[00:02] <theyranos> something is sending sigterm to anything that listens on port 22
[00:04] <ethicalhack3r> hi Guys, any one know how I install my remastered ubuntu server livecd to harddrive?
[00:05] <ethicalhack3r> is there a built in tool or external script I can use?
[00:05] <JasonMSP> I've got VSFTPD setup and working with TLS/SSL.  I'd like to configure it for virtual users on top of this.  I've tried a few solutions that have not worked.  I get login 530 error.  Does anyone have this working?
[00:09] <kevinash> Can anyone suggest a good backup script for MySQL 5.1 databases for use on Ubuntu Server 10.0.4? I've found a few on google but I'd like to know someone is using their chosen script with good results
[00:12] <Four2zero> hey guys, i was running samba for filesharing with windows 7 and now i installed ebox(zyntal) and was trying to configure it with samba but it woudn't work. now im stuck with multiple samba network on my windows 7 machine that will not go away, how can i make sure samba is still not running or installed ?
[00:13] <Four2zero> I ran top and i dont see any smb services
[00:24] <Skaag> can someone remind me how to tell apt not to upgrade a certain package?
[04:26] <JasonMSP> trying to setup virtual users with VSFTPD using this http://viki.brainsware.org/?en/Virtual_Users_simple I get this error 530 Login incorrect.
[04:26] <JasonMSP> Login failed.
[04:39] <_Techie_> JasonMSP, whats wrong with creating real users?
[04:43] <pedahzur> Not strictly server related, but this is on Ubuntu server, so....: Weird dependency problem: trying to install gitosis. Says it can't install because Depends: git-core (>= 1:1.5.0) but 1:1.7.3-0ppa2~hardy1 is installed.  Ideas?
[04:46] <JasonMSP> I was hoping for easy managemnt of users by only allowing them access using httpaswd to create users.  Basically it would keep the ftp users all in one place and not in the passwd file.
[04:47] <JasonMSP>  _Techie_: I expect as things progress on our server that we will have more and more FTP users.  I don't want them to have shell access, or any other access to the system.
[04:55] <shey> Question: I have a webcam on my Ubuntu Server in my office, and was wondering if there was a way I can install some sort of monitoring software so I can see my office, from home lets say. There is no X Server installed on it.
[04:55] <shey> Is something like this possible?
[04:56] <ecelis> yes it is posible
[04:57] <shey> Well, this is linux, I am sure everything is possible, care to elaborate?
[05:00] <ecelis> depending on your webcam, you might make it to take still images and store them somewhere an then write some PHP or web front end
[05:01] <ecelis> which can use imagemagik to create clips for  you to download
[05:01] <ecelis> or maybe an streaming solution such as videolan
[05:01] <ecelis> I have done the first one
[05:01] <ecelis> google a bit, there are lots of resources about this out there
[05:02] <erichammond> shey: zoneminder is available in Ubuntu and seems to be popular, but I recommend the  "motion" software which is much easier to set up and can create simple video files to peruse.
[05:35] <Datz> humm, just checking munin stats today. would there be any alarm seeing number of threads take a sudden jump? http://datzfunk.net/munin/localdomain/localhost.localdomain/threads.html
[05:36] <Datz> I just installed some updates today, I don't know if it would be related to any of them..
[05:47] <twister004> hi guys!...
[05:47] <twister004> hi guys!...
[05:47] <twister004> have a query on BIND in ubuntu... if i add a CNAME in the db.xx.xx file, do I need to restart/reload the BIND service on the DNS server?
[05:51] <ehcah> Hello. Can anyone help me with the following error: "could not create key from $KEY_HMAC: bad base64 encoding". I have been systematicaly working through errors on getting DDNS up and running.
[06:12] <SpamapS> ehcah: you're trying to do DDNS using rndc ?
[06:13] <SpamapS> twister004: reload, yes. Did you also make sure to change the serial number?
[06:14] <ehcah> yes
[06:14] <ehcah> I'm down to my final error.
[06:14] <ehcah> I've been at this for 8 hours.
[06:15] <RoAkSoAx> SpamapS: where's the HOWTO to set the loadbalancer in the cloud?
[06:16] <SpamapS> RoAkSoAx: still working on it. https://wiki.ubuntu.com/CloudLoadbalancingHowto
[06:16] <SpamapS> RoAkSoAx: I packaged clb today so it should make the instructions much simpler.
[06:17] <RoAkSoAx> SpamapS: ty :). Yeah I just saw it :). How do you feal about automatically starting/stoping webservers based on the load?
[06:17] <SpamapS> RoAkSoAx: I was discussing it with mathiaz though, and really, with the puppet stuff he's rolling out, it should be really easy to do puppet in the cloud.
[06:17] <SpamapS> RoAkSoAx: there are already a number of autoscaling solutions out there.
[06:18] <RoAkSoAx> SpamapS: oh I see
[06:18] <SpamapS> RoAkSoAx: I'm not really very proud of clb. It needs to switch to using basic AUTH.. and SSL
[06:19] <SpamapS> RoAkSoAx: but its purely a proof of concept at this point.
[06:19] <RoAkSoAx> SpamapS: well, that's how everything starts though :)
[06:21] <SpamapS> RoAkSoAx: my original goal was to make tools that worked just like the amazon ELB tools
[06:21] <SpamapS> actually i have some time this week, maybe I'll try that again
[06:22] <RoAkSoAx> SpamapS: well you can achieve that overtime
[06:22] <RoAkSoAx> SpamapS: if I knew how that works I'd help but unfortunately I don't since I don't use the cloud yet :(
[06:24] <RoAkSoAx> since I don't really have the resources for this now
[06:39] <sandGorgon> anybody know what is the upgrade path for postgres 8.4 -> postgres 9.0 is on Ubuntu ? The problem is that installing postgres-9.0 overwrites the pg_config binary and so my upgrade command fails : http://ubuntu.pastebin.com/0FXhQUZC
[08:18] <ttx> SpamapS: I think the graph is still buggy
[08:19] <SpamapS> ttx: maybe we should get one of those UV zappers for it?
[08:19] <ttx> The solid line points in the middle of the inprogress now
[08:19] <ttx> and the dotted line points to something HIGHER than  the total amount of WI
[08:20] <ttx> I'm not exactly sure where that extra amount comes from though
[08:20] <SpamapS> ttx: hmm, works fine on my personal page..
[08:20] <SpamapS> http://people.canonical.com/~pitti/workitems/maverick/u/clint-fewbar-ubuntu-10.10.html
[08:20] <SpamapS> so maybe its a team thing
[08:20] <ttx> SpamapS: http://people.canonical.com/~pitti/workitems/maverick/canonical-server-ubuntu-10.10.html
[08:20] <ttx> SpamapS: it's recent
[08:21] <SpamapS> hm maybe the paints are backwards for foreign/team
[08:21] <SpamapS> I think they are
[08:21] <ttx> rigth, therte shoudl be some foreign things
[08:21] <ttx> the graph don't show them anymore
[08:22] <SpamapS> ttx: I think they're behind the other stuff
[08:22] <ttx> so the three TODO's from hggdh don't add up on the bar
[08:23] <ttx> hmm, the difference is not "3" though
[08:24] <ttx> more like "1"
[08:24] <SpamapS> it may be cumulative
[08:25] <SpamapS> the foreign numbers are not missing in the previous milestone tho...
[08:25] <ttx> SpamapS: we need to fix it because it adds confusion on the meaning of the lines... I can have a shot at it if you want
[08:26] <ttx> hmm
[08:27] <ttx> I think it was introduced when I added TODOs on james plate
[08:27] <SpamapS> ttx: should be pretty simple.. just 3 coordinates to get right ;)
[08:28] <ttx> SpamapS: why is http://people.canonical.com/~pitti/workitems/maverick/u/james-page-ubuntu-10.10.html using the classic burndown ?
[08:28] <ttx> hmm
[08:28] <ttx> the team membership is not refreshed at every run, maybe
[08:29] <SpamapS> ttx: those are explicitly set in config
[08:29] <SpamapS> ttx: did not have an easy way to determine team membership
[08:29] <ttx> SpamapS: or maybe it's the 1 inprogress / foreign that screws it
[08:29] <SpamapS> or rather, did not take the time to do that. ;)
[08:29] <ttx> (from andresrl)
[08:31] <ttx> SpamapS: no need to duplicate effort -- want me to do it, or you're on it ?
[08:31] <ttx> (you should rather be sleeping, but hey)
[08:33] <SpamapS> ttx: Yeah, I'm waiting for a build of drizzle then I'll be sleeping
[08:33] <SpamapS> ttx: It looks to me like the stacking just isn't right
[08:34] <SpamapS> ttx: if I generate a chart w/o inverted, I see some foreign done/inprogress
[08:35] <ttx> SpamapS: ah! you broke it!
[08:35] <ttx> SpamapS: that used to work alright ! :P
[08:35] <ttx> (I agreee it's pretty brittle, though) :)
[08:36] <SpamapS> ttx: when it was 3 things (todo/done/postponed).. no foreign/team .. no blocked.. in progress.. inverted.. the manual chart build was a nice simple way to go
[08:36] <SpamapS> but now.. it needs some dynamism
[08:37] <ttx> SpamapS: yep, it could use a bit of abstraction
[08:37] <ttx> or crazy things like an object to represent the data.
[08:38] <SpamapS> no, remembering what number has "todo foreign" in it is easy
[08:39] <ttx> SpamapS: except that it tends to change when you add new "stuff"
[08:39] <ttx> so it makes regressions easier (like this one)
[08:40]  * SpamapS forgot to pay the bill on his lease for the sarcmark
[08:41] <ttx> SpamapS: heh, joking with me before I got my first coffee of the day is always dangerous :P
[08:48] <SpamapS> ttx: got it
[08:49] <ttx> SpamapS: what was the offender ?
[08:50] <SpamapS> ttx: bzr diff -c 228 lp:launchpad-work-items-tracker
[08:50] <SpamapS> >:
[08:50] <SpamapS> I was seriously just joking
[08:51] <ttx> ah
[08:56]  * ttx fetches coffee
[09:01]  * RoyK steals ttx's coffee
[09:47] <Name141> Will ubuntu-server tell me when it needs to restart in the terminal ?
[09:47] <Name141> (over ssh)
[09:47] <Name141> (after an upgrade..)
[09:55] <kklimonda> Name141: not after an upgrade (unless you are running a byobu) but after you log in, you will get a message in motd
[10:03] <Name141> kklimonda: So I wont notice anything after the upgrade but the new packages ?
[10:03] <Name141> kklimonda: but later on, it will tell me after a new login ?
[10:05] <kklimonda> Name141: yes
[10:06] <kklimonda> Name141: there aren't that many packages that require a restart - I think only kernel updates trigger the message
[10:07] <Name141> kklimonda: Ok, so instead of the disk and processes, etc.. I will see like "Hey baby, this OS needs a restart sugar.."
[10:08] <kklimonda> Name141: you will see *** System restart required *** at the bottom of the motd
[10:08] <Name141> OK.
[11:11] <baffle> Our VM management project: http://www.dropbox.com/gallery/1314391/1/Vilma%202010-09-22?h=9ba7cc
[11:52] <lau> hello, I am runnin exim4 4.69 on ubuntu box
[11:52] <lau> when I use the ubuntu init exim4 scrit and ps aux | grep exim I get
[11:52] <lau> /usr/sbin/exim4 -oP /var/run/exim4/eximqr.pid -q30m
[11:53] <lau> and the server is not listening on local if port 25
[11:53] <lau> If I start it manually via /usr/sbin/exim4 -bd -q30m
[11:54] <lau> the server is listenning on localhost 25
[11:54] <lau> why does the init script not start the daemon with the -bd option ?
[11:58] <google-fu> lau, change the init script to do what you want
[11:59] <joschi> lau: what's the content of /etc/default/exim4? (use a pastebin)
[12:13] <Daviey> jdstrand, Are you free to talk about libvirt i386?
[12:19] <lau> good catch joschi thanks
[12:55] <mdeslaur> ttx: what was the outcome of the apache2 problem?
[12:56] <ttx> mdeslaur: looks like we have a pile of bugs coming in each time we upgrade apache
[12:56] <ttx> mdeslaur: when people manually play with mods.available
[12:56] <mdeslaur> oh, yeah, the maintainer scripts are really fragile
[12:57] <ttx> mdeslaur: we should add some || true on the a2enmod call on the NEXT update
[12:57] <ttx> to make it less brittle
[12:57] <mdeslaur> ttx: AFAIK, the server team has some pending apache2 stuff to push to lucid -proposed, so add it to them
[12:57] <ttx> mdeslaur: but that's nothing specific to your update
[12:57] <mdeslaur> ttx: ok, cool :)
[12:58]  * mdeslaur wipes sweat from forehead
[12:59] <ttx> mdeslaur: I overreacted to those 3 bug reports over the course of a couple hours
[13:03] <ttx> zul: if you're on top of that lucid-proposed  apache2 stuff that mdeslaur is talking about, please add that bug to your queue
[13:26] <zul> ttx: acked
[13:37] <zul> we need to do an apache-mpm-itk update for hardy if its not done already
[13:38] <Daviey> zul, mdeslaur is on that one :)
[13:39] <zul> Daviey: ah ok i just woke up ;)
[13:39] <mdeslaur> zul: I'll push them out as soon as they finish building
[13:44] <jdstrand> Daviey: hey. what's up?
[13:44] <Daviey> jdstrand, OK.. regarding this issue we have been seeing with i386 with libvirt.. /me finds bug number
[13:45] <jdstrand> mdeslaur: I wonder if we could have a reminder in umt for apache2
[13:45] <jdstrand> bug #628055
[13:45] <jdstrand> yes
[13:45] <mdeslaur> jdstrand: yeah, I was thinking about that
[13:45] <Daviey> 628055
[13:45] <Daviey> yes!
[13:46] <Daviey> jdstrand, OK... pulling a euca libvirt.xml and trying to run it externally gives the same issue
[13:46] <jdstrand> mdeslaur: there are probably going to be other cases where we need to do funky stuff, so something general where we have '<package>:<commnet>' or something
[13:46] <Daviey> (on the same box, running a manual libvirt instances DOES work)
[13:46]  * jdstrand is just thinking otoh
[13:46] <Daviey> So i'm guessing it's something in the xml which is causing an explosion
[13:46] <jdstrand> Daviey: interesting
[13:47] <jdstrand> Daviey: before we go that route. do the instances *ever* start in euca on i386 or they always fail?
[13:47] <Daviey> jdstrand, I can do one of two things to help... you can ssh to the box i'm seeing this on... or i can try and tar up the whole shindig including disk image.
[13:48] <Daviey> jdstrand, Well i did get a report of running... but went directly to terminate afterwards... I'm certain it's a red herring - but i didn't check the logs.
[13:48] <jdstrand> Daviey: ok. that is good-- it consistently fails
[13:48] <Daviey> I am reasonably conformable saying it's always failing
[13:49] <jdstrand> Daviey: what is this 'libvirt.xml' file?
[13:49] <Daviey> jdstrand, pastebin on the way
[13:50] <hggdh> I thought I had had a mix of failures & starts, but it ended up I was on the wrong version
[13:51] <Daviey> jdstrand, http://pastebin.daviey.com/QycQ/raw/ <-- this is what euca throws at libvirt
[13:51] <Daviey> libvirt, naturally adds the defaults.. want to see what libvirt generates?
[13:52] <jdstrand> Daviey: so that is the domain xml for an individual machine, correct?
[13:52] <jdstrand> Daviey: yes
[13:52] <Daviey> yeah
[13:54] <jdstrand> so there is no uuid, which implies this is doing a 'create'
[13:55] <Daviey> jdstrand, http://pb.daviey.com/Q8Pm/raw/
[13:55] <Daviey> that is what libvirt throws out
[13:57] <mdlueck> Anyone having trouble with US 10.04.1 LTS failing to recognize USB HDD's and USB Flash Drives? I verified and it correctly works with a USB Keyboard, so USB is not totally dead.
[13:57] <jdstrand> Daviey: right, so euca takes the first, does a create and ends up with the second-- is that accurate?
[13:58] <jdstrand> Daviey: if you take the second on its own and define it, it works?
[13:58] <jdstrand> Daviey: (eg, if you change the name and uuid)
[13:59]  * jdstrand notices the 'sda' and thought euca was moving to 'vda'... (just an aside)
[13:59] <Daviey> jdstrand, no
[14:00] <Daviey> the second is what is in /etc/libvirt/qemu/
[14:00] <Daviey> i defined it, and tried to start
[14:00] <jdstrand> Daviey: that is what I meant
[14:00] <Daviey> (expecting it to work)
[14:00] <Daviey> but got the security labelling error
[14:00] <mdlueck> And what package should I even open a bug about USB drives not being recognized? The kernel itself, or... ???
[14:00] <Daviey> which means there is something in the xml it doesn't like
[14:01] <jdstrand> Daviey: to get the second, it is doing a 'virsh create', no?
[14:01] <jdstrand> Daviey: or is it creating the 2nd on its own somehow?
[14:01] <hggdh> mdlueck: open the bug against linux, yes
[14:02] <Daviey> jdstrand, yeah, from a define i think
[14:02] <jdstrand> Daviey: I don't understand (that was an either/or)
[14:02] <mdlueck> hggdh: http://packages.ubuntu.com/lucid/linux correct???
[14:03] <hggdh> mdlueck: 'linux' means the current kernel *whatever version it is).
[14:04] <mdlueck> hggdh: K, thanks, I think I got to the correct place. Will log away.
[14:04] <hggdh> mdlueck: so 'ubuntu-bug linux' will grab all data considered interesting for kernel bugs
[14:04] <mdlueck> hggdh: Extra thanks!
[14:04] <jdstrand> Daviey: can you do:
[14:05] <jdstrand> qemu-img info /var/lib/eucalyptus/instances//admin/i-49C008EB/disk
[14:05] <jdstrand> Daviey: you know what, I'll take that ssh access
[14:06] <Daviey> jdstrand, ok :)
[14:06] <Daviey> jdstrand, i'll get that organised now
[14:13] <jdstrand> Daviey: ok, thanks
[14:13] <jdstrand> Daviey: so, am I allowed to do anything I want on this machine to triage this?
[14:15] <jdstrand> Daviey: hmmm, that machine has no machines defined
[14:15] <Daviey> jdstrand, you can, yes
[14:15] <Daviey> it's a throwaway machine
[14:16] <jdstrand> Daviey: can you get it to the point where it has just thrown the error? I'm sorry, I know next to nothing about euca specifics
[14:18] <Daviey> jdstrand, ok.. sorry for the delay
[14:20] <Daviey> jdstrand, are you seeing that?
[14:20] <jdstrand> Daviey: yes
[14:20] <Daviey> okay..
[14:20] <jdstrand> Daviey: how are /etc/libvirt/qemu/*xml being created?
[14:20] <Daviey> jdstrand, In that case... i found the one euca throws out, and did a virsh define
[14:21] <jdstrand> ok
[14:21] <jdstrand> let me see what the problem is, then we can figure out what is causing it
[14:21] <Daviey> jdstrand, feel free :)
[14:21] <Daviey> jdstrand, if the box goes bang, it's no concerns
[14:22] <jdstrand> I shouldn't be doing anything that crazy :)
[14:23] <Daviey> jdstrand, passwd in ~/PASSWD.txt if sudo times out
[14:23] <jdstrand> ok :)
[14:32] <ehcah> Does anyone here have experience setting up dyndns without installing their update client locally?
[14:34] <pmatulis> ehcah: have you reason to believe that it's possible?
[14:35] <ehcah> I was running F12 and opendns without a client installed locally?
[14:35] <Daviey> jdstrand, looks like euca has done a garbage collect
[14:36] <jdstrand> Daviey: can we stop that? it is kinda a pain
[14:36] <Daviey> jdstrand, it "is stopped"... or at least the setting is made for this to be the case.
[14:36] <Daviey> :/
[14:36] <Daviey> jdstrand, I'll fire another instance over
[14:36] <jdstrand> Daviey: thanks
[14:37] <Daviey> done
[14:38] <jdstrand> Daviey: so, that one started?
[14:38] <Daviey> geez it has!
[14:38] <jdstrand> hold on
[14:39] <Daviey> hmm
[14:39] <Daviey> euca thinks it has started
[14:40] <Daviey> (doesn't look like it to me tho)
[14:46] <jdstrand> Daviey: did it garbage collect again? I need the i-433... or at least something that doesn't keep disappearing
[14:47] <hggdh> jdstrand: Euca will garbage-collect all terminated instances
[14:47] <Daviey> hggdh, I've set the value that should stop it! :(
[14:47] <Daviey> jdstrand, I think we might need to cp -R it.
[14:47] <Daviey> then if it does collect, we can put it back in place
[14:47] <hggdh> Daviey: and, of course, you started it CLEAN=1?
[14:48] <Daviey> i-4BFC08F4
[14:48]  * hggdh hopes this is it
[14:48] <Daviey> hggdh, hmm... no
[14:49] <jdstrand> Daviey: are you restarting libvirt?
[14:49] <Daviey> jdstrand, no
[14:49] <jdstrand> meh
[14:49] <Daviey> hggdh, MANUAL_INSTANCES_CLEANUP=1
[14:50] <Daviey> jdstrand, i need to go AWOL soon... can you make sure you have a snapshot of that instance incase the garbage collect happens again?
[14:50] <jdstrand> libvirt is hanging on me
[14:50] <hggdh> Daviey: and that goes where? In the -cc init?
[14:51] <jdstrand> ok, there we go
[14:51] <Daviey> hggdh, i put it in both :)
[14:51] <Daviey> hggdh, /etc/eucalyptus/eucalyptus.conf
[14:51] <jdstrand> Daviey: it is gone already
[14:51] <Daviey> but it's failing to do it
[14:51] <Daviey> *sigh*
[14:52] <jdstrand> I'll do it totally outside of euca
[14:52] <Daviey> one moment
[14:53] <Daviey> jdstrand, see byobu window "1"
[14:53] <Daviey> press f4
[14:53] <Daviey> that is ssh'd to the euca server
[14:53] <Daviey> if you press return on that, it will fire an instance
[14:54] <jdstrand> how do I get back?
[14:54] <Daviey> f3
[14:54] <jdstrand> no\
[14:54] <jdstrand> I hit something and it looks like another window was created
[14:54] <Daviey> yeah
[14:54] <Daviey> f2 = new window
[14:54] <Daviey> f3 = previous window
[14:55] <Daviey> f4 = next window
[14:55] <jdstrand> gotcha
[14:55] <Daviey> i think you hit f2 twice
[14:58] <Daviey> jdstrand, /me afk
[14:58] <jdstrand> k
[16:23] <zorton> update-grub results in: "/usr/sbin/grub-probe: error: unknown filesystem"
[16:33] <cdubya> I'm looking for recommendations on vulnerability scanners.....
[16:34] <zorton> nmap for single hosts, nessus has gone commercial but will scan a small number of hosts under the home license
[16:35] <zorton> nmap will also do full IP ranges now that I think of it, I suppose nessus is just more comprehensive
[16:44] <cdubya> zorton, I've used nmap on simpler scans internally (-F -T4 options) but I think that only scans the 100 most common ports. Is there a good reference on setting up "best practice" vulnerability scans that are comprehensive enough?
[16:46] <zorton> well, if you want more port coverage just increase the port range nmap is using, if you do the full port range it will take more time per host
[16:46] <zorton> plus if you add in UDP
[17:00] <SpamapS> zorton: nessus is to nmap as MI5 is to 007 ;)
[17:00] <zorton> heh, nice one
[17:01] <SpamapS> I haven't used nessus in like, 5 years. Is it still being updated with new signatures/tests?
[17:01] <zorton> i've had a pentester run everything with nmap and shell scripting
[17:02] <zorton> oh yeah, they've just gone comercial now so the plugin feed is free to home use but requires a full up copy for commercial/gov/more than 10 or so hosts
[17:03] <SpamapS> makes sense that was happening when I was playing with it
[17:03] <zorton> yeah, nothing really in the OSS world is keeping up and alive
[17:03] <zorton> writting the plugins takes time and really should be done by someone being paid to do it
[17:05] <SpamapS> Who says open source precludes being paid to work on it? ;)
[17:05] <\sh> does anybody has a clue how to prevent udev to overwrite an already created 70-persistent-net.rules file? whatever I do, it doesn't listen to the admin ;)
[17:08] <SpamapS> \sh: if its in /etc and has been changed, you should be prompted for what to do with it
[17:10] <\sh> SpamapS: no...what I mean is that /lib/udev/rules.d/75-persistent-net.rules always triggers the write_net_rules script, which overwrites a 70-persistent-net.rules file from /etc/udev/rules.d/ .. I don't talk about packages
[17:10] <zorton> really? I did a custom name for one of my network devices in that same file and didn't have any trouble
[17:11] <\sh> zorton: on jaunty that worked...now I can reproduce it all the time
[17:11] <\sh> now == lucid
[17:11] <zorton> yeah, this is on lucid
[17:11] <\sh> zorton: and I write it from scratch I don't change it manually when it was created by first startup
[17:12] <zorton> ahhh, that might be why it didn't get overwritten.  I'm too lazy to write it from scratch
[17:12] <\sh> zorton: well actually it's automatically written from scratch from out deployment system
[17:14]  * SpamapS is completely and totally confused now
[17:19] <\sh> SpamapS: why?
[17:37] <ehcah> ! Static IP
[17:39] <SpamapS> \sh: I don't mess with udev stuff. ;)
[17:40] <SpamapS> I've struggled mightily to avoid having anything in a server more complicated that 2 bonded ethernet ports. Network complexity belongs on routers/switches/firewalls only.. and I don't use Linux for any of those. ;)
[17:42] <zorton> i'm with you on that one, the only thing i'm using it for is to give pretty names to the ethernet interfaces, admin, management, recorder and such
[17:47] <SpamapS> zorton: interesting
[17:48] <zorton> makes life a little easier that's for sure
[17:49] <SpamapS> zorton: does it work for say,  tcpdump -i admin instead of tcpdump -i eth0 ?
[17:49] <zorton> yeah, it's accutaly the same mechanism that keeps eth0 bonded to the right mac address
[17:50] <zorton> code that accesses the device dosen't care, eth0 is the same as admin
[17:51] <zorton> just make sure you update /etc/network/interfaces with the right name first :)
[17:54] <RoyK> zorton: I just came in, but if you have replaced a nic, just reset the udev stuff to get eth0 back
[18:06] <zorton> RoyK: wan't the issue :)
[18:07] <RoyK> :)
[18:07] <zorton> thanks though, i'm currently fighting grub issues
[18:11] <ehcah> What would I search to find a tutorial on assigning specific clients, static IP's vs. DHCP. My LAN will have a combination of both. All my search's come back with makign eth0 or eth1 on my server static.  :(
[18:12] <zorton> take a peak at the manpage for interfaces
[18:12] <RoyK> ehcah: you can use dhcp to assign static IPs
[18:13] <RoyK> imho a good solution
[18:14] <RoyK> ehcah: example http://pastebin.com/DYYJ7B8c
[18:14] <ehcah> That is what I would like to do, but can't find the option. I know on my previous box, there were files under bind dns that contained host hames and host ip's etc...
[18:15] <RoyK> ehcah: you bind the mac address to a specific IP in the dhcp server
[18:15] <RoyK> outside of the dynamic scope
[18:15] <ehcah> RoyK: Thanks' for the pastebin.
[18:16] <RoyK> dunno what happens if you allocate an address in the dynamic scope, though, but I guess you'll be playing with matches and gasoline
[18:16] <ehcah> My last box wasn't ubuntu, (F12). Must have been done differently.
[18:16] <RoyK> ehcah: same dhcp server on the two
[18:16] <ehcah> I'll paste one of the files from the box I haven't turned off yet.
[18:16] <RoyK> ISC dhcpd
[18:16] <ehcah> just a sec.
[18:18] <ehcah> RoyK: This is an example. The host was already assigned at this point? http://paste.ubuntu.com/498635/
[18:20] <ehcah> RoyK: On second glance, those may be solely for the purpose of internal DNS?
[18:20] <RoyK> ehcah: that's just DNS - that won't give you a fixed IP if the PC is offline for some time
[18:20] <ehcah> k.
[18:20] <RoyK> if using rfc1918 addresses, just allocate a /24 part for static IPs
[18:21] <RoyK> say 192.168.0.0/24 for dynamic and 1.0/24 for static, set the netmask to 255.255.254.0 and they're on the same net
[18:22] <ehcah> I use webmin for management. I know it's not preferred for most of this group. Under DHCP, I have a Host & Host Group option that also allow me to assign a name, or IP via mac address?
[18:22] <ehcah> ok.
[18:22] <RoyK> !webmin
[18:23] <RoyK> !ebox
[18:25] <ehcah> Alright. Where it's description talks about not all modules being enabled, will it be difficult for me to install or enable the missing?
[18:25] <ehcah> Webmin has always been easy for me as it generally picks up everything that is installed on first opening.
[18:26] <RoyK> ehcah: webmin was written for redhat, but isn't really compatible with debian-based distros
[18:27] <ehcah> ok.
[18:27] <ehcah> I'm trying to install ebox now.
[18:30] <Daviey> jdstrand, Did you have any joy?
[18:33] <jdstrand> Daviey: I know what is failing. I don't know why yet
[18:33] <Daviey> jdstrand, OK... that is something :)
[18:33] <jdstrand> Daviey: please do not turn off that machine. I need the disk
[18:33] <jdstrand> Daviey: and am copying it over
[18:33] <Daviey> jdstrand, Okay.. that is fine.
[18:33] <jdstrand> I'm trying to reproduce locally
[18:34] <Daviey> jdstrand, I don't suppose you could, erm - run "ssh-import-lp-id davewalker" could you?  I've erm, sort of not made a note of the password.
[18:35] <jdstrand> INFO: Successfully authorized [davewalker]
[18:35] <Daviey> Even thought the machine is in arm distance... that just makes life easier :)
[18:35] <Daviey> thanks!
[18:35] <jdstrand> sure
[18:37] <jdstrand> Daviey: btw, I've marked the bug confirmed and assigned it to me
[18:38] <Daviey> jdstrand, Expect many drinks at UDS.
[18:38] <jdstrand> hehe
[18:39] <jdstrand> I know right where the failure happens in virt-aa-helper, I just need to reproduce locally, and then I can hopefully zero in on it
[18:40] <jdstrand> I think it may be related to all their recent disk probing changes (which virt-aa-helper is affected by), but I need to investigate more
[18:48] <ehcah> RoyK: Is ebox exclusively firefox or will Google Chrome work also?
[18:48] <ehcah> Royk: I really don't like firefox. It's a pig. IMO
[18:49] <RoyK> ehcah: I have no idea - I use the commandline for configuring servers :þ
[18:50] <ehcah> LOL
[18:52] <RoyK> ehcah: take some time reading some docs, playing with a system with the commandline, and you'll never go back to some fancy gui trying to do what's done in 30 seconds in the commandline
[18:52] <RyanP> I'm in need of some routing help. I have a machine with a bunch of OpenVPN tunnels into remote sites. I'd like to route traffic from the local office over the tunnels into the remote sites. I have the VPN tunnels set up, and I know how to setup routing on the local office computers. I'm not sure how to correctly allow the OpenVPN machine to forward packets over the tunnels. How do I do this?
[18:52] <ehcah> RoyK: Thank you for your assistance.
[18:53] <RoyK> ehcah: that last comment was not meant as arrogant - just honest
[18:54] <ehcah> I know. You're a real Linux guy! :)  I'm not, and not offended at all.
[18:54] <ehcah> I'm trying to do as much command line as possible. But, I need google's assistance for that!!!  LOL
[18:54] <RoyK> even my windoze-loving boss uses the commandline to configure linux
[18:55] <RoyK> sure, google (and man pages) are there to help
[18:56] <ehcah> In the meantime, I'l try to get eBox up and running. The repos installed version 1.5. It looks like there is a 2.0 available under a new name of Zental? Have i got those packages mixed up on the web?
[18:58] <RoyK> ehcah: not to be a bitch, but if you had spent five minutes looking through the dhcp config, you had been finished by now :P
[19:00] <ehcah> I actually made those changes before ebox. They were changes I could easily handle.
[19:01] <ehcah> ebox would work best for me when I'm in unfamilliar territory.
[19:03] <RoyK> I don't really agree - if you're on unfamiliar territory, explore it and it'll be quite familiar next time you go there
[19:12] <RyanP> It seems all I need to do is ask here, and then I can figure the question out myself. In answer to the OpenVPN question, it's just a simple NAT, but you have to be sure to actually type the commands correctly, like I wasn't doing.
[19:14] <cemc> what's the easiest/best way to turn a server install into a desktop install (and back) ?
[19:16] <nich0s> cemc: What?
[19:17] <cemc> say I've installed 10.04 server edition, but I want to turn it into desktop (install packages for X, gnome, etc)
[19:17] <cemc> is it enough to just install ubuntu-desktop package?
[19:18] <nich0s> I believe that package should cover just about everything.
[19:18] <nich0s> You'll want to go through updaterc.rc and clear out any daemons set to load on start.
[19:19] <nich0s> update-rc.d*
[19:19] <cemc> and can I go back to server after that? or going back is not se easy?
[19:20] <bobslaede> Hi, Im having a small problem with autofs and sshfs, as it seems that autofs doesnt recognize the pub_keys for the user, where sshfs does
[19:20] <bobslaede> and Im not fond of the idea of creating pub keys for the root user
[19:21] <nich0s> cemc: Ultimately you can remove any packages you want to.
[19:22] <nich0s> cemc: Server edition is simply a base version of Ubuntu set up to function more like a server than a personal computer.
[19:22] <cemc> nich0s: I know, I was looking for an easy way ;) I guess there isn't, I can't just remove ubuntu-desktop :)
[19:23] <cemc> or I can do a list of all installed packages when I installed ubuntu-desktop, and apt-get purge them
[19:23] <ScottK> cemc: In theory, sudo apt-get remove ubuntu-desktop && apt-get autoremove would do it.
[19:23] <ScottK> You can, although remember that even purge doesn't always put your system back exactly like it was before.
[19:23] <cemc> ScottK: you think all those packages will be marked 'not needed' because I removed ubuntu-desktop? ;)
[19:24] <ScottK> Since they are pulled in by ubuntu-desktop and not installed directly, that should be the case.
[19:24] <ScottK> No promised though.
[19:24] <ScottK> (I'd still keep the list)
[19:24] <nich0s> pipe it in to a install log.
[19:25] <RoyK> cemc: if you want server and desktop on the same box, just install desktop and disable X when you don't need it anymore
[19:25] <RoyK> cemc: ubuntu desktop and server is really the same thing
[19:26] <cemc> I know, I was merely wondering about the theory in doing server->desktop and back
[19:26] <cemc> thanks for the tips
[19:26] <RoyK> cemc: just install desktop - the only problem will be that it's not supported for 5 years, but only a mere 3 years
[19:27] <u92> what is a .listing file is it like a thumbs.db?
[19:27] <cemc> RoyK: I guess that's ultimately is decided for every package individually, right?
[19:28] <RoyK> cemc: not sure, but then, I don't use desktop distros for my servers
[19:29] <nich0s> bobslaede: Something like http://www.tjansson.dk/?p=84 should help you out.
[19:30] <cemc> if you install ubuntu desktop it doesn't mean you won't get ANY updates after 3 years, you should get updates for 'server packages' if you have any installed, maybe you won't get update for say firefox
[19:30] <cemc> but you should get updates for postfix, squid and the like, hm?
[19:30] <nich0s> cemc: That's going to depend on which repos you're pinging with apt-get.
[19:31] <nich0s> cemc: you should be able to uncomment the lines for universe, etc.
[19:31] <ScottK> nich0s: No.  All the server and desktop packages are in the same repository
[19:31] <RoyK> cemc: use a server distro for a server
[19:31] <nich0s> cemc: /etc/apt/sources.list, I believe.
[19:31] <lamont> nich0s: there's no diff between the server and desktop repos
[19:32]  * RoyK doesn't get why anyone would want X on a server
[19:32] <lamont> some people like guis on their servers.  I've never really understood that so much either
[19:32] <RoyK> well, X libs, of course, but then you can run commands over ssh
[19:32] <cemc> RoyK: I don't, it's just for argument's sake
[19:32] <RoyK> menus belong in restaurants :)
[19:33] <cemc> :)
[19:35] <ehcah> Anyone using ebox 2.01 (Zentyal) and reccomend bypassing 1.5?
[19:36]  * RoyK diverts ehcah to vimtutor
[19:37] <bobslaede> nich0s: thanks
[19:37] <bobslaede> nich0s: went with the root public key
[19:38] <RoyK> erm - do you allow root ssh logins?
[19:38]  * RoyK thinks that's a BAD idea
[19:56] <bobslaede> RoyK: root login? not really
[19:56] <bobslaede> RoyK: if that was for me anyway
[19:57] <bobslaede> RoyK: its just that autofs doesnt see the correct public keys, like sshfs on its own does
[19:57] <bobslaede> apparently still using the correct user to login
[19:57] <bobslaede> altho im not 100% sure
[19:57] <bobslaede> but it works now
[20:01] <cemc> say I'm logged in as a user and want to do sudo command > afile.txt in a directory I don't have write permissions to as a normal user, how can I work around this?
[20:02] <nich0s> sudo -i
[20:02] <nich0s> sudo command > /dir/you/have/write/access/to/something.log
[20:03] <cemc> I knew this ;) no other way?
[20:03] <cemc> that the logfile gets written directly to the current dir?
[20:03] <cemc> http://pastebin.ubuntu.com/498693/
[20:06] <_Techie_> cemc, sudo command > /tmp/afile.txt && sudo mv /tmp/afile.txt ./
[20:07] <cemc> :)
[20:07] <_Techie_> that should do the trick
[20:08] <cemc> I knew this trick ;) is there any trick that doesn't require this one? :)
[20:08] <_Techie_> not that i can think of at 7a
[20:08] <_Techie_> 7am*
[20:16] <cemc> it's insane how fast a 10.04 server boots as a kvm guest ;)
[20:18] <jdstrand> Daviey: you can destroy that instance. I have what I need now and can reproduce locally
[20:19] <Daviey> jdstrand, ok - great.. i'll turn them off, as they are making the room warm now :)
[20:20] <Daviey> jdstrand, thanks for looking into it
[20:22] <jdstrand> Daviey: sure
[20:23] <WinstonSmith> hi ppl :) . just updated an remote server. is there a command that would let me restart all the affected services (apache, mysql, etc) in 1 go? or restart the whole runlevel? thx
[20:24] <nich0s> WinstonSmith: You could run a for loop.
[20:24] <WinstonSmith> nich0s, but that would require me knowing all the affected services
[20:26] <nich0s> WinstonSmith: That would be helpful information to have.
[20:26] <WinstonSmith> i sure could write a script that starts & stops a lot of services with a loop but i was looking for a way to "restart the runlevel" so to speak
[20:28] <nich0s> I'm not familiar with a way to restart a runlevel outside of reboot.
[20:30] <WinstonSmith> nich0s, ok. another thing learned. thanks !
[20:30]  * WinstonSmith goes away to script 
[20:42] <guntbert> !runlevel | nich0s WinstonSmith  (just as a reminder)
[20:46] <WinstonSmith> guntbert, thank you. but in my (very) limited knowledge the implementation of upstart is far from complete?
[20:46] <guntbert> WinstonSmith: that seems to be true, but that doesn't alter the fact that !all runlevels except 0,1 and 6 are by default equal"
[20:46] <WinstonSmith> guntbert, cause it seems not all services can be controlled properly by the "service" command
[20:46] <guntbert> "all...
[20:47] <guntbert> WinstonSmith: yes, some services are still handled by sysv init - but no differences between runlevels
[20:48] <WinstonSmith> guntbert, yes ok
[20:49] <WinstonSmith> guntbert, just curious : how does on add a service to upstart like update-rc in sysv?
[20:49] <Four2zero> can ubuntu-server show the repos what's available such as version numbers of a software thats available without doing actual download first ?
[20:49] <Four2zero> is there a command for that ?
[20:49] <guntbert> WinstonSmith: not surem have a look at insserv please
[20:50] <WinstonSmith> ok cheers have a nice 1
[20:59] <g0tcha> anyone have a script that auto adds ip addresses if the server got rebooted by any chance?
[20:59] <guntbert> g0tcha: ?
[21:00] <g0tcha> guntbert, i have a HE tunnel setup on my ubuntu server, if it gets rebooted, i have to readd everything manualy
[21:00] <g0tcha> so i thought mayb someone has a script that does it automaticly or something
[21:01] <guntbert> g0tcha: aha - no idea, sorry
[21:02] <Four2zero> can ubuntu-server show the repos what's available such as version numbers of a software thats available without doing actual download first ?
[21:02] <Four2zero> is there a command for that ?
[21:02] <g0tcha> guntbert, np thanks for trying dued
[21:04] <guntbert> Four2zero: aptitude search ..., and aptitude show <package>
[21:05] <Four2zero> thank you guntbert
[21:05] <guntbert> Four2zero: you're welcome :-)
[21:09] <Kyle__> how do you setup a syslog server with rsyslogd?
[21:09] <Kyle__> There's no refrence to how to do it in the man page that Ic ould find.
[21:15] <savid> Hi, I have a script in /etc/cron.d,   but I'm not sure if it's running.  Is there a way to verify whether or not the script is being executed?  Is there a log somewhere I can check?
[21:16] <nakhlawi> savid: ps -ef | grep <script name>
[21:17] <savid> nakhlawi,  it'd be difficult to tell using ps when it's just a cron.  The command runs very quickly.
[21:18] <savid> nakhlawi,  I'm pretty sure the command isn't running.  Just can't figure out why.
[21:18] <nich0s> savid: Add a line to the end to the script which appends the time and date.
[21:18] <savid> nich0s, huh?  appends to what?
[21:19] <savid> If I place a crontab in /etc/cron.d,  that crontab should be processed, right?
[21:21] <nakhlawi> try this: grep -l CRON /var/log/*
[21:22] <guntbert> !details | Kyle__
[21:23] <nakhlawi> savid: also: grep CRON /var/log/*
[21:24] <savid> Ok, I see where cron logs things.  I don't see anything about my crontab that I entered.
[21:24] <nakhlawi> savid: it means your script didn't run
[21:24] <savid> http://dpaste.com/247544/
[21:25] <savid> nakhlawi,  yeah,  I'm trying to figure out _why_ it didn't run
[21:25] <savid> I have it set up for ever minute
[21:26] <guntbert> savid: did you give full paths to the files you want executed?
[21:26] <savid> guntbert, yes, but even if I didn't, shouldn't I see an error somewhere?
[21:26] <nakhlawi> savid: try adding it in /etc/crontab
[21:27] <guntbert> savid: not necessarily
[21:31] <savid> If I copy and paste the command into the shell it works fine
[21:34] <savid> guntbert,    shouldn't   /etc/crontab be referring to cron.d ?  I don't see it in there
[21:34] <guntbert> savid: in your shell there are search paths set wich are probably not set in the cron environment
[21:34] <savid> guntbert,  I'm using absolute paths
[21:35] <savid> guntbert,  http://dpaste.com/247544/
[21:37] <guntbert> savid: for a test: put a simple line there (use logger to write to the syslog) and try that
[21:38] <savid> guntbert, it's working when I crontab -e root's crontab.
[21:39] <guntbert> savid: I see - I have to tell you that I'm already to tired to think properly - sorry
[21:39] <savid> guntbert, tell me about it :-P
[21:39] <jdstrand> Daviey: I have the fix for bug #628055
[21:40] <guntbert> savid: then call it a day and try tomorrow :-)
[21:40] <savid> guntbert,  heh,  unfortunately I have deadlines :-P
[21:40] <savid> oh well,  I'll keep working at it
[21:40] <guntbert> savid: Good luck :-)
[21:48] <pretto> can anyone show me why this is not working using iptables dnat http://pastebin.com/aevKupST? thanks in advance
[21:51] <Ad0> hi
[21:51] <Ad0> I have issues compiling a v4l2 driver
[21:51] <Ad0>  media/v4l2-ioctl.h: No such file or directory
[21:57] <soren> pretto: What do you expect it to do?
[21:59] <Daviey> jdstrand, you are full of awesome
[22:04] <Ad0> I guess I have to run and set up the script
[22:04] <Ad0> oops
[22:05] <pretto> soren, just to redirect http from external port 8080 to internal server at port 80
[22:05] <ewook> hey, anyone had perl , 5.10 it seems start bailing out on ya?
[22:06] <pretto> soren, as you can see the firewall is just a router
[22:08] <ewook> upgraded to latest patches (10.04), rebooted, and now things are just plain broke. did a new apt-get upgrade, it started to "re-install" bunch of packages, but still, ie - irssi is busted for my regular user, but now at least runs fine for root.
[22:09] <soren> pretto: And what happens?
[22:17] <pretto> soren, from outside it shows filtered when i do a nmap -p 8080 serverip
[22:18] <pretto> soren, if i use the browse it ends in a timeout
[22:19] <soren> pretto: And you're sure a) something is listening on  192.168.0.30:80 and b) that you can actually reach port 8080 on that box from the outside?
[22:23] <hallyn> zul: around?
[22:24] <hallyn> zul: for bug 645082, should i propose the trivial patch adding gawk to depends in debian/control, or do you want to handle it with upstream some way?
[22:24] <pretto> soren, yes, I can even ssh the server from the outside
[22:24] <soren> pretto: On port 8080?
[22:26] <pretto> soren, no 22
[22:27] <pretto> the only one i want to redirect to another server is the port 80 soren
[22:27] <soren> And I'm asking whether you're sure you can reach that port from the outside.
[22:27] <soren> And you answer that you can reach another port.
[22:28] <soren> which isn't what I'm asking.
[22:28] <pretto> yes, i do
[22:28] <soren> win 61
[22:28] <soren> Whoops
[22:28] <pretto> soren, i can, the log shows it in the forward
[22:33] <jdstrand> Daviey, ttx: ok, uploaded fix for bug #628055. Can you guys follow up with whoever you need to to get that accepted? the debdiff is in the bug
[22:33] <Daviey> jdstrand, Thanks... i'll try and chase the release team - as it would be nice to get it on tomorrows ISO
[22:34] <jdstrand> Daviey: sounds great
[22:49] <ehcah> I'm going to get laughed at - again - but I can not get vnc working correctly. My xstartup file appears to be fine, but some files it references are not. I have an error opening the security policy, can init several files and finally a permission error. Does seem to matter if i start VNC as root or me?
[22:53] <_Techie_> is the release date for maverick 10/10/2010, or 11/10/2010 ?
[22:55] <EtienneG> _Techie_, the former
[23:28] <_Techie_> anyone wanna integrate a ssh server and some form of implementation of screen, so that i can re inatll my server without having to rip a graphics card out of my desktop to use the installer