=== CyberJacob is now known as CyberJacob|Away
[01:30]  * mwhudson is trying to enlist some midway nodes
[01:30] <mwhudson> and it's failing
[01:32] <mwhudson> uh
[01:32] <mwhudson> is this expected to work even
[01:32] <mwhudson> /etc/maas/templates/pxe/config.commissioning.template sure doesn't look like it will work on armhf
=== CyberJacob|Away is now known as CyberJacob
=== vladk|offline is now known as vladk
=== CyberJacob is now known as CyberJacob|Away
[08:26] <rvba> allenap: btw, if you have 10 minutes later today, I'd like to talk briefly about maas-test.
[08:26] <allenap> rvba: I will have 10 minutes later today :)
[08:26] <rvba> allenap: heh
[08:27] <allenap> rvba: I’ll ping you once I’ve finished writing the email I’m on.
[08:27] <rvba> allenap: cool, ta
[09:41] <gmb> rvba: Do you ahve time for a quickish pre-imp? I think that there are some improvements to be made to Andre’s code, but I’d like to run them past you in your post as Djangoist-in-chief.
[09:46] <rvba> gmb: sure
[09:47] <gmb> rvba: Cool.
[09:57] <rvba> gmb: you might want to link your branch to the bug Andres "fixed" with his change.  This way, it will get picked up by whoever does the release later this week.
[09:58] <gmb> rvba: Good idea! Thanks.
[09:58] <gmb> Oh!
[09:59] <gmb> rvba: Except there’s no bug linked to the original branch…
[10:01] <rvba> :/
[10:01] <gmb> *sigh*
[10:01] <rvba> I guess it should be linked to the FFe bug.  But let's ask Andres.
[10:03] <rvba> gmb: it's not obvious to me if your thing fits into the "component error" model.  If it does, then great.  If not, then you need to add a new key in the dict global_options() (src/maasserver/context_processors.py) returns.
[10:04] <gmb> rvba: I’ve just arrived at exactly the same conclusion :)
[10:04] <gmb> It doesn’t fit.
[10:04] <gmb> Well, it *sort of* does, but not enough for me to go to the effort to make it fit.
[10:04] <gmb> rvba: Adding a new key is much, much simpler.
[10:06] <rvba> gmb: now that I look at Andres' code, maybe the warning is intended to only be displayed on the node listing page.
[10:06] <gmb> rvba: It should at least be shown on the add Node page, too, surely. Mind you, that *is* the Node listing page with a JS form, isn’t it. Hmm...
[10:07] <rvba> Yes
[10:07] <gmb> rvba: So, simplest solution is just to a check for is_superuser and add some tests.
[10:07] <rvba> Yeah.
[10:07] <gmb> Simple wins.
[10:08] <rvba> Also, missing space before the 'if' in '{%if third_party_drivers and drivers %}'
[10:08] <rvba> While we're at it :)
[10:08] <gmb> I’d noticed :)
[10:12] <rvba> gmb: also, the url shown in the UI can be "improved" by adding '#third_party_drivers' to it.  This way it will point to the right anchor directly.
[10:12] <gmb> rvba: Ah, yes, good thinking.
[10:25] <gmb> Holy testfailures batman!
[10:47] <gmb> So, new thing I’ve learned: calling reverse() at import time (as opposed to in a function) causes ALL THE THINGS to blow up.
[10:52] <rvba> gmb: yes, Django does some funky stuff when an application is imported (registers the views, etc.).
[10:53] <rvba> gmb: btw, I'm thinking that the first part of the message (all the text but the link) applies to all users, not only superusers.
[10:53] <gmb> rvba: Do the users care though?
[10:54] <gmb> Eh, I supposed they might.
[10:55] <gmb> rvba: Easy to fix. I’ll get on it.
[12:50] <rvba> allenap: can you please have a look at https://code.launchpad.net/~rvb/maas/arm-bug-1307780/+merge/215861 ?
[12:52] <jjkuj> hej
[13:09] <allenap> rvba: Sure.
[13:11] <allenap> rvba: Done, +1.
[13:12] <rvba> allenap: ta
[13:12] <rvba> allenap: tell me when you have a couple of minutes to talk about maas-test.
[13:12] <rvba> (It's not urgent)
[13:13] <gmb> rvba, allenap: My test-addtion branch is now up for review: https://code.launchpad.net/~gmb/maas/add-tests-bug-1307906/+merge/215870
[13:18] <rvba> gmb: will have a look in a minute
[13:18] <gmb> Thanks.
[13:33] <gmb> rvba: Thanks for the review :)
[13:33] <rvba> np
[13:35] <gmb> allenap: I have a spurious test failure in the RPC tests: src/maasserver/rpc/tests/test_regionservice.py", line 667, in test_prepare_holds_startup_lock().
[13:36] <gmb> failed in a branch that was documentation changes only.
[13:36] <gmb> allenap: I’ll file a bug; want me to disable the test in the meantime?
[13:45] <allenap> gmb: If it’s blocking you, sure, do eet.
[13:45] <allenap> rvba: Wanna talk?
[13:45] <gmb> allenap: Well, it wasn’t blocking: failed once. But I’ve put up a branch to disable it anyway because we hates wobbly tests, my precioussssss.
[13:58] <rvba> allenap: sure, let's talk
[15:13]  * gmb -> afk, evening stuff. might be back later
=== roadmr is now known as roadmr_afk
=== vladk is now known as vladk|offline
=== CyberJacob|Away is now known as CyberJacob
=== roadmr_afk is now known as roadmr
[21:06] <mwhudson> rvba: thanks for fixing that armhf thing so quickly
[21:31] <mwhudson> oh yeah, another maas oddity: i got dns failures when i specified my proxy by name
[21:32] <mwhudson> it worked fine when i specified an ip address
[21:32] <mwhudson> does this sound like anything you have run into?
[21:40] <tych0> hi maas folks, i just did a clean install of current trusty maas on a fresh format of trusty, and i got this: http://paste.ubuntu.com/7257823/
[21:40] <tych0> (in /var/log/apache2/error.log)
[21:40] <tych0> any ideas?
[21:41] <tych0> ah ha :-)
[21:41] <tych0> i bet it is the stuff in /etc/maas
[21:41] <tych0> don't mind me, then :-)
=== CyberJacob is now known as CyberJacob|Away
=== CyberJacob|Away is now known as CyberJacob
[23:25] <mwhudson> how long is the lease time in maas's dhcp server by default?
[23:25] <mwhudson> i see this:
[23:25] <mwhudson>        class "PXE" {
[23:25] <mwhudson>           match if substring (option vendor-class-identifier, 0, 3) = "PXE";
[23:25] <mwhudson>           default-lease-time 30;
[23:25] <mwhudson>           max-lease-time 30;
[23:25] <mwhudson>        }
[23:25] <mwhudson> but i don't know what that really means :)
=== CyberJacob is now known as CyberJacob|Away