[13:54] <epic_> how do i get out of this one: https://bugs.launchpad.net/ubuntu/+source/maas/+bug/979630 ? :)
[13:54] <ubot5`> Launchpad bug 979630 in maas (Ubuntu) "Unable to re-install maas after truncating/flushing database" [Medium,Confirmed]
[14:03] <epic_> yay I'm stuck on a 1 year old confirmed bug
[14:04] <epic_> which hangs apt-get which gets me into all sorts of trouble (two locked dbs)
[14:05] <roaksoax> epic_: how did you get there in the first place?
[14:05] <roaksoax> epic_: i would suggest yo remove everything, its dependencies, and install everything again
[14:06] <epic_> i purged the maas packages
[14:06] <epic_> thats how I got there
[14:07] <roaksoax> epic_: in precise, quantal, raring? how did you try to reinstall? was dbconfig-common purged as well?
[14:08] <epic_> 13.04 i believe
[14:09] <epic_> now I cannot purge.. "E: dpkg was interrupted, you must manually run 'dpkg --configure -a' to correct the problem." and if i run that command it fails again
[14:09] <epic_> with "FATAL:  password authentication failed for user "maas"" which locks up apt-get again
[14:09] <epic_> round and round ;)
[14:09] <roaksoax> epic_: so I'd be great to see what are the steps you followed and put those in the bug report
[14:10] <roaksoax> epic_: + all of the answers on debconf
[14:10] <roaksoax> epic_: maybe one combination of answers actually broke your environment when it shouldn't have
[14:10] <roaksoax> epic_: but for what I read, is that MAAS cannot connect to the DB
[14:10] <roaksoax> because there's a wrong password
[14:11] <epic_> yeah I understand, but I have not changed it afaik
[14:12] <roaksoax> yeah dbconfig-common that's that
[14:12] <roaksoax> does*
[14:12] <epic_> I just wanted to purge all maas config to to a clean install
[14:12] <epic_> and packages
[14:13] <epic_> now all purging ends in apt-get lock, which i need to hardkill, which leads to messyness
[14:15] <epic_> ooh purging the dbconfig-common seems to get me through
[14:15] <epic_> once again going to try apt-get install maas maas-dhcp maas-dns
[14:17] <epic_> seems to have worked
[14:17] <epic_> thanks for the help :)
[14:18] <epic_> the steps i took to reproduce was the same as in the bug; tried to purge packages, said yes to remove everything
[14:21] <epic_> maas-import-pxe-files does not seem to do anything though, no network traffic
[14:26] <roaksoax> that might be other issue. maas-import-pxe-files wget's from ubutnu archives
[14:29] <epic_> yeah I changed the pxe importer config file to raring, that seemed to get it going but it failed with a 404 on http://ports.ubuntu.com/ubuntu-ports//dists/raring/main/installer-armhf/current/images/highbank/netboot//vmlinuz , i guess raring is not there yet
[14:29] <roaksoax> epic_: raring is there yet, but mainy not for arm
[14:30] <roaksoax> epic_: edit the architectures to only include amd64 and i396
[14:30] <roaksoax> epic_: edit the architectures to only include amd64 and i386
[14:30] <epic_> ah, why did it try to dl arm ? :)
[14:30] <epic_> can i remove that arch?
[14:30] <roaksoax> its default
[14:30] <roaksoax> epic_: yeah, so just specify amd64 and i386 without arm
[14:30] <epic_> sweet
[14:37] <epic_> some commands (such as import-pxe) should have built-in screen, didn't realise it would take this long, now I'm stuck at work until it finishes :P
[15:22] <epic_> is there a way to get the maas-dhcp to only listen to one interface?
[15:26] <roaksoax> epic_: so when you configure the DHCP/DNS you specify to what interface to server the DHCP requests
[15:27] <roaksoax> epic_: you can do it through the WebUI or maas-cli
[15:28] <epic_> yeah i checked  /var/lib/maas/dhcpd-interfaces  and it pointed to the correct one (the iface not pointing out to my corporate network) :)
[15:29] <epic_> It seems everything is up and running, and I am rebooting a server of mine (it tries PXE at boot), how can I see that my maas controller is getting the PXE request?
[15:30] <roaksoax> epic_: /var/log/maas/pserv.log
[15:30] <epic_> seems that is not there, so pserv is not started..
[15:31] <roaksoax> epic_: yeah probably
[15:31] <epic_> huh start: Job is already running: maas-pserv
[15:31] <epic_> 0:01 /usr/bin/python /usr/bin/twistd -n --uid=maas --gid=maas --pidfile=/run/maas-pserv.pid --logfile=/dev/null maas-pser --config-file=/etc/maas/pserv.yaml
[15:31] <epic_> logfile dev null ? :D
[15:32] <epic_> but my server reboots into it's linux anyway, so it is not caught by the PXE boot it seems
[15:32] <roaksoax> epic_: the logfile is being get from the config in /etc/maas/pserv.yaml
[15:33] <epic_> /var/log/maas/pserv.log ..
[15:34] <epic_> a service restart spawned the file
[15:34] <epic_> its now empty
[15:35] <roaksoax> epic_: so change the zone name from master to something else
[15:35] <roaksoax> and make sure you enable DNS/DHCP management
[15:35] <roaksoax> and that should do it
[15:41] <epic_> no hits in the logs
[15:41] <roaksoax> epic_: so probably DHCP is not working
[15:41] <epic_> it still boots to the HD
[15:41] <roaksoax> epic_: remove the cluster controller from maas webui
[15:41] <roaksoax> epic_: then sudo dpkg-reconfigure maas-cluster-controller
[15:41] <roaksoax> accept it again in MAAS
[15:41] <roaksoax> and configure DNS/DHCP once again
[15:42] <roaksoax> and make sure you change the zone "master" to something else
[15:42] <epic_> ok
[15:42] <epic_> ok
[15:42] <roaksoax> or you can try that first, changeing the zone name from "master" to something else
[15:42] <epic_> i did that :)
[15:42] <epic_> DNS zone name?
[15:42] <roaksoax> yeah
[15:53] <roaksoax> robbiew: ping
[15:53] <roaksoax> err
[15:53] <roaksoax> robbiew: sorry :) was meant for rvba`
[15:53] <roaksoax> rvba`: ping
[16:54] <epic_> I should not need to add the node before i reboot it into pxe right?
[16:55] <roaksoax> epic_: nope
[16:59] <epic_> the two machines are connected via a ethernet cable and I can ping/ssh to it when it boots into the current ubuntu installation, so the dhcp (configured on the same interface) should reach it..
[17:02] <roaksoax> epic_: so go to settings
[17:02] <roaksoax> then Cluster Controllers
[17:02] <roaksoax> edit the cluster controller
[17:02] <epic_> yep
[17:02] <roaksoax> then go to the interface you want it to manage DHCP/DNS
[17:02] <roaksoax> does it show something like? eth010.16.255.255/16Manage DHCP and DNS
[17:02] <epic_> em210.0.0.254/24Manage DHCP and DNS
[17:03] <roaksoax> then does "DNS zone name" is different from master?
[17:03] <epic_> which is the inteface which is connected to the machine
[17:03] <epic_> yes i changed the zone name
[17:03] <roaksoax> epic_: check /etc/maas/dhcpd.conf is created
[17:03] <roaksoax> epic_: and check maas-dhcp-server is running
[17:04] <epic_> first is there
[17:04] <epic_> second:
[17:04] <epic_> ps aux | grep maas-dhcp-server
[17:04] <epic_> gives nothing
[17:04] <roaksoax> epic_: isc-dhcpd
[17:04] <roaksoax> err
[17:04] <roaksoax> dhcpd
[17:05] <roaksoax> epic_: as in: dhcpd     1018  0.0  0.1  19136  4912 ?        Ss   May09   0:01 /usr/sbin/dhcpd -f -q -4 -pf /run/maas/dhcp/dhcpd.pid -cf /etc/maas/dhcpd.conf -lf /var/lib/maas/dhcp/dhcpd.leases eth0
[17:05] <epic_> yep usr/sbin/dhcpd -user dhcpd -group dhcpd -f -q -4 -pf /run/maas/dhcp/dhcpd.pid -cf /etc/maas/dhcpd.conf -lf /var/lib/maas/dhcp/dhcpd.leases em2
[17:06] <epic_> does this dhcp have any log output?
[17:06] <roaksoax> epic_: is /etc/bind/maas/<zone-file> created according to the name?
[17:06] <roaksoax> accroding to the zone name you configured?
[17:06] <epic_> yep
[17:07] <roaksoax> ok so everything seems to be working jsut fin
[17:07] <roaksoax> epic_: what's the output of the nodes when they try to PXE boot?
[17:07] <epic_> PXE trying to find DHCP ...
[17:08] <epic_> hehe, right now i cannot view the output (its remote, i can only reboot it), but earlier today this was the output i saw
[17:08] <epic_> and i have not changed the bios boot settings after i saw the PXE looking for the DHCP
[17:08] <epic_> but does the maas dhcp have any log output? maybe i can see some errors
[17:17] <roaksoax> epic_: syslog is your best bet