[00:00] <hazmat> genii-around, you can use the service command for either type.. but start/stop only work for upstart'ified services
[00:00] <genii-around> Ah
[01:30] <Tohuw> Can someone assist me with building a proper route table for my Ubuntu 12.04 server? I have two network interfaces (one with the public static IP, the other with a private IP), but I can see the route table is wrong, and I'm having trouble fixing it.
[01:31] <Tohuw> (The modem will plug into the "outside" interface, the "inside" interface will connect to a switch)
[02:22] <pukeko_> rdiff-backup restore problem here...  how do i restore a dir  such as /path/to a dir/with spaces/foo ?
[02:23] <pukeko_> normally its rdiff-backup -r /some/directory /someother/directory
[02:24] <pukeko_> ' /path/to a dir/with spaces/foo/'  doesn't work..
[02:26] <imbrandon> /a/path\ with\ some/spaces/
[03:11] <pukeko_> imbrandon: hi there - so i need to back slash the spaces huh.. i'll go try it ( its a Samba server - with windows users not OBEYING naming conventions ) m- cheers
[04:05] <jefimenko> do you have to reboot after reboot after changing the timezone on a server?
[04:06] <jefimenko> using dpkg-reconfigure tzdata
[05:00] <grendal-prime> i need a quick and dirty radius server
[05:03] <twb> grendal-prime: what for
[05:04] <grendal-prime> I need to create some accounting radius packets send them to a server..then send a stop packet
[05:05] <grendal-prime> doesnt need authentication..
[05:05] <grendal-prime> just accounting
[05:05] <twb> What kind of server is speaking radius instead of sasl or pam?
[05:06] <grendal-prime> i have to write those two scripts. A vpn server needs to send this info to an appliance that uses radius.
[05:06] <grendal-prime> I think most isps use radius.
[05:07] <grendal-prime> still
[05:07] <twb> What, for dialup?
[05:07] <grendal-prime> ?
[05:07] <grendal-prime> No dude...radius has been adapted for all connection mediums ..has been for some time
[05:08] <twb> Maybe if you like gooey insides
[05:08] <grendal-prime> ?
[05:08] <twb> Never mind.
[05:08] <grendal-prime> how do YOU connect to the internet
[05:08] <grendal-prime> cable?
[05:08] <grendal-prime> cause i can tell you this much..comcast uses it.
[05:09] <grendal-prime> cablevision ..time warner..they all use radius
[05:09] <grendal-prime> but i digress
[05:09] <twb> I have an ATM service over copper, called "ADSL2+".  Over that run a stupid PPPoA tunnel that exists purely for the ISP's billing purposes.
[05:10] <grendal-prime> Ive just never set up a  radius server.  Ya that billing business  thats radius.
[05:10] <grendal-prime> but anyway
[05:10] <twb> The only time I had to deal with radius myself, was for stupid 802.11 "enterprise", and for that I used the fake one built into hostapd
[05:11] <twb> Everything I could find about radius indicated it was an old and stupid alternative to SASL or PAM, and that if you used any method but EAP-TLS you were very very stupid indeed.
[05:11] <twb> Since I am not authenticating to my ISP with a client-side TLS cert, it's implied that they're in the "very very stupid" category...
[05:12] <twb> Personally I would prefer they just peg my ISP account to the PSTN service it's currently running on, and not have the PPP overhead of 7% or whatever it is
[05:14] <grendal-prime> right
[05:14] <grendal-prime> Anyway
[05:16] <samba35> after upgrade some package (apt-get upgrade ) mouse doesnt work (as system hang) but if i use super (window) key then i am able to access mouse  why is it so any idea ?
[05:17] <twb> samba35: we don't support GUIs here
[05:17] <twb> samba35: try #ubuntu
[05:17] <samba35> ok
[05:18] <grendal-prime> radius is a AAA server basically...from what i can gather the first really widely industry adopted trip A  so. its still used for alot of things.
[05:41] <tohuw> How can I make a script that will run each time I login to an Ubuntu server? I only want it to run at logon, and only when I am interacting with the session (as the script prompts for input)
[05:42] <greppy> tohuw: call the script at the end of your .bashrc ( or other shell env scripts )
[05:43] <twb> tohuw: write a cron job that runs @reboot
[05:43] <twb> tohuw: write an upstart job
[05:43] <twb> Oh sorry at *login time*
[05:43] <twb> Then yeah, it goes in your ~/.bash_profile (or .bashrc, depending)
[05:44] <tohuw> greppy, twb: thank you
[05:47] <grendal-prime> nevermind i got one working
[05:47] <twb> grendal-prime: which one
[05:47] <grendal-prime> it really doesnt matter they are all based on cistron
[05:48] <grendal-prime> i used xtradius because it seems to just use the default files and doesnt really expect you to connect a db.
[05:49] <grendal-prime> but it seems like freeradius does the same thing really i dont know..all i know is i can creat a user and authenticate so..i should be able to do accounting only.
[05:50] <grendal-prime> that and i need  an a&w root beet
[05:50] <grendal-prime> beer that is
[05:51] <grendal-prime> gonna try and do this with php5
[06:00] <tohuw> twb: I added the command to my bashrc, but now each time I create a new window in tmux, it executes those commands. Is there a way to prevent this?
[06:01] <twb> Use .bash_profile instead.
[06:02] <tohuw> twb: Oh. That was too easy. Thanks
[06:02] <andol> Or ~/.profile, if you already have one, which seems to be the Ubuntu default.
[06:02] <andol> Creating a ~/.bash_profile would lead to an existing ~/.profile not being read.
[06:03] <tohuw> andol: Ah, thank you! I do indeed have a .profile, and now I see how .profile and .bashrc are different. :) Thanks
[06:04] <twb> andol: oh, thanks.  I thought .bash_profile was the default
[06:04] <twb> that is, I thought Ubuntu defaulted to providing a .bash_profile in /etc/skel
[06:08] <tohuw> twb: My /etc/skel on Ubuntu Server 12.04 has not been altered, and contains only .profile, no .bash_profile
[06:09] <twb> Okey dokey
[07:10] <codeshepherd> what is the free ubuntu 12.04 AMI in ec2? what is the AMI - id ?
[07:24] <smb> codeshepherd, You can find daily builds here (http://uec-images.ubuntu.com/precise/current/).
[07:28] <codeshepherd> smb: is it better to use the latest build or should I use the AMI that the aws market place suggests?  - in other words - will the daily build have unexpected bugs?
[07:30] <smb> codeshepherd, Not sure what exactly the market place suggests. But I suspect that it would be the released version. While the daily, despite all efforts to prevent that, can have regressions, the released version for sure has known bugs the more time there is between release and "now".
[07:31] <codeshepherd> thanks smb
[07:31] <smb> You could as an alternative take the released version but then I'd at least update all security relevant changes
[07:39] <blendedbychris> hey guys i just switched to precise and am having trouble getting hyper-v "native" networking adapter drivers to work… my prior legacy adapter seems to work but doesn't have connectivity… i need non-legacy to work anyhow. any ideas?
[07:39] <blendedbychris> i have another vm that has precise running fine but i am not sure what is causing this machine to be wonk
[07:49] <twb> blendedbychris: what does lspci -nn report?
[07:49] <twb> (pastebin it)
[07:50] <twb> If you have a working precise VM, try comparing its lspci -nn output.  Compare their kernel versions (uname -a).  Compare their lsmod output.
[07:50] <twb> Compare their dmesgs, ip a / ip l / ip r output, &c &c
[07:58] <Thorn> hello
[08:01] <Thorn> how do I setup a full remote backup of an ubuntu server using rsnapshot? it appears that I need to grant root privileges to the user that rsnapshot connects with over ssh
[08:07] <twb> Thorn: that is correct
[08:08] <twb> Unless you want to run rsync without ssh, which would be very stupid
[08:08] <Thorn> that still sounds like a security risk
[08:08] <twb> Thorn: well it's going to have a copy of all the data the target machine had anyway
[08:09] <twb> So if the rsnapshot host is compromised you're mostly screwed anyway
[08:09] <twb> Anyway, what you CAN do is passphraseless SSH key + /usr/share/doc/rsync/examples/rrsync.gz in the forced command
[08:13] <Thorn> in this case it's the rsnapshot host that will keep backups of a VPS (which is obviously accessible from the net)
[08:14] <Thorn> I don't think it's possible to selectively disable password auth in openssh?
[08:14] <twb> AllowRoot without-password
[08:14] <twb> But you should not allow passwords at all in any case
[08:14] <twb> Nor should the root account *have* a password
[08:15] <Thorn> I mean, disable password and leave key auth
[08:15] <twb> Yes, of course you can do that
[08:15] <twb> PasswordAuthentication no
[08:17] <Thorn> that would disable it for all users
[08:19] <twb> OK there are two things you might be asking for, and I have given you both.
[08:19] <twb> If it's still not clear read sshd_config(5) and the default config file's comments
[08:20] <Thorn> sorry, I didn't phrase it correctly. I was asking about disabling password auth for a particular user only
[08:20] <Thorn> doesn't appear to be possible
[08:22] <andol> Thorn: See the Match directive
[08:23] <twb> or if that particular user is root, see above
[08:24] <Thorn> Match is nice
[08:24] <Thorn> okay, thanks much for help. looks like it's working.
[08:47] <RoyK> http://biebian.sourceforge.net/
[09:09] <thys> hi, I'm trying to update my 9.10 to 10.04 to get the LTS
[09:09] <thys> so I typed sudo apt-get install update-manager-core
[09:10] <thys> it got 404, failed to fetch
[09:10] <twb> 9.10 is end-of-lifed, so you need to change your sources.list
[09:10] <twb> I forget where the EOL stuff lives
[09:14] <popey> !eol
[09:14] <twb> Good thinking.
[09:16] <eutheria> i've got a weird issues using sssd to cache users from my ldap server
[09:17] <eutheria> i change to a local users, ie su - irc -s /bin/bash when i log out i get the message su: User not known to the underlying authentication module
[09:17] <eutheria> the log looks like http://pastie.org/4000576
[09:18] <eutheria> however local users aren't affected when sssd isn't running
[09:19] <eutheria> this has only started happened after upgrading from 11.10 to 12.04
[10:04] <brainysmurf> I'm getting "No such object" on ldapadd command. But I"m pretty sure I added the base correctly...
[10:05] <thys> so I'm getting ssh: connect to host warcamp.no-ip.org port 22: Connection refused after I upgraded to 10.04
[10:10] <greppy> thys: is sshd running?
[10:11] <thys> cant say, no access
[10:11] <thys> its on a serverfarm
[10:12] <Tm_T> thys: my basic rule is "make sure you have ssh installed before reboot"
[10:12] <Tm_T> but, it's quite possible that it just failed to boot
[10:12] <ikonia> brainysmurf: do an ldap query
[10:12] <thys> well, here is a site www.hallandstorget.se
[10:12] <ikonia> brainysmurf: confirm you can query the directory
[10:12] <ikonia> brainysmurf: then look at adding
[10:12] <Tm_T> thys: yes, rest could work just fine, just not ssh
[10:13]  * Tm_T is currently planning a trip to a server hall to find out what happened to one server that doesn't have ssh responing while everything else seems to work
[10:13] <Tm_T> ...after an upgrade (:
[10:14] <thys> fml
[10:15] <thys> so I might be planning a trip as well now. what should I do once I get there? or what should I tell someone to check or do?
[10:17] <Tm_T> I would start by checking ssh is installed, running, and responging in localhost, then checking if it responds to external connection
[10:17] <Tm_T> syslog and friends checked, ofcourse
[10:18] <Thorn> thys: can you ping your server?
[10:18] <thys> yes
[10:20] <brainysmurf> ikonia I can query the directory successfully but I am trying to add
[10:21] <brainysmurf> ikonia Do you mean add something else than what I'm already trying to add?
[10:21] <ikonia> brainysmurf: ok, so that's the good news, you do have a running directory service up and searchable
[10:37] <brainysmurf> ikonia: Bad news though is that I don't know how to add any users
[10:38] <ikonia> brainysmurf: ok, are you following a guide/document, using a tool etc
[10:41] <brainysmurf> I'm using https://help.ubuntu.com/10.04/serverguide/openldap-server.html and at populating the directory stage
[11:10] <brainysmurf> Hmmm, following command returns "no such object":
[11:10] <brainysmurf> ldapsearch -z 0 -b "o=ssis-suzhou.net" -D "cn=admin,dc=ssis-suzhou,dc=net" -W "(objectclass=*)"
[11:24] <zul> good morning
[11:24] <ikonia> brainysmurf: sorry I was just away, let me catch up now
[11:26] <thys> Im about to get billed by my hostingservice for logging in.. I only have a vague idea of what to tell him to type in to check why ssh is not working on the internet
[11:26] <thys> I was thinking pstree .. and looking for ssh
[11:27] <ikonia> ps -ef | grep ssh
[11:27] <ikonia> is that running yes/no
[11:27] <ikonia> step 2 "sudo service ssh start" - what is the error
[11:28] <ikonia> step 3 . ps -ef | grep ssh - is it now running
[11:30] <brainysmurf> THanks ikonia, I got it working on my own. I just needed to add a ou=people entry
[11:42] <ikonia> brainysmurf: ahh, well done
[12:34] <smoser> erichammond, i owe jbrown some work.
[12:36] <zul> smoser: you owe alot of things
[12:36] <smoser> i owe, i owe, its off to work i go
[12:42] <Captain_Proton> need some help with postfix. When I receive a email via smtp it passes to spamassion then it reject for some reason. here is a pastbin of it http://pastebin.com/3k8tDQfF
[12:51] <Captain_Proton> I found the problem anyone know how to fix this : Starting amavisd: Error in config file "/etc/amavis/conf.d/05-domain_id": Insecure directory in $ENV{PATH} while running with -T switch at /etc/amavis/conf.d/05-domain_id line 7.
[12:56] <ikonia> Captain_Proton: please don't cross-post the same question in multiple ubuntu channels
[12:57] <Captain_Proton> ikonia, sorry I know.. You got me :)
[12:57] <ikonia> Captain_Proton: not a problem
[12:59] <patdk-wk> and unless you manually used sa-learn, you would have had to turn on autolearn
[13:01] <Captain_Proton> ikonia, figure it out bug post https://bugs.launchpad.net/ubuntu/+source/amavisd-new/+bug/251377 my /usr/local was set to 777
[13:01] <Captain_Proton> :)
[13:13] <jcastro> adam_g: where you looking for me the other day?
[13:13] <jcastro> also, if you or jamespage could look at this: http://askubuntu.com/questions/141552/creating-volume-group-in-nova-volume-juju-charm
[13:14] <jamespage> jcastro, mm - I don't think the openstack charms work with the local provider - might be wrong
[13:25] <trond-> hi room. I am trying to install samba4 on a fresh 12.04 server, but I am getting permission denied when it is about to configure. install is done as root.
[14:47] <smoser> roaksoax, https://bugs.launchpad.net/ubuntu/+source/keepalived/+bug/160426
[14:48] <smoser> could you take a quick look at the most recent comment on there. it looks like someone reproduced on 10.04, possibly has a workaround/fix. you were last to look at that (recently re-opened) bug.
[14:48] <roaksoax> smoser: sure
[14:55] <jamespage> smoser: ping
[14:56] <smoser> here.
[14:57] <smoser> jamespage, ^
[14:57] <smoser> also, random question, as i'm triaging
[14:57] <smoser>  https://bugs.launchpad.net/ubuntu/+source/rabbitmq-server/+bug/968722
[14:57] <smoser> i would suggest that that bug is "triaged"
[14:59] <jamespage> smoser, agreed - if whomever wants to deal with that bug decides that its not required then they can 'Won't fix' it
[14:59] <jamespage> smoser, ping was re tracking of bugs for 12.04.1 for the server package set
[15:00] <smoser> ok.
[15:00] <jamespage> smoser: zul hacked the SRU report to focus on precise - http://people.canonical.com/~chucks/SRUTracker/reports/precise.html
[15:01] <jamespage> I'd like to progress that further so that we can use it both in the weekly team meeting to keep driving 12.04.1 forwards in terms of fixing and verification
[15:01] <jamespage> and to use for us to support updates in the 12.04.1 bi-weekly meeting
[15:02] <zul> people use stuff that i write? scarey!
[15:04] <smoser> jamespage, so you want separate sections that show things that are targetted to a milestone?
[15:05] <jamespage> smoser, I think that would be good
[15:06] <jamespage> that would allow us to see potential targets for 12.04.1 much easier
[15:10] <jamespage> smoser: let me spec something up in the next day or so and see if we can make it really useful
[15:11] <jamespage> smoser: in the meantime getting the team to target SRU candidates to 12.04.1 would be a good first step
[15:13] <smoser> jamespage, is there a class of things that should not be targetted to 12.04.1 ?
[15:13] <smoser> ie, things that should be SRU'd (targetted to distribution), but not immediately targetted to 12.04.1.
[15:14] <jamespage> smoser, hmm - probably not
[15:14] <smoser> other htan things that might not for some reason be able to make that milestone
[15:14] <jamespage> I guess the working assumption is that its 12.04.1 unless
[15:14] <jamespage> a) its know to be a different target date
[15:14] <jamespage> actually I think thats it
[15:49] <smoser> roaksoax, http://askubuntu.com/questions/130772/how-do-i-modify-the-pxe-config-in-maas
[15:49] <smoser> is that right?
[15:50] <smoser> you can't change that? or did you just chose to not mention /etc/cobbler/settings for non future-proofness.
[16:10] <roaksoax> smoser:he wants to edit the kernel args passed to a node and since we are getting rid of cobbler, I didn't mention anything about the commands
[16:10] <roaksoax> smoser: we don't really need to modify /etc/cobbler/settings
[16:11] <smoser> well... if you were trying to hide cobbler , then you should not have mentioned editing /var/lib/tftpboot necessarily.
[16:12] <roaksoax> smoser: my main point wasn't really to hide cobbler, but give the shortest path to modify the kernel args
[16:12] <roaksoax> smoser: doing with 'sudo cobbler' means first having to find out the system node name
[16:12] <roaksoax> and then passing the args
[16:12] <roaksoax> having to edit the pxe file directly, is much faster
[16:13] <smoser> in the case where it is just this specifric node that needs fixing, yes.
[16:13] <smoser> but he already knew he could edit the file :)
[16:16] <roaksoax> smoser: right.
[16:17] <roaksoax> smoser: well... you can provide your approach to the answer :)
[16:17] <smoser> well, i was thinking more generically he wanted to add parameters
[16:17] <jonathangf> good evening
[16:19] <roaksoax> smoser: yeah, well ultimately the ability to add parameters should go into maas
[16:32] <jonathangf> hi again, good evening
[16:35] <ea1het> good evening
[16:35] <ea1het> anyone with experience in KVM who can help me?
[16:42] <ea1het> anyone in the channel with experience with KVM (kernel virtual machine)
[16:42] <ea1het> ??
[16:42] <ikonia> you're going to have to be patient
[16:43] <ea1het> :)
[16:43] <ea1het> at least there is someone here in
[16:43] <ea1het> :)
[16:43] <ea1het> thanks
[16:43] <ea1het> for answering, i mean
[16:44] <ikonia> no problem
[16:45] <ikonia> ea1het: try to keep your questions to one channel though please rather than cross-posting the same question in other ubuntu channels
[16:45] <ikonia> ea1het: work out the best one for your needs, and query that one if possible.
[16:45] <ea1het> i will keep here in... no problem...
[16:46] <ikonia> cool, thanks
[16:47] <ea1het> you're welcome...
[16:47] <ea1het> :)
[16:51] <smoser> lynxman, you wanting to do rabbitmq merge?
[16:51] <smoser> https://bugs.launchpad.net/ubuntu/+source/rabbitmq-server/+bug/1004373
[16:51] <lynxman> smoser: sounds like a plan :)
[16:52] <smoser> can you upload that, lynxman ?
[16:52] <smoser> if not, i can review (and i can review even if you can upload)
[16:52] <lynxman> smoser: nope, it'll need to be sponsored, but I can do a branch merge request
[16:52] <lynxman> smoser: even if I could I'd be happy to have your review :)
[16:52] <smoser> or unhappy
[16:53] <ikonia> ea1het: I'm finishing off some stuff at the moment, if you are around when I'm done I'll try to help you with kvm
[16:53] <lynxman> smoser: unhappy and smoser don't go in the same phrase
[16:55] <ea1het> ikonia: good...
[16:55] <ea1het> i'm waiting the queue... :)
[17:07] <lynxman> smoser: 21 conflicts encountered <-- it'll take a while to merge :)
[17:10] <smoser> lynxman, right. we have a fairly large delta.
[17:10] <lynxman> smoser: yes we do
[17:10] <smoser> ideally, anything we are carrying would have a debian bug also unless it truely is ubuntu specific
[17:10] <lynxman> smoser: I'll do my best to merge and then try to contact the debian maintainer if that's okay
[17:11] <smoser> great.
[17:12] <roaksoax> lynxman: is this rabbitmq merge?
[17:12] <lynxman> roaksoax: indeed
[17:13] <lynxman> roaksoax: we merged all the plugin packages into the rabbitmq-server package (since it was pretty much there), that plus other supercool stuff that SpamapS added make the big delta
[17:13] <roaksoax> lynxman: i was working on it (was using grab-merge.sh instead because bzr wouldn't cooperate), and most of the conflicts are simple to resolve, and IIRC some of them are simply to take debian's version
[17:14] <lynxman> roaksoax: ah sorry, smoser just asked me to merge :) let me know if I can be of any help
[17:14] <roaksoax> lynxman: go for it :) I was working on it last week but didn't finish... :)
[17:14] <lynxman> roaksoax: ah cool, will continue then!
[17:14] <lynxman> roaksoax: thanks
[17:14] <smoser> ugh. sorry, roaksoax lynxman for adding confusion
[17:15] <lynxman> smoser: hey no problemo at all
[17:15] <roaksoax> smoser: no worries
[17:16] <roaksoax> :)
[17:42] <greppy> grrrrr
[17:42] <greppy> ( doh, wrong channel )
[17:51] <ea1het> ikonia: are you still here ?
[17:55] <JonEdney> Anyone aware of any server monitoring software?  I'm looking for something to monitor processes, and to be able to set certain limits, that will create an email.
[17:59] <greppy> JonEdney: take a look at xymon/hobbit
[17:59] <ea1het> ikonia: are you still here ?
[18:08] <jonconley> JonEdney: #zabbix
[18:12] <JonEdney> I was looking at Zabbix last night, but couldn't for the life of me get it installed on my server, so I decided to keep looking.
[18:12] <JonEdney> Maybe I'll give it another shot today.
[18:21] <rockets> What's the difference, in a cron, between 0 9,12,15,18,21 * * * and * 9,12,15,18,21 * * *
[18:21] <rockets> Does the second run every minute of all of those hours?
[18:24] <Pici> rockets: yes.
[18:24] <rockets> Pici, thanks.
[18:27] <jonconley> JonEdney: where were you running into issues?
[18:27] <JonEdney> I"m relatively new still to Linux, i don't recall exactly, I may have just not given myself enough time.
[18:29] <smoser> JonEdney, well, when you figure it out, you can "charm" it for juju
[18:29] <smoser> bug 972328
[18:35] <jamespage> JonEdney, the server guide references a few monitoring solutions - https://help.ubuntu.com/12.04/serverguide/monitoring.html
[18:44] <JonEdney> I found a web-based monitoring system that isn't free, but if you have 1-2 servers, they monitor for free.  Anyone heard of  CopperEgg?
[18:55] <jmedina> Hi there,
[18:55] <jmedina> I'm doing a research for integratin live migration in a virtual enviroment with KVM and ubuntu 12.04
[18:55] <jmedina> I like to use a iSCSI SAN with at least 6TB
[18:56] <jmedina> do you know or have experience with something like this?
[18:56] <jmedina> any model you could recomend?
[18:59] <freesbie> jmedina: depends on your iops need and budget really :)
[19:01] <jmedina> freesbie: actually we dont have those numbers, I want to compare features and price
[19:04] <freesbie> jmedina: iops is pretty damn important if youre gonna virtualize, unless its just storagemachines youre putting on it (which is a bit stupid) :)
[19:06] <jmedina> freesbie: I know, I just learning how to get those numbers, I hope I can get those numbers in the next weeks
[19:06] <jmedina> freesbie: any recomendation?, I have read a little about sysstat
[19:07] <freesbie> jmedina: you could easily build your own iscsi machine with ubuntu and use drbd to replicate to another machine. or you could go with the commercial ones, lefthand, hitachi, equallogic etc
[19:09] <jmedina> freesbie: yeap, I already built one using ubuntu server drdb+lvm and iscsi, but this customer is a big corporation and they want something "enterprise" with commercial support
[19:09] <jmedina> I just convince them to use KVM instead of VMWARE
[19:10] <freesbie> jmedina: depends on the utilization of your existing physical harddrives, and how much you want to pay to be able to give the same max iops or you expect 50% utilisation and so forth ..
[19:12] <jmedina> freesbie: they are all underutilized with sata discs, it is a small site, actually 12 vms
[19:13] <freesbie> ive been using hitachi and emc on fibrechannel with vmware on top earlier at the places ive worked. where im now im thinking about buying lefthand and using KVM
[19:13] <freesbie> try this site to calculate your iops http://www.wmarow.com/strcalc/
[19:14] <jmedina> vmware it is really expennsive for this small proyect, 3 times
[19:14] <jmedina> freesbie: thanks, I was about to ask for a tool like that :)
[19:21] <jmedina> freesbie: thanks for your time :)
[19:22] <freesbie> jmedina: yw
[19:56] <jamespage> SpamapS, around - seems we need to discuss ceph?
[19:59] <SpamapS> jamespage: probably yes. :)
[19:59] <jamespage> SpamapS, have time now?
[20:00] <SpamapS> jamespage: I'm in the middle of a conversation.. hopefully will be resolved soon.
[20:00] <jamespage> SpamapS, sure - ping me when you are done
[20:02] <Zanzacar> does anyone have any recommendations on pci raid controllers? I have no idea what to even look for.
[20:14] <rockets> Anybody know any decent alternatives to logcheck?
[20:14] <rockets> It doesn't appear to be maintained
[20:19] <ea1het> hi good night
[20:19] <ea1het> anyone here to talk?
[20:22] <mgw> e1het: about what?
[20:24] <ea1het> mgw: hi, thanks for answer. I would like to know.... i guess the utility of LVM for the further use of a KVM-based hypervisor
[20:24] <ea1het> i don't know if you understand...
[20:32] <jmedina> rockets: I use ossec
[20:34] <never2far> can someone explain me why is _U and not U_ ...since the failed device is the second from the list ...or U and _ is counted alphabetically ?! ...so here are the lines from /proc/mdstat
[20:34] <never2far> md1 : active raid1 sdb3[2] sda3[0](F)
[20:34] <never2far>       7157752 blocks super 1.1 [2/1] [_U]
[20:34] <never2far>       bitmap: 1/1 pages [4KB], 65536KB chunk
[20:35] <never2far> or in this example(from wiki.kernel.org)
[20:35] <never2far> md0 : active raid5 sda1[0] sdd1[2] sdb1[1]
[20:35] <never2far>      1465151808 blocks level 5, 64k chunk, algorithm 2 [4/3] [UUU_]
[20:36] <never2far> what device is down ? maybe sdc1 ...but how can i determine this from the staus of [UUU_] ?!
[20:56] <grendal-prime> is there a way to set the exit status of a script to 0 even though its not 0
[20:56] <colon_D> exit 0
[20:56] <smw_> grendal-prime, in what context?
[20:59] <SpamapS> jamespage: here now, tho time is limited. :)
[21:08] <stgraber> SpamapS: do you think you can review the lxc SRU this week? I know it has a long changelog and the diff isn't exactly short so it seems to have scared most SRU reviewers this week, but I'd really like to see it hit proposed soonish (as I already have the next batch of fixes stacking here).
[21:13] <SpamapS> stgraber: I've spent > 8 hours doing SRU's this week
[21:14] <SpamapS> stgraber: I will try to get to it.
[21:15] <iToast> Can i get help wiuth the server installer --_
[21:15] <iToast> -_-*
[21:15] <stgraber> SpamapS: thanks and thanks again for all the work on the SRU queue this week, good to see it fit on a single page again!
[21:16] <iToast> so...
[21:16] <iToast> Can I get help...
[21:17] <iToast> Who ever made the server 10.4 installer forgot people will USB boot every here and there....
[21:17] <iToast> It wants me to mount a non existent cd-rom.
[21:17] <iToast> How do i make it goto the usb its booted from
[21:19] <ScottK> iToast: Does it work with 12.04?
[21:20] <iToast> ScottK: Thats like throwing a match into a gas tanker, no.
[21:22] <ScottK> Not really.
[21:22] <ScottK> A lot of bugs got fixed between 10.04 and 12.04.
[21:23] <ScottK> Unless there's a reason you really, really need 10.04, I'd try the newer release.
[21:29] <freesbie> scottk: new bugs have also been introduced, so maybe thats a reason to stick with the older and still supported release :)
[21:29] <ScottK> The general trend is, IME, towards better.  Particularly in the installer.
[21:30] <ScottK> Two years ago, booting USB was much less common than it is now, so it got a lot broader testing and more attention during 12.04 development than it would have for 10.04.
[21:30] <ScottK> I have a vague recollection of a specific bug like that which I know isn't around anymore.
[21:30] <ea1het> can anyone explain me the MaaS concept ??
[21:31] <freesbie> well, ive never had any troubles booting 10.04 from usb .. only had troubles with unsupported raidcontrollers ..
[21:34] <ScottK> I can't recall what release I had it on.
[21:37] <RoyK> ScottK: I never choose a x.x release for a server, better wait for x.x.1 ;) that's also the reason do-release-upgrade won't upgrade 10.04 to 12.04 until 12.04.1 ;)
[21:37] <RoyK> ea1het: basically similar control over hardware as that over virtual machines
[21:38] <ScottK> RoyK: The upgrade bit has more to do with testing the LTS to LTS upgrade path than anything about new installs.
[21:38] <ScottK> I didn't upgrade my 10.04 servers yet, but if I was doing a new install, I'd do 12.04.
[21:38] <RoyK> doesn't matter, x.x.1 is, im my experience, far better tested, which is rather logial
[21:38] <RoyK> or logical
[21:39] <ea1het> RoyK: Maas = PaaS  ????
[21:39] <RoyK> meaning: If I'm to install a file server or something I *need* to keep stable, I'd use 10.04 for now, and perhaps another year.
[21:39] <ScottK> No.
[21:40] <ScottK> https://wiki.ubuntu.com/ServerTeam/MAAS
[21:41]  * RoyK greets good night and is off for some hourzzzz
[21:41] <ea1het> ScottK: i was right there reading.. but i don't see it clear....
[21:41] <ScottK> Honestly I don't either.
[21:41] <ea1het> hehehehehe
[21:44] <ihashacks> MaaS: setup MaaS "master", PXE boot a bunch of physical servers which will launch Ubuntu installer, join said servers to the MaaS, those servers then shutdown and wait for the master to summon them for some kind of deployment
[21:45] <ihashacks> MaaS + Juju is a neat example of a provisioning method
[21:46] <ihashacks> instead of "gimme a VM with foo service on it", you get, "gimme a physical server with foo service on it"
[21:47] <ihashacks> then one of your MaaS nodes gets wake-on-lan'ed, says "yes sir" and starts setting up foo service on it
[21:52] <ihashacks> MaaS is likely useless if you're a small shop with relatively few physical servers but probably awesome if you have a large purchase of racked equipment you need to bring online in a pinch.
[22:19] <grendal-prime> smw
[22:21] <grendal-prime> soooo i have this script.  It gathers some info and then sends off a radius packet using radclient..problem is the radius server never responds (which is OK)
[22:21] <grendal-prime> but the script (its a openvpn client connect script exits with a 1 and then the opevpn server will not let the client connect.
[22:30] <SpamapS> ihashacks: MaaS is also intended to be useful for the case where you have just a few servers. What about it do you think would make it inadequate for that case?
[23:11] <SpamapS> stgraber: hey are you here? I am looking at the LXC upload to precise-proposed.. only one question.
[23:13] <SpamapS> stgraber: there is one change which I can't find a bug for.. its minor, but "Make sure /etc/resolv.conf is valid before running any apt command."
[23:14] <SpamapS> ahh, no actual report, I see that now in the changelog
[23:14] <SpamapS> well it looks like somethign that would work as an SRU
[23:14] <ihashacks> SpamapS: perhaps s/useless/less usefull/
[23:15] <ihashacks> SpamapS: If MaaS is designed to provision servers in a sleeping state unti you WoL them, and you only have a couple of servers to begin with, then you have servers that only sit there until you summon them from MaaS ...
[23:15] <ihashacks> ...unless I'm missing a valid use case here?
[23:16] <SpamapS> stgraber: n/m .. it all looks good. Thanks!
[23:16] <SpamapS> ihashacks: thats such a tiny piece of what i does :)
[23:16] <SpamapS> s/ i / it /
[23:17] <SpamapS> ihashacks: just automating that bit is useful with a few servers if you are going to test deploying something over and over.
[23:18] <ihashacks> SpamapS: touchee
[23:18] <ihashacks> SpamapS: I suppose I was thinking more post-testing, in-production - why, if you only have a few physical servers to begin with, would you have them off, waiting to be summoned by MaaS?
[23:18] <ihashacks> That is definitely a use case I did not think about.
[23:20] <SpamapS> ihashacks: well if you only EVER plan to have those few servers, why have any automation at all? :) MaaS should make it easier to grow.
[23:32] <ihashacks> perhaps I should have said, "MaaS is likely less usefull if you're a small shop with relatively few physical servers and will likely always have said limited amount of physical hardware to deploy."
[23:32] <ihashacks> ;-)
[23:35] <SpamapS> ihashacks: said shop is likely a boring shop to work in. ;)
[23:35] <SpamapS> unless its like, a circus.. or maybe a gun range... nothing but laughs in a gun range.
[23:41] <txomon|home> has anyone used cobbler here?
[23:42] <txomon|home> I am trying to add ubuntu distros to the cobbler installation, but I don't succeed in adding the netboot files (I don't want to mount/import CDs, as I have all in a mirror)