[00:00] <gary0> yep.
[00:00] <gary0> that's what it's for  =)
[00:01] <JanC> right  :-)
[00:01] <gary0> i've got a hardware RAID card with 4 drives, and room for 4 more. i wanna store my uncompressed HD on here so i can access it via OpenVPN from work.
[00:02] <gary0> but i'm not good with deleting, so i'll prob need to add drives over time to the full 8.
[00:02] <JanC> do you also use LVM?
[00:03] <Bilge> LUKS doesn't partition drives, it formats them
[00:03] <gary0> i didn't think i'd need to, since this will be the only partition on the RAID array.
[00:03] <Bilge> You might be getting confused between LUKS and LVM
[00:03] <gary0> i encrypted the standalone drive, too, and used LVM to cut it up into a few ext3 partitions for some VMs.
[00:04] <JanC> I think LVM is what you would need to be able to enlarge "logical disks"
[00:04] <gary0> yeah, LVM let's you resize live partitions dynamically.
[00:04] <gary0> but since i'm only going to have the one partition on the RAID, i don't *think* i need it.
[00:05] <gary0> the only think that made me pause was when i realized that i wasn't sure if i could add more free space to the encrypted partition.
[00:07] <gary0> linux's handling of partitions is new to me... with windows, it's all much simpler.  =P
[00:07] <JanC> gary0: it's not simpler, you just have less options  ;)
[00:07] <gary0> exactly.
[00:08] <gary0> i mean, if i just drop encryption and format the drive as XFS, i can add new free space to the XFS partition.
[00:09] <JanC> I'm not an LVM expert, but I think you'd want to use it for the purpose you're describing
[00:09] <gary0> maybe.
[00:09] <gary0> i first heard of LVM 2 days ago, when i started researching all this.
[00:10] <JanC> windows has something like LVM too, it's just less powerful  ;)
[00:10] <JanC> and on linux/unix, people like to put things into layers
[00:11] <gary0> i noticed.
[00:11] <gary0> in the installer, i took the standalone drive and set it as a 'physical volume for encryption'
[00:12] <gary0> which crated a new drive.
[00:12] <gary0> then i took that and created an LVM group in that.
[00:12] <gary0> then in the group i created the actual partitions.
[00:12] <gary0> it certainly makes sense.
[00:13] <JanC> once you get used to it
[00:13] <gary0> and i'm pretty sure i found info on how to add another physical drive to the encrypted volume... which i can then include in the VG, and then in the VPs
[00:14] <gary0> but i couldn't find anything on adding existing free space on the existing physical drive.
[00:14] <gary0> or maybe i did and just didn't know that i'd found what i was looking for.
[00:14] <JanC> except that currently your data on the RAID isn't inside a LVM layer?
[00:14] <gary0> well i don't have any data on the RAID yet, even.
[00:14] <gary0> i'm still at the installation screen  =P
[00:14] <JanC> oh, in that case
[00:15] <JanC> not sure about the installation screen
[00:15] <gary0> right now, i've partitioned the system drive... and i was about to partition the RAID when this worry came up.
[00:15] <gary0> ubuntu server's automated installer that comes on the CD
[00:15] <gary0> well, the CD image on the site.
[00:16] <JanC> I'd certainly make sure you use LVM in that case, if not possible during the install, then afterwards (or set up from the live CD before)
[00:16] <gary0> http://kuparinen.org/martti/comp/ubuntu/en/cryptolvm.html
[00:17] <gary0> that's the guide i found for setting up an encrypted partition.
[00:17] <gary0> i guess using LVM doesn't hurt regardless of anything.
[00:17] <gary0> it adds a layer, but as long as i can figure it out, it's not an issue.
[00:19] <JanC> every layer slows things down a very little bit, but if you want encryption, that slows things down a lot...  ;-)
[00:20] <gary0> well since i'm gonna be accessing this over the LAN, even with jumbo frames i won't get transfer rates higher than 60-70MB/s
[00:20] <gary0> so the access time isn't an issue.
[00:21] <JanC> transer rates != access times, but whatever  ;-)
[00:21] <gary0> however, i was hoping to also use the server (Phenom II x2) to do transcoding and run a few VMs, so if encryption increases the cpu usage a lot, that will definitely affect me.
[00:23] <gary0> i just meant that i can't imagine 4 drives in a RAID5, even with a slowdown due to on-the-fly encrypt/decryption, dropping below the 70MB/s read throughput.
[00:23] <gary0> unless you meant slow things down as in the system itself...
[00:28] <JanC> gary0: that's exactly the sort of thing only you can determine; every layer slows things down, and encryption slows things down considerably, but if the slow-down  is acceptable for your purpose, that's okay! ツ
[00:29] <JanC> it's a cost/benefit decision
[00:30] <gary0> i have no idea how much of an effect it'll have.
[00:30] <gary0> i guess i'll have to try and see.
[00:30] <gary0> can i remove the encryption once it's in?
[00:30] <gary0> or would i need to back up the data and repartition everything from scratch?
[00:31] <JanC> last option, I fear
[00:32] <JanC> if you have spare diskspace, you can move things around of course
[00:33] <JanC> I doubt encryption would be an issue though
[00:36] <gary0> i guess we'll see..  thanks for the help  =)
[02:15] <luddite33> hi all - just wondering if anyone is running 9.10 beta with a apache/mysql setup up thats in production yet?
[02:27] <jmarsden> Why would anyone with any sanity run a beta OS on a production server? :)
[02:36] <KurtKraut> jmarsden, the high adoption of Windows on servers (and even desktops) shows how people are tolerant with system failures.
[03:07] <selje> ebox users?
[03:12] <uvirtbot`> New bug: #285439 in backuppc (main) "backuppc does not play nicely with updatedb" [Wishlist,New] https://launchpad.net/bugs/285439
[03:30] <DKcross> hello, some know any tool with i can read what is the distro of any server?
[03:30] <DKcross> I'm trying with nmap
[03:31] <DKcross> sudo nmap -O  200.30.138.227
[03:39] <qman__> DKcross, nmap is about the best you can get, since that information is (for good reason) not explicitly reported by remote hosts
[03:40] <qman__> properly configured firewalls further obscure the information
[03:40] <DKcross> qman__, thanks, my dude is for one reason.... i have hosting server, but i'm not sure what distro use...
[03:40] <qman__> DKcross, if you have shell access, check /etc/issue
[03:40] <DKcross> is linux but i don't know if centOS or redhat..i'm no sure
[03:40] <qman__> and uname -a
[03:41] <qman__> centOS and redhat are quite similar
[03:41] <DKcross> qman__, Linux pikachu 2.6.29.6-aufs2-grsec-v1.4-i686-opt #10 SMP Sat Oct 3 10:27:32 PDT 2009 i686 GNU/Linux
[03:42] <DKcross> this line is about uname -a
[03:42] <qman__> yeah
[03:42] <qman__> probably a custom kernel, I don't know of any distros code-named pikachu
[03:42] <qman__> cat /etc/issue, if it's there
[03:42] <qman__> might tell you
[03:43] <DKcross> yes!
[03:43] <DKcross> is debian server
[03:43] <DKcross> [pikachu]$ cat /etc/issue
[03:43] <DKcross> Debian GNU/Linux 3.1 \n \l
[03:44] <qman__> nice
[03:44] <qman__> personally, I really dislike redhat and derivatives
[03:44] <DKcross> :o
[03:44] <DKcross> what is your favorite o prefere?
[03:45] <qman__> ubuntu, naturally
[03:45] <qman__> but I like the debian way far better than the redhat way
[03:45] <qman__> I'll even take BSD over redhat
[03:46] <DKcross> good
[03:46] <DKcross> what about open suse like server?
[03:46] <DKcross> yes i know is similar to red hat.. because is a derivatives..
[03:47] <qman__> I've actually never used suse
[03:47] <DKcross> ohh ok.
[03:47] <qman__> I've only used redhat 9, fedora, RHEL and centOS
[03:47] <qman__> didn't much like any of them
[03:48] <DKcross> RPM systems
[03:51] <qman__> I much prefer apt/dpkg to yum and rpm, and I like the way debian does init scripts and conf files over redhat's tools
[03:52] <qman__> and just a number of other bits about how the systems are configured
[03:58] <DKcross> qman__,  +1
[03:58] <DKcross> :D
[06:58] <proteusguy> does Ubuntu offer a system I can run locally that let's me build my own private cloud supporting EC2 apis? i want to be able to migrate back&forth between my own cloud and EC2.
[07:37] <erichammond> proteusguy: Take a look at Eucalyptus in Karmic.
[07:47] <MazuGrazus> Hellow, does ubuntu fully run cloud computing? what can i do using ubuntu cloud?
[08:05] <proteusguy> erichammand: this is what i've heard of but does it let me migrate my environment between my local cloud & ec2?
[08:13] <Brylie> sommer: thanks for writing the ubuntu server guide :-)
[09:02] <uvirtbot`> New bug: #454521 in eucalyptus (main) "Eucalyptus installer should skip range-of-public-ip's question in favour of DHCP" [Wishlist,Triaged] https://launchpad.net/bugs/454521
[09:02] <uvirtbot`> New bug: #454540 in eucalyptus (main) "DHCP requests for public IPs should use DHCP options" [Wishlist,Triaged] https://launchpad.net/bugs/454540
[09:16] <uvirtbot`> New bug: #455246 in eucalyptus (main) "Unnecessary depend on "ecj" adds too many packages to server CD" [Wishlist,Triaged] https://launchpad.net/bugs/455246
[10:27] <blinkiz> Hi. Am trying to automate installation of a package from repository. It asks three questions with debconf style. I know I can pre define these answers with debconf-set-selections. problem is, I don't know the string to set with debconf-set-selections. How can I find out what the package are questioning about? I guess it's in the source package and debian directory?
[10:29] <blinkiz> oh, found it. debconf-get-selections.. Okay. That was not so hard ;)
[10:33] <nijaba> ttx: I have finished testing/updating UECInstall, should we proceed doing the switch with UEC?
[10:34] <nijaba> that is on help.ubuntu.com/community/ of course
[10:36] <ttx> nijaba: can't look right now, but that seems like a good idea, if the right pointers are present
[10:36] <nijaba> ttx: right pointers?
[10:37] <ttx> The "UEC" doc should stay to cover multi-component install ?
[10:37] <ttx> (granted, it needs some fixes right now)
[10:38] <ttx> maybe rename UEC -> UECmanualinstall and UECinstall -> UEC ?
[10:42] <acalvo> hi
[10:42] <acalvo> anyone know if the keyserver.ubuntu.com is down?
[10:43] <acalvo> been trying to get a gpg key for about an hour or so and still not able to
[10:44] <nijaba> ttx: renaming to UEC/PackageInstall
[10:45] <nijaba> ttx: pointer is already in place
[10:45] <ttx> nijaba: ok, cool
[10:47] <soren> acalvo: I've had problems as well. Try another one instead.
[10:47] <soren> acalvo: Like subkeys.pgp.net or something.
[10:52] <nijaba> ttx: done
[10:56] <acalvo> soren: thanks
[11:18]  * soren lucnhes
[11:18] <soren> ..and lunches
[11:19] <_ruben> sounds like a plan
[11:26]  * nijaba lunches and goes get a haircut :P
[11:44] <J_P> hi all
[11:45] <J_P> people, how I change in clock time of BRT to BRST?
[11:53] <soren> J_P: You don't. It happens automatically.
[11:54] <J_P> soren: yes, if as configured corectaly when was intalled. So agaim, how I change the timezone?
[11:55] <soren> J_P: You don't.
[11:55] <soren> It happens automatically.
[11:55] <J_P> soren: here not happen
[11:55] <soren> You don't have to manually  take care changing anything when daylight savings time kicks in.
[11:55] <J_P> soren: here not happen
[11:55] <soren> J_P: What does "date" say?
[11:56] <J_P> soren: Seg Out 19 07:55:54 BRT 2009
[11:57] <J_P> soren: in other machine I have right clock: Seg Out 19 08:56:40 BRST 2009
[11:57] <soren> J_P: Which version of tzdata do you have installed?
[11:57] <J_P> soren: I don't know, where I see?
[11:57] <soren> dpkg -l tzdata
[11:58] <J_P> soren: ii  tzdata                            2007e-0ubuntu0.7.04               Time Zone and Daylight Saving Time Data
[11:58] <soren> You're running Feisty?
[11:59] <J_P> soren: yes, feisty = 7.04
[11:59] <soren> Feisty has not been supported for almost a year.
[12:00] <soren> That's your problem.
[12:00] <J_P> soren: yes, I know
[12:00] <J_P> soren: is a joke, I need updagrade just because tzdata is not upgrade
[12:00] <soren> Your tzdata version does not have correct information about when Brazil has daylight savings.
[12:00] <soren> What?
[12:01] <J_P> soren: yes, But I can update just tzdata, and not entire distribuition right
[12:01] <soren> J_P: Feisty is not supported anymore. At all.
[12:01] <J_P> soren: ok, thanks
[12:01] <soren> You're likely vulnerable to a whole bunch of security problems.
[12:01] <soren> You really, /REALLY/ should upgrade.
[12:01] <soren> But yes, you can probably grab tzdata from a later ubuntu version and install that.
[12:02] <soren> But you really, really, really should upgrade ASAP.
[12:02] <J_P> soren: ASAP?
[12:02] <soren> As Soon As Possible.
[12:02] <J_P> ahh ok
[12:02] <soren> Like... Right now.
[12:02] <soren> Yesterday, if you can.
[12:03] <J_P> soren: I a server, is more complicated..
[12:03] <J_P> soren: because this , Now I just use ONLY LTS in servers
[12:03] <soren> The version of tzdata you have is not even the most current one for Feisty.
[12:05] <soren> http://old-releases.ubuntu.com/ubuntu/pool/main/t/tzdata/tzdata_2008g-0ubuntu0.7.04_all.deb
[12:05] <soren> Try installing that one and see if it fixes it.
[12:05] <soren> J_P: ..but you should still upgrade.
[12:06] <J_P> soren: ok
[12:27] <J_P> soren: I just copy /etc/locatime from 8.04 to 7.04 and works :-)
[12:29] <J_P> soren: but I will upgrade it.. :-)
[12:36] <Bilge> How could I get a file from a package without installing the entire package
[12:36] <Bilge> For example just the init.d script
[13:00] <zul> morning
[13:22] <sommer> mornings :)
[13:36] <Omahn> Hello all. Would this be the correct please to ask someone to follow up on a sponsorship request for bug 341256?
[13:36] <uvirtbot`> Launchpad bug 341256 in unattended-upgrades "RFE: unattended-upgrades should identify host." [Medium,Fix released] https://launchpad.net/bugs/341256
[13:36] <Omahn> That one :-)
[13:36] <UnixDawg> morning any of the server devs here
[13:36] <Omahn> Ah, cool. Any takers? :-)
[13:36] <UnixDawg> I need help with some php mods I am not fimding in the repos
[13:37] <UnixDawg> php5-xml php5-json php5-xmlwriter php5-xmlreader php5-posix php5-iconv php5-hash php5-simplexml php5-pcre php5-filter php5-dom php5-session php5-spl
[13:37] <acalvo> UnixDawg: apt-cache search <string>
[13:37] <acalvo> is what I usually do to find packages
[13:37] <UnixDawg> I did they are not there
[13:37] <acalvo> or use aptitude
[13:38] <UnixDawg> thats the issue
[13:38] <acalvo> UnixDawg: I'd do apt-cache search php xml
[13:39] <greenkernel1> hello
[13:39] <UnixDawg> dont find them and I need them
[13:40] <greenkernel1> I would like to run a file server. could someone help me please?
[13:40] <UnixDawg> use nfs or samba
[13:40] <acalvo> greenkernel1: take a look at the ubuntu community help about samba
[13:40] <UnixDawg> your chooice
[13:41] <greenkernel1> I'm not using ubuntu server edition currently. I'm using desktop version. but, I like to add a software to help my machine runs like a server.
[13:42] <greenkernel1> is that possible?
[13:42] <acalvo> UnixDawg: did you take a look at php5-cli? it includes several packages
[13:42] <UnixDawg> yes
[13:42] <acalvo> greenkernel1: in fact is easier if you're using the desktop version
[13:42] <UnixDawg> I have it installed
[13:42] <acalvo> just create a folder
[13:42] <acalvo> and check share permissions
[13:42] <acalvo> that'd run as a file server
[13:43] <greenkernel1> okay .. what software should i install?
[13:43] <UnixDawg> but looking in the/usr/lib/php 1/2 the modules are not listed in the extensions.ini
[13:43] <acalvo> greenkernel1: nothing, just try to do that and it'd tell you what needs to be installed
[13:44] <greenkernel1> try to do what? sorry .. I'm quite new to networking.
[13:44] <acalvo> greenkernel1: just create a folder and try to share it
[13:44] <acalvo> it will pop-up a dialog about options
[13:44] <greenkernel1> share with who?
[13:44] <acalvo> share across the network
[13:44] <UnixDawg> none of these php mods show up
[13:44] <UnixDawg> php5-xml php5-json php5-xmlwriter php5-xmlreader php5-posix php5-iconv php5-hash php5-simplexml php5-pcre php5-filter php5-dom php5-session php5-spl
[13:45] <UnixDawg> none of them show up in the repo list or in the php modules dir
[13:45] <UnixDawg> they are missing
[13:46] <greenkernel1> okay .. I think I need to tell you what I want to do exactly first. I want to share my files with my friends.
[13:46] <zul> ttx: have you seen the likewise open evenlog bug? bug 454164
[13:46] <acalvo> across the network or across the internet?
[13:46] <uvirtbot`> Launchpad bug 454164 in likewise-open5 "likewise-open5-eventlog doesn't properly install" [Undecided,New] https://launchpad.net/bugs/454164
[13:46] <acalvo> greenkernel1: https://help.ubuntu.com/community/SettingUpSamba
[13:46] <greenkernel1> and I don't want to upload my files to another server like hosting services.
[13:46] <VK7HSE> anyone have a LAMP installed in Karmic? if so what is the result from ...   php -r 'var_dump(function_exists("gzopen"));'
[13:46] <ttx> zul: yes, I saw that. I need to look into it
[13:46] <greenkernel1> across the internet.
[13:47] <UnixDawg> well this makes deb unuseable for the project
[13:47] <UnixDawg> this sucks
[13:47] <UnixDawg> to many pkgs missing
[13:47] <acalvo> greenkernel1: ok then, you'll need to install a ftp server
[13:47] <zul> ttx: just making sure
[13:47] <acalvo> UnixDawg: look http://packages.ubuntu.com/jaunty/php5-cgi
[13:48] <greenkernel1> I've already installed vsftpd and tested .. it didn't work for me.
[13:48] <acalvo> this modules are already in this package: bcmath bz2 calendar ctype date dba dom exif filter ftp gettext hash iconv json libxml mbstring mime_magic openssl pcre posix Reflection session shmop SimpleXML soap sockets SPL standard sysvmsg sysvsem sysvshm tokenizer wddx xml xmlreader xmlwriter zip zlib.
[13:48] <acalvo> greenkernel1: https://help.ubuntu.com/8.04/serverguide/C/ftp-server.html
[13:49] <UnixDawg> I have installed all the php5 libs that are in the repo and still the modules/extensions I have klisted do not show up
[13:50] <greenkernel1> I just want to share my files to others (no write access to my drive). some people said it's better to run a http server.
[13:50] <greenkernel1> is that right?
[13:50] <acalvo> greenkernel1: ftp means file transfer protocol, so I'd point to use FTP instead of HTTP
[13:51] <greenkernel1> and before we discuss further, some people said I'm behind a NAT. is it possible to run a ftp server for me?
[13:51] <UnixDawg> so deb lacks alot of the php extensions
[13:52] <UnixDawg> O well
[13:52] <acalvo> greenkernel1: it is, if you have access to the policies of your router and you can map the port 21 to your machine (or any other port you want to use)
[13:53] <UnixDawg> root@ubuntupbx:/usr/lib/php5/20060613+lfs# ls curl.so    mysqli.so  pdo_mysql.so  pdo_sqlite.so  tidy.so mcrypt.so  mysql.so   pdo_odbc.so   sqlite3.so     xmlrpc.so mhash.so   odbc.so    pdo.so        sqlite.so      xsl.so
[13:53] <greenkernel1> I think I don't have access to change the policies of the router.
[13:53] <greenkernel1> how could I check if I have access to the router?
[13:53] <UnixDawg> is all the php extensions that are installed
[13:53] <UnixDawg> php5-xml php5-json php5-xmlwriter php5-xmlreader php5-posix php5-iconv php5-hash php5-simplexml php5-pcre php5-filter php5-dom php5-session php5-spl  are missing
[13:54] <UnixDawg> and I have looked for them
[13:54] <acalvo> UnixDawg: why don't you try to create a test file with the modules you need and see if they're already there?
[13:54] <UnixDawg> and with out these extensions the php app wont work
[13:55] <acalvo> greenkernel1: you should know the IP of your router, and the user/password to access the router configuration
[13:55] <UnixDawg> I am a bsd person  trying to learn deb server . I dont know how to make pkgs on deb
[13:55] <greenkernel1> okay .. how could I find the IP of the router?
[13:55] <UnixDawg> there not there
[13:55] <acalvo> UnixDawg: I don't think you'll need to create deb packages...
[13:55] <UnixDawg> they would be listed in the extensions dir if they where
[13:56] <uvirtbot`> New bug: #454263 in backuppc (main) "Update of backuppc (server) crashed on update (dup-of: 445154)" [Undecided,New] https://launchpad.net/bugs/454263
[13:56] <UnixDawg> installed
[13:56] <Bilge> How could I get a single file from a package without installing the entire package first?
[13:56] <pmatulis> Bilge: 'apt-get source foo'
[13:57] <UnixDawg> cd /usr/src
[13:57] <Bilge> Does source mean source code only or would it include package maintainer's files such as init.d scripts
[13:57] <Guest76209> greenkernel1: Run "route" and look at the default route's IP
[13:57] <pmatulis> Bilge: everything needed to build the package
[13:58] <pmatulis> Bilge: what package and what file are you after?
[13:58] <Bilge> znc as it happens
[13:58] <Bilge> init.d script
[13:58] <Bilge> Does it make any difference?
[13:58] <acalvo> UnixDawg: from what I can see, almost all your packages are or inside the -cli, -cgi or standalone (like -dom or -json)
[13:58] <UnixDawg> well I would need a .deb for each module so I can install them on other servers when we build
[13:58] <ttx> smoser: re:"Low" : will be fixed, but not in karmic
[13:58] <smoser> right.
[13:59] <greenkernel1> okay .. this is my route result >> http://paste.ubuntu.com/296819/ .. can you take a look please?
[13:59] <acalvo> greenkernel1: ifconfig should tell you more about your network
[13:59] <pmatulis> Bilge: i'm curious
[13:59] <pmatulis> Bilge: what release?
[13:59] <acalvo> greenkernel1: should be 10.64.64.64
[13:59] <smoser> oh, and one other poke i forgot to put in that response... "if your product claims itself as 'EC2 compatible cloud'" then bugs like "this doesn't act like EC2" are kind of important, ttx
[13:59] <UnixDawg> I have searched the system for the extensions they dont exist
[14:00] <greenkernel1> okay .. how should I connect to that address?
[14:00] <UnixDawg> so there for it means they where not installed
[14:00] <UnixDawg> so they are not in any deb pkg
[14:00] <ttx> smoser: yes, I just wasn't sure how common it was to crawl the metadata service like this
[14:00] <Bilge> pmatulis: 8.04
[14:00] <smoser> probably not too common. admittedly.
[14:00] <acalvo> UnixDawg: I wouldn't say that
[14:00] <acalvo> UnixDawg: check the forums, the community, google...
[14:00] <Guest76209> greenkernel1: Try browsing to it in a web browser - http://10.64.64.64/
[14:01] <zul> jdstrand: ping about bug #453453 (libvirt and pulsearmor) can that be closed now since the fix 453329 has been uploaded?
[14:01] <uvirtbot`> Launchpad bug 453453 in libvirt "libvirt sometimes hangs when using pulseaudio" [Undecided,New] https://launchpad.net/bugs/453453
[14:02] <greenkernel1> I can't connect to that address >> the connection was reset.
[14:03] <Guest76209> greenkernel1: Are you at home or on a bigger network managed by someone else?
[14:03] <greenkernel1> I'm at home.
[14:03] <uvirtbot`> New bug: #455293 in eucalyptus (main) "UEC management interface still has Eucalyptus as title" [Wishlist,Triaged] https://launchpad.net/bugs/455293
[14:03] <greenkernel1> but, I'm using a mobile broadband.
[14:03] <UnixDawg> I have been I am not finding them anywehere
[14:04] <UnixDawg> I have spent 4 days searching for answers and very little true help
[14:04] <Guest76209> greenkernel1: Is it Telstra (Aus)?
[14:04] <UnixDawg> with these php extensions
[14:04] <greenkernel1> no .. it's StarHub (Singapore).
[14:04] <Bilge> pmatulis: I don't think it has any init.d script > http://packages.ubuntu.com/hardy/amd64/znc/filelist
[14:05] <VK7HSE> is there any love for Bug 439407
[14:05] <uvirtbot`> Launchpad bug 439407 in zlib "Abort class-pclzip.php : Missing zlib extensions" [Undecided,Confirmed] https://launchpad.net/bugs/439407
[14:05] <Guest76209> greenkernel1: OK, well with Telstra the way they have it set up is all the mobile users join a massive "LAN", with a private IP address (10.x.x.x). This makes it impossible for you to run a server on that connection and have it accessible from the outside.
[14:06] <Guest76209> 10.64.64.64 would be a machine managed by your ISP
[14:06] <greenkernel1> I have inet addr: 10.xx.xx.xx.xx when using ifconfig.
[14:06] <greenkernel1> does that mean I can't run a server?
[14:07] <Guest76209> That doesn't mean anything
[14:09] <pmatulis> Bilge: yeah, i was looking at the same
[14:09] <pmatulis> Bilge: interesting problem, let me know how you make out
[14:09] <greenkernel1> some people said 10.xx.xx.xx is a private network and usually behind a NAT. that machine could not run as a server. is it true?
[14:09] <UnixDawg> I have search the repos and the full system for the extensions
[14:09] <UnixDawg> they are not installed and they are not in the repos
[14:09] <acalvo> UnixDawg: post in the forums, mailing lists
[14:10] <UnixDawg> so with ot them it makes ubuntu useless for our project
[14:10] <Guest76209> 10.64.64.64 could be an externally accessible server, but your computer will not be thanks to your ISP's configuration
[14:10] <VK7HSE> greenkernel1: as long as you set up port forwarding in your router it will work...
[14:11] <Guest76209> He doesn't have a router - it's wireless
[14:11] <greenkernel1> yes .. i have no router.
[14:11] <VK7HSE> Oh! ok...
[14:11] <VK7HSE> forget I spoke! ;)
[14:11] <Guest76209> You'd need a land connection with a static IP
[14:12] <greenkernel1> is there any solution to switch my current network to be a static one?
[14:12] <Guest76209> No
[14:12] <VK7HSE> Guest76209: or a DynDNS (or similar service) to forward your IP ...
[14:12] <Bilge> pmatulis: it's not a problem, I've written init.d scripts before, I just thought I'd save myself some time, but all I did was waste it instead
[14:13] <Guest76209> VK7HSE: Won't work, he's only joining a bigger LAN which is using a 10.x.x.x gateway, which is managed by his ISP
[14:13] <pmatulis> Bilge: by problem I meant finding the script online or otherwise
[14:13] <greenkernel1> That means I can't run a server with my current network?
[14:13] <Bilge> pmatulis: I assume it doesn't exist
[14:14] <Guest76209> greenkernel1: Correct
[14:14] <Bilge> pmatulis: http://doomclaw.de/index.php/2009/08/18/run-znc-at-boot-with-an-init-script/
[14:14] <VK7HSE> Guest76209: I was referring to your comment "You'd need a land connection with a static IP"
[14:14] <Guest76209> Oh
[14:15] <VK7HSE> I don't have a fixed IP (from ISP) but I use DynDNS instead! ;)
[14:16] <greenkernel1> So, no one is able to connect to me directly?
[14:16] <Guest76209> I manage a company's network as part of my work... and their ISP stopped their landline ADSL connection and said they couldn't supply it any more. They told them they can run servers on their wireless product (3 year contract), which they can't. I had to move their mailserver offsite and now I can't even remotely managed their network.
[14:17] <Guest76209> Correct
[14:17] <aubre> are there daily ISO updates? where can I get them?
[14:17] <VK7HSE> Guest76209: was that a Telstra suggestion?
[14:17] <Guest76209> Yes
[14:18] <VK7HSE> Gahh! typical... maybe have a look at Internode they are very good for business plans...
[14:18] <greenkernel1> can you connect to me directly if you know my IP address?
[14:18] <Guest76209> Their ADSL started dropping out. They reconnected it multiple times but it kept dropping out almost immediately. Then they said they can't do it, despite them having used it for years before.
[14:18] <pmatulis> aubre: for karmic?
[14:19] <Guest76209> greenkernel1: No
[14:19] <aubre> pmatulis: yes, I was wanting to test the latest UEC CD install
[14:19] <pmatulis> aubre: http://cdimage.ubuntu.com/daily/current/
[14:19] <aubre> grazi
[14:19] <Guest76209> Internode wouldn't work either. They're right on the edge of the exchange reach
[14:19] <greenkernel1> okay .. you can't connect to me .. but, can I connect to you if I know your IP address?
[14:20] <Guest76209> Yep
[14:20] <greenkernel1> ham .. I can connect to you .. but why can't you connect to me?
[14:21] <Guest76209> Because when you connect out your ISP's router remembers that you sent a request. When it gets a response for that request it sends it back to you.
[14:21] <Guest76209> If I try to connect to you, I reach your ISP's router and it doesn't know who I actually want to connect to, so it won't work that way.
[14:23] <greenkernel1> you have my IP address to connect with, right?
[14:23] <Guest76209> No
[14:24] <Guest76209> If you went to www.whatismyip.com it would tell you the external IP of your ISP's router
[14:24] <Guest76209> Another user on your ISP's mobile service would get the same IP
[14:25] <Guest76209> And your computer's IP is a private address (starts with 10.x.x.x), so I can't connect to it
[14:25] <greenkernel1> so how does our ISP's router know what the IP address of my machine is?
[14:26] <Guest76209> You connect to it when you turn on your wireless device
[14:26] <greenkernel1> can you connect to me using that device's address?
[14:26] <Guest76209> No, it's a private address
[14:27] <greenkernel1> private address means one way traffic .. outgoing only?
[14:28] <Guest76209> Sort of
[14:28] <Guest76209> There are a few IP ranges which are officially private addresses and shouldn't be assigned publically
[14:28] <Guest76209> One is the 10.x.x.x range
[14:28] <Guest76209> My own network could use a 10.x.x.x range
[14:29] <Guest76209> And if your public IP was 10.1.2.3, how would my computer know whether I mean 10.1.2.3 in my own network, or the public address?
[14:30] <Guest76209> Answer: it wouldn't, which is why private addresses aren't allowed to be assigned publically
[14:31] <greenkernel1> can you assign your machine that 10.1.2.3 is not your network but, another private network?
[14:31] <Guest76209> No
[14:32] <Guest76209> Would be nice if you could though
[14:32] <nijaba> ttx: soren: still get "EC2 responded with an error for DescribeInstances" in ElasticFox connecting to a 1.6 clc...
[14:33] <ttx> nijaba: euca version ?
[14:33] <Guest76209> To connect to a specific machine in a different network you'd connect to their external address (router) on a port, say port 80 which is web traffic
[14:33] <Guest76209> Then your router forwards port 80 to a machine in the network, which is configured in the router
[14:34] <nijaba> ttx: 1.6~bzr931-0ubuntu3
[14:34] <Guest76209> But your ISP manages their router and they won't configure a port forward just for one user
[14:34] <ttx> nijaba: that one should be alright
[14:35] <ttx> nijaba: Neil Soman upstream has been testing this, maybe he could give some insight
[14:35] <greenkernel1> okay ... so, there is no way to connect to me directly?
[14:35] <nijaba> ttx: is he in #eucalyptus?  what nick?
[14:36] <ttx> nijaba: no clue, you can use email though
[14:36] <Guest76209> greenkernel1: Correct, unless you got a different type of internet connection such as landline ADSL
[14:36] <Guest76209> I have to go to bed
[14:36] <Guest76209> Bye
[14:37] <nijaba> ttx: ok, I will
[14:37] <greenkernel1> bye Guest76209
[14:37] <greenkernel1> have a good dream.
[14:37] <greenkernel1> and good night.
[14:37] <greenkernel1> thanks a lot for your help.
[14:37] <Guest76209> Oh yes, I dream about internet routing and IP addressing
[14:37] <greenkernel1> ^^
[14:37] <uvirtbot`> greenkernel1: Error: "^" is not a valid command.
[14:41] <aubre> I think the latest version is 1.6~bzr931-0ubuntu5
[14:47] <uvirtbot`> New bug: #454566 in chkrootkit (main) "False positive for SucKit" [Low,Incomplete] https://launchpad.net/bugs/454566
[14:50] <jdstrand> zul: 453453 is not fixed. apparmor didn't cause the problem. fixing #453329 only makes #453453 noticeable. I talked to kirkland about it at length on friday. he said he was able to reproduce it, though I don't recall how
[14:50] <zul> k
[14:51] <zul> so im gonig to set it to confirmed
[14:51] <jdstrand> zul: that sounds fine. I invalidated the apparmor task
[14:51] <zul> thanks for the heads up
[14:51] <jdstrand> np
[15:06] <ttx> kirkland: Depending on "ecj" instead of "libecj-java" was introducing all GCJ on the server CD, extra 25Mb. I uploaded a new eucalyptus to fix that
[15:07] <ttx> kirkland: also the release team asked me to fix urgently the transitional packages in qemu-kvm (that also resulted in breaking UEC Node install completely)
[15:07] <ttx> kirkland: bothe fixes are in, new ISo spin should arrive in the next hour
[15:08] <kirkland> ttx: thanks so much
[15:08] <ttx> kirkland: hopefully everything RC should be fixed in that one -- until the next issue is found :)
[15:10] <ttx> kirkland: when you'll have caught up with email, I'd welcome a short session on euca bugs with you, to validate the RC/targets-of-opportunity nominations we have so far.
[15:11] <kirkland> ttx: sounds good, give me a few minutes to catch up on email, and make some coffee
[15:11] <ttx> kirkland: i.e. what we need to solve, and what we could solve before release
[15:11] <Omahn> Any server devs free to sponsor me on bug 341256?
[15:11] <uvirtbot`> Launchpad bug 341256 in unattended-upgrades "RFE: unattended-upgrades should identify host." [Medium,Fix released] https://launchpad.net/bugs/341256
[15:16] <ttx> Omahn: I can try to do it tomorrow if noone else gets to it before
[15:17] <Omahn> ttx: That would be great, thanks. I have an evening ahead planned on reading about the MOTU processes and structure so I might have a better idea tomorrow of the procedure anyway :-)
[15:32] <nijaba> ttx: are we aware that the cloud controller services do not restart correctly after a reboot?
[15:32] <ttx> nijaba: define "do not restart correctly" ?
[15:33] <nijaba> ttx: got error trying to connect to the admin interface, add to manually restart the services for it to work
[15:33] <ttx> nijaba: bug 444352 ?
[15:33] <uvirtbot`> Launchpad bug 444352 in eucalyptus "DB deadlock on reboot prevents UEC from working, temporarily" [Low,Triaged] https://launchpad.net/bugs/444352
[15:33] <ttx> nijaba: please look for DEADLOCK / Resetting hard messages in cloud-output.log to confirm
[15:34] <nijaba> ttx: this is what I am doing
[15:34] <ttx> nijaba: if you reproduce, try to connect 6 times in a row and see if that persists
[15:35] <ttx> nijaba: and if that's the same issue, I'd raise prri on this bug (so far I was the only oen to reproduce it)
[15:35] <nijaba> ttx: yep, that's the case.  Trying to reprodice
[15:35] <ttx> s/prri/rpio
[15:35] <ttx> prio
[15:35] <ttx> grmbl
[15:35] <nijaba> yep
[15:36]  * ttx shouldn't try to type on two keyboards at the same time
[15:37] <ttx> nijaba: please comment on that bug if it looks the same.
[15:37] <nijaba> ttx: grr... can't reproduce
[15:37] <nijaba> will try again
[15:38] <ttx> nijaba: did you get those DEADLOCK things in your logs ?
[15:38] <nijaba> ttx: yep I did
[15:38] <nijaba> ttx: was rebooting after an apt-get upgrade though
[15:38] <ttx> nijaba: even if you can't reproduce, please comment on the bug, so that I don't look like the only idiot with that bug.
[15:39] <nijaba> ttx: nobody would ever consider you are an idiot!!
[15:39] <ttx> nijaba: yeah, right...
[15:43] <kirkland> ttx: okay, bug scrub?
[15:43] <ttx> kirkland: yay
[15:43] <ttx> kirkland: at that point there is no RC-targeted bugs for eucalyptus (release blockers), and a few karmic nominations (targets of opportunity in those frozen times): bug 453456, bug 444352 and bug 455293
[15:43] <uvirtbot`> Launchpad bug 453456 in eucalyptus "excessive logs in /var/log/eucalyptus" [Low,Triaged] https://launchpad.net/bugs/453456
[15:43] <uvirtbot`> Launchpad bug 444352 in eucalyptus "DB deadlock on reboot prevents UEC from working, temporarily" [Low,Triaged] https://launchpad.net/bugs/444352
[15:43] <uvirtbot`> Launchpad bug 455293 in eucalyptus "UEC management interface still has Eucalyptus as title" [Wishlist,Triaged] https://launchpad.net/bugs/455293
[15:43]  * kirkland opens
[15:43] <ttx> I was about to remove the deadlock one when nijaba reproduced it
[15:43] <ttx> so I'm not sure about it
[15:43] <ttx> I think at that point it would be a SRU
[15:44] <kirkland> ttx: yeah, i think we should leave that one as a placeholder for now
[15:44] <kirkland> ttx: i think there is probably something wrong
[15:44] <kirkland> ttx: it's probably just complex to reproduce
[15:44] <ttx> Removing as a karmic target then
[15:44] <kirkland> ttx: we may get users reporting it
[15:44] <kirkland> ttx: agreed
[15:44] <ttx> kirkland: you agree that the other two may still get fixed ?
[15:44] <kirkland> ttx: regarding logs ...
[15:45] <kirkland> ttx: nurmi says that eucalyptus rolls its own logs
[15:45] <kirkland> ttx: i didn't see anything in logrotate.d so I didn't know that
[15:45] <kirkland> ttx: he says that they definitely *won't* grow indefinitely
[15:45] <ttx> yes, they do... but they are still too noisy by default
[15:45] <kirkland> ttx: right, the noisy thing is related, and should also be fixed
[15:46] <kirkland> ttx: i don't think it's release critical
[15:46] <ttx> keeping it as a target of opportunity
[15:46] <kirkland> ttx: if there was *no* rotation, then i think it's more importatn
[15:46] <ttx> oh its definitely not release-critical
[15:46] <ttx> but we could still attempt to fix it
[15:47] <ttx> since a saner conffile default is good to have out of the box
[15:47] <kirkland> ttx: okay, i agree with that
[15:47] <ttx> bug 455293 is harmless and we may have another branding round anyway, so keeping it as a target of opportunity as well
[15:47] <uvirtbot`> Launchpad bug 455293 in eucalyptus "UEC management interface still has Eucalyptus as title" [Wishlist,Triaged] https://launchpad.net/bugs/455293
[15:49] <ttx> kirkland: anything in https://bugs.launchpad.net/ubuntu/+source/eucalyptus that we might still try to fix before release ?
[15:49] <ttx> I think anything else is quite risky at that stage.
[15:50] <kirkland> ttx: yeah, i was hoping friday's uploads were the *last*
[15:50] <ttx> kirkland: sorry :)
[15:50] <kirkland> ttx: i don't see anything as critical as the tmpfile or the network/ip address issues we solved on friday
[15:50] <kirkland> ttx: the branding stuff is perhaps 'critical' from a business perspective
[15:51] <kirkland> ttx: at least that is low risk
[15:51] <kirkland> ttx: as for the ecj issue, my apologies
[15:51] <kirkland> ttx: at some point during my testing i realized that i only needed libecj-java
[15:51] <kirkland> ttx: and i built a test package with that, but forgot to commit that into the source tee
[15:51] <kirkland> tree
[15:51] <uvirtbot`> New bug: #361169 in xinetd (main) "[jaunty] xinetd upgrade invokes itox incorrectly" [Low,Won't fix] https://launchpad.net/bugs/361169
[15:52] <kirkland> ttx: my fault, very sorry
[15:52] <ttx> kirkland: just test the heck out of 20091019.1 when it's out
[15:52] <ttx> kirkland: since I couldn't do any testing today
[15:52] <kirkland> ttx: okay
[15:52] <kirkland> ttx: yeah, our roles have reversed :-)
[15:53] <kirkland> ttx: and thanks for the qemu-kvm fix; i just took that one from james_w
[15:53] <ttx> kirkland: ok, so nothing else in that list that we might /try/ to fix before release ?
[15:54] <ttx> kirkland: yes, not your fault :)
[15:54] <nijaba> kirkland: just added my comment and log to the deadlock bug
[15:54] <kirkland> ttx: i'm looking...
[15:54] <ttx> james_w can make errors, apparently. I used to think he was immune to that.
[15:54] <kirkland> ttx: the dpkg-stateoverrides one is clearly wrong, easy to fix, high risk of regression; defer to lucid
[15:55] <uvirtbot`> New bug: #359473 in vm-builder (universe) "vmbuilder crashed with NameError in deploy()" [Low,Incomplete] https://launchpad.net/bugs/359473
[15:55] <uvirtbot`> New bug: #378639 in vm-builder (universe) "firstlogin script can't be executed because the permission of /root directory is 0700 in the virtual machine created by python-vm-builder " [Low,Incomplete] https://launchpad.net/bugs/378639
[15:55] <kirkland> ttx: we do need an NC upstart script; but also should defer that to lucid
[15:55] <ttx> kirkland: if they are not nominated to karmic, they are lucid targets now
[15:56] <kirkland> ttx: okay
[15:56] <kirkland> ttx: the one-NIC bug kinda sucks
[15:56] <kirkland> ttx: but i don't think we should touch the installer right now
[15:56] <kirkland> ttx: i had a look at fixing that one; it's harder than it sounds
[15:56] <ttx> kirkland: definitely *not* a good idea :)
[15:57] <ttx> kirkland: on another subject, you have bug 443080 nominated to karmic
[15:57] <uvirtbot`> Launchpad bug 443080 in mountall "ecryptfs mount does not support -f (fake mount)" [Medium,Fix committed] https://launchpad.net/bugs/443080
[15:57] <ttx> is it still something you expect to land in karmic ?
[15:57] <kirkland> ttx: i'm not fixing that for karmic
[15:57] <ttx> ok, removing nomination then
[15:58] <kirkland> ttx: it's marked won't-fix
[15:58] <ttx> ah
[15:58] <ttx> for some reason it's appearing in your buglist
[15:58] <kirkland> ttx: how do i un-nominate?
[15:58] <ttx> you did it right, it's the mountall part that fucks the buglist up
[15:59] <ttx> kirkland: see http://qa.ubuntu.com/reports/team-assigned/canonical-server-assigned-bug-tasks.html
[15:59] <kirkland> ttx: yeah, i'll fix that upstream, at some point
[15:59] <kirkland> ttx: before lucid, but not right now
[15:59] <ttx> kirkland: it appears in that list
[15:59] <ttx> as Wishlist/Triaged/KK
[16:00] <ttx> probably a bug in the report.
[16:01] <ttx> kirkland: I opened bug 455411, and its still a target of opportunity for karmic
[16:01] <uvirtbot`> Launchpad bug 455411 in qemu-kvm "Conffiles from kvm are left around on upgrade from Jaunty" [Low,Triaged] https://launchpad.net/bugs/455411
[16:01] <ttx> kirkland: its a leftover of the transition packages migration as well
[16:02] <ttx> kirkland: slangasek asked me to fix urgently the RC portion of it, so I left that part out
[16:02] <kirkland> ttx: okay
[16:02] <kirkland> ttx: thanks for the tracker
[16:02] <ttx> kirkland: just so you know it's still a target of opportunity
[16:02] <kirkland> ttx: right
[16:02] <ttx> definitely not release critical, since its harmless
[16:03] <kirkland> ttx: i spent the second half of friday trying to fix sound in qemu-kvm
[16:03] <ttx> 20091019.1 is out
[16:03] <uvirtbot`> New bug: #408298 in postfix (main) "package postfix 2.5.5-1.1 failed to install/upgrade: le sous-processus post-installation script a retourné une erreur de sortie d'état 1" [Low,Incomplete] https://launchpad.net/bugs/408298
[16:03] <kirkland> ttx: syncing
[16:04] <ttx> kirkland: doublecheck MD5sums since sometimes the current/ link is outdated
[16:05] <ttx> I should have time to do basic testing before I stop for the day.
[16:12] <lamont> zul: 408298 is, near as I've been able to figure out, someone nuking /etc/postfix and then upgrading postfix and wondering why it doesn't work
[16:12] <zul> lamont: ok *sigh* ;)
[16:12] <uvirtbot`> New bug: #327348 in kvm (universe) "keep losing ability to type in guest" [High,Fix released] https://launchpad.net/bugs/327348
[16:18] <uvirtbot`> New bug: #321049 in ntp (main) "clock skew causes ntp failure on boot" [Low,Incomplete] https://launchpad.net/bugs/321049
[16:18] <nijaba> Congratulations, server team, new Gartner linux market analysis from Gartner lists us among the leading distro!
[16:19] <nijaba> Geoges Weiss to present this Gartner Datacenter Conference at the start of December
[16:21] <zul> nifty
[16:22] <ttx> nijaba: \o/
[16:48] <LordBurrito> G'day.  Got some Dell 1750 rack-mount servers with PERC 4/Di RAID controllers.  Any show-stoppers on this platform with 8.04 LTS?
[16:51] <uvirtbot`> New bug: #386619 in multipath-tools (main) "*** stack smashing detected ***: /sbin/mpath_prio_hds_modular terminated" [Low,Invalid] https://launchpad.net/bugs/386619
[16:54] <addisonj> most likely no, use google to find out though
[16:54] <addisonj> only possible problem might be the raid controllers
[16:54] <iBLuCKy>  [HowTo] Fix Ubuntu Jaunty (9.04) Intel Graphics Problem and pules audio problem when i install graphic drivers?? please help
[16:55] <iBLuCKy> i tryed the other room and no one is helping me wtf
[16:55] <addisonj> thats because no one know how to help you
[16:56] <addisonj> go back there and specify what your actual problem is
[16:56] <addisonj> the more info, the more likely you are to get help
[16:56] <iBLuCKy> thats the thing i new to linux
[16:56] <iBLuCKy> i dont know how to say it
[16:57] <iBLuCKy> everything is fine untill i install the graphics
[16:57] <iBLuCKy> then i reboot
[16:57] <iBLuCKy> and it make me run in low graphic
[16:57] <LordBurrito> addisonj: I'll just have to give it a go.  It's the MegaRAID controller, so it *shouldn't* be a problem, I'm told.
[16:57] <addisonj> most likely no, ubuntu has lots of raid support drivers
[16:58] <SyL> am I able to use a 2.6.28-6-server kernel in karmic?
[16:58] <addisonj> and since it is hardware raid you should be able to use normal install, software raids need the alternate install disc
[16:59] <LordBurrito> *nod*
[16:59] <LordBurrito> addisonj: Tnx
[17:05] <ttx> kirkland, nijaba: just hit bug 452556 in ISO testing
[17:05] <uvirtbot`> Launchpad bug 452556 in eucalyptus "euca-authorize default failing" [Medium,Confirmed] https://launchpad.net/bugs/452556
[17:05] <nijaba> ttx: just do a run-instance, it will work right after !!
[17:06] <ttx> yes, I saw that
[17:06]  * soren pauses for dinner
[17:07] <nijaba> was there a known install problem for NC on 20091019?  I have a journalist that is reporting an issue
[17:07]  * nijaba does not know how to view history on iso.qa.ubuntu.com
[17:09] <ttx> nijaba: yes
[17:09] <ttx> 20090919.1 is the one to use now
[17:09] <nijaba> ttx: ok, thanks
[17:09] <ttx> nijaba: bug 455114 preventing Node install to succeed
[17:09] <uvirtbot`> Launchpad bug 455114 in qemu-kvm "builds uninstallable package 'kvm'" [Critical,Fix released] https://launchpad.net/bugs/455114
[17:10] <ttx> kirkland: amd64 tests passed / added my results to the ISO tracker
[17:21] <ruben23> hi
[17:28] <_BuBU> Hi there
[17:29] <_BuBU>  I've a problems with number of group apache users belong under ubuntu... don't know if that's an issue for apache or for ubuntu
[17:29] <_BuBU>  cat /proc/sys/kernel/ngroups_max
[17:29] <_BuBU> 65536
[17:29] <_BuBU> and I've several user/groups, basically 1 per virtualhost
[17:29] <_BuBU> and apache belongs to all that groups
[17:29] <_BuBU> rights on html files are user_xxxx:grp_xxxx and 0640 same for all images (jpg, ...)
[17:29] <_BuBU> but seems first users sites are working fine.. and not after a certain number
[17:29] <_BuBU> and if I update the group to have only apache to belongs to that non working user... it is working ! :(
[17:29] <_BuBU> any idea ?
[18:16] <spydmobile> Hi folks I am still trying to get my UEC operational, the backstory is here: http://paste.ubuntu.com/296974/ Can anyone point me to a next logical step?
[18:25] <Family_> Is there anywhere I can study/learn the use of ubuntu server, other than ubuntu server guide
[18:25] <Family_> or any linux manuals that would help understand using linux
[18:26] <Family_> help in ubuntu server usage
[18:26] <pmatulis> Family_: books (there is an official ubuntu server book) and/or online courses (offered by canonical) are options
[18:28] <Family_> how much for the book, where from and do the courses cost
[18:29] <pmatulis> Family_: courses: http://shop.canonical.com/index.php?cPath=21&osCsid=034c980130591292bc3c0a2dcc42aa2f
[18:30] <spydmobile> Family_: More here: http://www.ubuntu.com/training
[18:35] <Family_> Theres a couple of books about ubuntu sever
[18:36] <Family_> which would be the best for me with no experience of linux
[18:37] <MagicFab> Family_,    http://search.barnesandnoble.com/Official-Ubuntu-Server-Book/Kyle-Rankin/e/9780137036035
[18:38] <Family_> u sure?
[18:40] <Family_> what about http://www.amazon.com/Beginning-Ubuntu-LTS-Server-Administration/dp/1430210826/ref=sr_1_5?ie=UTF8&s=books&qid=1255973777&sr=1-5
[18:41] <pmatulis> Family_: i wouldn't go for a server-specific book if you're just starting out with Linux
[18:42] <Family_> yes but i must use a server specific OS
[18:42] <Family_> so isint it best to get a book that is also one
[18:42] <spydmobile> Family_: If you are targeting Ubu Server, dont settle for one book, get a few, these mentioned above for sure....
[18:43] <pmatulis> Family_: well, learn about the basics of Linux and then move on to server-specific material
[18:43] <spydmobile> Family_: also if you are new to it all, then perhaps a third edition on server 8.xx is a better approach than a first edition on 9.xx
[18:45] <Family_> but im using 9.04, does things for 8.xx work for 9.04
[18:45] <MagicFab> pmatulis, Family_ the book I cited starts from zero, literally
[18:45] <MagicFab> check the ToC or sample chapters.
[18:46] <pmatulis> MagicFab: thank you
[18:51] <nijaba> kirkland: regarding your email, https://help.ubuntu.com/community/UEC/PackageInstall would be the closest we have to that, but would need quite a bit of rewriting to cover multi-cluster
[20:20] <holddigga> Has anyone here successfully used KVM with a VMWare disk image? I tried to boot one up and it just pegged the host CPU at 100% and never booted (I left it going overnight). I was using Server 9.04 (Intrepid).
[20:26] <EtienneG> mathiaz, kirkland: do you know if we still need to have an entry for "mailhost" on the CLC for the email to be sent correctly in Eucalyptus 1.6?
[20:26] <mathiaz> EtienneG: hm - no
[20:26] <mathiaz> EtienneG: technically eucalyptus tries first localhost, then mailhost
[20:27] <mathiaz> EtienneG: however postfix is installed by default now
[20:27] <mathiaz> EtienneG: so the first try with localhost should succeed
[20:27] <EtienneG> mathiaz, ok good - you had to on 1.5 (the clc was forwarding mail through an MX by the name of "mailhost")
[20:27] <EtienneG> mathiaz, on 1.5, it would not even try localhost
[20:27] <mathiaz> EtienneG: reading through the code, it should try localhost first
[20:28] <EtienneG> so just having postfix installed was not even, you had to add an entry to /etc/hosts
[20:28]  * mathiaz nods
[20:28] <EtienneG> mathiaz, that's good news (although it should really be configurable, if you ask me)
[20:28]  * mathiaz nods
[20:29] <EtienneG> mathiaz, while there, what are the tags you should use on a eucalyptus-related bug?
[20:29] <EtienneG> like, on reported against another package but related to eucalyptus
[20:30] <mathiaz> EtienneG: hm - if it's a problem with an image, use uec-images
[20:30] <mathiaz> EtienneG: if it's a problem with euclyptus
[20:30] <mathiaz> EtienneG: use eucalyptus
[20:30] <EtienneG> mathiaz, ok, thanks, reporting a wishlist bug against d-i regarding the installer integration
[20:31] <mathiaz> EtienneG: right - try to use the eucalyptus tag then
[20:35] <zul> jdstrand: thanks for testing the sru (vblade) ;)
[20:35] <EtienneG> it's weird that the cluster controller installer ask for the public IP range, but not for a private one
[20:35] <EtienneG> I guess it defaults to 172.16.0.0/16?
[20:35] <mathiaz> EtienneG: 172.19.0.0/24
[20:35] <jdstrand> zul: heh-- I spent too much time trying to get vblade/aoe working in my hardy multipurpose vm for regression testing and didn't want someone else to do the same
[20:36] <mathiaz> EtienneG: it's a debconf question - priority medium
[20:36] <mathiaz> EtienneG: so won't be shown by default during the install
[20:36] <EtienneG> mathiaz, ok, I guess the default is exotic enough not to be a problem
[20:36] <jdstrand> zul: there was an article on it in Linux Journal recently, so I also updated https://help.ubuntu.com/community/ATAOverEthernet for Hardy as well
[20:36] <zul> cool thanks
[20:37] <jdstrand> zul: though perhaps an SRU for adding /etc/udev/rules.d/60-aoe.rules to aoetools wouldn't be a bad idea. it isn't needed for jaunty, but aoe won't create /dev/etherd without it
[20:37] <jdstrand> zul: see my entry in the wiki for details
[20:37] <Qwell> So, apparently I'm supposed to be asking here..  I'm a little confused about why Karmic will be shipping with an RC version of Asterisk 1.6.2.  Anybody able to comment on/discuss that?
[20:38] <soren> Daviey: ^^
[20:39] <soren> Qwell: Daviey's the mastermind behind asterisk here :)
[20:43] <zul> jdstrand: good idea
[20:47] <EtienneG> mathiaz, kirkland: the cluster controller install target ask for the private net interface *after* the dialog about the installation being finished.  Is that a known bug, or should I report it?
[20:47] <EtienneG> (have not seen it in my bugmail, but I am not subscribed to d-i)
[20:48] <mathiaz> EtienneG: *after* ?
[20:48] <mathiaz> EtienneG: like just before rebooting?
[20:48] <mathiaz> EtienneG: are you installing from an iso?
[20:48] <EtienneG> mathiaz, yes.  You answer the question, and then it goes on to reboot.  very weird.
[20:49] <EtienneG> yes, from today's ISO
[20:49] <EtienneG> mathiaz, I understand this is an unreported bug then ...
[20:49] <mathiaz> EtienneG: yes - I think so
[20:49] <EtienneG> lemme try it again in a vm
[20:50] <aubre> When I use the latest UEC iso on the cloud controller it configures the private interface to use dhcp - is it expecting a dhcp server of some kind to be running on the private network?
[20:50] <aubre> or do I need to manually set a static ip for it? I ask because I am trying to follow http://help.ubuntu.com/community/UEC
[20:50] <aubre> and this issue is not addressed
[20:50] <mathiaz> EtienneG: hm - I think I know why
[20:51] <mathiaz> EtienneG: do you have 2 network interfaces on the system right?
[20:51] <EtienneG> mathiaz, of course
[20:51] <aubre> EtienneG: mathiaz : I noticed this as well
[20:52] <mathiaz> aubre: how many physical network interfaces do you have on your CC?
[20:52] <aubre> mathiaz: 6 but I am only using 2 atm
[20:52] <mathiaz> aubre: right
[20:53] <mathiaz> aubre: well - the default for the installer is to setup a dhcp interface
[20:53] <mathiaz> aubre: independently of UEC
[20:53] <aubre> mathiaz: now I know I can set up my CC to use static private IP addressing, and I can set up static IPs on my nodes, but it needs to be clearer in the documentation
[20:53] <EtienneG> aubre, the debian installer always configure only the primary interface.  with the UEC installation target, the primary is considered to be the public facing one, and the interfaces are left unconfigured.  I *think* this is of no consequence (the CC will run a dhcp server on that interface just for instances, using the eucalyptus private net range)
[20:54] <mathiaz> aubre: right.
[20:54] <aubre> mathiaz: can we make it clearer in the docs/online install ?
[20:54] <Qwell> Daviey: Please nick highlight me when you're around, if you could.  I'm going to try to ignore this channel.
[20:55] <aubre> mathiaz: or give the user the option to set up a static address on eth1 for example?
[20:55] <Qwell> (until then, anyways)
[20:55] <mathiaz> aubre: at this time of the karmic cycle documentation is the only option we have to work on
[20:55] <aubre> mathiaz: ok
[20:57] <mathiaz> aubre: hm - well - the UEC install will actually always set the private interface to run on dhcp if the private interface is different from the public interface
[20:58] <aubre> mathiaz: also want to be sure to edit /etc/sysctl.conf and uncomment net.ipv4.ip_forward=1 too right? or is that not necessary?
[20:58] <aubre> mathiaz: ok, maybe we just need to make sure that if that isn't going to work for some people that it shows up in the documentation
[20:58] <mathiaz> EtienneG: could you file a bug about your issue?
[20:59] <mathiaz> aubre: no - it's not needed anymore
[20:59] <aubre> mathiaz: good
[20:59] <mathiaz> aubre: ipv4 forward is automatically taken care of when the CC starts
[20:59] <aubre> mathiaz: thanks for the help
[21:01] <EtienneG> mathiaz, yep, reproducing it first to confirm
[21:01] <mathiaz> EtienneG: right - I see why it does so
[21:01] <EtienneG> mathiaz, I am starting to doubt the wisdom of defaulting to MANAGED-NOVLAN
[21:02] <mathiaz> EtienneG: I'm not sure however why the prompt comes *after* the message to reboot is given
[21:02] <EtienneG> SYSTEM was so much more straightforward ...
[21:02] <mathiaz> EtienneG: I think the main issue with MANAGED-NOVLAN is to get the network environment setup properly
[21:02] <EtienneG> mathiaz, I trust this bug will be shallow to cjwatson
[21:03] <EtienneG> mathiaz, indeed ... but moreover, there are external dependencies (not running DHCP on the "private" lan, etc)
[21:03] <EtienneG> mathiaz, this seems like a recipe for disaster
[21:03] <EtienneG> mathiaz, and there is a couple of "architectural" decision to be made that really do not have a good one-size-fit-all default
[21:03] <mathiaz> EtienneG: well - the default configuration assumes that the private and the public interfaces are the same
[21:04] <mathiaz> EtienneG: right - the more experience we get know the better decision we'll be able to make for Lucid :)
[21:05] <EtienneG> mathiaz, you're so Zen!
[21:05] <mathiaz> EtienneG: I guess that what happens when you turn 30 ;)
[21:06] <EtienneG> mathiaz, was it your birthday recently?
[21:06] <mathiaz> EtienneG: yeah - like today :p
[21:07] <EtienneG> QUICK SOMEONE, SEND ONE OF THOSE CHEESY ANSI-COLORED BIRTHDAY CAKE!!!11
[21:07] <EtienneG> mathiaz, hey, happy birthday!
[21:07] <EtienneG> you are not in Mtl today, right?
[21:07] <mathiaz> EtienneG: I am - ask cr3
[21:08] <EtienneG> damn, i have something tonight - beer would have been good otherwise
[21:14] <EtienneG> mathiaz, I am faberglasted.  Installing a node, after the partition table change prompt, there are no other prompts (not even the last one, "installation finished, remove cd, blah, blah").  It just goes on and then reboot.
[21:14] <EtienneG> is that intended?
[21:15] <mathiaz> EtienneG: yes
[21:15] <mathiaz> EtienneG: the prompt for the private interface only goes on the CC
[21:15] <mathiaz> EtienneG: the assumption (I guess here) is that the NC are supposed to be on their private LAN anyway
[21:15] <mathiaz> EtienneG: like a CC
[21:15] <mathiaz> EtienneG: hm - like a cluseter
[21:16] <mathiaz> EtienneG: *cluster*
[21:16] <EtienneG> mathiaz, ok, I see, but not even "installation done, rebooting" prompt?  I mean, that's kind of unsettling
[21:16] <mathiaz> EtienneG: well - that's experience we wanted to have IIRC
[21:16] <mathiaz> EtienneG: as few prompts are possible
[21:17] <EtienneG> mathiaz, hurmm ok
[21:17] <mathiaz> EtienneG: that being said it may be a bug/whishlist to have the 'installation done, rebooting' prompt
[21:18] <mathiaz> EtienneG: file a bug :)
[21:18] <Daviey> Qwell: ping?
[21:18] <Qwell> Daviey: one sec
[21:18] <EtienneG> mathiaz, I got a few to file, and little time.  I will skip that one.  Surely someone else will report it
[21:19] <mathiaz> EtienneG: well if everyone would think that way, we'd have zero bugs in LP :)
[21:19] <EtienneG> mathiaz, imagine, a bug-free Linux distro!
[21:19] <Qwell> Daviey: I wanted to figure out why Karmic will be shipping with an RC of Asterisk
[21:20] <mathiaz> EtienneG: well - I would probably loose my job then :D
[21:21] <Daviey> Qwell: Well i'm really hoping it hits stable before release.  If it doesn't, i'm hoping to propose an SRU for when it is.
[21:21] <Qwell> Daviey: It won't
[21:21] <EtienneG> mathiaz, that is just one step removed from saying you are purposefull yintroducing bugs for the sake of job security - I would be careful!
[21:21] <EtienneG> :)
[21:22] <Daviey> Qwell: well that was the *hope*.. Do you have any idea when 1.6.2 will be stable?
[21:22] <Qwell> When it's ready. :)
[21:23] <Qwell> I *really* don't want to have to deal with the bug reports that will come to the bug tracker if a major distro ships with an RC.
[21:23] <EtienneG> bad news is that karmic will be shipping with an RC of Asterisk.  Good news is that the Asterisk RC will get tested.  Oh well!
[21:24] <Daviey> it is my understanding that most people who use Ubuntu+Asterisk actually build from source.. which i dislike :(
[21:24] <Qwell> sure, it's great for ubuntu.  They don't have to deal with the flurry of reports of the same (known and fixed...) issues.
[21:24] <Qwell> We, however, will.  That doesn't exactly please me. :)
[21:24] <mathiaz> Qwell: if the issues are fixed, they can be fixed in Stable Release Updates
[21:25] <Daviey> Qwell: The installation does strictly state that reports should be reported to Ubuntu fwiw.
[21:25] <mathiaz> Qwell: https://wiki.ubuntu.com/StableReleaseUpdates
[21:25] <Qwell> Daviey: in theory that's great, but in practice...  they go to upstream.
[21:26] <Daviey> Qwell: What do you feel is the timeframe for 1.6.2 being tagged stable?
[21:26] <Qwell> Daviey: when it's ready...  I can't say anything beyond that, because we simply don't (and can't) know
[21:27] <erichammond> ttx, smoser: I regularly browse the EC2 meta-data tree manually to see what values are available.  I would miss this ability if I were using Eucalyptus.
[21:28] <Daviey> Qwell: Well the options are, either ship rc* and propose an SRU when it hits stable / or revert to an earlier stable release.
[21:29] <Daviey> It is worth noting, that the -server team feels that non LTS versions are "technology previews" for an LTS..
[21:29] <Daviey> So many suggest that non-lts releases aren't used for critical tasks
[21:30] <Qwell> the -server team doesn't have to handle all the bug reports upstream gets. :)
[21:30] <Daviey> All my PBX's are nstill Hardy.
[21:30] <Daviey> Qwell: What timezone are you on btw?
[21:31] <Qwell> central US
[21:33] <Daviey> Qwell: The banner is customised to be:
[21:33] <Daviey> http://bazaar.launchpad.net/~ubuntu-voip/ubuntu-voip/asterisk/annotate/head%3A/debian/patches/ubuntu-banner
[21:34] <Daviey> So, i ereally was trying to shield you from bugs..
[21:34] <Daviey> The hope is to triage them in Ubuntu, and forward them upstream when it is confirmed, and the correct information is provided.
[21:35] <Daviey> Qwell: I'm keen to try and get a good working ethos with upstream on this.
[21:36] <Daviey> I should really dash, I'm on UTC+1 and family commitments mean i'm getting my ear bending at the moment.
[21:36] <Daviey> So if we can thrash this out tommorrow, I would really appreicate it.
[21:37] <Qwell> Daviey: that's fine..  you can msg me or I'll be in #asterisk-dev
[21:38] <smoser> erichammond, but do you know of any tools that would expect that index on / or /<api-version>
[21:39] <erichammond> smoser: Yes, wget, curl, GET, etc.
[21:39] <erichammond> :)
[21:40] <erichammond> I'm not saying it should be more important than "Low", but I'm agreeing that API compatible should be compatible.
[21:40] <aubre> whee after setting up static IPs, nodes were detected no problem by euca_conf
[21:42] <aubre> I like the new services page :)
[21:46] <aubre> sweet, the new setup was super easy, except that I had to set up my static IPs on my private network. Good work folks :)
[21:46] <aubre> I'll be testing EBS/SC support tomorrow
[21:47] <jjohansen> smoser: I have uploaded some new test kernels using just minimal config changes Bug #428692
[21:47] <uvirtbot`> Launchpad bug 428692 in linux-ec2 "ec2 kernel needs CONFIG_BLK_DEV_LOOP=y and other config changes" [Medium,Triaged] https://launchpad.net/bugs/428692
[21:48] <aubre> I'll wait till tomorrow to register images :)
[21:56] <aubre> I could ask Mark Spencer about Asterisk. He's a friend.
[21:59] <aubre> I dropped Mark an email. I'll let y'all know what he says.
[22:00] <smoser> jjohansen, thanks.
[22:02] <ruben23> root@Dbase:/usr/src/astguiclient/trunk/www# cp /usr/src/astguiclient/trunk/www/agc/ /var/www
[22:02] <ruben23> cp: omitting directory `/usr/src/astguiclient/trunk/www/agc/'
[22:04] <qman__> ruben23, -r
[22:04] <ruben23> qman__:  ive done it with -a
[22:12]  * soren calls it a day
[22:13] <ruben23> hi anyne done performance tuning of kernel to optimize ubuntu server.>?
[22:46] <uvirtbot`> New bug: #452900 in asm2 (main) "Please run testsuite during build" [Medium,Triaged] https://launchpad.net/bugs/452900
[22:46] <uvirtbot`> New bug: #452926 in c3p0 (main) "Warnings during build" [Medium,Triaged] https://launchpad.net/bugs/452926
[22:46] <uvirtbot`> New bug: #452939 in cglib2.1 (main) "Please run testsuite during build" [Medium,Triaged] https://launchpad.net/bugs/452939
[23:12] <nijaba> First version of https://help.ubuntu.com/community/UEC/ElasticFox completed. Feel free to add to it.
[23:12]  * nijaba -> bed
[23:17] <cjwatson> mathiaz: is anyone in a position to test a fix for Etienne's question ordering bug, if I provide one? It would involve some manual messing around in d-i
[23:17] <mathiaz> cjwatson: yes - I should be able to test such a fix
[23:17] <mathiaz> cjwatson: however - how important is this to get it fixed in karmic for release?
[23:18] <cjwatson> that's up to you guys I think
[23:18] <cjwatson> it's a wart
[23:18] <cjwatson> I wouldn't expect it to be a fatal wart
[23:19] <mathiaz> cjwatson: right - It seems to be cosmetic
[23:19] <mathiaz> cjwatson: it doesn't break things
[23:20] <cjwatson> well, if you decide you're interested, branch pushing up to lp:~cjwatson/eucalyptus/question-ordering at the moment
[23:21] <mathiaz> cjwatson: oh cool
[23:21] <cjwatson> ... actually, I was just writing out directions and realised that it's painful to test
[23:21] <mathiaz> cjwatson: right - a new iso needs to be respun IIUC
[23:22] <cjwatson> not actually true
[23:22] <cjwatson> but it's more fiddly than usual
[23:22] <phoenixz> Anybody here who could tell me if ubuntu can use fibreoptic cards?
[23:22] <phoenixz> To be specific, lspci gave me this : Fibre Channel: Brocade Communications Systems, Inc. 415 4Gb/815 8Gb PCIe FC HBA (rev 01)
[23:23] <cjwatson> I think what you can do is build the package, put the udeb somewhere wgettable, write a preseed file that says 'd-i preseed/early_command string wget http://url/to/eucalyptus-udeb_1.6~bzr931-0ubuntu7_i386.udeb && udpkg -i eucalyptus-udeb_1.6~bzr931-0ubuntu7_i386.udeb'
[23:23] <phoenixz> I have multiple of those and need to get them to work with powerpath
[23:23] <cjwatson> but then the tricky bit
[23:24] <cjwatson> you have to start the installer in expert mode, run through it until right before the "Select cloud installation mode" menu item, *but then* scroll down the menu and select "Download debconf preconfiguration file"
[23:24] <cjwatson> after that, you can select "Change debconf priority", set it back to high, and continue as normal
[23:24] <cjwatson> (obviously I meant 0ubuntu8 not 0ubuntu7, you get the idea)
[23:24] <mathiaz> cjwatson: right
[23:25] <mathiaz> cjwatson: thanks for the instructions - I'll paste them in teh bug
[23:25] <mathiaz> cjwatson: and will try to find hardware to test the fix
[23:25] <Kentuky> Can anybody help me with streaming my music collection?
[23:25] <cjwatson> mathiaz: if you regard it as purely cosmetic, though, perhaps we shouldn't waste time on it
[23:26] <mathiaz> cjwatson: well - I'll discuss this with kirkland and ttx
[23:26] <mathiaz> cjwatson: and we'll decide whether it's worth getting into the release
[23:26] <Kentuky> I'd really like an answer that'll allow my windows pc and linux laptop access my ubuntu server.
[23:29] <cjwatson> mathiaz: branch finished pushing
[23:29] <mathiaz> cjwatson: great - thanks!
[23:39] <uvirtbot`> New bug: #455873 in apache2 (main) "mod proxy causes duplicate query strings when nocanon option is used" [Undecided,New] https://launchpad.net/bugs/455873
[23:40] <Kentuky> anybody here use gnump3d or mpod?
[23:45] <The2morrowMan> Does Ubuntu 9.10 promise to bring anything to the non-cloud server arena over 9.04?
[23:56] <The2morrowMan> Anyone?