[00:29] <smoser> bigjools_, "do you know a way of sending some PXE config that will make the machine power off again?"
[00:29] <bigjools_> hey smoser
[00:30] <smoser> ipmipower -h IPMI_ADDR -u USER -p PASSWORD --off
[00:31] <smoser> ^ is that what you were wanting bigjools_ ?
[00:31] <bigjools_> smoser: not exactly
[00:31] <bigjools_> it's easy to recognise the node is in the wrong state and execute a power command
[00:31] <bigjools_> but I just wondered if there was a pxe way
[00:32] <bigjools_> either way it would be nice to have something on the node's console so users are not surprised
[00:33] <smoser> oh. pxe.
[00:33] <smoser> duh.
[00:33] <smoser> you saidthat, i read ipmi
[00:33] <smoser> we have somthing in plae that does that.
[00:33] <smoser> i think.
[00:33] <smoser> right?
[00:35] <smoser> maybe i'm confused as to what youe're wanting.
[00:35] <smoser> while possibly more heavy, one solution is to just boot it into commissioning environment with user-data that says:
[00:35] <smoser> ==== this node is not supposed to be powered on ===
[00:36] <smoser> ==== it will power off in 120 seconds ====
[00:36] <smoser> ==== good-bye ====
[00:37] <bigjools_> ah that would work
=== bigjools_ is now known as bigjools
[01:55] <bigjools> roaksoax: I think we should move maas-enlist into maas trunk and build a binary package from it there
[01:56] <roaksoax> bigjools: yeah already on my todo for this week :)
[01:56] <bigjools> \o/
=== bigjools_ is now known as bigjools
[13:32] <smoser> gray--, so... you're sure that nothing has changed in routing since you ran that ?
[13:32] <smoser> ie, since you saw this issue:
[13:32] <smoser> http://pastebin.com/Wkguv4Fg
[13:33] <gray--> there has… i trashed the machine and rebuilt it
[13:33] <smoser> ah.
[13:33] <smoser> :)
[13:33] <gray--> if it's not reproducable, it's probably my error, not maas's error :)
[13:33] <smoser> ok. so thre is definitely a bug in that script, that woudl result in you seeing what you saw
[13:33] <smoser> (the stack trace there in that pastebin)
[13:33] <smoser> if you had no default route
[13:34] <smoser> where "that script" = maas-region-controller.postinst
[13:34] <gray--> i did have a default route, was one i checked
[13:34] <gray--> actually, wait, i take that back, i don't think i checked
[13:34] <gray--> balls, oh well
[13:35] <smoser> so your isse now, what is that?
[13:35] <gray--> i actually have the maas console in my browser, how about that? :)
[13:36] <gray--> so now… http://192.168.10.3/MAAS/settings/, that page seems to time out
[13:36] <gray--> i also have this in the apache error.log:
[13:36] <gray--> [Tue Jun 04 14:22:06 2013] [alert] (2)No such file or directory: mod_wsgi (pid=1267): Unable to change working directory to '/home/maas'.
[13:37] <gray--> after the actual install, i saw this: [Tue Jun 04 14:12:30 2013] [error] [client 192.168.10.1] DBusException: org.freedesktop.DBus.Error.FileNotFound: Failed to connect to socket /var/run/dbus/system_bus_socket: No such file or directory
[13:37] <gray--> but a reboot (eugh) fixed that
[13:38] <gray--> so, right now, the settings page is timing out
[13:39] <gray--> hrm, doesn't like being proxied to it seems
[13:39] <gray--> proxypassreverse could be my friend here, rather than an ssh -L to the server in question
[13:42] <gray--> i think this is probably turning away from a maas issue, towards an apache issue
[13:43] <gray--> smoser: thanks for your assistance thus far, i'll look at the apache thing for now, cheers
[13:47] <smoser> gray--, hm.. i've used ssh -L many times.
[13:47] <smoser> to remote maas.
=== kentb-out is now known as kentb
[14:01] <gray--> remote server address is 192.168.0.3, i hit up localhost in the browser, and hitting the banner that says 'Some cluster controllers are missing boot images.', it tries to send me to 192.168.0.3/MAAS/settings/
[14:01] <gray--> http://localhost/MAAS/settings works fine
[14:02] <smoser> hm..
=== Daviey_ is now known as Daviey
=== chuck_ is now known as zul
=== kentb is now known as kentb-out
[22:38] <mwhudson> is https://bugs.launchpad.net/maas/+bug/1177932 going to be fixed any time soon?
[22:38] <ubot5> Launchpad bug 1177932 in MAAS "Unable to select which pxe files to download by both series and architecture." [High,Triaged]
[22:38] <mwhudson> i could even spend an hour or two on it myself if it's going to be easy i suppose
[22:38] <bigjools> mwhudson: not soon, but scheduled for a few weeks away
[22:39] <bigjools> if you want to fix it, I'd not complain :)
[22:39] <mwhudson> bigjools: maybe we can have a quick skype/similar about how hard it would be this afternoon?
[22:42] <bigjools> mwhudson: sure - I need to be suitably caffeinated first and try to think how I'd fix it