[00:10] I've put a series of bonding configurations in my /etc/rc.local. What I'm noticing is that it seems only the first line of them are executed. [00:11] what might be causing rc.local to short [00:18] I know Synaptic has a menu choice to force reinstall of a package. Where is the equivalent in aptitude? [00:18] Or must I force reinstall via dpkg? [00:20] mdlueck, apt-get install --reinstall [00:21] don't know what it would be for aptitude but it should have an option like it in the manual [00:21] qman__, yes that would work too. No wat via aptitude? [00:21] then I will just apt-get, thanks [00:28] stgraber: all good reasons [00:28] Daviey: yep [00:42] Hey [00:45] I would like to setup a new Xen guest with 12.04 ... apparently this is not easily possible using the dist option [00:45] how would you recommend to do it? [01:02] mklappstuhl, easy, install 12.04 with the ec2 kernel [01:03] same way it's been done since 10.04 [01:06] patdk-lap: I am not using ec2 but it reads like this is not a requirement [01:07] what are you talking about? you asked for xen, the answer is ec2 [01:07] patdk-lap: can you point me to some more information? [01:07] sure, the kernel-ec2 config file [01:08] apt-get show linux-ec2? === Lcawte is now known as Lcawte|Away [01:12] and I dont understand how ec2 is related to the problem and also there is now package as kernel-ec2 or linux-ec2 [01:13] hmm [01:13] your failing is that ubuntu doesn't maintain a xen base image [01:13] the next failing is that, xen guest stuff was merged into the kernel [01:13] and the 3rd is that, ec2 is based on xen [01:14] therefor, to have a xen compatable image, ec2 is the 1 stop shop [01:14] who cares if you use amazon or not, you want xen, install ec2 kernel [01:14] I know that ec2 builds up on top of xen but [01:15] I can install ubuntu versions on my hypervisor .. it works fine with debian and lucid and many others [01:16] if your using a hypervisor, then you don't need xen support, cause your using qemu [01:16] in my understanding the hypervisor is the Dom0 [01:16] And I am also not looking for "xen-support" [01:17] for paravirtual [01:17] for real virtual machines, hvm mode, it uses qemu [01:17] I want to install 12.04 as easy as the other debian based distributions by using the --dist flag [01:17] what --dist flag? [01:17] * patdk-lap doesn't remember any --dist flag [01:17] but there are not instructions shipped for the version I am running xen and probably that will take a while since ubuntu 12.04 is not even released [01:18] xen-create-image --hostname bla --dist lucid [01:18] something like the abote [01:18] oh, xen-tools, that isn't a part of xen :) [01:20] haha okay [01:20] but it has xen in its name :D === juicy is now known as _- [03:03] Hi [03:03] Thoughts on Webin for ubuntu [03:03] Webmin** [03:03] ? [03:04] !webmin [03:04] webmin is no longer supported in Debian and Ubuntu. It is not compatible with the way that Ubuntu packages handle configuration files, and is likely to cause unexpected issues with your system. [03:07] Really? [03:08] !webmin [03:08] webmin is no longer supported in Debian and Ubuntu. It is not compatible with the way that Ubuntu packages handle configuration files, and is likely to cause unexpected issues with your system. [03:08] Very interesting [03:08] any ideas why this is the case/ [03:08] I'm having some issues managing my web sevrer [03:08] any other graphical front ends [03:08] ? [03:08] make my life easier? [03:13] Webmin is a lame web-based interface for unsafe system administration for Unix. Check it out at http://webmin.com/ Remember, dondelelcaro *hates* webmin. "i'd rather sit on the floor shoving table knives into live electrical outlets than run webmin on an exposed server." Removed from Debian post-Sarge, see http://bugs.debian.org/343897 . The Debian package from webmin.com is of poor quality. See cp> for alternatives. [03:14] ^^ that's the Debian version of that info entry. [03:14] twb: Error: "^" is not a valid command. [03:14] You can /msg dpkg to ask about it if you are in #debian [03:17] Perl web host control panel? [03:17] I am not interesting in discussing it with you further. [03:17] *interested [03:17] :( [03:17] how come? [03:18] I guess the commad line way is the best. [03:18] Thanks [04:32] Why does /lib/udev/devices need a loop0 ? [04:32] It is pissing me off because when copied to a solaris/ZFS backup server, I get this: [04:32] $ du --apparent-size /tank/hosted-backup/backups/cybersource/alpha.cyber.com.au:./.zfs/snapshot/2012-03-28T14:05:01Z/lib/udev/devices/loop0 [04:32] 9007199254740992 /tank/hosted-backup/backups/cybersource/alpha.cyber.com.au:./.zfs/snapshot/2012-03-28T14:05:01Z/lib/udev/devices/loop0 [04:34] I was just going to instruct the backup to be --one-file-system, but it turns out /lib/udev/devices isn't (as I had assumed) a separate pseudofs === _- is now known as juicy === dcramer_ is now known as dcramer [09:22] hi all [10:34] morning o/ [10:41] New bug: #980682 in bind9 (main) "package bind9 1:9.8.1.dfsg.P1-3 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/980682 [11:05] If I go to http://mysite.com, my browser downloads the index.php file instead of displaying it. However if I go to http://mysite.com/index.php, it works correctly. What causes that problem? [11:21] New bug: #980712 in samba (main) "Package winbind 2:3.6.3-2ubuntu1 failed to install/upgrade: ErrorMessage: subprocess new pre-removal script returned error exit status 143" [Undecided,New] https://launchpad.net/bugs/980712 === dendrobates is now known as dendro-afk [11:55] question about lstp server, i notice that the next version of zentyal (aka zentyal 3.0) will be providing a very nice lstp configuration module for lstp.. so my question is this : 20 workstations running a gnome shell desktop with thunderbird and chrome, libre office etc does the actual video rendering processing happen on the workstation or the server ? [11:56] i mean LTSP === koolhead11 is now known as koolhead17 [12:31] New bug: #980758 in samba (main) "new buffer overflow attack on samba 3.6.3 -> enables unauthenticated remote root access" [Undecided,Invalid] https://launchpad.net/bugs/980758 [12:32] zul, around [12:33] koolhead17: no [12:33] zul, hehe. :) [12:35] * koolhead17 wonders if there has been some major changes in dash yesterday [12:35] dash == horizon [12:50] koolhead17: why? [12:51] zul, because some one told me upgade asks you for DEBUG = False [12:52] and new package python-django-horizon [12:52] koolhead17: ok yes the configuration has changed to turn off debug [12:52] zul, and that breaks the css apparently which it should not. [12:52] what? [12:52] can you please open up a bug then? [12:54] zul, even in my un-upgraded horizon pkg if am doing DEBUG = False it breaks [12:54] koolhead17: ok can you please open up a bug then [12:55] am doing it right away [12:55] thanks === bladernr_afk is now known as bladernr_ [13:02] koolhead17: if you set it back to True then it works like normal right? [13:02] yes [13:03] zul, https://bugs.launchpad.net/horizon/+bug/980787 [13:03] Launchpad bug 980787 in horizon "openstack-dashboard breaks with option DEBUG =False in config file" [Undecided,New] [13:31] New bug: #980798 in bind9 (main) "incorrect owner '644' for /etc/bind/named.conf.options" [Undecided,New] https://launchpad.net/bugs/980798 === dendro-afk is now known as dendrobates [14:15] utlemming: for bug 979996, does your fix still work with precise? doesn't look like it should, becuase you unconditionally switch to $release-*, but in precise it's still ubuntu-12.04* ? [14:15] Launchpad bug 979996 in lxc "ubuntu-cloud template can't find .img" [High,Confirmed] https://launchpad.net/bugs/979996 [14:16] hallyn: I tested it...but let me double check again to be sure. [14:17] Daviey: the horizon i did this morning fixes a regression from the last upload [14:20] utlemming: while I'm converting lxc-test to pyunit, I'm gogn to add ubuntu-cloud to the tests [14:20] hallyn: nice [14:21] just playing whack-a-bug with my typos right now :) [14:26] hallyn: I just re-confimred the patch works with precise released (beta images) and re-confirmed it works with dailies. [14:30] zul, the fresh upload of horizon add deps of python-memcache and memcahed right ? [14:30] *adds [14:30] no it fixes the regression [14:30] koolhead17: ill get the memcached in for an SRU [14:31] zul, it be cool to have CACHE_BACKEND = 'memcached://127.0.0.1:11211/' [14:31] line added inside vim /etc/openstack-dashboard/local_settings.py [14:31] koolhead17: yes i know [14:31] :) [14:41] utlemming: so just 'lxc-create -t ubuntu-cloud -n c1' works? === sergevn is now known as zz_sergevn === zz_sergevn is now known as sergevn === bladernr_ is now known as bladernr_afk === bladernr_afk is now known as bladernr_ === Lcawte|Away is now known as Lcawte [15:20] NAMES #ubuntu-devel [15:20] /NAMES #ubuntu-devel [15:21] /N #ubuntu-devel [15:21] /NAMES #ubuntu-devel [15:21] New bug: #980902 in lxc (universe) "lxc-destroy -f needs to lxc-wait, or use lxc-shutdown" [Low,Confirmed] https://launchpad.net/bugs/980902 [15:21] New bug: #980905 in lxc (universe) "lxc-shutdown -h calls undefined 'help' function" [Low,Confirmed] https://launchpad.net/bugs/980905 [15:21] hm [15:21] oh, sorry for the noise, didn't realize it caches the last channel I wrote to :) [15:22] I'm having problems with an /etc/rcX.d startup script ... it's not starting the service , altho I followed the readme instructions for invoking the script at startup. Is the /etc/rcX.d file the right place or have things changed? [15:23] BluesKaj: Did you use the update-rc.d script? [15:23] hey genii-around , I did update-rc.d [15:26] BluesKaj: upstart has the /etc/init/rc-sysinit.conf which should still start any old stuff which is still using the init.d way. Is it giving any error message or just not starting at all? [15:26] jamespage: I'm getting jenkins.JenkinsException: Error in request. Possibly authentication failed [403] using python-jenkins on an internal jenkins instance with open access. Do you happen to know what this might be? Do I need to create users with permissions? [15:26] New bug: #980910 in lxc (universe) "Merge 0.8.0~rc1-4 from debian" [Low,Confirmed] https://launchpad.net/bugs/980910 [15:27] rbasak, auth in jenkins is a bit of a back art [15:27] do you have users/perms setup? [15:27] jamespage: no, none at all [15:27] genii-around, it's the no-ip dns app which tracks the dynamic IP automatically ...no error message , but it tries to start kate and fails , then freezes the desktop [15:27] rbasak, are you passing any using python-jenkins? [15:27] I'm passing None, None [15:27] But wait [15:28] Looks like others have created some users [15:30] jamespage: I have three people but no sign of any passwords or anywhere to set them up === matsubara is now known as matsubara-lunch [15:52] jamespage: where did you say the jenkins PPA was? I'm going to try and upgrade from the oneiric version. [15:53] rbasak, ppa:hudson-ubuntu/backports [15:53] thanks! [15:53] sorry - just been debugging something else [15:53] BTW 403's are a guess when it comes to authentication - it might actually be a different issue [15:54] hmm [15:55] I think I needed to update anyway - the build publisher plugin wasn't working and I'll need that [15:55] New bug: #980930 in nova (main) "nova client does not respect regions" [Undecided,New] https://launchpad.net/bugs/980930 [15:59] When you do "sudo -i" which pam modules are being looked? [16:06] SpamapS: what day are you coming to UDS? [16:06] New bug: #980940 in euca2ools (main) "euca2ools does not respect the --region option correctly" [Undecided,New] https://launchpad.net/bugs/980940 [16:06] SpamapS: we need to make sure the mysql sessions are on the day you are there. [16:15] morning! [16:19] jcastro: Tuesday for sure [16:20] jcastro: thats the day I registered for [16:20] jcastro: I will also try to make it for Thursday === jkyle_ is now known as jkyl === jkyl is now known as jkyle [16:20] SpamapS: ok and you told launchpad you can only be there on tuesday and mark yourself required? [16:21] (on the mysql blueprint?) [16:27] jcastro: yes [16:27] SpamapS: have you submitted the blueprint for approval? [16:27] few questions on maas. does the maas server just tie into cobbler? can I manage my nodes via cobbler and they'll register with maas? [16:27] jcastro: bug report for launchpad.. need to be able to specify non-contiguous days ;) [16:29] SpamapS: yeah, I only care about the blueprint right now, I need to manually schedule it anyway, is it submitted for approval or still drafting? [16:29] roaksoax: ^^ do you know the answer to that? [16:29] roaksoax: re jkyle's maas question [16:29] jkyle: That wouldn't really make much sense.. ;) [16:30] jkyle: cobbler mostly is just used for pxe/pre-seed .. [16:33] SpamapS: right, but I was thinking maybe maas got the list of nodes from cobbler's list. currently deleting nodes is bugged [16:33] https://bugs.launchpad.net/ubuntu/+source/maas/+bug/978706?comments=all [16:33] Launchpad bug 978706 in maas "Internal Server Error when deleting a node (interface needs to be defined)" [Undecided,Confirmed] [16:33] so, if deleting them with cobbler was registered by maas...it'd be a work around :P [16:34] SpamapS: ok it looks like it's on a tuesday: http://summit.ubuntu.com/uds-q/2012-05-08/display [16:34] SpamapS: ok good, I just wanted to make sure we weren't scheduling it when you were not there, etc. [16:40] phew, debootstraps seem to be failing bc hte archive is updating so fast right now (or something) [16:43] ok, I have a maas setup. I entered the mac addresses of a couple of nodes, booted them with pxe enabled. dnsmasq detects them and offers an ip on the right range. But the node never pxe boots [16:44] jkyle: nope you cant [16:45] jkyle: (manage nodes in cobbler) [16:45] jkyle: and do you have several nic's on the MAAS server? [16:46] roaksoax: yes, 4 nic's. 2x1gb, 2x10gb with both pairs bonded (one mode 4, one mode 1) [16:46] then, each of the bonds have multiple vlan trunks [16:47] that's the maas node [16:47] jkyle: ok, so I'm guessing the interface that connects to the external network, or has the default route is not the interface you use for PXE booting? [16:47] the clients currently run one bonded pair with no trunking, and the 10gb pair has two vlan's [16:48] roaksoax: correct. I noted the default conf is listening on all nics. I'll tweak that later (currently securing it via iptable rules) [16:48] jkyle: ok so the problme might be that MAAS has autodetected the default IP address to be one different from the one you are PXE'ing from [16:49] gotcha [16:50] jkyle: so I just filed related bug [16:50] jkyle: but to fix, please sudo dpkg-reconfigure cobbler [16:50] that would be here? /etc/maas/maas_local_settings.py [16:51] jkyle: and set the correct IP address for cobbler Boot and PXE Server IP address [16:51] jkyle: and yes, also set the correct IP address for DEFAULT_MAAS_URL [16:51] k, I'm very comfortable with dnsmasq, pxe, etc. but wasn't familiar with cobbler [16:52] I currently have it on a public ip...reconfigured for https..for convenience [16:52] does the DEFAULT_MAAS_URL have to match the address of the primary management ip? [16:52] jkyle: right, but if you sudo dpkg-reconfigure cobbler, it will first prompt you to enter cobbler's password, you can leave that blank [16:52] jkyle: but it will ask you for the IP address for Boot and PXE. This has to be the IP address of the nic on the MAAS srver that the clients can access to for PXE booting [16:53] you were correct, it was listening on teh public ip [16:53] jkyle: and DEFAULT_MAAS_URL can be any IP address of the MAAS server *as long as* the clients can access it [16:53] jkyle: obviously it is recommended it to be the samve as above [16:53] k, seems practical that this should be my management ip than [16:54] jkyle: exaclty, so, I'll file a bug and fix it so we can easily change that with sudo dpkg-reconfigure maas [16:55] now, teh DEFAULT_MAAS_URL seems to be unrelated to the web ui url that apache serves [16:55] Hi, just wondering how I should manage something on my server... I need a user to be able to run a shell script which moves a folder into /var/. However, folder doesn't have the permissions for this by default... Is the only option to basically remove the write protection on that directory? [16:56] roaksoax: I wonder if this misconfiguration is related to the bug I linked...I'll test and see :P [16:57] jkyle: aesome, you might wanna run sudo cobbler sync to update the PXE address after changing it with dpkg-reconfigure cobbler === matsubara-lunch is now known as matsubara [16:58] ok, so far. same behavior [16:58] I restarted the maas-pserv too [16:59] pxe still timing out though [17:01] jkyle: sudo cobbler sync [17:02] I did, lemme do it again [17:03] * jkyle cycles server [17:04] hggdh: do you still experience bug #977609 [17:04] Launchpad bug 977609 in maas "MAAS barfs with 'Internal server error' when clicking on Settings" [Undecided,New] https://launchpad.net/bugs/977609 [17:05] roaksoax: yes, still happens [17:05] yeah, still not serving pxe images on that interface [17:06] jkyle: can you pastebin /var/lib/tftpboot/pxelinux.cfg/default ? [17:07] I verified the /etc/cobbler/settings:next_server value as the ip of the interface I'm listening on [17:07] roaksoax: sure, sec [17:09] roaksoax: https://gist.github.com/2378412 [17:09] log host is set to the public ip, probably not good. the url= arg is st to the proper interface though [17:10] * jkyle wonders where its getting the log host value..it doesn't appear in /etc/cobbler/ [17:10] jkyle: if url=...192.168.112.10.. is set correctly, then there maybe something else [17:11] that's the correct ip. and dnsmasq is serving the specified mac address an ip [17:12] jkyle: can you check what's in /etc/cobbler/settings in ^server: and ^next_server: [17:12] jkyle: and you should be able to update it by running: sudo maas-import-isos --update-settings [17:12] next_server points to the 192.168.112.10 addy [17:13] server: points to the 192.168.112.10 addy [17:13] jkyle: ok so run "sudo maas-import-isos --update-settings" [17:13] jkyle: and check that the default file provides the correct IP address [17:14] jkyle: and check if there';s any dnsmasq output when you runn it (as in started successfully) [17:15] the default looks good on teh url's..though its log_host is off on a couple of menu items. been watching the dnsmasq logs. nothing looks bad there. it serves up the ip on request [17:17] hm, maybe it's a firewall issue. our rules were set up for a more vanilla environment [17:18] maybe, cause if it can get the PXE file, might be DHCP issue [17:19] dropping firewall for a bit [17:21] nope, that's not it. same behavior when wide open [17:21] jkyle: can you psatebin the output of sudo cobbler sync? [17:22] sure [17:41] New bug: #975655 in glance "Migration 012_id_to_uuid attempts to convert IDs twice for non-sqlite databases" [Undecided,Fix committed] https://launchpad.net/bugs/975655 [17:41] Question - 11.04, DHCP, I run sudo /etc/init.d/dhcp3-server start, NO error, yet its still - when I run service --status-all === Lcawte is now known as Lcawte|Away [18:17] roasted: what do the log say? [18:17] *logs === dendrobates is now known as dendro-afk [18:20] guntbert: nothing [18:20] literally, no errors for dhcp [18:20] it may be windows related [18:21] I was just told the server was part of the domain via likewise-open. a controller went down and they shuffled things around, and magically ubuntu stopped working [18:21] far too werid to be inconvenient timing === dendro-afk is now known as dendrobates [18:51] hallyn, the workaround to bug 974584 (rmdir /dev/shm from existing instances) is not working for us (same symptoms, ENOSYS). I looked at the description of your fix in the upcoming package and made a symlink (rootfs/dev/shm -> /run/shm), but still no fix. What is the proper workaround? the mount worked, but I'd like to verify that the new approach also works for us. [18:51] Launchpad bug 974584 in sysvinit "Semaphores cannot be created in lxc container" [Undecided,Confirmed] https://launchpad.net/bugs/974584 [18:51] gary_poster: in precise contaienrs? [18:52] hallyn, lucid [18:52] gary_poster: i should comment in the bug [18:52] in lucid, the fix is to add dev/shm entry to the contaienr's fstab [18:52] hold on lemme comment [18:52] hallyn, cool, thanks === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [19:07] so im trying to remove a word off mulitple files.. ex: processed_1234, processed_4567.. I just want to remove the processed_ off all the files [19:07] is this possible? [19:11] No sysbench in repos for hardy, any suggestions on where to get a .deb? [19:20] ok, so the maas stack. What apps provide dhcp & tftp. I'm poking around packages and it looks like dnsmasq for dhcp...but I see the tftpd-hpa package is also installed [19:20] so dnsmasq is doing dhcp only? [19:20] Slammed on the debs from lucid, seems to have done the trick :-) [19:21] jkyle: I believe dnsmasq is a dns / dhcp server, not sure if it has any ftp stuff, but I really don't know [19:21] what I'm seeing when I start cobbler up, is that dnsmasq + cobblerd are started. But tftpd-hpa is not [19:21] chmac: tftp, pxe boot :) [19:21] jkyle: Double dutch to me :-) [19:25] adam_g / zul: I just thought, the horizon css issue with Debug.. there is a convention in Django to serve static media through django app server if Debug = True.. This doesn't scale. What should happen is that /media should be directly served through apache vhost config, bypassing django. Is this the issue? [19:26] sounds like it [19:46] is it possible to shut off a nic so that the link light goes dead on the back of the server? [19:46] I'm trying to identify the cord i do not need plugged in [19:48] nm ethtool has an id tool i can use. thanks [19:50] this bug now says fixed: released => https://bugs.launchpad.net/ubuntu/+source/maas/+bug/978706?comments=all [19:50] Launchpad bug 978706 in maas "Internal Server Error when deleting a node (interface needs to be defined)" [Undecided,Fix released] [19:50] does this mean the package is updated or I need to pull source down? [19:52] Daviey: well spotted. adding an alias to the apache config fixes the issue and lets debug = false. is it worth another upload? [19:52] adam_g: yes [19:52] Daviey: k [19:54] * koolhead17 is famous today!! :) [19:55] koolhead17: trouble maker ! :) [19:55] koolhead17: will you be at ODS? [19:55] Daviey, yes sir. very much. am all ser [19:55] set [19:56] Daviey, i mailed you, seems like your mailbox is full or gmail goes to spam folder [19:56] :P [19:56] koolhead17: When did you send that? I've ben somewhat overwhelmed by email this week. [19:56] roaksoax: ping [19:57] Daviey, i think last week. I thought better mail you rather troubling you here:) [19:57] * koolhead17 checks his mail box to find send date [19:57] *sent [19:58] Daviey, 3ed apr :) [19:58] i suck so bad. [19:59] Daviey, you were busy with release so no issue :D [19:59] i just needed your suggestion 4 head on collision and if its good. :D hehe [20:00] koolhead17: ok, lets catch up next week. [20:00] Daviey, just buzz me whenever you have sometime. We still have around 15 days before UDS :D [20:01] koolhead17: sorry, will you be at *ODS* [20:01] lol. no no. [20:01] UDS [20:01] i will be sir [20:02] ok, cool [20:02] althought i would have loved to b there TBH :( [20:02] so many awesome peeps coming in [20:02] * Daviey blushes. [20:04] Daviey, and espacially that gartner lady i was interested to meet. cloudpundit :P [20:04] she had so much issue with openstack [20:04] hahaha [20:05] jcastro, hey, do you have any docs on how to deploy openstack using juju? [20:06] matsubara, i thought there already was one [20:06] not sure i its on ubuntu server blog page [20:06] adam_g, hi there [20:06] Daviey: horizon_2012.1-0ubuntu5 [20:07] koolhead17: ole [20:07] adam_g, was wondering if we should write a keystone-GUI [20:07] :P [20:07] or keystone browser [20:07] koolhead17: hmm [20:07] too lazy to add the values [20:07] koolhead17: horizon basically is that [20:07] koolhead17: well, for the user at least [20:07] koolhead17: initial config, you mean? of endpoints, services, etc? [20:07] koolhead17, could you point me to it? [20:08] adam_g, horizon comes once keystone gets done with its action :P [20:08] matsubara, 1 sec [20:08] adam_g, yes [20:09] and we can easily see and modify stuff from our browser [20:09] matsubara: i have one [20:09] that way we have complete control on whats happening with keystone [20:09] koolhead17: wouldnt be difficult, the keystoneclient library is easy to use stuff [20:09] adam_g, yes sir. [20:09] matsubara: as if by magic, https://help.ubuntu.com/community/UbuntuCloudInfrastructure :) [20:10] thanks Daviey for that :P [20:10] courtesy of adam_g [20:10] :P [20:10] jkyle: pong [20:10] thanks DavidLevin [20:10] adam_g, and for my understanding i added each tenat/service/user/role one by one [20:10] err, thanks Daviey [20:15] Daviey: disregard that upload, it was rejected? [20:16] New bug: #981121 in samba (main) "package winbind 2:3.6.3-2ubuntu2 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/981121 [20:16] adam_g: i didn't see it hit the queue [20:17] adam_g: Ah, is it in ubuntu-server-dev pakcageset? [20:18] Daviey: it should be, i uploaded it yesterday [20:20] adam_g: don't think it is NOW. [20:20] sigh. [20:21] whmmv [20:21] roaksoax: the problem was the dnsmasq.conf did not specify a tftp-root path [20:22] $ edit_acl.py --person=gandelman-a --series=precise --source=horizon check [20:22] Adam Gandelman (gandelman-a) cannot upload horizon to Precise [20:22] Daviey: where can i get that script? [20:23] jkyle: I see [20:23] Daviey: i pushed update to lp:~ubuntu-server-dev/horizon/essex in the meantime [20:23] roaksoax: tried to put a tftp-root=/var/lib/tftpboot in the /etc/cobbler/dnsmasq.conf.template, but now it's not expanding the cobbler variables. I assume this is a maas debinstaller issue? [20:24] adam_g: lp:ubuntu-archive-tools [20:24] adam_g, did you fixed bug i filed :P [20:24] i had pointed zul about it [20:24] jkyle: maas doesn't really do anything to dnsmasq than just set 3 things, 1. domain, 2. network range, 3. gateway to hand out to the clients [20:25] koolhead17: yeah, as daviey spotted the apache config file for that dashboard needed an update [20:26] ooh okey. in order to have disabled DEBUG [20:26] and about the memcached/python-memcache dependency ? === dendrobates is now known as dendro-afk [20:28] hm, interesting [20:29] koolhead17: i didn't see a bug about memcache, but perhaps it should be added too [20:29] adam_g, horizon breaks if it does not have memcached and python-memcache [20:29] let me pastebin what all i found compulsary [20:29] :P [20:30] koolhead17: i believe you [20:30] but [20:30] python-keystoneclient, [20:30] python-memcache, [20:30] python-django-nose [20:30] its listed as a Depends of python-django-horizon === dendro-afk is now known as dendrobates [20:30] adam_g, and python-memcache depands on memcached? [20:31] correct me if am wrong? [20:32] koolhead17: no, it does not [20:32] koolhead17: it Suggests memcache [20:32] +d [20:32] adam_g, well then we need to pull in memcached as well [20:34] * koolhead17 rechecks his doc [20:42] adam_g, yes we need memcached python-memcache both and then the local_settings.py with CACHE_BACKEND = 'memcached://127.0.0.1:11211/' [21:35] New bug: #980956 in glance (main) "No module named MySQLdb from glance-registry" [Undecided,Invalid] https://launchpad.net/bugs/980956 === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [22:23] join cobbler [22:27] so, with maas...how can I script the adding of nodes? can I use cobbler or is there an api reference? [22:29] jkyle: there's an API you can access with curl [22:29] jkyle: though would be better if you use maas-enlist [22:29] jkyle: so all the nodes in the network will automatically enlist themselves into maas [22:32] roaksoax: ah, so you just serve a pxe image that selects teh maas-enlist debian-installer option at boot [22:33] jkyle: yes basically [22:33] ok, that'll work [22:34] jkyle: so basically, it pxe boots, downloads an image, maas-enlist-udeb during install, enlists itself into maas, then powers off [22:34] jkyle: and in maas you have to accept the enlistment and it will be ready for you to use [22:34] k, so, fixed that glitch in the dnsmasq. now when the nodes boot, it loads up the pxe initrd.gz...................ready [22:34] and pauses. I assume it's in a waiting state for provisioning by juju? [22:35] jkyle: what maas version are you using? [22:35] lemme check, fresh 12.04 [22:35] jkyle: it might be outputting to serial console instead of VGA [22:35] ii maas 0.1+bzr415+dfsg-0ubuntu2 Ubuntu MAAS Server [22:35] jkyle: ah yes, then the output is to serial console, that's why you are not seeing anythjing on VGA [22:36] jkyle: that's fixed in the latest release [22:36] roaksoax: him, it's been there for quite a while..hours [22:36] roaksoax: should I pull from the source or use the 12.04 packages? [22:36] jkyle: packages [22:37] jkyle: upgrade to the latest in archive, then run 'sudo maas-import-isos' [22:37] yep, there's the updates [22:43] hm, looking at the diffs there's a lot of default password and empty password action going on [22:43] jkyle: passwords should be updated automatically, you *must* accept overwritting the config files [22:43] ah, good to know [22:44]