need to link something outside the chroot.
[06:40] <ader10> qman__: http://www.pureftpd.org/project/pure-ftpd says that symbolic links can be followed even if outside the chroot
[06:41] <qman__> it must pre-process them then
[06:41] <qman__> sure the links are valid?
[06:41] <qman__> I make them backwards by accident a lot
[06:41] <ader10> I'll double check
[06:42] <ader10> Yes, the link works
[06:42] <qman__> ok
[06:47] <qman__> --with-virtualchroot: usually, when a user is chrooted (-A and -a options), it's impossible to go out of his home directory. Enabling that feature makes it possible: symbolic links are always followed, even if they are pointing to directories not located in the user's home directory. ... This feature isn't enabled by default.
[06:47] <qman__> in the compilation switches
[06:47] <qman__> I'm guessing the built in package doesn't have that switch
[06:47] <ader10> ugh, dumb compilers
[06:48] <qman__> it's a security feature
[06:48] <qman__> mount -o bind will work around it
[06:48] <twb> http://mywiki.wooledge.org/FtpMustDie
[06:48] <qman__> that too
[06:55] <ader10> thank you qman__
[06:55] <ader10> wait, it's still not showing
[06:57] <ader10> The mount --binded folder works, but it doesn't show in the listing
[06:57] <qman__> that's a bit strange
[06:57] <qman__> maybe it's got a single-filesystem limitation or something
[06:58] <Hilikus> can someone help me with sftp? my connection is dropping right after the user is authenticated successfully
[06:58] <Hilikus> here are the logs of the client and server
[06:58] <Hilikus> http://pastebin.ca/1905802
[06:59] <twb> Hilikus: how are you invoking the client?
[06:59] <Hilikus> sftp user@server
[06:59] <Hilikus> -v for debugging
[06:59] <twb> Hilikus: try that on mazinger itself
[07:00] <Hilikus> same thing twb
[07:00] <qman__> Hilikus, I've run into that in two cases--it'll do that if the user does not have a valid shell, and it'll do that if their home directory doesn't exist and they're chrooted
[07:01] <twb> Hilikus: pastebin "iptables-save -c" on mazinger
[07:01] <Hilikus> that user does not have a valid shell. i read that for sftp only i could leave the shell /bin/false
[07:01] <Hilikus> i only want to give the user sftp access
[07:02] <twb> Hilikus: that's not how you do it
[07:02] <Hilikus> damn
[07:02] <ader10> Strange, the mount --binded folder doesn't show in FileZilla but it does show in Windows Explorer
[07:02] <qman__> users need a valid shell to sftp
[07:02] <qman__> chroot them with the Match block and force-command internal-sftp
[07:02] <twb> Yep
[07:03] <Hilikus> http://www.minstrel.org.uk/papers/sftp/builtin/
[07:03] <Hilikus> step 8
[07:03] <Hilikus> bullet 2
[07:03] <Hilikus> so that's wrong then
[07:03] <qman__> yes
[07:03] <Hilikus> ok, so i'll change the user's shell
[07:03] <twb> Hilikus: anything more complicated than what qman__ just said is the Old Way
[07:04] <qman__> even with the Old Way, they still needed a valid shell
[07:04] <qman__> I used the one included with jailkit
[07:04] <twb> Actually that URL is using Match, so it isn't far wrong
[07:04] <qman__> but with the New Way, it doesn't matter
[07:08] <qman__> also, I'm fairly certain that bit about root needing to own the home directory is wrong
[07:09] <qman__> the user needs to own his own home directory
[07:13] <Hilikus> thats my next problem
[07:14] <Hilikus> if what the website says is true then it makes sense, cause right now the dir is now owned by root
[07:14] <Hilikus> and i get this error when chrooting
[07:14] <Hilikus> Jul 22 02:12:29 mazinger sshd[14872]: fatal: bad ownership or modes for chroot directory component "/tmp/"
[07:14] <Hilikus> the user's home is in /tmp/test
[07:15] <Hilikus> and ChrootDirectory /tmp/test/
[07:15] <qman__>  /tmp does have bad modes
[07:15] <qman__> it's 777
[07:15] <Hilikus> man does say that it should be owned by root
[07:15] <qman__> the parent directory to the home directory must not be world writable
[07:16] <qman__> and should be owned by root
[07:16] <Hilikus> This path, and all its components, must be root-             owned directories that are not writable by any other user or group
[07:16] <qman__> but the home directory itself should be owned by the user
[07:16] <Hilikus> that's where i'm, confused. why would i need to chown the user's home to root
[07:16] <Hilikus> it says THIS path
[07:17] <qman__> you shouldn't
[07:17] <qman__> but the directory containing the user's home must be root
[07:17] <qman__> and all directories above it
[07:17] <qman__> and must not be writable by anyone else
[07:17] <qman__> but the user's home should be owned by the user, and writable by the user
[07:17] <qman__> it wouldn't make a lot of sense if it wasn't
[07:18] <qman__> the user then couldn't create any files or directories
[07:19] <Hilikus> ok, so its every other dir in the path that shoyld be owned by root, except the actual, last dir in the path
[07:19] <qman__> yes
[07:19] <silentwhisper> how to confine a user to its own home directory only?
[07:20] <qman__> silentwhisper, chroot
[07:20] <Hilikus> yes, and in sftp only
[07:24] <spartan07> FYI: * * * * * echo "$(date)" >> /home/test.sh >> /home/spartan7/test.log 2>&1  gave me the datestamp and output of the script on the log.
[07:25] <spartan07> quick question. if everything is running good on ubuntu 8.04 server should I install 10.04?
[07:25] <qman__> spartan07, that's appending the date to test.sh
[07:26] <spartan07> qman__, its running the script though
[07:27] <qman__> if it is, that's a pretty bizarre parsing
[07:27] <qman__> I wouldn't rely on it
[07:27] <qman__> * * * * * /home/test.sh >> /home/spartan7/test.$(date).log is better
[07:28] <spartan07> ok I wont then. ok
[07:28] <qman__> or, if you want the date in the log, add it to the script
[07:28] <qman__> rather than a dated log
[07:29] <spartan07> is there a way to pipe output of the script to a script that adds the date?
[07:29] <qman__> sure, but that's a bit of a waste
[07:29] <spartan07> i'll figure it out. I need to learn to bash script
[07:30] <spartan07> for now I'll do what you mentioned and make a dated lof file
[07:30] <spartan07> *log
[07:30] <qman__> to make the script add the date, just add echo $(date) to the end of it
[07:31] <spartan07> ok , let me try. thanks!!!
[07:31] <qman__> oh, also
[07:31] <qman__> what I gave you won't work as-is, you need to put it in quotes
[07:31] <qman__> * * * * * /home/test.sh >> "/home/spartan7/test.$(date).log" 2>&1
[07:32] <qman__> since the default output of date has spaces in it
[07:32] <imyousuf> I have installed UEC and downloaded the lucid lynx 64 bit version from the store. Now when I am trying to run the image I am getting - http://paste.ubuntu.com/467353/ My machine configuration is core i7, 4G RAM, 1G VGA, Lucid Lynx 64-bit. Any idea how I could solve it?
[07:36] <silentwhisper> is this correct ?
[07:37] <spartan07> qman__, thank you that exactly what I have been trying to do all afternoon! im gonna idle here, maybe I can learn something LOL
[07:41] <Hilikus> is it not legal to hard link to a dir?
[07:51] <silentwhisper> sudo chroot /home/edwin
[07:52] <silentwhisper> very slow internet connection i need to restart brb
[07:57] <Hilikus> qman__: i think that all the dirs in the path should be owned by root
[07:57] <Hilikus> check this out
[07:57] <Hilikus> http://www.debian-administration.org/articles/590
[07:57] <Hilikus> its really strange though, why would the user's home dir be owned by root
[07:57] <Hilikus> ChrootDirectory /home/%u
[07:58] <Hilikus> which is the user's home dir
[07:58] <Hilikus> The directory in which to chroot() must be owned by root.
[08:01] <twb> Isn't there a %x variable specifically used for the home dir?
[08:02] <twb> yeah, %d
[08:04] <Hilikus> yea, but still, whatever is the dir of the jail, it should be owned by root
[08:04] <imyousuf> I have installed UEC and downloaded the lucid lynx 64 bit version from the store. Now when I am trying to run the image I am getting - http://paste.ubuntu.com/467353/ My machine configuration is core i7, 4G RAM, 1G VGA, Lucid Lynx 64-bit. Any idea how I could solve it? (re-run)
[08:05] <Hilikus> so i don't understand how to trap a user in its own home
[08:05] <Hilikus> since its own home would have to be owned by root
[08:24] <Roxyhart0> HI there somebody know how i can create a virtual interface but not associated
[08:24] <Roxyhart0> to any ethX?
[08:25] <larsemil> imyousuf: dont you need like three computers to use UEC?
[08:25] <imyousuf> larsemil: I made the same computer as node control and cluster control
[08:26] <imyousuf> do I still need 3 computers? I am absolutely new to UEC, so please forgive my ignorance or stupidities.
[08:29] <larsemil> imyousuf: i did not use it myself, sorry
[08:31] <imyousuf> no problem larsemil, in that case I am just did one step more than you :)
[08:31] <larsemil> i think you need two nodes atleast
[08:32] <imyousuf> hmm
[08:32] <imyousuf> where do I find documentation on that?
[08:32] <imyousuf> because its first time I heard that I need *at least* 2 computers :(
[08:58] <smoser> TeTeT, i responded to https://answers.launchpad.net/ubuntu/+source/eucalyptus/+question/118429
[08:58] <smoser> i think the user has a funky environment in some way.
[08:58] <smoser> note that the program they're running is in /usr/local/bin
[08:59] <TeTeT> smoser: ah, ok, I don't have 9.10 any longer around, otherwise I'd have tested it
[09:00] <smoser> i didn't see anywhere we the opener mentioned the client (where euca2ools was being run)
[09:00] <smoser> the server is seemingly carmic
[09:00] <smoser> karmic
[09:11] <overrider> What is (argueably) the most used tool to monitor my servers logfiles, and send me e-mail on things that look suspicious? On my OpenBSD Machines i like to use logsentry; is that also the tool of choice on Ubuntu Server? Thanks
[09:20] <larsemil> nagios maybe?
[10:00] <_ruben> overrider: we use SEC (simple event correlator) for that
[10:34] <binBASH> moin \sh
[10:35] <\sh> hey binBASH
[10:35] <binBASH> Gruss aus LU :P
[11:26] <silentwhisper> what mail server is usually used in accessing through web
[11:27] <Carachi> hello, i try to install postfix but i have this problem: when i test postfix via telnet  (ehlo) it  not write AUTH LOGIN PLAIN and AUTH=LOGIN PLAIN . how can i solve it?
[11:38] <qman__> Carachi, it isn't supposed to, or at least none of mine do
[11:39] <qman__> the process is telnet, ehlo hostname, user username
[11:40] <Carachi> qman__ i do ehlo hostname
[11:40] <Carachi> 250-kalilab.it
[11:40] <Carachi> 250-PIPELINING
[11:40] <Carachi> 250-SIZE 10240000
[11:40] <Carachi> 250-VRFY
[11:40] <Carachi> 250-ETRN
[11:40] <Carachi> 250-STARTTLS
[11:40] <Carachi> 250-ENHANCEDSTATUSCODES
[11:40] <Carachi> 250-8BITMIME
[11:40] <Carachi> 250 DSN
[11:41] <qman__> !pastebin
[11:41] <qman__> but yes, that's the same thing I get
[11:41] <Carachi> and how can i solve this problem?
[11:42] <qman__> I wasn't aware that's a problem
[11:42] <qman__> my servers send mail just fine
[11:55] <qman__> ah, pretty sure what you're looking for is smtpd_sasl_auth_enable in /etc/postfix/main.cf
[12:05] <Carachi> yes is enable
[12:23] <progre55> hi people! I've got a server that's using the CEST timezone.. how can I change it to UTC? and will it affect the applications and services running on it? like, mysql
[12:28] <sommer> morning all
[12:28] <progre55> day
[12:52] <qman__> progre55, timestamps and such use UTC, so there's nothing to worry about there
[12:53] <qman__> that is, file timestamps and database transactions
[12:53] <qman__> log timestamps will change
[12:54] <qman__> the command is `sudo dpkg-reconfigure tzdata`
[12:54] <progre55> qman__: thanks. but will I have to reboot then?
[12:54] <qman__> probably not
[12:55] <qman__> I haven't tried it but I don't see why you would need to
[12:57] <progre55> qman__: I've been using puppets, and my server shows me "Thu Jul 22 13:55:34 CEST 2010" while my client has "Thu Jul 22 11:55:40 UTC 2010". and I'm getting errors while connecting.. I was assuming there was a problem with timezones, as the connection is over SSL
[12:58] <qman__> progre55, timezone is not likely the problem
[12:58] <qman__> run `date -u` to ensure they agree on UTC
[12:59] <progre55> qman__: sec
[12:59] <progre55> yeah, they're the same..
[12:59] <progre55> qman__: familiar with puppets by any chance? =)
[12:59] <qman__> no, I'm not
[12:59] <qman__> I mean, it might be a problem, but that'd be rather silly of it
[13:00] <progre55> qman__: well, I'll try to google more then =)
[13:00] <progre55> qman__: thanks man, appreciate
[13:00] <qman__> if the systems agree within the minute to UTC it _should_ make any time-based authentication systems work
[13:00] <qman__> no problem
[13:13] <psteyn> Hi guys, I've removed two NIC's from a server and left only the 1.  it's still called eth2 even on reboot...is there a way to 'rescan' / 'remap' it so that it's eth0?  This is ubuntu 9.10
[13:17] <qman__> psteyn, remove the entries in /etc/udev/rules.d/70-persistent-net.rules
[14:46] <hggdh> Daviey: euca 2.0 i386 is in the archive, but not amd64
[15:58] <urbis> moin
[15:58] <urbis> little help with proftp needed
[16:00] <SandGorgon> hi guys.. quick question. I read somewhere that EXT3 filesystems give better performance than EXT4 for MySQL. Is that true ? What filesystem are you guys using _in production_ ?
[17:27] <Hilikus> is it safe to do security based on user's groups? i have two groups: backup and sftp. one gives it access to reading the backups, the otherone gives them limited, chrooted access through sftp. sshd makes this distinction based on the user's group. is this safe?
[19:55] <Jeeves_> Anyone here in the UK?
[20:19] <rcsheets> I am trying to install Windows Server 2008R2 using virt-manager on a 64-bit lucid server. The ISO boots up fine, but as soon as I click "next" on the first installer screen, I get this: http://i.imgur.com/Tomty.png
[20:34] <dominicdinada> why is the servers asking about grub configs after the updates... i do believe this crap is bugged.... Update? Crash my servers ? not good
[20:41] <pmatulis> rcsheets: someone was saying the same thing yesterday. i believe they filed a bug
[20:41] <rcsheets> i see
[20:41] <rcsheets> thanks. i'll look around launchpad.
[20:43] <pmatulis> rcsheets: it you're around tomorrow i can let you know exactly, i have it in my irc client logs
[20:44] <rcsheets> cool
[20:44] <rcsheets> i'll try to stop by
[20:47] <rcsheets> i see a bug about >4 vCPUs
[21:03] <dominicdinada> Drupal = Trash
[21:07] <koolhead17> kees: ping
[22:20] <bogeyd6> how do i disconnect an iscsi connection? (i am the initiator)
[22:26] <lowridah> anyone familiar with kickstart options for ubuntu installs?
[22:27] <lowridah> I've got a %pre directive with a script that switches to TTY6, takes input for variables that get exported to /tmp/includes which is referenced earlier in the KS file, so I can set IPs and such from one menu prior to install
[22:27] <lowridah> this works great in Fedora/CentOS but in Ubuntu there doesn't seem to be a /tmp folder for ramdisk, does anyone know where ubuntu's chroot ramdisk is at during install?
[22:29] <stinger05> hi there
[22:30] <stinger05> does ubuntu 10.04 come with a GUI after installing it ?
[22:30] <lowridah> not server, no.
[22:31] <lowridah> if you need a gui i suggest desktop
[22:31] <stinger05> well actually i wanted a server with a GUI
[22:31] <nealmcb> stinger05: but there are various options: https://help.ubuntu.com/community/ServerGUI
[22:31] <lowridah> red the arguments against part real good though
[22:32] <stinger05> ok thanks guys
[22:32] <stinger05> later.
[22:49] <lowridah> really odd, I can have %pre touch /tmp/includes but the code ran in %pre won't write to /tmp/includes
[23:00] <jord> How can I tell my ubuntu 10.04 install to continue booting even if it can't mount one of my (non-boot) hard drives?
[23:00] <jord> Now, it just hangs if it fails to mount /dev/md0
[23:07] <talcite> hey guys, I'm trying to start up an iscsi target using tgtd, over an infiniband connection, but I'm getting an error "cannot initialize RDMA; load kernel modules?". The modules for iSER are loaded.
[23:08] <talcite> I'm also using IPoIB though, is it possible that you can't have both at the same time?
[23:10] <FunnyLookinHat> Anyone know how to restart Subversion on 10.04 ?
[23:28] <rcsheets> pmatulis: i found the bug. it's not quite the problem i'm seeing. thanks though.