[02:31] <StevenK> wgrant: Bad you for using Equal/NotEqual for None
[02:32] <wgrant> StevenK: Where?
[02:32] <wgrant> In my IBB rework?
[02:33] <StevenK> No, your log fix for r16750
[02:33] <wgrant> That was cjwatson, I just reviewed it and missed that :)
[02:35] <StevenK> % bzr grep 'assertEqual.*None' | wc -l
[02:35] <StevenK> 512
[02:35] <StevenK> So I can't complain too much. :-)
=== jamesh_ is now known as jamesh
=== tasdomas_afk is now known as tasdomas
[08:10] <stub> Who's up for a Swift branch review?
[08:49] <cjwatson> StevenK: I had to use assert[Not]Equal there, because that mixin method is used in test case classes that derive from TrialTestCase rather than from the LP TestCase variants, and that don't have assertIs[Not]
[08:50] <cjwatson> StevenK: (I tried using the more correct methods first, of course)
[08:50] <cjwatson> No doubt we should fix that, but I didn't feel like pulling in a complete test rework
[08:52] <stub> I just use self.assert_(foo is None), even if the test fail message is suckier.
[08:53] <stub> Or maybe I abuse assertEqual too... sucks this is all so fragmented.
[08:54] <cjwatson> I was following usage in nearby tests, anyway.
[08:54] <cjwatson> W: Conflicting distribution: http://gb.archive.ubuntu.com devel Release (expected devel but got saucy)
[08:54] <cjwatson> Oh dear
[08:54] <cjwatson> I did not predict that :-(
[09:09] <wgrant> Huh, really.
[09:31] <mpt> Wow, dogfood.launchpad.net still exists, separately from staging.launchpad.net
[11:58] <lifeless> mpt: yes, dogfood was an entirely separate cluster IIRC
[12:44] <wgrant> "cluster"
[12:50] <StevenK> cjwatson: Ah ha, Trial impacts again.
[12:50] <StevenK> lifeless: "is"
[12:51] <lifeless> wgrant: "pedant"
[12:52] <StevenK> lifeless: Duh? That's wgrant means.
[12:53] <lifeless> StevenK: EPARSE :)
[12:54] <StevenK> lifeless: As in duh, wgrant is just another word for pedant
[12:56] <lifeless> StevenK: Ah :).
[13:20] <cjwatson> wgrant: Hey, it has builders attached.  That makes it a cluster, right? :)
[13:21] <wgrant> Heh, I guess it does now.
=== tasdomas is now known as tasdomas_afk
=== smartboyhw_ is now known as smartboyhw
=== BradCrittenden is now known as bac
[20:06] <cjwatson> wgrant: I'm seeing a lot of tracebacks like this in buildd-manager:
[20:06] <cjwatson> https://pastebin.canonical.com/96660/
[20:06] <cjwatson> is that new?
[21:13] <lifeless> cjwatson: private bin :(; probably a -ops channel question?
[22:08] <cjwatson> sorry, got confused about where I was
[22:08] <cjwatson> http://paste.ubuntu.com/6045865/
[22:40] <lifeless> cjwatson: I don't remember seeing that before
[22:43] <cjwatson> wgrant did quite a lot of rearrangement of buildd-manager that went into the last deployment, and I suspect a regression from that
[22:43] <cjwatson> unless it's one of my buildd-manager changes, but I don't think it is
[22:51] <lifeless> seems likely
[23:43] <wgrant> Hi
[23:44] <wgrant> Hmm
[23:44] <wgrant> Probably a regression from my stuff, but I tested TTBJs.
[23:46] <wgrant> I think it only happens when they fail.
[23:46] <wgrant> And anyway, TTBs aren't critical, and this doesn't seem to have wide impact, so can probably wait until Monday.
[23:46] <wgrant> cjwatson: Thanks for poking.
[23:48] <cjwatson> ok