=== Daviey_ is now known as Daviey [02:35] bigjools, thanks [02:38] evilnickveitch: pleasure as always === mwhudson is now known as zz_mwhudson [07:03] jtv: reviewing your branch now. [07:03] Thanks! === CyberJacob|Away is now known as CyberJacob [07:15] jtv: I assigned the "drop old relation" card to myself. I've got it in progress but of course it's still blocked for now. As I said in the email I sent you, I'll run the tests from time to time so we can see how far we are with the refactoring. [07:16] I did the same. [07:16] Okay :) [07:17] What upcomming changes are coming to MAAS with Ubuntu 14? [07:19] Lord_Set: a start is in the bugs list for MAAS on Launchpad, where you can select the 14.04 milestone: https://launchpad.net/maas/+milestone/14.04 [07:20] The Fix Committed / Fix Released ones are done. [07:20] Awesome thank you [07:20] There's also the blueprints; I'm looking up the link. [07:20] I was mainly curious about the new features if any :) But bug fixes are always welcome [07:21] Lots of grey area between them! But the features are in blueprints; see https://blueprints.launchpad.net/maas/ [07:21] and hopefully improved Openstack deployment with Juju [07:21] Or just the ones for 14.04: https://blueprints.launchpad.net/maas/1.5 [07:22] :) [07:30] Is Juju under different development than MAAS? [07:38] Yes, different groups. [07:38] They hang out in #juju. :-) [08:20] jtv: (talking about the "update "node" filter in networks listing" work) I'm wondering if we still want to pass node system IDs to the network listing API method. We probably want to pass MAC addresses now right? [08:20] rvba: not entirely sure tbh... [08:21] I wouldn't be surprised if we'd end up wanting both. [08:22] I was thinking to wait until we hear that somebody needs it. :) [08:24] jtv: okay, nodes are probably more central to MAAS than MACs so I'll just adapt the existing method. [08:26] Right. [08:26] The main thing is that we don't leave it to block the abolition of the old linking able. [08:27] Yep. That's the only remaining thing AFAIK. [08:30] It's how I found it. :) [09:18] factory.make_networks() still needs an option for either "only VLANs please" or "at most one non-virtual network please." [09:19] Otherwise we're going to have trouble when we start enforcing the rules for how MACs connect to networks. [09:19] It'll be easier though if we only enforce that at the form level... [09:21] jtv: I'm getting this from time to time: http://paste.ubuntu.com/6969880/ [09:22] Would you mind having a look? [09:22] That's what I'm talking about. [09:22] The fix is landing. [09:22] Ah, ok, cool :) [09:23] jtv: I've got a branch up for review if you have time for it: https://code.launchpad.net/~rvb/maas/update-node-filtering/+merge/207603 [09:25] rvba: careful — the new relationship introduces an additional way for a call like this to go wrong. [09:26] I'll look at it after the call. [09:26] jtv: hum, I see what you mean, let me add a test… [10:00] jtv: I've improved one of the tests on https://code.launchpad.net/~rvb/maas/update-node-filtering/+merge/207603 [10:00] jtv: please have a look when you have some time. [10:00] OK. [10:02] Diff's still updating... weird how that's usually very fast now, and then sometimes slow like this. [10:02] jtv: the improvement is minor, you can review what you see now. [10:03] nm it just updated [10:06] Done [10:07] Thanks. [12:18] rvba: my branch is up for review. After that, ditching the old linking table still breaks maasserver.views.tests.test_networks.NetworkDetailViewTest.test_network_detail_displays_node_count [12:29] It looks easy to fix, but there is a pitfall of double-counting. === sfeole` is now known as sfeole === arges_ is now known as arges === medberry is now known as Guest5611 === Guest5611 is now known as med___ === med___ is now known as med_ === CyberJacob is now known as CyberJacob|Away === medberry is now known as med_ === freeflying is now known as freeflying_away