| === bigjools-afk is now known as bigjools | |
| [06:09] <lifeless> bigjools: which bug # for clarity ? | |
| [06:10] <bigjools> lifeless: https://bugs.launchpad.net/juju/+bug/945505 | |
| [06:10] <ubot5> Ubuntu bug 945505 in juju "Use ipAddress instead of dnsName now that txaws supports it" [High,In progress] | |
| [06:11] <bigjools> esp see comment 8 | |
| [06:11] <bigjools> dunno if that' all aws-specific though | |
| [06:11] <bigjools> (on second viewing) | |
| [06:14] <lifeless> bigjools: its all AWS specific | |
| [06:14] <bigjools> then we're all good | |
| [06:15] <lifeless> bigjools: more details in mail | |
| [09:02] <rvba> mrevell: want to talk about the hexr stuff? | |
| [09:16] <mrevell> rvba, Sure! Was just on another call. I'll set up a hangout | |
| [12:23] <melmoth> hmm, at wich stage is dnsmasq suppose to have set a new entry for the new host so that its name is resolvable ? | |
| [12:24] <melmoth> i have just bootsraped juju on a maas environment, it s using the only node i defined right now: zookeeper.mydomain.com (enlisted manually) | |
| [12:24] <melmoth> and while the system is being installed, i cannot resovled its name. | |
| [12:24] <melmoth> so, juju will not able to log in it, will it ? | |
| [12:27] <melmoth> i see the entry for the node in /var/lib/misc/dnsmasq.leases, but still, the name cannot be resolved | |
| [12:43] <rbasak> allenap: I don't understand https://bugs.launchpad.net/maas/+bug/1044393/comments/4 | |
| [12:43] <ubot5> Ubuntu bug 1044393 in MAAS "Tests fail with "HTTPError: HTTP Error 503: Service Unavailable"" [Low,Triaged] | |
| [12:43] <rbasak> allenap: is this against the wrong bug? | |
| [12:48] <melmoth> i try the workarond mentionned in https://bugs.launchpad.net/ubuntu/+source/maas/+bug/1043121 , but i still have the same behaviour | |
| [12:48] <ubot5> Ubuntu bug 1043121 in maas (Ubuntu) "deployed node cannot be looked up with dnsmasq on MAAS" [Undecided,New] | |
| [12:48] <melmoth> anyone knows exactly at wich stage is dnsmasq suppose to resolve the name ? | |
| [13:09] <allenap> rbasak: Nope, correct bug. I can reproduce the error, and that, I think, is the root cause. | |
| [13:10] <allenap> Well, on the way to the root cause. | |
| [13:15] <rbasak> allenap: ok, I've tested and commented on the bug | |
| [13:15] <rbasak> allenap: (your step to reproduce doesn't work for me) | |
| [13:16] <allenap> rbasak: Interesting. I hate bugs. | |
| [13:16] <allenap> like this. | |
| [13:39] <melmoth> is there a way to bootstrap on a node of my choice (the same way i can jitsu deploy-to a specific node) ? | |
| [13:40] <jtv> You can pass a "name" constraint that must map the node's hostname. | |
| [13:49] <melmoth> hmmm | |
| [13:50] <allenap> melmoth: In 12.10 there will be much richer placement/hardware constraint support. Obviously that's not useful right now, but if you're just evaluating then it may be worth coming back to it then. | |
| [13:51] * melmoth wonders if 'juju bootstrap --constraints "instance-name=zookeeper" ' is already suppose to work | |
| [13:53] <melmoth> may be maas-name=zookeeper | |
| [14:09] <melmoth> juju bootstrap --constraints "maas-name=zookeeper.mydomain.com" | |
| [14:09] <melmoth> works ! \o/ | |
| [15:09] <allenap> roaksoax: Hello! I have some questions about packaging. To start with: is there an easy way to attempt packaging with my current branch, instead of with whatever's at the top of the changelog? | |
| [15:12] <allenap> roaksoax: I've tried `bzr builddeb --export-upstream=...` but I'm just waving my arms in the dark really. | |
| [15:12] <roaksoax> allenap: your branch on lp:~allenap etc etc or a different revision? | |
| [15:13] <allenap> roaksoax: I want to be able to use a local branch, just the one I'm working on. I'm trying to write a script to build my current branch and run tests on it. | |
| [15:14] <roaksoax> allenap: uhmm maybe try modifying debian/rules | |
| [15:14] <roaksoax> allenap: the other one is that you create an .orig.tar.gz yourself | |
| [15:16] <allenap> roaksoax: So, walk me through creating the .orig.tar.gz. I can bzr export the branch into a tarball, but what do I call it so that the builddeb finds it and uses it? | |
| [15:17] <roaksoax> allenap: what i would do is simply tar zxvf .orig.tar.gz && cp debian maas-<whatever> and then: cd maas-<whatever> && debuild -S -sa | |
| [15:17] <roaksoax> allenap: note that the tarball has to be the same version as what it is in the changelog | |
| [15:18] <roaksoax> allenap:btw.. how can I add kernel parameters to a system? | |
| [15:19] <roaksoax> allenap: i don't want to modify provisioningserver/pxe/config.template | |
| [15:19] <allenap> roaksoax: File a bug :) | |
| [15:19] <allenap> We will move those templates to somewhere easier to customise once installed, but... time. | |
| [15:19] <roaksoax> allenap: heh.. ok.. I thought we had agreed that it was important for us to be able to modify kernel parameters for the system | |
| [15:20] <roaksoax> allenap: as in, being able to add kernel parameters from the API/WebUI | |
| [15:20] <allenap> It is. What's your use case? | |
| [15:20] <roaksoax> and IIRC< that was a request we made | |
| [15:21] <roaksoax> allenap: right now not my particular case, but for example, sabdlf just lost serial interface output due to not having the kernel parameters. And really there's no easy way to add them without having to modify those config files | |
| [15:22] <allenap> roaksoax: Find or file a bug, add a comment and mark it critical. | |
| [15:24] <roaksoax> allenap: will do | |
| [15:25] <allenap> roaksoax: I'd like to give you a better answer, but we don't have any slack right now. | |
| [15:27] <roaksoax> allenap: yeah we are all like that | |
| [15:31] <roaksoax> allenap: Anyways, if there's nothing else I can help you with, I'll go resume my holiday :) | |
| [15:31] <allenap> roaksoax: Ah, I didn't realise you were away. Sorry about that, have a good one :) | |
| [15:32] <roaksoax> allenap: hehe no worries, either way I have to do school work so I'm kinda around | |