[00:06] ScottK, ping [00:22] kirkland: do you have a libvirt daemon running? [00:22] kirkland: or you just use a plain kvm? [00:23] mathiaz: both [00:23] root 6136 1 0 Oct13 ? 00:00:00 /usr/sbin/libvirtd -d [00:23] nobody 6213 6136 0 Oct13 ? 00:00:00 dnsmasq --keep-in-foreground --strict-order --bind-interfaces --pid-file --conf-file --listen-address 192.168.122.1 --except-interface lo --dhcp-leasefile=/var/lib/libvirt/dhcp-default.leases --dhcp-range 192.168.122.2,192.168.122.254 [00:23] kirkland: ok - so with libvirt you need to update the configuration of the guest [00:24] mathiaz: k, how so? [00:24] kirkland: http://paste.ubuntu.com/57644/ [00:24] kirkland: ^^ this is the configuration of a running guest [00:24] mathiaz: where does this go? [00:25] kirkland: the important part is when you define the interface [00:25] kirkland: [00:25] kirkland: try to use the virsh command [00:25] kirkland: virsh dumpxml guest-vm-name [00:25] kirkland: or start with [00:25] kirkland: virsh list [00:25] kirkland: to get the list of running guest. [00:26] kirkland: then dump the xml configuration to a local file [00:26] kirkland: virsh dumpxml guest-name > libvirt.xml [00:26] kirkland: edit libvirt.xml to set the network interface model to virtio [00:26] kirkland: and reload the configuration with [00:26] mathiaz: virsh list shows nothing [00:26] mathiaz: but i have two kvm's running from the command line, with kvm ..... [00:26] kirkland: virsh define guest-vm-name [00:27] kirkland: hm - try: virsh list --all [00:27] kirkland: if nothing is listed, then you don't have any vm guest defined in libvirt [00:27] mathiaz: okay, now i see a few ***ancient*** vm's i created in virt-manager, long ago [00:27] kirkland: I don't think that libvirtd knows about the kvm machines started from the command line [00:27] mathiaz: should i do this testing through virt-manager? [00:28] kirkland: you could - I tend to use virsh [00:28] kirkland: virsh is just a command line/shell front-end to libvirtd while virt-manager is the GUI front-end to libvirtd [00:29] kirkland: I don't know if you can set a virtio network interface with virt-manager [00:29] mathiaz: okay, fair enough [00:29] mathiaz: okay, so i have a vm in virsh that I can overwrite, or replace the disk or whatever [00:30] kirkland: ok - so dump the xml configuration to a local file [00:30] kirkland: virsh dumpxml vm-name > new_vm.xml [00:30] mathiaz: http://paste.ubuntu.com/57647/ === samuraip1nguin is now known as samuraipenguin [00:31] kirkland: ok - edit the xml file to add " [00:32] kirkland: to the network interface section [00:32] kirkland: below source network='default'/> [00:32] My ubuntu 7.10 server has ceassed allowing XWin (via cygwin) to connect. XWin can connect to other Linux X-Windows server. [00:33] mathiaz: done [00:33] kirkland: all right - now reload the configuration with [00:33] kirkland: virsh define new_vm.xml [00:33] more ... so not firewall issue, unless a firewall has grown in Ubuntu? [00:33] mathiaz: done [00:34] kirkland: now you can start the vm with: [00:34] kirkland: virsh start vm-name [00:34] more ... this all worked last month, and then ceased (maybe an ugrade introduced new security?) [00:34] mathiaz: okay. [00:34] mathiaz: and to connect to it? [00:34] more.. Has anyone had a similar experience, or a suggested solution? (over) [00:35] kirkland: if you can get the vncdiplay with the vncdipslay command [00:35] kirkland: virsh vncdisplay vm-name [00:36] mathiaz: nope [00:36] mathiaz: can i just get a console? [00:36] kirkland: try ttyconsole then [00:36] kirkland: virsh ttyconsole vm-name [00:37] mathiaz: didn't work either [00:37] kirkland: try virsh dumpxml vmname [00:37] kirkland: there should be the vnc port there [00:37] mathiaz: hang on [00:38] mathiaz: kernel panic'd [00:38] mathiaz: not enough memory [00:38] mathiaz: it's booting now [00:38] kirkland: on the host or on the guest? [00:38] mathiaz: i connected to it through virt-manager [00:38] mathiaz: guest [00:39] kirkland: ok - you can use virt-manager to connect to it also - it uses the vnc display IIRC [00:39] mathiaz: okay network is working [00:39] mathiaz: how can i verify that i'm virtio-networking? [00:40] mathiaz: lsmod | grep virt shows: virtio_net, virtio_pci, virtio_ring, virtio [00:42] kirkland: on the host try to get the kvm command line [00:42] kirkland: the list of modules seems correct too [00:42] mathiaz: ? [00:42] mathiaz: kvm command line? [00:43] kirkland: on the host, libvirtd starts kvm to run your guest [00:43] kirkland: so if you grep the kvm processes you'll see the one that runs your virtio guest [00:43] mathiaz: ah, looking good [00:43] kirkland: it should show that the network interface is started with a virtio driver [00:43] /usr/bin/kvm -S -M pc -m 512 -smp 1 -name intrepid -monitor pty -no-acpi -boot c -drive file=/local/virt/intrepid-virtio.img,if=ide,index=0,boot=on -net nic,macaddr=00:16:3e:32:80:31,vlan=0,model=virtio -net tap,fd=11,script=,vlan=0,ifname=vnet1 -serial none -parallel none -usb -vnc 127.0.0.1: [00:44] kirkland: right - so I think it all looks good [00:44] kirkland: you've got a guest running with a virtio network device. [00:44] mathiaz: excellent, i think this one is put to bed ;-) [00:45] mathiaz: and TheMuso has committed the remaining fix for virtio disks [00:45] mathiaz: now, I just need to go through the dozen or so KVM CVE's that kees threw at me :-) [00:45] kirkland: right - to be correct the virtio network bug was in the installer [00:45] kirkland: the installer wasn't able to load the virtio network modules [00:46] mathiaz: right [00:46] mathiaz: i just mean that you've verified hardy host, me intrepid host now [00:46] kirkland: have you booted an install? [00:47] hrm, no [00:47] kirkland: well, many of those aren't kvm [00:47] mathiaz: ah, i see what you mean [00:47] kees: i know, i was just lighting you up :-) [00:47] haha [00:48] kirkland: I've done a couple of test install today and they all completed correctly (all my guests are setup to use virtio networking by default) [00:48] kirkland: so I'm confident the installer work is working correctly. [00:53] mathiaz: sounds good [00:54] mathiaz: i need to respond to a couple of ecryptfs and raid bugs that i've neglected while focusing on virtio over the last few days [00:54] mathiaz: btw ... i moved my blog [00:54] mathiaz: we'll need to update the ubuntu-server pull [00:54] mathiaz: new location is at blog.dustinkirkland.com [00:54] mathiaz: i imported all of my old articles from wordpress [00:55] mathiaz: planet.ubuntu.com is currently broken [00:55] kirkland: noticed that. I'll update the feed on ubuntuserver. [00:59] kirkland: is there a feed for the Ubuntu-Server label? [01:00] kirkland: I don't see an obvious link to just pull post labeled with Ubuntu-Server [01:00] mathiaz: one sec [01:01] mathiaz: http://blog.dustinkirkland.com/feeds/posts/default/-/ubuntu-server [01:05] kirkland: awesome - thanks. [01:05] kirkland: note that the url is case sensitive [01:06] XDMCP/GDM: problem connecting to Ubunut 7.10 w/ XWin under cygwin. It worked last week. Won't work this week! Has latest upgrade added some firwall security? Any ideas on where to lok for a solution? [01:06] kirkland: http://blog.dustinkirkland.com/feeds/posts/default/-/Ubuntu-Server [01:06] kirkland: ^^ that works [01:07] Baryon: I'm not sure that here is the best place to ask for X related questions. [01:07] Baryon: you may wanna try #ubuntu instead. [01:08] mathiaz: I thought that it may be a server security issue - e.g blocked port. [01:09] mathiaz: But I will try #ubuntu. Thanks. [01:11] Baryon: an update in a stable release won't introduce such a new feature. [01:12] mathiaz: Then I am at a loss, and don't know how I broke XDMCP! [01:18] mathiaz: If you are right, then that leaves but one other possibility. Thanks! [01:48] XDMCP or GDM or port problem. It worked last week. Cannot login to Ubuntu Server 7.10 via thin client (XWin under cygwin on XP64). I don't know what changed. Any ideas? [02:09] zul: is the -virtual kernel working in a xen environment? [03:47] New bug: #283534 in openssh (main) "Segmentation Fault with SSH" [Undecided,New] https://launchpad.net/bugs/283534 [03:59] Hiya, never done this, so I thought I'd ask before I break some stuff seriously. Have a ubuntu server in production that is running samba without domain logons. All users are connected as guests. I want to migrate to it being a domain server, but I don't want to break everything immediately... [04:00] I want to turn domain logons on, but before I do, will this kill all the clients, require accounts or any such major impact, or can I just turn it on and the clients will be able to connect as they have as a guest? [04:01] * owh is slowly bringing this client into the 21st century :) [04:03] Where might I locate a process that will allow me to migrate from this essentially peer-to-peer environment into one where the server is a pdc? === bdmurray_ is now known as bdmurray [04:50] h === mcasadevall is now known as NCommander [05:05] NCommander: Pong [05:06] ScottK, I need your 2 cents on a backport which is somewhat unique [05:06] OK [05:07] ScottK, https://bugs.edge.launchpad.net/hardy-backports/+bug/283468 [05:07] Launchpad bug 283468 in hardy-backports "[hardy] kdesvn not compatible with svn 1.5 - please backport (import) kdesvn from debian unstable to hardy" [High,Incomplete] [05:07] I'm all for it. [05:07] Ok, I confirmed it build, and installs [05:08] How's it work with svn 1.5? [05:08] I haven't confirmed the original but I"m about to test to make sure it works with the version from SVN [05:08] I am testing that now ;_) [05:09] ScottK, I'm suprised you said yes, I was expecting a no since it requires backporting a sid package [05:10] NCommander: It's not really different than sticking KDE3.5.10 in hardy-backports when we have no KDE3 in Intrepid. [05:10] I think KDE3 in Intrepid is a special case. [05:11] This is part of that because if we hadn't gone to KDE4 in Inrepid, then you'd be able to backport from there. [05:11] Well, we can't fix the problem in intrepid since thats frozen, and kdesvn won't qualify for a FFe. and it will be fixed in Jaunty via an import from Debian, I don't see the huge issue [05:12] In Intrepid we'd want the KDE4 version anyway. [05:12] Which there is one. [05:12] Ah [05:12] That makes sense [05:13] This bug must also exist in intrepid then ... [05:13] Unless the KDE4 version is an updated one. [05:13] I know it's packaged, but not if it works with 1.5 [05:14] It's the same version [05:14] (in both intrepid and hardy) [05:14] The KDE4 one is kdesvn-kde4 [05:15] Dunno why we didn't transition it. [05:15] Probably got overlooked [05:15] So it's a bug in Intrepid. [05:15] Yeah. [05:15] Probably should be fixed now [05:15] kdesvn just crashes for me [05:16] Lemme see if I can find someone to sponsor it for you. [05:18] NCommander: There you go (#kubunut-devel) [05:18] For some better spelling of kubuntu [05:18] lol [05:18] seems accurate to me [05:18] ScottK, so when are we going to create intrepid-backports? [05:19] NCommander: That's up to jdong, but it can be created as soon as the release happens. [05:20] Ah [05:21] ScottK, I'm confirming both the bug, and that backporting kdesvn fixes it [05:22] NCommander: OK. See nixternal about fixing Intrepid and then we'll deal with the backport. [05:22] ScottK, the package is in the PPA, but I'm unsure I got the versioning right (I assume its normal, but this might be a special case since its a backport from sid vs intrepid/jaunty) [05:24] NCommander: What version are you pulling in? 0.14.6-1 or 1.0.4-2? [05:32] 1.0.4-2 from Debian sid [05:32] Which works [05:41] OK. [05:41] That's a KDE3 version still. [05:41] The KDE4 version is some higher number. [05:50] ah [06:39] is there any tool to tset the squid performance === freaky_t is now known as fReAky_t === fReAky_t is now known as freaky_t [07:10] Is there a large resource loss installing ubuntu-desktop on top of a ubuntu-server? [07:11] talking about the ubuntu-desktop package... CLI is to time assuming... [07:12] install xfce [07:12] if you're going to use a gui [07:12] I mean, most of the tools you use on a server are console based anyway [07:13] The main reason i want a gui is to run tools like gbind or QG (GQ?) [07:19] is there any tool to tset the squid performance [07:20] <_ruben> Scix: best way is to just install the xlibs and run an X server on your desktop .. then use ssh -X to tunnel your X traffic [07:22] xlibs and that's it? Sounds realy easy :) [07:23] <_ruben> and any other dependencies you might need for gbind .. shouldnt require a full blown X install [07:24] _ruben: tanks :) I'll try it out [07:33] Do i need to install gbind on the server, or at the workstation? === mdz_ is now known as mdz [10:02] please anyone tell me, how to get the "switchable on-the-fly I/O schedulers " in kernel 2.6.26 [10:16] idea for intrepid+1 - bring up SSH daemon when in single user mode, since network is already up :/ [10:38] <[Solars]> ivoks that makes too much sense [10:42] i have a situation where server is couple of thousands of miles away [10:43] and it has filesystem error and no ipmi and no any remote management [10:43] but it did bring network up, i can ping it and everything [10:43] i just can't do anything about it :/ === freaky_t is now known as fReAky_t === fReAky_t is now known as freaky_t [12:51] hi could some one have a look at bug #258192 [12:51] Launchpad bug 258192 in dhcp3 "problem with paths and binding to ldap server" [Medium,Confirmed] https://launchpad.net/bugs/258192 [12:59] zul: would you consider bug 280428 a bug or a feature ? [12:59] Launchpad bug 280428 in openvpn "Prompted for VPN (openvpn) user info at boot-up" [Undecided,Invalid] https://launchpad.net/bugs/280428 [13:00] Koon: just a sec [13:01] Koon: bug imho [13:02] zul: it's a debian "fix" we merged [13:02] zul might also cause bug 278784 [13:02] Launchpad bug 278784 in openvpn "openvpn configuration with token (pkcs11 provider) blocks the boot" [Medium,New] https://launchpad.net/bugs/278784 [13:02] since according to the reporter it did not occur with hardy [13:03] Koon: *sigh* yeah I saw I still think its a bug though [13:03] zul: so we should revert the change and add the < /dev/null again ? [13:03] I think so [13:04] zul: I wanted confirmation that it would fix 278784 too, but I won't have it before tomorrow [13:04] Koon: ok hmmm.... [13:04] I'll prepare the fix today and we'll see how we can slip it before freeze tomorrow ? [13:05] or just fix it [13:05] slip it [13:13] ubuntu.com lets me download an alternative desktop cd, is there an alternative server cd available as well? [13:13] I would like to install ubuntu server to a software raid with lvm [13:15] TrioTorus: there is not a alternate server disk. but google ubuntu lvm raid, and you should find something to help you :) [13:16] TrioTorus: You do not need an alternate cd, but I would recommend waiting for Intrepid as kirkland as worked a lot to make it easier to set this up [13:19] ok, anywhere I can see a 'what's new' page for intrepid server? [13:30] I guess a 'what's new is only for after the release. Makes sense. Thanks for the pointers. [14:26] im doing all my server updates today -- is there a way i can make it do an update, upgrade and all in one command without me having to come back to it and tell it yes i want to do it? [14:34] W8TAH: you can try sudo apt-get update && sudo apt-get -y upgrade [14:36] cool [14:36] thanks [15:14] if i specify dns-nameservers in /etc/network/interfaces will ubuntu overwrite /etc/resolv.conf on boot? [15:31] New bug: #283807 in freeradius (universe) "freeradius with openssl support doesn't compile" [Undecided,New] https://launchpad.net/bugs/283807 [15:43] anyone have any recommendations about the best way to prevent people from sending e-mail from addys on my domain? [15:44] I've tried using SPF records, but it seems to still be happening [15:44] stop using smtp? [15:45] sorry, there really isn't any good cures for spam out there [15:46] brewmaster, there is nothing you can do. It is totally up to the receiving mail server to check and enforce your SPF [15:47] brousch, meaning the people that the mail's are being sent to? [15:48] brewmaster: yes [15:49] brewmaster, not even the recipients, it is up to the recipient's mail server admin. I don't think there is any SPF checking for mail clients [15:51] brousch, I'm just a little distraught that I can do this: http://pastebin.ca/1227864 [15:52] if I run that code on a completely different server, gmail sees no problem with it [15:52] they even give an SPF pass [15:53] Gmail is one of the few that that checks SPF [15:53] SPF records are in a server's DNS, no? [15:53] right [15:53] But most recipient mail servers don't care about SPF [15:54] Gmail checks SPF, but I don't think they reject based on it. I think they just send it to spam [15:56] If Gmail is passing SPF from what should be incorrect servers, then your SPF may be set up wrong [16:01] brousch, weird, openspf tells me everything is fine with my spf records [16:01] it rejects from the IP i'm testing [16:01] yet gmail labels it as a "pass" [16:02] that is unexpected [16:04] just tested somewhere else, it gave me a "softfail" [16:06] wait a minute... [16:06] "However, the domain is still testing its SPF policy, so the message should not have been rejected." [16:06] i didn't realize my domain was testing the SPF policy.... [16:06] (and I have no idea what that mean :P) [16:07] I have no idea either [16:09] brousch, if I'm getting rejection letters saying that messages from "spammer@yourdomain.com" are being rejected, then that is a good thing, no? [16:10] It means people are rejecting spam that is using your domain name. Whether that is good or not is subjective. ;) [16:21] zul: posted openvpn fix for bug 280428, will you handle it or should I subscribe main-sponsors ? [16:21] Launchpad bug 280428 in openvpn "Prompted for VPN (openvpn) user info at boot-up" [Medium,Confirmed] https://launchpad.net/bugs/280428 [16:22] Koon: ill do it right now [16:22] this one behaves like hardy did [16:23] brousch, good that it's recognized as spam since it's not me who's sending it, bad that people are trying to send spam through my domain [16:27] Koon: done [16:27] * Koon hugs zul [16:34] i'm trying to install and use ltsp on ubuntu server, which i thought was available with tasksel. however tasksel does not execute. [16:34] what am i doing wrong? [16:37] I am trying to use the break=top option on the Ubuntu Hardy Server CD. Should this work like it typically does? [16:38] It's just going straight to the install screen without dropping me into busybox [16:41] New bug: #283844 in sg3-utils (universe) "missing scsi_reserve script" [Undecided,New] https://launchpad.net/bugs/283844 [16:44] hi there. i've a question about how ubuntu server releases updates. does it only send out patches like rhel or new releases aswell? [16:44] say.. will there wver be a php 5.3 package in 8.04LTS? === freaky_t is now known as fReAky_t [16:57] New bug: #277213 in apache2 (main) "Apache 2.2.9 strange logs" [Undecided,Invalid] https://launchpad.net/bugs/277213 [17:08] I am trying to install Ubuntu Hardy on an IBM Netfinity 5100 and it is failing with all sorts of different issues. I just tried pure Debian and it seems to be working fine. [17:11] With Ubuntu server it gets through part of the screens until it stops and errors about reading the cd. The cd has been checked and is fine... [17:11] with the Ubuntu hardy live cd it spits out IO errors [17:14] rrittenhouse: was this an official cd or one your burned? [17:19] greenfly: One that was burned. I just used it in another machine and it worked fine. [17:22] rrittenhouse: I've seen things like that some times when I burn a CD at a fast rate. it seems some CD-ROM drives can read it fine and others can't [17:22] perhaps try burning a new CD-ROM at a much slower speed, or alternatively, on a different CD-R [17:34] greenfly: Weve tried 3 different cd-rom drives and a new ide cable [17:35] greenfly: It always seems to get to the same step with the server cds when it fails [17:35] greenfly: For some reason the Debian cd worked like a charm [17:37] Did you try the alternate CD? It uses the same installer that Debian uses. [17:37] ScottK: Yeah with the same results. [17:38] Odd. [17:44] ScottK: Like right now for instance it's stuck on binutils-static-udeb and it ends up erroring out. [17:44] Sorry. No ideas here. [17:44] ok [17:44] thanks anyhow [17:53] ScottK: https://bugs.launchpad.net/ubuntu/hardy/+source/linux/+bug/148466 Appears to be the same problem were having here. [17:53] Launchpad bug 148466 in linux "Won't boot on a DELL Poweredge 2500" [Undecided,Fix released] === fReAky_t is now known as freaky[t] === ogra_ is now known as ogra [19:30] Hey guys, quick question. How can I tell if the ubuntu version of 32bit or 64bit? [19:31] uname -a [19:32] No, tried that. [19:32] On a 32bit install: Linux jayoffice 2.6.24-21-generic #1 SMP Mon Aug 25 17:32:09 UTC 2008 i686 GNU/Linux [19:33] dammit [19:33] Linux terra 2.6.24-19-server #1 SMP Wed Aug 20 18:43:06 UTC 2008 x86_64 GNU/Linux [19:33] Just noticed the second-to-last word... [19:33] ivoks: Thank you and sorry! [19:34] np [19:34] ivoks: You should have to do VERY little custom configuration of amavisd-new in Intrepid. The Ubuntu defaults will look very familiar to you. Thanks. [19:34] ScottK: :D [19:35] For Jaunty I think we're well positioned now to automate integration with postfix, clamav, and spamassassin. [19:35] ScottK: we could push quarantine system there too [19:36] I think so, but I haven't looked at the specifics. [19:36] the main problem is that there's lots of products [19:36] but they don't blend very nice... [19:36] Is it possible to deinstall Apache & PHP cleanly so that no traces of the software are left and I can reinstall them from scratch? [19:36] we should just take the code and make one final solution :) [19:37] xbxb: apt-get --purge remove apache.* php.* [19:37] thanks... will this also get rid of configuration files, etc? [19:37] xbxb: yes [19:37] ivoks: thank you [19:38] xbxb: unless you created your own configs in /etc/apache2/conf.d/* or /etc/php5/conf.d/* [19:38] xbxb: you'll have to remove those manualy [19:38] but it will delete every file of the software that was created automatically, yes? [19:38] ivoks: Then I guess we ought to pick one and make it play nice. [19:39] ScottK: or write something from scratch :/ [19:39] Yes, but we should avoid that where it's reasonable to do so. [19:40] ScottK: i told you i'm thinking of starting a project inside my company which should result with something like that [19:40] ScottK: in the end, it would be open source, and pushed into ubuntu [19:40] .. [19:40] ScottK: but this is like... 2 years time frame [19:41] anyway... time to grab and work on ebox [19:41] ivoks: is this correct? [19:41] xbxb: yes [19:41] k [19:41] xbxb: man apt-get [19:42] Use purge instead of remove for anything that would be removed. [19:42] oh... [19:42] there's apt-get purge now [19:42] nice... [19:42] didn't know about that :D [19:53] hi. i installed ubuntu server 8.04 with postgresql, but postgresql does not seem to have the fuzzystrmatch functions. why is that_ [19:56] Level15: I don't know, but #postgresql may be a better place to ask [19:56] xbxb: im asking there now [20:31] New bug: #283982 in samba (main) "Windbindd crashes while handling something with trusted domains" [Undecided,New] https://launchpad.net/bugs/283982 [20:54] Hi all. I have a problem with Torrentflux-b4rt. I just set it up and got it running now, the problem is, I can only access it locally on my network (with a local ip, e.g 192.168.1.10) and not with my dns (or the ip that I get from my provider, which is the same). When I try to do this I get a "404 error, file not found". I suspect it might have something to do with my routers ability to be remotely configured. It's a Linksys WRT54gl router. Th [21:01] anlif: what port is it running on, and have you forwarded that in your router? [21:22] ScottK: kewl [21:22] (integration fun) [21:22] ;-) === ajmitch_ is now known as ajmitch