[00:16] Any body know how to fix this: http://paste.uni.cc/17831 [00:16] could someone help me look at the error: http://paste.uni.cc/17831 [00:16] I have already ask [00:16] ed [00:16] I try to upgrade from 7.04 - > 7.10 [00:16] thank you [00:16] so sad [00:39] Wahahaha!!!!! [00:39] wawawwawawawa [00:40] * fujin_ blows up pschulz02 [00:40] Wait foe it.. [00:40] root@newmoon:~# ldapaddgroup --help [00:40] Successfully added group --help to LDAP [00:40] WTF! [00:40] for it. Kamping_Kaiser is about to post something [00:40] posted [00:40] honestly. wth. [00:41] * kgoetz files bug [00:41] man ldapaddgroup [00:41] That's hardly worth filing a bug. [00:42] what the hell even provides ldapaddgroup? [00:42] that's not in ldap-utils [00:42] how is it not? [00:42] ldapscripts [00:42] What's the bet that 'ldapdeletegroup' accespts the --help flag? [00:42] hehe [00:43] root@newmoon:~# ldapdeletegroup --help [00:43] Successfully deleted group cn=--help, from LDAP [00:43] NAME [00:43] ldapaddgroup - adds a POSIX group entry to LDAP. [00:44] SYNOPSIS [00:44] ldapaddgroup [gid] [00:44] OPTIONS [00:44] [00:44] The name of the group to add. [00:44] learn2manpage [00:44] That's no bug. [00:44] That's you being an idiot. [00:44] why? [00:44] you didn't read the manpage? [00:44] and why is it not a bug? in your clearly (not so) humble opinion? [00:44] the expected behaviour is for ldapaddgroup to try and add --help as a group [00:45] no, i go for --help before i go for man [00:45] There's the idiocy. [00:45] *why* [00:45] you're doing it wrong? [00:45] manpage clearly states that there isn't a --help [00:46] * kgoetz gives up. [00:46] thank fuck for that [00:46] !tell fujin_ about coc [00:46] !tell kgoetz i don't give a fuck [00:46] oh snap [00:47] !tell kgoetz about man [00:47] your in ubuntu chanel here, not debian... [00:47] *an ubuntu [00:48] regardless of where I am, I believe 'rtfm' explains what I was trying to achieve. [01:07] fujin_, That isn't acceptable here. [01:07] cool [01:08] what are you going to do about it? [01:09] If you can't abide by the Ubuntu Code of Conduct and the Ubuntu IRC Channel guidelines, I think it would be best if you left. [01:09] We don't call people idiots [01:10] I do [01:10] and I don't intend on leaving [01:10] see, I wast just educatin [01:21] infinity, ping [01:22] Ow. === joerlend_ is now known as XiXaQ [02:26] Is setfcaps in Ubuntu? [02:27] Is it a package? [02:32] somerville32: No. It's part of the linux kernel capabilities framework. I am testing different methods of privisioning a tap device for use with kvm. setfcaps is not in libcap-bin or any other package that I can find. [02:49] * antdedyet submitted the question to launchpad's answers section [02:59] You must have the Reductive build library in your RUBYLIB. [02:59] ergh [03:06] 'lo dendrobates [03:07] antdedyet: hi [03:08] * antdedyet is searching for setfcaps in Ubuntu [03:16] Nothing I've found so far in Ubuntu is both properly creating & provisioning a user owned tap device that can be used with kvm. [03:17] s/creating \& provisioning/creates \& provisions/ [03:21] antdedyet: hmm, I don't use tap with kvm, but I'll look into it. [03:26] dendrobates: a basic test to run, which fails shortly (before the qemu/kvm window pops up) for me and at least for someone on the community docs is this: 'kvm -net nic -net tap -hda foobar.img -boot c' [03:26] replace whatever's proper after the tap ... [03:28] I have tunctl also that I found with uml-utilities, but kvm won't use the device I've setup with it, using 'sudo tunctl -b -u myuser' [03:29] I tracked it down to being that the tuntap kernel driver had NET_CAP_ADMIN capability required since sometime around 2.6.17 [03:30] 'setfcaps' is to add the capability for any user to on the system to use tun/tap devices to some degree. [03:31] But I never found setfcaps and that's where my testing stopped about 5 minutes ago. [03:35] I use tap for vpn's without needing setfcaps. [03:35] it is probably something else. [03:39] dendrobates: There is also the ownership and mode of /dev/net/tun ... In the beginning, I changed it's mode from 0660 to 0666. [03:41] antdedyet: are you using gutsy? [03:41] dendrobates: hardy daily ... :) [03:42] dendrobates: and of course, on the only machine my processor has the kvm extensions [03:44] antdedyet: I'm very glad to hear your thoughts and experiences - exactly what errors are you getting? I've run into tap/kvm problems also but attributed them to never actually bringing my network down and making my regular eth0 into a bridged device or something like that, so the tap could work with it [03:45] my testing is on gutsy [03:47] e.g. "can't add tap0 to bridge eth2: Operation not supported" - which i think was because at the time my main network device (eth2) was not bridgeable [03:49] using directions like this: http://compsoc.dur.ac.uk/~djw/qemu.html [03:50] soren just updated kvm, there may be a regression. [03:53] nealmcb: I have a multiple scenarios that I have been through. Testing one at a time, with the outcome of none of them working, I started by setting up a bridging interface slaved to either of my outgoing ethernet connections, one is cabled, the other wifi (which I use most of the time and really wanted to use - FUD aside). Also an evdo ppp interface at times. [03:53] nealmcb: basic error was this: [03:54] warning: could not open /dev/net/tun: no virtual network emulation [03:54] Could not initialize device 'tap' [03:56] nealmcb: I followed those directions too, but started with this page: https://wiki.ubuntu.com/kvm then moved on to for networking input: https://help.ubuntu.com/community/KVM [03:56] dendrobates: running an update now [03:57] it was a couple days ago, so you are probably using the newest one, but it has not been tested yet. [03:57] dendrobates: ah yes, I got that one I believe: 55 [03:58] soren asked me to let him in on what I came across since I have amd hardware [03:59] although, I don't think this is cpu specific. [03:59] he uploaded 56, maybe it has hit the repo's yet. [04:00] dendrobates: ahh, ok. I use mirrors.kernel.org ... let me change my repo. [04:05] antdedyet: have you tried gutsy? [04:07] nealmcb: I don't have a gutsy box with kvm cpu extensions. I could test qemu with a bit of setup, but I haven't done that yet. [04:07] right - I forgot that [04:07] but qemu would be an interesting test [04:07] I don't see 56 on mirrors.anl.gov either. [04:07] * antdedyet nods [04:08] hmm: http://packages.ubuntu.com/hardy/misc/kvm-source [04:16] Is there somewhere I'm not looking at to find the 56 package? [04:16] * antdedyet doesn't see it in deb sid either ... [04:31] * antdedyet shrugs [05:19] * antdedyet stretches ... I can't wait to install Ubuntu on more client machines. :) [05:22] None of which are desktop machines, all servers. [05:29] nealmcb: same error for qemu on gutsy: [05:29] warning: could not open /dev/net/tun: no virtual network emulation [05:29] Could not initialize device 'tap' [05:32] ah: http://kvm.qumranet.com/kvmwiki/ChangeLog [05:34] dendrobates tells no lies! [05:40] -net tap,vlan=0,ifname=tap0 \ [05:41] after 'sudo tunctl -b -u myuser' # when no other tun/tap devices are built/present [05:41] gives this on kvm init: [05:41] SIOCSIFADDR: Permission denied [05:41] SIOCSIFFLAGS: Permission denied [05:41] SIOCSIFFLAGS: Permission denied [05:41] can't add tap0 to bridge br0: Operation not permitted [06:17] After manually populating tap0 with a static address and pointing the dhcp server's listen statement to it, my kvm session gets an IP that is reachable on the host, even with perm denied errors. [06:34] antdedyet: interesting! [07:42] jdstrand: congrats on becoming an motu! [07:57] moin [08:10] nealmcb: Thanks for pointing me at the ubuntu-jeos-builder script. I'm not yet down to 90 seconds, but I'm at 180, which is a whole lot better than 15 minutes for a copy/paste/fixup or 45 minutes for a bare .ISO install. Now all I need to do is combine the options from the various versions so I can specify the hostname, directory, ip address and not have an opt drive, but so far, so good. [08:19] antdedyet: I apparantly never got round to uploading kvm 56.. I'll do that today. [08:23] soren: nice! [08:23] <_ruben> any vmware gurus around? .. my ubuntu guest is spamming my sles host with "host clock rate change request" messages .. i've added "clock=pit" to the ubuntu guest's grub but it doesnt seem to make a diff [08:24] however, I didn't see anything in the upstream changelog that would lessen the bootstrap aches. [08:24] antdedyet: This was about tap devices, was it? [08:24] * antdedyet will be happy to test it nevertheless [08:24] soren: Yes, it was about tap devices. [09:12] antdedyet: Uploading kvm 56. [09:14] soren: Spectacular. [09:16] I've _really_ got to get more clients on Ubuntu so I can develop more with it. It may get difficult to justify why testing out new servers settings takes awhile if they find out I'm running them all inside a Ubuntu virtual machine. :) [09:22] :) [10:02] Is it possible to configure your ubuntu server to use the smtp server of your ISP to send mail? [10:03] If so, is there any good tutorial for this? Searched the web, but cant find a clear and simple tutorial. [10:04] installing een mta eg exim4 wil ask you a few questions about how to configure the mailserver [10:05] one of the options is, smtp with smarthost, where smarthost is the smtp-server of your provider [10:08] avatar_: Thx, that cleared up a little bit, can i paste some dutch output of the setup wizard? [10:10] Im a bit confused what to fill in [10:13] sergevn: lets switch to #ubuntu-nl, i'm dutch too :) [10:24] any one have ubutu as mail server ? [10:28] RCOMALTA: we use ubuntu in our mailsetup [10:28] RCOMALTA: Sure. [10:30] 6x mx, 4x filter (virus) 6x spamd, 3x maildrop, 4x smtp [10:30] all running exim [10:30] configuration from ldap [10:31] avatar is smtp and pop 3 mail setup [10:32] RCOMALTA: we are running mail voor a few thousand domains, so we use quite a bit of servers. Running mail for a single lowtraffic domain can happily run smtp/pop/imap on one singel server [10:34] avatar, your mail servers are all with ubuutu [10:49] i will do it [10:49] and the end computers are windows xp or vista [11:44] avatar the end users are with windows xp and vista [13:14] nealmcb: thanks! :) [14:01] ScottK: morning, time for a question? [14:02] ScottK: I was wondering about https://wiki.ubuntu.com/MOTU/Clamav [14:29] wonderful news - they've finally figured out that our voting equipment is deeply flawed - http://www.bradblog.com/?p=5451 [14:30] so I'll miss the server team meeting - I'll be be down at the hearing today [14:30] nealmcb: heh... I was just reading an article on ArsTechnica about the Ohio voting machines [14:30] I've been pointing this stuff out for 4 years now.... finally getting more traction.... [14:31] is it too late to change before 08 elections? [14:33] wow actually they're talking about the same machines used in ohio [14:33] nealmcb: good luck with that... seems like a big mess to me === newbie23_ is now known as newbie23 [15:00] is there a meeting going on now? [15:00] It's in an hour, I think. [16:00] Meeting in #ubuntu-server. [16:00] Meeting in #ubuntu-meeting, I mean. [16:02] <_ruben> heh === Delvien_ is now known as Delvien [16:41] !mta [16:41] A Mail Transfer Agent (MTA) is the server software that sends and queues mail. The default MTA (and !MDA) on Ubuntu is !postfix ("exim" is also officially supported). See also !MailServer and !MUA === juliux_ is now known as juliux [16:59] Fantastic. I upload kvm 56, and 6 hours later, kvm 57 is released. [17:21] who makes the best USB keyboard/mouse kvm's, w/ dvi? [17:30] sorry for the meeting :/ [18:33] * ScottK is curious if anyone has comments on the message I just sent to the Ubuntu Server ML. [18:45] ScottK: I'm for it... instructions have been added to the Mail Filtering seciton :-) [18:46] sommer: Great. To answer your clamav question, I haven't had much time for it recently. [18:46] sommer: Now for the fun part is we are about to get libclamav3 in the version that was just released yesterday so we need to worry that library transition first. [18:47] cool, I was looking at packaging again and remembered the backport project [18:47] OK. We still need to do it, but we need to fix Hardy first. [18:47] sure [18:48] I also realized that I didn't understand exactly what a "backport" is [18:48] I actually backported the new api to the existing source instead of backporting the new package... doh [18:48] it was fun though... so I can't complain too much [18:49] sommer: If you can figure the other way around to backport a clamav with the old API, that would solve the whole Dapper problem right away. [18:49] ScottK: heh... I really don't know that much C :/ [18:49] We can just patch the current clamav to use the old API, upload that to dapper-backports and declare victory. [18:50] Well if you figured it out one way, you ought to be able to do the reverse, right??? [18:50] mmmMMMM... that really doesn't sound too hard [18:51] there's documentation for the other way though :) [18:51] I'll do some digging and see where it takes me [18:54] sommer: Cool. That's solve a huge problem in Dapper. [18:57] ScottK: another question regarding python-policyd-spf. The order of execution is postfix filters a message through dkim-milter and python-spf then sends it to Amavisd? [18:58] or amavisd scans it first then postfix pipes it through any other "outside" filters? [18:58] sommer: It's a function of how you set up your system. [18:58] okay, I docuemnted it the first way [18:58] trying to explain how all the peices fit together [18:59] Generally avamisd-new will run after queue. [18:59] Running before queue filters can be tricky. [18:59] that's what I was thinking after reading the postfix book and amavisd docs [18:59] Policy servers will run before queue as will milters (although milters are more complex). [19:00] cool that's the one part of documenting the postfix+amavisd+spamassassin+clamav+spf+dkim I wasn't quite sure of [19:01] thanks... if there are any issues with the mail filtering section you find just let me know [19:11] sommer: Sure. I guarantee you will have to keep harassing me to look at it. [19:31] ScottK: for dapper clamav as you say to backport the clamav api .. [19:32] ScottK: can this be to replace clamav's libclamav with ubuntu backported api libclamav ?? [19:32] I'd put it a little differently. [19:33] I'd take the current libclamav and patch it to use the old API and backport that. [19:33] this way there won't be dependency problems ? [19:37] Exactly. [19:40] current you mean dapper's .88 or new .92 ? [19:47] I'm stumped here, my thin clients that have been working for a month just stopped letting me login since Sunday. The only thing that changed is that I finished setting up a chillispot/freeradius server. [19:47] I've already tried running the ltsp-update-sshkeys and I do see the client connecting in syslog and I don't see any error messages when login is attempted. [19:56] leonel: It'd be to make the current (still 0.91.2) libclamav look like the 0.8x one to the rest of the packages in the system. [20:18] ScottK: Ok i got it ... [20:23] leonel: Just checking to see did you get /msg I just sent you? [20:24] yes [21:38] leonel: 0.92~dfsg-0build1 uploaded. Now we wait for the archive admins. [21:39] ScottK: great ! [21:40] ScottK: the next is do a merge to hardy then backport to gutsy right ? [21:40] Actually it looks like a sync. I think Debian has all our needed changes now. [21:41] Because of the soname change we've got the same problems backporting to Gutsy and Feisty that we've already had with Dapper/Edgy. [21:41] * leonel goes to read the diff in merge and sync ... [21:51] what's the preferred way to 'git' on Ubuntu? [21:51] cogito? [22:25] git-core package === Gamble61 is now known as gamble6x [23:53] I need to create a gui interface for some server configuration stuff and I was wondering what the ubuntu preferred tools were. [23:54] I've been looking at gtkdialog, but the included examples don't all work and the manual is pretty skinny.