[05:59] <bigjools> rvba: replied to your comment.
[06:07] <bigjools> rvba: could do with a pre-imp chat if you're awake now
[06:08] <rvba> bigjools: sure
[06:08] <bigjools> I'll call
[06:51] <bigjools> rvba: https://bugs.launchpad.net/bugs/1319644 is possibly a dupe
[06:56] <rvba> bigjools: that's what I thought but I cannot find a dupe…
[06:57] <bigjools> it might be under a wildly different title but I'm sure I've seen it
[06:57] <bigjools> tag search for maas-cli?
[06:57] <bigjools> GAH we've got cli AND maas-cli as tags
[06:57] <bigjools> fuxacke
[06:57] <rvba> bigjools: we really need to fix the cli (that's why I triagged that bug 'High').  The bugs are piling up.
[06:58] <bigjools> rvba: you're conflating "importance" and "priority"
[06:58] <bigjools> we *need* to fix *all* bugs, but .... :)
[06:58] <bigjools> low is realistic
[06:59] <rvba> bigjools: yeah, I know.
[07:01] <rvba> bigjools: it's just that I have the feeling that a lot of these bugs are related.  If we start fixing some of them others will be fixed as side-effects.
[07:01] <bigjools> rvba: well, we all have spare evenings right? :)
[07:01] <rvba> bigjools: heh
[07:19] <bigjools> rvba: bugger, get_or_create is screwing me over :(
[07:19] <bigjools> one of its args is not a key field so it makes a new object when it should not
[07:22] <bigjools> this is what happens when tests make too many damn assumptions.
[07:22] <bigjools> oh well I will go and cook steak and come back pumped later
[09:06] <bigjools> rvba: if you would care to do the review since we pre-imped it? https://code.launchpad.net/~julian-edwards/maas/supported-subarches-api/+merge/219664
[09:07] <rvba> bigjools: sure
[09:07] <bigjools> actually let me change ooooone line in the test
[09:07] <bigjools> oh actually never mind
[09:07] <bigjools> see if you spot it :)
[09:08] <rvba> okay :)
[09:08] <bigjools> back in about 15m, relocating
[09:39] <jtv> bigjools: in order to fix the proxy settings on the nodes, we'll need to tweak preseeds for both d-i and curtin.
[09:40] <bigjools> figures
[09:41] <bigjools> is that straightforward?  I caught the tail end of your conversation with scott on scrollback
[09:43] <jtv> It's probably straightforward, if we can figure out how, but it signals a potential need to unify interfaces for preseeding.
[09:43] <bigjools> well - we might get rid of d-i
[09:43] <bigjools> so could be moot :)
[09:43] <jtv> Whether by ditching d-i, or adding the proverbial One More Layer.
[09:44] <bigjools> preseeds are complicated, but given that curtin is new-ish and easy to change for us, we could make it more modular
[09:49] <bigjools> rvba: huh, I didn';t know about defaults
[09:50] <bigjools> rvba: also, you noticed what I was going to change in the test :)
[09:50] <bigjools> well done!
[09:50] <bigjools> lol
[09:50] <rvba> \o/
[12:53] <jtv> gmb: I'll be available to pick up bootresources-related jobs tomorrow.  Just pick something off the board?
[12:53] <gmb> jtv: Yar. I’ll email the devel list at my EoD with what’s RTC.
[12:53] <jtv> OK
[13:23] <rvba> gmb: I've started working on the API side of things.
[13:40] <gmb> rvba: Excelltn, thanks.
[15:40] <gmb> allenap, rvba: What’s the name of the maas dhcp service again? maas-dhcp doesn’t exist, nor do several other permutations… I assume I’m being forgetful.
[15:41] <allenap> gmb: maas-dhcp-server
[15:41] <gmb> GAH!
[15:41] <gmb> Thanks
[15:42] <gmb> IT’S ALIVE
[15:42] <gmb> Blinkenlights