[00:05] <Psi-Jack> SpamapS: heh, so far, so good. There was a depenancy conflict because it was looking for a slightly newer version of pacemaker-dev, but I think it built the debs. :d
[00:07] <SpamapS> Psi-Jack: if it all works out, you should submit it as a backport.
[00:08] <Psi-Jack> Heh yeah, prolly. :)
[00:08] <Psi-Jack> First now I gotta finish setting up pacemaker enough to allow me to use it. ;)
[01:04] <CppIsWeird> how do you control which user a service starts up using?
[01:06] <constantin_mike> hello. i'm back with the "/etc/init.d/apache2 start" problem. when i do that, i get this "/etc/default/rcS: 1: [Desktop: not found". can anyone help me?
[01:30] <JuJuBee> Can someone help me get nfs-kernel-server started ?  It currently takes about 5 minutes or so and fails.  I purged it and re-installed it 2x already
[03:19] <faileas> I'm attempting to set up openvpn on ubuntu server using this guide - https://help.ubuntu.com/community/OpenVPN - i am quite sure i have the port open (but i need to drop by to that location to check) but can't seem to connect from my client. I'm wondering if there's any way that i can check if the server's listening/connectable before i start tearing apart my client side setup
[03:22] <faileas> client side logs - http://pastebin.ca/1926431 server side logs - http://pastebin.ca/1926432
[03:24]  * ball is confused
[03:25] <faileas> by me or something else?
[03:25] <ball> faileas: Life generally.
[03:26] <ball> brb
[03:32] <ball> Is it possible to partition an "md" software RAID array?
[04:23] <qiyong> what's the standard tool to configure rc.d ? rcconf isn't shipped. so what tools else?
[04:49] <itsux2bu> so when i installed ubuntu-server i was presented with a list of server related apps to install..  like lamp/openssh/samba etc.. i just ran something called Taslsel.. it looks just like the list shown at install time..  its cool..
[04:49] <itsux2bu> *tasksel
[05:32] <itsux2bu> is there anything in 32-bit ubuntu server that captures screen output during boot up?  i see a red dot but screen clears too quickly to see text next to it..  or does ubuntu keep any boot logs?
[06:29] <MatBoy> what is the way to upgrade my testserver to the testing branche ?
[08:08] <UnGoWyzre> Olla brodderz
[08:09] <UnGoWyzre> ikonia brotha hola
[08:09] <UnGoWyzre> bazhang brotha hello
[08:39] <droogie> I've just installed iRedmail following their instructions and I encountered a problem. I know this is perhaps not the correct channel to ask the question, but I'll try anyway. Everything except phpldapadmin works fine. The problem with phpldapadmin is that it is unable to base dn and thus object navigation is broken. I've verified the dn I use to login with in slapd.conf and it is correct. I can search the ldapdatabase using phpldapa
[10:56] <ikonia> off to the gym please keep an eye out for ranjan ban dodging and goat-re was kicked as chipped_cpu earlier
[10:56] <ikonia> oops
[12:44] <Psi-Jack> Wow, pacemaker is effing awesome when you understand it. :D
[12:51] <Psi-Jack> What do I need to have installed for pacemaker to use ocf:pacemaker:controld?
[12:51] <Psi-Jack> For the dlm system to work.
[12:53] <wastl> hi
[12:54] <Guest56018> hello
[12:55] <Guest56018> can i make a question about my ubuntu server, which is running on amazon ec2?
[12:55] <Psi-Jack> You just did.
[12:55] <wastl> lol
[12:55] <wastl> indeed
[12:55] <wastl> *g*
[12:56] <Psi-Jack> Hmmm
[12:56] <Psi-Jack> Trying to figure out what I need to install to get dlm working with ocf:pacemaker:controld.
[12:57] <Psi-Jack> I installed cman even though I'm using pacemaker, but it still gives a dlm: not installed rc status.
[13:10] <Guest56018> I have found an AMI with Ubuntu 10.04 installed, it is EBS and i configured everything. The performance is on c1.small
[13:10] <Guest56018> What can i do to increase the performance to e.g. c1.xlarge
[13:10] <Daviey> "performance"?
[13:11] <Guest56018> http://aws.amazon.com/de/ec2/instance-types/
[13:11] <Daviey> Guest56018: You essentially want to upgrade your instance to a higher spec?
[13:11] <Guest56018> Amazon is offering different performance types for their virtual machines.
[13:12] <Guest56018> Thats right, but with the current data on my volume
[13:15] <Daviey> Guest56018:  smoser might be able to help.  It's not really a Ubuntu specific issue :(
[13:25] <RoyK> apt-get install \*
[13:26] <RoyK> :D
[13:31] <RoyK> hi all. I have this machine used as a number cruncher - it seems the kernel is leaking memory somehow. See http://pastebin.com/ZFbxG3Ni for free/slabinfo/ps. any idea what may cause this?
[13:35] <Psi-Jack> Hmm interesting.
[13:36] <Psi-Jack> I upgraded to the PPA for lucid-cluster for pacemaker stuff, which got me dlm controld for it all, but the ocf:pacemaker:controld does not have the parameter for "daemon".
[13:36] <pmatulis> RoyK: why do you say there is a leak?
[13:46] <Psi-Jack> Heh, interesting. Though the ocf:pacemaker:countrold doesn't have the "daemon" paremeter, it checks the resource-name instance for if it has [gG][fF][sS] or [dD][lL][mM] in it to determine if it's a gfs or dlm instance. ;)
[13:49] <smoser> Guest56018, your big issue is with architecture. the largest you can go in i386 is c1.medium.
[13:49] <smoser> If you want to move to c1.medium, its easy.
[13:50] <Guest56018> wow, nice! OK; i have to make a whole new AMI (like ubuntu 64bit), and do everything again
[13:51] <smoser> you stop the instance (ec2-stop-instances), and then start the instance with a new --instance-type (ec2-modify-instance-attribute --instance-type c1.medium <iid>) , ec2-start-instances
[13:51] <ttx> hggdh, kirkland: ping me when around. need status on server-maverick-uec-testing and server-maverick-uec-liveusb
[13:52] <Guest56018> Thank you, very helpful! Do there exist different package versions of 32 and 64 bit? I use apache, php, exim4 and so on
[13:52] <smoser> i have never tried it, but it may be possible for you to boot an i386 root filesystem on an x86_64 arch, you'd have to get the x86_64 modules installed in the root filesytem though.  (you probably dont want to do this)
[13:52] <Guest56018> I cannot migrate my 32 bit EBS to a 64 bit AMI
[13:53] <smoser> well, maybe you coudl. its just not goig nto be easy or ideal.
[13:53] <smoser> http://uec-images.ubuntu.com/server/releases/10.04/
[13:53] <smoser> those are our released AMI ids.  we should have refreshed ones later today.
[13:53] <smoser> just pick a x86_64 instance, basically everything will be the same unless you have some arch specific code.
[13:54] <Guest56018> OK understood everything! Thank you!! I do it from scratch again with your AMIs and hope that everything that worked in the 32 bit now works again
[13:54] <smoser> there really is very little "works on i386 doesn't on x86_64"
[13:55] <Guest56018> Final question: This is one possibility to "scale up". Another possibility is to scale up with multiple instances and a load balancer. i have now 5 instances running of one AMI. but where do i have to attach my EBS volume to?
[13:55] <Guest56018> i don't have RDS nor S3, only an apache website with CPU-needing programs on it.
[14:01] <mamac> sisi
[14:01] <mamac> partyboy77, hello
[14:02] <partyboy77_> Mamac are you here ??
[14:03] <RoyK> pmatulis: because I'm blind :þ
[14:03] <mamac> partyboy77, yes I am
[14:10] <partyboy77_> Hello, someone knows the book "The Official Ubuntu Server Book" ??
[14:12] <ewook> I belive that might not be updated to the latest, if it resides within the community.
[14:12] <ewook> more then that, I do not know.
[14:15] <RoyK> partyboy77_: never heard of it :)
[14:16] <RoyK> partyboy77_: the review here was quite positive http://millionchimpanzees.blogspot.com/2009/08/official-ubuntu-server-book.html
[14:17] <RoyK> hm
[14:17] <RoyK> ven going through just this chapter, I got the impression that the book is geared for the newbie (or wannabe) Linux server administrator. If you have some experience in the Linux server (or just the server) admin arena, the challenge for you might only be getting up to speed with Ubuntu's tips and tricks
[14:17] <RoyK> s/ven/even/
[14:23] <partyboy77> I would know which is the best site to begin with Ubuntu Server ??
[14:23] <Pici> !serverguide
[14:24]  * wastl redirects partyboy77 to #ubuntu-server
[14:24] <wastl> *g*
[14:24] <Pici> wastl: thats here
[14:24] <wastl> oops
[14:24] <wastl> :m
[14:24] <partyboy77> :)
[14:24] <mamac> wastl, :-)
[14:25] <partyboy77> alcool is so sad.....
[14:25] <wastl> am on too many chans and am too busy here...
[14:25]  * wastl is still at work...
[14:26]  * wastl gives himself a double m) for this
[14:28] <partyboy77> Ok for Server Guide but is it possible to find it in pdf format ??
[14:29] <mamac> http://www.google.se/url?sa=t&source=web&cd=1&ved=0CBgQFjAA&url=https%3A%2F%2Fhelp.ubuntu.com%2F9.04%2Fserverguide%2FC%2Fserverguide.pdf&rct=j&q=ubuntu%20server%20guide%20pdf&ei=ob13TOnkLs-gOK3FkbsG&usg=AFQjCNHHiygn5_2a3DJ7OhOmUfMC7f_lkw&cad=rja
[14:29] <pmatulis> partyboy77: asking for the best starting point for learning about a CLI environment and then wanting it in PDF format is a bad start
[14:30] <partyboy77> okok
[14:30] <partyboy77> .....
[14:42] <MTecknology> what's the cheapest possible vps?
[14:43] <ewook> your own?
[14:52]  * Psi-Jack sighs.
[14:52] <Psi-Jack> Still on that eh, MTecknology?
[14:53] <MTecknology> Psi-Jack: I just +b -> ##fix_your_connection
[14:53] <Psi-Jack> Okay, and?
[14:53] <mamac> what does vps stand for?
[14:53] <MTecknology> If they come back I want to direct them to using a vps if they can't keep a reliable connection
[14:54] <Psi-Jack> mamac: Virtual Private Server
[14:54] <MTecknology> virtual private server
[14:54] <Psi-Jack> MTecknology: Heh. That's a bit overkill isn't it? :p
[14:55] <Psi-Jack> Some people's internet connections become temporarily unstable while, for example, their ISP is doing maintenance, or thunderstorms causing interuptions in service, or ... Well, various levels of possabilities.
[14:55] <MTecknology> Psi-Jack: it's the last thing I'll suggest to them - but I'd like to know in advance
[14:55] <MTecknology> It's been consistent for a few months
[14:55] <MTecknology> 200 joins/parts this month
[14:56] <MTecknology> sometimes 3 or more nicks in the channel
[14:56] <Psi-Jack> Sometimes, that's even freenode's IRC servers being unstable.
[14:58] <MTecknology> this is their own connection though
[15:02] <MTecknology> $6.95/mo for a VPS and then you'd also get to host a website
[15:02] <MTecknology> probably a php-free website
[15:06] <patdk-wk> MTecknology, you banning me? :)
[15:07] <MTecknology> patdk-wk: yes, you! get out!
[15:07] <MTecknology> patdk-wk: in lastlog I don't even see a join/part from you
[15:07] <patdk-wk> hmm, only two of me in here right now
[15:08] <Psi-Jack> heh, patdk-wk.
[15:08] <patdk-wk> heh, short lastlog list :)
[15:08] <patdk-wk> look for patrickdk :)
[15:08] <patdk-wk> isp died for just over an hour yesterday
[15:09] <MTecknology> patdk-wk: oh, that's still nothing
[15:09] <MTecknology> patdk-wk: http://dpaste.com/234841/]
[15:09] <MTecknology> I ran /names 3x
[15:10] <patdk-wk> those are fun
[15:10] <Psi-Jack> Heh, well, so far I got my pacemaker running drbd with gfs2 successfully. :)
[15:10] <Psi-Jack> Now all I gotta do is figure out how the frack to get clvm working with it. heh
[15:10] <patdk-wk> hmm doesn't know if it's wise to hack you pacemaker
[15:11] <patdk-wk> are you scared of killing you heart?
[15:11] <Psi-Jack> heh
[15:31] <iceflatline> hi all, I'm installing ubuntu server on vmware; any recommendations for minimum mem allocation and hard disk space?
[15:31] <thesheff17_> I did aptitude install multipath-tools and it didn't create an /etc/multipath.conf file.  Where can I find this conf file?
[15:34] <thesheff17_> iceflatline: runs fine w/ 512 megs of ram.  Just scale up what you are running on the server from there.
[15:35] <iceflatline> thesheff17, gotcha, thanks much.
[15:36] <hggdh> iceflatline: depends on what you want to do; for simple tests, I usually boot a 256M instance; for heavier tests, usually a 512
[15:38] <iceflatline> hggdh: small home svn and web server - nothing too killer.  First time with VMware though...
[15:41] <patdk-wk> iceflatline,  what are you going use it for?
[15:41] <patdk-wk> hmm, web could use some mem, all depends
[15:41] <patdk-wk> vmware is good at swapping mem though
[15:41] <MTecknology> iceflatline: I had my first web server on about 16MB RAM - It didn't go fast.. but it worked - now I'd suggested a minimum of ~64MB for Ubuntu server + Nginx
[15:42] <MTecknology> no.. it was 8MB..
[15:42] <MTecknology> I think, I know I decked it out
[15:42] <patdk-wk> smallest I could get ubuntu server to use was 24megs ram :)
[15:42] <iceflatline> patdk-wk: thanks. I've got plenty of sys mem, so would prefer to avoid swapping if I can. I was considering bumping virtual mem to 1024
[15:43] <patdk-wk> iceflatline, is this vmware workstation? or esxi?
[15:43] <iceflatline> patdk-wk: server
[15:43] <patdk-wk> esxi, just overallocate like nuts
[15:43] <iceflatline> :)
[15:43] <patdk-wk> hmm, dunno about server, thought it was dead :)
[15:44] <iceflatline> k, I'll start with the (vmware) recommended 512 and see how it feels. Thanks for the help guys.
[15:46] <ttx> hggdh: around ?
[15:50] <hggdh> ttx: here
[15:51] <ttx> hggdh: about the uec-testing spec. The items on your shoulders are still up for completion by next week ?
[15:51]  * hggdh goes to look at the spec
[15:51] <newz2000> help, I have a job that won't finish unless I give it more ram, can I take a running kvm and add more ram to it w/out restarting the jobs?
[15:54] <hggdh> ttx: of the 3 I have: TAP -- I will mark done; elastic IP -- will do it this week; 2.0 code drops -- I think we can mark this as done?
[15:55] <ttx> hggdh: yes, please do now
[15:55] <hggdh> ttx: also, I noticed kirkland has the basic EBS testing -- Sylvain's script does it, so maybe we can mark it done?
[15:55] <ttx> I need updated status for the release meeting starting in ... 4 min
[15:56] <ttx> hggdh: sync that with kirkland, but i'd say yes, do it
[15:57] <hggdh> ttx: updated, including kirkland's task. I will sync with him later on
[15:57] <ttx> hggdh: ack, thanks
[16:01] <Psi-Jack> Hmm blast-it all to heck.
[16:02] <Psi-Jack> Seems like I'm going to have to use both cman and pacemaker just to get clvm working. :/
[16:02] <Psi-Jack> At least I don't have to use rgmanager though. :0
[16:12] <smoser> ttx, with no ebs root support in eucalyptus, i dont see any huge value in euca2ools 1.3
[16:12] <smoser> so i would say stick with 1.2
[16:13] <hggdh> smoser: I agree, but I would still like to run (for a test) 1.3
[16:13] <ttx> smoser: that is quickly becoming our common position
[16:38] <Psi-Jack> Is it even possible to run clvm without cman?
[16:43] <smoser> Hey. i'm looking for suggestions...
[16:44] <Psi-Jack> Suggestion: Ask your question.
[16:44] <smoser> I have been running, for quite a while, tests from lp:~smoser/+junk/ec2-test
[16:44] <smoser> that really needs to move to the ubuntu-on-ec2 project or somewhere, but ignore that for the moment.
[16:44] <smoser> what i'm looking for suggestions on is that it creates logs of the test results.
[16:45] <smoser> I'd like to store those logs somewher other than my laptop :)
[16:45] <smoser> a bzr branch along side of the test suite, seems to be a reasonable place, but the logs are somewhat large (I currenlty have 86M of logs)
[16:46] <G> Psi-Jack: I'm pretty sure no
[16:46] <smoser> they're text, so they compress well, but compressing them before checking them in means that a.) they're less useful and b.) they're binary blobs in a revision control system.
[16:46] <G> Psi-Jack: I worked w/ cman a lot w/ my old job, and I'm pretty sure clvmd has to use cman for locking
[16:46] <smoser> hggdh, i think that you are checking in logs larger than that, though, is that right ?
[16:47] <G> Psi-Jack: I'll just have a quick look
[16:47] <Psi-Jack> G: Actually, I just did. Just that the ubuntu clvmd init script checks for /etc/cluster/cluster.conf and cman_tool status, but cman communicates through openais and dlm_controld for /everything/, which I already have pacemaker bringing up dlm_controld and it's of course using openais.
[16:48] <Psi-Jack> Err... clvmd communicates through...
[16:50] <Psi-Jack> So far though, now that I took out those initial checks in clvmd, it's running and appearing to function without cman at all.
[16:53] <Psi-Jack> It's just like people think gfs2 requires cman to work, because of gfs_controld, but it's not required, as I'm initiating gfs_controld from pacemaker. ;)
[16:53] <G> Psi-Jack: well the reason for that is because it's the only solution that Red Hat officially supports (why people think it's the only way)
[16:54] <Psi-Jack> Yep. :)
[16:54] <Psi-Jack> Pacemaker's definitely come a looooooong way.
[16:54] <Psi-Jack> As has it's documentation, for once I finally understand it!
[17:00] <Psi-Jack> And there's even an ocf provider lvm2 which has clvmd. ;)
[17:10] <SpamapS> Psi-Jack: how did your tests go with the pacemaker client?
[17:10] <Psi-Jack> SpamapS: It works great!
[17:10] <Psi-Jack> I didn't know pacemaker-mgmt was so nice, but it's pretty decent. Much better than pacemaker-pygui used to be.
[17:12] <SpamapS> Psi-Jack: Do you know how to upload to a PPA? Would be awesome to have one available since we probably won't be syncing it into Maverick.
[17:12] <Psi-Jack> SpamapS: I don't know how to, presently, no.
[17:13] <hggdh> smoser: oh yes. Even compressed, I checked in logs adding up to ~80M
[17:13] <SpamapS> Psi-Jack: https://help.launchpad.net/Packaging/PPA
[17:13] <smoser> and this is what you ended up settling on, eh?
[17:14] <Psi-Jack> SpamapS: Alrighty. I'll look into that in a bit. ;)
[17:14] <hggdh> smoser: yes, but much against my will. I found nothing else...
[17:14] <smoser> yeah. and you check in binaries, right ?
[17:15] <smoser> the .tar.gz rather than the source
[17:15] <hggdh> I check in tar.gz
[17:15] <smoser> yeah
[17:15] <hggdh> of course, every new checkin is a brand new checkin...
[17:15] <smoser> why did you choose to do that ?
[17:15] <smoser> versus checking in the tar
[17:15] <smoser> err.
[17:15] <smoser> not the tar, but the text output
[17:15] <hggdh> because of the time it would take to upload them all
[17:16] <hggdh> each long-running test adds up to 300M of logs
[17:16] <smoser> hm..
[17:16] <smoser> is that true ?
[17:16] <hggdh> well, I am playing it safe, so I save *all* logs
[17:16] <smoser> a bzr push should send the repo, not the checked out files, and the repo may be compressed for you
[17:17] <smoser> git can do wonderful things in that regard.
[17:17] <smoser> lifeless, ?
[17:17] <hggdh> oh
[17:17] <hggdh> bloody hell, I forgot that bzr does a good part of it
[17:17] <hggdh> :-(
[17:17] <smoser> is there reason for me to compress (ie, create a .tar.gz) before checking something in ? will that gain me anything ? versus checking in a large bunch of text files
[17:18] <smoser> git does a fantastic job. i really dont have any experience with bzr in this regard.
[17:18] <RoyK> smoser: it'll break svn diff
[17:18] <smoser> well, i dont think svn diff would work anyway :) (its bzr)
[17:18] <smoser> :)
[17:18] <RoyK> k
[17:19] <smoser> but yeah, i am just wondering if there is *any* benefit to doing it.
[17:19] <hggdh> and, in my case, every new log upload is a Brand New Directory, so no diffs
[17:19] <smoser> at first blush, it seems like you're compressing something, so that will smaller.
[17:19] <hggdh> smoser: I can try
[17:19] <smoser> hggdh, but even the new directory may be compressed.
[17:20] <smoser> ie, you could store *everything* in .bzr in a packed/compressed format
[17:20] <hggdh> right now I tar -cz every thing I upload
[17:20] <smoser> that is what git does.
[17:20] <smoser> right.
[17:20] <smoser> i'm just saying, that may buy you nothing savings in local storage.
[17:20]  * RoyK sniggers http://karlsbakk.net/top-24.png
[17:21] <hggdh> yeah. I have to test it
[17:22] <smoser> above that should have said "nothing but savings in local storage of the checked out branch"
[17:25] <hggdh> which might be enough for me, with each test run adding in a few hundreds M
[17:26] <hggdh> still, I will check it
[17:34] <smoser> hggdh, my queries in bzr indicate that you should check in the full log files
[17:34] <zul> smoser: couldnt you store the diff?
[17:34] <smoser> well, in this case (hggdh's and mine) there is no diff, really.
[17:35] <smoser> and in both cases, even if there were, thats basically what bzr is trying to generically do
[17:37] <hggdh> smoser: I will try it on next save
[17:40] <squidly> anyone hwere use libvirtd? I'm tying to do a couple of tests mostly moving servers between kvm hosts.. but I keep getting an error about the monitor not shoing up
[17:40] <squidly> this happens if I do it with a win VP or an ubuntu VM
[17:47] <n0a1ias> hey, I want a server, and i want it to run ubuntu, but i dont know what i want to do with it
[17:47] <n0a1ias> what are some of the things i could do with ubuntu server
[17:47] <mickster04> n0a1ias: lol, what a question
[17:48] <squidly> n0a1ias: what ever you want
[17:48] <n0a1ias> mickster04, yea i know, sorry ha
[17:48] <mickster04> n0a1ias: web server ssh server, samba, lamp, the list is endless
[17:48] <n0a1ias> shh? lamp? sambaa?
[17:48] <mickster04> n0a1ias: you could install a steam dedicated sever for many of its games, infact any online game will prolly have a server program that you can install]
[17:49] <mickster04> n0a1ias: woah...ok you need to go to the ubunt server site and do some reading...
[17:49] <n0a1ias> yea im on it now as well as here
[17:49] <mickster04> if you don't know what it can do or what you wanna use it for why have it, you can do most things with ubuntu anyway
[17:50] <squidly> why not start out with something basic.. like a webserver
[17:50] <n0a1ias> because i have an old computer and dont know what to do with it
[17:50] <n0a1ias> ok i dont have a website, i guess i better make one
[17:52] <n0a1ias> once i have a website how would i put it on the server?
[17:53] <n0a1ias> like im using webs.com
[17:53] <smoser> n0a1ias, or maybe you want to learn assembly programming, or write documentation, or learn python/C/perl/. web hosting is not the only thing you can do.
[17:54] <mickster04> n0a1ias: with all due respect this is a help channel not a hold your hand channel. there are plenty of good guides out there...
[17:54] <smoser> figuring out what you *want* to do is probably a good place to start :)
[17:55] <n0a1ias> haha, im not into programing, in taking cico classes and im gonna be a computer engineer, so is there anything i would need for that?
[17:55] <squidly> smoser: that only place to start really.. unless you want a server there doing nothing except warming your house and using more power
[17:55] <smoser> cico == cisco ?
[17:55] <n0a1ias> wel im am cold....
[17:56] <webwurst> n0a1ias: and here is the link regarding ubuntu-server http://doc.ubuntu.com/ubuntu/serverguide/C/ :)
[17:56] <n0a1ias> yes, im on a netbook with a small keyboard, sorry
[17:56] <mickster04> n0a1ias: yeah, when my room gets cold I play some games to make it warm me upD
[17:56] <n0a1ias> ha
[17:56] <smoser> if you're interested in networking, there are loads of things you can do. you could run an VPN on it, play with routing, learn ipv6.
[17:57] <webwurst> n0a1ias: you could setup a webserver yourself as a start http://doc.ubuntu.com/ubuntu/serverguide/C/httpd.html
[17:58] <n0a1ias> i like the idea of vpn, but i use my home network for experimentation.
[17:58] <squidly> and that is just for starters
[17:59] <n0a1ias> well here, lets start at the begining
[17:59] <n0a1ias> what is a server?
[18:00] <webwurst> n0a1ias: pls read the server-guide and ask what you dont' understand ;)
[18:00] <n0a1ias> ok.
[18:00] <mickster04>  n0a1ias again this isn't a kindergarten of servers....
[18:00] <webwurst> n0a1ias: you know wikipedia?
[18:01] <n0a1ias> i know
[18:01] <n0a1ias> yea, i know what a server is, im just putting the deffinition out there so i can find out what i what it to do
[18:02] <shauno> n0a1ias, wrong way around imho.  if you decide it's a hammer now, you're just going to bend a lot of nails
[18:02] <shauno> figure out what you want to do, and then worry about getting the tools to play along
[18:03] <n0a1ias> isnt that what im doing? i think i might go for a shh but im not shure btw
[18:04] <mickster04> n0a1ias: you really don't have any need of a server therefore it is a waste of electricity
[18:04] <webwurst> n0a1ias: yeah, just do it http://doc.ubuntu.com/ubuntu/serverguide/C/openssh-server.html
[18:04] <mickster04> n0a1ias: you can install normal ubuntu on older hardware too, or look at things like puppy linux
[18:04] <squidly> pick something to learn and learn it
[18:05] <n0a1ias> i know but i want a server, if for no other reason than that im still doing it
[18:06] <n0a1ias> could i have it do multiple things?
[18:06] <mickster04> n0a1ias: if you want to install a frontend onto the server (a gui) you may aswell install ubuntu and install the server type packages oontop of that#
[18:06] <webwurst> n0a1ias: do it! and then ask if it does not work
[18:06] <mickster04> n0a1ias: of course
[18:06] <n0a1ias> ok thanks, im going with shh, and a proxy server
[18:07] <mickster04> n0a1ias: have fun
[18:13] <smoser> hggdh, so, a little test and results here:
[18:13] <smoser> $ du -hs .bzr
[18:13] <smoser> 4.1M    .bzr
[18:13] <smoser> $ ls -lh *.tar.gz
[18:13] <smoser> -rw-r--r-- 1 smoser smoser 420K 2010-08-27 13:12 hardy-server-daily-20100827.tar.gz
[18:13] <smoser> -rw-r--r-- 1 smoser smoser 433K 2010-08-27 13:12 karmic-server-daily-20100826.tar.gz
[18:13] <smoser> -rw-r--r-- 1 smoser smoser 975K 2010-08-27 13:12 lucid-server-daily-20100827.tar.gz
[18:13] <smoser> $ du -hs *2[0-9]
[18:13] <smoser> 3.2M    hardy-server-daily-20100827
[18:13] <smoser> 3.2M    karmic-server-daily-20100826
[18:13] <smoser> 7.4M    lucid-server-daily-20100827
[18:14] <smoser> hte bzr repo is tracking the directories. i just created the tarballs to see what size they would be themselves.
[18:14] <smoser> (ie, if i checked them in).
[18:14] <smoser> so, the checkout is 7.4 + 3.2*2 = 13.8 M.
[18:15] <smoser> the tarballs would be 1.8M
[18:15] <smoser> and the .bzr dir is 4.1M
[18:15] <mickster04> is there a non-caching proxy for ubuntu?
[18:16] <smoser> mickster04, i would guess that you could turn caching off on squid or robably any other proxy.
[18:17] <smoser> hggdh, when i checked those tarballs into a new repo, the .bzr dir was 1.9M.
[18:17] <smoser> so, this is just "initial state", but the .tar.gz checkins do benefit you at least there. maybe later in repo life you'd get better delta storage.
[18:17] <mickster04> smoser: well lasst tim i tried when a user was denied access it still mention cache rather than proxy
[18:18] <smoser> i really odnt have any experience with it, sorry. it just would seem like you could set the cache size to 0 or very low and essentially kill it
[18:19] <mickster04> smoser: yeah thats what i thought. i wonder if theres a way of testing it
[18:19] <tynot> por favor, fresh 10.04.1 server install. setup local ip, sn & dns (off-site). can ping everywhere, but dns not working. couldn't download any packages. help?
[18:20] <smoser> tynot, your /etc/resolv.conf has a dns ?
[18:21] <tynot> on a macbook pro in the same local network. best way to term into it to check?
[18:22] <tynot> smoser, i thought i entered it.
[18:24] <smoser> 'term into it' ? ssh ip.address
[18:24] <mickster04> smoser: does mac have an ssh server?
[18:24] <smoser> my guess is that you put in a bad dns server, or cannot reach the dns server.
[18:24] <smoser> i have no idea. but my guess is yes.
[18:38] <squidly> yes it does
[18:41] <mickster04> squidly: what does?
[18:46] <zul> smoser: there isnt a <name>-initrd-virtual anymore in maverick?
[18:46] <smoser> silly you
[18:46] <smoser> that was gone in lucid
[18:46] <smoser> uec-publish-tarball is what you want to use.
[18:46] <zul> smoser: its in the readme though
[18:47] <smoser> what README ?
[18:47] <smoser> does it not say optional ?
[18:47] <smoser> http://bazaar.launchpad.net/~ubuntu-on-ec2/vmbuilder/automated-ec2-builds/annotate/head%3A/README.files ?
[18:47] <zul> ah yes...i didnt read that part
[18:47] <smoser> :)
[18:56] <lifeless> smoser: not sure what you're asking
[18:58] <smoser> well, basically, hggdh is checking into a bzr repo a bunch of .tar.gz files that contain a directory full of logs (text)
[18:58] <smoser> he's doing that in an effort to save disk and bandwidth
[18:59] <smoser> i had hoped/thought that it not actually save .bzr/ space or bandwidth, and thus he could simply check in uncompressed text logs.
[19:00] <smoser> my test shows that bzr does an OK job (gets some compression) but not a great job.  in the initial commit the .bzr directory is smaller by checking in .tar.gz files than by checking in dir/*.txt files.
[19:00] <smoser> before testing it, i was asking you if checking in .tar.gz files was pointless.
[19:00] <smoser> my test did only initial check in, i'm not sure if future commits would see better deltas with text than with tar
[19:00] <smoser> that all make sense ?
[19:01] <lifeless> checking in .tar.gz is pointless
[19:01] <lifeless> it is in fact counterproductive because zlib fucks with the data so that its extremely hard to compress it further
[19:02] <lifeless> bzr and git use the same underlying compressor; we do different things on top, but I'd expect pretty similar results
[19:02] <smoser> bzr did better out of the gate than .git did.
[19:02] <smoser> but git repack came away ahead
[19:03] <smoser> again, this is only after the initial commit.
[19:03] <lifeless> sure
[19:03] <smoser> its not completely counterproductive, as my numbers show above
[19:03] <lifeless> smoser: du -sh .bzr is often an inaccurate count
[19:04] <lifeless> particularly if you're trying to assess bandwidth and space over time
[19:04] <axisys> anyone know the ubuntu channle name for ubuntu virtualization talk ?
[19:04] <smoser> axisys, either here or ubuntu-cloud
[19:04] <lifeless> if you want to measure active storage you need to exclude .bzr/repository/{obsolete-packs,upload}
[19:05] <axisys> smoser: i think it something like #ubuntu-virt.. or something.. i forgot
[19:05] <smoser> axisys, http://www.mail-archive.com/ubuntu-server@lists.ubuntu.com/msg04352.html
[19:05] <axisys> smoser: no wonder :-)
[19:05] <axisys> smoser: thanks
[19:06] <axisys> smoser: i was trying to go to #ubuntu-virt.. should have redirected me to #ubuntu-server .. may be it did and I did not pay attention
[19:07] <smoser> axisys, it did
[19:07] <smoser> you're here :)
[19:07] <axisys> smoser: i usually login to #ubuntu-server by default.. may be that is why i did not notice.. thanks
[19:09] <axisys> i have a laptop with intel core 2 duo 2.4ghz, 8gb ram and 750gb hard drive.. i want to setup my private cloud in here while still have gnome (after all this is my laptop) .. what path is recommended for that ?
[19:09] <axisys> should i install desktop and then use vbox to install ubuntu cloud ?
[19:10] <axisys> or should i do desktop + kvm and do the cloud using kvm ?
[19:10] <axisys> too many choices.. :-)
[19:12] <SpamapS> can we teach uvirtbot things?
[19:12] <SpamapS> !uvirtbot
[19:12] <SpamapS> uvirtbot: can you learn?
[19:12] <SpamapS> uvirtbot: launchpad
[19:12] <Pici> Why? ubottu is here.
[19:12] <SpamapS> Oh I meant ubottu
[19:12] <SpamapS> uvirtbot: sorry
[19:12] <Pici> !usage
[19:13] <SpamapS> should really teach ubottu what happend to ubuntu-virt ;)
[19:16] <axisys> SpamapS: probably a good idea or smoser have to keep posting that mail for all newbies like me :-)
[19:17] <axisys> s/have to/has to/
[19:18] <SpamapS> smoser: ^^
[19:20] <smoser> ?
[19:20] <smoser> i have no idea how to teach that finicky old ubuttu
[19:20] <smoser> ubuttu, read the above and teach yourself.
[19:23]  * SpamapS admits his puny attempt at delegation failed
[19:26] <squidly> what happened to ubuntu-virt?
[19:27] <soren> squidly: The IRC channel?
[19:27] <soren> squidly: It was folded into this one.
[19:39] <squidly> ahh okies
[19:45] <n0a1ias_> so would a samba server be a secure print server?
[19:47] <mdeslaur> http://www.serverwatch.com/trends/article.php/3900711/The-Top-10-Linux-Server-Distributions.htm
[19:48] <ivoks> mdeslaur: yay!!!
[19:49] <n0a1ias_> so would a samba server be a secure print server?
[19:50] <ivoks> no
[19:50] <ivoks> cups is a secure print server
[19:50] <ivoks> even windows can print on cups shard printers
[19:51] <n0a1ias_> oh, then whats samba?
[19:53] <ivoks> samba is file sharing
[19:53] <ivoks> it can do printer sharing too
[19:53] <n0a1ias_> oh.. thanks
[20:47] <n0a1ias_> whats a good vpn server that i can install on ubuntu normal?
[20:47] <zash> n0a1ias_: openvpn
[20:47] <n0a1ias_> thanks
[20:57] <Fredrick> hey, my other nick is n0a1ias, i had to move to the computer im installing openvpn on
[20:58] <Fredrick> who was the one who suggested that to me?
[21:06] <RoyK> freeflying: zash
[21:07] <RoyK> freeflying: ops - wrong
[21:07] <zash> whowhatnow?!
[21:07]  * zash said "openvpn"
[21:09] <SpamapS> phew.. 12 bugs triaged.. i think thats a new single day record for me
[21:10] <zash> SpamapS: yay, gratz! \o/
[21:17] <n0a1ias> is there an open vnc room?
[21:17] <n0a1ias> cuz im lost?
[21:27] <patdk-wk> yay, loading launchpad, googlemaps rejected me :)
[21:27] <patdk-wk> "The Google Maps API server rejected your request. The "client" parameter specified in the request is invalid."
[21:30] <ScottK> You can turn the map off.
[21:31] <patdk-wk> http://gawker.com/5537406/how-to-sneak-a-dirty-joke-into-the-new-york-times
[21:31] <patdk-wk> scottk, I know
[21:58] <n0a1ias> how do i get my ubuntu to be an anonymous proxy server like freeproxyserver.com?
[22:02] <mickster04> apt-get dist-upgrade will update to a newer version of ubuntu right (think 9.10 to 10.04)
[22:06] <panfist> i was in here asking about umasks for nfs shares, and i guess the general answer is that the system umask applies to local and nfs file systems
[22:06] <panfist> in general, i want an nfs share where all users on my network can collaborate, regardless of group membership
[22:06] <panfist> but, i don't want to change the umask of each system that will compromise a user's own files
[22:06] <panfist> so what's the correct (or a good) practice in ubuntu for creating a shared network space where any users are free to read/write each other's files, just in that one space and not in general?
[22:07] <Nafallo> ivoks_away: happy birthday! :-)
[22:19] <patdk-wk> mickster04, nope, not at all
[22:22] <mickster04> patdk-wk: oh what is then
[22:25] <patdk-wk> do-release-upgrade
[22:47] <binBASH> hi patdk-wk
[22:47] <patdk-wk> hey
[22:54] <n0a1ias> anybody ever use squid-cache?
[22:54] <patdk-wk> sure
[22:54] <n0a1ias> great!
[22:55] <n0a1ias> you cool with pm?
[22:55] <patdk-wk> now that you used your 1 question allocation, move along :)
[22:55] <n0a1ias> that way all the info is in the same place
[22:55] <n0a1ias> hah
[22:55] <n0a1ias> wait really?
[22:56] <n0a1ias> hello?
[22:58]  * patdk-wk doesn't know pm
[22:59] <patdk-wk> power management?
[22:59] <n0a1ias> no, privet message
[22:59] <patdk-wk> those are evil
[22:59] <n0a1ias> oh, ok
[22:59] <n0a1ias> so no.
[22:59] <n0a1ias> anyway
[23:00] <n0a1ias> i just installed squid
[23:00] <n0a1ias> how do i control it?
[23:00] <patdk-wk> depends on your definition of control
[23:01] <patdk-wk> /etc/squid/squid.conf
[23:01] <patdk-wk> the squid cgi?
[23:01] <n0a1ias> yea the cgi
[23:01] <patdk-wk> dunno, I don't use the cgi
[23:01] <n0a1ias> what do you use? cli?
[23:02] <patdk-wk> the config file :)
[23:02] <n0a1ias> it says command not found
[23:03] <n0a1ias> what is squid used for?
[23:04] <n0a1ias> like i know its for taking pages and saveing them to make internet faster, but how does it make you more secure?
[23:04] <patdk-wk> I dunno how it makes you more secure
[23:05] <n0a1ias> oh
[23:05] <patdk-wk> unless you tell it, never to allow anyone to download exe files
[23:05] <patdk-wk> but that isn't much security
[23:05] <n0a1ias> why do you use it?
[23:07] <n0a1ias> why do you use squib?
[23:07] <n0a1ias> and if i download it, then dont configure it, will it do anything at all?
[23:08] <patdk-wk> it will do some, but not much
[23:08] <n0a1ias> hmm
[23:08] <patdk-wk> I use it to cache the crap out of .deb, .iso, all images, and stuff
[23:08] <patdk-wk> gives the work here about a 15% speed boost
[23:08] <n0a1ias> i dont understand
[23:09] <patdk-wk> I don't know why you are bothering with squid
[23:09] <n0a1ias> me either but now i want to know
[23:09] <patdk-wk> cause you can totally screw your internet up with it (atleast websites)
[23:10] <n0a1ias> so when you need a .iso, you use the cached one and its faster then downloading and saving one?
[23:10] <patdk-wk> only if you already downloaded it before
[23:10] <n0a1ias> hmm
[23:11] <n0a1ias> ok well i save all my disks so i guess i wouldnt really use it
[23:11] <n0a1ias> thanks for the help
[23:11] <patdk-wk> now, caching image files helps everyone though
[23:11] <n0a1ias> ??
[23:12] <patdk-wk> all the pictures on websites people browse
[23:12] <patdk-wk> those are reused a crapload
[23:12] <n0a1ias> o those
[23:12] <patdk-wk> but you get into issues there
[23:12] <patdk-wk> can your harddrive seek and load them, faster than you can download them :)
[23:12] <n0a1ias> and does it do that with out being configured?
[23:12] <patdk-wk> it will, but I forget the default times it uses for them
[23:13] <patdk-wk> I have mine set to pretty long
[23:13] <patdk-wk> if I download a picture, it goes int othe cache, and it wont bother to ask if it has changed even, for 12hours
[23:13] <patdk-wk> by default, I think it's 30min or something
[23:13] <patdk-wk> but it still asks if it changed
[23:14] <n0a1ias> oh, ok so i should change that
[23:14] <n0a1ias> but id set it to 6 hours
[23:14] <n0a1ias> so to do that i cd into the config folder?
[23:15] <patdk-wk> na, you edit the config, and restart squid
[23:15] <n0a1ias> wheres the config?
[23:18] <n0a1ias> wheres the config
[23:28] <n0a1ias> this is ubuntu-server?
[23:47] <Hopsa> Hi people! My Ubuntu Lucid is fully updated and dhcpd segfaults. Valgrind sais its due to an "Invalid read of size 4". Is this a known issue or ?
[23:48] <n0a1ias> this is ubuntu-server, you want ubuntu
[23:48] <n0a1ias> try /join #ubuntu
[23:49] <Hopsa> ?
[23:49] <n0a1ias> type this
[23:49] <n0a1ias>  /join #ubuntu
[23:49] <Hopsa> go away please
[23:49] <n0a1ias> only without the space infrint
[23:49] <Hopsa> Have you a fritz :)
[23:50] <ScottK> n0a1ias: Please don't assume people aren't running servers.
[23:50] <ScottK> Hopsa: "Go away please" isn't very nice either.
[23:50] <ScottK> Come on.
[23:51] <Hopsa> See it from my side of the planet ;)
[23:52] <n0a1ias> oh, sorry
[23:52] <n0a1ias> are you?
[23:52] <n0a1ias> didnt mean to sound mean
[23:52] <ScottK> According to previous discussion on #ubuntu-devel, yes.
[23:53] <n0a1ias> once again, sorry
[23:53] <Hopsa> ScottK. Any chance that dhcpd will be updated so it has a chance to function properly on Lucid ?
[23:54] <ScottK> Hopsa: I don't know.  It's not a package I personally pay a lot of attention to.  The advice you've already gotten to file a bug is good advice.
[23:55] <Hopsa> Hmm, id pay attention to any critical package if i where you.
[23:55] <ScottK> Why should I?  It's not critical for me.
[23:58] <Hopsa> Inhabitateem est
[23:58] <Hopsa> Corriculaare
[23:59] <Hopsa> ScottK: Do you know youre in #ubuntu-server and doesnt feel a server segfaulting is bad ?