[01:53] <qman__> I'm setting up postfix as an internet mail server, and it's asking me if I want to use procmail for local delivery. What's the difference?
[01:54] <qman__> I've set up postfix for internal mail plenty of times, but I've never run into this question
[01:54] <qman__> guessing it's new in lucid
[01:56] <qman__> also, it's not covered in the server guide
[01:56] <qman__> they just give a list of answers without the questions
[01:57] <qman__> and they don't match up with the current state of the debconf questions
[02:07] <enzoplex> I'm using irssi. Is there a way to remove all of the join/parts ?
[02:09] <qman__> sorry about that, DNS quit working properly
[02:51] <pmatulis> enzoplex: yes, but ask in #irssi
[03:38] <lamont> qman__: you have your debconf priority set lower than you have in the past - the questions haven't changed in a very long time.
[03:39] <lamont> that is, just hit return and take the default, and you'll see what you would have seen if the debconf priority was higher.
[03:40] <qman__> lamont, that makes sense, but following the guide precisely gives a different set of questions than the answers given follow
[03:41] <qman__> this is a fresh install, nothing else getting in the way
[03:41] <lamont> qman__: that's because normally (and for whoever wrote the guide), the debconf priority was higher, so they didn't see that question and it just got given the default answer without prompting
[03:41] <lamont> interesting
[03:42] <lamont> the procmail question only comes up if procmail is installed, but I had thought it was low
[03:43] <lamont> yeah.  procmail is a low priority question, so unless you've done something to make debconf start asking everything (like an error during the install...), you wouldn't get asked that question
[03:43] <qman__> I simply did `sudo dpkg-reconfigure postfix` as shown
[03:44] <qman__> for reference the procmail question comes after the listen addresses and before the mailbox size limit
[03:45] <lamont> sounds about right from my recollection of writing those questions several years ago
[03:46] <lamont> so... it's possible that the default for dpkg-reconfigure was changed to be low instead of high, dunno
[03:46] <lamont> but the postfix debconf questions haven't changed in an eon
[03:46] <qman__> not a big deal, just thought it might help if the questions were listed as well to avoid that problem
[03:46] <lamont> might not hurt, esp if the default priority changed
[04:23] <MTecknology> !info varnish
[04:30] <hggdh> yeah, seems interesting. See the Communications of the ACM, July 2010
[04:52] <cs1> guys
[04:52] <cs1> i need help here
[04:52] <cs1> is there any default LDAP password??
[05:00] <enav> Hola gente estoy buscando una alternativa al artisteer para editar plantillas de joomla... alguna buena sugerencia???????
[05:08] <cs1> guys
[05:08] <cs1> how do i trace my LDAP password??
[05:08] <cs1> is there anyway i can see the LDAP password??
[07:44] <netwidget>  hi, new to Ubuntu, installed ubuntu server 10.04 without domain name for local domain.  how do I do this after install?
[10:46] <joschi> hi, I have a problem with the a dynamic file template in rsyslog. it seems rsyslogd wants to write to the log files in /var/log/%HOSTNAME% before the directory was created.
[10:46] <joschi> at least in normal mode. in debug mode, it creates the directory just fine.
[10:46] <joschi> I've posted details about my configuration, the error message and part of the debug output under http://pastie.org/1048389
[10:46] <joschi> it could be related to Bug #484336 (https://bugs.launchpad.net/ubuntu/+source/rsyslog/+bug/484336). changing the owner of /var/log to the user 'syslog' works, but I'd rather avoid that
[11:40] <tarvid> why does the kernel add to cache when that causes increase in swap used?
[11:42] <tarvid> is there a way to control how aggressively  memory is cached?
[11:42] <tarvid> is there a way to reduce total swap?
[11:45] <joschi> tarvid: you can control that behaviour with sysctl parameter vm.swappiness
[11:45] <joschi> tarvid: also see https://help.ubuntu.com/community/SwapFaq#What%20is%20swappiness%20and%20how%20do%20I%20change%20it?
[11:47] <tarvid> thanks - current value is 60
[11:47] <tarvid> is there such a thing as cachinerr?
[11:47] <tarvid> is there such a thing as cachiness?
[12:02] <qman__> tarvid, what are you trying to do, exactly?
[12:03] <tarvid> avoid terrible situations where swap grows out of control and cached is large
[12:04] <qman__> the system will never use swap to cache
[12:04] <qman__> it only uses free memory to cache the disk
[12:04] <tarvid> there seems to be some "tipping point" where a sudden demand brings the server to its knees
[12:04] <tarvid> i'd rather not cache than swap
[12:05] <qman__> caching should not mean swapping, ever
[12:05] <qman__> it would be fairly pointless to cache the disk in swap, which is just another disk (or often the same disk)
[12:05] <eventyret> Got some issues with permissions denied when running echo server1.example.com > /etc/hostname
[12:05] <eventyret> /etc/init.d/hostname restart
[12:06] <eventyret> :/
[12:06] <tarvid> root@helen:/usr/sbin# free -m
[12:06] <tarvid>              total       used       free     shared    buffers     cached
[12:06] <tarvid> Mem:          3711       2865        845          0        302       1589
[12:06] <tarvid> -/+ buffers/cache:        973       2737
[12:06] <tarvid> Swap:        10691        148      10542
[12:06] <joschi> eventyret: forgot `sudo`?
[12:06] <eventyret> joschi: naah i used sudo as well though i made a thing here it was hostname.sh start
[12:06] <eventyret> following this guide http://ubuntuforums.org/showthread.php?t=1197883
[12:07] <qman__> tarvid, lower your swappiness
[12:07] <qman__> it will then do less caching to free more memory
[12:07] <tarvid> like 10?
[12:07] <joschi> eventyret: you used sudo in the wrong place? ;)
[12:08] <qman__> yeah, try 10
[12:08] <joschi> eventyret: `sudo echo foo > bar` for example won't work the you'd expect
[12:08] <qman__> if it's still swapping unnecessarily, lower further
[12:08] <eventyret> well
[12:08] <qman__> of course, you already have more than enough free to hold what's in swap
[12:08] <eventyret> joschi: this gives me an error as well echo ns1.YOURDOMAINNAMEHERE.com > /etc/hostname
[12:08] <qman__> which simply means it decided that memory won't be needed any time soon
[12:09] <joschi> eventyret: yes, of course. /etc/hostname is only writable by root.
[12:09] <eventyret> Gives me -bash: /etc/hostname: Permission denied
[12:09] <eventyret> *doh*
[12:09] <qman__> eventyret, using redirection doesn't quite work with sudo
[12:10] <eventyret> hehe ^
[12:10] <qman__> the first part will be root, but the redirection won't be
[12:10] <eventyret> might be that
[12:10] <eventyret> its a VPS so :)
[12:10] <eventyret> but even their guide sucked
[12:10] <eventyret> so only root can change hostnames and such
[12:10] <qman__> I just use sudo -i, then do whatever I needed
[12:10] <eventyret> what is sudo -i ? :)
[12:10] <qman__> there is probably a way to quote it into a single command but I don't know
[12:10] <joschi> eventyret: `man sudo`
[12:10] <qman__> sudo -i invokes a root shell
[12:12] <eventyret> ahhh ok :)
[12:12] <eventyret> *shurgs* how to exit manual ><
[12:12] <eventyret> I know im fresh but i though it was Ctrl X like in nano ><
[12:12] <eventyret> q :D
[12:12] <eventyret> worked :P
[12:13] <tarvid> thanks joschi qman__  it takes a week or so to test but I am monitoring with cacti
[12:13] <qman__> tarvid, it's far more likely that a memory leak is causing your issue, but that will make it wait longer to swap things
[12:14] <qman__> also ten gigs of swap is an awful lot
[12:14] <tarvid> I didn't choose that, it was the default and a lot less would be preferable
[12:14] <tarvid> is there a safe way to reduce the size of the swap partition?
[12:15] <qman__> swapoff, resize the partition, swapon
[12:15] <qman__> but that will just leave that space unused
[12:15] <tarvid> pretty standard lamp+postfix+mailman installation
[12:16] <tarvid> got lots of space
[12:18] <qman__> on systems with plenty of RAM, I generally just don't do swap at all
[12:19] <tarvid> still on 9.10, php issues with drupal and python issues with google apps
[12:20] <tarvid> I'll take a look at 10.4.1
[12:20] <qman__> I'm keeping my desktop on 9.10 for a while, a lot of people had trouble with it, but my upgrade went smooth and I've got it just the way I like it
[12:21] <tarvid> me too, some issues like transcode
[12:22] <tarvid> i ran across a NID system that could work on IPtables but I forget what it was
[12:22] <tarvid> My server gets hammered periodically
[12:23] <qman__> try the iptables recent module
[12:23] <qman__> great way to limit the request rate per IP
[12:25] <tarvid> any tricks to install? prerequisites?
[12:26] <qman__> I think it's included in the default install for ubuntu
[12:27] <qman__> a good explanation of how to use it: http://www.debian-administration.org/articles/187
[12:28] <qman__> the example situation is solved better other ways, but the information there is good anyway
[12:31] <tarvid> not just ssh, i watched msnbot banging user/register repeatedly the other night in spite of robots.txt and rejections by the server
[12:31] <qman__> right
[12:31] <qman__> just change the port specified and the counts and such
[12:31] <qman__> it works for any type of connection
[12:31] <tarvid> that means I have to think
[12:32] <qman__> of course
[12:32] <qman__> but you only have to think once
[12:32] <qman__> script it, and be done
[12:32] <tarvid> I have the Gerald Ford problem and much is floating around in my mind
[12:33] <tarvid> wherever I left it
[12:40] <qman__> and yeah, msnbot purposely ignores robots.txt and is especially brutal
[12:41] <tarvid> looking through the auth log and I have a number of failed attempts on vsftpd
[13:28] <eventyret> does this channel cover questions about DNS stuff :)
[13:32] <eventyret> exit
[14:01] <gnoob> Hi guys. I have a ubuntu-server with 4 identical disks. 1Tb each. Running RAID5. Each disk has a partition of 6Gb free space before raid-partitions start. sda1 (6Gb) has Ubuntu on it.  I would like to make something substituting RAD1 using dd to clone sda1 to sdb1.
[14:01] <gnoob> Is this possible?
[14:02] <gnoob> want the server to keep running from sdb1 if sda fails..   But sending me a e-mail :)
[14:02] <gnoob> anyone with good advices?  (my first time doing this :))
[14:04] <gnoob> I would have to set boot priority in bios to sda first and if no sda -> sdb I guess..   BUt afraid dd, and identical partitions (and boot sectors) can mess things up a bit
[14:05] <gnoob> Raid1 is probably a better option?  but had some problems with grub..   Any suggestions?
[14:09] <joschi> gnoob: I'd say go with a RAID-1
[14:16] <ivo_> hi guys I am having a small problem I have soft RAID5 and it meant one of the drives is failed, which is not, because all drives' SMART is ok
[14:16] <ivo_> I am now rebuilding it
[14:16] <ivo_> but I should be able to use it
[14:16] <gnoob> Ok raid1 over sda1 and sdb1 then,  but it will slow down performance a little or not? And grub will need to be installed on sdb also right? :)  Thanks for advice joschi
[14:16] <ivo_> but It is not avaibe to mount in the disk utility
[14:16] <ivo_> I tried to mount it manually
[14:17] <ivo_> but because the array is encrypted I get this:
[14:17] <ivo_> do mount -o ro /dev/md0 /mnt/tempstore/
[14:17] <ivo_> mount: unknown filesystem type 'crypto_LUKS'
[14:17] <ivo_> can you help me how to proceed
[14:17] <gnoob> I tried raid1 at first,  but had some grub problems..  (but used old disks and IDE over one channel =  slooooow  :p)
[14:17] <joschi> gnoob: yes, it'll marginally cost performance and yes, you'll have to install grub in each mbr
[14:18] <ivo_> a link to a manpage will be aso ok
[14:19] <gnoob> Ok thanks a  lot joschi!!  Ill start getting hands dirty then :)      I will probably return to ask for how the e-mail thing works later :p    want to be noticed when a disk have trouble
[14:19] <ivo_> I am not new to linux but have used other unix and unix like systems for a very long time
[14:50] <BeeBuu> anyone help me?
[14:51] <BeeBuu> i follow the article on https://help.ubuntu.com/community/UEC/CDInstall ,when i do the part 5 of STEP 7, it failed~~~
[16:56] <tdr112> can you get a 32 bit verstion ?
[17:27] <alison> hello, one of my servers with a gigabit nic keeps connecting and 10 or 100 Mbps, and very occasionally at 1000. Would anyone know why this is??
[17:28] <alison> oh and its connected to a 1000Mbps port on the switch
[17:35] <soren> alison: Line noise, maybe.
[17:35] <soren> alison: bad cable or something.
[17:42] <mestre> hola
[17:42] <mestre> me pueden ayudar a configurar el ldap con samba
[17:43] <mestre> para autenticacion de usuarios?
[17:49] <mestre> somebody?
[17:51] <mestre> english or spanish?
[17:57] <Trixboxer> Hi, one small question regarding UEC, in "MANAGED" mode why to use dhcp server.. I got succeed in "SYSTEM" mode for creating instance but IP assignment is not working correctly.. any pointers ?
[19:04] <MrPicard> Hi Guys, where could i find the most easiest ftp server to setup?
[19:05] <MrPicard> The other question i have is, Where is the config file for proftpd?
[19:05] <remix_tj> /etc/proftpd/proftpd.conf afaik
[19:06] <MrPicard> Ta remix shall give it a go.
[19:08] <MrPicard> Ok, i type cd /etc/proftpd but it comes up as having no directory
[19:09] <MrPicard> any alternative ftp systems?
[19:16] <MrPicard> when i now connect upto ftpd im given this error.
[19:16] <MrPicard> Connection attempt failed with "ECONNREFUSED - Connection refused by server".
[22:47] <Italian_Plumber> at startup, is there a way to cancel a disk check in progress?
[22:49] <hackeron> hey, does anyone know a way to get 16 audio input channels into one PC? - I dont need a mixer or anything, basic phono/composite or 3.5mm microphone jack will do
[22:51] <palt> Italian_Plumber: Try pressing c
[22:52] <cloakable> hackeron: sixteen soundcards?
[22:53] <cloakable> or there's likely some specialised kit with that many input channels
[22:53] <hackeron> cloakable: tried 2 creative ones and even that caused stability problems
[22:54] <ksoviero> What are the differences between the generic kernel and the server kernel?  The FAQ points me here: http://www.ubuntu.com/products/whatisubuntu/serveredition/features/kernel, but it is a 404...
[22:58] <ksoviero> hello?
[23:01] <jpds> Hi.
[23:01] <jpds> The server kernel has less modules built-in I believe.
[23:02] <ksoviero> Well, for example, I know it has a PAE extension, that allows up to 64GB or RAM in the 32bit version...
[23:02] <ksoviero> does it have any enhancements for something like asynchronous I/O?
[23:42] <cwillu_at_work> any slab guru's around?
[23:42] <cwillu_at_work> trying to troubleshoot an oom that shouldn't be happening
[23:42] <cwillu_at_work> small allocation is failing even though there's 50+MB of ram free (on an embedded server with 256mb ram)
[23:43] <cwillu_at_work> slabtop is reporting an insane number of size-2048 slabs, all in use, and the number increase at a rate of about 1 per second
[23:43] <cwillu_at_work> disconnecting the network cable stops the increase :)
[23:43] <cwillu_at_work> but I'm trying to figure out how to reclaim the space, if such a thing is possible
[23:43] <cwillu_at_work> or figure out exactly where the memory is being allocated
[23:44] <cwillu_at_work> (rmmoding the network driver doesn't reclaim anything)
[23:45] <soren> cwillu_at_work: Try in #ubuntu-kernel. They're more likely to be able to help.