| === CyberJacob|Away is now known as CyberJacob | |
| === melmoth_ is now known as melmoth | |
| [08:47] <jtv> allenap, bigjools: my current test failure is a test timeout — | |
| [08:47] <jtv> Unfired and/or unhandled EventualResult(s); | |
| [08:47] <jtv> Unfired/unhandled EventualResult | |
| [08:47] <jtv> It was connected to a Deferred. | |
| [08:47] <jtv> Are you getting the same thing? | |
| [08:47] <allenap> jtv: Interesting. Do you want me to take a look at your branch? | |
| [08:47] <jtv> Oh yes please. | |
| [08:48] <jtv> Let me push it up. | |
| [08:50] <jtv> allenap: https://code.launchpad.net/~jtv/maas/configure-dhcpv6/+merge/232179 | |
| [08:51] <jtv> Conflicts. Fix0ring. | |
| [09:03] <jtv> allenap: diff has updated. If you look at TestConfigureDHCPv6, you'll see the test that's breaking for me. (It's still got an undefined placeholder in there, rpc_configure_dhcpv6, but that's past the failure.) | |
| === CyberJacob is now known as CyberJacob|Away | |
| [09:19] <gmb> allenap: Why would I be seeing this in my testcase for the DHCP probe service: | |
| [09:19] <gmb> AttributeError: 'TestDHCPProbeService' object has no attribute 'expectThat' | |
| [09:20] <gmb> allenap: More correctly, full traceback here: http://pastebin.ubuntu.com/8148330/ | |
| [09:20] <jtv> Wrong base class for your test case? | |
| [09:20] <gmb> jtv: I thought so, but I'm using the same base class (PservTestCase) as bigjools used for his TimerService subclass. | |
| [09:21] <jtv> Outdated dependencies then? | |
| [09:21] <gmb> Ah, that could be it. | |
| [09:23] <gmb> jtv: That's got it, thanks. | |
| [09:23] <jtv> \o/ | |
| [09:28] <jtv> allenap: do my tests need to provide their own implementation for the RPC function I'm calling on the fixture? | |
| [09:32] <allenap> jtv: I think I’ve figured your branch out. I’ve switched to the “live” fixture, but there’s a bug in it. | |
| [09:33] <jtv> From what I read I think the live fixture would do me just fine. | |
| [09:33] <jtv> Of course that's a bit like buying parts for an experimental fusion reactor: it all sounds just fine. But the eventual explosion won't surprise me either. | |
| === jpds_ is now known as jpds | |
| [10:00] <allenap> jtv: http://paste.ubuntu.com/8148612/ is what I have so far. | |
| [10:02] <allenap> jtv: Yes. For each command you pass into makeCluster() a Mock is provided on the returned protocol instance. You can customise these. E.g. protocol.ConfigureDHCPv6.return_value = defer.succeed({…}) | |
| [10:03] <jtv> allenap: thanks! Looking... | |
| [10:19] <jtv> allenap: tests pass! | |
| [10:19] <jtv> That's with your changes, and the stuff that was wrong in my original tests fixed up. | |
| [10:26] <gmb> allenap: Can I trouble you to take a look at https://code.launchpad.net/~gmb/maas/probe-dhcp-2.0/+merge/232191 when you have a chance? | |
| [10:27] <gmb> I think that's got all the bits now. | |
| [10:55] <allenap> jtv: I’m a dolt; waiting for the handshake is a solved problem, because getClientFor can take a timeout: http://paste.ubuntu.com/8148963/ | |
| [10:57] <jtv> Even better. :) | |
| [11:15] <jtv> allenap: updated MP is at https://code.launchpad.net/~jtv/maas/configure-dhcpv6/+merge/232192 — it has a small prerequisite branch to keep the diff down. | |
| [12:23] <jtv> allenap: is the package not reading the common celery config any more? A custom package I built hits errors where app_or_default().conf only seems to hold rabbit settings and such. | |
| [13:01] <allenap> jtv: I don’t know. I don’t recall changing anything there... | |
| [13:01] <jtv> Maybe it's just a different execution environment... | |
| [13:01] <jtv> Anyway, not all that interesting a question given that we're ditching Celery. I put up a branch to make those settings variables. | |
| [13:01] <jtv> Regular variables in the module that needs them. | |
| === jfarschman is now known as MilesDenver | |
| [14:42] <gmb> allenap: Re: your suggestion about adding an errback… is this the kind of thing you meant? https://plus.google.com/hangouts/_/canonical.com/gavin-graham | |
| [14:42] <gmb> Er | |
| [14:42] <gmb> http://paste.ubuntu.com/8150461/ | |
| [14:42] <gmb> allenap: ^^ | |
| [14:44] <allenap> gmb: Yeah, that works. As you’re using inlineCallbacks you could use try:…except: around the yield self._probe_dhcp(…) too. But either is just fine. | |
| [14:44] <gmb> allenap: I’ll stick with the errback; less levels of indentation. | |
| [14:44] <gmb> Thankee! | |
| [15:47] <allenap> gmb: I see your branch just bounced with some lint and something else. I think my branch (here: https://code.launchpad.net/~allenap/maas/race-on-start-region-service/+merge/232233) will fix that. | |
| [15:48] <gmb> allenap: Ah, ta! | |
| [16:08] <gmb> allenap: Rabbit shat all over your branch | |
| [16:15] <allenap> gmb: It knows its days are now numbered in double digits. | |
| [16:16] <allenap> gmb: It landed second time around, so give your branch another go. | |
| [16:26] <matsubara> gmb, allenap: You wouldn't be saying that with such impunity if it was the one from Caerbannog | |
| === matsubara is now known as matsubara-lunch | |
| [16:31] <gmb> lol | |
| [16:55] * gmb -> outta here | |
| === matsubara-lunch is now known as matsubara | |
| [17:49] <allenap> matsubara, gmb: Indeed! Henceforth the RPC thing shall be known as the Holy Hand Grenade of Antioch. | |
| [17:56] <matsubara> hehe | |
| === roadmr is now known as roadmr_afk | |
| === jfarschman is now known as MilesDenver | |
| [18:37] <LaTaPiaT> e ai matsubara | |
| [18:37] <LaTaPiaT> ta vendo como soube te achar kkkkk | |
| === roadmr_afk is now known as roadmr | |
| === roadmr is now known as roadmr_afk | |
| === roadmr_afk is now known as roadmr | |
| === jfarschman is now known as MilesDenver | |
| === allenap_ is now known as allenap | |
| === jfarschman is now known as MilesDenver | |
| [23:26] <mattrae> hi guys, i have nodes that need to pxe on a particular vlan. when enlistment boots though it doesnt' appear to be using the vlan. is there a way to specify a vlan to use during enlistment? | |
| [23:26] <mattrae> using maas 1.5 | |
| [23:58] <bigjools> for the benefit of others watching, maas doesn't support tagged vlans for pxe | |