[00:03] yes :) [00:08] mmkay, just making sure :P [01:07] Hey all I'm working with ubuntu 12.04 server and trying to setup ldap with kerberos and when I run the command sudo krb5_newrealm I keep getting an error message "kdb5_util: Password mismatch while reading master key from keyboard" I've run this command 3 times and keep getting the same error. Is there a file I need to delete so I can rerun the command and have it work? Thanks. [01:15] How can i prevent Ubuntu from routing between vlan's using the "ip route" command? [01:18] it won't do that unless you explicitly configure it to [01:19] Hey all I'm working with ubuntu 12.04 server and trying to setup ldap with kerberos and when I run the command sudo krb5_newrealm I keep getting an error message "kdb5_util: Password mismatch while reading master key from keyboard" I've run this command 3 times and keep getting the same error. Is there a file I need to delete so I can rerun the command and have it work? Thanks. [01:19] !patience | bearly230 [01:19] bearly230: Don't feel ignored and repeat your question quickly; if nobody knows your answer, nobody will answer you. While you wait, try searching https://help.ubuntu.com or http://ubuntuforums.org or http://askubuntu.com/ [01:22] qman__: Talking to me? [01:22] yes [01:23] ubuntu does not route out of the box [01:23] you have to set it up to do that [01:24] if you want to enable routing but filter what gets routed, you need to set up iptables [01:24] and if you want to filter traffic as it passes through a bridge, you need ebtables === n0ts_off is now known as n0ts [01:38] Hey all I'm working with ubuntu 12.04 server and trying to setup ldap with kerberos and when I run the command sudo krb5_newrealm I keep getting an error message "kdb5_util: Password mismatch while reading master key from keyboard" I've run this command 3 times and keep getting the same error. Is there a file I need to delete so I can rerun the command and have it work? Thanks. [01:38] bearly230: 19 minutes isn't much time to wait... [01:39] How long would you suggest post once and wait 24 hours? 20 mins is more than a reasonable time between posts. [01:40] three times in thirty minutes is definitely too much :) try every half-hour, tops... [01:41] don't overlook askubuntu.com -- I've usually preferred irc over websites, but I think the stackexchange folks figured out how to make 'web forum' question-and-answer sites not suck. It's a different audience, might work well. :) [01:41] The purpose of not flooding a channel is so everyone can talk. With as quite as this room is. You never know when someone will actually come on that can answer the question. [01:42] I've already googled the error and didn't locate any answer. [01:42] that's a good point, but I've got two new joins and one guy re-joining :) hehe [01:44] irc isn't like it use to be. [01:44] indeed. :/ [01:44] to quite these days ( [01:47] qman_: I prevented routing between vlans using iptables, just woundering if it is possible using "ip route" command. [01:47] ? === blinkin_ is now known as blinkin [01:49] bearly230: man, I can't see anything in the krb5 code that would give you that error condition except a mis-typed password. :( [01:51] sarnold: kind of hard to believe I would mistype the password each time I ran the command, 5 times now. But I am human. So has to be computer, can't be me I'm perfect. hehe [01:51] bearly230: indeed. :) [01:51] bearly230: can you strace the silly thing? (I'm not sure if it is setuid or something, might forbid it...) [01:52] or try with a password you cannot mistype? 'a' for example? :) [01:53] sarnold: Just prompted me again. This time I did the password with my eyes shut. And it worked. [01:53] bearly230: hahaha [01:53] sarnold: guess it want to make sure the pw was so secure that I didn't even see nothing echo to the screen. [01:54] so either your PW was typed incorrectly when you set it up, or you type better when you are blind. [01:54] and the PW is never shown on screen ;P [01:54] nor the number of characters ;P [01:54] bearly230: reminds me of a story I heard of someone who couldn't log in while standing but could log in while sitting.. something about keycaps being fiddled with, and touch-typing while sitting but look-typing while standing. :) [01:55] yeah, I'm upset at many gui tools showing the number of chars in passwords, that drastically reduces the search space for brute-forcing. [01:55] and now you know why most stuff I do is CLI [01:55] including logging in on this system :P [01:55] * TheLordOfTime manually starts X [01:55] well, that, and it often sucks less. :) [01:56] true statement! [01:56] hehe, I miss inittab's :5: vs :3: configuration... It's been long enough that I even forget the details. :( === n0ts is now known as n0ts_off [03:43] hey, I just edited my /etc/network/interfaces to have a static ip for eth0, now I want to change my default dns server, but it says to not manually edit /etc/resolv.conf - how do I do this in ubuntu? [03:48] Hello Everyone, on 12.04 TLS server looking for this package libffi5 ? [03:48] ruby1.9.2 : Depends: libffi5 but it is not installable [03:48] ant help thank you [03:49] any === n0ts_off is now known as n0ts === cpg|away is now known as cpg [06:21] when I log into my ubuntu server and do man (less is throwing the error) [06:21] I am getting "WARNING: terminal is not fully functional" === yofel_ is now known as yofel === philballew_ is now known as philballew [08:31] New bug: #1075456 in dovecot (main) "wrong path given in README" [Undecided,New] https://launchpad.net/bugs/1075456 === mcclurmc_away is now known as mcclurmc [09:47] <_ruben> hm .. ipv6 address of nl.archive.ubuntu.com appears unreachable [09:49] <_ruben> and it's back [09:51] New bug: #1075484 in mysql-5.1 (main) "upgrade of mysql-server-5.1 to 5.1.66-0ubuntu0.10.04.1 removed InnoDB datafiles" [Undecided,New] https://launchpad.net/bugs/1075484 [10:09] hey, does anyone know how to set up mail forwarding with dovecot/postfix/mysql? === n0ts is now known as n0ts_off === Malediction_ is now known as Malediction [11:53] hi, I have a question about lxc [11:54] is there a way to make a specific block device available inside the container? [11:54] or should I create it manually, with mknod, or perhaps using a loop device attached to a file i create locally? [11:54] with manually, I actually meant manually and internally [11:54] inside the container [11:56] I see the config has lxc.cgroup.devices.deny = a [11:56] and then allows apparently all block devices [11:56] lxc.cgroup.devices.allow = b *:* m === n0ts_off is now known as n0ts [11:56] but there is this comment right above that: # Allow any mknod (but not using the node) [11:56] "not using the node", I don't understand that bit === coldbyte is now known as sdfwsfas === sdfwsfas is now known as coldbyte === cpg is now known as cpg|away === n0ts is now known as n0ts_off === Ursinha_ is now known as Ursinha === n0ts_off is now known as n0ts === kInOzAwA2 is now known as kInOzAwA === n0ts is now known as n0ts_off === gary_poster|away is now known as gary_poster [13:37] I installed maas on quantal, and I get a timeout acquiring lock in errors.log when trying to access the website [13:38] stgraber: libvirt is adding its own fuse /proc/mem support for lxc. (duplication everywhere) [13:38] if i'm reading this right, they're actually implementing the fuse bits themselves [13:43] ok my problem was avahi-daemon not running [13:45] ahasenack: yeah you would have to update your cgroups with the block major/minor and then use lxc-attach in the container (i think)( [13:46] btw, is maas suited as a general cobbler replacement, where you can modify the preseeding and whatnot [13:46] *generic, duh === dendrobates is now known as dendro-afk [13:49] zul: lxc-attach? didn't know about that one, for now I'm just passing "dm-12" to the openstack.cfg that the charm uses, and closely watching to see if that index changes during reboots. There may be a better way, still experimenting [13:49] no, not lxc-attach [13:50] i was using a pre-mount hook to set that up, one sec [13:50] ahasenack: zul: i did http://s3hh.wordpress.com/2012/10/22/easily-making-a-blockdev-available-to-a-container/ [13:51] ahasenack: zul: stgraber is going to implement a simpler lxc- command to do it [13:51] (this cycle) [13:51] hallyn: cool...still doesnt help libvirt thoguh ;) [13:51] d'oh [13:51] missed that bit === n0ts_off is now known as n0ts [13:51] you said 'lxc-attach' :) [13:52] it's cool, I'm experimenting [13:52] going step by step, see where it fails each time, and improving [13:53] hallyn: the hook, that's in the lxc config, right [13:53] lxc.hook.pre-mount [13:53] among many others [13:56] ahasenack: yes, but again that's not with libvirt [13:56] libvirt has it's own device attach stuff [13:56] ok [13:57] still, those hooks might prove useful [14:08] hallyn: yay, more duplication! (especially as we already have the code for the fuse filesystem...) === lynxman_ is now known as lynxman === JamesJRH_ is now known as JamesJRH [14:21] stgraber: yup. === kInOzAwA is now known as Guest32197 [14:53] Hi! [14:54] I'm trying to setup nagios3 on Ubuntu 12.10, but can't get it to work due missing "generic-switch" template. I placed the switch.cfg in "/etc/nagios3/objects" and i followed this guide... [14:54] http://nagios.sourceforge.net/docs/3_0/monitoring-routers.html === n0ts is now known as n0ts_off [15:23] hello, i run 10.04 as a guest on virtualbox. How would i do a fsck without using a live cd? [15:23] running headless === Guest32197 is now known as kInOzAwA [15:25] vagy, reboot? [15:27] patdk-wk: i can't see or interfere with virtualbox while booting this vm (i run it on a headless 10.04 as well) [15:28] so a headless 10.04 is host to a headless 10.04 guest [15:28] is there some obvious way for me to determine if I'm running LTS (or what ever) on some random server out of a hundred to which I connect? [15:29] well, I can't help you with virtualbox [15:30] bc, cat /etc/lsb-release [15:30] patdk-wk: thank you; lsb_release -sd [15:30] that too [15:31] patdk-wk: just found the page when you replied. :) [15:31] New bug: #1075589 in clamav (main) "package clamav-freshclam 0.97.6+dfsg-1ubuntu0.12.04.1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 128 (dup-of: 1074708)" [Undecided,New] https://launchpad.net/bugs/1075589 [15:31] smoser: any chance of having this MR reviewed : https://code.launchpad.net/~louis-bouchard/ubuntu/oneiric/grub2/grub2-lp669481-oneiric/+merge/131625 [15:31] SpamapS, alive? [15:32] smoser: I thought I had assigned it to you but doesn't look like it [15:35] i'll look at that today. [15:35] thank you [15:42] New bug: #1064748 in quantum "a port which is not able to delete is made when floatingip create fails." [Undecided,Confirmed] https://launchpad.net/bugs/1064748 [15:42] New bug: #1069966 in quantum "rootwrap error with L3 agent" [Undecided,Confirmed] https://launchpad.net/bugs/1069966 [15:42] New bug: #1070335 in quantum "l3 db operations failed in multiple transactions" [Undecided,Confirmed] https://launchpad.net/bugs/1070335 [15:42] New bug: #1072713 in quantum "Linux bridge port update causes exception" [Undecided,Confirmed] https://launchpad.net/bugs/1072713 [15:43] New bug: #1051744 in quantum "remove default value of 'local_ip' of 10.0.0.3 in ovs_quantum_plugin.ini " [Undecided,Confirmed] https://launchpad.net/bugs/1051744 [15:43] smoser: thanks [15:43] New bug: #1053364 in nova/folsom "Add SIGPIPE handler to subprocess execution in rootwrap and utils.execute" [Medium,In progress] https://launchpad.net/bugs/1053364 [15:43] New bug: #1056720 in quantum "L3 agent external network flag" [Undecided,Confirmed] https://launchpad.net/bugs/1056720 [15:43] New bug: #1057558 in quantum "I can't add interface to router, if there is another port in non-shared network of other tenant" [Undecided,Confirmed] https://launchpad.net/bugs/1057558 [15:43] New bug: #1059393 in quantum "Ryu plugin does not support Security Groups" [Undecided,Confirmed] https://launchpad.net/bugs/1059393 [15:43] New bug: #1060559 in quantum "l3_agent not disabling namespace use" [Undecided,Confirmed] https://launchpad.net/bugs/1060559 [15:43] New bug: #1061391 in quantum "nvp_plugin does not update device_id in nvplib" [Undecided,Confirmed] https://launchpad.net/bugs/1061391 [15:43] New bug: #1061781 in quantum "get_network in nvp plugin didn't return subnet information" [Undecided,Confirmed] https://launchpad.net/bugs/1061781 [15:43] New bug: #1064070 in quantum "tap device added to integration bridge without tag" [Undecided,Confirmed] https://launchpad.net/bugs/1064070 [15:43] New bug: #1064235 in quantum "router create with external_gateway_info fails with 500 always." [Undecided,Confirmed] https://launchpad.net/bugs/1064235 [15:43] New bug: #1064261 in quantum "Port create with invalid IP data creates port." [Undecided,Confirmed] https://launchpad.net/bugs/1064261 [15:45] TheLordOfTime: just now online. Wassup? [15:46] SpamapS, just some minor bugchecking, need someone to check my research on the php5 package. [15:46] SpamapS, https://bugs.launchpad.net/ubuntu/+source/php5/+bug/723480 <--- I believe this was fixed in 5.3.6-11, there's a debian changelog suggesting that [15:46] Launchpad bug 723480 in php5 "php5-fpm uses too high value for pm.max_children by default" [Low,Confirmed] [15:47] wanted someone to check the research before i mark that as "Fixed" and quoting the changelog [15:48] (Oneiric has that fix afaict, but Lucid does not... not sure that matters) [15:48] (if it does, that needs a Lucid SRU) [15:49] I'm trying to setup nagios3 on Ubuntu 12.10, but can't get it to work due missing "generic-switch" template. I placed the switch.cfg in "/etc/nagios3/objects" and i followed this guide... [15:49] (15:54:32) halvors: http://nagios.sourceforge.net/docs/3_0/monitoring-routers.html === mcclurmc is now known as mcclurmc_away [15:54] TheLordOfTime: well perhaps you can share the changelog bit you intend to quote? [15:54] lemme pastebin [15:54] zul: no utlemming to be seen. he's next on the schedule to chair mtg. which i assume happens in 5 mins? [15:55] *sigh* [15:55] SpamapS, this'd be my closure message: http://pastebin.com/gYwgMKGc [15:55] i thought he was around [15:55] maybe ttx would like to run the meeting :) [15:55] is the server team meeting in 5 minutes? ;P [15:55] s/;P// [15:56] (apparently that was leftover from an incomplete message) [15:56] i thin kit's in 5 mins [15:56] 1600 utc [15:56] that should be in about 4 minutes, unless i'm off by an hour [15:57] * TheLordOfTime thinks this through: Eastern US standard time = UTC-0500, 1600 - 0500 = 11:00... [15:57] polom [15:57] yep, 3 minutes if time servs me right. [15:58] SpamapS, i did check, Lucid won't have that fix, though. is php5 still "updated" for lucid? [15:58] i.e. are SRUs still accepted for it. [16:16] smb, you are needed in #ubuntu-meeting [16:18] rbasak: #ubuntu-meeting please [16:18] ^ that [16:18] TheLordOfTime: Error: "that" is not a valid command. [16:18] STOP NINJAING ME ZUL! [16:18] :P === dendro-afk is now known as dendrobates [16:21] zul++ [16:22] * TheLordOfTime casts "Slow" on zul. zul loses 50% speed. [16:23] okay, enough with my offtopicness. [16:23] * TheLordOfTime pokes SpamapS to make sure SpamapS hasn't died. === mcclurmc_away is now known as mcclurmc [16:29] SpamapS, did you check that pastebin? [16:30] TheLordOfTime: yes, sorry distracted.. [16:30] no problem :) [16:31] TheLordOfTime: re the second question, yes, lucid SRU's are accepted, though only for high/critical bugs [16:32] ah, so this would not need lucid fixing :P [16:32] TheLordOfTime: Lucid server packages are supported for 2.5 more years [16:35] SpamapS, if only high/crit bugs are being fixed in Lucid, then excluding that all currently supported releases have the fix for that bug. [16:35] TheLordOfTime: so the default appears to be 5, instead of 50, so, it does seem like thats Fixed [16:35] i'll mark that :) [16:36] SpamapS, that's the conclusion I came to as well, but i wasnt sure about whether it needed fixing in Lucid first :p [16:36] hence asking for a research-check ;) [16:37] TheLordOfTime: we never fix in old releases first [16:37] TheLordOfTime: dev release first, then backport to wherever it might make sense [16:37] SpamapS, this had already been fixed in dev and the other releases [16:37] in *this* circumstance that's the case [16:37] TheLordOfTime: thats almost always the case.. that its fixed in newer releases first [16:37] SpamapS, i've done a few SRUs before, i'm aware it gets fixed release -> older in reverse-chrono order. [16:37] s/release/devRelease/ [16:38] the question was whether lucid factored into this one bug [16:38] yeah ok we're just talking about nothing. +1 go forth, fix released that bug, and continue your awesome help with PHP :) [16:38] yep. [16:38] this was me just sifting through old bugs that were likely fixed :P [16:39] * TheLordOfTime yawns as he realizes his local builders failed again [16:54] hallyn: hey, that blog post of yours (http://s3hh.wordpress.com/2012/10/22/easily-making-a-blockdev-available-to-a-container/), it's missing a link to the actual script [16:54] hallyn: "So I use this script to set a container up to use a particular block device." <-- "this script" was meant to be a link probably [16:54] it changes color when I hover over it :) [16:54] me mess up a blog posting? impossible! [16:55] (j/k i hate blogging) [16:57] ahasenack: thanks! updated [16:57] hallyn: thanks! :) [17:04] hallyn: you blog great [17:04] aw shucks, thanks [17:20] stgraber: if you get bored at some point, my initial notes, in preparation for an email to the kernel team (or for a patch :) for a syslog ns design are at https://wiki.ubuntu.com/LxcSyslogNs [17:20] smb: ^ [17:27] New bug: #1031119 in nova "nova: proxy floating ip calls to quantum" [Medium,Fix committed] https://launchpad.net/bugs/1031119 [17:27] hallyn: nice. Looks good. Only quickly read through but do you expect anything to be needed to make attaching to the syslog ns to work? [17:28] hallyn: thinking of the case where you start a daemon using lxc-attach, attaching to all namespaces from the outside, if that daemon uses the log buffer, it'd be nice to have that land in the right syslog ns [17:32] stgraber: hm, trying to parse that after thinking about cgroups, one sec... [17:32] stgraber: in that case, which would you call the right syslog ns? [17:33] i can definately say this much: netfilter printks would go to the container syslog, [17:33] bc it would have lxc-attached to the netns, and netns would be 'owend' by the container syslogns [17:33] what i'm not sure about is whether we also have a syslogns pointer in nsproxy. that depends on whether we need it, or whether we can get what we need with pointers from other namespaces to the syslogns [17:34] i should probably go to lkml with this soon, bc this design could easily be rejected. the inter-relations between namespaces becomes ever more complex. [17:34] in fact, as i type, i wonder if it's better to restrict and simplify it - so only a user namespace can point to a syslog ns [17:34] as it is, every ns and task is owned by a user ns. this would keep the relations impler [17:35] it would also mean that a container without its own userns could not have its own syslogns [17:35] (i'm done :) [18:06] hallyn: those patches work for me [18:07] smoser: around ? [18:07] here [18:08] smoser: hey. So I was asking about the cloud-init-nonet upstart job. Specifically I've seen references to it in a bug, but its not in the cloud-image filesystem tarball, so I'm a little confused about *where* it can be found. [18:10] lifeless, [18:10] ubuntu@quantal$ dpkg -S /etc/init/cloud-init-nonet.conf [18:10] cloud-init: /etc/init/cloud-init-nonet.conf [18:10] ubuntu@quantal$ dpkg-query --show cloud-init [18:10] cloud-init 0.7.0-0ubuntu2 [18:10] it should be in precise also. and maybe even oneiric. [18:11] smoser: *blink*. Clearly I fail. [18:11] smoser: thank you. [18:11] no problem. :) [18:12] massive problem for my geek cred... but I can deal :) [18:12] zul: cool === schmidtm_ is now known as schmidtm [18:36] is there something new in lxc in quantal, related to apparmor? [18:36] I can't even apt-get install lxc inside an lxc container [18:37] apparmor_parser: Unable to replace "/usr/bin/lxc-start". Permission denied; attempted to load a profile while confined? [18:37] I don't know when it was introduced, but yes, lxc containers are contained with apparmor [18:39] I see there are two lxc profiles at least [18:39] /etc/apparmor.d/lxc/lxc-default-with-nesting and /etc/apparmor.d/lxc/lxc-default [18:39] the -with-nesting one seems the right one, as it also allows mounting of cgroup [18:39] not sure how either one is selected, or if I have to do something [18:39] * ahasenack debugs [18:47] I'm also getting apparmor errors when trying to mount cgroup inside lxc [18:47] so I guess nested lxc doesn't work [18:47] but I think it used to, I vaguely remember it working in precise [18:47] but I could be wrong [18:48] [15799.661604] type=1400 audit(1352226887.742:122): apparmor="DENIED" operation="mount" info="failed type match" error=-13 parent=31318 profile="lxc-container-default" name="/sys/fs/cgroup/cpuset/" pid=31326 comm="mount" fstype="cgroup" srcname="cgroup" flags="ro" [18:48] not using the lxc-default-with-nesting profile then [19:12] New bug: #1066115 in cloud-init "The landscape client plugin doesn't create the config file with the proper permissions" [Undecided,New] https://launchpad.net/bugs/1066115 === mcclurmc is now known as mcclurmc_away [19:46] Hello, I'm trying to install phantom.js on remote ubuntu 10.04 x64 server. I'm accessing it using gitbash from my pc. I did download the .bz2 file, using "wget". how do I install it now? what should be done? do I need to "run" it ? thanks. [19:54] danieltlx: probably just scp it to your server, bunzip2 the file, put it in place in your web hierarchy, chown it to the correct owner, and chmod to permissions that forbid writing [20:01] New bug: #1064771 in ubuntu-se-server "Frigöra diskutrymme på adder" [Medium,Confirmed] https://launchpad.net/bugs/1064771 [20:11] New bug: #1070345 in cloud-init "need to restart landscape after updating config" [Medium,Fix committed] https://launchpad.net/bugs/1070345 === cpg|away is now known as cpg [21:11] New bug: #1075736 in lxc (universe) ""lxc-create -u ubuntu" fails if /var/cache/lxc too old?" [Undecided,New] https://launchpad.net/bugs/1075736 === cpg is now known as cpg|away === cpg|away is now known as cpg === mcclurmc_away is now known as mcclurmc [21:58] Hey, can anyone tell me if it's possible to run a maas cluster without any "nodes" [21:58] like, add the master server as a node? [21:59] what are you actually trying to achieve? [22:00] I guess I was thinking I'd use maas as an easy way to manage vms, is that not quite what it is? [22:00] no, it's not, maas manages real hardware [22:01] you can set up VMs as nodes but it's mainly for developer testing [22:02] makes sense, sorry for the dumb question then :) [22:02] that does still sound pretty useful so maybe I'll leave it installed on this server === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha